Articles. Funding Newcastle University and Bloodwise.

Transcription

1 De-escalation of tyrosine kinase inhibitor dose in patients with chronic myeloid leukaemia with stable major molecular response (DESTINY): an interim analysis of a non-randomised, phase 2 trial Richard E Clark, Fotios Polydoros, Jane F Apperley, Dragana Milojkovic, Christopher Pocock, Graeme Smith, Jenny L Byrne, Hugues de Lavallade, Stephen G O Brien, Tony Coffey, Letizia Foroni, Mhairi Copland Summary Background Discontinuation of tyrosine kinase inhibitor (TKI) therapy is feasible for some patients with chronic myeloid leukaemia with deep molecular responses; however, patients with stable major molecular response (MMR), but not MR4, have not been studied, nor has the effect of treatment de-escalation rather than outright cessation. We aimed to examine the effects of treatment de-escalation as a prelude to complete cessation, not only in patients with MR4 or greater, but also in those with MMR but not MR4. Methods We did this interim analysis of a non-randomised, phase 2 trial at 2 hospitals in the UK. We recruited patients (aged 18 years) with chronic myeloid leukaemia in first chronic phase who had received TKI for 3 years or more and were either in stable MR4 (BCR-ABL1:ABL1 ratio < 1%; MR4 cohort) or in stable MMR (BCR-ABL1:ABL1 ratio consistently < 1%) but not MR4 (MMR cohort) for 12 months or longer. Participants received half their standard TKI dose (imatinib 2 mg daily, dasatinib 5 mg daily, or nilotinib 2 mg daily) for 12 months. Molecular recurrence was defined as loss of MMR (BCR-ABL1:ABL1 ratio > 1%) on two consecutive samples. The primary endpoint of this interim analysis was the proportion of patients who lost MMR on de-escalation and regained MMR on TKI resumption. Analyses were by intention to treat. This study is registered with ClinicalTrials.gov, number NCT Findings Between Dec 16, 213 and April 1, 215, we enrolled 174 patients into the MMR cohort (n=49) or the MR4 cohort (n=125). During the 12 months of half-dose therapy, 12 patients (7%) had molecular recurrence, all of whom regained MMR within 4 months of full-dose TKI resumption (median time to recovery 77 days). Recurrence was significantly lower in the MR4 cohort (three [2%; 9% CI 2 4 8] of 121 evaluable patients) than in the MMR cohort (nine [19%; 9% CI ] of 48 evaluable patients; hazard ratio 12, 9% CI 4 37; p= 7), but was unrelated to previous TKI or TKI therapy duration. Adverse events (eg, lethargy, diarrhoea, rash, and nausea) improved during the first 3 months of de-escalation, though not thereafter. 16 serious adverse events were reported, including one fatality due to worsening pre-existing peripheral arterial occlusive disease in a patient who had received only imatinib. Interpretation TKI de-escalation is safe for most patients with excellent responses to TKI therapy, and is associated with improvement in symptoms. These findings show that lower TKI doses might maintain responses in these patients, implying that such patients could be unnecessarily overtreated. Studies of more ambitious de-escalation are warranted. Funding Newcastle University and Bloodwise. Introduction The advent of daily tyrosine kinase inhibitor (TKI) therapy for chronic myeloid leukaemia has transformed the outlook for patients with this disease, most of whom can now expect to live a normal lifespan. 1 However, the adverse effects 2 and cost 3 of TKI therapy have led specialists and patients to ask whether the disease might be sufficiently suppressed after a few years of therapy to permit treatment discontinuation. Several studies 4,5 have established that some patients with enduring deep molecular responses to TKI therapy (defined as a BCR-ABL1:ABL1 ratio consistently < 1%, also called MR4 [molecular response 4 logs below the standard arbitrary baseline]) can discontinue treatment. Early studies 6 8 were confined to patients with BCR-ABL1 transcripts that were undetectable by standard RT-PCR, and molecular recurrence was defined as the reappearance of BCR-ABL1 transcript positivity. More recent studies 9 11 have specified recurrence as the loss of major molecular response (MMR; defined as a BCR-ABL1:ABL1 transcript ratio of > 1%, also called MR3 [molecular response of 3 logs below an arbitrary standard baseline]). With this definition of recurrence, 24 months after TKI cessation the A-STIM 9 and KID 1 studies reported that 58 64% of patients were recurrence-free, and the large EUROSKI study 11 of 868 patients