NATIONAL INSTITUTE FOR HEALTH AND CLINICAL EXCELLENCE. Single technology appraisal (STA)

Transcription

1 NATIONAL INSTITUTE FOR HEALTH AND CLINICAL EXCELLENCE Single technology appraisal (STA) Everolimus (Afinitor) in combination with exemestane for the treatment of advanced or metastatic HER2 negative, hormone receptor positive breast cancer after prior endocrine therapy Final 22 November 2012 Specification for manufacturer/sponsor submission of evidence Page 1 of 292

2 Contents NATIONAL INSTITUTE FOR HEALTH AND CLINICAL EXCELLENCE... 1 Single technology appraisal (STA)... 1 Contents... 2 List of tables and figures... 3 List of abbreviations... 8 Instructions for manufacturers and sponsors Executive summary Section A Decision problem Description of technology under assessment Context Equality Innovation Statement of the decision problem Section B Clinical and cost effectiveness Clinical evidence Cost effectiveness Section C Implementation Assessment of factors relevant to the NHS and other parties References Appendices Internal Model QA: Specific Validation Checks Related procedures for evidence submission Specification for manufacturer/sponsor submission of evidence Page 2 of 292

3 List of tables and figures Figures Figure A1 Sequential systemic endocrine therapy for women with HR+ breast cancer, as defined by NICE Figure B1 PRISMA diagram of included and excluded studies in the systematic review Figure B2 CONSORT diagram for BOLERO-2 at the 7-month interim analysis Figure B3 CONSORT diagram for the TAMRAD trial at 24 months Figure B4 PFS at 7 months of follow-up based on local (a) and central (b) assessment and at 18 months based on local (c) and central (d) assessment in BOLERO Figure B5 PFS across various subgroups at the 7-month interim analysis in BOLERO Figure B6 Overall survival at 16 months in BOLERO Figure B7 Change in bone turnover markers at 6- and 12-weeks in BOLERO Figure B8 Incidence of disease progression in bone in (a) the overall population and (b) the subgroup of patients with bone metastases at baseline in BOLERO Figure B9 PFS in the subgroup of patients with bone-only metastases at the 18-month analysis in BOLERO Figure B10 Time to definitive deterioration (MID 5%) in EORTC QLQ-C30 Global Health Status score in BOLERO Figure B11 Time to progression in the intention-to-treat population in the TAMRAD trial Figure B12 Overall survival in the intention-to-treat population in the TAMRAD trial Figure B13 Evidence network of trials used to inform the indirect treatment comparison Figure B14 Forest plot of log hazard ratios for PFS/TTP comparison of fulvestrant (250 mg) versus exemestane Figure B15 PFS or TTP (months) reported for a) endocrine monotherapy or b) chemotherapy in phase III studies in women with HER2 advanced breast cancer Figure B16 Model structure Figure B17 Overall survival, Weibull parametric curve: base case (everolimus in combination with exemestane vs exemestane alone) Figure B18 Overall survival, all parametric functions fitted: everolimus arm Figure B19 Overall survival, all parametric functions fitted: exemestane arm Figure B20 Progression-free survival, Weibull parametric curve (base case) Figure B21 Progression free survival, all parametric functions fitted: exemestane arm Specification for manufacturer/sponsor submission of evidence Page 3 of 292

4 Figure B22 Progression free survival, all parametric functions fitted: exemestane arm Figure B24 Naïve chained comparison of chemotherapy with everolimus 160 Figure B24 Everolimus in combination with exemestane vs exemestane alone Markov Trace Figure B25 Exemestane alone Markov Trace Figure B26 Tamoxifen Markov Trace Figure B27 Doxorubicin Markov Trace Figure B28 Fulvestrant Markov Trace Figure B29 Capecitabine Markov Trace Figure B30 Docetaxel Markov Trace Figure B31 Everolimus in combination with exemestane vs exemestane alone Cumulative QALYs Figure B32 Exemestane alone Cumulative QALYs Figure B33 Tamoxifen Cumulative QALYs Figure B34 Doxorubicin Cumulative QALYs Figure B35 Fulvestrant Cumulative QALYs Figure B36 Capecitabine Cumulative QALYs Figure B37 Docetaxel Cumulative QALYs Figure B38 Probabilistic sensitivity analysis plane (everolimus in combination with exemestane versus exemestane alone: WTP 30K) Figure B39 Cost-effectiveness acceptability curve (everolimus in combination with exemestane versus exemestane alone) Figure C1 WinBUGS model used to generate hazard ratios for indirect treatment comparison Specification for manufacturer/sponsor submission of evidence Page 4 of 292

5 Tables Table A1 Unit costs of technology being appraised Table A2 Overall survival according to stage of disease in women with breast cancer Table A3 Key Assumptions and flow for calculating Advanced Breast Cancer Population (ABC): Table A4 Key Assumptions and flow for calculating total eligible population for everolimus from Advanced Breast Cancer population Table B1 Eligibility criteria used in search strategy Table B2 List of relevant RCTs Table B3 Comparative summary of methodology of the RCTs Table B4 Eligibility criteria in the RCTs Table B5 Characteristics of participants in the RCTs across randomised groups Table B6 Primary and secondary outcomes of the RCTs Table B7 Analyses for BOLERO Table B8 Summary of statistical analyses in RCTs Table B9 Patient disposition at the 18-month interim analysis of BOLERO Table B10 Quality assessment results for RCTs Table B11 Summary of the BOLERO-2 18-month efficacy data Table B12 PFS based on local and central assessment at 7-, 12- and 18- month analyses in BOLERO Table B13 Overall survival at 7-, 12- and 18-month interim analyses in BOLERO Table B14 CBR and ORR at 7-, 12- and 18-month interim analyses (local assessment) in BOLERO Table B15 HRQoL in BOLERO-2, measured by EORTC QLQ-C30, at 7-, 12- and 18-month interim analyses Table B16 Clinical benefit rate in full population and subgroups in the TAMRAD trial Table B17 Time to progression at median 24-month follow-up in the TAMRAD trial Table B18 Overall survival at median follow-up of 24-month in the TAMRAD trial Table B19 Details of trials included in the mixed treatment comparison Table B20 Details of methodology for trials selected for the mixed treatment comparison Table B21 Quality assessment of studies included in the mixed treatment comparison Table B22 Summary of results from the trials included in the indirect treatment comparison Table B23 Identified chemotherapy reviews Table B24 Summary of the trials used to conduct the indirect comparison Table B25 Hazard ratios for PFS or TTP in studies selected for indirect treatment comparison Table B26 Hazard ratios for OS in studies selected for indirect treatment comparison Specification for manufacturer/sponsor submission of evidence Page 5 of 292

6 Table B27 Indirect treatment comparison for PFS and TTP Table B28 Indirect treatment comparison for OS Table B29 Indirect treatment comparison for PFS/TTP of fulvestrant (250 mg) versus exemestane Table B30 Grade 3/4 adverse events across randomised groups in BOLERO Table B31 Adverse events across randomised groups in the TAMRAD trial Table B32 Duration of exposure to study treatment Table B33 Comparison of incidence of grade 3/4 AEs during therapy with everolimus in combination with exemestane and with chemotherapy regimens in women with advanced breast cancer Table B34 Key features of analysis Table B35 Summary of variables applied in the economic model Table B36 Statistics of parameter-fitting to observed OS Kaplan Meier curves Table B37 Statistics of parameter-fitting to observed PFS Kaplan Meier curves Table B38 Preferred sources of data for each comparison Table B39 Hazard ratios versus everolimus in combination with exemestane Table B40 Model assumptions Table B41 Summary of details for studies reporting utility values for postmenopausal women with HR+ advanced or metastatic breast cancer receiving second-line therapy Table B42 Quality assessment of studies reporting utility values for postmenopausal women with HR+ advanced or metastatic breast cancer receiving second-line therapy Table B43 Summary details for other studies reporting utility values for postmenopausal women with HR+ advanced or metastatic breast cancer receiving treatment for advanced disease Table B44 Quality assessment other studies reporting utility values for postmenopausal women with HR+ advanced or metastatic breast cancer receiving treatment for advanced disease Table B45 Summary of utility values for health states used in the costeffectiveness analysis Table B46 Summary of utility values for adverse events used in the costeffectiveness analysis Table B47 Unit costs associated with the technology in the economic model Table B48 Additional costs associated with specific health states used in the economic model Table B49 List of adverse events and summary of costs that could be included in the economic model Table B50 Probabilistic sensitivity analysis Table B51 Everolimus in combination with exemestane vs exemestane alone Table B52 Model outputs by clinical outcomes Table B53 Summary of QALY gain by health state Specification for manufacturer/sponsor submission of evidence Page 6 of 292

7 Table B54 Summary of costs by health state Table B55 Summary of predicted resource use by category of cost Table B56 Base-case results Table B57 Probabilistic ICERs and probability of cost effectiveness: everolimus vs comparators Table B58 Results of scenario analysis Table C1 Key assumptions and flow for calculating advanced breast cancer population Table C2 Size of advanced breast cancer population over next 5 years 224 Table C3 Key assumptions and flow for calculating total eligible population for everolimus from advanced breast cancer population Table C4 Estimated population eligible for treatment with everolimus in combination with exemestane over the next 5 years Table C5 Market share assumptions for everolimus in combination with exemestane Table C6 Costs, without everolimus Table C7 Costs, first 3 years Table C8 Incremental costs, first 3 years Table C9 Inclusion and exclusion criteria for the review of clinical evidence and adverse effects Table C10 Quality assessment of BOLERO-2 18, Table C11 Quality assessment of TAMRAD 22, Table C12 Quality assessment of CONFIRM Table C13 Quality assessment of EFECT Table C14 Quality assessment of SoFEA Specification for manufacturer/sponsor submission of evidence Page 7 of 292

8 List of abbreviations AE adverse event ABC advanced breast cancer AI aromatase inhibitors BNF63 British National Formulary 63 rd edition BSAP bone-specific alkaline phosphatase CBR clinical benefit rate CDSR Cochrane Database of Systematic Reviews CG clinical guidelines CHF congestive heart failure CHMP Committee for Medicinal Products for Human Use CI confidence interval CMH Cochran Mantel Haenszel CoA ER-coactivator CR complete response CT computed tomographic CTX c-telopeptide of type 1 collagen CYP cytochrome P450 DARE Database of Abstracts of Reviews of Effects DCIS ductal carcinoma in situ ECOG Eastern Cooperative Oncology Group EGFR epidermal growth factor receptor EORTC QLQ-C30 European Organization for Research and Treatment of Cancer Quality of Life Questionnaire EQ-5D five-domain EuroQoL questionnaire ER oestrogen receptor ESMO European Society of Medical Oncology EU European Union FACT-B Functional Assessment of Cancer Therapy Breast FAS full analysis set FDA Food and Drug Administration GFR growth factor receptor Specification for manufacturer/sponsor submission of evidence Page 8 of 292

9 HaFS hand-foot syndrome HEED Health Economic Evaluations Database HER2 human epidermal growth factor receptor 2 HR hormone receptor HRQoL health-related quality of life HTA Health Technology Assessment ICER incremental cost-effectiveness ratio ICTRP International Clinical Trials Registry Platform IGF1R insulin-like growth factor 1 receptor ITT intention-to-treat IVRS interactive voice response system IWRS interactive web response system LCIS lobular carcinoma in situ LYG life-years gained MA megestrol acetate MAPK mitogen-activated protein kinase MEK mitogen-activated protein kinase kinase MID minimal important difference MRI magnetic resonance imaging mtor mammalian target of rapamycin N/A not applicable NCCN National Comprehensive Cancer Network NHS National Health Service NHS EED NHS Economic Evaluation Database NICE National Institute for Health and Clinical Excellence NICE CG81 NICE clinical guideline 81 Advanced breast cancer: diagnosis and treatment NR not reported NS not significant NSAI non-steroidal aromatase inhibitor ONS Office of National Statistics ORR overall response rate OS overall survival P1NP amino-terminal propeptide of type 1 collagen Specification for manufacturer/sponsor submission of evidence Page 9 of 292

10 PD PFS PgR PI3K PNET PR PRISMA PSA PSN PSS QALY QoL RCT RECIST SAE SD SMC STA TOI TOT TTO TTP VAS VEGF progressive disease progression-free survival progesterone receptor phosphatidylinositol 3-kinase primitive neuroectodermal tumours partial response preferred reporting items for systematic reviews and meta-analyses probabilistic sensitivity analysis peripheral sensory neuropathy personal social services quality-adjusted life-year quality of life randomised controlled trial Response Evaluation Criteria in Solid Tumours serious adverse event stable disease Scottish Medicines Consortium single technology appraisal Total Outcome Index time on treatment time trade-off time to progression visual analogue scale vascular endothelium growth factor Specification for manufacturer/sponsor submission of evidence Page 10 of 292

11 Instructions for manufacturers and sponsors This is the specification for submission of evidence to the National Institute for Health and Clinical Excellence (NICE) as part of the single technology appraisal (STA) process. It shows manufacturers and sponsors what information NICE requires and the format in which it should be presented. NICE acknowledges that for medical devices manufacturers particular sections might not be as relevant as they are for pharmaceuticals manufacturers. When possible the specification will refer to requirements for medical devices, but if it hasn t done so, manufacturers or sponsors of medical devices should respond to the best of their ability in the context of the question being addressed. Use of the specification and completion of appendices 1 to 13 (sections 10.1 to 10.13) are mandatory (when applicable), and the format should be followed whenever possible. Reasons for not following this format must be clearly stated. Sections that are not considered relevant should be marked N/A and a reason given for this response. The specification should be completed with reference to the NICE document Guide to the methods of technology appraisal ( particularly with regard to the reference case. Users should see NICE s Guide to the single technology appraisal (STA) process ( for further details on some of the procedural topics referred to only briefly here. If a submission is based on preliminary regulatory recommendations, the manufacturer or sponsor must advise NICE immediately of any variation between the preliminary and final approval. A submission should be as brief and informative as possible. It is expected that the main body of the submission will not usually exceed 100 pages excluding the pages covered by the template. The submission should be sent to NICE electronically in Word or a compatible format, and not as a PDF file. The submission must be a stand-alone document. Additional appendices may only be used for supplementary explanatory information that exceeds the level Specification for manufacturer/sponsor submission of evidence Page 11 of 292

12 of detail requested, but that is considered to be relevant to the submission. Appendices are not normally presented to the Appraisal Committee. Any additional appendices should be clearly referenced in the body of the submission and should not be used for core information that has been requested in the specification. For example, it is not acceptable to attach a key study as an appendix and to complete the clinical-effectiveness section with see appendix X. Clinical trial reports and protocols should not be submitted, but must be made available on request. Trials should be identified by the first author or trial ID, rather than by relying on numerical referencing alone (for example, Trial 123/Jones et al. 126 rather than One trial 126 ). For information on submitting cost-effectiveness analysis models, disclosure of information and equality and diversity, users should see Related procedures for evidence submission, section 11. If a patient access scheme is to be included in the submission, please refer to the patient access scheme submission template available on request. Please submit both documents and ensure consistency between them. Specification for manufacturer/sponsor submission of evidence Page 12 of 292

13 Executive summary Please provide an executive summary that summarises the key sections of the submission. All statements should be directly relevant to the decision problem, be evidence-based when possible and clearly reference the relevant section of the submission. The summary should cover the following items. Everolimus (Afinitor ) was granted marketing authorisation by the European Commission in July 2012 for treatment of hormone receptor positive, human epidermal growth factor receptor 2 negative (HER2/neunegative [HER2 ]) advanced breast cancer, in combination with exemestane, in postmenopausal women without symptomatic visceral disease after recurrence or progression following a non-steroidal aromatase inhibitor (NSAI). This submission considers everolimus in combination with exemestane as treatment for advanced, hormone receptor positive (HR+), HER2 breast cancer in accordance with this licence and its use in the pivotal clinical breast cancer trials of oral everolimus-2 (BOLERO-2). Everolimus is a protein kinase inhibitor that inhibits the mammalian target of rapamycin (mtor), a key component of the phosphatidylinositol 3- kinase/akt/mtor (PI3K/Akt/mTOR) signalling pathway that controls cell survival, growth and proliferation. 1-3 This pathway is believed to be important in allowing tumour cells expressing hormone receptors to escape from hormone dependence and hence develop resistance to endocrine therapy, the cornerstone of management of advanced or metastatic disease in postmenopausal women with HR+ tumours. Patients developing resistance to endocrine therapy fail to benefit from further endocrine therapy and have limited effective treatment options. Indeed, current UK and international guidelines on management of breast cancer each state that on failure of a first-line aromatase inhibitor (AI), the optimal treatment for on-going patient management is unclear, 4-7 and recent NICE guidance (NICE clinical guideline 81 Advanced breast cancer: diagnosis and treatment ; NICE CG81) stated that there is a need to establish Specification for manufacturer/sponsor submission of evidence Page 13 of 292

14 effective endocrine therapy for postmenopausal women with HR+ tumours who progress on treatment with an AI. 8 This is particularly relevant as approximately 50% of women with HR+ breast cancer present with de novo resistance to AIs and hence need an effective alternative therapy, while nearly all initial responders to endocrine therapy develop resistance at some point, resulting in disease progression Furthermore, around 50 to 60% of women in whom first-line endocrine therapy fails will not respond adequately to subsequent lines of endocrine therapy, leaving limited treatment options. 14 In addition, with increasing use of AIs in the adjuvant setting, more patients who progress to metastatic disease have already been exposed to AIs and may therefore have already developed resistance to these agents. 7 The significant improvements in PFS demonstrated for everolimus suggest that this agent may alter the paradigm for management of HR+ advanced/metastatic breast cancer in postmenopausal women. Everolimus has been shown to restore the sensitivity of breast cancer cells to endocrine therapy. 15 Furthermore, everolimus has been demonstrated to prevent the emergence of hormone-independent tumour cells, suggesting that early intervention with combined endocrine therapy and mtor inhibition may prevent or delay endocrine resistance. 16 Thus everolimus targets directly the pathway that is believed to be central to the development of resistance to endocrine therapy in patients with HR+ metastatic breast cancer. By restoring sensitivity to endocrine therapy, everolimus may be expected to extend the benefits of endocrine therapy in women with HR+ tumours and address the unmet need for effective therapies for postmenopausal women with HR+ tumours failing on endocrine therapy, thereby delaying the need to progress to chemotherapy. It should be noted that this patient population represents a very small subset of all breast cancer patients with an estimated 1,548 patients being eligible for everolimus treatment in this setting. Specification for manufacturer/sponsor submission of evidence Page 14 of 292

15 Clinical effectiveness and safety of everolimus for treatment of postmenopausal women with HR+, HER2 advanced breast cancer The clinical effectiveness of everolimus, in combination with exemestane, for treatment of postmenopausal women with HR+, HER2 metastatic breast cancer that have progressed or recurred following treatment with NSAIs has been demonstrated in a large, multicentre, double-blind, placebo-controlled, randomised study, BOLERO-2 (see table) BOLERO-2 was conducted in 724 postmenopausal women with HR+, HER2 advanced breast cancer that had recurred or progressed following prior treatment with the NSAIs letrozole or anastrozole, and compared the combination treatment of everolimus 10 mg/day in combination with exemestane 25 mg/day with that of placebo in combination with exemestane 25 mg/day. The primary endpoint was progression-free survival (PFS) based on local and central radiological assessment. Secondary endpoints included overall survival (OS), clinical benefit rate (CBR) (defined as the proportion of patients with a complete response [CR], partial response [PR] or stable disease [SD]), objective response rate (ORR), quality of life outcomes evaluated using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30), safety and bone turnover markers. Data are available for three analyses at median follow-up times of 7 months, months 17 and 18 months. 19 BOLERO-2 demonstrated that everolimus in combination with exemestane substantially reduces the risk of disease progression, more than doubling median PFS compared with exemestane alone at all analysis time points The combination increased PFS from approximately 3 or 4 months for exemestane alone to between 7 and 11 months, corresponding to a reduction in the risk of disease progression of 62% to 64% (central assessment) and 55% to 57% (local assessment).. Results were consistent and statistically significant (p < 0.001) across all analyses (local and central) and time points. A statistically significant benefit for the combination was observed for all subgroups analysed. The combination also resulted in improved OS; at 16 months, 112 patients (23%) in the Specification for manufacturer/sponsor submission of evidence Page 15 of 292

16 everolimus in combination with exemestane group versus 70 patients (29%) in the exemestane alone group had died (p = 0.046). At each assessment the CBR achieved with everolimus in combination with exemestane was approximately double that achieved with placebo in combination with exemestane and at 18 months clinical benefit was achieved in 51.3% of patients receiving everolimus in combination with exemestane versus 26.4% of patients receiving exemestane alone (p < ). Significant differences were also seen between treatment arms in terms of ORR (12.6% for everolimus in combination with exemestane versus 1.7% for exemestane alone at 18 months, p < ). An assessment of markers of bone resorption suggested that everolimus was able to reverse the increase in bone resorption and suppress bone turnover associated with exemestane. The combination also reduced the rate of fractures at 18 months (2.3%) compared with placebo in combination with exemestane (3.8%), and reduced the rate of disease progression in bone both in the overall population and in patients with bone metastases at baseline. 20 Improvements in PFS, OS, CBR and bone health with everolimus in combination with exemestane were achieved without compromising health-related quality of life (HRQoL). 17,18,21 Further supporting evidence for the efficacy of everolimus in combination with endocrine therapy is provided by the tamoxifen plus everolimus (TAMRAD) phase II study, which compared everolimus in combination with tamoxifen with tamoxifen monotherapy in postmenopausal women with HR+, HER2, AI-resistant metastatic breast cancer. 22 Patients were randomised to receive everolimus 10 mg/day in combination with tamoxifen 20 mg/day, versus tamoxifen 20 mg/day alone. The primary endpoint was CBR at 6 months. Secondary outcomes included time to progression (TTP), OS, ORR and toxicity. Results have been reported for a median follow-up of 24 months. 22 Everolimus in combination with tamoxifen provided a significant increase in CBR at 6 months compared with tamoxifen monotherapy (61% versus 42%, p = 0.045), and this benefit was observed in all subgroups analysed. Specification for manufacturer/sponsor submission of evidence Page 16 of 292

17 The combination also significantly increased TTP compared with tamoxifen monotherapy (8.6 months versus 4.5 months) corresponding to a 46% reduction in risk of progression (p = ), and was associated with a 55% reduction in the risk of death compared with tamoxifen alone (p = 0.007). The improvements in TTP and OS with everolimus combination therapy were particularly marked in patients with secondary hormone resistance. 23 The median PFS and TTP reported for everolimus in combination with exemestane in BOLERO-2 and for everolimus in combination with tamoxifen in the TAMRAD study, respectively, compare favourably with those reported in previous studies for single-agent endocrine therapy (tamoxifen, fulvestrant or exemestane) or chemotherapy (doxorubicin, docetaxel or capecitabine) in patients in whom NSAIs are failing In addition, everolimus in combination with tamoxifen significantly improved OS over tamoxifen alone while most studies of endocrine therapies in the second-line setting have failed to demonstrate a significant OS benefit These improvements in PFS/TTP and OS can be expected to benefit patients, especially given the favourable safety profile of everolimus combination therapy. In both studies everolimus was generally well tolerated. 18,22 Most adverse events (AEs) were mild or moderate in severity. The most frequently reported grade 3/4 AE in both studies was stomatitis, a known class effect of mtor inhibitors, which in most cases was managed by dose reductions or treatment interruptions. The only other grade 3/4 AEs reported in more than 5% of patients receiving everolimus were anaemia (in BOLERO-2) and fatigue, anorexia, infections and pain (in the TAMRAD trial). Noninfectious pneumonitis and increased risk of infection are other recognised class effects of mtor inhibitors. Grade 3 or 4 non-infectious pneumonitis was observed in 2% (TAMRAD) to 3% (BOLERO-2) of patients and the incidence of grade 3 or 4 infections was (CIC information removed) in BOLERO-2 (versus (CIC information removed) in the exemestane alone group) and 7% (versus 5% in the control group) in the TAMRAD trial. Few Specification for manufacturer/sponsor submission of evidence Page 17 of 292

18 patients discontinued therapy due to AEs in BOLERO-2: 9.1% (combination) versus 3.3% (exemestane alone) at 18 months. Thus everolimus provides a valuable therapeutic option in patients for whom AI therapy fails. In summary everolimus in combination with exemestane represents a step-change innovation in the management of advanced/metastatic HR+ breast cancer in postmenopausal women. This is an area of hitherto unmet clinical need and has been described by clinicians as a quantum shift in the future of endocrine therapy in this patient population. Summary of the BOLERO-2 18-month efficacy data Outcome Median PFS, months Central assessment Local assessment OS (deaths), % (at 16 months) Everolimus in combination with exemestane Placebo in combination with exemestane p-value < < ORR, % < CBR, % < Piccart et al., ; Data on file. 35 CBR, clinical benefit rate; PFS, progression-free survival; ORR, overall response rate; OS, overall survival. Economic evaluation The licensed indication of everolimus is the treatment of HR+, HER2 advanced breast cancer, in combination with exemestane, in postmenopausal women without symptomatic visceral disease after recurrence or progression following a NSAI. The pivotal study, BOLERO- 2, 18 focused on this patient group, and forms the basis for the economic evaluation. Specification for manufacturer/sponsor submission of evidence Page 18 of 292

19 Everolimus, which is administered orally, has a recommended dosage of 10 mg once daily, and is available as follows (see also Table A4): 10 mg x 30 tablets: 2970 A Markov model was used to represent the progressive nature of the disease; the model structure has been used in previous NICE submissions (for example, eribulin [TA 250] 36 and fulvestrant [TA 239] 37 ), and reflects the main outputs of the pivotal BOLERO-2 study. The model comprises three mutually exclusive health states: stable disease, progressive disease and death. Progression following endocrine therapy is represented by patients moving from the stable disease state to the progressive disease state; patients can die from any cause, and at any time. To allow for changes in the probability of progression over time, a partitioned survival model was used. As described in the 2009 NICE CG81, 8 there is no clear pathway of care following failure of NSAI therapy thus highlighting the unmet clinical need for these patients. Options include hormonal therapy (tamoxifen, exemestane or fulvestrant), chemotherapy and chemotherapy followed by maintenance hormonal therapy; the NSAI therapies anastrozole and letrozole are not generally used after NSAI failure (see section 2.7). The comparators for the economic analysis are therefore tamoxifen, exemestane, fulvestrant and the chemotherapeutic agents doxorubicin, capecitabine and docetaxel, which reflects current UK clinical practice. Clinical parameters for everolimus in combination with exemestane were taken from BOLERO-2 (see section 7.3.7). The base-case analysis used parametric survival curves based directly on PFS data from the BOLERO month follow-up, and on 16-month OS data. For the comparators, hazard ratios for PFS and OS were derived from the BOLERO-2 and TAMRAD 22 studies in the case of exemestane and tamoxifen, from the SoFEA, 26 EFECT 24 and CONFIRM 25 trials for fulvestrant (see section 6.7 Specification for manufacturer/sponsor submission of evidence Page 19 of 292

20 for details of the indirect comparison of endocrine therapies), and from a systematic review of chemotherapy versus endocrine therapy for doxorubicin, capecitabine and docetaxel. 38 The comparison with exemestane was used as the base case for the model. The age-adjusted utility scores for the Lloyd et al. study were used in the base-case analysis. 39 To be consistent with the 5-dimension EuroQol questionnaire UK tariff, the average age of respondents to the original study (ie 47 years) was used to calculate utility values as per the Liverpool Reviews and Implementation Group recommendation in previous appraisals. 37,40 The weighted age-adjusted utility value for stable disease has been reported as (fulvestrant ERG report 37 ) and that of treatment response as (eribulin ERG report 40 ). Thus the average utility value for stable disease and treatment response was calculated to be (based on the CBR of the everolimus arm from BOLERO-2 of 51.3%). For patients in the progressed state, a health state age-adjusted utility value of was used. As there is limited published information on resource use in metastatic breast cancer, resource use data for the base-case analysis were based on NICE CG81, 8 using Package 1 for the stable disease state and Package 2 for the progressive disease state, with unit costs taken from the Personal Social Services Research Unit report Unit costs of health and social care The results of the cost-effectiveness analysis are shown in the following table. Everolimus in combination with exemestane has an incremental cost-effectiveness ratio of 32,417 compared with exemestane and 29,109 compared with tamoxifen and 27,147 compared with fulvestrant. For all the other comparators, the ICER is less than 25K/QALY gained. Specification for manufacturer/sponsor submission of evidence Page 20 of 292

21 Base-case cost-effectiveness results Everolimus in combination with Incremental Incremental exemestane compared with: costs QALYs ICER Exemestane 27, ,417 Tamoxifen a 34, ,109 Doxorubicin 25, ,253 Fulvestrant 20, ,147 Capecitabine 29, ,362 Docetaxel 13, ,000 a Based on TAMRAD trial. ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life-year. Other considerations Everolimus is an oral treatment that can be conveniently taken by patients at home. The benefits of oral therapy over chemotherapy (docetaxel and doxorubicin) and fulvestrant are not fully captured by the costeffectiveness model, because the same utility values for stable disease are used for all comparators. Societal benefits are not accounted for in the reference case but as everolimus is likely to delay chemotherapy treatment and its detrimental impact on productivity we have provided a sensitivity analysis to explore the impact of productivity losses. Everolimus also slows down the rate of disease progression in bone, reducing the risk of bone metastases and fractures, both of which adversely affect HRQoL (see section 6.5.3) and can result in costly consequences. The beneficial effects of everolimus on bone are likely to be even more important in the longer term both in terms of positive impact on quality of life and reducing the costs of treating bone complications such as fractures. Budget impact The licensed population (see section 8.1) represents a very small subset of all breast cancer patients estimated to be around 1548 patients in Engalnd and Wales. Therefore the budget impact is estimated to be relatively low, with additional drug costs offset to some extent by reduced Specification for manufacturer/sponsor submission of evidence Page 21 of 292

22 costs due to AEs. The estimated net impact of everolimus over the first 5 years of use is a budget increase for England and Wales of 9.04 million. Conclusion Everolimus directly addresses the hitherto unmet clinical need for effective treatment options for the relatively small patient population of postmenopausal women with advanced, HR+, HER2 tumours who progress on treatment with an AI, as highlighted in NICE CG81. 8 In the pivotal BOLERO-2 study, everolimus in combination with exemestane increased OS, and more than doubled median PFS and CBR compared with exemestane alone. Everolimus represents a step change in the management of breast cancer for a patient group with limited effective treatment options, and has been described as a quantum shift in the future of endocrine therapy for the licensed patient population. 42 Specification for manufacturer/sponsor submission of evidence Page 22 of 292

23 Section A Decision problem Manufacturers and sponsors will be requested to submit section A in advance of the full submission (for details on timelines, see the NICE document Guide to the single technology appraisal (STA) process A (draft) summary of product characteristics (SPC) for pharmaceuticals or information for use (IFU) for devices, a (draft) assessment report produced by the regulatory authorities (for example, the European Public Assessment Report (EPAR)), and a (draft) technical manual for devices should be provided (see section 9.1, appendix 1). 1 Description of technology under assessment 1.1 Give the brand name, approved name and, when appropriate, therapeutic class. For devices, provide details of any different versions of the same device. Everolimus (Afinitor ) is a protein kinase inhibitor, a member of the pharmacotherapeutic class: antineoplastic agents, protein inhibitor; ATC code: L01XE What is the principal mechanism of action of the technology? Everolimus is a protein kinase inhibitor that inhibits the mammalian target of rapamycin (mtor), a key component of the phosphatidylinositol 3- kinase/akt/mtor ( PI3K/Akt/mTOR) signalling pathway that controls cell 1-3 survival, growth and proliferation. This mechanism of action is thought to be important in allowing everolimus to overcome resistance to endocrine therapy in patients with advanced or metastatic hormone receptor positive (HR+) breast cancer. 2,3,16 Endocrine therapy, also known as anti-oestrogen therapy, is the cornerstone of management of advanced or metastatic breast cancer in postmenopausal women with tumours expressing hormone receptors (ie HR+ tumours) (see section 2.5). These tumours are initially dependent on Specification for manufacturer/sponsor submission of evidence Page 23 of 292

24 oestrogen, which inhibits apoptosis and promotes cell proliferation, angiogenesis and metastasis through an HR-mediated genomic response. 2 This is mediated by the nuclear HR, which then binds to specific DNA promoter regions. HRs near the cytoplasmic membrane may also promote cell proliferation through interactions with growth factor signalling pathways such as the PI3K/Akt/mTOR pathway. 2 A therapeutic strategy of oestrogen deprivation typically effected through use of inhibitors of oestrogen production (such as non-steroidal aromatase inhibitors [NSAIs]) or agents that antagonise HR function (such as tamoxifen), can be successful in inhibiting tumour growth. However, over time, most of these HR+ tumours adapt to hormone deprivation and become resistant to endocrine therapy. 2,9 This acquired resistance, or escape from hormone dependence thus renders the tumour unresponsive to currently available therapeutic strategies for management of advanced or metastatic disease. In breast cancer, tumour resistance is associated with activation of the PI3K/Akt/mTOR pathway which provides an alternative tumourigenic signalling pathway to overcome the inhibited, HRmediated pathways that would otherwise drive growth and proliferation. 2,3,44 Everolimus inhibits the activity of the PI3K/Akt/mTOR pathway, as demonstrated in preclinical studies with breast cancer cells. Boulay et al. showed that everolimus inhibits the proliferation of aromatase-expressing breast cancer cells, and when combined with the NSAI letrozole, inhibitory effects were found to be synergistic. 1 In a second study, everolimus was found to restore the sensitivity of breast cancer cells to endocrine therapy, 15 Furthermore, inhibition of the PI3K/Akt/mTOR pathway can prevent the emergence of hormone-independent tumour cells, suggesting that early intervention with combined endocrine therapy and mtor inhibition may prevent or delay endocrine resistance. 16 Thus everolimus targets directly the pathway that is believed to be central to the development of resistance to endocrine therapy in patients with HR+ metastatic breast cancer. Specification for manufacturer/sponsor submission of evidence Page 24 of 292

25 1.3 Does the technology have a UK marketing authorisation/ce marking for the indications detailed in this submission? If so, give the date on which authorisation was received. If not, state current UK regulatory status, with relevant dates (for example, date of application and/or expected approval dates). A European regulatory submission was made on 9 November 2011 to the Committee for Medicinal Products for Human Use (CHMP; European Medicines Agency). A positive opinion was received on 21 st June Marketing authorisation in the European Union (EU) was received on 22 th July Describe the main issues discussed by the regulatory organisation (preferably by referring to the [draft] assessment report [for example, the EPAR]). If appropriate, state any special conditions attached to the marketing authorisation (for example, exceptional circumstances/conditions to the licence). No special conditions are attached to the marketing authorisation for everolimus in this indication. 45 Of note, the EU licence for everolimus in this indication excludes patients with symptomatic visceral disease whereas the US Food and Drug Administration (FDA) licence does not exclude such patients. 43, What are the (anticipated) indication(s) in the UK? For devices, provide the (anticipated) CE marking, including the indication for use. Everolimus is indicated for the treatment of hormone receptor-positive, HER2/neu- negative advanced breast cancer, in combination with exemestane, in postmenopausal women without symptomatic visceral disease after recurrence or progression following a non-steroidal aromatase inhibitor. 43 Specification for manufacturer/sponsor submission of evidence Page 25 of 292

26 1.6 Please provide details of all completed and ongoing studies from which additional evidence is likely to be available in the next 12 months for the indication being appraised. The following studies have investigated or are investigating everolimus for the treatment of postmenopausal women with HR+, HER2, metastatic breast cancer that has progressed or recurred following treatment with NSAIs, and may provide additional evidence in the next 12 to 24 months. Two randomised studies have reported their primary outcomes and are described in greater detail in section B of this submission. A further, openlabel, single-arm study of everolimus in combination with exemestane is ongoing. In brief, the phase III breast cancer trials of oral everolimus (BOLERO-2) study (NCT ; evaluates the combination treatment of everolimus 10 mg/day in combination with exemestane 25 mg/day versus placebo in combination with exemestane 25 mg/day in postmenopausal women with HR+, HER2, advanced breast cancer who have progressed after treatment with NSAIs. The primary endpoint of the study is progression-free survival (PFS). The results of three analyses have been reported and are described in section No further analyses of PFS data are planned, but results of a final analysis after 392 survival events are anticipated in late The tamoxifen plus everolimus (TAMRAD) study (NCT ; provides additional data for everolimus in combination with endocrine therapy. This phase II study compares treatment with everolimus 10 mg/day in combination with tamoxifen 20 mg/day versus tamoxifen 20 mg/day alone in postmenopausal women with HR+, HER2, AI-resistant metastatic breast cancer. Results for a median follow-up of 24 months have been reported and are described in section The TAMRAD study is still ongoing and further updates will be available in next 12 months. Specification for manufacturer/sponsor submission of evidence Page 26 of 292

27 The 4EVER study (NCT ; NCT ?term=4+EVER&rank=2) is a multi-centre, open-label, single-arm study which aims to evaluate the efficacy and safety, quality of life and health resources utilization of the combination of everolimus and exemestane in a broader patient population compared with BOLERO-2, i.e. without limitations as to the number of previous chemotherapy lines, the time point of progression after NSAI therapy, and previous endocrine therapy. The estimated primary completion date is in May The 4EVERUK trial (CRAD001YGB11, EUDRACT ) is a phase 4 multi-centre, open-label, single-arm study of postmenopausal women with oestrogen receptor-positive locally advanced or metastatic breast cancer treated with everolimus (RAD001) in combination with exemestane, and will evaluate efficacy, safety, and quality of life measures, with exploratory epigenetic marker analysis. Planned recruitment is 50 patients and the trial will be conducted in 10 UK centres. Recruitment will commence in December 201 for a period of 1 year. 1.7 If the technology has not been launched, please supply the anticipated date of availability in the UK. The UK launch for everolimus in the treatment of postmenopausal women with HR+, HER2, advanced breast cancer that has progressed or recurred following treatment with NSAIs occurred on 18 th of September Does the technology have regulatory approval outside the UK? If so, please provide details. Everolimus was approved by the US FDA for this indication on 20 th July Marketing authorisation in the EU was received on 22 th July As of July 2012, everolimus has also been approved in Argentina, Bangladesh, Paraguay, Philippines, Russia and Columbia, and has received positive opinions in other countries (El Salvador, Guatemala) for this indication. Specification for manufacturer/sponsor submission of evidence Page 27 of 292

28 1.9 Is the technology subject to any other form of health technology assessment in the UK? If so, what is the timescale for completion? Novartis intends to make a submission to the Scottish Medicines Consortium (SMC) for everolimus in the indication described in this submission in the first half of For pharmaceuticals, please complete the table below. If the unit cost of the pharmaceutical is not yet known, provide details of the anticipated unit cost, including the range of possible unit costs. Table A1 Unit costs of technology being appraised Pharmaceutical formulation Acquisition cost (excluding VAT) 10 mg: 2970 per month (30 tablets Method of administration Doses Dosing frequency Average length of a course of treatment Average cost of a course of treatment Anticipated average interval between courses of treatments Anticipated number of repeat courses of treatments Dose adjustments Oral tablet 10 mg, 5 mg; the recommended dose of everolimus is 10 mg once daily Once daily Treatment should continue until disease progression or intolerable adverse events are experienced 16,335 based on an average duration of therapy of 5.5 months at a dose of 10 mg from BOLERO 2 trial Patients receive a single course of continuous treatment (based on BOLERO-2) Not applicable: patients receive continuous treatment Management of adverse reactions may require dose reduction (suggested dose is 5 mg) or dosing to be temporarily withheld 1.11 For devices, please provide the list price and average selling price. If the unit cost of the device is not yet known, provide details of the anticipated unit cost, including the range of possible unit costs. Not applicable Are there additional tests or investigations needed for selection, or particular administration requirements for this technology? Specification for manufacturer/sponsor submission of evidence Page 28 of 292

29 It is anticipated that no additional tests or investigations will be needed for selection of patients. Therapy is administered orally Is there a need for monitoring of patients over and above usual clinical practice for this technology? There is no need for monitoring over and above usual clinical practice for this technology What other therapies, if any, are likely to be administered at the same time as the intervention as part of a course of treatment? Exemestane is administered with everolimus as part of treatment in the indication described. Specification for manufacturer/sponsor submission of evidence Page 29 of 292

30 2 Context In this background section the manufacturer or sponsor should contextualise the evidence relating to the decision problem. 2.1 Please provide a brief overview of the disease or condition for which the technology is being used. Include details of the underlying course of the disease. Pathology Breast cancer, a malignant cellular growth in the tissues of the breast, is a heterogeneous disease, diverse in its natural history and pathology, and in its responsiveness to treatments. 47 There are several histological and molecular subtypes of breast cancer, with the main molecular subtypes being classified based on gene expression profiles. 47,48 Breast cancers may be HR positive (HR+) or negative (HR ) tumours, distinguished by whether or not they express oestrogen or progesterone receptors, or both, on their cell surface, and HER2 positive (HER2+) or negative (HER2 ), according to their level of expression of HER2 (also known as c-erbb2 or neu). The hormone receptors mediate the physiological effects of oestrogen and progesterone, which include the regulation of cell growth, differentiation and homeostasis in eukaryotic cells, and are also associated pathologically with an increased risk for breast and endometrial cancers. 49 It is estimated that approximately 75% to 84% of breast cancer tumours are HR+. 50,69 HER2 is also involved in the regulation of cell growth. Amplification of the HER2 gene or over-expression of its product, HER2, is linked with cancer development, 51 and is seen in 20 to 30% of breast cancers. 52,53 Approximately 55% of breast cancers are HR+ and HER2, and around 50% occur in postmenopausal women. 54 Prognosis A number of clinical and pathological factors impact on the clinical behaviour of tumours and may be important for predicting disease prognosis and for the selection of therapy. 48,55 These include age and Specification for manufacturer/sponsor submission of evidence Page 30 of 292

31 menopausal status of the patient, stage of disease, histologic and nuclear grade of the primary tumour, HR status of the tumour, the presence or absence of overexpression of HER2 and proliferative capacity. The stage of disease (see below for details) is particularly influential, with higherstage cancers being associated with a worse prognosis. HR status is important for choice of therapy with HR+ tumours being more aggressive, but also being more likely to respond to endocrine therapy. Similarly, HER2 status of the tumour influences likely outcome and responsiveness to treatment, with HER2+ tumours potentially being responsive to anti- HER2 therapies, and HER2 tumours associated with a less favourable prognosis. 7,51 Stages of disease Stage of disease is defined according to the size of the primary tumour, the extent of lymph node involvement and the presence of metastases. 56 In terms of treatment, breast cancer can be broadly classed into four groups, namely: 1) the pure non-invasive carcinomas, or stage 0 disease cancers, which include ductal carcinoma in situ (DCIS) and lobular carcinoma in situ (LCIS); 2) operable, local-regional invasive carcinomas with or without associated non-invasive carcinoma (disease stages I, II and most stage III tumours); 3) inoperable local-regional invasive carcinoma with or without associated non-invasive carcinoma (some stage III tumours); 4) metastatic (stage IV) or recurrent carcinoma The stage of disease strongly influences treatment options and prognosis, with treatments for early stage disease being curative in intent, whereas those for metastatic or recurrent disease are generally palliative. In this submission, advanced breast cancer is considered to be represented by stage IV disease. This is in accordance with NICE clinical guideline 81 Advanced breast cancer: diagnosis and treatment (NICE CG81), 59 which defines advanced breast cancer as invasive adenocarcinoma of the breast of clinical stage IV. In the BOLERO-2 study, % of patients had metastatic (stage IV) disease, and the clinical evidence presented in Specification for manufacturer/sponsor submission of evidence Page 31 of 292

32 this submission is therefore relevant to a stage IV population; this is also similar to the population described in the EU licence for everolimus. The EU licence excludes patients with symptomatic visceral disease, reflecting the BOLERO-2 study population. There are no national data on the incidence of advanced breast cancer; however regional data from the West Midlands Cancer Intelligence Unit indicates that approximately 5% of patients diagnosed with breast cancer between 1992 and 1994 had metastases at the time of their primary diagnosis (stage IV) and a further 35% of all those with a primary diagnosis of early disease went on to develop advanced breast cancer in the 10 years following diagnosis. 61 Survival Prognosis and survival among patients with breast cancer is related to stage of disease. Those with advanced disease tend to have a very poor prognosis, with patients typically surviving between 1 and 3 years. Survival statistics according to stage are available for one area of England, for women diagnosed in the early 1990s, and are summarised in Table A2. 62 The data show that while 90% of women with stage I cancers lived more than 5 years (Table A2), for patients with metastatic disease (ie stage IV disease), 5-year survival rates were 13%, 62 close to similar reports in the literature of 5-year survival for stage IV disease of 18%. 63 Table A2 Overall survival according to stage of disease in women with breast cancer 62 Stage of disease 5-year overall survival 10-year overall survival 1 > 90% > 85% 2 > 70% > 60% 3 50% 40% 4 13% 10% Specification for manufacturer/sponsor submission of evidence Page 32 of 292

33 Symptoms and health-related quality of life (HRQoL) Maintenance or improvement in quality of life, together with a reduction in the burden posed by symptoms associated with the cancer and its treatment, are important treatment goals in advanced breast cancer. 6 The symptoms of metastatic breast cancer typically result from the spread and growth of tumour cells within distant tissues and all are likely to significantly impact on patient health-related quality of life (HRQoL). 64 For example, more that 50% of patients who have stage IV breast cancer at diagnosis have bone metastases, which are recognised as an important cause of morbidity and mortality and can lead to irreversible skeletalrelated events such as fracture or spinal cord compression In addition to the physical pain caused by these events, patients can experience a reduction in functional independence and ability to participate in normal daily activities, leading to an overall decrease in their quality of life. 66 Lung metastases can lead to shortness of breath and liver metastasis is often linked with a loss appetite or weight loss. Neurological pain or weakness, headaches, confusion or seizures, may be associated with brain metastasis, and are further debilitating symptoms associated with advanced disease Please provide the number of patients covered by this particular therapeutic indication in the marketing authorisation and also including all therapeutic indications for the technology, or for which the technology is otherwise indicated, in England and Wales and provide the source of the data. Advanced breast cancer This submission concerns the use of everolimus in postmenopausal women with HR+, HER2 metastatic breast cancer without symptomatic visceral disease whose disease has recurred or progressed following treatment with NSAIs. An estimation of the number of eligible patients is described in detail in section 8.1 and is summarized in Table A3 and Table Specification for manufacturer/sponsor submission of evidence Page 33 of 292

34 A4. It is estimated that there are 1548 eligible patients in England and Wales. Table A3 Key Assumptions and flow for calculating Advanced Breast Cancer Population (ABC): Description Proportio n Populatio n Source Female population 15 with invasive breast cancer Women with early and locally advanced invasive breast cancer Women with advanced invasive breast cancer Women presenting with early breast cancer that die before disease progresses Women with early and locally advanced breast cancer progressing to advanced stage Total number with Advanced Breast Cancer 0.18% 42,658 NICE, costing templates and 95.00% 40,525 clinical guidelines (CG) 80 and 5.00% ,68, % 12, % 10,451 12,584 Table A4 Key Assumptions and flow for calculating total eligible population for everolimus from Advanced Breast Cancer population Description Proportion 2012 Population Post menopausal Women (aged 55yrs) 68 69% 8831 Women with hormone receptor positive breast 84% 7418 cancer 70 (83.8%) Women with HER2 negative breast cancer 68 75% 5563 Women with Asymptomatic Visceral metastases 75% 4172 (without Visceral Crisis) 71 (74.6%) Women with hormone receptor-positive ABC for 70% 2921 whom (hormonal) therapy is appropriate 59 Women in whom disease progresses or relapses while on, or after receiving a NSAI 72 53% (52.8%) 1548 Specification for manufacturer/sponsor submission of evidence Page 34 of 292

35 Other indications Everolimus is also indicated for the treatment of unresectable or metastatic, well- or moderately differentiated neuroendocrine tumours of pancreatic origin (PNET) in adults with progressive disease and for treatment of patients with advanced renal cell carcinoma whose disease has progressed on or after treatment with VEGF-targeted therapy. 43 In England and Wales, it is estimated that 1542 patients with renal cell carcinoma and 46 patients with PNET would be eligible to receive everolimus. 2.3 Please provide information about the life expectancy of people with the disease in England and Wales and provide the source of the data. Five-year overall survival (OS) results for patients with advanced breast cancer (stage IV disease) are approximately 13%, 62,63 and data from the West Midlands Cancer Intelligence Unit indicate that median OS for these patients is approximately 12 months Please give details of any relevant NICE guidance or protocols for the condition for which the technology is being used. Specify whether any specific subgroups were addressed. There are four NICE guidance documents relevant to everolimus and the indication described in this submission. NICE clinical guideline 81 Advanced breast cancer: diagnosis and treatment (NICE CG81) was published in 2009 and describes the current therapeutic management of postmenopausal women with HR+ tumours. 8 This guidance states that there is good evidence to support the use of AIs for treatment of postmenopausal women with HR+ tumours, and recommends offering an AI to postmenopausal women with HR+ breast cancer who have received no prior endocrine therapy or have previously received tamoxifen. The guidance however also states that there is little evidence for the best sequence of alternative endocrine therapies on disease progression and thus clinical trials are needed to investigate the most effective endocrine Specification for manufacturer/sponsor submission of evidence Page 35 of 292

36 therapy, or sequence of cancer therapies, to employ in patients who progress whilst on treatment with an AI. Three further NICE guidance documents relating to the management of metastatic breast cancer have been issued since publication of NICE CG81. These are TA 239 (fulvestrant), 37 TA 214 (bevacizumab) 73 and TA 250 (eribulin). 36 The use of these three drugs is not recommended for the treatment of metastatic breast cancer. 2.5 Please present the clinical pathway of care that depicts the context of the proposed use of the technology. Explain how the new technology may change the existing pathway. If a relevant NICE clinical guideline has been published, the response to this question should be consistent with the guideline and any differences should be explained. Current clinical pathway of care For the majority of advanced or metastatic tumours, including metastatic breast cancer, the treatment goals are palliative, centred on prolonging PFS and providing symptomatic relief, comfort and the maintenance or improvement of patient quality of life. 4,6 According to the most recent UK guidelines, 7 together with guidance published by the European Society of Medical Oncology (ESMO) 6 and the National Comprehensive Cancer Network (NCCN) in the USA 4 and recent international consensus guidelines, 5 anti-oestrogen therapies (endocrine therapies) such as NSAIs or tamoxifen, are the option of choice for most postmenopausal women with HR+ advanced breast cancers who do not have immediate lifethreatening visceral disease, as they are generally well tolerated. Such patients have inoperable disease and most have stage IV disease, though some with stage III disease may also be inoperable. As cytotoxic agents are associated with substantial toxicity, chemotherapy is not considered except for patients with immediately life-threatening visceral disease who require a rapid tumour response or on failure of endocrine therapy. 4,6,7 Specification for manufacturer/sponsor submission of evidence Page 36 of 292

37 Endocrine therapies such as anti-oestrogens that antagonise HR function (eg tamoxifen) or which inhibit oestrogen production (eg the AI drug class) have been extensively developed for the treatment of HR+ metastatic breast cancer and together have been the main form of treatment for advanced breast cancer for several decades. 74 Today, third-generation AIs including the NSAIs such as letrozole (Femara ) and anastrozole (Arimidex ) and the steroidal AI exemestane are the accepted standard of care for adjuvant therapy in postmenopausal women and for first-line treatment of metastatic disease in postmenopausal women. 4-7 There is also the option to use tamoxifen in advanced breast cancer, and tamoxifen together with ovarian suppression (using goserelin [Zoladex ]) is a recommended first-line treatment for metastatic disease in premenopausal women, but its efficacy has been shown to be inferior to AIs in postmenopausal women. 4-7 Another HR antagonist, fulvestrant (Faslodex ), although approved for the treatment of HR+ metastatic breast cancer in postmenopausal women with disease progression following anti-oestrogen therapy, has not been recommended by NICE for use in UK practice. 37 The population that fulvestrant is indicated for is different than that of everolimus in combination with exemestane, as fulvestrant is only licensed for use in patients with relapse on or after antioestrogen therapy and not for patients with relapse after NSAI therapy. 75 Each of these guidelines states that on failure of first-line AI, the optimal treatment is unclear. The NCCN for example recommends three lines of consecutive endocrine therapy, although evidence suggests that patients derive diminishing benefits with each additional line. Options for patients include the use of another class of AI (eg a steroidal AI, if a NSAI was used previously), but reported objective response rates (ORRs) are low. While there is some evidence from the tamoxifen and arimidex randomised group efficacy and tolerability (TARGET) trial that responses to tamoxifen may occur in patients relapsing after AIs, current UK guidelines note that the role for tamoxifen following the use of AIs is also uncertain. 7 Fulvestrant is not recommended by NICE within its licensed indication, as an alternative to aromatase inhibitors for the treatment of oestrogen- Specification for manufacturer/sponsor submission of evidence Page 37 of 292

38 receptor positive, locally advanced or metastatic breast cancer in postmenopausal women whose cancer has relapsed on or after adjuvant anti-oestrogen therapy, or who have disease progression on antioestrogen therapy. UK clinical consensus is that the use of fulvestrant following AI failure is a valid clinical option for post menopausal patients. 7 Another option post-ai, for responders to sequential endocrine therapy, is to treat with oral oestrogens and progestins, including megestrol acetate. 5-7 Role of everolimus As stated in the NICE Guideline CG81 on diagnosis and treatment of advanced breast cancer, there is a need to establish effective endocrine therapy for postmenopausal women with HR+ tumours who progress on treatment with an AI. 8 This is particularly relevant since approximately 50% of women with HR+ breast cancer present with de novo resistance to AIs and hence need an effective alternative therapy, while nearly all initial responders to endocrine therapy develop resistance at some point, resulting in disease progression Furthermore, around 50 to 60% of women in whom their first-line endocrine therapy fails will not respond adequately to subsequent lines of endocrine therapy, leaving limited treatment options. 14 In addition, with increasing use of AIs in the adjuvant setting, more patients who progress to metastatic disease have already been exposed to AIs and may therefore have already developed resistance to these agents. 7 In clinical studies, the addition of everolimus to endocrine therapy (exemestane or tamoxifen) has been shown to induce tumour responses and significantly prolong PFS and OS (as discussed in section 6). 18,22 The st international consensus guidelines for advanced breast cancer and the updated 2012 NCCN guidelines both acknowledge the recent, emerging evidence that the addition of everolimus to treatment with an AI improves outcomes in patients with endocrine resistance. 4,5 Specification for manufacturer/sponsor submission of evidence Page 38 of 292

39 2.6 Please describe any issues relating to current clinical practice, including any variations or uncertainty about best practice. There are a number of relevant issues relating to the use of endocrine therapy in postmenopausal women with advanced/metastatic HR+ breast cancer. Currently, endocrine therapy is recommended for postmenopausal women with advanced/metastatic HR+ breast cancer and AIs are recommended over tamoxifen. However, when considering lines of endocrine therapy, there is little evidence to indicate the best sequence in which to use the available endocrine therapies, which include both steroidal and NSAIs, as well as tamoxifen. 5,7,59 Thus patients eligible to receive everolimus in combination with exemestane may have received various different endocrine therapies prior to progression of advanced/metastatic disease. Endocrine therapy may be used both in the adjuvant setting and for advanced/metastatic disease. Failure of NSAI therapy may have occurred in either setting. Thus patients with metastatic disease may receive everolimus as first-line therapy in the metastatic setting (if NSAI therapy previously failed in the adjuvant setting) or as second/third-line therapy in the metastatic setting (ie following failure of NSAIs in the metastatic setting). Treatment options following failure of AI therapy are unclear. Possible options include switching to the other class of AI (ie use of a steroidal AI if a NSAI was previously used, and vice versa), the use of tamoxifen, the use of oral oestrogens or progesterones in hormone-responsive patients, or treatment with fulvestrant a product licensed following failure of an anti-oestrogen but not recommended by NICE. Thus there is no standard of care for postmenopausal women with advanced/metastatic HR+ breast cancer following failure of AI therapy. 5,7,59 Specification for manufacturer/sponsor submission of evidence Page 39 of 292

40 It should also be noted, as stated in current NICE guidance, that current practice and availability of certain treatments and procedures varies across the UK. 5,7, Please identify the main comparator(s) and justify their selection. The indication considered in this submission is treatment of HR+, HER2 advanced breast cancer in postmenopausal women, after failure following NSAIs, with everolimus given in combination with exemestane. 43 The following comparators, as monotherapy, are considered in this submission, in agreement with the NICE final scope: the endocrine therapies exemestane (a steroidal AI), fulvestrant, tamoxifen, and chemotherapy (the latter including those options described in NICE CG81 8 ). The choice of comparators is based on the recent UK guidance for the treatment of metastatic breast cancer and the 1 st international consensus guidelines for advanced breast cancer. 5,7 Both of these evidence-based guidelines recommend endocrine therapy as the preferred approach for most postmenopausal women with HR+ advanced disease and further favour AIs as first-line endocrine therapy. On failure of first-line AI therapy, these guidelines recommend that subsequent treatment lines employ tamoxifen, another AI (ie steroidal AI if NSAIs used previously), fulvestrant (although this therapy is not recommended by NICE following failure of antioestrogen therapy) or the progestogen megestrol acetate. This is consistent with the views of clinical experts at the NICE scoping workshop who agreed that patients failing on NSAI therapy generally do not receive further NSAI therapy and thus the NSAI therapies anastrozole and letrozole were not considered relevant comparators within this submission. Chemotherapy is only recommended on failure of endocrine therapy or when a rapid treatment response is required. Specification for manufacturer/sponsor submission of evidence Page 40 of 292

41 2.8 Please list therapies that may be prescribed to manage adverse reactions associated with the technology being appraised. Non-infectious pneumonitis is a class effect of rapamycin derivatives and has been observed in approximately 12% of patients treated with everolimus. 43 Dose adjustments may be indicated in patients with moderate symptoms, with treatment interruption and the use of corticosteroids indicated until clinical symptoms resolve. Everolimus has immunosuppressive properties and treatment may predispose patients to an increased risk of infections. Treatment with appropriate antifungal, antibacterial, antiviral or antiprotozoal agents may be required to manage infections. Mouth ulcers, stomatitis and oral mucositis have been observed in patients treated with everolimus and in such cases, topical treatments may be required to treat these conditions. 2.9 Please identify the main resource use to the NHS associated with the technology being appraised. Describe the location of care, staff usage, administration costs, monitoring and tests. Provide details of data sources used to inform resource estimates and values. Everolimus is an oral therapy and requires no special monitoring requirements necessitating additional resource use Does the technology require additional infrastructure to be put in place? No. Specification for manufacturer/sponsor submission of evidence Page 41 of 292

42 3 Equality NICE is committed to promoting equality of opportunity, eliminating unlawful discrimination and fostering good relations between people with particular protected characteristics and others. For further information, please see the NICE website ( 3.1 Identification of equality issues Please let us know if you think that this appraisal: could exclude from full consideration any people protected by the equality legislation who fall within the patient population for which [the treatment(s)] is/are/will be licensed; could lead to recommendations that have a different impact on people protected by the equality legislation than on the wider population, e.g. by making it more difficult in practice for a specific group to access the technology could lead to recommendations that have any adverse impact on people with a particular disability or disabilities Please provide us with any evidence that would enable the Committee to identify and consider such impacts. There are no equality issues or equality legislation pertinent to the technology How has the analysis addressed these issues? Not applicable. Specification for manufacturer/sponsor submission of evidence Page 42 of 292

43 4 Innovation Discuss whether and how you consider the technology to be innovative in its potential to make a significant and substantial impact on health-related benefits, and whether and how the technology is a step-change in the management of the condition. Scientifically and clinically, everolimus is an innovative technology with the potential to make a significant and substantial impact on the management of a patient group for whom there have been limited effective treatment options. A major concern in the treatment of advanced breast cancer in postmenopausal women is the lack or loss of response to endocrine therapy the therapeutic strategy of choice. Patients who develop resistance to endocrine therapy are a very difficult-to-treat patient population. In 2009, NICE guidance (NICE CG81) recommended more research to discover the most effective endocrine therapy for postmenopausal women with HR+ tumours who progress on treatment with AIs (Figure A1) thus highlighting the unmet clinical need for these patients. 8 Everolimus directly addresses this unmet need for effective treatment options for postmenopausal women with HR+ tumours who progress on treatment with an AI. Specification for manufacturer/sponsor submission of evidence Page 43 of 292

44 Figure A1 Sequential systemic endocrine therapy for women with HR+ breast cancer, as defined by NICE NICE CG81 8 Everolimus is the first treatment to redefine the role of endocrine therapy in advanced disease, and achieves this through actions effected at a molecular level in breast cancer cells. Research has revealed that the development of endocrine therapy resistance is associated with activation of the PI3K/Akt/mTOR pathway, which occurs frequently in specific cancers and is often associated with a poor prognosis. 1-3,10 Everolimus targets the P13K/Akt/mTOR pathway and through this mechanism of action, restores sensitivity to endocrine therapy (see section 1.2). Everolimus is the first agent of the mtor-inhibitor class to be licensed in breast cancer. Clinically, this means that everolimus, when combined with AIs, restores sensitivity to the the therapeutic strategy of choice (endocrine therapy) for postmenopausal patients with advanced breast cancer and provides a valuable therapeutic option in patients in whom AI therapy is failing. In Specification for manufacturer/sponsor submission of evidence Page 44 of 292

45 postmenopausal women with HR+, HER2 advanced breast cancer without symptomatic visceral disease after recurrence or progression following a NSAI, treatment with everolimus, in combination with exemestane, extends the benefits of endocrine therapy, helping to prolong survival and delay the need for use of chemotherapy. The PFS benefits of everolimus exceed those that have been observed with endocrine therapy and chemotherapy. For example, PFS in BOLERO-2 compares favourably with the results reported for fulvestrant and fulvestrant plus anastrozole 26 as well as for a recent trial of capecitabine and taxane/anthracycline chemotherapy in patients with HR+ disease. 18,76 Furthermore, PFS is extended without compromising patient quality of life (see section 6.5). 77 Everolimus therefore represents a step-change in the management of advanced/metastatic HR+ breast cancer in postmenopausal women, and has been described by clinicians as a quantum shift in the future of endocrine therapy in this patient population Discuss whether and how you consider that the use of the technology can result in any potential significant and substantial health-related benefits that are unlikely to be included in the qualityadjusted life year (QALY) calculation. Everolimus is an oral treatment that can be conveniently taken by patients at home. The improved HRQoL expected for patients receiving endocrine therapy compared with chemotherapy may translate into increased productivity as many women with advanced breast cancer aim to continue in employment. This potential benefit of everolimus is not captured in the economic model. In addition to sparing patients the adverse effects of chemotherapy, everolimus in combination with exemestane may also reduce overall resource use as chemotherapy has been shown to be associated with Specification for manufacturer/sponsor submission of evidence Page 45 of 292

46 higher resource use compared with endocrine therapy. In particular, in a study of resource use for postmenopausal women with HER2, HR+ advanced breast cancer treated at centres in five European countries, chemotherapy was reported to be associated with more monitoring, more hospitalisation and a greater use of complimentary therapies. 78 The analysis assumes the same level of resource use irrespective of treatment option and therefore might be underestimating the resource use savings associated with everolimus in combination with exemestane. Bone metastases and fractures are recognised to be an important cause of morbidity and mortality in patients with breast cancer, leading to physical pain and reduction of functional independence and quality of life Everolimus slows down the rate of disease progression in bone, thereby reducing the risk of fractures and the risk of bone metastases, both of which adversely affect HRQoL (see section for clinical outcomes and HRQoL data). 65,66 The beneficial effects of everolimus on bone are likely to be even more important in the longer term both in terms of positive impact on quality of life and reducing the costs of treating bone complications such as fractures Please identify the data you have used to make these judgements, to enable the Appraisal Committee to take account of these benefits. The benefits of oral treatment, although not captured by the QALY, are an important consideration, as was acknowledged and taken into account by the Committee in the appraisal of abiraterone for prostate cancer (TA259). 79 These could be captured indirectly by assigning a lower evidence based utility value for the non oral treatment options in the analysis. However, there are limited HRQoL data for treatment options such as chemotherapy in this setting. Specification for manufacturer/sponsor submission of evidence Page 46 of 292

47 5 Statement of the decision problem In this section the manufacturer or sponsor should specify the decision problem that the submission addresses. The decision problem should be derived from the final scope issued by NICE and should state the key parameters that the information in the evidence submission will address. Final scope issued by NICE Decision problem addressed in the submission Rationale if different from the scope Population Postmenopausal women with HER2- negative, oestrogen receptor-positive locally advanced or metastatic breast cancer whose disease has recurred or progressed after prior therapy which has included a nonsteroidal aromatase inhibitor. Postmenopausal women with HER2 negative, hormone receptor-positive, advanced breast cancer without symptomatic visceral disease after recurrence or progression following a nonsteroidal aromatase inhibitor (NSAI). To reflect the licence Intervention Everolimus in combination with an aromatase inhibitor Everolimus in combination with exemestane To reflect the licence Comparator(s) Exemestane Tamoxifen Fulvestrant Chemotherapy (in accordance with NICE guidance) Exemestane Tamoxifen Fulvestrant Chemotherapy (in accordance with NICE guidance, specifically docetaxel, capecitabine and doxorubicin) n/a Outcomes Overall survival Progression free survival Response rate Adverse effects of treatment Health-related quality of life Overall survival Progression free survival Response rate Adverse effects of treatment Health-related quality of life n/a Specification for manufacturer/sponsor submission of evidence Page 47 of 292

48 Final scope issued by NICE Decision problem addressed in the submission Rationale if different from the scope Economic analysis The reference case stipulates that the cost effectiveness of treatments should be expressed in terms of incremental cost per quality-adjusted life year. The reference case stipulates that the time horizon for estimating clinical and cost effectiveness should be sufficiently long to reflect any differences in costs or outcomes between the technologies being compared. Costs will be considered from an NHS and Personal Social Services perspective. Cost-effectiveness presented as incremental cost per quality-adjusted life year (QALY) Time horizon: lifetime (10 years) base case in line with other late stage cancer models Perspective: NHS and Personal Social Services n/a Subgroups to be considered None stated None Special considerations, including issues related to equity or equality Guidance will only be issued in accordance with the marketing authorisation. None n/a Specification for manufacturer/sponsor submission of evidence Page 48 of 292

49 Section B Clinical and cost effectiveness When estimating clinical and cost effectiveness, particular emphasis should be given to adhering to the reference case (see the NICE document Guide to the methods of technology appraisal Reasons for deviating from the reference case should be clearly explained. Particularly important features of the reference case include those listed in the table below. Element of health technology assessment Defining the decision problem Comparator(s) Reference case Section in Guide to the methods of technology appraisal The scope developed by NICE and Therapies routinely used in the NHS, including technologies regarded as current best practice and Perspective costs NHS and PSS to Perspective benefits All health effects on individuals to Type of economic Cost-effectiveness analysis and evaluation Synthesis of Based on a systematic review 5.3 evidence on outcomes Measure of health QALYs 5.4 effects Source of data for measurement of HRQL Reported directly by patients and carers 5.4 Source of preference data for valuation of changes in HRQL Discount rate Equity weighting Representative sample of the public An annual rate of 3.5% on both costs and health effects An additional QALY has the same weight regardless of the other characteristics of the individuals receiving the health benefit HRQL, health-related quality of life; NHS, National Health Service; PSS, personal social services; QALY(s), quality-adjusted life-year(s) Specification for manufacturer/sponsor submission of evidence Page 49 of 292

50 6 Clinical evidence Manufacturers and sponsors are requested to present clinical evidence for their technology in the following sections. This section should be read in conjunction with NICE s Guide to the methods of technology appraisal, sections 3 and to Identification of studies Describe the strategies used to retrieve relevant clinical data, both from the published literature and from unpublished data that may be held by the manufacturer or sponsor. The methods used should be justified with reference to the decision problem. Sufficient detail should be provided to enable the methods to be reproduced, and the rationale for any inclusion and exclusion criteria used should be provided. Exact details of the search strategy used should be provided in section 10.2, appendix 2. A systematic literature review was performed to identify all studies evaluating the clinical effectiveness, costs and safety of everolimus in combination with another endocrine treatment (for example, exemestane, fulvestrant, tamoxifen) as second-line therapy or beyond for HR+, HER2 advanced breast cancer in postmenopausal women. The databases searched included Medline, Medline In-Process, EMBASE and the Cochrane Library. Two reviewers independently applied the predefined inclusion criteria to select studies for inclusion based on titles and abstracts; any discrepancies were resolved through discussion or by consulting a third reviewer. All publications meeting the inclusion criteria based on titles and abstracts were obtained as full documents and reassessed against the inclusion criteria. 6.2 Study selection Describe the inclusion and exclusion selection criteria, language restrictions and the study selection process. A justification should Specification for manufacturer/sponsor submission of evidence Page 50 of 292

51 be provided to ensure that the rationale is transparent. A suggested format is provided below. Studies needed to meet the inclusion and exclusion criteria detailed in Table B1 to be included in the review. Table B1 Eligibility criteria used in search strategy Inclusion criteria Exclusion criteria Clinical effectiveness Population: postmenopausal women with HR+, HER2 locally advanced or metastatic breast cancer whose disease had recurred or progressed following endocrine therapy, including treatment with non-steroidal aromatase inhibitors Intervention: everolimus in combination with exemestane, fulvestrant or tamoxifen Comparator: exemestane, fulvestrant or tamoxifen Outcomes: clinical benefit rate, response rate (complete, partial, stable disease), overall survival, progression-free survival or time to progression, AEs and discontinuations due to AEs, health-related quality of life, time to treatment discontinuation Study design: RCTs of any duration and crossover RCTs if data were presented at crossover; non-randomised comparative and uncontrolled studies reporting AEs were also eligible for inclusion Language: there was no language restriction applied to the search; studies with English abstracts, but whose full reports were in languages other than English were not extracted but were listed for information only Publication status: published, unpublished and grey literature was eligible; studies published as abstracts or conference presentations were included if an associated published full paper could not be found and adequate data were presented None specified AE, adverse event; HR, hormone receptor; HER2, human epidermal growth factor receptor 2; RCT, randomised controlled trial. Specification for manufacturer/sponsor submission of evidence Page 51 of 292

52 6.2.2 A flow diagram of the numbers of studies included and excluded at each stage should be provided using a validated statement for reporting systematic reviews and meta-analyses such as the QUOROM statement flow diagram ( The total number of studies in the statement should equal the total number of studies listed in section Figure B1 PRISMA diagram of included and excluded studies in the systematic review PRISMA, preferred reporting items for systematic reviews and meta-analyses When data from a single RCT have been drawn from more than one source (for example, a poster and a published report) and/or when trials are linked (for example, an open-label extension to an RCT), this should be made clear. Where data for a single randomised controlled trial (RCT) are taken from multiple sources these are clearly referenced. Specification for manufacturer/sponsor submission of evidence Page 52 of 292

53 Complete list of relevant RCTs Provide details of all RCTs that compare the intervention with other therapies (including placebo) in the relevant patient group. The list must be complete and will be validated by independent searches conducted by the Evidence Review Group. This should be presented in tabular form. A suggested format is presented below. Two relevant RCTs were identified by the systematic review: BOLERO-2 and TAMRAD. These are summarised in Table B2. The BOLERO-2 phase III study was conducted in postmenopausal women with HR+, HER2 advanced breast cancer that had recurred or progressed following prior treatment with letrozole or anastrozole. Everolimus in combination with exemestane was compared with placebo in combination with exemestane. 18 The TAMRAD phase II trial was conducted in postmenopausal women with HR+, HER2 metastatic breast cancer who had received prior AI therapy in the adjuvant or metastatic setting and developed progressive disease. Everolimus in combination with tamoxifen was compared with tamoxifen monotherapy. 22 Specification for manufacturer/sponsor submission of evidence Page 53 of 292

54 Table B2 List of relevant RCTs Trial no. (acronym) BOLERO-2 NCT TAMRAD NCT Intervention Comparator Population Primary study ref. Everolimus (10 mg/day) in combination with exemestane (25 mg/day) Everolimus (10 mg/day) in combination with tamoxifen (20 mg/day) Placebo in combination with exemestane (25 mg/day) Oral tamoxifen (20 mg/day) Postmenopausal women with HR+, HER2 advanced breast cancer whose disease was refractory to prior letrozole or anastrozole Postmenopausal women with HR+, HER2 metastatic breast cancer with prior exposure to AIs, and experiencing progressive disease Baselga et al., NEJM 2012; 366: Bachelot et al., JCO 2012; 30: AI, aromatase inhibitor; BOLERO, breast cancer trials of oral everolimus; HR, hormone receptor; HER2, human epidermal growth factor receptor 2; RCT, randomised controlled trial; TAMRAD, tamoxifen in combination with everolimus Please highlight which of the RCTs identified above compares the intervention directly with the appropriate comparator(s) with reference to the decision problem. If there are none, please state this. BOLERO-2 is directly relevant to the decision problem because it compares the intervention of interest, everolimus in combination with exemestane, with one of the comparators of interest, exemestane (in combination with placebo). The BOLERO-2 patient population (described in section 6.3.4) is also relevant to the decision problem, because all patients would be eligible to receive everolimus according to the licence for breast cancer. The TAMRAD trial provides supporting evidence as it compares everolimus in combination with tamoxifen with tamoxifen monotherapy, also in patients corresponding to the licensed indication. Tamoxifen alone is one of the comparators of interest, while the combination provides additional evidence of the efficacy and safety of everolimus in combination with endocrine therapy, although this is not an approved combination for this indication. Specification for manufacturer/sponsor submission of evidence Page 54 of 292

55 6.2.6 When studies identified above have been excluded from further discussion, a justification should be provided to ensure that the rationale for doing so is transparent. For example, when studies have been identified but there is no access to the level of trial data required, this should be indicated. Both of the RCTs identified are discussed. List of relevant non-rcts Please provide details of any non-rcts (for example experimental and observational data) that are considered relevant to the decision problem and a justification for their inclusion. Full details should be provided in section 6.8 and key details should be presented in a table; the following is a suggested format. No non-rcts meeting the inclusion criteria were identified by the systematic review. 6.3 Summary of methodology of relevant RCTs As a minimum, the summary should include information on the RCT(s) under the subheadings listed in this section. Items 2 to 14 of the CONSORT checklist should be provided, as well as a CONSORT flow diagram of patient numbers ( It is expected that all key aspects of methodology will be in the public domain; if a manufacturer or sponsor wishes to submit aspects of the methodology in confidence, prior agreement must be requested from NICE. When there is more than one RCT, the information should be tabulated. Specification for manufacturer/sponsor submission of evidence Page 55 of 292

56 Methods Describe the RCT(s) design (for example, duration, degree and method of blinding, and randomisation) and interventions. Include details of length of follow-up and timing of assessments. The following tables provide a suggested format for when there is more than one RCT. The comparative methodology for the two identified RCTs is summarised in Table B3. BOLERO-2 was a multicentre, international, double-blind, phase III trial in which 724 patients were randomised to receive exemestane (25 mg/day) in combination with oral everolimus (10 mg/day) or exemestane (25 mg/day) in combination with matched placebo (10 mg/day). 18 Patients were randomised 2:1 and stratified according to the presence of visceral metastasis and previous sensitivity to endocrine therapy. This was defined as, at least 24 months of endocrine treatment in the adjuvant setting prior to recurrence, or either a response or disease stabilisation lasting for at least 24 weeks with endocrine therapy given for advanced disease. Treatment continued until disease progression, unacceptable toxicity, or withdrawal of consent. No crossover between treatment arms was allowed. The TAMRAD trial was an open-label, phase II study, carried out in multiple centres in France, in which 111 patients were randomised to receive tamoxifen (20 mg/day) in combination with everolimus (10 mg/day) or tamoxifen alone (20 mg/day). 22 Patients were randomised 1:1 and stratified based on hormone resistance (primary or secondary). Primary hormone resistance was defined as relapse during or within 6 months of stopping adjuvant AI therapy or progression within 6 months of starting AI treatment for metastatic disease. Secondary resistance was defined as relapse more than 6 months after stopping adjuvant AI therapy or previous response to AI treatment for at least 6 months in the metastatic setting. Treatment continued until disease progression, unacceptable toxicity or Specification for manufacturer/sponsor submission of evidence Page 56 of 292

57 withdrawal of consent. No crossover between treatment arms was planned or allowed. The primary and secondary outcomes reported in the two trials are commonly used measures of efficacy in trials of interventions for advanced breast cancer, and included PFS, overall response rate (ORR), clinical benefit rate (CBR; defined as complete response [CR] plus partial response [PR] plus stable disease [SD]), OS and time to progression (TTP). 18,22 Specification for manufacturer/sponsor submission of evidence Page 57 of 292

58 Table B3 Trial no. (acronym) Comparative summary of methodology of the RCTs BOLERO-2 18,60,80 TAMRAD 22,81 Location International (189 centres in 24 countries) France Design Duration of study Method of randomisation Method of blinding (care provider, patient and outcome assessor) Randomised, double-blind, phase III trial Treatment continued until disease progression, the development of unacceptable toxicity or withdrawal of consent The protocol provided detailed guidelines for dose interruptions for up to 4 weeks in response to grade 2 to 4 AEs, and for dose reductions for everolimus and matched placebo after resolution of AEs. From an initial dosage of 10 mg daily, reduction to 5 mg daily and subsequently to 5 mg every second day were permitted Patients who interrupted therapy for more than 4 weeks were withdrawn from the study Randomisation at a 2:1 ratio in favour of the everolimus in combination with exemestane group and stratified according to the presence of visceral metastasis and previous sensitivity to endocrine therapy No crossover was allowed Patients were assigned to each treatment arm by centralised allocation (ie, interactive web response system/interactive voice response system) Patients, investigator staff, persons performing the assessments, all Novartis personnel and individuals at central laboratories (including central imaging) were to remain blinded to the identity of the treatment from the time of randomisation until database lock Randomised, open-label, phase II trial Treatment continued until disease progression, unacceptable toxicity or withdrawal of consent In the event of any grade 3/4 AEs possibly related to the drug, tamoxifen doses could be interrupted for up to 14 days In the event of any grade 3/4 AEs possibly related to the drug, everolimus dose reduction to 5 mg/day or interruption for up to 14 days was allowed For both study drugs, if AEs did not resolve in 14 days, or recurred after study drug reintroduction, treatment was terminated In the tamoxifen in combination with everolimus arm, if tamoxifen was stopped, the investigators could decide to either stop or continue everolimus Randomisation at a 1:1 ratio was stratified according to primary or secondary hormone resistance No crossover was planned or allowed Patients were not blinded Specification for manufacturer/sponsor submission of evidence Page 58 of 292

59 Trial no. (acronym) Intervention(s) and comparator(s) BOLERO-2 18,60,80 TAMRAD 22,81 Everolimus (10 mg/day) in combination with exemestane (25 mg/day) (n = 485) Everolimus (10 mg/day) in combination with tamoxifen (20 mg/day) (n = 54); Primary outcomes (including scoring methods and timings of assessments) Exemestane (25 mg/day) in combination with matched placebo (10 mg/day) (n = 239) Progression-free survival based on local and central assessment Pre-planned analyses of PFS were to be undertaken after 317 and 528 local PFS events Tumour assessments based on the RECIST v1.0 criteria 82 were carried out locally every 6 weeks until progression Tamoxifen (20 mg/day) (n = 57) Clinical benefit rate at 6 months Tumour response in measurable lesions was assessed according to the RECIST v1.0 criteria 82 at baseline and every 2 months until the 6-month visit, followed by every 3 months up until 18 months Secondary outcomes (including scoring methods and timings of assessments) Duration of follow-up Overall survival Overall response rate Clinical benefit rate Time to response Duration of response Safety (AEs, biomarker analysis, vital signs, time to deterioration of ECOG performance status) Quality of life, evaluated using the EORTC QLQ-C30 and breast cancer module BR23 PFS analyses: 1 st interim analysis 7 months 2 nd interim analysis 12 months Final PFS analysis 18 months Overall survival analysis 16 months Time to progression Overall survival Objective response rate Toxicity safety assessments were performed monthly until the 6-month visit, then every 3 months until 18 months, and included AE monitoring and recording, haematologic and clinical biochemical levels (laboratory evaluations), vital sign measurements and physical examinations Median follow-up (range): Everolimus in combination with tamoxifen: 23.7 months (2.6 to 32.7 months); Tamoxifen: 24.2 months (0.9 to 36.2 months) Bachelot et al., ,81 ; Baselga et al., ,80 ; Therasse et al., 2000; 82 BOLERO-2 7-month CSR. 60 AE, adverse event; ECOG, Eastern Cooperative Oncology Group; EORTC QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire; PFS, progression-free survival; RECIST, response evaluation criteria in solid tumours. Specification for manufacturer/sponsor submission of evidence Page 59 of 292

60 Participants Provide details of the eligibility criteria (inclusion and exclusion) for the trial. The following table provides a suggested format for the eligibility criteria for when there are more than one RCT. Highlight any differences between the trials. Inclusion and exclusion criteria for the two RCTs are listed in Table B4. BOLERO-2 included postmenopausal women with HR+, HER2 advanced breast cancer that was either refractory to letrozole or anastrozole in the adjuvant setting or progressed following treatment with letrozole or anastrozole in the metastatic setting; these agents did not have to be the most recent treatment. 18 Patients were also required to have at least one measurable lesion or mainly lytic bone lesions in the absence of measurable disease. Exclusion criteria included a history of brain metastases or other central nervous system metastases and previous treatment with exemestane or mtor inhibitors. 60 The TAMRAD trial had similar inclusion and exclusion criteria to BOLERO Postmenopausal women with HR+, HER2 metastatic breast cancer were eligible for enrolment. Patients had to be refractory to AI therapy and could have primary or secondary resistance (see section for definitions). Previous treatment with tamoxifen was an exclusion criterion specific to the TAMRAD trial. Specification for manufacturer/sponsor submission of evidence Page 60 of 292

61 Table B4 Trial no. (acronym) Eligibility criteria in the RCTs Inclusion criteria Exclusion criteria BOLERO- 2 18,60 Postmenopausal women with HR+, HER2 advanced breast cancer Disease refractory to previous letrozole or anastrozole, defined as recurrence during or within 12 months after the end of adjuvant treatment or progression during or within 1 month after the end of treatment for advanced disease At least one measurable lesion or mainly lytic bone lesions in the absence of measurable disease ECOG performance status of 2 Adequate organ and haematological function History of brain or other central nervous system metastases Prior treatment with exemestane or mtor inhibitors. TAMRAD 22,81 Postmenopausal women aged 18 years HR+, HER2 metastatic breast cancer Previously treated with an AI in the adjuvant or metastatic setting Presence of at least one measurable or evaluable metastatic lesion Progressive disease as assessed by local investigator ECOG performance status 2 Adequate organ and haematological function Symptomatic brain metastases, pulmonary carcinomatous lymphangitis involving > 50% of the lungs, hepatic metastases in more than one third of the liver Previous tamoxifen treatment unless given in the adjuvant setting and terminated more than 1 year before metastatic relapse History of other malignancy within 5 years before randomisation with the exceptions of adequately treated in situ uterine carcinoma or nonmelanomatous skin cancer Previous treatment with or known sensitivity to mtor inhibitor, or treatment with molecules that interfere with isoenzyme CYP3A Any severe or uncontrolled medical condition AI; aromatase inhibitor; CYP, cytochrome P450; ECOG, Eastern Cooperative Oncology Group; HR, hormone receptor; HER2, human epidermal growth factor receptor type 2; mtor, mammalian target of rapamycin; PgR, progesterone receptor; RCT, randomised controlled trial. Adapted from Pharmaceutical Benefits Advisory Committee (2008) Guidelines for preparing submissions to the Pharmaceutical Benefits Advisory Committee (Version 4.3). Canberra: Pharmaceutical Benefits Advisory Committee Specification for manufacturer/sponsor submission of evidence Page 61 of 292

62 6.3.4 Describe the patient characteristics at baseline. Highlight any differences between study groups. The following table provides a suggested format for the presentation of baseline patient characteristics for when there is more than one RCT. Baseline demographics and disease-related characteristics are summarised in Table B5. In BOLERO-2, the patients were well balanced between treatment groups and typical of a UK population. 18,26 The majority of the patients (84%) were sensitive to prior endocrine therapy and 75% of patients had received a NSAI as their most recent treatment. Approximately half of the population were heavily pre-treated, having received three or more prior therapies. Twenty-six percent of the patients had received chemotherapy for metastatic disease and 56% of the patients had visceral involvement. In the TAMRAD trial, the patients were generally well balanced between treatment groups (except for the combination group having a better ECOG performance status) with similar baseline and disease characteristics to those included in BOLERO Patients were stratified based on endocrine resistance; 49% of patients had primary resistance and 51% of patients had secondary resistance. Specification for manufacturer/sponsor submission of evidence Page 62 of 292

63 Table B5 Characteristics of participants in the RCTs across randomised groups Trial no. (acronym) BOLERO-2 18,60 (N = 724) TAMRAD 22 (N = 111) Baseline characteristic Everolimus in combination with exemestane (n = 485) Placebo in combination with exemestane (n = 239) Everolimus in combination with tamoxifen (n = 54) Tamoxifen (n = 57) Age, median (range) 62 years (34 93) 61 years (28 90) 63 years (41 81) 66 years (42 86) Ethnic background, n (%) White 361 (74) 186 (78) NR NR Black 13 (3) 3 (1) NR NR Asian 98 (20) 45 (19) NR NR Other 13 (3) 5 (2) NR NR Performance status, n (%) ECOG (60) 142 (59) 32 (59) 23 (40) ECOG (36) 84 (35) 18 (33) 28 (49) ECOG 2 9 (2) 7 (3) 3 (6) 3 (5) Missing 9 (2) 6 (3) 1 (2) 3 (5) Disease-free interval, median a (range) Duration of metastatic disease, median (range) Hormone receptor status, n (%) 58 months (1 340) 57 months (5 316) NR NR NR NR 1.1 years ( ) 1.2 years ( ) ER+, PgR+ 351 (72) 173 (72) 35 (65) 41 (72) ER+, PgR 134 (28) 66 (28) 19 (35) 15 (26) ER, PgR (2) HER2, n (%) 485 (100) b 239 (100) b 53 (98) 53 (93) c Specification for manufacturer/sponsor submission of evidence Page 63 of 292

64 Trial no. (acronym) BOLERO-2 18,60 (N = 724) TAMRAD 22 (N = 111) Baseline characteristic Number of metastatic sites, n (%) Everolimus in combination with exemestane (n = 485) Placebo in combination with exemestane (n = 239) Everolimus in combination with tamoxifen (n = 54) (32) 69 (29) NR NR (31) 81 (34) NR NR Tamoxifen (n = 57) (36) 89 (37) 13 (24) 16 (28) Type of metastatic sites, n (%) Visceral 281 (58) 143 (60) 31 (57) 28 (49) Lung 140 (29) 79 (33) NR NR Liver 160 (33) 72 (30) NR NR Bone 369 (76) 184 (77) 41 (76) 45 (79) Bone only 105 (22) 50 (21) 16 (30) 14 (25) Measurable disease, n (%) 338 (70) 162 (68) 21 (39) 31 (55) Previous sensitivity 409 (84) 201 (84) NR NR to endocrine therapy, n (%) Primary resistance NR NR 26 (48) 28 (49) Secondary NR NR 27 (50) 29 (51) resistance Missing NR NR 1 (2) 0 Prior treatment with aromatase inhibitor (NSAI in BOLERO-2), n (%) Adjuvant 142 (29) d 57 (24) e 21 (39) h 24 (42) h Metastatic 343 (71) f 182 (76) g 37 (69) h 37 (65) h Specification for manufacturer/sponsor submission of evidence Page 64 of 292

65 Trial no. (acronym) BOLERO-2 18,60 (N = 724) TAMRAD 22 (N = 111) Baseline characteristic Prior chemotherapy, n (%) Everolimus in combination with exemestane (n = 485) Placebo in combination with exemestane (n = 239) Everolimus in combination with tamoxifen (n = 54) Tamoxifen (n = 57) Adjuvant 211 (44) 95 (40) 25 (46) k 32 (56) k Metastatic 125 (26) i 61 (26) j 13 (24) k 15 (26) k Prior anti-oestrogen treatment, n (%) Any 276 (57) 140 (59) NR NR Tamoxifen 230 (47) 118 (49) 18 (33) l 24 (42) l Fulvestrant 80 (17) 39 (16) NR NR Number of prior therapies m, n (%) 1 76 (16) 42 (18) NR NR (30) 71 (30) NR NR (54) 126 (53) NR NR Purpose of most recent treatment Adjuvant therapy (21) (16) NR NR Treatment of advanced/ metastatic disease (79) (84) NR NR Bachelot et al., ,81 ; Baselga et al., ,80 ; 7-month BOLERO-2 CSR. 60 a Disease-free interval is defined as the time from diagnosis of breast cancer to first relapse in patients who received adjuvant therapy (308 patients in the combination therapy group and 153 patients in the exemestane alone group); b Data on HER2 status of two patients were missing; c One patient in the tamoxifen in combination with everolimus arm with HER2+ status had a tumour that only slightly overexpressed HER2 and therefore was not considered a major protocol violation; d Five patients were miscoded as having received NSAI therapy for prevention only; e Two patients were miscoded as having received NSAI therapy for prevention only; f Twenty patients received NSAI therapy in both adjuvant and metastatic settings; g Twelve patients received NSAI therapy in both adjuvant and metastatic settings; h Some patients received AI therapy in both adjuvant and metastatic settings; i 58 patients received chemotherapy in both adjuvant and metastatic settings; j 38 patients received chemotherapy in both adjuvant and metastatic settings; k Some patients received chemotherapy in both adjuvant and metastatic settings; l Therapy received in adjuvant setting; m Previous therapies include those used in the adjuvant setting or to treat advanced disease. ECOG, Eastern Cooperative Oncology Group; ER, oestrogen receptor; HER2, human epidermal growth factor receptor type 2; NR, not reported; PgR, progesterone receptor Adapted from Pharmaceutical Benefits Advisory Committee (2008) Guidelines for preparing submissions to the Pharmaceutical Benefits Advisory Committee (Version 4.3). Canberra: Pharmaceutical Benefits Advisory Committee Specification for manufacturer/sponsor submission of evidence Page 65 of 292

66 Outcomes Provide details of the outcomes investigated and the measures used to assess those outcomes. Indicate which outcomes were specified in the trial protocol as primary or secondary, and whether they are relevant with reference to the decision problem. This should include therapeutic outcomes, as well as patient-related outcomes such as assessment of health-related quality of life (HRQL), and any arrangements to measure compliance. Data provided should be from pre-specified outcomes rather than posthoc analyses. When appropriate, also provide evidence of reliability or validity, and current status of the measure (such as use within UK clinical practice). The following table provides a suggested format for presenting primary and secondary outcomes when there is more than one RCT. The primary and secondary outcomes for BOLERO-2 and the TAMRAD trial are summarised in Table B6. In BOLERO-2, the primary outcome was PFS, based on local and central radiological assessment. 18 Secondary outcomes included OS, CBR, ORR, time to overall response, duration of overall response, time to deterioration of Eastern Cooperative Oncology Group (ECOG) performance status, safety and quality of life. Tumour assessments were based on computed tomographic (CT) scanning or magnetic resonance imaging (MRI) of the chest, abdomen and pelvis performed at baseline and every 6 weeks until disease progression. Tumour response was assessed using Response Evaluation Criteria in Solid Tumours (RECIST) version In addition, bone scans were performed every 6 weeks using radiography, CT scanning or MRI. HRQoL was evaluated every 6 weeks using the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30) and the breast module, BR23. 83,84 Time to definitive Specification for manufacturer/sponsor submission of evidence Page 66 of 292

67 deterioration ( 5%), in the global health status on the EORTC QLQ-C30 scale as well as in each of the three functional scales (emotional, physical and social functioning), was compared between the two treatment groups. An analysis of the bone turnover markers bone-specific alkaline phosphatase (BSAP; a marker for osteoclast metabolism), amino-terminal propeptide of type 1 collagen (P1NP; a bone formation marker) and c- telopeptide of type 1 collagen (CTX; a bone resorption marker) was conducted at 6 and 12 weeks. 20,85 The primary outcome in the TAMRAD trial was 6-month CBR. 22 Scans were assessed locally and a central review of all radiologic reports was also performed. Tumour response in measurable lesions was assessed using RECIST v1.0. In non-measurable lesions, a CR was defined as the complete disappearance of a lesion and normalisation of cancer antigen 15-3 levels. Progressive disease was defined as the appearance of at least one new lesion or unequivocal progression of an existing lesion. CT scans were used for tumour assessments and were performed at baseline and every 2 months until the 6-month visit, then every 3 months until 18 months. The secondary outcomes included TTP, OS, ORR and toxicity based on AEs and laboratory measures. Specification for manufacturer/sponsor submission of evidence Page 67 of 292

