Ovarian Cancer: New insights into biology and treatment

Transcription

1 Ovarian Cancer: New insights into biology and treatment 2018 Master Class Course Ursula A. Matulonis, MD Director, Gynecologic Oncology Brock-Wilson Family Chair Dana-Farber Cancer Institute Professor of Medicine Harvard Medical School Boston, MA

2 Disclosure Disclosure I have nothing to disclose. Off Label/Investigational Discussion In accordance with Annenberg Center policy, faculty have been asked to disclose discussion of unlabeled or unapproved use(s) of drugs or devices during the course of their presentations.

3 Question 1 Mrs. S is a 65 year-old woman with newly diagnosed advanced epithelial ovarian cancer; the histology is high grade serous. Germline testing is negative for BRCA1 or BRCA2. Which of the following statements is FALSE regarding next recommendations? A. Olaparib as maintenance therapy post carboplatin/paclitaxel is not FDA approved for newly diagnosed patients, regardless of BRCA status. B. Administration of heated intraperitoneal cisplatin chemotherapy (HIPEC) provides a survival benefit in newly diagnosed patients. C. Addition of bevacizumab to carboplatin and paclitaxel provides a progression free survival advantage but no survival benefit compared to carboplatin and paclitaxel alone. D. The ICON8 trial showed that weekly carboplatin and weekly paclitaxel has a worse OS and PFS compared to every 3 week carboplatin and paclitaxel. E. Neoadjuvant chemotherapy is equivalent to upfront chemotherapy with respect to OS and PFS, but patients have reduced post-operative mortality when chemotherapy is used neoadjuvantly.

4 Question 2 Ms. S is a 50 year-old woman with recurrent platinum-sensitive high-grade serous ovarian cancer. She initially received carboplatin and paclitaxel, and has now recurred 14 months after completing this treatment. Germline panel testing reveals a deleterious BRCA1 mutation. You decide to give her carboplatin/liposomal doxorubicin followed by olaparib maintenance. Which of the following statements is FALSE? A. Olaparib can lead to elevated serum creatinine values B. Olaparib s most common hematologic toxicity is anemia C. Niraparib could also be used, rather than olaparib, as a maintenance PARP inhibitor for this patient D. Carboplatin and liposomal doxorubicin has a higher incidence of carboplatin hypersensitivity reactions compared to carboplatin and paclitaxel E. Olaparib s FDA maintenance approval is for both BRCA wild-type as well as BRCA mutated cancers.

5 Ovarian Cancer 2018 Incidence in U.S. women 22,240 new cases estimated for 2018 Remains the second most common gynecologic tumor in the U.S. (endometrial cancer is the most common) ~1.4% lifetime risk of developing ovarian cancer Mortality in U.S. women 14,070 estimated deaths in th most common cause of cancer death in women 5-year survival Very modestly improved over time 36% ( ); 38% ( ); 43% ( ) Siegal et al, 2018 SEER.gov

6 Ovarian Cancer Risk Factors Increased Risk Advanced age Inherited high risk genetic mutation Family history Nulliparity Estrogen replacement Pelvic inflammatory disease Living in industrialized Western countries Being of Jewish descent Decreased Risk Lactation Oral contraceptives Parity Reproductive surgery (tubal ligation oophorectomy/hysterectomy) Martin VR. In: Yarbro CH, et al, eds. Cancer Nursing: Principles and Practice. 6th ed. Jones and Bartlett; Ozols RF, et al. In: Hoskins WJ, et al, eds. Principles and Practice of Gynecologic Oncology. Philadelphia, Pa: Lippincott Williams & Wilkins; Lacey JV Jr, et al. JAMA. 2002;288: Dibble SL, et al. Oncol Nurs Forum. 2002;29:E1-E7. Fairfield KM, et al. Int J Cancer. 2004;110: , Norquist et al, JAMA Oncology 2016;2:

7 Hereditary Ovarian Cancer Increasingly recognized that hereditary mutations may contribute to ovarian cancer risk Most commonly associated genes: BRCA1 and BRCA2 BRCA1 associated with ~40% lifetime risk of ovarian cancer BRCA2 associated with ~15% lifetime risk of ovarian cancer 9 other genes have been identified as high risk: MSH2, MLH1, PMS2, MSH6, PALB2, BARD1, RAD51C, RAD51D, and BRIP1 ~10-15% of ovarian cancer may have BRCA1 or BRCA2 mutations Up to 40% of patients with germline BRCA1/2 mutations may not have a suggestive family history All women with a diagnosis of ovarian cancer are recommended to have genetic counseling regarding germline high risk gene testing: panel testing is preferred SGO guidelines, NCCN genetics guidelines Norquist et al, JAMA Oncology 2016;2:

8 Screening and Diagnosis Why is ovarian cancer so deadly? No effective screening tests currently exist for ovarian cancer Women often diagnosed with advanced-stage disease Ovarian cancer, especially high grade serous histology, is very sensitive to platinum but majority of women recur with increasingly treatment-refractory cancer Symptoms are usually non-specific Abdominal bloating, pain; other GI complaints Shortness of breath Urinary tract symptoms (frequency, urgency) Work-up for suspected ovarian cancer Transvaginal ultrasound CT scans are useful to evaluate upper abdomen, detect ascites/pleural effusions, and help ascertain appropriateness for upfront surgery Cancer markers: CA125, CEA, CA19-9 Pathology: biopsy or surgical specimen needed for diagnosis

9 Can screening help us detect ovarian cancer sooner? PLCO (JAMA 2011) # of women entered 78, ,638 Patient population Ages 55 to 74 Ages 50 to 74 UK Collaborative Trial of Ovarian Cancer screening (UKCTOCS) (Lancet 2016) Randomized Groups 1) Annual screening (39,105): Yearly CA125 for 6 years and TVU for 4 years Results 2) No screening (39,111) 1) Ovarian cancer stage distributions same for both groups 2) No reduction in mortality in the screened group 1) ROCA (50,640) [divides pts into yearly follow-up, rpt CA125 in 3 mos or abnormal and repeat CA125/TVU needed] 2) TVU (50,639) [divided pts into normal, unsatisfactory with rpt in 3 mos, and abnormal with rpt in 6 weeks 3) No screening (101,359) 1) Lower stage cancers (I, II and IIIA) diagnosed more frequently in ROCA group compared to no screening 2) Ovarian cancer diagnosed in 0.6% in #2 and #3 and 0.7% in #1 2) Mortality was not reduced in any groups TVU = transvaginal ultrasound ROCA = risk of ovarian cancer algorithm To date, no effective screening mechanism for ovarian cancer has been identified.

10 The fallopian tube is a site of origin for some ovarian cancers Many ovarian cancers may start in the fimbriae of the fallopian tube Up to 50% of cancers in BRCA mutation carriers at prophylactic surgery originate in fallopian tube Research has shown a potential evolution from precursor lesion to invasive carcinoma Trials now in progress to investigate whether removal of fallopian tubes can prevent ovarian cancer in high-risk women Standard of care in U.S. remains oophorectomy/fallopian tube removal Jones and Drapkin, Frontiers in Oncology, 2013

11 Classification of ovarian cancer is evolving to include molecular data Serous Mucinous Endometrioid Clear cell

12 Classification of ovarian cancer is evolving to include molecular data High grade serous or endometrioid Low grade endometrioid Low grade serous Clear Cell Mucinous Genetic characteristics Up to 50% with alterations in HR Associated with TP53 and BRCA mutations PTEN, ARID1A, PIK3CA alterations May have MSI KRAS, BRAF mutations PIK3CA, ARID1A, PTEN KRAS Clinical characteristics Increased plat-sensitivity PARP inhibitors with potential activity in HRD tumors Potentially more responsive to hormonal therapy, although not established Hormonal therapies Potentially MEK inhibitors Often resistant to initial plat-based therapy Targeted or immunooncology agents being explored Often chemotherapyinsensitive