reported a similar rate of 52% at the Lancet Haematol 217; 4: e31 16 Published Online May 26, S (17)366-2 See Comment page e33 Institute of Translational Medicine (Prof R E Clark MD) and Cancer Research UK Liverpool Cancer Trials Unit (F Polydoros MSc, T Coffey), University of Liverpool, Liverpool, UK; Centre for Haematology, Imperial College London at Hammersmith Hospital, London, UK (Prof J F Apperley MD, D Milojkovic PhD, Prof L Foroni PhD); East Kent Hospitals, Canterbury, UK (C Pocock PhD); Department of Haematology, St James s Hospital, Leeds, UK (G Smith MD); Faculty of Medicine and Health Sciences, University of Nottingham, Nottingham, UK (J L Byrne PhD); Department of Haematological Medicine, Kings College Hospital, London, UK (H de Lavallade MD); Northern Institute for Cancer Research, Newcastle University, Newcastle, UK (Prof S G O Brien PhD); and Paul O Gorman Leukaemia Research Centre, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK (Prof M Copland PhD) Correspondence to: Prof Richard E Clark, Department of Haematology, Royal Liverpool University Hospital, Liverpool L7 8XP, UK clarkre@liverpool.ac.uk Vol 4 July 217 e31

2 Research in context Evidence before this study Discontinuation of previous lifelong tyrosine kinase inhibitor (TKI) therapy is of interest to patients with chronic myeloid leukaemia and their clinical attendants. The seminal trial of stopping imatinib (STIM) was published in The Lancet Oncology in 21. Our literature search of PubMed, using no language restrictions and the terms stopping TKI/therapy with CML or chronic myeloid leukaemia, and which also included abstracts of the American Society of Hematology and the European Haematology Association up to Nov 3, 216, identified results from nine further studies (STIM2, ALLG CML8, A-STIM, ISTAV, EUROSKI, STOP 2G TKI pilot, ENESTfreedom, ENESTop, and DADI), some interim and some only in abstract form so far, reporting recurrence-free rates of 45 64%. However, all these studies have been confined to the very best TKI responders namely, patients with stable MR4 (molecular response 4 logs below the standard arbitrary baseline, also known as a deep molecular response) (BCR-ABL1:control gene ratio of 1%). None of these studies have included patients with quite good, but not perfect, major molecular responses (MMR; but not MR4 ie, ratios of 1%, but not always below 1%), even though such patients might also have a normal life expectancy and for whom the notion of stopping treatment is of equal interest. Furthermore, no studies have examined whether treatment dose de-escalation rather than stopping might be feasible for a greater number of patients than outright stopping. Added value of this study To our knowledge, the DESTINY study is the first to include patients with chronic myeloid leukaemia in MMR but not MR4, in addition to patients in stable MR4. Here we report the effect of de-escalation of TKI to half the standard dose for a period of 12 months, a strategy not previously investigated. Implications of all the available evidence We present novel data showing that treatment de-escalation is feasible and safe for almost all (98%) patients in stable MR4, and also for 81% of patients in stable MMR but not MR4. In line with previous studies, we confirm that patients who undergo molecular recurrence are rapidly restored to good molecular control on resumption of full-dose TKI. Treatment de-escalation also improves adverse events and saves money. The data imply for the first time that such patients could be unnecessarily overtreated. For the trial protocol see DESTINY_Protocol.pdf See Online for appendix same timepoint. However, these studies were confined to patients in stable MR4 at entry. Although anecdotal reports exist of successful treatment cessation for a few months in patients with stable MMR, but not MR4 eg, during pregnancy 12 such patients have not been formally studied in a stopping trial. Additionally, dose reduction in patients with good responses to TKI therapy, but also adverse events, can ameliorate these events yet maintain deep molecular response. 