Randomized Phase 2 Open-Label Study of R-CHOP ± Bortezomib in Patients (Pts) with Untreated Non-Germinal Center B-Cell-like (Non-GCB) Subtype Diffuse Large Cell Lymphoma (DLBCL): Results from the Pyramid Trial (NCT00931918)

Background Addition of bortezomib to R-CHOP (VR-CHOP) may overcome the less favorable prognosis of non-GCB subtype lymphoma (Ruan et al. JCO 2011). We report results of a prospective open-label, randomized, phase 2 study evaluating the efficacy and safety of frontline R-CHOP vs VR-CHOP in pts with non-GCB DLBCL.

Methods Adult pts with previously untreated non-GCB DLBCL who had ≥1 site of measurable disease, ECOG performance status 0-2, and adequate hematologic, hepatic, and renal function were eligible. Confirmation of non-GCB subtype using the Hans immunohistochemical (IHC) algorithm was required. Hans IHC non-GCB testing was performed in real time at a central US laboratory (48-72 hr turnaround from arrival of FFPE sample). Several centers demonstrated consistent scoring with the central laboratory and were permitted to enroll pts based on local non-GCB subtyping, with retrospective central laboratory confirmation. All pts received 6 cycles of standard R-CHOP in 21-d cycles (rituximab 375 mg/m², cyclophosphamide 750 mg/m², doxorubicin 50 mg/m², vincristine 1.4 mg/m² [max 2 mg], all IV on d 1, and prednisone 100 mg PO on d 1-5). In the VR-CHOP arm, pts also received bortezomib 1.3 mg/m² IV on d 1 and 4 of each cycle. Follow-up was every 3 mos for up to 2 yrs after enrollment of the last pt. The primary endpoint was progression-free survival (PFS). Secondary endpoints included: overall survival (OS), overall response rate (ORR) and complete response (CR) rate after cycles 2 and 6, and safety. Response/disease progression were investigator-assessed by CT and FDG-PET scan at the end of cycles 2 and 6, and in follow up per 2007 Revised Response Criteria for Malignant Lymphoma. Adverse events (AEs) were graded by NCI-CTCAE v3.0. Sample size (n=~190) was determined to provide 80% power to detect an improvement in 2-yr PFS from 62% with R-CHOP (Fu et al. JCO 2008) to 77% (1-sided log rank; significance level 0.05). Here we present preliminary findings (data cut-off: Jun 2015).

Results 206 pts were randomized at 69 sites; of these, 183 (91 R-CHOP, 92 VR-CHOP) had centrally confirmed non-GCB DLBCL and received ≥1 dose of study drug (modified intent-to-treat population). 86% (R-CHOP) and 85% (VR-CHOP) of pts completed study treatment per protocol. 60% of pts received study drug within 4 wks after tumor sample collection. Baseline characteristics were (R-CHOP vs VR-CHOP): male 58% vs 49%; median age 62 vs 65 yrs (>65 yrs 44% vs 46%); AJCC stage III/IV disease 73% vs 72%; extranodal disease 46% vs 52%; bone marrow involvement 11% vs 14%; IPI low/low-int/high-int/high 24/25/38/12% vs 26/28/33/13%. After a median follow-up of 31.5 mos, 2-yr PFS was 77% vs 82% (HR 0.77; 90% CI: 0.45, 1.30; p=0.70). In pts with high-int/high IPI, 2 yr PFS was 64% vs 72% (HR 0.66; 90% CI: 0.34, 1.28; p=0.294), whereas in pts with low/low-int IPI the HR was 1.13 (90% CI: 0.46, 2.75; p=0.821). At data cut-off, 15% and 11% of pts in the R-CHOP and VR-CHOP arms had died (HR: 0.65; 90% CI: 0.32, 1.29); median OS was not estimable in either arm and 2 yr OS rates were 80% vs 82%. In pts with high-int/high IPI, 2 yr OS rates were 79% (R-CHOP) vs 92% (VR-CHOP); in pts with low/low-int IPI, 2-yr OS rates were 98% in both arms. In 86 R-CHOP and 90 VR-CHOP response-evaluable pts, ORRs were 98% vs 92% (52% vs 54% CR). After 2 yrs, 73% of R-CHOP pts and 76% of VR-CHOP pts had not yet received a subsequent anti-lymphoma therapy. The safety population comprised 100 R-CHOP and 101 VR-CHOP pts. In both arms, pts received a median of 6 cycles of therapy (range 1-6). In the R-CHOP and VR-CHOP arms, 71% and 79% of pts had a G≥3 AE, and 31% and 34% had serious AEs, and 55% and 68% reported drug-related G≥3 AEs, the most common of which were neutropenia (34% vs 28%) and thrombocytopenia (8% vs 20%).G≥3 peripheral neuropathy rates were 1% (R-CHOP) and 5% (VR-CHOP).

Conclusions These preliminary data suggest no significant efficacy advantage with the addition of bortezomib to R-CHOP in pts with previously untreated non-GCB DLBCL. This may be due to lack of bortezomib effect, time required for Hans IHC testing, IHC missclassification, or pt selection (R-CHOP alone pts had better outcomes/lower event rate than expected). These results have important implications for upcoming studies of new therapeutic strategies in DLBCL that target pt subsets based on cell of origin.