13 Newly diagnosed ovarian cancer management: 2018 Basic principles Surgery by a gynecologic oncologist is associated with improved survival 1 Extent of cytoreduction (debulking) is classified by residual disease following surgery Suboptimal: >1 cm of residual tumor Optimal: 1 cm of residual tumor NED (R0): No gross residual disease Considerations in the approach to initial disease treatment Role of neoadjuvant chemotherapy Dose-dense versus every 3 week treatment (and weekly treatment for all agents) Intraperitoneal chemotherapy 1 Cliby WA, Gynecol Oncol :11-7

14 Neoadjuvant Chemotherapy Upfront surgery Vergote et al, NEJM 2009 Kehoe et al, NEJM 2015 Upfront chemotherapy (NACT) Upfront surgery # of patients Upfront chemotherapy (NACT) Stage IIIC Stage IV 257 (75.6%) 77 (22.9%) 253 (75.7%) 81 (24.3%) 206 (75%) 70 (25%) 206 (75%) 68 (25%) Median PFS 12 months 12 months 12 months 10.7 month Median OS 29 months 30 months 22.6 months 24.1 months Death within 28 days post surgery 8 (2.5%) 2 (0.7%) 14 (6%) 1 (<1%) Venous thromboembolism 8 (2.5%) 0 (0%) 5 (2%) 0 (0%) Recent data showed that increased rates of use neoadjuvant chemotherapy led to survival improvement compared to regions that did not increase their uptake of neoadjuvant chemotherapy (Melamed et al, BMJ 2018)

15 Dose-dense Chemotherapy Study Study Arms Median PFS, Mos Median OS, Mos JGOG ,2 Carboplatin/paclitaxel every 21 days Carboplatin every 21 days/paclitaxel weekly (dose dense) GOG Carboplatin/paclitaxel every 21 days +/- bevacizumab Carboplatin every 21 days/paclitaxel weekly +/- bevacizumab GOG (no bevacizumab patients; 16% of total trial patients) Carboplatin/paclitaxel every 21 days 10.3 NR Carboplatin every 21 days/paclitaxel weekly 14.2 NR ICON8 4 Carboplatin/paclitaxel every 21 days Carboplatin every 21 days/paclitaxel weekly Weekly carboplatin and weekly paclitaxel Katsumata et al., Lancet 2009; 2 Katsumata et al., Lancet Oncol Chan et al., N Engl J Med 2016, 4 ICON8, ESMO 2017

16 Intraperitoneal Chemotherapy Chemotherapy delivered directly into the abdominal space Higher concentration of drug delivery to cells in the abdominal space Only effective if most of the disease burden has been surgically removed (i.e., optimally cytoreduced) Overall survival benefit initially demonstrated in 3 large studies NCI issued alert in 2006 stating that IP chemotherapy should be considered the preferred regimen for treatment ***GOG252 has raised question of optimal setting for IP chemotherapy use; lowered dose of cisplatin to 75 mg/m 2 has same PFS as IV chemotherapy SGO 2016