2 Therefore, we aimed to investigate whether some patients who might have molecular recurrence on stopping TKI might be able to safely decrease the dose without an increase in tumour burden. We did the DESTINY study to examine the effects of treatment de-escalation as a prelude to complete cessation, not only in patients with deep molecular responses of MR4 or greater, but also in those with MMR but not MR4. The cessation phase is ongoing. Methods Study design and participants We did this interim analysis of a non-randomised, phase 2 trial at 2 hospitals in the UK (appendix). We included patients aged 18 years or older with positive BCR-ABL1 transcripts, with either an e13a2, e14a2, or e19a2 fusion transcript, who were in the first chronic phase of chronic myeloid leukaemia and had either received the same TKI (imatinib, dasatinib, or nilotinib) since diagnosis or had switched only once for intolerance to the initial drug. We excluded patients who showed resistance to previous TKIs and switched to another TKI. Previous interferon treatment was not an exclusion criterion as long as it had finished at least 12 months before entry; Philadelphia chromosome positivity was not mandatory. Participants must have received TKI for at least 3 years, and all PCR tests (minimum of three) in the 12 months before trial entry must have been 1% or lower (ie, MMR), each with 1 or more ABL1 control transcripts. Patients who had previously received more than 4 mg daily of imatinib, 1 mg daily of dasatinib, or 4 mg daily of nilotinib were ineligible, unless their high dose arose from participation in a previous clinical trial in which high doses were being compared with standard doses. Recipients of bosutinib or ponatinib at any point were ineligible. All entrants provided written informed consent, and the trial was done in line with the Declaration of Helsinki, sponsored jointly by the University of Liverpool, and Royal Liverpool and Broadgreen University Hospitals Trust (Liverpool, UK). Ethics approval was granted by the North West Liverpool East Committee of the UK National Research Ethics Service. The trial protocol is available online. Procedures Participants decreased their entry TKI dose to half the standard dose for 12 months: imatinib 2 mg daily, dasatinib 5 mg daily, or nilotinib 2 mg daily. Monitoring was done monthly in a central laboratory at Imperial Molecular Pathology at Hammersmith Hospital, London, UK, and we expressed all BCR-ABL1 ratios according to the international scale. Any result of e311 Vol 4 July 217

3 more than 1% prompted an urgent alert to the site to request a further confirmatory sample, typically done within 2 weeks of the alerting sample. Molecular recurrence was defined as loss of MRR (> 1%) on these two consecutive samples and timed as the date of the first of these samples. In the event of recurrence, all patients were required to resume the standard dose of their entry TKI. We continued molecular monitoring monthly until MMR was reached again. Outcomes The primary endpoint for this interim analysis was the proportion of patients who lost MMR on de-escalation and regained MMR on TKI resumption. The primary endpoint for the DESTINY study is the proportion of patients who can first de-escalate their treatment (to half the standard dose of their TKI) for 12 months, and then stop treatment completely for a further 2 years, without losing MMR, and the results of this endpoint will be reported in a future publication. The secondary endpoints will also be published at a later date. They include molecular relapse-free survival, progression-free survival, overall survival, event-free survival, time to MMR recovery, proportion of patients registered as MMR and MR4 in confirmed MR4.5 before being enrolled in the study, proportion of patients who develop a BCR-ABL1 kinase domain mutation at a level of more than 2% of all BCR-ABL1 transcripts on at least two consecutive occasions, quality of life, health economic assessment, and laboratory studies to define subsets of patients who are more likely to relapse on de-escalation or cessation. Formal adverse event reporting was not attempted, although the presence and severity of TKI-related symptoms were recorded at each monthly visit, both by verbal reporting and by the formal quality-of-life instruments EuroQol 5 dimensions (EQ-5D) 13 and FACT- BRM. 14 Additionally, each patient was asked to complete a diary of symptoms that arose between scheduled visits. Statistical analysis We structured the trial as two parallel cohorts, one comprising patients who were in MR4 for all assessments in the 12 months before entry (MR4 cohort), and the other for patients who were partly or wholly in MMR but not stable MR4 during this time (MMR cohort). Because the MMR cohort could be regarded as more experimental, this arrangement provided a mechanism for the independent Data Monitoring Committee