68 Table B6 Primary and secondary outcomes of the RCTs Trial no. (acronym) Primary outcome(s) and measures Reliability/ validity/current use in clinical practice PFS is a recognised outcome measure for assessment of treatments for metastatic breast cancer; 86 unlike OS, it is not influenced by therapy or relapse and therefore can be a more accurate measure of treatment efficacy RECIST v1.0 criteria are approved for measurement of tumour response 82 Secondary outcome(s) and measures Reliability/validity/ current use in clinical practice BOLERO PFS based -2 18,60,80 on local and central assessment of radiographic studies, with tumour response assessed using RECIST v1.0. OS ORR (CR, PR, SD and PD; unknown or too early) assessed using RECIST v1.0 criteria CBR Time to response Duration of response Health-related quality of life (EORTC QLQ-C30 and BR23) Time to deterioration of ECOG performance status Safety (adverse events, biomarker analysis, vital signs) Bone turnover markers OS TTP ORR Toxicity OS is a well-recognised outcome measure for treatments of metastatic breast cancer since therapies aim to prolong survival ORR and CBR are recognised measures of efficacy in metastatic breast cancer and reflect clinical benefit The EORTC QLQ-C30 is a patient-reported outcome instrument which is used widely to assess health-related quality of life in patients with cancer TAMRAD 22 CBR a (method of assessment not reported) CBR is a wellrecognised measure of efficacy in patients with metastatic breast cancer; it is accepted that therapies leading to disease stabilisation are likely to lead to improved OS OS is a well-recognised outcome measure for treatments of metastatic breast cancer since therapies aim to prolong survival TTP and ORR are recognised measures of efficacy in metastatic breast cancer and reflect clinical benefit; a recent study in breast and colorectal cancer found that the time to disease progression is strongly correlated with an increase in overall survival 87 a CBR was defined as the percentage of all patients with a complete or partial response or stable disease at 6 months. CBR, clinical benefit rate; CR, complete response; ECOG, Eastern Cooperative Oncology Group; EORTC QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; PD, progressive disease; PR, partial response; RECIST, Response Evaluation Criteria in Solid Tumours; SD, stable disease; TTP, time to progression. Specification for manufacturer/sponsor submission of evidence Page 68 of 292

69 Statistical analysis and definition of study groups State the primary hypothesis or hypotheses under consideration and the statistical analysis used for testing hypotheses. Also provide details of the power of the study and a description of sample size calculation, including rationale and assumptions. Provide details of how the analysis took account of patients who withdrew (for example, a description of the intention-to-treat analysis undertaken, including censoring methods; whether a perprotocol analysis was undertaken). The following table provides a suggested format for presenting the statistical analyses in the trials when there is more than one RCT. The statistical analyses used in the two RCTs are summarised in Table B8. The objective of BOLERO-2 was to evaluate the efficacy and safety of the combination of everolimus and exemestane in patients with HR+ advanced breast cancer refractory to NSAIs. A power calculation determined that a total of 528 PFS events were required for the final analysis to detect a hazard ratio of 0.74 with 90% power with the use of a log-rank test and a two-look Lan DeMets group sequential design with an O Brien Flemingtype boundary at a one-sided cumulative 2.5% level of significance. Assuming a median PFS of 3.7 months in the everolimus alone group, 18 months of recruitment, a 10% rate of loss to follow-up, and a 2:1 randomisation ratio in favour of the everolimus in combination with exemestane group, 705 patients were to be randomly assigned. In addition to the final analysis, an interim analysis was to be performed after 317 PFS events (60%). An intention-to-treat (ITT) analysis was used. Patients who discontinued one or both study treatments for any reason other than progression were required to follow the same schedule of assessments until progression. Three OS analyses were planned: at the time of the interim PFS analysis; after 173 deaths; and after 392 deaths (Table B7). ORR (defined as the proportion of patients with best overall response of either CR or PR) and CBR (defined as the proportion of patients with best Specification for manufacturer/sponsor submission of evidence Page 69 of 292

70 overall response of CR, PR, or SD 24 weeks) were summarised along with the exact 95% confidence intervals (CIs). A stratified Cochran Mantel Haenszel (CMH) test was used to compare the two treatment groups with respect to ORR and CBR at the one-sided 2.5% level of significance using the same stratification information that was used for randomisation. The objective of the TAMRAD trial was to evaluate the efficacy and safety of everolimus in combination with tamoxifen in patients with metastatic breast cancer who were resistant to AIs. A power calculation determined that 53 evaluable patients were required in the everolimus in combination with tamoxifen group in order to detect a 20% gain in CBR over the tamoxifen monotherapy group with 90% power assuming a CBR of 50% for tamoxifen monotherapy. All randomly assigned patients who received at least one dose of study medication (ITT population) were assessed for efficacy and safety. CBR was analysed using Fisher s exact test, TTP and OS were analysed using the Kaplan Meier method, and the log-rank test was used to compare the survival curves for exploratory purposes. Cox proportional hazards regression models were used to estimate hazard ratios. Table B7 Analysis First PFS interim analysis Analyses for BOLERO-2 Follow-up, months Cut-off date Feb No. PFS events No. OS events 359 (68%) 83 (21%) PFS Update a 12 8-Jul (87%) 137 (35%) Second OS Oct- Not 182 (46%) b interim analysis 2011 performed Final PFS Dec analysis a 2011 Final OS analysis month BOLERO-2 CSR. 60 ; a not preplanned, requested by FDA; b data on file 35 Specification for manufacturer/sponsor submission of evidence Page 70 of 292

71 Table B8 Summary of statistical analyses in RCTs Trial no. (acronym) Hypothesis objective Statistical analysis The primary efficacy analysis was PFS based on local assessment and determined using a log-rank test stratified according to visceral metastases and previous hormone sensitivity Sample size, power calculation A total of 528 PFS events were required for the final analysis, in order to detect a hazard ratio of 0.74 with 90% power with the use of a log-rank test and a two-look Lan DeMets group sequential design with an O Brien Flemingtype boundary at a one-sided cumulative 2.5% level of significance Assuming a median PFS of 3.7 months in the everolimus alone group, 18 months of recruitment, a 10% rate of loss to followup, and a 2:1 randomisation ratio in favour of the everolimus in combination with exemestane group, 705 patients were to be randomly assigned Data management, patient withdrawals Patients who discontinued one or both study treatments for any reason other than progression were required to follow the same schedule of assessments until progression The full analysis set consisted of all randomised patients Following the ITT principle, patients were analysed according to the treatment and stratum they were assigned to at randomisation; data from the full analysis set were the primary basis for all efficacy analyses The safety set consisted of all patients who received at least one dose of the study treatment and who had at least one valid postbaseline safety assessment All randomly assigned patients who received at least one dose of study medication (ITT population) were assessed for efficacy and safety BOLERO-2 18,60 To evaluate the efficacy and safety of the combination of everolimus and exemestane in patients with hormonereceptorpositive advanced breast cancer refractory to nonsteroidal aromatase inhibitors TAMRAD 22 To evaluate the efficacy and safety of everolimus in combination with tamoxifen in patients with metastatic breast cancer who were resistant to aromatase inhibitors Although no formal statistical between-group comparison was planned, Fisher s exact test was used to compare between-group CBRs, and the log-rank test was used to compare the survival curves for exploratory purposes; Cox proportional hazards regression models were used to estimate hazard ratios A Simon two-stage minimax design was used; considering a 20% gain in the CBR as the minimum required to warrant further study of the combination, and assuming a CBR of 50% in the tamoxifen group, it was determined that 53 evaluable patients were needed in the tamoxifen in combination with everolimus group with 90% power and α = 5% CBR, clinical benefit rate; ITT, intention-to-treat; PFS, progression-free survival. Specification for manufacturer/sponsor submission of evidence Page 71 of 292

72 6.3.7 Provide details of any subgroup analyses that were undertaken and specify the rationale and whether they were pre-planned or posthoc. Subgroup analyses of the primary outcome measure of PFS in BOLERO-2 were prespecified. 18 A total of 17 subgroup analyses were performed based on patient and disease characteristics and prior therapies. The categories included: age (less than 65 years and 65 years or older); geographical region (Asia, Europe, North America, other); race (Asian, Caucasian, other; analysis also conducted for Japanese or non- Japanese); baseline ECOG status (0, 1 or 2); sensitivity to previous hormonal therapy (yes or no); visceral metastases (yes or no); number of organs involved (1, 2, 3 or more); measurable disease (yes or no); number of previous therapies (1, 2, 3 or more); prior use of hormonal therapy other than NSAI (yes or no); most recent therapy (AI, anti-oestrogen, other); purpose of most recent therapy (adjuvant therapy, treatment for advanced or metastatic disease); previous treatment with fulvestrant (yes or no); previous chemotherapy (yes or no, neoadjuvant or adjuvant therapy only, treatment for metastatic disease [with or without neoadjuvant or adjuvant therapy]); positive PgR status (yes or no); bone lesions at baseline (yes or no). Safety subgroup analyses based on age (less than 65 years and 65 years or older) and race (Japanese or non-japanese) were also conducted. Subgroup analyses in the TAMRAD trial were all exploratory. CBR was assessed based on visceral metastases (yes or no), previous adjuvant tamoxifen (yes or no), prior metastatic chemotherapy (yes or no) and hormone resistance (primary or secondary). 23 TTP and OS subgroup analyses were also performed based on primary or secondary hormone resistance. 23 Specification for manufacturer/sponsor submission of evidence Page 72 of 292

73 Participant flow Provide details of the numbers of patients who were eligible to enter the RCT(s), randomised, and allocated to each treatment. Provide details of, and the rationale for, patients who crossed over treatment groups and/or were lost to follow-up or withdrew from the RCT. This information should be presented as a CONSORT flow chart. The CONSORT flow chart for BOLERO-2 at a median follow-up of 7 months is shown in Figure B2. 18,80 The rates of discontinuation were lower in the everolimus in combination with exemestane group compared with the placebo in combination with exemestane group (53.2% versus 71.1%, respectively). The most frequent reason for discontinuation was disease progression in both groups, and the rate was lower in the everolimus in combination with exemestane group (37.3%) compared with the placebo in combination with exemestane group (65.7%). Discontinuation as a result of AEs was low in both groups; 6.6% in the everolimus in combination with exemestane group and 2.9% in the placebo in combination with exemestane group. The median duration of exposure was 14.6 weeks to everolimus and 12.0 weeks to placebo. The median duration of exposure to exemestane was 17.4 weeks in the everolimus in combination with exemestane group and 12.0 weeks in the placebo in combination with exemestane group. Specification for manufacturer/sponsor submission of evidence Page 73 of 292

74 Figure B2 CONSORT diagram for BOLERO-2 at the 7-month interim analysis Baselga et al., ,80 The disposition of the patients in BOLERO-2 at the 18-month analysis is shown in Table B9 and is similar to that reported in the 7-month analysis. 19 Discontinuation rates were higher at 18 months than at 7 months, and remained lower in the everolimus in combination with exemestane group compared with the placebo in combination with exemestane group (83.3% versus 95.8%, respectively). 19 The most frequent reason for discontinuation at 18 months was disease progression; this was lower in the everolimus in combination with exemestane group compared with the placebo in combination with exemestane group (61.9% versus 88.7%, respectively). A higher percentage of patients discontinued treatment in the everolimus in combination with exemestane arm compared with the placebo in combination with exemestane because of AEs (9.1% versus 3.3%, respectively) and withdrawal of consent (9.5% versus 2.9%, respectively). Specification for manufacturer/sponsor submission of evidence Page 74 of 292

75 Table B9 Patient disposition at the 18-month analysis of BOLERO-2 Disposition Everolimus in combination with exemestane (n = 485), % Protocol therapy ongoing Discontinued Disease progression Adverse event Consent withdrawal Death New cancer therapy Protocol deviation Treatment duration completed per protocol Entered post-treatment evaluation a Discontinued post-treatment evaluation b Patient withdrew consent Lost to follow-up Placebo in combination with exemestane (n = 239), % Piccart et al., a The percentages of patients who entered post-treatment evaluation use the number discontinued from treatment as the denominator. All subsequent percentages use the number who entered post-treatment evaluations as the denominator. b Per Study Evaluation Completion page. The CONSORT flow chart for the TAMRAD trial is shown in Figure B3. 22 At a median follow-up of 24 months, the majority of patients had discontinued treatment. The discontinuation rates were lower in the everolimus in combination with tamoxifen group (87.0%) compared with the tamoxifen alone group (98.2%). Disease progression was the most common reason for discontinuation and this occurred more frequently in the tamoxifen alone group than in the everolimus in combination with tamoxifen group (86.0% versus 59.3%, respectively). The rate of discontinuation due to AEs was higher in the everolimus in combination with tamoxifen group (22.2%) compared with the tamoxifen alone group (7.0%). The median treatment duration was 6.2 months (range, 0.7 to 31 months) for the everolimus in combination with tamoxifen group and 4.8 months (range, 0.7 to 27 months) for the tamoxifen alone group. Specification for manufacturer/sponsor submission of evidence Page 75 of 292

76 Figure B3 CONSORT diagram for the TAMRAD trial at 24 months Bachelot et al., Critical appraisal of relevant RCTs The validity of the results of an individual study will depend on the robustness of its overall design and execution, and its relevance to the decision problem. Each study that meets the criteria for inclusion should therefore be critically appraised. Whenever possible, the criteria for assessing published studies should be used to assess the validity of unpublished and part-published studies. The critical appraisal will be validated by the ERG. The following are the minimum criteria for assessment of risk of bias in RCTs, but the list is not exhaustive. Was the method used to generate random allocations adequate? Was the allocation adequately concealed? Were the groups similar at the outset of the study in terms of prognostic factors, for example, severity of disease? Were the care providers, participants and outcome assessors blind to treatment allocation? If any of these people were not Specification for manufacturer/sponsor submission of evidence Page 76 of 292

77 blinded, what might be the likely impact on the risk of bias (for each outcome)? Were there any unexpected imbalances in drop-outs between groups? If so, were they explained or adjusted for? Is there any evidence to suggest that the authors measured more outcomes than they reported? Did the analysis include an intention-to-treat analysis? If so, was this appropriate and were appropriate methods used to account for missing data? Please provide as an appendix a complete quality assessment for each RCT. See section 10.3, appendix 3 for a suggested format If there is more than one RCT, tabulate a summary of the responses applied to each of the critical appraisal criteria. A suggested format for the quality assessment results is shown below. Quality assessment of the two RCTs is described in Table B10. The critical appraisal of BOLERO-2 was based on the clinical study report and update. 60,88 Details of the trial methods fully address random sequence generation for participant allocation, concealment of allocation, and blinding of the care providers, participants and outcome assessors. The patient groups were well balanced in major baseline characteristics. More patients discontinued in the exemestane in combination with placebo group compared with the exemestane in combination with everolimus group, as might be expected given that the main reason for discontinuation was disease progression. Patients were analysed according to ITT and all patients were accounted for in the primary analysis and those secondary analyses using Kaplan Meier methods. Adequate detail was included in the clinical study report and interim analyses to suggest there was no selective reporting. The critical appraisal of the TAMRAD trial was based on a published paper. 22,81 This did not report sufficient details of the trial methodology to Specification for manufacturer/sponsor submission of evidence Page 77 of 292

78 fully address the quality criteria evaluated; in particular, it was unclear how randomisation was carried out. The trial was described as open label and thus was not blinded by design. The patient groups were generally well balanced in terms of baseline characteristics except for performance status where the authors acknowledged that an ECOG performance status of zero was more common in the tamoxifen in combination with everolimus group (59% versus 40%) and an ECOG status of one was more common in the tamoxifen alone group (49% versus 33%). There were a higher percentage of drop-outs in the combined treatment group than the tamoxifen alone group due to AEs (22% versus 7%). The trial analyses were based on the principle of ITT; per protocol results were also reported. Adequate detail was included in the final publication to suggest there was no selective reporting. Specification for manufacturer/sponsor submission of evidence Page 78 of 292

79 Table B10 Quality assessment results for RCTs Trial no. (acronym) BOLERO-2 18,60,80,88 TAMRAD 22,81 Was randomisation carried out appropriately? Was the concealment of treatment allocation adequate? Were the groups similar at the outset of the study in terms of prognostic factors? Were the care providers, participants and outcome assessors blind to treatment allocation? Were there any unexpected imbalances in dropouts between groups? Is there any evidence to suggest that the authors measured more outcomes than they reported? Did the analysis include an intentionto-treat analysis? If so, was this appropriate and were appropriate methods used to account for missing data? Yes, patients were assigned to each treatment arm by centralised allocation Yes, randomisation data were kept strictly confidential until the time of unblinding and were not accessible by anyone involved in the study Yes, baseline characteristics were similar between groups Yes, patients, investigator staff, persons performing the assessments, all Novartis personnel and individuals at central laboratories (including central imaging) were to remain blinded to the identity of the treatment from the time of randomisation until database lock No No, all outcomes were reported in results Yes. The full analysis set (FAS) consisted of all randomised patients. Following the intentionto-treat (ITT) principle, patients were analysed according to the treatment and stratum they were assigned to at randomisation; data from the FAS were the primary basis for all efficacy analyses Not clear, no specific description of how random allocation was generated included in the manuscript N/A: study was open label Baseline characteristics were similar between groups except for ECOG performance status No; study was open label No No, based on primary manuscript, there is no evidence of outcomes in methods that were not reported in results Yes, all patients were followed up and included in ITT analysis Bachelot et al., ,81 ; Baselga et al., ,80 ; 7-month BOLERO-2 CSR 60 ; 12-month BOLERO-2 CSR 88 Adapted from Centre for Reviews and Dissemination (2008) Systematic reviews. CRD s guidance for undertaking reviews in health care. York: Centre for Reviews and Dissemination. Specification for manufacturer/sponsor submission of evidence Page 79 of 292

80 6.5 Results of the relevant RCTs Provide the results for all relevant outcome measure(s) pertinent to the decision problem. Data from intention-to-treat analyses should be presented whenever possible and a definition of the included patients provided. If patients have been excluded from the analysis, the rationale for this should be given. If there is more than one RCT, tabulate the responses The information may be presented graphically to supplement text and tabulated data. If appropriate, please present graphs such as Kaplan Meier plots For each outcome for each included RCT, the following information should be provided. The unit of measurement. The size of the effect; for dichotomous outcomes, the results ideally should be expressed as both relative risks (or odds ratios) and risk (or rate) differences. For time-to-event analysis, the hazard ratio is an equivalent statistic. Both absolute and relative data should be presented. A 95% confidence interval. Number of participants in each group included in each analysis and whether the analysis was by intention to treat. State the results in absolute numbers when feasible. When interim RCT data are quoted, this should be clearly stated, along with the point at which data were taken and the time remaining until completion of that RCT. Analytical adjustments should be described to cater for the interim nature of the data. Other relevant data that may assist in interpretation of the results may be included, such as adherence to medication and/or study protocol. Discuss and justify definitions of any clinically important differences. Specification for manufacturer/sponsor submission of evidence Page 80 of 292

81 Report any other analyses performed, including subgroup analysis and adjusted analyses, indicating those pre-specified and those exploratory. Overview of BOLERO-2 results BOLERO-2 demonstrated that everolimus in combination with exemestane substantially reduced the risk of disease progression in postmenopausal women with HR+, HER2 advanced breast cancer previously treated with a NSAI compared with exemestane alone Improvements in PFS were observed across all pre-planned subgroup analyses.. The addition of everolimus to exemestane also improved secondary outcome measures including OS, ORR and CBR. An assessment of markers of bone resorption, which is known to increase in patients treated with AIs such as exemestane, 89 suggested that addition of everolimus was able to reverse the increase in bone resorption and suppress bone turnover. Three analyses have been performed to date at a median of 7, 12 and 18 months of follow-up, corresponding to 359, 457, and 510 PFS events, respectively The data from these analyses are discussed below and key data from the most recent analysis are summarised in Table B11. Table B11 Summary of the BOLERO-2 18-month efficacy data Outcome PFS median, months Central assessment Local assessment Everolimus in combination with exemestane Placebo in combination with exemestane p-value < < OS (deaths), % a ORR, % < CBR, % < a OS analysis was conducted at 16 months. 35 Piccart et al., CBR, clinical benefit rate; PFS, progression-free survival; ORR, overall response rate; OS, overall survival. Specification for manufacturer/sponsor submission of evidence Page 81 of 292

82 Progression-free survival (PFS) Everolimus significantly increased PFS across all analyses Everolimus in combination with exemestane more than doubled median PFS compared with placebo in combination with exemestane at all analysis time points (Table B12) Across the three analyses, the addition of everolimus to exemestane reduced the risk of disease progression by 55% to 57% based on local assessment. Central assessment was generally in good agreement with local evaluation, and indicated a reduction in the risk of disease progression of 62% to 64%. At 18 months, this translated to an improvement in median PFS of 6.9 months (central assessment) and 4.6 months (local assessment).. 19 Table B12 PFS based on local and central assessment at 7-, 12- and Reference Baselga et al., NEJM Hortobagy i et al., SABCS Piccart et al., ASCO month analyses in BOLERO-2 Followup (months) Assessor PFS, median (months) Hazard ratio (95% CI) Everolimus in combination with exemestane Placebo in combination with exemestane 7 Central (0.27 to 0.47) Local (0.35 to 0.54) 12 Central (0.28 to 0.45) Local (0.36 to Central (0.31 to 0.48) Local (0.38 to 0.54) CI, confidence interval; PFS, progression-free survival. p-value < < < < < < The Kaplan Meier plots for PFS at the 7- and 18-month analyses are shown in Figure B4. Specification for manufacturer/sponsor submission of evidence Page 82 of 292

83 Figure B4 PFS at 7 months of follow-up based on local (a) and central (b) assessment and at 18 months based on local (c) and central (d) assessment in BOLERO-2 Baselga et al., ; Piccart et al., Specification for manufacturer/sponsor submission of evidence Page 83 of 292

84 The PFS benefits of everolimus were consistent across all subgroups Everolimus in combination with exemestane benefited all patients regardless of demographic characteristics, endocrine therapy sensitivity, and extent of disease or prior therapy, and no differences were seen between the 7-, 12- and 18-month analyses with regard to efficacy in any of the subgroups The pre-specified PFS subgroup analysis at 7 months (which is the most comprehensive presentation of the pre-planned subgroup analyses) is shown in Figure B5. Further PFS subgroup analyses were presented for the 18 month analysis by Piccart et al. 19 which were consistent with the 7 month analysis. No statistically significant differernce was observed between any of the pre-sepecified subgroups and when compared to the overall PFS analysis. The 3 PFS analysis are all consistent consistent with each other with regards to this Specification for manufacturer/sponsor submission of evidence Page 84 of 292

85 Figure B5 PFS across various subgroups at the 7-month interim analysis in BOLERO-2 Baselga et al., Specification for manufacturer/sponsor submission of evidence Page 85 of 292

86 Overall survival Everolimus in combination with exemestane improved OS compared with placebo in combination with exemestane (Table B13) The protocol specified up to three OS analyses; at the time of the interim PFS analysis (corresponding to the 7-month analysis), after 173 deaths (corresponding to the 16-month analysis) and after 392 deaths (not yet reached); a further analysis was carried out at the time of the 12 month PFS update (see Table B7). To date, OS has been assessed at three time points: after 7 months (83 deaths), 12 months (137 deaths) and 16 months (182 deaths). At each time point OS was numerically greater for everolimus in combination with exemestane versus placebo in combination with exemestane, (Table B13). At 16 months, 112 patients (23%) in the everolimus in combination with exemestane group had died, compared with 70 patients (29%) in the placebo in combination with exemestane group (p = 0.046; Figure B6). 35 Table B13 Overall survival at 7-, 12- and 18-month analyses in Reference BOLERO-2 Followup, months Number of deaths Total Everolimus in combination with exemestane (n = 485), n (%) Placebo in combination with exemestane (n = 239), n (%) Baselga et al., (10.7) 31 (13.0) 2.3 NEJM a 18 Hortobagyi et al., SABCS a (17.2) 54 (22.7) 5.5 Pre-planned analysis; data on file. 35 Difference, % (23) 70 (29) 6; p = b a n-values for number of deaths in each group calculated from percentages; b One-sided P- value obtained using log-rank test stratified by sensitivity to prior hormonal therapy and presence of visceral metastasis. Figure B6 Overall survival at 16 months in BOLERO-2 Specification for manufacturer/sponsor submission of evidence Page 86 of 292

