Results from a subgroup analysis of the phase 3 ENDEAVOR study and a phase 2 study in diffuse large B cell lymphoma patients were presented at the annual meeting of the American Society of Hematology.
In a session on new agents and combination treatments for myeloma, Philippe Moreau, MD, from the University of Nantes, France, presented subgroup analysis from the randomized phase 3 ENDEAVOR Study, during the annual meeting of the American Society of Hematology.
The ENDEAVOR study, recently published in the Lancet Oncology, demonstrated that the doublet of carfilzomib and dexamethasone (Kd) significantly improved progression-free survival (PFS) compared with bortezomib (BTZ) and dexamethasone (Vd) (median PFS, 8.7 vs 9.4 months; hazard ratio [HR], 0.53; 95% CI, 0.44—0.65; P <.0001) in relapsed multiple myeloma (RMM). The subgroup analysis compared Kd or Vd after first relapse vs 2 or more prior lines of therapy
The randomized phase 3 study evaluated 929 adult patients with RMM who had received 1 to 3 prior lines of therapy. Patients were randomized 1:1 to Kd or Vd. Patients in the Kd arm received carfilzomib (30-min intravenous [IV] infusion) on days 1, 2, 8, 9, 15, and 16 (20 mg/m2 on days 1 and 2 of cycle 1; 56 mg/m2 thereafter) and dexamethasone 20 mg on days 1, 2, 8, 9, 15, 16, 22, and 23 of a 28-day cycle. Patients in the Vd arm received BTZ 1.3 mg/m2 (IV or subcutaneously) on days 1, 4, 8, and 11 and dexamethasone 20 mg on days 1, 2, 4, 5, 8, 9, 11, and 12 of a 21-day cycle. Treatment continued till disease progression or unacceptable toxicity. Primary end point was progression-free survival (PFS) and secondary end points included overall survival, overall response rate (ORR), duration of response, rate of grade 2 or higher peripheral neuropathy, and safety.
The analysis found median PFS for patients who had received 1 prior line was 22.2 months (95% CI, 17.7-not estimable [NE]) for Kd vs 10.1 months (95% CI, 8.8-12.7) for Vd (HR, 0.45; 95% CI, 0.33-0.61). Patients who had received at least 2 lines of therapy, however, did not respond as well to Kd—median PFS among these patients was significantly lower at 14.9 months (95% CI, 10.2-NE) and for Vd-treated patients it was 8.4 months (95% CI, 6.5-10.2. Additionally, the analysis found that prior exposure to either BTZ or lenalidomide reduced median PFS in both the Kd and Vd treatment groups. Similar trends were noted with ORR in these subgroups.
Grade 3 and greater adverse events (AEs) were significantly higher with Kd compared with Vd: 69.8% and 63.9%, respectively, in patients with 1 prior treatment, and 76.6% and 69.9%, respectively, in patients with at least 2 prior treatments. Grade 3 or higher hypertension, dyspnea, and cardiac failure were more common in the Kd group.
With these results, Moreau concluded that Kd treatment in patients with RMM yielded clinically meaningful improvement in PFS, regardless of the number of prior lines of therapy. “However, this improvement was more significant in patients who had received 1 previous line of therapy,” he said. Further, PFS benefit with either combination was recognized regardless of prior exposure to specific agents. “Our results show that the combination of carfilzomib and dexamethasone has a favorable benefit-risk profile in RMM, irrespective of prior treatment, and this 2-drug combination should be considered in patients who have progressed on lenalidomide maintenance,” Moreau proposed.
Diffuse large B cell lymphoma
In a phase 2 open-label trial of bortezomib plus R-CHOP (Rituximab, Cyclophosphamide, doxorubicin [Hydroxydaunomycin], vincristine, and Prednisone) therapy, investigators tested whether the including bortezomib would improve response rates in patients with diffuse large B cell lymphoma (DBLCL) of the non-GCB (ie, germinal center B cell) subtype. Investigators randomized patients who were previously untreated to receive VR-CHOP (n = 95) or standard R-CHOP (n = 95). In both trial arms, patients received at least 6 cycles of therapy. Researcher evaluated primary and secondary end points of PFS, and secondary end points of overall survival (OS), ORR, complete response (CR), and safety. This study was powered to detect a 15 percentage-point difference in response rate at 2 years with VR-CHOP versus RCHOP. Patients were evaluated at the end of cycles 2 and 6 using a fluorodeoxyglucose (FDG) positron emission tomography (PET) scan, and through computed tomography (CT) scan.