to close that group if its recurrence rate was unacceptably high, without prejudicing the MR4 cohort. We calculated the minimum required sample size (n=168) on the basis of the maximum width of the 9% CI for a wide range of values of the proportion of relapsing patients; we required the maximum width for the smallest group to be smaller than 28. We did all statistical analyses on an intention-to-treat basis with R (version 3.3.1). No adjustment for multiple testing or missing data was incorporated. We estimated the proportion of patients relapsing together with 9% CIs, and survival distribution with Kaplan Meier curves. In the calculation of TKI drug costs, we used the UK National Health Service list price for each TKI dose, with no local discounts or value added tax (currently 2% in the UK). Role of the funding source The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication. Results Between Dec 16, 213, and April 1, 215, 174 patients were enrolled into the MMR cohort (n=49) or the MR4 cohort (n=125; figure 1). At entry, 148 patients (85%) were receiving imatinib, 16 (9%) were receiving nilotinib, and ten (6%) were receiving dasatinib (table 1). Five patients (three receiving imatinib, one receiving nilotinib, and one receiving dasatinib) were already on the half-dose at trial entry because of toxicity at standard or intermediate doses (their entry was not excluded in the protocol). These patients all continued on half-dose treatment and were included in all the analyses. The median duration of TKI therapy was 7 7 years in the MMR cohort and 6 5 years in the MR4 cohort, although this difference 335 patients assessed for eligibility 174 enrolled 49 in MMR cohort 125 in MR4 cohort 1 dropped out 9 relapsed 1 ended study 161 ineligible 7 dropped out 3 relapsed 4 ended study 39 completed de-escalation phase 118 completed de-escalation phase 3 did not proceed to stopping phase 36 proceeded to TKI stopping phase 1 did not proceed to stopping phase 117 proceeded to TKI stopping phase Figure 1: Trial profile MMR=major molecular response (BCR-ABL1:ABL1 ratio consistently < 1%). MR4=deep molecular response (BCR-ABL1:ABL1 ratio < 1%). TKI=tyrosine kinase inhibitor. Vol 4 July 217 e312

4 was not significant; baseline characteristics were otherwise broadly similar (table 1). During the 12 months of half-dose therapy, 12 patients (7%) had molecular recurrence (loss of MMR), all of whom were receiving imatinib (table 2). Molecular recurrence was significantly lower in the MR4 cohort (three [2%] of 121 evaluable patients; 9% CI 2 4 8) than in the MMR cohort (nine [19%] of 48 evaluable patients; ; hazard ratio 12, 9% CI 4 37; p= 7; figure 2A, table 2). Time to relapse was significantly shorter in the MMR cohort than in the MR4 cohort (median 4 4 months vs median 8 7 months; figure 2A). The probability of molecular recurrence on de-escalation was not related to age, sex, weight, performance status, BCR-ABL1 transcript type, or duration of TKI therapy (median 6 9 years overall; table 2). In this regard, no recurrences were observed in the quartile with the shortest duration of previous TKI treatment (<4 8 years), compared with five (12%) of 43 patients in the second quartile ( years), four (9%) of 43 patients in the third quartile ( years), and three (7%) of 44 patients in the fourth quartile ( years). Too few data were available from diagnosis on the components (especially spleen size) of the Sokal, EURO/Hasford, or more recent scoring systems to investigate whether these might predict molecular recurrence. Moreover, too few patients were receiving dasatinib or nilotinib to allow comparison of recurrence rates between imatinib and second-generation TKI recipients. The probability of molecular recurrence was not associated with simple blood-count measures (data not shown). Five patients (3%; one in the MMR cohort and four in the MR4 cohort) did not complete 12 months of deescalation for various reasons (poor protocol adherence [n=2], relocation [n=1], pregnancy [n=1], and intercurrent illness [n=1]). 