17 Summary of Intraperitoneal Chemotherapy Studies Study Study Arms Median PFS, Mos Median OS, Mos SWOG 8501/GOG Cisplatin IV 100mg/m2 + IV Cyclophosphamide 600mg/m Cisplatin IP 100mg/m2 + IV Cyclophosphamide 600mg/m GOG IV Paclitaxel 135mg/m2 x 24 hrs (D1) + IV Cisplatin 75mg/m2 (D2) IV Carboplatin AUC9 x 2 cycles IV Paclitaxel 135mg/m2 x 24 hrs (D1) + IP Cisplatin 100mg/m2 (D2) GOG IV Paclitaxel 135mg/m2 x 24 hrs (D1) + IV Cisplatin 75mg/m2 (D2) IV Paclitaxel 135mg/m2 x 24 hrs (D1) + IP Cisplatin 100mg/m2 (D2) + IP Paclitaxel 60mg/m2 (D8) GOG IV Paclitaxel 80mg/m2 (D1,8,15) + IV Carboplatin AUC6 (D1) + Bevacizumab 15mg/kg (D1 from Cycle 2 onwards) HIPEC 5 (all pts received neoadjuvant chemo) IV Paclitaxel 80mg/m2 (D1,8,15) + IP Carboplatin AUC6 (D1) + Bevacizumab 15mg/kg (D1 from Cycle 2 onwards) IV Paclitaxel 135mg/m2 x 3 hrs (D1) + IP Cisplatin 75mg/m2 (D2) + IP Paclitaxel 60mg/m2 (D8) + Bevacizumab 15mg/kg (D1 from Cycle 2 onwards) NR 28.7 NR 27.8 NR No HIPEC added to carboplatin/paclitaxel Heated cisplatin (HIPEC) after 3 cycles of neoadjuvant carbo/paclitaxel Alberts et al., N Engl J Med 2006; 2 Markman et al., J Clin Oncol Armstrong et al., N Engl J Med 2006; 4 Walker et al., 2016 SGO Meeting 5 van Driel et al, N Engl J Med 2018

18 Chemotherapy for newly-diagnosed Ovarian Cancer Neoadjuvant chemotherapy can result in similar outcomes to upfront surgery and adjuvant chemotherapy Defer to gynecologic oncology surgeon regarding when neoadjuvant chemotherapy is appropriate Regimens that can be considered for upfront therapy Carboplatin and paclitaxel q3 weeks Carboplatin q3weeks and paclitaxel weekly (dose-dense) Weekly carboplatin and weekly paclitaxel Intraperitoneal chemotherapy although exact benefit margin, best regimen, and appropriate population unclear; HIPEC cisplatin shows OS benefit in one phase 3 study Note: Addition of bevacizumab to carboplatin and paclitaxel results in PFS but no OS benefit 1,2, and is not approved by FDA for use in this setting though is currently under review by the FDA 1 Burger et al, NEJM 2011;365: Perren et al, NEJM 2011;365:

19 Recurrent Ovarian Cancer 2018 The majority of patients with advanced stage ovarian cancer will recur Recurrences (in US) most often detected by rising CA125 Follow-up typically every 3 months for first 2 years, every 6 months for next 3 years following initial treatment Recurrences defined by time since initial platinum-based therapy Platinum-refractory: failed to achieve at least a partial response to therapy/growth on platinum Platinum-resistant: recurrence within 6 months of last platinum-based therapy Platinum-sensitive: recurrence more than 6 months after last platinum-based therapy Recent advances Bevacizumab in conjunction with chemotherapy PARP inhibitors Salani R, Am J Obstet Gynecol. 2011;204:466-78

20 Current recommended therapies for recurrent ovarian cancer (NCCN Clinical Practice Guidelines Preferred Agents) Platinum-sensitive disease Carboplatin Carboplatin/docetaxel Carboplatin/gemcitabine Carboplatin/gemcitabine/ bevacizumab (category 2B) Carboplatin/liposomal doxorubicin (category 1) Carboplatin/paclitaxel (category 1) Carboplatin/paclitaxel (weekly) Cisplatin Cisplatin/gemcitabine Targeted therapy Bevacizumab Olaparib Platinum-resistant disease Docetaxel Oral etoposide Gemcitabine Liposomal doxorubicin Liposomal doxorubicin/ bevacizumab Paclitaxel (weekly) Paclitaxel (weekly)/bevacizumab Topotecan Topotecan/bevacizumab Targeted therapy Bevacizumab Olaparib NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines ): Ovarian Cancer including Fallopian Tube Cancer and Primary Peritoneal Cancer, version

21 Deciding on the appropriate recurrence regimen Toxicities from prior therapies CA125 marker elevation only? Evidence of symptoms? Availability of clinical trials/suitability for clinical trials Patient convenience and choice Platinum-sensitivity Platinum-based regimens not used in primary platinum-refractory cancers Platinum-based doublets not clearly beneficial in platinum-resistant cancers Presence of somatic or germline BRCA1/2 mutation Potential for incorporating PARP inhibitors Allergies to chemotherapy drugs (paclitaxel, platinum)