A total of 206 patients were randomized to receive treatment, with 103 patients in each arm of the trial. However, of patients who actually received at least 1 dose of study medication (the modified intent-to-treat population), only 91 patients qualified for this population in the R-CHOP group versus 92 patients in the VR-CHOP group. Further reducing this number, the evaluable patients in each arm (R-CHOP and VR-CHOP) were 86 patients and 90 patients, respectively.
In terms of demographic characteristics, patients had a median age of 64 years, and population characteristics were evenly balanced across risk groups. Three-fourths of all patients had stage III or stage IV disease upon randomization. Of patients receiving R-CHOP, 86% completed all 6 cycles of treatment, comparable to the 85% of patients receiving VR-CHOP for all 6 cycles. The median dose intensity in this trial was >98% of the full therapeutic dose, indicating very few dose reductions due to adverse events. Following at least 6 weeks of therapy, patients were followed for a median of 34 months.
No significant difference in PFS was observed between the 2 regimens. After 2 years of therapy, 78% of patients receiving R-CHOP and 82% of patients receiving VR-CHOP met the PFS end point (P = .611). In subanalyses of low- and high-risk groups, PFS curves were identical to those with high-risk, or intermediate/high-risk disease. Similarly, overall survival was unaffected, with 88% of patients receiving R-CHOP and 93% of patients receiving VR-CHOP surviving at the 2-year end point (P = .78). These negative results indicate no benefit to adding bortezomib to treatment of patients with DBLCL without the GCB genotype.
Although this negative finding may indicate that bortezomib is not effective in these patients, it is possible that the method of selection of patients (the Hans IHC algorithm) did not adequately select patients with the non-GCB genotype. In addition, it is possible that patients randomized in a prospective analysis (as in this study) selects out patients with very severe disease, who tend to drop out of clinical trials. This bias in prospective studies versus retrospective studies may explain why retrospective studies show a benefit with bortezomib, whereas prospective studies do not.
In another prospective randomized controlled trial of bortezomib added to R-CHOP therapy in patients with DLBCL, researchers from the United Kingdom reported similarly negative results. This study, known by the acronym REMoDL-B, showed the results of treatment with targeted therapy for DBLCL based on a real-time gene expression profiling data.
Rather than focusing solely on the non-GCB population, researchers separated patients into 3 groups: patients with the GCB genotype, the activated B cell (ABC) genotype, and patients with DBLCL that could not be categorized. In patients with the ABC genotype, the NF-kappa-B pathway is known to be constitutively active. Due to this constitutive activity, the goal of bortezomib therapy is inhibition of this pathway to improve outcomes in patients with the ABC genotype.
A total of 1085 patients were eligible for this study and were randomized to treatment. Of patients enrolled in the study, 248 had the ABC genotype, 477 had the GCB genotype, and 201 patients were unclassifiable. For a total of 130 additional patients, screening failed for a variety of reasons, and 29 patients did not receive results due to technical equipment failures. Patients with both ABC and GCB genotypes and unclassifiable patients were randomized to receive R-CHOP or VR-CHOP, and baseline demographics in each group were generally well matched.
Across treatment groups, rates of adverse events of grade 3 or higher severity were similar, and rates of neuropathy were broadly similar, with the exception of grade 3 or higher neuropathy, which was more common in the VR-CHOP arm than in the R-CHOP arm (3.0% vs 0.9%). CR rates as determined by CT scan were nearly equivalent in all groups treated, with response rates hovering around 60%. No significant differences were detected across groups, even after multiple subanalyses.
Over a median follow-up of 16.3 months, researchers observed 245 progression events, 138 deaths (12.7% of the population). Most of these deaths were the result of progressive disease. For the full population sample, PFS was 79.0% at 12 months and 71.7% at 24 months. Although the data from this study remains immature, the results are unpromising, with PFS curves for all patient subgroups that virtually overlap. These results confirm that real-time gene expression profile is possible in a clinical trial, with low failure rates due to equipment failure. However, addition of bortezomib to R-CHOP chemotherapy does not appear to affect rates of early treatment failure in patients with DBLCL of any subtype.