22 patients entered the trial on lower than usual TKI doses (table 3). Two recurrences (11%) occurred among the 18 patients receiving less than imatinib 4 mg daily, which is similar to the ten recurrences (8%) among the 13 patients (8%) entering on 4 mg daily (table 3). We cannot comment on the effect of lower entrance doses for dasatinib (n=2 for lower than 1 mg daily) or nilotinib (n=2 for lower than 6 mg daily) because no secondgeneration TKI recipient had molecular recurrence. We observed no progression to advanced phase or loss of cytogenetic response. No tyrosine kinase domain mutation was detected at the time of molecular relapse in any of seven patients analysed to date by next-generation sequencing. All 12 patients with molecular recurrence regained MMR within 4 months of resumption of fulldose TKI (median time to recovery 77 days; figure 2B). MMR (n=49) MR4 (n=125) Overall (n=174) Demographic characteristics Age (years) 57 (45 66) 61 (51 68) 59 (5 68) Sex Male 25 (51%) 73 (58%) 98 (56%) Female 24 (49%) 52 (42%) 76 (44%) Physical findings Weight (kg) 79 (7 89) 81 (72 94) 81 (72 92) Missing data 1 1 ECOG performance status 42 (86%) 113 (9%) 155 (89%) 1 7 (14%) 1 (8%) 17 (9%) 2 1 (1%) 1 (1%) 3 1 (1%) 1 (1%) 4 Clinical characteristics BCR-ABL1:ABL1 47 ( 2 9) 1 ( 3 2) 1 ( 6 3) transcript ratio* Medical history Total time on TKI (years) 7 7 ( ) 6 5 ( ) 6 9 ( ) Missing data 1 1 Medication Imatinib 43 (88%) 15 (84%) 148 (85%) Nilotinib 2 (4%) 14 (11%) 16 (9%) Dasatinib 4 (8%) 6 (5%) 1 (6%) Data are median (IQR), n (%), or n. MMR=major molecular response (BCR-ABL1:ABL1 ratio consistently < 1%). MR4=deep molecular response (BCR-ABL1:ABL1 ratio < 1%). ECOG=Eastern Cooperative Oncology Group. TKI=tyrosine kinase inhibitor. *BCR-ABL1 data are the centralised results at trial entry. Table 1: Baseline demographic and clinical characteristics Molecular recurrence Yes (n=12) No (n=162) p value Molecular cohort 7 MMR 9 (75%) 4 (25%).. MR4 3 (25%) 122 (75%).. Sex 77 Male 6 (5%) 92 (57%).. Female 6 (5%) 7 (43%).. ECOG performance status (1%) 143 (88%) (12%).. Age (years) 6 (46 69) 59 (5 68) 84 Weight (kg) 84 (74 9) 8 (71 92) 9 Time on TKI (years) 7 6 ( ) 6 8 ( ) 36 Time in MMR (years) 5 1 ( ) 5 5 ( ) 68 Missing data 1.. BCR-ABL1 transcript type 28 e13a2 5 (42%) 29 (18%).. e14a2 5 (42%) 72 (44%).. Other or both 2 (17%) 31 (19%).. Unknown 3 (19%).. Data are n (%) for categorical variables or median (IQR) for continuous variables, unless otherwise stated. Statistical tests are Fisher s exact test for categorical variables and the Mann Whitney U test for continuous variables. MMR=major molecular response (BCR-ABL1:ABL1 ratio consistently < 1%). MR4=deep molecular response (BCR-ABL1:ABL1 ratio < 1%). ECOG=Eastern Cooperative Oncology Group. TKI=tyrosine kinase inhibitor. Table 2: Effect of different variables on molecular recurrence e313 Vol 4 July 217

5 36 patients (73%) in the MMR cohort and 117 (94%) in the MR4 cohort have proceeded to the ongoing stopping phase of the trial (figure 1). Many patients described symptoms present at trial entry, either in verbal reporting at scheduled visits or in their diaries. Individual side-effects (lethargy, diarrhoea, rash, nausea, periorbital oedema, and hair thinning) all improved in the first 3 months of de-escalation, but not thereafter (appendix). However, little further improvement was observed in the subsequent 9 months (appendix). The appendix provides details of symptoms not present at trial entry that arose during de-escalation. 53 new musculoskeletal symptoms were reported by 36 patients (21%), of which 43 were assessed as grade 1 (not interfering with the patient s usual function), and ten as grade 2 (enough discomfort to interfere with usual activity; appendix). The episodes were described as cramps, arthritis, or musculoskeletal pain. No grade 3 or 4 episodes were recorded (appendix). A similar pattern, albeit at a lower frequency, was observed for other common TKI adverse events (appendix). All 12 recurrences occurred in the 138 patients who did not report any musculoskeletal withdrawal symptoms (ie, recurrence rate 9%). During the course of the trial, 16 serious adverse events were reported (table 4); all were assessed by sites as unrelated to the TKI or the underlying chronic myeloid leukaemia. These serious adverse events included one death due to worsening pre-existing peripheral arterial occlusive disease in a patient who had received only imatinib. Formal quality-of-life assessments by EQ-5D and FACT-BRM were of marginal use because the mean scores for each instrument were similar to that of a healthy control population at trial entry and did not change during de-escalation (data not shown). Savings in drug costs occurred during the 12 months of de-escalation (table 3). In the UK, although the cost of imatinib is directly proportional to the dose used, this is not the case for nilotinib (4 mg daily and 3 mg daily cost the same) or dasatinib (1 mg daily and 8 mg daily cost the same). Among the 12 patients undergoing molecular recurrence, the timing of relapse and of the subsequent resumption of standard TKI dose was variable, which resulted in saved TKI costs varying from 338 to per relapsing patient (table 3). Overall, 46 7% ( ) was saved from an expected TKI budget (without de-escalation) of If considering the MR4 cohort alone, the saving was 47 7% ( from an expected budget of ). Similarly, in the MMR cohort alone, the saving was 44 2% ( from an expected budget of ). Discussion Although several studies of TKI cessation have been reported, little is known about the feasibility of treatment de-escalation in patients with stable molecular responses. Investigators of a single-arm study, 15 commenced in 28, Relapse-free survival (%) Number at risk MMR MR4 MMR recovery (%) Number at risk MMR MR4 A MMR HR 12, 9% CI 4 37; p= 7 MR B Figure 2: Relapse-free survival (A) and time to MMR recovery (B) MMR=major molecular response (BCR-ABL1:ABL1 ratio consistently < 1%). MR4=deep molecular response (BCR-ABL1:ABL1 ratio < 1%). HR=hazard ratio of 76 patients aged 65 years or older who had received imatinib for at least 2 years and in stable complete cytogenetic response (for at least 1 year), examined intermittent imatinib (1 month on alternating with 1 month off). Almost all entrants were also in MMR. With a minimum follow-up of 4 years, 27 patients (36%) lost MMR (and 13 [17%] lost cytogenetic remission), although all patients with adequate follow-up regained MMR within 7 months. 15 In the present study, we show three principal findings. First, with a molecular recurrence rate of 2% after 12 months, de-escalation is clearly safe for patients in stable MR4 or deeper remission. Similarly, because 81% of patients in stable MRR, but not MR4, remained recurrence-free at 12 months, the offer of de-escalation to such patients is also clinically reasonable. The absence of stable MR4 has been suggested to be a red-light warning against treatment cessation. 4 Our study is confined to treatment de-escalation so does not allow us to comment Time (months) Vol 4 July 217 e314

6 Daily dose at trial entry TKI Dose Cohort TKI cost without de-escalation ( ) Time to recurrence (months) Time at half dose (months) Actual TKI cost ( ) TKI cost saved ( ) Patients with molecular recurrence UPN 8 Imatinib 4 mg MMR Imatinib 3 mg MMR Imatinib 4 mg MMR Imatinib 4 mg MMR Imatinib 4 mg MMR Imatinib 3 mg MMR Imatinib 4 mg MMR Imatinib 4 mg MMR Imatinib 4 mg MMR Imatinib 4 mg MR Imatinib 4 mg MR Imatinib 4 mg MR Patients without molecular recurrence Number of patients 12 Imatinib 4 mg Imatinib 35 mg Imatinib 3 mg Imatinib 25 mg Imatinib 2 mg Nilotinib 4 mg Nilotinib 3 mg Nilotinib 225 mg Nilotinib 2 mg Dasatinib 1 mg Dasatinib 8 mg Dasatinib 5 mg All costs are from the UK National Health Service price list, exclusive of value added tax. Costs for imatinib are for commercial Glivec or Gleevec (Novartis, Camberley, UK) before the UK patent expired in December, 216. TKI=tyrosine kinase inhibitor. UPN=unique patient number. Table 3: Financial implications of de-escalation on this suggestion, but does suggest that as long as the patient is in stable MMR, de-escalation could be a reasonable option. Also, our findings cannot be generalised to patients with excellent responses to a TKI given as second-line treatment after initial resistance, since we excluded these patients. Notably, all 12 patients with molecular recurrence were among the 148 patients receiving imatinib, whereas no recurrences were reported in the recipients of second-generation TKI; this difference was not statistically significant. All relapsing patients promptly returned to MMR or better within 4 months of resumption of full-dose TKI; therefore, de-escalation as the standard of care for such patients is plausible. Second, this practice-changing view is reinforced by the general improvement of adverse events in both the Grade 1 Grade 2 Grade 3 Grade 4 Grade 5 Myocardial infarction (2) or syncope (1) MMR MR4 1; n=1 2; n=2 Abdominal pain (1) MMR MR4 1; n=1 Pain in lower limbs (1)* MMR MR4 1; n=1 Gallbladder pain (1) MMR MR4 1; n=1 Allergic reaction (1) MMR MR4 1; n=1 Sepsis (3), urinary tract infection (1), or skin Infection (1) MMR 1; n=1 MR4 2; n=1 2; n=2 Bone pain (1) or other (1) MMR MR4 2; n=2 Urinary retention (1) MMR MR4 1; n=1 Total MMR 1; n=1 MR4 5; n=4 8; n=7 1; n=1 Data in parentheses show number of serious adverse events; other data are number of serious adverse events; n=number of patients. Grading is according to the National Cancer Institute Common Toxicity criteria. *The single fatality. Table 4: Serious adverse events during de-escalation MMR and MR4 cohorts, with only mild (none with grade 2 or worse severity) and transient evidence of the musculoskeletal symptoms that have been described on complete TKI withdrawal. 