87 Data on file. 35 Clinical benefit rate and overall response rate Everolimus in combination with exemestane significantly increases CBR and ORR compared with exemestane alone CBR and ORR were substantially higher in patients treated with everolimus in combination with exemestane compared with placebo in combination with exemestane (Table B14). At each assessment the CBR achieved with everolimus in combination with exemestane was approximately double that achieved with placebo in combination with exemestane; at the 18-month follow-up the CBRs were 51.3% and 26.4%, respectively (p < ) Similarly the ORR achieved with everolimus in combination with exemestane was significantly greater than for placebo in combination with exemestane at all time points; at 18 months, the ORRs were 12.6% and 1.7%, respectively (p < ) Specification for manufacturer/sponsor submission of evidence Page 87 of 292

88 Table B14 CBR and ORR at 7-, 12- and 18-month analyses (local Reference Baselga et al., NEJM ,60 Hortobagyi et al., SABCS c,17 Piccart et al., ASCO assessment) in BOLERO-2 ORR, n (%) a CBR, n (%) b Followup, Everolimus Placebo in Everolimus Placebo in in combinatio in combinatio months combinatio n with combinatio n with n with exemestane n with exemestane exemestane (n = 239) exemestane (n = 239) (n = 485) (n = 485) 7 46 (9.5)* 1 (0.4) 162 (33.4)** 43 (18.0) (12.0)** 3 (1.3) 245 (50.5)** 61 (25.5) (12.6)** 4 (1.7) 249 (51.3)** 63 (26.4) a ORR defined as patients with either CR or PR; b CBR defined as CR, PR or SD at 24 weeks; c n-values calculated from percentages *p < 0.001; **p < CBR, clinical benefit rate; CR, complete response; ORR, overall response rate; PR, partial response; SD, stable disease. Time to response and duration of response At the 7-month interim analysis, time to response ranged from 5.1 to 37.1 weeks for the exemestane in combination with everolimus group compared with 7.4 weeks for the single patient with a response in the placebo in combination with exemestane group. Duration of overall response ranged from 6.0 to 66.1 weeks for the everolimus in combination with exemestane group compared with 12.1 weeks for the single patient in the placebo in combination with exemestane group (responses were ongoing at the time of the data cut-off). 18,60 Bone marker analysis Everolimus in combination with exemestane reduces bone resorption and fractures compared with exemestane alone Bone resorption is known to increase during treatment with AIs such as exemestane and can increase the risk of fractures. 89 Analysis of bone Specification for manufacturer/sponsor submission of evidence Page 88 of 292

89 markers showed that the addition of everolimus to exemestane suppressed bone turnover and reversed the increase in bone resorption associated with exemestane. 20,85,89 Levels of the bone turnover markers BSAP, P1NP and CTX all decreased from baseline in patients treated with everolimus in combination with exemestane (at both 6 and 12 weeks), whereas these levels all increased from baseline in patients treated with placebo in combination with exemestane (Figure B7); these trends remained true regardless of the presence of bone metastasis or use of bisphosphonate therapy at baseline. 20,85 Figure B7 Change in bone turnover markers at 6- and 12-weeks in BOLERO-2 Gnant et al BSAP, bone-specific alkaline phosphatase; CTX, c-terminal cross-linking telopeptide of type I collagen; P1NP, amino-terminal propeptide of type 1 collagen. Everolimus in combination with exemestane also reduced the rate of fractures compared with placebo in combination with exemestane. 20 In the 18-month analysis, 2.3% of the patients receiving everolimus in combination with exemestane had fractures compared with 3.8% of patients receiving placebo plus exemestane. Furthermore, everolimus in combination with in combination with exemestane decreased the incidence of disease progression in bone compared with placebo in combination with exemestane. 20 At 18 months, the reduction in disease progression in bone with everolimus in combination with exemestane compared with placebo in Specification for manufacturer/sponsor submission of evidence Page 89 of 292

90 combination with exemestane was statistically significant both in the overall population (p = )) and in the subgroup of patients with bone metastases at baseline (p = ) (Figure B8 and Figure B9). Figure B8 Incidence of disease progression in bone in (a) the overall population and (b) the subgroup of patients with bone metastases at baseline in BOLERO-2 Gnant et al. ASCO Figure B9 PFS in the subgroup of patients with bone-only metastases at the 18- month analysis in BOLERO-2 Campone et al. ESMO Health-related quality of life Treatment with everolimus in combination with exemestane did not compromise patient quality of life compared with placebo in combination with exemestane at 7-, 12- or 18-month follow-ups (Table B15). 17,21,77 Specification for manufacturer/sponsor submission of evidence Page 90 of 292

91 Table B15 HRQoL in BOLERO-2, measured by EORTC QLQ-C30, at 7-, 12- and 18-month analyses Reference Baselga et al., ESMO Hortobagyi et al., SABCS Beck et al., ASCO Followup, months EORTC QLQ-C30 time to deterioration of global health status/qol domain score ( 5%), months Everolimus in combination with exemestane Exemestane in combinatio n with placebo Hazard ratio (95% CI) (0.68 to 1.20) (0.62 to 1.06) (0.58 to 0.95) p-value EORTC QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire; HRQoL, health-related quality of life. Everolimus in combination with exemestane extends time to deterioration in HRQoL compared with exemestane alone At the 7-month interim analysis, the median time to definitive deterioration of the global health status/hrqol domain score of EORTC QLQ-C30, analysed using the generally established and accepted definition for minimal important difference (MID; 5%), was numerically greater for exemestane in combination with everolimus than with placebo in combination with exemestane (Table B15). 77 The difference between groups in median time to definitive deterioration of global health status increased with further follow-up and the difference between treatment groups was statistically significant at 12 and 18 months. 17,21 At the 18- month analysis, the median time to definitive deterioration ( 5%) of the global health status/hrqol domain score of EORTC QLQ-C30 was 8.3 months (95% CI, 7.0 to 0.97) for everolimus in combination with exemestane versus 5.8 months (95% CI, 4.2 to 7.2) for placebo in combination with exemestane (p = ) (Table B15 and Figure B10). 21 Specification for manufacturer/sponsor submission of evidence Page 91 of 292

92 Figure B10 Time to definitive deterioration (MID 5%) in EORTC QLQ- C30 Global Health Status score in BOLERO-2 Beck et al., EORTC QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire; MID, minimal important difference; QoL, quality of life. Using a more conservative definition of MID ( 10%), median time to deterioration in global health status/hrqol domain score was numerically greater for the everolimus in combination with exemestane group (11.7 months [95% CI, 9.7 to 13.3]) than with the placebo in combination with exemestane group (8.4 months [95% CI, 6.6 to 12.5]) at 18 months, but did not reach statistical significance (p = 0.102). 21 These observations suggest that any potential HRQoL deficits resulting from AEs experienced by patients treated with everolimus are at least balanced by the HRQoL benefits gained by reducing disease progression. Specification for manufacturer/sponsor submission of evidence Page 92 of 292

93 Overview of the TAMRAD trial results Data reported for a median follow-up of 24 months indicate that the addition of everolimus to tamoxifen significantly improved CBR, TTP and OS compared with tamoxifen alone in postmenopausal women with HR+, HER2, AI-resistant metastatic breast cancer. 22 Clinical benefit rate and overall response rate Everolimus in combination with tamoxifen provided a significant increase in CBR compared with tamoxifen monotherapy CBR at 6 months in the ITT population (primary endpoint) was significantly higher for patients treated with everolimus in combination with tamoxifen (61%; 95% CI, 47% to 74%) compared with tamoxifen alone (42%; 95% CI, 29% to 56%, p = 0.045, Table B16). 22 Similar results were reported for the per protocol population. The clinical benefit of adding everolimus to tamoxifen was observed in all subgroups analysed, including patients with and without visceral metastases (Table B16). Particular benefit was observed in patients with secondary hormone resistance. For this population, the CBR in patients treated with everolimus in combination with tamoxifen was 74%, compared with 48% in those treated with tamoxifen alone. 22 Specification for manufacturer/sponsor submission of evidence Page 93 of 292

94 Table B16 Clinical benefit rate in full population and subgroups in the TAMRAD trial N Clinical benefit rate, n/n (%) p-value Everolimus in Tamoxifen combination with tamoxifen Total population /54 (61.1) 24/57 (42.1) Visceral metastases 59 19/31 (61.3) 11/28 (39.3) NR No visceral metastases 52 14/23 (60.9) 13/29 (44.8) NR Prior adjuvant tamoxifen 42 12/18 (66.7) 9/24 (37.5) NR No prior adjuvant tamoxifen 69 21/36 (58.3) 15/33 (45.3) NR Prior metastatic 28 6/13 (46.2) 4/15 (26.7) NR chemotherapy No prior metastatic 83 27/41 (65.9) 20/42 (47.6) NR chemotherapy Primary hormone 54 12/26 (46.2) 10/28 (35.7) NR resistance Secondary hormone resistance 56 20/27 (74.1) 14/29 (48.3) NR NR, not reported Bachelot et al., ; Bourgier et al., ORRs were similar for both treatment groups. 22 In the ITT population, 5 patients (9%) achieved a complete or partial response in the everolimus in combination with tamoxifen group, as did 5 patients (9%) in the tamoxifen only group. In patients with measurable lesions, the ORRs were similar in the everolimus in combination with tamoxifen group (14%) compared with the tamoxifen only group (13%). Time to progression Everolimus in combination with tamoxifen significantly increased TTP compared with tamoxifen monotherapy The addition of everolimus to tamoxifen significantly increased TTP from 4.5 months (95% CI, 3.6 to 8.7) to 8.6 months (95% CI, 5.9 to 13.9), corresponding to a 46% reduction in risk of progression (hazard ratio, 0.54; 95% CI, 0.36 to 0.81; p = ) (Table B17; Figure B11). 22 Specification for manufacturer/sponsor submission of evidence Page 94 of 292

95 Figure B11 Time to progression in the intention-to-treat population in the TAMRAD trial Bachelot et al., The improvement in TTP with everolimus in combination with tamoxifen was particularly marked in patients with secondary hormone resistance. The TTP for this subgroup increased from 5.5 to 14.8 months, corresponding to a reduction in risk of 54% (hazard ratio, 0.46; 95% CI, 0.26 to 0.83; p = ) (Table B17). This was greater than the 30% risk reduction observed in patents with primary hormone resistance (hazard ratio, 0.70; 95% CI, 0.40 to 1.21; not significant). 22 Specification for manufacturer/sponsor submission of evidence Page 95 of 292

96 Table B17 Time to progression at median 24-month follow-up in the Population TAMRAD trial Median time to progression, months Everolimus in combination with tamoxifen Tamoxifen Hazard ratio (95% CI) p-value All patients (n = (0.36 to 0.81) ) a Primary hormone (0.40 to 1.21) NS resistance (n = 54) Secondary hormone resistance (n = 56) (0.26 to 0.83) Bachelot et al., a Hormone resistance status of one patient was missing. CI, confidence interval; OS, overall survival; NS, not significant; TTP, time to progression. Overall survival Everolimus in combination with tamoxifen significantly prolonged overall survival compared with tamoxifen monotherapy Everolimus in combination with tamoxifen also significantly improved OS compared with tamoxifen alone. 22,23 In total, 47 patients had died at the time of the updated survival analysis (September 2011); 16 (30%) in the everolimus in combination with tamoxifen group and 31 (54%) in the tamoxifen group (Table B18). The median OS was not reached in the everolimus in combination with tamoxifen group and was 32.9 months in the tamoxifen alone group. 22 Everolimus in combination with tamoxifen was associated with a 55% reduction in the risk of death compared with tamoxifen alone (hazard ratio, 0.45; 95% CI, 0.24 to 0.81, p = 0.007) (Table B18; Figure B12). The improvement in OS achieved with everolimus in combination with tamoxifen was particularly marked in patients with secondary hormone resistance. In this subgroup everolimus in combination with tamoxifen was associated with a reduction in risk of death of 79% compared with tamoxifen alone (p = 0.002) (Table B18). 23 Specification for manufacturer/sponsor submission of evidence Page 96 of 292

97 Table B18 Overall survival at median follow-up of 24-month in the TAMRAD trial Population All patients (n = 111) Primary hormone resistance (n = 54) Secondary hormone resistance (n = 56) Overall survival number of deaths, n (%) Everolimus in Tamoxifen combination with tamoxifen Bachelot et al., ; Bourgier et al., HR (95% CI) 16/54 (29.6) 31/57 (54.4) 0.45 (0.24 to 0.81) 12/26 (46.2) 15/28 (53.6) 0.73 (0.34 to1.55) 4/27 (14.8) 16/29 (55.2) 0.21 (0.07 to 0.63) p-value Figure B12 Overall survival in the intention-to-treat population in the TAMRAD trial Bachelot et al., Specification for manufacturer/sponsor submission of evidence Page 97 of 292

98 6.6 Meta-analysis When more than one study is available and the methodology is comparable, a meta-analysis should be undertaken. This section should be read in conjunction with NICE s Guide to the methods of technology appraisal, sections to The following steps should be used as a minimum when presenting a meta-analysis. Perform a statistical assessment of heterogeneity. If the visual presentation and/or the statistical test indicate that the RCT results are heterogeneous, try to provide an explanation for the heterogeneity. Statistically combine (pool) the results for both relative risk reduction and absolute risk reduction using both the fixed effects and random effects models (giving four combinations in all). Provide an adequate description of the methods of statistical combination and justify their choice. Undertake sensitivity analysis when appropriate. Tabulate and/or graphically display the individual and combined results (such as through the use of forest plots). No meta-analysis was conducted If a meta-analysis is not considered appropriate, a rationale should be given and a qualitative overview provided. The overview should summarise the overall results of the individual studies with reference to their critical appraisal. The systematic review identified only two relevant head-to-head trials. One trial, BOLERO-2, 18 compared everolimus in combination with exemestane with exemestane alone, while the other trial, TAMRAD, 22 compared everolimus in combination with tamoxifen with tamoxifen alone. Neither of Specification for manufacturer/sponsor submission of evidence Page 98 of 292

99 these two trials contained a common comparator. Thus performance of a meta-analysis was not considered appropriate If any of the relevant RCTs listed in response to section (Complete list of relevant RCTs) are excluded from the metaanalysis, the reasons for doing so should be explained. The impact that each exclusion has on the overall meta-analysis should be explored. No meta-analysis was conducted. Specification for manufacturer/sponsor submission of evidence Page 99 of 292

100 6.7 Indirect and mixed treatment comparisons Data from head to head RCTs should be presented in the reference-case analysis, if available. If data from head to head RCTs are not available, indirect treatment comparison methods should be used. This section should be read in conjunction with NICE s Guide to the methods of technology appraisal, sections to Describe the strategies used to retrieve relevant clinical data on the comparators and common references both from the published literature and from unpublished data. The methods used should be justified with reference to the decision problem. Sufficient detail should be provided to enable the methods to be reproduced, and the rationale for any inclusion and exclusion criteria used should be provided. Exact details of the search strategy used should be provided in section 10.4, appendix 4. Fulvestrant is one of the comparators included in the model. No head-tohead studies of fulvestrant compared with everolimus in combination with exemestane were indentified in the systematic review. Further literature searches were therefore undertaken to identify relevant studies that would allow indirect comparisons of the efficacy of everolimus in combination with exemestane versus fulvestrant to be performed. A series of literature searches were performed to identify systematic reviews and trials of interventions in advanced metastatic breast cancer. The databases searched included Medline, EMBASE, the Cochrane Library, clinicaltrials.gov and the International Clinical Trials Registry Platform (ICTRP). Additional searches were conducted to identify systematic reviews and health technology assessments for interventions in advanced or metastatic breast cancer in the Cochrane Database of Systematic Reviews (CDSR), Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment (HTA) database. Searches were also performed on the National Horizon Scanning Centre and NICE websites (see Appendix 10.4 for details.) Specification for manufacturer/sponsor submission of evidence Page 100 of 292

101 Reviews were assessed for potential RCTs that provided links between fulvestrant and everolimus in advanced or metastatic breast cancer. All publications meeting the inclusion criteria based on titles and abstract were obtained as full documents and reassessed against the inclusion criteria. In total, four trials were identified, which formed a network permitting indirect comparisons; an overview of these trials is provided in Table B19. Table B19 Details of trials included in the mixed treatment comparison Trial no. (acronym) BOLERO-2 Intervention Comparator Population Primary study ref. Everolimus (10 mg/day) in combination with exemestane (25 mg/day) Placebo in combination with exemestane (25 mg/day) Postmenopausal women with HR+, HER2 advanced breast cancer whose disease was refractory to prior letrozole or anastrozole Baselga et al., NEJM 2012; 366: CONFIRM Fulvestrant (500 mg/month) Fulvestrant (250 mg/month) Postmenopausal women with ER+ locally advanced or metastatic breast cancer DiLeo et al., JCO 2010; 28: EFECT Exemestane (25 mg/day) Fulvestrant (250 mg/month) Postmenopausal women with HR+ locally advanced or metastatic breast cancer Chia et al., JCO 2008; 26: SoFEA Fulvestrant (250 mg/month) in combination with anastrozole (1 mg/day) or Fulvestrant (250 mg/month) Exemestane (25 mg/day) Postmenopausal women with HR+ locally advance or metastatic breast cancer Johnston EBC Conference Please follow the instructions specified in sections 6.1 to 6.5 for the identification, selection and methodology of the trials, quality assessment and the presentation of results. Provide in section 10.5, Specification for manufacturer/sponsor submission of evidence Page 101 of 292

102 appendix 5, a complete quality assessment for each comparator RCT identified. Methodology of relevant RCTs The comparative methodology for the four identified RCTs is summarised in Table B20. Study design The included studies were randomised, double-blind, multicentre, phase III trials. Patients The studies involved between 693 and 736 patients. All patients were postmenopausal with advanced, locally advanced or metastatic breast cancer and all had HR+ tumours. The BOLERO-2 18 study was conducted specifically in HER2 patients; over half of the patients in the SoFEA trial 26 were HER2 patients; and the comparison of fulvestrant 250 mg and 500 mg in postmenopausal women with oestrogen receptor positive advanced breast cancer progressing or relapsing after previous endocrine therapy (CONFIRM) 25 and EFECT 24 studies did not report the proportions of HER2+ and HER2 patients. In all studies patients had received prior therapy and in three studies, patients were required to have received prior treatment with NSAIs. Interventions and comparators The BOLERO-2 study compared everolimus in combination with exemestane with exemestane monotherapy. The EFECT 24 and SoFEA 26 trials compared exemestane with fulvestrant (and fulvestrant in combination with anastrozole), and the CONFIRM study 25 compared two separate doses of fulvestrant. Specification for manufacturer/sponsor submission of evidence Page 102 of 292

103 Outcomes All studies reported disease progression, measured either as PFS or TTP, as their primary outcomes except for TAMRAD. Secondary outcomes typically included OS, CBR, ORR, quality of life and safety assessments. Specification for manufacturer/sponsor submission of evidence Page 103 of 292

104 Table B20 Details of methodology for trials selected for the mixed treatment comparison Trial no. (acronym) Design BOLERO-2 18,80 CONFIRM 25 EFECT 24 SoFEA 26 Randomised, doubleblind, placebocontrolled, multicentre, phase III trial Randomised, doubleblind, placebocontrolled, parallel group, multicentre, phase III trial Randomised, doubleblind, placebocontrolled, multicentre, phase III trial Patients N % HER2 93 to 98 NR NR 50.2 to 61 % HR Prior treatment with aromatase inhibitor, % Adjuvant Metastatic Prior chemotherapy, % Adjuvant Metastatic Randomised, partially double-blind, placebocontrolled, multicentre, phase III trial Specification for manufacturer/sponsor submission of evidence Page 104 of 292

105 Trial no. (acronym) Intervention(s) and comparator(s) BOLERO-2 18,80 CONFIRM 25 EFECT 24 SoFEA 26 Everolimus (10 mg/day) in combination with exemestane (25 mg/day) (n = 485) Fulvestrant (500 mg/month a ) (n = 362) Exemestane (25 mg/day) in combination with matching placebo (n = 342) Fulvestrant (250 mg/month a ) in combination with anastrozole (1 mg/day) (n = 243) Exemestane (25 mg/day) in combination with matched placebo (10 mg/day) (n = 239) Fulvestrant (250 mg/month a ) in combination with placebo (monthly injection) (n = 374) Fulvestrant (250 mg/month c ) in combination with matching placebo (n = 351) Fulvestrant (250 mg/month a ) in combination with placebo (once daily) (n = 231) Exemestane 25 mg/day (n = 249) Outcomes Primary PFS PFS TTP PFS Secondary OS ORR ORR ORR ORR CBR CBR CBR CBR Safety Quality of life Duration of response and clinical benefit OS Safety Quality of life Duration of response Time to response OS Safety Quality of life OS Safety a Given from month two onwards. Specification for manufacturer/sponsor submission of evidence Page 105 of 292

106 Critical appraisal of included RCTs Quality assessment of the four RCTs selected for the mixed treatment comparison is described in Table B21 with a more detailed assessment in Appendix A critical appraisal of BOLERO-2 is described in section 6.4. In most studies, baseline characteristics were well balanced between the treatment groups, although many studies did not report sufficient data to adequately assess randomisation, concealment of treatment allocation and blinding. All studies used an intent-to-treat analysis, and there was little evidence of outcome reporting bias. Table B21 Quality assessment of studies included in the mixed treatment comparison Quality assessment element Was randomisation carried out appropriately? Was the concealment of treatment allocation adequate? Were the groups similar at the onset of the study in terms of prognostic factors? Were the care providers, participants and outcome assessors blind to treatment allocation? Were there any unexpected imbalances in dropouts between groups? Is there any evidence to suggest that authors measured more outcomes than they reported? Did the analysis include an ITT analysis? If so, was this appropriate and were appropriate methods used to account for missing data? BOLERO- CONFIRM 2 18,80 25 Low risk of bias Low risk of bias Low risk of bias Low risk of bias Unclear risk of bias Low risk of bias Low risk of bias Unclear risk of bias Unclear risk of bias Low risk of bias Unclear risk of bias Low risk of bias Low risk of bias Low risk of bias EFECT 24 SoFEA 26 Unclear risk of bias Unclear risk of bias Low risk of bias Unclear risk of bias Unclear risk of bias Low risk of bias Low risk of bias Unclear risk of bias Unclear risk of bias Low risk of bias Unclear risk of bias Unclear risk of bias Unclear risk of bias Low risk of bias Specification for manufacturer/sponsor submission of evidence Page 106 of 292

107 Results of selected RCTs The results of the four RCTs selected for the mixed treatment comparison are summarized in Table B22. Progression-free survival Everolimus in combination with exemestane led to a significant improvement in PFS compared with exemestane alone in BOLERO-2. 18,80 By contrast, the EFECT 24 and SoFEA 26 trials showed no significant difference in TTP and PFS, respectively, between exemestane and fulvestrant. Overall survival Among the studies that reported OS, most reported no significant difference in deaths between treatment groups, although some did not test this endpoint statistically. Clinical benefit rate and overall response rate CBR and ORR were substantially higher in patients treated with everolimus in combination with exemestane compared with exemestane alone in BOLERO-2. 18,80 The other studies either reported no significant differences in ORR and CBR between study groups, or did not test for significance of the difference between these groups. Specification for manufacturer/sponsor submission of evidence Page 107 of 292

