Progression-free survival in relapsed myeloma doubled among patients treated with carfilzomib, rather than bortezomib, in combination with dexamethasone, a randomized trial showed.
Meletios A. Dimopoulos, MD
Progression-free survival (PFS) in relapsed myeloma doubled among patients treated with carfilzomib (Kyprolis), rather than bortezomib (Velcade), in combination with dexamethasone, a randomized trial showed.
The carfilzomib-based regimen, administered at a higher dose than currently approved, led to a median PFS of 18.7 months versus 9.4 months with the bortezomib-dexamethasone combination. The objective response rate was significantly higher with carfilzomib, and twice as many patients achieved complete responses and very good partial responses or better with the newer proteasome inhibitor, as reported at the ASCO meeting in Chicago.
“The combination of carfilzomib and dexamethasone was superior to bortezomib and dexamethasone regardless of age or prior bortezomib exposure and represents a new standard of care,” said Meletios A. Dimopoulos, MD, chair of clinical therapeutics at the University of Athens in Greece.
“Although patients treated with carfilzomib and dexamethasone remained on study treatment longer, treatment discontinuation due to adverse events and on-study deaths due to adverse events were comparable between groups.”
The combination of bortezomib and dexamethasone is a standard-of-care regimen recommended by the NCCN for patients with relapsed multiple myeloma. Carfilzomib has approval in the United States and elsewhere as single-agent therapy in relapsed and refractory myeloma.
The approved dosage for carfilzomib is 20 mg/m2 on days 1 and 2 of the first cycle, followed by 27 mg/m2 thereafter (20/27 mg/m2), infused over 2 to 10 minutes. A phase Ib/II trial demonstrated a maximum tolerated dose of 20/56 mg/m2, using a 30-minute infusion duration. The higher dose was associated with higher response rates as compared with lower doses, said Dimopoulos.
On the basis of the phase Ib/II data, investigators initiated a first-ever phase III randomized comparison of carfilzomib and bortezomib, each combined with dexamethasone. Patients with relapsed myeloma and a treatment history of as many as three prior regimens were eligible.
Prior exposure to the two study drugs was allowed if a patient had achieved at least a partial response, had at least a 6-month protease inhibitor treatment-free interval, and did not discontinue because of adverse events.
Investigators randomized 929 patients to carfilzomib 20/56 mg/m2 by 30-minute infusion or bortezomib 1.3 mg/m2 (by intravenous or subcutaneous administration). Patients in both arms received dexamethasone 20 mg/day. The carfilzomib regimen was administered in 28-day cycles versus 21-days for the bortezomib regimen. Treatment in both groups continued until disease progression or development of unacceptable toxicity.
The primary endpoint was PFS. Secondary endpoints include overall survival, objective response rate, duration of response, incidence of grade ≥2 peripheral neuropathy, and safety.
The treatment cohorts had a median age of 65, and about 15% of each group was 75 or older. All but 7% of patients had ECOG 0 or 1 performance status (about 50% ECOG 0), and about 20% of the patients had high-risk cytogenetic by fluorescence in situ hybridization.
About 80% of the patients had baseline creatinine clearance of at least 50. Grade 1 peripheral neuropathy was present at baseline in about 30% of patients, and an additional 2% had grade 2 neuropathy. Half of the patients had received one prior regimen, 30% to 35% had received two prior regimens, and 15% to 20% had received three prior regimens.
Half the patients had prior exposure to bortezomib, 38% had received lenalidomide, and almost half had exposure to thalidomide. Fewer than 1% had a treatment history that included carfilzomib.
The primary analysis showed a difference in PFS that translated into a 47% reduction in the hazard for disease progression or death (P <.0001). The advantage for carfilzomib was consistent regardless of patient age, performance status, creatinine clearance, history of neuropathy, cytogenetic risk group, International Staging System risk category, and treatment history (number or types of regimens).
Twice as many patients achieved complete responses with carfilzomib (13% vs 6%). The rate of very good partial response or better was 54% with carfilzomib and 29% with bortezomib. The overall response rate was 77% in the carfilzomib arm and 63% in the bortezomib arm (P <.0001).
Median overall survival was 24.3 months in the bortezomib group but has yet to be reached in the carfilzomib arm (HR =0.79; P = .066, carfilzomib vs bortezomib). The survival data are immature for a meaningful comparison of overall survival, said Dimopoulos.
The median treatment duration was 40 months with carfilzomib and 27 months with bortezomib. Grade 3 adverse events occurred more often with carfilzomib (73% vs 67%), as did serious adverse events (48% vs 36%). Dose reductions associated with adverse events occurred more often in the bortezomib arm (48% vs 23%). Four deaths associated with adverse events occurred in the carfilzomib group, as did three in the bortezomib arm.
Grade ≥3 hematologic adverse events occurred in a similar proportion of patients in both groups, including anemia, thrombocytopenia, neutropenia, upper respiratory infection, and pneumonia.
No grade ≥3 nonhematologic events occurred in more than 8% of patients in either group. The most frequent grade ≥3 events were diarrhea, fatigue, dyspnea, pyrexia, constipation, and insomnia. Peripheral neuropathy occurred in 5% of patients treated with bortezomib and 1.3% of those in the carfilzomib arm.
The proportion of patients with grade ≥2 peripheral neuropathy was significantly higher in the bortezomib arm: 32% versus 6% (P <.0001).
Dimopoulos MA, Moreau P, Palumbo A, et al. Carfilzomib and dexamethasone (Kd) vs bortezomib and dexamethasone (Vd) in patients (pts) with relapsed multiple myeloma (RMM): results from the phase III study ENDEAVOR. J Clin Oncol. 2015;(suppl; abstr 8509).