Can R-CHOP Plus Bortezomib Improve Outcomes in Patients With Non-GCB DLBCL?

In patients with newly diagnosed diffuse large B-cell lymphoma (DLBCL), adding the proteasome inhibitor bortezomib to standard R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone) did not significantly improve outcomes, compared with R-CHOP alone in patients with non–germinal center B-cell–like (GCB) DLBCL. John P. Leonard, MD, of Weill Cornell Medicine and New York Presbyterian Hospital, and co-authors published the findings in the Journal of Clinical Oncology.

The open-label, multicenter, prospective, randomized, phase II PYRAMID (Personalized Lymphoma Therapy: Randomized Study of Proteasome Inhibition in Non-GCB DLBCL) trial enrolled 206 patients from 69 U.S. sites between October 2009 and July 2013. Adult patients with untreated DLBCL with one or more measurable tumor masses and no central nervous system lymphoma were included. People were excluded from the trial if they had any other malignancy within two years of first dose, grade ≥2 peripheral neuropathy, and heart disease or uncontrolled cardiovascular conditions.

Patients were stratified by International Prognostic Index (IPI) risk group (low, low-intermediate, high-intermediate, or high) and randomized 1:1 to receive R-CHOP (n=103; median age = 61 years; range = 22-85 years) or bortezomib plus R-CHOP (VR-CHOP; n=103; median age = 65 years; range = 20-83 years). Dosing in each cycle was:

  • rituximab 375 mg/m2 intravenously (IV) on day 1
  • cyclophosphamide 750 mg/m2 IV on day 1
  • doxorubicin 50 mg/m2 IV on day 1
  • vincristine 1.4 mg/m2 (maximum of 2 mg) IV on day 1
  • prednisone 100 mg orally on days 1-5
  • bortezomib 1.3 mg/m2 IV on days 1 and 4

One-hundred patients receiving R-CHOP and 101 receiving VR-CHOP were included in the safety analysis. Common grade ≥3 adverse events (AEs) included neutropenia (53% vs. 49%, respectively), thrombocytopenia (13% vs. 29%), anemia (7% vs. 15%), leukopenia (26% vs. 25%), and neuropathy (1% vs. 5%). Serious AEs occurred in 31 percent of R-CHOP–treated patients and 34 percent of VR-CHOP–treated patients. Four patients (4%) discontinued R-CHOP and six (6%) discontinued VR-CHOP.

Two patients in each cohort died: one from treatment-related septic shock in the R-CHOP group and the other from acute cardiopulmonary arrest in the VR-CHOP group.

Progression-free survival (PFS; primary endpoint) was evaluable in 183 patients with centrally confirmed non-GCB DLBCL who received one or more study doses: 91 in the R-CHOP cohort and 92 in the VR-CHOP cohort.

After a median follow-up of 34.3 months in the R-CHOP cohort and 34.4 months in the VR-CHOP cohort (ranges not provided), median PFS was not reached in either group. Two-year PFS rates were 77.6 percent and 82.0 percent, respectively, and were similar across disease risk group:

  • 1% for high-intermediate vs. 72.4% for high (hazard ratio [HR] = 0.67; 90% CI 0.34-1.29)
  • 0% for low vs. 88.9% for low-intermediate (HR=0.85; 90% CI 0.35-2.10)

“The similarity of these rates for
R-CHOP and VR-CHOP suggest that
prolonged follow-up of patients … is
unlikely to reveal a difference.”

—John P. Leonard, MD

Median overall survival (OS) was also not reached in either cohort: 14 patients (15%) in the R-CHOP cohort and 11 (12%) in the VR-CHOP cohort died (HR=0.75; 90% CI 0.38-1.45) during follow-up. The two-year OS rates were 88.4 percent and 93.0 percent, respectively (HR=0.75; 90% CI 0.38-1.45; p=0.763). Again, two-year OS rates were similar regardless of disease stage:

  • 79.2% for high-intermediate vs. 92.1% for high (HR=0.62; 90% CI 0.25-1.42; p=0.638)
  • 97.7% for low vs. 93.8% for low-intermediate (HR=1.02; 90% CI 0.34-3.27; p=0.999)

“The similarity of these rates for R-CHOP and VR-CHOP suggest that prolonged follow-up of patients on this trial is unlikely to reveal a difference,” the authors noted. “A potential reason for the lack of benefit with VR-CHOP was that only two doses of bortezomib were given per 21-day cycle,” they wrote.

Another possible explanation was the observed discrepancies in the Hans algorithm used for DLBCL subtype classification, suggesting “that the Hans algorithm may be suboptimal for this purpose,” the researchers wrote, noting this as a limitation.

Millennium Pharmaceuticals, a subsidiary of Takeda, supported the study.

The authors report financial relationships with Celgene, AbbVie, Juno Therapeutics, Compass Oncology, TG Therapeutics, Gilead Sciences, Seattle Genetics, Acerta Pharma, CTI BioPharma, Genentech, Janssen Oncology, Pharmacyclics, Novartis, and GlaxoSmithKline.


Leonard JP, Kolibaba KS, Reeves JA, et al. Randomized Phase II Study of R-CHOP With or Without Bortezomib in Previously Untreated Patients With Non-Germinal Center B-Cell-Like Diffuse Large B-Cell Lymphoma. J Clin Oncol. 2017;35:3538-46.