22 Use of Anti-angiogenics in Recurrent Ovarian Cancer Multiple anti-angiogenic agents have demonstrated single-agent activity in recurrent ovarian cancer in Phase 2 trials Response rate 10-20% Progression-free survival up to 4-5 months Combining anti-angiogenics with chemotherapy followed by maintenance therapy in newly diagnosed ovarian cancer with PFS benefit of 2-4 months, no OS benefit Not approved in the U.S. for this indication Multiple trials have explored combining anti-angiogenics with chemotherapy in recurrent ovarian cancer Platinum-sensitive: OCEANS, GOG213 Platinum-resistant: AURELIA

23 Bevacizumab + platinum-based chemotherapy improves PFS in platinum-sensitive recurrent ovarian cancer OCEANS GOG-213 Study Regimen(s) PFS OS OCEANS 1 (plat-sens) GOG213 2 (plat-sens) PFS OS PFS OS (A) Carboplatin/gemcitabine (CG) (B) CG/bevacizumab + bevacizumab maintenance (A) Carboplatin/paclitaxel (CP) (B) CP/bevacizumab + bevacizumab maintenance 8.4 mos 12.4 mos (p=0.0001) 10.4 mos 13.8 mos (p<0.0001) 35.2 mos 33.3 mos (p=0.65) 37.3 mos 42.2 mos (p=0.056) Aghajanian et al., J Clin Oncol 2012; Coleman et al., SGO 2015 and Lancet Oncology 2017

24 Bevacizumab + platinum-based chemotherapy improves PFS in platinum-sensitive recurrent ovarian cancer OCEANS GOG-213 Study Regimen(s) PFS OS OCEANS 1 (plat-sens) GOG213 2 (plat-sens) PFS OS PFS OS (A) Carboplatin/gemcitabine (CG) (B) CG/bevacizumab + bevacizumab maintenance 8.4 mos 12.4 mos 35.2 mos 33.3 mos Bevacizumab (A) Carboplatin/paclitaxel approved (CP) by FDA in December 10.4 mos2016 in conjunction with carboplatin/paclitaxel (B) CP/bevacizumab + or bevacizumab carboplatin/gemcitabine maintenance 13.8 mos chemotherapy (p<0.0001) in recurrent platinum-sensitive ovarian cancer 37.3 mos 42.2 mos (p=0.056) Aghajanian et al., J Clin Oncol 2012; Coleman et al., SGO 2015 Aghajanian et al., J Clin Oncol 2012; Coleman et al., SGO 2015 and Lancet Oncology 2017

25 Bevacizumab + chemotherapy in recurrent platinum-resistant ovarian cancer: the AURELIA trial Stratified by chemotherapy, PFI (< 3 mos vs 3-6 mos), prior anti-angiogenesis Pts with measurable OC that progressed < 6 mos from platinum-based chemotherapy; 2 prior therapies; no bowel involvement (N = 361) Nonplatinum Chemotherapy Nonplatinum Chemotherapy + Bevacizumab Disease progression Chemotherapy options: Paclitaxel 80 mg/m 2 on Days 1,8,15, 22 every 28 days Topotecan 4 mg/m 2 on Days 1, 8,15 every 28 days or Topotecan 1.25 mg/m 2 on Days 1-5 every 21 days Pegylated liposomal doxorubicin (PLD) 40 mg/m 2 on Day 1 every 28 days Pujade-Lauraine et al., J Clin Oncol 2014 and 2015

26 Bevacizumab + chemo improved PFS and QoL in platinumresistant ovarian cancer Patient-reported (QOL) outcome benefit seen in patients with pre-existing abdominal symptoms Bevacizumab approved by FDA in December 2014 in conjunction with chemotherapy in platinumresistant ovarian cancer Pujade-Lauraine et al., J Clin Oncol 2014 and 2015

