A Global, Randomized, Placebo-Controlled, Phase 3 Study of Ibrutinib Plus Rituximab, Cyclophosphamide, Doxorubicin, Vincristine, and Prednisone (RCHOP) in Patients with Previously Untreated Non-Germinal Center B-Cell-like (GCB) Diffuse Large B-Cell Lymphoma (DLBCL)

Introduction

DLBCL is an aggressive disease that can be classified by immunohistochemistry (IHC) into GCB and non-GCB or by gene expression profiling (GEP) into GCB, activated B-cell (ABC), and unclassified subtypes. RCHOP is the standard frontline treatment (tx) for DLBCL with a 60% cure rate. Ibrutinib (ibr), an oral covalent Bruton's tyrosine kinase inhibitor, showed activity in relapsed/refractory ABC DLBCL. This study aimed to analyze if the addition of ibr to RCHOP would improve efficacy in previously untreated patients (pts) with non-GCB DLBCL or the ABC subtype.

Methods

Pts with centralized IHC-confirmed non-GCB DLBCL were randomized 1:1 to standard RCHOP with either ibr (560 mg/d, po) or placebo (pbo) on a 21-day cycle for 6 or 8 cycles. Pts were stratified by revised International Prognostic Index, region (US/Western Europe vs rest of world), and number of prespecified RCHOP cycles (6 or 8). ABC DLBCL was retrospectively identified using GEP (HTG EdgeSeq cell of origin assay). Primary end point was event-free survival (EFS), defined as time from randomization to progression, relapse from complete response (CR), initiation of tx for PET-positive or biopsy-proven residual disease after ≥ 6 cycles of RCHOP, or death, in the intent-to-treat (ITT) or ABC population. Secondary end points included progression-free survival (PFS), CR rate, overall survival (OS), and safety.

Results

Overall, 838 pts were randomized to ibr + RCHOP (n = 419) or pbo + RCHOP (n = 419). Median age was 62 years (yrs) with 58.5% < 65 yrs; 67.7% had ABC subtype. Baseline prognostic factors were balanced; slightly more pts ≥ 65 yrs in the ibr + RCHOP arm (45% vs 38%). In the ibr + RCHOP and pbo + RCHOP arms, 80.8% and 90.7% of pts received ≥ 6 cycles of RCHOP; 22.4% and 13.6% discontinued tx (all drugs) with adverse event (AE) being the most common reason (12.2% vs 5.3%). All deaths rate was 16.5% vs 17.2% in the ibr + CHOP vs pbo + RCHOP arm, respectively. In pts < 65 yrs, 90.4% vs 91.1% of pts received ≥ 6 cycles of RCHOP in the ibr + RCHOP vs pbo + RCHOP arm, respectively. Fewer pts ≥ 65 yrs in the ibr + RCHOP arm (69.0%) received ≥ 6 cycles of RCHOP compared with the pbo + RCHOP arm (90.0%). Ibr pharmacokinetics with RCHOP was similar to other studies in B-cell malignancies where ibr was used as a single agent or in combination. Median follow-up was 34.8 months.

Ibr + RCHOP did not improve EFS in pts with non-GCB (by IHC) or ABC (by GEP) DLBCL: The hazard ratio (HR) for EFS was 0.934 (95% confidence interval [CI], 0.726-1.200) for the ITT and 0.949 (95% CI, 0.704-1.279) for the ABC subset. Multivariate analysis showed a significant interaction (based on SAP prespecified 1-sided α level p < 0.1) between tx effect and age as a continuous or categorical variable. Efficacy was a function of age: stable benefit seen in younger pts, with a decline with increasing age, and benefit no longer favorable in pts ≥ 65 yrs.

Efficacy outcomes favored ibr + RCHOP-treated pts < 65 years with statistically significant risk reduction in EFS (30%), PFS (33%), and OS (43%; Table, Figure). A similar trend was seen in the ABC subset. Multivariate analyses, adjusting for other baseline covariates, confirmed the tx benefit in younger but not elderly pts.

Grade ≥ 3 AE rates were 89.9% and 87.1% in the ibr + RCHOP and pbo + RCHOP arms, respectively. Serious AEs (SAEs) were greater in the ibr + RCHOP vs pbo + RCHOP arm (53.1% vs 34.0%), particularly febrile neutropenia and pneumonia, as were AEs leading to RCHOP discontinuation. In pts ≥ 65 yrs, there were more SAEs (67.4% vs 40.6%) and pts who received < 6 cycles of RCHOP (31.0% vs 10.0%) in the ibr + RCHOP vs pbo + RCHOP arm. In pts < 65 yrs, grade ≥ 3 AEs were similar; SAEs were numerically higher (41.5% vs 29.8%) in the ibr + RCHOP vs pbo + RCHOP arm, but the proportion of pts who received < 6 cycles of RCHOP was similar (9.6% vs 8.9%) between arms.

Conclusions

While the addition of ibr to RCHOP did not improve efficacy in the ITT population, there was a significant interaction between tx and age. Among pts ≥ 65 yrs, unexpected increased toxicity associated with ibr + RCHOP resulted in reduced RCHOP exposure, which may explain in part the worse clinical benefit/risk profile of pts in the ibr + RCHOP vs pbo + RCHOP arm. However, in pts < 65 yrs, the addition of ibr showed clinically meaningful improvement in EFS, PFS, and OS with an acceptable safety profile.

Funding Source

Sponsored by Janssen. Writing assistance was provided by Liqing Xiao of PAREXEL and funded by Janssen.

View largeDownload slide

View largeDownload slide

Close modal