Abstract: S483

Type: Oral Presentation

Presentation during EHA20: From 13.06.2015 16:15 to 13.06.2015 16:30

Location: Room A7

Background
Subcutaneous rituximab (R^SC) offers improved patient (pt) convenience and healthcare resource savings versus intravenous R (R^IV), with similar efficacy and safety outcomes in the SABRINA study in follicular lymphoma (FL). R^SC is approved in Europe and other countries for use in FL and diffuse large B-cell lymphoma (DLBCL).

Aims
To confirm the efficacy and safety of R^SC versus R^IV plus CHOP (cyclophosphamide, doxorubicin, vincristine and prednisone) for the treatment of DLBCL.

Methods
The open-label, randomised MABEASE study (NCT01649856) investigated the efficacy of R^SC versus R^IV plus CHOP in untreated DLBCL. Pts were randomised 2:1 to receive R^SC (R^IV 375 mg/m² cycle 1; R^SC 1400 mg fixed dose cycles 2–8) or R^IV (R^IV 375 mg/m² cycles 1–8), plus 6 or 8 cycles of CHOP every 14 or 21 days. Responses (complete response [CR], unconfirmed CR [CRu], partial response [PR], progressive disease [PD]) were determined by investigator assessment (Cheson 1999 criteria). Administration-related reactions (ARRs) were defined as R-related adverse events (AEs) that occurred ≤24 hours of administration. The primary endpoint was response rate at end of treatment (EOT). Safety was a secondary endpoint.

Results
In total, 576 pts were randomised (R^SC n=381; R^IV n=195; intent-to-treat [ITT] population); baseline characteristics were balanced between arms. A total of 572 pts (R^SC n=378; R^IV n=194) received ≥1 dose of R. Nine R^SC pts discontinued after cycle 1, due to AE (n=5), withdrawn consent (n=2), PD and protocol violation (both n=1); as both arms received R^IV during cycle 1, these pts were included in the R^IV arm for safety analyses (R^SC n=369; R^IV n=203). At EOT (R^SC n=342; R^IV n=177), response rates were similar between arms in the ITT population overall (R^SC 52%; R^IV 51%) and when pts were stratified by age (~50% for all groups; Table). AE rates were comparable between the R^SC and R^IV arms overall (92% vs 91%) and during cycle 1 (61% vs 64%); the most common AE overall was neutropaenia (R^SC 36%; R^IV 33%). During cycles 2–8, R^SC and R^IV pts had similar rates of grade ≥3 AEs (53% vs 50%) and serious AEs (SAEs; 34% vs 31%); the most frequent grade ≥3 AE was neutropaenia (R^SC 22%; R^IV 20%) and the most common SAE was febrile neutropaenia (R^SC 11%; R^IV 6%). ARR rates were similar between R^SC and R^IV pts overall (28%/arm), during cycle 1 (11% vs 14%) and cycles 2–8 (22% vs 20%). From cycle 2–8, the most common ARR by System Organ Class (SOC) was general disorders and administrative site conditions (R^SC 10%; R^IV 4%). Rates of grade ≥3 ARRs was similar between arms overall (2% vs 5%), during cycle 1 (0% vs 2%) and cycles 2–8 (2% vs 4%); from cycles 2–8, the most common grade ≥3 ARR by SOC was blood and lymphatic system disorders (R^SC 1%; R^IV 3%). In total, 52% (R^SC) and 54% (R^IV) of pts had R-related AEs; the most frequent R-related AE was neutropaenia (R^SC 14%; R^IV 16%). Overall, 8% (R^SC) and 9% (R^IV) of pts discontinued R due to AEs, most commonly infections (R^SC 2%; R^IV 3%). In total, 63 pts (11%) died (R^SC 10%; R^IV 13%). AEs led to death in 6% (R^SC) and 7% (R^IV) of pts, most commonly due to infections (2%/arm).

Summary
R^SC and R^IV plus CHOP had similar efficacy and safety outcomes in previously untreated DLBCL. R^SC offers convenience for pts and efficient utilisation of healthcare resources.

Keyword(s): CHOP, Diffuse large B cell lymphoma, Rituximab



Session topic: Optimization and innovation in treating aggressive lymphomas