27 PARP inhibitors display synthetic lethality in the setting of homologous recombination loss Silver and Iglehart, NEJM, 2009

28 PARP inhibitors are active agents in ovarian cancer and likely work via multiple mechanisms of action Up to 50% of high-grade serous ovarian cancers with alterations in homologous recombination (HR) genes Clinical activity of PARP inhibitors in multiple settings in ovarian cancer BRCA mutated (somatic or germline) relapsed disease BRCA non-mutated relapsed disease Maintenance therapy Konstantinopoulos et al., Cancer Discov 2015

29 PARP inhibitors have activity in relapsed BRCA-mutated ovarian cancer Olaparib approved by FDA in December 2014 for relapsed germline BRCA mutated ovarian cancer after 3 or more prior therapies. Rucaparib approved by FDA in December 2016 for relapsed germline or somatic BRCA mutated ovarian Olaparib 1 Rucaparib 2 Niraparib 3 Veliparib 4 No of pts # of lines of prior therapy At least 3 prior lines At least 2 prior lines (43% had 3 or more) NA 1 prior: 32% 2 or 3: 68% Objective RECIST RR 34% 54% (IRR 5 42%) 40% 26% RR based on platinum sensitivity NA for this population; other data available Plat sens 66% Plat resist 25% Plat refract 0% Plat sens 50% Plat resist 33% Plat refract 0% Plat sens 35% Plat resist 20% Median Duration of response 7.4 mos 9.2 mos (IRR months) 12.9 mos 8.18 mos (reported as median PFS) 1 Olaparib FDA package insert; 2 Rucaparib FDA package insert; 3 Sandhu, Lancet Oncol 2013; 4 Coleman, Gyn Onc 2015, 5 IRR = independent radiology review

30 PARP inhibitor toxicities Gastrointestinal toxicities include: Nausea Vomiting Decreased appetite Diarrhea Constipation Abdominal pain Dyspepsia Bone marrow suppression: Drop in RBC, WBC, platelets, and small risk ~1-2% risk of AML and/or MDS Others: Fatigue HTN, tachycardia, palpitations (niraparib) Liver function test elevations (rucaparib) Serum creatinine elevation (olaparib, rucaparib) FDA PI for niraparib, olaparib and rucaparib

31 PARP inhibitors have equivalent activity in germline and somatic BRCA-mutated ovarian cancers ARIEL2 trial demonstrated similar response rate in tumors with germline and somatic BRCA mutations Data from additional trials of PARP inhibitors demonstrate similar efficacy of both niraparib and olaparib between somatic and germline BRCAmutated tumors Swisher et al, Lancet Oncology 2017

32 What predicts likelihood of response to PARP inhibitors? Presence of a BRCA mutation (either germline or somatic) Platinum sensitivity Number of prior lines of treatment Presence of homologous recombination deficiency Tests to look at markers of genomic scarring have some correlation with PARP inhibitor activity Multiple problems exists with these tumor tests Olaparib response by plat-sensitivity and # of prior lines Matulonis, Ann Oncol 2016

33 PARP inhibitors as maintenance therapy Ovarian cancer (newly diagnosed or recurrent) with high relapse rate Therapies with low toxicity profile that can delay recurrence and/or improve survival of greatest interest Study 19 was a randomized phase II study that showed benefit of maintenance therapy post platinum response 1 3 Phase 3 studies of PARP inhibitor maintenance with evidence of PFS benefit ENGOT-OV16/NOVA (niraparib) 2 SOLO2/ENGOT-OV21 (olaparib) 3 ARIEL3 (rucaparib) 4 1 Study 19, NEJM NOVA NEJM SOLO2 Lancet Oncology ARIEL3, Lancet 2017

34 ENGOT-OV16/NOVA Trial KEY INCLUSION CRITERIA Histologically diagnosed ovarian cancer Predominantly high-grade serous Completed at least two previous courses of platinum-containing therapy Platinum-sensitive to the penultimate platinum regimen, and remain in response to platinum Mirza, N Engl J Med 2016

