Résumé
Background : Ibrutinib is a once-daily Bruton's tyrosine kinase inhibitor approved in the US and EU as either a single-agent therapy or in combination with rituximab (R) for treatment of patients (pts) with Waldenström's macroglobulinemia (WM) across all lines of therapy. In the phase 3 iNNOVATE study (PCYC-1127; NCT02165397), ibrutinib demonstrated superior progression-free survival (PFS) in combination with R (IR) vs placebo plus R in pts with WM after a 26.5 mo median follow-up in the primary analysis (Dimopoulos, N Engl J Med 2018). After a 30.4 mo median follow-up, IR continued to show superiority over R in treatment-naive (TN) and previously treated pts with WM, regardless of genomic factors (Buske, ASH 2018). Here, we present results from the final analysis of the randomized arms of iNNOVATE.
Methods : Pts with confirmed symptomatic WM requiring treatment were randomized to once-daily ibrutinib 420 mg or placebo plus R (375 mg/m2/week IV at weeks 1-4 and 17-20). Pts could be TN or previously treated; pts with prior R therapy had to have achieved at least a minor response (MR) to their last R-based regimen. Endpoints included PFS and response rates per Independent Review Committee (IRC), overall survival (OS), hemoglobin (Hgb) improvement, time to next treatment (TTNT), and safety.
Results : Of the 150 pts randomized (75 per arm), baseline characteristics were well balanced between arms. Median follow-up was 50 mo (range 0.5+ to 63). Median PFS was not reached (NR; 95% CI 57.7 mo to not estimable) with IR vs 20.3 mo (95% CI 13.0-27.6) with R (hazard ratio [HR] 0.25 [0.15-0.42]; P<0.0001) (Figure 1A). PFS rates were 68% vs 25% at the 54 mo landmark timepoint. Patients treated with IR also had a PFS benefit over those treated with R, regardless of their prior treatment status (HR [95% CI]: TN, 0.32 [0.14-0.70]; previously treated, 0.22 [0.11-0.43]) or genotype (HR [95% CI]: MYD88L265P/CXCR4WT, 0.18 [0.08-0.43]; MYD88L265P/CXCR4WHIM, 0.27 [0.12-0.62]; MYD88WT/CXCR4WT, 0.29 [0.07-1.19]) (Figure 1B). PFS benefit of IR over R was also observed across prespecified subgroups including baseline age, sex, serum immunoglobulin, Hgb, and IPSSWM. Major response rate (≥partial response) was 76% with IR vs 31% with R (P<0.0001), and overall response rates (≥MR) were 92% vs 44% (P<0.0001). High response rates were observed with IR regardless of prior treatment (91% [31/34] in TN; 93% [38/41] in previously treated) and across genotypes (94% [30/32] in MYD88L265P/CXCR4WT; 100% [26/26] in MYD88L265P/CXCR4WHIM; 82% [9/11] in MYD88WT/CXCR4WT). A greater proportion of pts receiving IR vs R had sustained Hgb improvement (77% vs 43%; P<0.0001). Median OS was NR in either treatment arm; at the 54 mo landmark timepoint, the OS rate was 86% with IR vs 84% with R. Thirty-five pts (47%) on R crossed over to single-agent ibrutinib after IRC-confirmed progressive disease. Median TTNT was NR with IR and 18 mo with R; 87% of pts receiving IR and 29% receiving R had not received subsequent treatment at 54 mo. Treatment has ended for the R arm; median treatment duration was 16 mo and safety results have been previously reported (Buske, ASH 2018). Median treatment duration for IR was 48 mo. Overall, the safety profile of IR is consistent with previous reports; minimal differences (1-4%) in common (≥10%) adverse event (AE) rates were observed after 24 months of additional follow-up since the primary analysis. The most common grade 3-4 AEs (percent in final analysis [difference from primary analysis]) were atrial fibrillation (16% [4%]), hypertension (15% [1%]), neutropenia (13% [4%]), and anemia (12% [1%]). Of 12 pts with grade 3-4 atrial fibrillation, 9 (75%) remained on treatment; no other ibrutinib discontinuations due to common (≥10%) grade 3-4 AEs occurred. After study closure, 68 pts (45%) remained on treatment (32 enrolled in a treatment extension study and 36 continued to receive ibrutinib in a non-trial setting).
Conclusions : With up to 5 years of follow-up, IR showed ongoing superiority across clinical outcomes in pts with WM regardless of genotype, prior treatment, and key pt characteristics. Patterns of response with IR were similar for both TN and previously treated pts. IR maintained a manageable safety profile, with no new safety signals identified and minimal increases in common AEs after an additional 24 months of follow-up.