Rapid Prospective Identification of Non-Germinal Center B Cell-Like (GCB) Diffuse Large B-Cell Lymphoma (DLBCL) Patients for Targeted Trials: Early Results From PYRAMID, A Phase 2 Randomized Study of R-CHOP ± Bortezomib In Newly Diagnosed Non-GCB DLBCL.

Abstract 1792 Background: With increasing knowledge of cancer pathogenesis and availability of target-specific novel agents, it is expected that therapy will be increasingly tailored to individual biology. This is particularly true in lymphoma because translational research has characterized the molecular basis of the clinical heterogeneity in major lymphoma types, and many new agents are in development. However, a major challenge for clinical studies of specific lymphoma patient subgroups is the real-time testing of patient material in a way that enables prospective targeted clinical trials. Here we describe early patient selection results from a phase 2 trial of newly diagnosed DLBCL that prospectively identifies and enrolls only patients with the non-GCB subtype for randomization to rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) with or without the proteasome inhibitor bortezomib (VELCADE®). The non-GCB subtype exhibits inferior outcome compared to the GCB subtype after therapy with either CHOP or R-CHOP. Both preclinical and clinical data suggest that inhibition by bortezomib of the nuclear factor-κB pathway, which is frequently activated in non-GCB tumors, specifically benefits these patients (Dunleavy K et al, Blood 2009). Methods: This multicenter phase 2 study is enrolling patients at academic centers (n=10), major hospitals (n=13), and community practices (n=32), providing insight that may be broadly informative with respect to both patient outcomes and future targeted trial design. Obstacles to non-GCB-specific patient selection in this setting include change in standard treatment practice for DLBCL, as well as the challenges of collecting a tumor sample, running the assay, and reporting the subtype in a timely manner. All clinical centers in this study have agreed to release diagnostic pathology blocks to a College of American Pathologists/Clinical Laboratory Improvement Amendments-certified pathology lab for centralized subtyping. The pathology samples are subject to non-GCB testing via a 3-marker immunohistochemistry panel (CD10, bcl-6, mum-1; Hans CP et al, Blood 2004), and the results are reported directly to the clinical site. Results: To date, 45 patients have provided consent and have been subtyped. The mean turnaround time from receipt of sample to reporting of subtype was 1.2 business days (range 1–2 business days). All pathology blocks have been returned to the clinical sites; the mean time elapsed was 4.6 business days (range 4–6 business days). Rapid return of these samples was cited by clinical centers as an important aspect of the study design. There is no indication that pathology block release and central laboratory subtyping has led to delays in patient treatment. Of the 42 patients with results available to date, 5 samples were inevaluable, primarily due to limited tumor cells in the biopsy. Of the 37 samples successfully subtyped, 22 were GCB (59.5%) and 15 were non-GCB (40.5%); this is generally consistent with the ∼50% prevalence of the non-GCB subtype reported in retrospective studies of DLBCL. Fourteen patients have been randomized to therapy. Conclusions: Updates to turnaround time, patient subtype frequency, and enrollment will be presented. These data indicate that many centers are interested in targeted trials that offer new and potentially tailored therapies for their DLBCL patients. These centers can overcome their reluctance to release diagnostic pathology samples and actively enroll patients to randomized studies of frontline therapy. Overall, these results suggest that with careful trial design and attention to key logistical issues, prospective enrichment of specific patient subtypes is feasible in lymphoma. The PYRAMID trial (Clinicaltrials.gov: NCT00931918) continues to enroll patients. Disclosures: Off Label Use: Discussion of Velcade in NHL subtypes other than mantle cell lymphoma is included. Flinn:Millennium Pharmaceuticals, Inc: Research Funding. Noga:Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Millennium Takeda: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Cephalon: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Ortho-Centicor: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau. Papish:Genentech: Consultancy, Honoraria, Speakers Bureau; Sanofi Aventis: Consultancy; MD Advantage: Consultancy; Genomic Health: Speakers Bureau. Reeves:Celgene: Equity Ownership. Parasuraman:Millennium: Employment, Equity Ownership. Corvez:Millennium: Employment, Equity Ownership. Mulligan:Millennium Pharmaceuticals, Inc.: Employment. Leonard:Millennium Pharmaceuticals, Inc.: Consultancy, Honoraria.