35 ENGOT-OV16/NOVA Trial Schema Platinum-Sensitive Recurrent High-Grade Serous Ovarian Cancer Treatment with at least 4 Cycles of Platinum-based Therapy Response to Platinum Treatment gbrcamut (N = 203) Non-gBRCAmut (N = 350) 2:1 Randomization 2:1 Randomization Niraparib 300 mg once daily Placebo Niraparib 300 mg once daily Placebo Treat until Progression of Disease Treat until Progression of Disease Primary Endpoint: PFS by central, blinded review: results for both gbrca and non-gbrca groups analyzed simultaneously Mirza, N Engl J Med 2016

36 ENGOT-OV16/NOVA Primary Analysis demonstrates PFS improvement across all patients subsets Primary Efficacy Populations gbrcamut Non-gBRCAmut HRDpos Non-gBRCAmut Overall PFS (%) PFS (%) PFS (%) Median PFS (months) Months Since Randomization HR=0.27 Niraparib:21.0 Placebo: Months Since Randomization HR=0.38 Niraparib:12.9 Placebo: Months Since Randomization HR=0.45 Niraparib: 9.3 Placebo: 3.9 Niraparib Placebo Mirza, N Engl J Med 2016

37 Niraparib effect on PFS in patients without germline BRCA mutations most notable in HRD-positive subset Exploratory Analyses HRD-positive HRD-negative sbrcamut BRCAwt PFS (%) PFS (%) PFS (%) Months Since Randomization Months Since Randomization Months Since Randomization HR=0.27 HR=0.38 HR=0.58 Median PFS (months) Niraparib: 20.9 Placebo: 11.0 Niraparib: 9.3 Placebo: 3.7 Niraparib: 6.9 Placebo: 3.8 Niraparib Placebo Mirza, N Engl J Med 2016

38 SOLO2/ENGOT-OV21 Study Schema Patients BRCA1/2 mutation Platinum-sensitive relapsed ovarian cancer At least 2 prior lines of platinum therapy CR or PR to most recent platinum therapy Randomized 2:1 Olaparib 300 mg bid n=196 Placebo n=99 Primary endpoint Investigator-assessed PFS Pujade-Lauraine, SGO 2017

39 SOLO2/ENGOT-OV21 confirms olaparib maintenance significantly improves PFS in BRCA-mutated platinum-sensitive ovarian cancer PFS by investigator assessment Progression-free survival (%) No. at risk Olaparib Placebo Months since randomization Median follow-up was 22.1 months in the olaparib group and 22.2 months for placebo Olaparib Placebo 29 6 Olaparib (n=196) 3 0 Placebo (n=99) Events (%) 107 (54.6) 80 (80.8) Median PFS, months HR % CI 0.22 to 0.41 P< August 2017: Olaparib is now also approved by US FDA for maintenance therapy in ovarian cancer patients with recurrent cancer and who are responding to platinum Pujade-Lauraine, based SGO 2017 Pujade-Lauraine, SGO 2017 chemotherapy. New FDA approval came also with a change in formulation using tablets at

40 ARIEL3: STUDY DESIGN Patient eligibility Stratification High-grade serous or endometrioid epithelial OC, primary peritoneal, or fallopian tube cancers Sensitive to penultimate platinum Responding to most recent platinum (CR or PR)* Excludes patients without assessable disease following second surgery CA-125 within normal range No restriction on size of residual tumour ECOG PS 1 No prior PARP inhibitors Randomisation 2:1 HRR status by NGS mutation analysis BRCA1 or BRCA2 Non-BRCA HRR gene None of the above Response to recent platinum CR PR Progression-free interval after penultimate platinum 6 to <12 months 12 months Rucaparib 600 mg BID n=375 Placebo BID n=189 *CR (defined by RECIST v1.1) or PR (defined by RECIST v1.1 and/or a GCIG CA-125 response [CA-125 within normal range]) maintained until entry to ARIEL3 ( 8 weeks of last dose of chemotherapy). ATM, ATR, ATRX, BARD1, BLM, BRIP1, CHEK1, CHEK2, FANCA, FANCC, FANCD2, FANCE, FANCF, FANCG, FANCI, FANCL, FANCM, MRE11A, NBN, PALB2, RAD50, RAD51, RAD51B, RAD51C, RAD51D, RAD52, RAD54L, RPA1. 40 HRR, homologous recombination repair; NGS, next-generation sequencing. Lancet 2017

