Real-World Evidence of the Use of Approved Carfilzomib Regimens in Patients Previously Exposed or Refractory to Lenalidomide: Updated Results from a Prospective Observational Study

Background: With approvals of lenalidomide (Len) based therapy for the treatment of transplant ineligible patients (pts) with newly diagnosed multiple myeloma, and for pts receiving maintenance following transplant, there is an increasing use of Len as induction and maintenance therapy. As a result, treatment options for pts who become Len-refractory (LR) has become an emerging critical challenge. We report updated results from a large real-life cohort of pts by LR status, who received approved K regimens (K, lenalidomide and dexamethasone [KRd] at the dose of 27 mg/m² twice-weekly or Kd at the dose of 56 mg/m² twice-weekly).

Methods: In 11 participating countries across the EU and Israel, adults with relapsed/refractory multiple myeloma (RRMM), who received ≥1 dose of a K regimen in routine clinical practice were included in the prospective cohort study (NCT03091127). Cytogenetic risk status was as assessed by the physician and frailty at K initiation was calculated according to Palumbo et al., Blood 2015;125:2068-74 and Facon et al., Blood 2015;126:4239. Pts were classified as LR if they met at least one of the following 3 criteria to any prior regimen containing Len, as defined by the International Myeloma Workshop Group (Rajkumar et al., Blood 2011;117:4691-95): best response was stable or progressive disease; progression was the reason for stopping any regimen; or/and date of relapse/progression was after the start date and within 60 days after the last dose of Len.

Results: At the data cut-off of 17 March 2020, the study included 382 KRd pts and 273 Kd pts. At K initiation, nearly one-fifth (19%, n=74) of KRd pts and over half (55%, n=151) of Kd pts were LR.

In the KRd cohort, LR pts were older than non-LR (NLR) pts (median: 66 vs 64 years). At KRd initiation, characteristics of LR pts were comparable with NLR pts in terms of cytogenetic risk (over 40% at high-risk of 34% reported overall) but were more frail (43% vs 28% of 67% reported) and with higher ISS stage (ISS 3: 28% vs 22% of 30% reported). LR pts had a median of 3 prior lines of therapy vs 1 for NLR pts and were more likely to have received a monoclonal antibody (mAb) (20% and 6%, respectively). Most LR pts (60%) became refractory in their last prior therapy of whom 55% became refractory from their 2^nd line (2L) therapy.

In the Kd cohort, LR pts were older than NLR pts at K initiation (median: 69 vs 67 years). At Kd initiation, more LR pts had a high cytogenetic risk than NLR pts (71% vs 33% of 18% reported) but were more fit (22% vs 7%, of 71% reported) and had a lower ISS stage (ISS 1: 41% vs 31% of 35% reported). LR pts had received a median of 3 prior lines vs 2 for non-LR pts and were more likely to have received a mAb (41% and 12%, respectively). In LR pts, 43% became refractory in their last prior therapy of whom 52% became refractory from their 2L therapy.

Among 93% of pts with an assessment overall, high overall response rates (ORR) were achieved in NLR pts for both KRd and Kd (Table). LR pts achieved lower ORRs in both cohorts although there are differences in pt population and treatment history. Despite this, almost half of LR pts in the KRd cohort and 30% in the Kd cohort achieved a very good partial response or better. The clinical benefit rate for LR pts was 72% and 62% in the KRd and Kd cohorts, respectively.

The mean K relative dose intensity was comparable for LR and NLR pts (90% vs 88%) in the KRd cohort and slightly lower for LR (74%) vs NLR (80%) in the Kd cohort. The Kaplan-Meier estimate of median treatment duration was shorter for the LR vs NLR subgroups (KRd: 11.1 [95% CI: 5.7-13.9] vs 16.2 [95% CI: 13.5 – 16.9] months; Kd: 6.3 [95% CI: 4.7-7.7] vs 9.7 [95% CI: 7.4-11.6] months).

The occurrence of adverse events of ≥grade 3 (Gr3+ AEs) was comparable between KRd subgroups (49% vs 47%, LR and NLR). Kd NLR pts experienced less Gr3+ AEs than LR pts (39% vs 50%). LR pts in both cohorts experienced more SAEs and fatal events than NLR pts (SAEs: KRd: 38% vs 28%; Kd: 39% vs 33%; fatal events: KRd: 7% vs 3%; Kd: 9% vs 7%), although there are differences in pt population and treatment history.

Conclusion: In this real-world cohort, LR pts had received more prior lines of therapy and were exposed to more agents than NLR pts. Fair responses were achieved for this difficult to treat population in this unadjusted analysis, which are consistent with published subgroup analyses of the clinical trials. This real-world data further supports the use of carfilzomib for RRMM and emphasizes the strength of Kd as a strong back-bone treatment for LR pts.