108 Table B22 Summary of results from the trials included in the indirect treatment comparison Trial BOLERO-2 19 Everolimus in combination with exemestane Placebo in combination with exemestane CONFIRM 25 Fulvestrant 500 mg Fulvestrant 250 mg EFECT 24 Followup, months 18 months Median PFS or TTP, months Central: 11.0*** Local: 7.8*** Central: 4.1 Local: 3.2 OS deaths, n (%) [median] NR 6.5* [25.1 months] 5.5 [22.8 months] 112 (23)* 61 (12.6)*** ORR, n CBR, n (%) a (%) b 249 (51.3)*** 70 (29)* 4 (1.7) 63 (26.4) 33 (9.1) 165 (45.6) 38 (10.2) 148 (39.6) Exemestane c 18 (6.7) 108 (31.5) months Fulvestrant 3.7 c 20 (7.4) 113 (32.2) SoFEA 26 Fulvestrant in NR 4.4 combination with anastrozole Fulvestrant 4.8 Exemestane 3.4 a ORR defined as patients with either CR or PR; b CBR defined as CR, PR or SD at 24 weeks; c Time to progression rather than PFS.; * - OS results presented are based on the 16 month analysis. *p < 0.05; **p < 0.001; ***p < NR, not reported Specification for manufacturer/sponsor submission of evidence Page 108 of 292

109 Review of chemotherapy studies A rapid search of the Cochrane Library was undertaken to identify Cochrane reviews and other recent systematic reviews in the Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment (HTA) databases to retrieve well conducted systematic reviews with data on chemotherapy treatment for metastatic or advanced breast cancer. The search strategy was designed to be sensitive in order to identify all systematic reviews and health technology assessments about advanced or metastatic breast cancer. The results were then sifted to remove those reviews that were not about drug interventions for advanced or metastatic breast cancer: surgery, radiotherapy, non-drug treatments, screening, prevention, etc. The search was also limited in DARE and HTA to retrieve only those reviews published in the last few years ( ). The search showed that the reviews shown in contained the most relevant data on chemotherapy. Table B23 Identified chemotherapy reviews Study Valachis Wilcken NHSC (iniparib) Type of chemotherapy All cytotoxic chemotherapy regimens were considered eligible, provided that the same drugs were given at the same dose in all study arms. Conventional cytotoxic chemotherapy (with or without colony stimulating factors but excluding cytokines or monoclonal antibodies) used alone, as well as high-dose chemotherapy requiring stem-cell support. Gemcitabine plus carboplatin. Of the three reviews above the Wilcken et al review was considered appropriate in establishing a link between everolimus and chemotherapy assuming that the outcomes for tamoxifen were equivalent to those of all endocrine therapies. In the Wilcken study, 3 of the 10 studies included tamoxifen as a comparator, and those studies were used as a basis for linking chemotherapy to everolimus (see section ). It should be noted that the Wilcken study provides data only for overall survival and therefore the chained link established in section is for OS. For PFS, the Specification for manufacturer/sponsor submission of evidence Page 109 of 292

110 efficacy of tamoxifen from TAMRAD was assumed to be the same as that of chemotherapy (see section ). It should be further noted that the Wilcken et al (2003) review was not connected to the indirect comparison network, but was included in the cost effectiveness analysis as a comparison of chemotherapy with everolimus using a naïve chained indirect analysis (see section ) Provide a summary of the trials used to conduct the indirect comparison. A suggested format is presented below. Network diagrams may be an additional valuable form of presentation. Four trials were identified for the indirect treatment comparison; these are shown in relation to their treatment arms in Table B24. An evidence network of the trials is shown in Figure B13, demonstrating the connections between comparators used in the analysis. Specification for manufacturer/sponsor submission of evidence Page 110 of 292

111 Table B24 Summary of the trials used to conduct the indirect comparison No. trial s References of trials Everolimus + exemestane Fulvestrant 500mg Fulvestrant 250mg Exemestane 1 BOLERO-2 1 CONFIRM 2 EFECT SOFEA Figure B13 Evidence network of trials used to inform the indirect treatment comparison Specification for manufacturer/sponsor submission of evidence Page 111 of 292

112 6.7.4 For the selected trials, provide a summary of the data used in the analysis. The indirect treatment comparison analysis focused on two outcomes: PFS or TTP (data available for all four trials) OS (data available for three trials). Although TTP and PFS are different outcomes, the indirect treatment comparison assumed that TTP is similar to PFS in diseases with short survival times. In the case of metastatic breast cancer, which has a median survival of approximately 2 years, 62 most deaths would be disease-related and therefore count towards progression. The hazard ratios and CIs are shown in Table B25 for PFS or TTP, and in Table B26 for OS. Table B25 Hazard ratios for PFS or TTP in studies selected for indirect treatment comparison Trial Treatment Comparator BOLERO-2 (central assessment) 19 CONFIRM 25 Everolimus in combination with exemestane Fulvestrant (500 mg) Hazard ratio 95% CI Exemestane Fulvestrant (250 mg) EFECT 24 Fulvestrant Exemestane SoFEA 26 Fulvestrant (250 mg) Exemestane Specification for manufacturer/sponsor submission of evidence Page 112 of 292

113 Table B26 Hazard ratios for OS in studies selected for indirect treatment comparison Trial Treatment Comparison BOLERO-2 19 Everolimus in combination with exemestane CONFIRM 25 Fulvestrant (500 mg) SoFEA 26 Fulvestrant (250 mg) Hazard ratio 95% CI Exemestane Fulvestrant (250 mg) Exemestane Please provide a clear description of the indirect/mixed treatment comparison methodology. Supply any programming language in a separate appendix. The treatments evaluated in the indirect treatment comparison were connected by one or more other treatments, although they failed to form a closed loop and they did not all have a common comparator. As a result, some treatments are connected by a longer path (Figure B13), and their comparison will have lower precision. Data for the indirect treatment comparison were taken from hazard ratios for PFS/TTP and OS from the four included studies. Log hazard ratios and their precision, or reciprocal of the variance, were calculated from these hazard ratios and CIs (Table B25 and Table B26). These formed the inputs for the model. A Bayesian fixed effect model was used for the analysis. Because exemestane is used in clinical practice and is the treatment with the most information in the network, it was adopted as the base treatment for the model. It should be noted that not all treatments have direct evidence relative to the baseline treatment. The basic parameters of the model are the log hazard ratios with respect to exemestane. Specification for manufacturer/sponsor submission of evidence Page 113 of 292

114 For each trial, the log hazard ratio was assumed to have a normal distribution, with the mean value equal to the true log hazard ratio and variance equal to the variance observed in the trial. The true log hazard ratio is modelled as the difference between the true log hazard ratios for the treatment and comparator with respect to exemestane. The basic parameters were given vague priors. Further details on the indirect treatment comparison methodology are provided in section Please present the results of the analysis. Results of the indirect treatment comparison analysis are summarised in Table B27 and Table B28. For PFS and TTP, everolimus in combination with exemestane was found to perform better than all other comparators, and this difference was statistically significant compared with exemestane, fulvestrant (both 250 mg and 500 mg) and tamoxifen. For the outcome of OS, everolimus in combination with exemestane was found to perform better than exemestane and fulvestrant (both 250mg and 500mg). Table B27 Indirect treatment comparison for PFS and TTP Comparator treatment Hazard ratio vs exemestane (95% CI) Hazard ratio vs everolimus in combination with exemestane (95% CI) Fulvestrant 250 mg (AIC information removed) (AIC information removed) Fulvestrant 500 mg (AIC information removed) (AIC information removed) HRs refer to the risk of the event, ie progression or death. In this particular analysis a high HR represents an increased risk of an event and therefore decreased PFS/OS for the comparator. Therefore a HR > 1 indicates worse effectiveness of the comparator treatment compared with everolimus in combination with exemestane. Specification for manufacturer/sponsor submission of evidence Page 114 of 292

115 Table B28 Indirect treatment comparison for OS Comparator treatment Hazard ratio vs exemestane (95% CI) Hazard ratio vs everolimus in combination with exemestane (95% CI) Fulvestrant 250 mg (AIC information removed) (AIC information removed) Fulvestrant 500 mg (AIC information removed) (AIC information removed) HRs refer to the risk of the event, ie progression or death. In this particular analysis a high HR represents an increased risk of an event and therefore decreased PFS/OS for the comparator. Therefore a HR > 1 indicates worse effectiveness of the comparator treatment compared with everolimus in combination with exemestane Please provide the statistical assessment of heterogeneity undertaken. The degree of, and the reasons for, heterogeneity should be explored as fully as possible. For this analysis, it is not possible to perform a complete statistical assessment of heterogeneity. There is only one pair-wise comparison for the PFS/TTP endpoint that is supported by evidence from more than one trial (fulvestrant 250 mg versus exemestane). For the endpoint of OS, there are no pair-wise comparisons that are supported by evidence from more than one trial. For PFS/TTP, the EFECT 24 and SoFEA 26 trials compared fulvestrant 250 mg with exemestane. The hazard ratios for the two trials are in close agreement (0.96 for EFECT compared with 0.95 for SoFEA; Table B29). The log hazard ratios were combined using the inverse variance method and the resulting I 2 value of 0% indicated very little heterogeneity between trials (Figure B14). Table B29 Indirect treatment comparison for PFS/TTP of fulvestrant (250 mg) versus exemestane Trial Hazard ratio 95% CI Log HR Precision EFECT SoFEA Overall (AIC information (AIC information (AIC information (AIC information removed) (AIC information removed) (AIC information removed) (AIC information removed) (AIC information removed) (AIC information Specification for manufacturer/sponsor submission of evidence Page 115 of 292

116 Trial Hazard ratio 95% CI Log HR Precision removed) removed) removed) removed) Figure B14 Forest plot of log hazard ratios for PFS/TTP comparison of fulvestrant (250 mg) versus exemestane (AIC information removed) If there is doubt about the relevance of a particular trial, please present separate sensitivity analyses in which these trials are excluded. No irrelevant trials that required a separate sensitivity analysis were identified Please discuss any heterogeneity between results of pairwise comparisons and inconsistencies between the direct and indirect evidence on the technologies. There are no pair-wise comparisons that have both direct and indirect evidence; the network does not contain any loops. As a result, inconsistency is not applicable to this network analysis. Specification for manufacturer/sponsor submission of evidence Page 116 of 292

117 6.8 Non-RCT evidence Non-RCT, both experimental and observational, evidence will be required, not just for those situations in which RCTs are unavailable, but also to supplement information from RCTs when they are available. This section should be read in conjunction with NICE s Guide to the methods of technology appraisal, sections to If non-rct evidence is considered (see section 6.2.7), please repeat the instructions specified in sections 6.1 to 6.5 for the identification, selection and methodology of the trials, and the presentation of results. For the quality assessments of non-rcts, use an appropriate and validated quality assessment instrument. Key aspects of quality to be considered can be found in Systematic reviews: CRD s guidance for undertaking reviews in health care ( Exact details of the search strategy used and a complete quality assessment for each trial should be provided in sections 10.6 and 10.7, appendices 6 and 7. No relevant non-rcts were identified. Specification for manufacturer/sponsor submission of evidence Page 117 of 292

118 6.9 Adverse events This section should provide information on the adverse events experienced with the technology in relation to the decision problem. Evidence from comparative RCTs and regulatory summaries is preferred; however, findings from non-comparative trials may sometimes be relevant. For example, postmarketing surveillance data may demonstrate that the technology shows a relative lack of adverse events commonly associated with the comparator, or the occurrence of adverse events is not significantly associated with other treatments If any of the main trials are designed primarily to assess safety outcomes (for example, they are powered to detect significant differences between treatments with respect to the incidence of an adverse event), please repeat the instructions specified in sections 6.1 to 6.5 for the identification, selection, methodology and quality of the trials, and the presentation of results. Examples for search strategies for specific adverse effects and/or generic adverse-effect terms and key aspects of quality criteria for adverseeffects data can found in Systematic reviews: CRD s guidance for undertaking reviews in health care ( Exact details of the search strategy used and a complete quality assessment for each trial should be provided in sections 10.8 and 10.9, appendices 8 and 9. No studies were identified which specifically aimed to capture safety evidence. Safety data are reported in section for both BOLERO-2 and the TAMRAD trial Please provide details of all important adverse events for each intervention group. For each group, give the number with the adverse event, the number in the group and the percentage with the event. Then present the relative risk and risk difference and associated 95% confidence intervals for each adverse event. A suggested format is shown below. Specification for manufacturer/sponsor submission of evidence Page 118 of 292

119 Everolimus was generally well tolerated in combination with exemestane or tamoxifen Results of the phase III BOLERO-2 study and the phase 2 TAMRAD trial demonstrate that everolimus was generally well tolerated in postmenopausal women with HR+, HER2 advanced breast cancer. 18,22,80 The overall pattern of AEs was consistent between the studies, with stomatitis and anaemia being the only grade 3/4 AEs observed in more than 5% of patients receiving everolimus in combination with exemestane. The rate of treatment withdrawal due to AEs was higher in the combination groups compared with exemestane or tamoxifen alone. This is likely to reflect the increased time on study drug and lower rates of withdrawal due to disease progression observed in the everolimus groups (discussed in section 6.9.3). The incidences of grade 3 or 4 AEs in the two studies are summarised in Table B30 and Table B31. Specification for manufacturer/sponsor submission of evidence Page 119 of 292

120 Table B30 Grade 3/4 adverse events across randomised groups in BOLERO-2 System organ/ class/adverse events 7-month follow-up, n (%) 18,60,80 18-month follow-up, n (%) 19,a Everolimus and Exemestane (N = 482) Placebo and Exemestane (N = 238) Everolimus and Exemestane (N = 482) Placebo and Exemestane (N = 238) Death, n (%) b 51 (10.7) 31 (13.0) 122 (25.4) 77 (32.2) SAEs, n (%) 110 (22.8) 29 (12.2) Grade 3 or 4 AEs, 211 (43.8) 61 (25.6) n (%) Withdrawal due to AEs, n (%) 32 (6.7) 7 (2.9) 44 (9.1) 8 (3.3) Non-haematological grade 3/4 AEs ( 2% of everolimus-treated patients), n (%) c Stomatitis 37 (7.7) 2 (0.8) (8) (<1) Gamma-glutamyltransferase level increased (CIC information removed) (CIC information removed) Dyspnoea 19 (3.9) 3 (1.2) Fatigue 18 (3.7) 6(26.1)/ (4) (1) 2 (0.8) Pneumonitis 15 (3.1) 0 (0) (3) (0) Alanine aminotransferase 15 (3.1) 4 (1.7) level increased Aspartate 13 (2.7) 3 (1.3) aminotransferase level increased Hypokalaemia (CIC information removed) (CIC information removed) Diarrhoea 10 (2.1) 2 (0.8) (2) (<1) Rash 4 (0.8) 0 (0) (1) (0) Haematological grade 3/4 AEs, ( 2% of everolimus-treated patients), n (%) Anaemia (CIC information removed) (CIC information removed) Thrombocytopenia 15 (3.1) 1 (0.4) Neutropaenia (CIC information removed) (CIC information removed) Baselga et al., ,80 ; BOLERO-2 7-month CSR 60 ; Piccart et al., a Only AEs occurring in 25% of patients or of particular interest (pneumonia) are reported for the 18-month analysis. b A total of 83 and 200 deaths occurred before the 7-and 18-month analyses; one death occurred before treatment commenced, and is not included here. c At the 7-month follow-up, a total of 5% of patients in the everolimus plus exemestane group, and 1.7% in the exemestane alone group, had experienced grade 3/4 infections no single category of infection exceeded 1%. AE, adverse event; SAE, serious adverse event. Adapted from European Public Assessment Reports published by the European Medicines Agency Specification for manufacturer/sponsor submission of evidence Page 120 of 292

121 Table B31 Adverse events across randomised groups in the TAMRAD trial System organ/ 24-month follow-up 22 class/adverse events Everolimus and Tamoxifen (N = 54) Tamoxifen alone (N = 57) Death, n (%) 16 (30) 31 (54) SAEs, n (%) a 17 (32) 18 (32) Grade 3 or 4 AEs, n (%) Withdrawal due to AEs, n (%) 12 (22) 4 (7) Non-haematological AEs ( 10% of everolimus-treated patients), n (%) any grade/grade 3 or 4 Stomatitis 30 (56)/ 6 (11) 4 (7)/ 0 Rash 24 (44)/ 2 (4) 4 (7)/ 0 Fatigue 39 (73)/ 3 (6) Diarrhoea 21 (39)/ 1 (2) Nausea 19 (35)/ 2 (4) Vomiting 9 (17)/ 0 Anorexia 23 (43)/ 4 (7) Constipation 9 (17)/ 0 Pneumonitis 9 (17)/ 1 (2) Infection 19 (35)/ 4 (7) Hot flashes 12 (22)/ 0 Pain 44 (82)/ 5 (9) 30 (53)/ 6 (11) 6 (11)/ 0 20 (35)/ 0 7 (13)/ 2 (4) 10 (18)/ 2 (4) 13 (23)/ 0 2 (4)/ 2 (4) 11 (19)/ 3 (5) 19 (33)/ 0 49 (86)/ 10 (18) Haematological AEs ( 10% of everolimus-treated patients), n (%) any grade/grade 3 or 4 Decreased haemoglobin 37 (69)/ 1 (2) Decreased leukocyte count 29 (54)/ 1 (2) Decreased lymphocyte count 26 (48)/ 1 (2) PNNs 26 (48)/ 1 (2) 20 (35)/ 2 (4) 10 (18)/ 0 12 (21)/ 2 (4) 11 (19)/ 3 (5) Bachelot et al AE, adverse event; SAE, serious adverse event. Adapted from European Public Assessment Reports published by the European Medicines Agency Specification for manufacturer/sponsor submission of evidence Page 121 of 292

122 6.9.3 Give a brief overview of the safety of the technology in relation to the decision problem. Exposure to study treatment In BOLERO-2 and the TAMRAD trial, the duration of exposure to treatment was longer in patients treated with everolimus in combination with exemestane or tamoxifen, respectively, compared with the exemestane or tamoxifen alone groups (Table B32). Indeed at the 12-month analysis for BOLERO-2, the duration of exposure to everolimus was approximately double that for placebo (24 weeks versus 13 weeks), possibly reflecting the lower rate of disease progression in the everolimus group. The difference in duration of exposure therefore may well have influenced the incidence of AEs and the rate of discontinuation for AEs reported in the two studies. Table B32 Duration of exposure to study treatment Exposure to study drug (exposure to exemestane/tamoxifen) Study Everolimus group Exemestane/tamoxifen alone group BOLERO-2 a 7-month analysis 14.6 weeks (17.4 weeks) 12.0 weeks (12.0 weeks) 12-month analysis 23.9 weeks (26.6 weeks) 13.4 weeks (14.1 weeks) TAMRAD 24-month analysis 6.2 months b 4.8 months Bachelot et al., ; Baselga et al., ; BOLERO-2 12-month CSR. 88 a Patients who discontinued everolimus/placebo for any reason other than progression could continue to receive exemestane. b 12 patients discontinued the combination due to an AE: 6 discontinued everolimus only, 1 tamoxifen only and 5 both treatments. Adverse events In BOLERO-2 and the TAMRAD trial, most AEs were grade 1 or 2 and the incidence of grade 3 or 4 haematological AEs was low (Table B30 and Table B31, respectively). 18,22 Stomatitis was the most frequently reported grade 3 or 4 non-haematological AE. Grade 3 or 4 non-infectious pneumonitis was observed in 2% to 3% of patients; this is a recognised class effect of mtor inhibitors. Everolimus is also associated with an increased risk of infections, however the incidence of grade 3 or 4 infections was (CIC information removed) in BOLERO-2 (versus (CIC Specification for manufacturer/sponsor submission of evidence Page 122 of 292

123 information removed) in the exemestane alone group) and 7% (versus 5% in the tamoxifen alone group) in the TAMRAD trial. BOLERO-2 The incidence of grade 3 and 4 AEs was low during treatment with everolimus in combination with exemestane In BOLERO-2, grade 3 or 4 AEs were reported in 44% of patients who received everolimus plus exemestane (34% attributed to study treatment) and in 26% of patients who received placebo in combination with exemestane (8% attributed to study treatment). The longer duration of exposure to study treatment may have contributed to the higher incidence in the everolimus group. The incidence of grade 4 AEs was low in both groups (7.3% and 5%, respectively). The most frequently reported grade 3 or 4 AE with everolimus in combination with exemestane therapy was stomatitis (7.7% versus 0.8% in the placebo in combination with exemestane group) (Table B30) and anaemia was the only other grade 3 or 4 AE reported in at least 5% of patients receiving everolimus in combination with exemestane. Other grade 3 or 4 AEs reported in more than 3% of patients receiving everolimus were dyspnoea, hyperglycaemia, fatigue, pneumonitis, increased alanine aminotransferase and gamma-glutamyl transferase levels and thrombocytopenia; the incidence of these AEs was approximately 2% to 4% higher than that observed in the placebo in combination with exemestane group. 18 Serious adverse events (SAEs) were reported in 23% of patients who received everolimus in combination with exemestane (11% attributed to study treatment) and in 12% of patients who received placebo in combination with exemestane (1% attributed to study treatment). 18 Based on available data, a similar difference in incidence of AEs was observed at 18 months. 19 Specification for manufacturer/sponsor submission of evidence Page 123 of 292

124 Few patients discontinued everolimus due to AEs Rates of discontinuation due to AEs were low in both the everolimus in combination with exemestane and placebo in combination with exemestane groups at the 7-month, 12-month and 18-month analyses (6.6% versus 2.9%, 8.0% versus 3.3% and 9.1% versus 3.3%, respectively, at 7, 12 and 18 months) (7- and 18-month data are summarised in Table B30) Discontinuation rates were higher in the everolimus group, probably reflecting in part the longer time on treatment in the everolimus in combination with exemestane group (Table B32). Over the median followup of 7 months, 58% of patients receiving everolimus required dose interruptions or reductions due to AEs, most commonly (in 5% of patients), stomatitis (CIC information removed) pneumonitis (CIC information removed) and thrombocytopenia (CIC information removed) At the 7-month analysis, mean dose intensity for everolimus was 7.89 mg/day and the relative dose intensity was 0.79 for everolimus and for exemestane. At 12 months, the mean dose intensity for everolimus was (CIC information removed) and the relative dose intensity was (CIC information removed) for everolimus and (CIC information removed) for exemestane. Based on the 7-month data, there were seven deaths (1%) attributed to AEs that occurred during treatment or within 28 days after stopping treatment in the everolimus in combination with exemestane group: two from sepsis, one each from pneumonia, tumour haemorrhage, cerebrovascular incident, renal failure and suicide. In the placebo in combination with exemestane group, one death due to pneumonia was reported during treatment. 18 The TAMRAD trial As observed in BOLERO-2, the most frequently reported grade 3 or 4 AE with everolimus in combination with tamoxifen therapy was stomatitis (11% versus 0% in tamoxifen-only group) (Table B31). 22 Other grade 3 or 4 AEs reported in more than 5% of patients receiving everolimus were fatigue, anorexia, infection and pain. The incidence in the everolimus in Specification for manufacturer/sponsor submission of evidence Page 124 of 292

125 combination with tamoxifen group was 2% to 4% higher than the tamoxifen alone group for grade 3 or 4 rash, nausea, anorexia and infection, whereas the incidence of grade 3 or 4 fatigue and pain was 5% and 8% higher in the tamoxifen alone group, respectively. The overall incidence of SAEs was 32% in both treatment groups. Treatment-related AEs led to everolimus dose reductions in 11 patients (20%). The main reasons for these dose reductions were stomatitis (n = 4), rash (n = 1), thrombocytopenia (n = 1), and pneumonitis (n = 1). Of these 11 patients, only three later stopped everolimus because of additional AEs before disease progression. More patients discontinued treatment due to AEs in the everolimus in combination with tamoxifen group compared with the tamoxifen alone group (22% versus 7%), possibly reflecting the longer duration on treatment in the everolimus group (Table B32). Reasons for discontinuing both treatments in the everolimus in combination with exemestane group were venous thrombosis (n = 2), acute renal insufficiency (n = 1), cardiac failure (n = 1), and haemorrhagic risk (n = 1), a further 6 patients discontinued everolimus while maintaining tamoxifen after experiencing pneumonitis (n = 2), rash (n = 2), hyperglycaemia (n = 1), and stomatitis (n =1) and one patient discontinued tamoxifen while maintaining everolimus following venous thrombosis. Reasons for discontinuing tamoxifen were venous thrombosis (n = 2), pulmonary embolism (n = 1), and asthenia (n = 1) in the tamoxifen alone group. Specification for manufacturer/sponsor submission of evidence Page 125 of 292