41 ARIEL3: Investigator-Assessed Progression-Free Survival BRCA mutant HRD ITT Median (months) 95% CI Rucaparib (n=130) Placebo (n=66) HR, 0.23; 95% CI, ; P< Median (months) 95% CI Rucaparib (n=236) Placebo (n=118) HR, 0.32; 95% CI, ; P< Median (months) 95% CI Rucaparib (n=375) Placebo (n=189) HR, 0.36; 95% CI, ; P< At risk (events) Rucaparib 130 (0) 93 (23) 63 (46) 35 (58) 15 (64) 3 (67) 0 (67) Placebo 66 (0) 24 (37) 6 (53) 3 (55) 1 (56) 0 (56) Rucaparib, 48% censored Placebo, 15% censored At risk (events) Rucaparib 236 (0) 161 (55) 96 (104) 54 (122) 21 (129) 5 (134) 0 (134) Placebo 118 (0) 40 (68) 11 (95) 6 (98) 1 (101) 0 (101) Rucaparib, 43% censored Placebo, 14% censored Visit cutoff date: 15 April At risk (events) Rucaparib 375 (0) 228 (111) 128 (186) 65 (217) 26 (226) 5 (234) 0 (234) Placebo 189 (0) 63 (114) 13 (160) 7 (164) 2 (167) 1 (167) 0 (167) Rucaparib, 38% censored Placebo, 12% censored Lancet 2017

42 PARP inhibitors are also in testing for newly diagnosed ovarian cancer Study Study Arms CT.gov # SOLO1 (BRCAm only) N=397 GOG 3005 (BRCAm and wildtype) N=1100 PAOLA1 N=612 PRIMA (BRCAm and wildtype) N=330 Olaparib maintenance Placebo maintenance Carboplatin/paclitaxel/placebo + placebo maint. Carboplatin/paclitaxel + veliparib + placebo maint. Carboplatin/paclitaxel + veliparib + veliparib maint. Plat/taxane chemotherapy/bev, then bev maintenance Plat/taxane chemotherapy/bev, then bev/olaparib maintenance Niraparib maintenance Placebo maintenance NCT NCT NCT NCT

43 New Agents and Strategies for Ovarian Cancer: 2018 Antibody drug conjugates -Mirvetuximab sorvanstine 1 : anti-folate receptor α ADC, phase I/II (novel combinations with bev and IO) and phase III (FORWARD1 study) PARP inhibitor combinations Olaparib and Cediranib (anti-vegfr TKI)- now in phase III studies Multiple others (PD1, HSP90, ATR, and others) Immunotherapy Multiple trials ongoing and planned adding checkpoint blockade to chemotherapy for newly diagnosed and recurrent ovarian cancer; novel IO and biologic combinations are also being tested 1 Moore et al, JCO 2017

44 Summary: Ovarian Cancer 2018 Remains a leading cause of cancer death in women in the U.S. No effective screening test has been identified Growing recognition of importance of molecular classifications of ovarian cancer Hereditary ovarian cancer increasingly recognized All women with a diagnosis of ovarian cancer should have genetic counseling Upfront therapy for ovarian cancer continues to evolve Recent advances in recurrent ovarian cancer Bevacizumab in combination with chemotherapy Approved in platinum-sensitive and platinum-resistant recurrent setting PARP inhibitors for recurrent ovarian cancer Approved in women with relapsed BRCA-mutated ovarian cancer (rucaparib for 3 rd line; olaparib for 4 th line) Approved as maintenance therapy in platinum-sensitive recurrence after response to platinum (niraparib and olaparib approved)