ODAC Recommends Delaying Approval of Selinexor

/by

Committee: Oncologic Drugs Advisory Committee (ODAC)

Indications: the treatment of patients with relapsed refractory multiple myeloma who have received at least three prior therapies and whose disease is refractory to at least one proteasome inhibitor, at least one immunomodulatory agent, and an anti-CD38 monoclonal antibody

Question 1: Should the approval of selinexor be delayed until results of the randomized phase 3 trial, BOSTON, are available?

Vote Results: Yes 8 No 5

Drug/Device: Drug

Sponsor: Karyopharm Therapeutic, Inc.

FDA Documents: 

Selinexor Background:

Most patients being treated for multiple myeloma are likely to relapse and develop refractory disease. In general, the duration of remission shortens with each subsequent line of therapy, and patients who become refractory to the major classes of available anti-myeloma therapies suffer from poor outcomes. The applicant submitted an accelerated New Drug Application (NDA) for selinexor tablets to be used in combination with dexamethasone to treat patients with RRMM who have received at least three prior therapies and whose disease is at least refractory to one proteasome inhibitor (i.e. bortezomib and/or carfilzomib), at least one immunomodulatory agent (i.e. lenalidomide and/or pomalidomide), and an anti-CD38 monoclonal antibody (i.e. daratumumab). The primary outcome being evaluated was the overall response rate (ORR) – defined as the proportion of patients with a partial response (PR) or better. Key secondary endpoints included duration of response (DOR), progression-free survival (PFS), and overall survival (OS). The evidence of support came from a phase 2b single-arm (STORM) trial of a combination regimen of selinexor and dexamethasone.

 

Unique Challenges:

Study design and efficacy:

The applicant submitted evidence from a single arm trial to support the accelerated NDA. FDA routinely considers single arm trials with a disease response rate as the primary endpoint, supported by duration of response – both of which were reported by the applicant in their primary and key secondary endpoints. However single arm trials cannot adequately characterize time-to-event endpoints such as progression-free survival (PFS) and overall survival (OS), and risk-benefit assessment in single arm trials is challenging due to the lack of counterfactuals.

Although the primary endpoint of STORM achieved ORR of 25.4%, treatment effect of selinxor alone could not be isolated. For one, selinexor failed to demonstrate single agent activity in the preceding phase 1 (STOMP) trial. There was also historical evidence of dexamethasone in treatment of RRMM with ORR of 10-27% when administered at high doses.

The application also included a retrospective observational study through use of the Electronic Health Record (a.k.a. Real-World Data) characterizing the survival distribution of patients similar to that of the STORM. The purpose of the retrospective study was to form an indirect comparator for the trial. However, without having reviewed and consented the protocol and SAP, FDA can not be certain that protocol and SAP were pre-specified and unchanged during the data selection and analyses. Also for inclusion/exclusion criteria, there were big differences between STORM and RWD; the FDA considered RWD are not comparable to those obtained in STORM. Therefore, upon reviewing of submitted evidence, the agency concluded that the study is inadequately designed and conducted to serve as the comparison group to STORM due to lack of pre-specified analyses and various methodological issues such as selection bias, misclassification and immortal time bias.

Toxicity/risk:

Treatment with selinexor was associated with significant toxicity. All of the trial participants experienced at least one episode of treatment-emergent adverse event (TEAE), and nearly 2/3 of the participants experienced serious adverse events (SAE). Most (88.6%) participants required dose modification, and over 1/4 (28.6%) of the participants stopped treatment early due to TEAE. In another RCT (SOPRA) that investigated the effect of selinexor vs physician’s choice (PC) in treatment of refractory acute myeloid leukemia (AML), patients in the selinexor arm exhibited higher remission and worsened overall survival compared with PC; underscoring the need of RCT to fully characterize risk-benefit of selinexor. Based on overwhelming evidence of toxicity and lack of single agent activity of selinexor, the ODAC voted against accelerated approval of the applicant’s NDA for selinexor in treatment of triple-class refractory RRMM.

Next Steps: