Favorable Outcomes in CLL Pts with Alternate Kinase Inhibitors Following Ibrutinib or Idelalisib Discontinuation: Results from a Large Multi-Center Study

Introduction: B cell receptor signal transduction kinase inhibitors (KI) represent a paradigm shift in CLL management; however, there are limited data regarding real world practice patterns of KI discontinuation and response to subsequent therapies. Two centers reported outcomes of ibrutinib (Ibr) failure patients (pts) treated on clinical trials (n=33, n=71) documenting an extremely poor prognosis (Jain 2014, Maddocks 2015). No analogous data in idelalisib (Ide) failure pts are available. Given the impressive activity of KI and the rarity of discontinuation events, 10 large US cancer centers collaborated to capture the experience of 123 CLL pts who discontinued Ibr- or Ide-based regimens focusing on reasons for discontinuation and response to subsequent KI therapies.

Methods: We conducted a multicenter, retrospective analysis of CLL pts who discontinued Ibr- or Ide-based therapy for any reason. We examined demographics, reason for discontinuation, responses, survival data, toxicity, and post KI therapies. Primary endpoint was post KI PFS (time from post KI treatment to progression or last f/u) as determined by the Kaplan Meier method. Comparisons were made using log rank test. Each institution received IRB approval.

Results: A total of 123 KI discontinuation pts (Ibr=93/Ide=30) were identified. Table 1 describes baseline characteristics at start of KI. Interestingly, 10% and 32% of Ibr (5% 140 mg, 5% 280 mg daily) and Ide pts (32% 100 mg BID), respectively, were initiated at doses less than FDA labeled dose. Further, 23% and 42% of Ibr pts and 30% and 65% Ide pts had doses modified or held, respectively, prior to discontinuation. Overall median time on KI therapy was 5 months (mos) (4.8 Ibr, 5.5 Ide). ORR to KI was 63% (CR 15%, PR 39%, PR-L 9%). Most common reasons for KI discontinuation were toxicity (58% Ibr, 60% Ide), CLL progression (24% Ibr, 30% Ide), and RT DLBCL (8% Ibr, 7% Ide). Table 2 describes the most common toxicities leading to KI discontinuation. Median PFS from KI initiation for the cohort was 9 mos (77 events, median f/u 6 mos, Figure A). When stratified by discontinuation reason (Figure B), PFS was significantly inferior in RT pts (5 mos RT vs 8 mos progression vs 9 mos toxicity, P=0.04). Median OS for the entire cohort has not been reached (34 deaths, med f/u 12 mos). Median OS was inferior in RT pts (10 mos in RT vs. not reached in CLL progression/toxicity pts; P=0.01). At the time of analysis, 66 (54%) pts following KI discontinuation (21 progression, 36 toxicity, 7 RT, 2 SCT) received a first salvage regimen: 20% R-chemotherapy, 18% Ibr-based, 17% Ide-based, 15% anti-CD20 mab, 14% BCL2-inhibitor, 5% cellular therapy, 5% experimental KI, 3% IMID, 3% other. In evaluable pts, ORR to non-KI therapies following Ibr/Ide discontinuation was 40%. ORR to Ide-based therapy (n=12) following Ibr discontinuation was 50% (42% PR, 8% PRL). ORR to Ibr-based therapy following Ide discontinuation (n=13) was 77% (54% PR, 23% PRL). Responses to alternate KI were similar in pts who discontinued KI for toxicity (ORR 60% PR+PRL) and progression (ORR 67% PR + PRL). PFS (median f/u 5 mos, Figure C) for pts treated with an alternate KI therapy (Ibr → Ide, Ide → Ibr) has not been reached vs 7 mos in non-KI therapies (Figure D).

Conclusions: We describe the largest combined experience of practice patterns and outcomes post KI discontinuation in CLL. The majority of pts discontinued KI therapy due to toxicity or progression, not RT. For the first time, we demonstrate that the majority of pts who discontinue a KI due to toxicity or progression respond to other therapies, particularly alternate KI therapy. Collectively, these data provide supporting evidence that: (1) reason for KI discontinuation is toxicity in the majority of cases; (2) alternate KI therapy following KI discontinuation is efficacious; (3) non-overlapping toxicity profiles permit utilization of alternate KI following discontinuation due to toxicity; and (4) mechanisms of KI resistance may not overlap.

Figure 1.

View largeDownload slide

Figure 1.

View largeDownload slide

Close modal