1,16 Although these and other symptoms generally improved during de-escalation, this association was not of sufficient strength to be detectable by the quality-of-life assessment tools used here, which emphasises their inappropriateness for well controlled patients with chronic myeloid leukaemia in the TKI era. 17,18 Quality-of-life data with a symptom assessment tool more specific to patients with chronic myeloid leukaemia treated with TKI 19,2 would be of interest in future studies of TKI de-escalation or discontinuation. Although the trial protocol required patients with molecular recurrence to resume their TKI at full dose until MMR was re-attained, future studies should assess whether subsequent resumption of reduced-dose treatment without molecular recurrence is a possibility. e315 Vol 4 July 217

7 Finally, de-escalation in the MR4 cohort alone, the MMR cohort alone, or the combined population is associated with a saving of almost half their expected TKI costs. The exact magnitude of these savings is dependent on local base prices, including any taxes, and for imatinib, these prices are currently falling as generic alternatives are introduced. Although we used individual patient TKI consumption data, the data do not take account of the increased PCR monitoring and associated clinical visits that are advisable in patients undergoing de-escalation (or complete cessation). These prices appear very unlikely to have an impact on the impressive savings in TKI costs, though require further study in trials incorporating detailed pharmacoeconomic evaluation of TKI deescalation or cessation. In summary, in patients with chronic myeloid leukaemia with stable MRR or better, a reduction in TKI treatment to half the standard dose appears safe, and is associated with improvement in TKI-related side-effects. This finding implies that many patients with stable responses might be able to maintain their responses on lower TKI doses. De-escalation is also associated with substantial financial savings. Studies of more ambitious de-escalation are warranted. Contributors REC is the chief investigator; contributed to the study design, data collection, and data interpretation; and wrote the manuscript. FP is the study s statistician and contributed generally, but focused on data analysis and co-wrote the manuscript. TC is the study coordinator, was in charge of site set up, sponsor liaison, and administrative aspects of study administration, and contributed to central data collection. LF was responsible for all the PCR laboratory studies. The study management group comprised REC, MC, SGO B, LF, TC, and FP. All authors contributed to study design and commented on and approved the manuscript. REC, JFA, DM, CP, GS, JLB, HdL, and MC were all principal investigators at their various sites and recruited patients. Declaration of interests REC reports other grants from Bloodwise during the conduct of the study; and grants and personal fees from Novartis, Bristol-Myers Squibb, and Pfizer, and personal fees from Ariad/Incyte, outside the submitted work. JFA reports grants and personal fees from Ariad/ Incyte and Pfizer, and personal fees from Bristol-Myers Squibb and Novartis, outside the submitted work. DM reports personal fees from Novartis, Bristol-Myers Squibb, Pfizer, and Ariad/Incyte, outside the submitted work. GS reports personal fees from Pfizer, Bristol-Myers Squibb, and Novartis, outside the submitted work. JLB reports personal fees from Novartis, Pfizer, Ariad/Incyte, and Bristol-Myers Squibb, outside the submitted work. HdL reports grants and personal fees from Ariad/Incyte, grants from Bristol-Myers Squibb, and personal fees from Novartis and Pfizer, outside the submitted work. SGO B reports grants from Ariad during the conduct of the study; and grants and personal fees from Bristol-Myers Squibb and personal fees from Pfizer, outside the submitted work. MC reports grants from Bloodwise during the conduct of the study; and personal fees from Ariad/Incyte and Pfizer, and personal fees and non-financial support from Novartis Pharma and Bristol-Myers Squibb, outside the submitted