Carfilzomib and dexamethasone (Kd) vs bortezomib and dexamethasone (Vd) in patients (pts) with relapsed multiple myeloma (RMM): Results from the phase III study ENDEAVOR.

8509 Background: ENDEAVOR (NCT01568866) is comparing Kd with Vd in pts with RMM. The primary endpoint is progression-free survival (PFS). Secondary endpoints include overall survival (OS), overall response rate (ORR), rate of peripheral neuropathy (PN), and safety. Methods: Adults with RMM and 1-3 prior treatments were eligible; planned enrollment was 888 pts. Pts were randomized 1:1 and stratified by prior K or V (yes vs no), prior lines of treatment (1 vs 2-3), ISS stage (1 vs 2-3), and intended route of V (IV vs SC). The Kd arm received K (30-min IV infusion) on days (D) 1, 2, 8, 9, 15, and 16 of a 28-day cycle (20 mg/m2 on D1 and 2 [cycle 1]; 56 mg/m2 thereafter) and dexamethasone (dex; 20 mg) on D1, 2, 8, 9, 15, 16, 22, and 23. The Vd arm received V (1.3 mg/m2; IV or SC on D1, 4, 8, and 11 of a 21-day cycle) and dex (20 mg) on D1, 2, 4, 5, 8, 9, 11, and 12. Cycles were repeated until disease progression or unacceptable toxicity. Results: Data are presented for Kd then Vd. In total, 929 pts (Kd: 464; Vd: 465) from 27 countries were randomized. In the Vd arm, 83.6% of pts received SC V. At the preplanned interim analysis, median treatment exposure was 39.9 and 26.8 weeks. Kd showed a significant improvement in median PFS vs Vd (18.7 months [mo] vs 9.4 mo; hazard ratio = 0.53; P< .0001). OS data were immature (75 and 88 deaths) and continue to be followed. ORRs were 76.9% and 62.6% (P< .0001); 54.3% and 28.6% had a very good partial response or better, and 12.5% and 6.2% of pts had a complete response or better. Treatment discontinuation due to an adverse event (AE) occurred in 14.0% and 15.7% of pts. On-study death due to an AE occurred in 3.9% and 3.4% of pts. AEs of interest (grade ≥ 3) included hypertension (preferred term; 8.9% vs 2.6%), dyspnea (high-level term; 5.6% vs 2.2%), cardiac failure (grouped term; 4.8% vs 1.8%), and acute renal failure (grouped term; 4.1% vs 2.6%). Rates of grade ≥ 2 PN (grouped term) were 6.3% vs 32.0% (P< .0001). Conclusion: Kd demonstrated statistically significant and clinically meaningful superiority over Vd in RMM, with a two-fold improvement in median PFS. In addition, Kd had a favorable benefit-risk profile. These data suggest that K is a potential best-in-class agent for RMM. Clinical trial information: NCT01568866.