126 6.10 Interpretation of clinical evidence Please provide a statement of principal findings from the clinical evidence highlighting the clinical benefit and harms from the technology. Efficacy Everolimus in combination with exemestane provided clinically meaningful improvements in PFS compared with exemestane monotherapy in postmenopausal women with HR+ metastatic breast cancer refractory to NSAI therapy The addition of everolimus to exemestane more than doubled median PFS compared with placebo in combination with exemestane from approximately 3.5 months to between 7 and 11 months, corresponding to a reduction in the risk of disease progression of 55 to 57% (local assessment) and 62 to 64% (central assessment). 18,19 Results were consistent and statistically significant (p < 0.001) across all analyses including at the three time points (ie at a median follow-up of 7, 12 and 18 months) and for both local and central assessment. A statistically significant benefit for the addition of everolimus was observed for all subgroups analysed, including in the elderly ( 65 and > 70 years of age) and in patients with or without visceral metastases. PFS assessments were based on radiological assessment of tumour burden and for each time point responses were assessed both locally and centrally. PFS was defined as the primary endpoint in the study and was assessed by local and independent central review (the latter is usually considered to be superior due to higher stringency and lower risk of bias). Results for the addition of everolimus to tamoxifen, as reported for the TAMRAD trial, also indicated that the addition of everolimus to endocrine therapy provides a clinically meaningful reduction in the risk of disease progression. The addition of everolimus was associated with a reduction in the risk of disease progression of 46% for the total study population and Specification for manufacturer/sponsor submission of evidence Page 126 of 292

127 was particularly marked in patients with secondary hormone resistance (risk reduction of 54%). 22 The latter suggests that everolimus is able to circumvent acquired resistance mediated by the ER pathway and is consistent with the suggested mechanism of action of everolimus. PFS observed for everolimus in combination with exemestane compared favourably with that reported for single-agent endocrine therapy or chemotherapy in patients in whom NSAIs are failing The median PFS reported for everolimus in combination with exemestane in BOLERO-2 compares favourably with that reported previously for endocrine therapy used second-line after progression on NSAIs (Figure B15a). The median PFS for everolimus in combination with exemestane in BOLERO-2 was 7.8 months by local assessment and 11.0 months by central assessment, while the corresponding values for exemestane monotherapy were 3.2 months and 4.1 months. 19 These exemestane monotherapy PFS results are in good agreement with those reported in two other studies that have investigated endocrine monotherapy in this patient population (the SoFEA study, the EFECT study). 24,26 PFS for tamoxifen monotherapy has been reported in two studies and ranged from 4.5 months 22 to 5.8 months, 92 and PFS for fulvestrant (250 or 500 mg) has been reported in three studies with results ranging from 3.7 to 6.5 months Taken together, these results suggest that the PFS benefit achieved with everolimus in combination with exemestane exceeds that achieved by switching to another endocrine therapy second-line. Chemotherapy can be an alternative option for patients with advanced disease in whom endocrine therapy is failing. The PFS of 11.0 months (central assessment) achieved with everolimus in combination with exemestane in BOLERO-2 compares favourably with results reported for various chemotherapy regimens in the setting of advanced disease (Figure B15b). In a phase III study from 1999 comparing docetaxel with doxorubicin, the median TTP was 26 weeks for docetaxel and 21 weeks for doxorubicin. 27 and subsequent analysis of this study to determine the effects of these chemotherapeutic agents in HER2 patients reported that Specification for manufacturer/sponsor submission of evidence Page 127 of 292

128 the respective median TTP for docetaxel was 5.0 months and for doxorubicin was 5.9 months. 28 Similar findings have been reported in other clinical studies with docetaxel in advanced disease after anthracycline failure (median TTP 6.3 months) 29 and for doxorubicin as first-line treatment of metastatic breast cancer (PFS 7.8 months). 30 Capecitabine monotherapy has also been studied first-line in HER2 metastatic breast cancer and recent reviews identify four phase III studies which reported median TTP/PFS for capecitabine monotherapy as first-line therapy ranging from 2.8 months to 7.1 months. 31 Specification for manufacturer/sponsor submission of evidence Page 128 of 292

129 Figure B15 PFS or TTP (months) reported for a) endocrine monotherapy or b) chemotherapy in phase III studies in women with HER2 advanced breast cancer. BOLERO-2 (Central analysis) 19 ; TAMRAD 22 ; CONFIRM 25 ; EFECT 24 ; SoFEA 26 a TTP reported in Sjostrom et al and Di Leo et al Chan et al and Di Leo et al ; Sjostrom et al ; O Brien ; O Shaughnessy 2012 (review) 31 Specification for manufacturer/sponsor submission of evidence Page 129 of 292

130 Everolimus in combination with endocrine therapy improves survival over endocrine monotherapy The prolongation of PFS achieved with the addition of everolimus to exemestane was accompanied by a numerical reduction in the number of deaths in BOLERO-2. The difference in incidence of death in the two treatment groups increased progressively with increasing follow-up for BOLERO-2 (from 2.3% at 7 months to 6.8% at 18 months follow-up) suggesting that a significant OS advantage for the addition of everolimus may be reported when the data are more mature. 18,19 This is supported by the observation that a highly statistically significant improvement in OS was reported for the addition of everolimus to tamoxifen in the TAMRAD trial at a median follow-up of 24 months (median OS not reached in the everolimus in combination with tamoxifen group versus 32.9 months in the tamoxifen alone group; p = 0.007). 22 This contrasts with the results of most studies of endocrine therapies in the second-line setting, which failed to achieve a significant OS benefit (Table B22) Clinically and statistically significant improvements in clinical benefit rate (CBR) were achieved with everolimus in combination with endocrine therapy over endocrine monotherapy CBR (defined as CR, PR or SD at 24 weeks) is also considered a clinically meaningful endpoint A statistically significant improvement in CBR of approximately twofold at 12 and 18 months was observed with everolimus in combination with exemestane over exemestane alone. 19,22 The CBR at the18-month follow-up was 51.3% in BOLERO-2 for the combination, 19 which compares favourably with the CBRs reported for other second-line endocrine therapies which range from 32.2% for fulvestrant and 31.5% for exemestane in EFECT, 24 through to 39.6% for fulvestrant 250 mg in the CONFIRM study. 25 The addition of everolimus to tamoxifen in TAMRAD also significantly increased the CBR from 42% to 61%. 22 The improvements in CBR achieved with everolimus in combination with endocrine therapy in BOLERO-2 and the TAMRAD trial can be expected to result in improvements in survival Specification for manufacturer/sponsor submission of evidence Page 130 of 292

131 Everolimus slows down the rate of progression in the bone It is estimated that 40 to 70% of patients with advanced breast cancer have bone metastases. 93 Bone metastases commonly result in debilitating pain and fractures; additionally, patients often develop hypercalcaemia and spinal cord compression, and many require radiotherapy for treatment of metastases and/or surgery for management of fractures. All of these can dramatically impair patients QoL. 65 Furthermore, it has been shown that cancer patients consume progressively more hospital resources when they develop bone metastases and subsequent skeletal-related events. 94 Everolimus in combination with exemestane significantly reduced the risk of disease progression in the bone compared with treatment with exemestane alone (p = 0.036). 20,85 The cumulative incidence rate of progressive disease in bone was lower for patients receiving everolimus, both in the total study population and for the subgroup with bone metastases at study entry. This reduction in the risk of disease progression in the bone can therefore be expected to result in an improvement in HRQoL and survival, and may reduce overall treatment costs. Everolimus reversed the increase in bone reabsorption associated with exemestane and reduced the risk of fractures In addition to effects on bone metastases, everolimus was also found to reduce the risk of fractures. AIs such as exemestane are associated with bone loss and an increased risk of fractures. 89 Such fractures and their management can adversely affect a patient s HRQoL; hence the impact of treatment on bone strength can be an important factor in the choice of therapy for patients with advanced disease. In BOLERO-2, everolimus reversed the increase in bone resorption associated with exemestane and reduced the incidence of fractures. 20,85 Thus everolimus in combination with exemestane may be expected to have important benefits for patients in terms of effects on bone health as well as anti-tumour effects. Specification for manufacturer/sponsor submission of evidence Page 131 of 292

132 Everolimus extended PFS without compromising quality of life Improvements in PFS, OS, CBR and bone health with everolimus in combination with exemestane were achieved without compromising HRQoL. 77 Over the course of the study, HRQoL decreased in many patients as would be expected, however the time to definitive deterioration of QoL was statistically significantly longer in the everolimus in combination with exemestane group compared with the placebo in combination with exemestane group at the 18-month follow-up (8.3 versus 5.8 months; p = ). 21 This suggests that the benefits of treatment with everolimus in combination with exemestane outweigh any negative effects of additional AEs on HRQoL. Previous studies of second-line endocrine therapies have reported no significant differences in HRQoL for treatment with exemestane versus fulvestrant 24 or for fulvestrant 250 mg and 500 mg 25 suggesting that everolimus in combination with exemestane may be associated with a better HRQoL than exemestane alone or fulvestrant. Specification for manufacturer/sponsor submission of evidence Page 132 of 292

133 Safety Everolimus has a predictable safety/tolerability profile that compares favourably with that of chemotherapy Everolimus was generally well tolerated in both BOLERO-2 and the TAMRAD trial. In both studies, most AEs were grade 1 or 2 in severity and few grade 4 AEs were reported in BOLERO-2. The most frequently reported grade 3/4 AE in both studies was stomatitis, a known class effect of mtor inhibitors, and in most cases was managed by dose reductions or treatment interruptions. The only other grade 3/4 AEs reported in more than 5% of patients receiving everolimus were anaemia (in BOLERO-2) and fatigue, anorexia, infections and pain (in the TAMRAD trial). The safety profile observed in both studies was consistent with that established in the other two approved indications for everolimus, namely renal cell carcinoma and pancreatic neuroendocrine tumour In both studies, rates of discontinuation for AEs were somewhat higher for patients receiving everolimus compared with the exemestane or tamoxifen alone groups. This probably reflects the duration of exposure to therapy in the everolimus groups which was fold longer than in the exemestane or tamoxifen alone groups. At the 18-month follow-up, 9.1% of patients had discontinued treatment with everolimus in combination with exemestane in BOLERO-2 (versus 3.3% in the exemestane alone group) and 22% had discontinued therapy with everolimus in combination with tamoxifen (versus 7% in the tamoxifen alone group) in the TAMRAD trial at 24 months. At a median follow-up of 7 months, 58% of patients receiving everolimus in combination with exemestane required dose interuptions or reductions due to AEs but the relative dose intensity for everolimus was 0.79 at this followup and remained at (CIC information removed) at a follow-up of 12 months. 60,88 The safety profile of everolimus in combination with exemestane compared favourably with that reported for chemotherapy regimens used in this setting. 27,29-31,98 As summarised in Table B33 the incidence of grade 3/4 Specification for manufacturer/sponsor submission of evidence Page 133 of 292

134 diarrhoea, nausea and vomiting are all higher with chemotherapy than with everolimus in combination with exemestane, and doxorubicin and docetaxel are associated with higher rates of grade 3/4 neutropenia (in excess of 80%), other haematological adverse events and higher rates of grade 3/4 asthenia than have been reported for everolimus in combination with exemestane. Doxorubicin treatment is also associated with cardiac toxicity as noted in the study by Chan et al which reported that 3.7% of patients treated with doxorubicin developed congestive heart failure (CHF), 27 and in the study by O Brien et al. where 14.1% of patients discontinued doxorubicin therapy due to cardiac toxicity. 30 Reporting of capecitabine-related adverse events from phase III studies notes rates of grade 3/4 hand and foot syndrome of between 16.0% and 26% and rates of diarrhoea and vomiting greater than those for everolimus in combination with exemestane. 31 By contrast, stomatitis, reported in 8% to 11% of patients, is the most frequently reported grade 3/4 AE associated with everolimus. 18,22 Specification for manufacturer/sponsor submission of evidence Page 134 of 292

135 Table B33 Comparison of incidence of grade 3/4 AEs during therapy with everolimus in combination with exemestane and with chemotherapy regimens in women with advanced breast cancer Non-haematological grade 3/4 AEs, % Stomatitis Docetaxel Doxorubicin Capecitabine a Everolimus and exemestane b 12.3 b 0, NR, < c 2.0 d 14.1 b , 2.0, NR b b 4.3 b - b 12.3 b 16.1 b Fatigue 2.0 d 3.0, 7.0, Diarrhoea 10.7 b 10.0 c , 12.0, Nausea 3.1 b 6.0 c 5.0 d Vomiting 3.1 b 12.3 b 4.0 d Skin toxicity/rash 1.9 b 2.0 c 0.0 d Nail disorder/toxicity 2.5 b 5.0 c 0.0 Allergy 2.5 b 2.0 c Infection 2.5 b 26.0 c HaFS 16.0, 26.0, 19.0 Alopecia - b 74.0 c 0.0 d Dyspnoea 3.9 Hyperglycaemia 4.3 Pneumonitis 3.1 Decreased <1.0 d 1.0 appetite/anorexia Decreased weight 1.0 Peripheral 3.0 c 1.0 oedema/fluid retention Asthenia 14.5 b c 1.0 d Haematological grade 3/4 AEs, % Anaemia 4.4 b 2.0 c ~2.0 d Neutropenia 93.5 b 88.9 b 8.0 d Thrombocytopenia 1.3 b 7.5 b 3.0 c <1/0 d Leucopenia 77.0 c 9.0 d PSN peripheral sensory neuropathy; HaFS hand-foot syndrome; NR not reported a values reported in three phase III studies reviewed by O Shaughnessey et al Note the Ribbon-1 study comparing capecitabine monotherapy vs capecitabine in combination with bevacizumab did not report information on capecitabine-related AEs. 31,98 b reported in Chan et al c reported in Sjostrom et al d reported in O Brien Specification for manufacturer/sponsor submission of evidence Page 135 of 292

136 Please provide a summary of the strengths and limitations of the clinical-evidence base of the intervention. Strengths Evidence for the efficacy and safety of everolimus in combination with exemestane is based on results from BOLERO-2, a large, randomised, double-blind, controlled multicentre, international study. 18 The results of this study provide robust evidence for the clinical efficacy and safety of everolimus in combination with exemestane in postmenopausal women with HR+ breast cancer refractory to NSAI therapy based on the rigorous design of the study, inclusion of appropriate endpoints, and the duration of follow-up. The study involved 724 women from 189 centres in 24 countries, and was double-blind with respect to treatment with everolimus or placebo. PFS was the primary endpoint and the study was powered for significance. PFS is particularly relevant in this setting as it is correlated with OS, 86 and is not confounded by subsequent treatments after disease progression. The trial was also powered to detect significant differences in OS between treatments groups and OS was included as a secondary endpoint. No crossover between treatment groups was allowed during the study; therefore OS would not be confounded. Tumour response was determined using RECIST v1.0 criteria by both local and central assessment, and HRQoL was assessed using the validated EORTC QLQ-C30 tool at 6-week intervals. Data for all endpoints are available for three analyses, performed at a median follow-up of 7, 12 and 18 months, and provide confirmation of the efficacy of treatment as well as an indication of long-term outcome. The TAMRAD trial, an open-label study involving 111 patients from centres in France, provides robust comparative evidence for the efficacy and safety of tamoxifen monotherapy, one of the comparators of interest, and provides additional evidence for the efficacy of everolimus in combination with endocrine therapy in the patient population of interest. Results for the two studies are in good agreement regarding the efficacy and safety profile of Specification for manufacturer/sponsor submission of evidence Page 136 of 292

137 everolimus in this setting. Furthermore the TAMRAD trial provides a longer follow-up of 24 months and analysis of efficacy according to type of hormone resistance. 22 Evidence for the safety profile of everolimus is further supported by the results of studies in the other two approved indications which are in good agreement with those of BOLERO-2 and the TAMRAD trial, and ongoing clinical experience with everolimus in these indications. Limitations Limitations regarding the evidence for everolimus in combination with exemestane in postmenopausal women with HR+ breast cancer refractory to NSAI therapy relate to the duration of follow-up for BOLERO-2. While the final planned PFS analysis for BOLERO-2 has been completed, the final planned analysis of OS (at 398 deaths) has yet to be performed. Longer follow-up of this study is required for OS data to mature and hence to determine the impact of everolimus on OS. Long-term data are available for the TAMRAD trial but this study was smaller (N = 111), open label in design and the regimen investigated everolimus in combination with tamoxifen does not correspond to the approved indication for everolimus. A further short-coming of the available clinical data relating to this submission is that there are only limited data regarding the comparators, fulvestrant, tamoxifen and chemotherapy in this setting. There are no studies directly comparing everolimus in combination with exemestane with fulvestrant or chemotherapy. However, a systematic review identified four RCTs, including BOLERO-2, which provided robust data on the comparative efficacy of endocrine therapy in the relevant setting. These four studies allowed a robust indirect comparison to be performed for everolimus in combination with exemestane versus fulvestrant, as described in section 6.7. Specification for manufacturer/sponsor submission of evidence Page 137 of 292

138 Please provide a brief statement of the relevance of the evidence base to the decision problem. Include a discussion of the relevance of the outcomes assessed in clinical trials to the clinical benefits experienced by patients in practice. Data from BOLERO-2 are directly relevant to the decision problem. BOLERO-2 provides evidence for the comparative efficacy and safety of everolimus in combination with exemestane (the therapy of interest) and one of the comparators, exemestane. Patients are representative of those who would be eligible to receive everolimus in combination with exemestane in clinical practice in the UK and 6 UK centres (13 patients) were involved in the study. The outcome measures of PFS and OS provide an accurate assessment of the clinical benefit of therapy and are used in the cost-effectiveness model. Additional evidence for the clinical benefit of everolimus in this indication is provided by data from the TAMRAD trial. Specification for manufacturer/sponsor submission of evidence Page 138 of 292

139 Identify any factors that may influence the external validity of study results to patients in routine clinical practice; for example, how the technology was used in the trial, issues relating to the conduct of the trial compared with clinical practice, or the choice of eligible patients. State any criteria that would be used in clinical practice to select patients for whom treatment would be suitable based on the evidence submitted. What proportion of the evidence base is for the dose(s) given in the SPC? The findings of BOLERO-2 have high external validity. The criteria used in clinical practice to select patients for whom everolimus in combination with exemestane would be suitable are similar to the inclusion and exclusion criteria of the phase III BOLERO-2 study, ie HR+ status, menopausal status, response status to NSAIs, performance status and disease burden. The doses of everolimus and exemestane used in BOLERO-2 correspond to the licensed doses. In BOLERO-2, two dose modifications an initial reduction to 5 mg daily and a subsequent reduction to 5 mg every other day were permitted. Recommendations for clinical practices are to reduce the dose to 5 mg daily or temporarily stop therapy for 1 week 43 and thus correspond closely to the dose adjustments used in BOLERO-2. Treatment was continued until disease progression or unacceptable toxicity this corresponds to current clinical practice for endocrine therapy. All evidence for the efficacy and safety of everolimus in HR+ metastatic breast cancer in postmenopausal women refractory to AI, as reported for BOLERO-2 and the TAMRAD trial, is for the licensed dose of 10 mg daily, with dose reductions as required for management of AEs. Specification for manufacturer/sponsor submission of evidence Page 139 of 292

140 7 Cost effectiveness 7.1 Published cost-effectiveness evaluations Identification of studies Describe the strategies used to retrieve relevant cost-effectiveness studies from the published literature and from unpublished data held by the manufacturer or sponsor. The methods used should be justified with reference to the decision problem. Sufficient detail should be provided to enable the methods to be reproduced, and the rationale for any inclusion and exclusion criteria used should be provided. The search strategy used should be provided as in section 9.10, appendix 10. A range of databases indexing published research were searched for studies about the cost-effectiveness of everolimus in postmenopausal women with HR+, HER2, advanced (locally advanced, metastatic) breast cancer who have already received endocrine therapy. The databases searched included: MEDLINE, MEDLINE In-Process, EMBASE, EconLit and the NHS Economic Evaluation Database (NHS EED). The search strategy used did not include an economic search filter because scoping searches had indicated that the amount of literature for everolimus is very small. This made it possible to search for only the drug name in combination with search terms for advanced or metastatic breast cancer. No date or language limits were applied. Full details of the search strategies, and databases and resources searched are provided in section 10.10, Appendix 10. Description of identified studies Provide a brief overview of each study, stating the aims, methods, results and relevance to decision-making in England and Wales. Each study s results should be interpreted in light of a critical appraisal of its methodology. When studies have been identified and not included, justification for this should be provided. If more Specification for manufacturer/sponsor submission of evidence Page 140 of 292

141 than one study is identified, please present in a table as suggested below. No economic evaluations of everolimus were identified Please provide a complete quality assessment for each costeffectiveness study identified. Use an appropriate and validated instrument, such as those of Drummond and Jefferson (1996) 1 or Philips et al. (2004) 2. For a suggested format based on Drummond and Jefferson (1996), please see section 9.11, appendix 11. No economic evaluations of everolimus were identified. 7.2 De novo analysis Patients What patient group(s) is (are) included in the economic evaluation? Do they reflect the licensed indication/ce marking or the population from the trials in sections 1.4 and 5.3.3, respectively? If not, how and why are there differences? What are the implications of this for the relevance of the evidence base to the specification of the decision problem? For example, the population in the economic model is more restrictive than that described in the (draft) SPC/IFU and included in the trials. The patient group considered in the base case is postmenopausal women with HR+ HER2 metastatic breast cancer who have progressed on therapy with a NSAI. The base-case uses data from BOLERO-2. This patient population is considered representative of the patients who will receive everolimus in the UK. 1 Drummond MF, Jefferson TO (1996) Guidelines for authors and peer reviewers of economic submissions to the BMJ. The BMJ Economic Evaluation Working Party. British Medical Journal 313 (7052): Philips Z, Ginnelly L, Sculpher M, et al. (2004) Quality assessment in decision-analytic models: a suggested checklist (Appendix 3). In: Review of guidelines for good practice in decision-analytic modelling in health technology assessment. Health Technology Assessment 8: 36. Specification for manufacturer/sponsor submission of evidence Page 141 of 292

142 Model structure Please provide a diagrammatical representation of the model you have chosen. The model structure is shown in Figure B16. Figure B16 Model structure Please justify the chosen structure in line with the clinical pathway of care identified in section 2.4. A Markov model was used to represent the progressive nature of the disease. The model structure is typical in modelling metastatic cancers, and has been used in many previous NICE submissions (for example, eribulin [TA 250] 36 and fulvestrant [TA 239] 37 ). It captures the progressive nature of the disease, and reflects the main outputs produced by the pivotal sources of evidence (the BOLERO-2 and TAMRAD trials) Please define what the health states in the model are meant to capture. The health states in a Markov model are typically defined based on a patient s possible health states specific to the disease. The states are mutually exclusive and collectively exhaustive; that is, patients can only be Specification for manufacturer/sponsor submission of evidence Page 142 of 292