work. All other authors declare no competing interests. Acknowledgments This study was funded by Newcastle University and Bloodwise (grant number 132). We are indebted to the Liverpool Cancer Trials Unit, who coordinated the trial, and to the clinical teams and their patients at all 2 sites for their enthusiasm. We acknowledge the support of Newcastle University Clinical Trials Unit. JFA and LF acknowledge support of the NIHR Imperial College Biomedical Research Centre. References 1 Bower H, Björkholm M, Dickman PW, Höglund M, Lambert PC, Andersson TM. Life expectancy of patients with chronic myeloid leukemia approaches the life expectancy of the general population. J Clin Oncol 216; 34: Steegmann JL, Baccarani M, Breccia M, et al. European LeukemiaNet recommendations for the management and avoidance of adverse events of treatment in chronic myeloid leukaemia. Leukemia 216; 3: Experts in Chronic Myeloid Leukemia. The price of drugs for chronic myeloid leukemia (CML) is a reflection of the unsustainable prices of cancer drugs: from the perspective of a large group of CML experts. Blood 213; 121: Hughes TP, Ross DM. Moving treatment-free remission into mainstream clinical practice in CML. Blood 216; 128: Saußele S, Richter J, Hochhaus A, Mahon FX. The concept of treatment-free remission in chronic myeloid leukemia. Leukemia 216; 3: Mahon FX, Réa D, Guilhot J, et al; Intergroupe Français des Leucémies Myéloïdes Chroniques. Discontinuation of imatinib in patients with chronic myeloid leukaemia who have maintained complete molecular remission for at least 2 years: the prospective, multicentre Stop Imatinib (STIM) trial. Lancet Oncology 21; 11: Etienne G, Guilhot J, Rea D, et al. Long-term follow-up of the French Stop Imatinib study (STIM1) in patients with chronic myeloid leukemia. J Clin Oncol 217; 35: Ross DM, Branford S, Seymour JF, et al. Safety and efficacy of imatinib cessation for CML patients with stable undetectable minimal residual disease: results from the TWISTER study. Blood 213; 122: Rousselot P, Charbonnier A, Cony-Makhoul P, et al. Loss of major molecular response as a trigger for restarting tyrosine kinase inhibitor therapy in patients with chronic-phase chronic myelogenous leukemia who have stopped imatinib after durable undetectable disease. J Clin Oncol 214; 32: Lee SE, Choi SY, Song HY, et al. Imatinib withdrawal syndrome and longer duration of imatinib have a close association with a lower molecular relapse after treatment discontinuation: the KID study. Haematologica 216; 11: Richter J, Mahon FX, Guilhot J, et al. Stopping tyrosine kinase inhibitors in a very large cohort of European chronic myeloid leukemia patients: results of the EURO-SKI trial. European Haematology Association 21st Annual Meeting; Copenhagen, Denmark; June 1, 216. S Kuwabara A, Babb A, Ibrahim A, et al. Poor outcome after reintroduction of imatinib in patients with CML who interrupt therapy on account of pregnancy without having achieved an optimal response. Blood 21; 116: Brooks R. EuroQol: the current state of play. Health Policy 1996; 37: Bacik J, Mazumdar M, Murphy BA, et al. The functional assessment of cancer therapy-brm (FACT-BRM): a new tool for the assessment of quality of life in patients treated with biologic response modifiers. Qual Life Res 24; 13: Russo D, Martinelli G, Malagola M, et al. Effects and outcome of a policy of intermittent imatinib treatment in elderly patients with chronic myeloid leukemia. Blood 213; 121: Richter J, Söderlund S, Lübking A, et al. Musculoskeletal pain in patients with chronic myeloid leukemia after discontinuation of imatinib: a tyrosine kinase inhibitor withdrawal syndrome? J Clin Oncol 214; 32: Hahn EA, Glendenning GA, Sorensen MV, et al. Quality of life in patients with newly diagnosed chronic phase chronic myeloid leukemia on imatinib versus interferon alfa plus low-dose cytarabine: results from the IRIS Study. J Clin Oncol 23; 21: Aziz Z, Iqbal J, Aaqib M, Akram M, Saeed A. Assessment of quality of life with imatinib mesylate as first-line treatment in chronic phase-chronic myeloid leukemia. Leuk Lymphoma 211; 52: Williams LA, Garcia Gonzalez AG, Ault P, et al. Measuring the symptom burden associated with the treatment of chronic myeloid leukemia. Blood 213; 122: Efficace F, Baccarani M, Breccia M, et al. International development of an EORTC questionnaire for assessing health-related quality of life in chronic myeloid leukemia patients: the EORTC QLQ-CML24. Qual Life Res 214; 23: Vol 4 July 217 e316