Survival among patients with relapsed/refractory diffuse large B cell lymphoma treated with single-agent selinexor in the SADAL study

Abstract

Patients with RR DLBCL who have received ≥ 2 lines of therapy have limited treatment options and an expected overall survival (OS) of < 6 months. The SADAL study evaluated single-agent oral selinexor in patients with RR DLBCL and demonstrated an overall response rate (ORR) of 29.1% with median duration of response (DOR) of 9.3 months. The analyses described here evaluated a number of subpopulations in order to understand how response correlates with survival outcomes in order to identify patients who could most optimally benefit from selinexor treatment. Median age was 67 years; 44.8% of patients were ≥ 70 years of age. The median OS was 9.0 months (95% CI 6.2, 13.7) at a median follow-up of 14.8 months. The median OS was not reached in patients with a CR or PR, while patients who did not respond have a median OS of 4.9 months (p < 0.0001). Patients < 70 years had an OS of 11.1 months compared with 7.8 months in patients ≥ 70 years. Among patients with or without prior ASCT, the median OS was 10.9 and 7.8 months, respectively. Among patients with disease refractory to the most recent DLBCL treatment regimen, the median OS was 7.0 months compared with 11.1 months for disease not refractory to the most recent treatment. In a patient population in which survival is expected to be < 6 months, treatment with single-agent oral selinexor was associated with a median survival of 9 months. Increased median OS observed in patients responding to selinexor was consistent across subgroups regardless of age, prior ASCT therapy, or refractory status. Randomized studies of selinexor in combination with a variety of other anti-DLBCL agents are planned. This trial was registered at ClinicalTrials.gov (NCT02227251) on August 28, 2014. https://clinicaltrials.gov/ct2/show/NCT02227251.

To the editor

Despite recent advances, nearly 50% of patients diagnosed with diffuse large B cell lymphoma (DLBCL) will succumb to their disease, with older age and comorbidities increasing risk of death and a median OS (~ 6 months) with relapsing disease after ≥ 2 prior therapies [1, 2]. XPO1 inhibition by selinexor, a first-in-class selective inhibitor of nuclear export compound results in cell cycle arrest; cells with DNA damage, including cancer cells, undergo apoptosis while sparing normal cells [3, 4]. Single-agent oral selinexor is approved for treatment of patients with DLBCL after ≥ 2 prior therapies [5]. Here, we have analyzed subgroups from the SADAL trial to understand how response correlates with survival outcomes following selinexor treatment.

Overall median OS was 9.0 months after a median follow-up of 14.8 months (95% CI: 13.2,21.7). Median OS in patients < 70 trended longer than patients > 70 (11.1 vs 7.8 months); HR 0.72 (0.46,1.13), p = 0.155. Patients with a best response of CR or PR on selinexor had a markedly longer median OS of 29.7 months, compared to those who did not respond (4.9 months) (p < 0.0001) (Table 1). Patients with lower baseline R-IPI (0–2) compared to R-IPI (3–5) had a significantly longer median OS (15.1 vs 4.6 months; HR 0.38 [0.24, 0.60], p < 0.0001) (Additional file 1: Figure S1). The majority (72%) of responding patients had lower R-IPI scores at baseline.

Table 1 Overall survival in subgroups

Patients who received selinexor after ASCT compared to those who did not had a median OS of 10.9 and 7.8 months, respectively; HR 1.39 (0.85,2.28), p = 0.185. Regarding response to most recent systemic therapy, patients who had a CR or PR trended toward longer OS (HR 0.71 [0.44, 1.17], p = 0.18; medians 11.1 and 7 months) than those who did not respond (Fig. 1; Additional file 1: Table S2). A significantly shorter median OS was observed in patients with relapse < 1 year from diagnosis (5.2 vs 13.1 months). Median OS was 6.6 and 10.9 months in patients that had relapsed/refractory disease < 6 or ≥ 6 months from last use of rituximab, respectively (HR = 1.30, p = 0.30) (Additional file 1: Figure S2). Four patients who had not achieved CR on their most recent prior therapy achieved CR on selinexor and two patients who did not have a CR or PR to any prior therapy had a CR (n = 2) or PR (n = 2) with selinexor. Median PFS for all patients was 2.6 months (summarized for subgroups-Additional file 1: Table S1).

Fig. 1

Overall survival by prior regimens and response. Kaplan–Meier curves according to A prior ASCT status; B prior use of lenalidomide; and C response to prior therapy: PR or CR and failure to reach PR or CR

The median OS associated with selinexor is consistent across the majority of the analyzed subgroups and also with the novel mechanism of action and the lack of apparent cross-resistance with this and other mechanisms. Results corroborate a retrospective study on patients with RR DLBCL after ASCT, which showed that the median OS was 6.6 months with cytotoxic chemotherapy compared to 11.3 months with novel agents [6]. While combination tafasitamab and lenalidomide showed an ~ 60% ORR for relapsed/refractory DLBCL and median OS was NR after a median follow-up of 19.6 months [7], it is difficult to compare to SADAL since 50% of patients had only one prior line of therapy compared to 3% on SADAL. In addition, only 18% of patients had primary refractory disease (relapse < 6 months of frontline therapy) compared to 47% of patients with available data on SADAL. Furthermore, 44% were refractory (i.e., relapsed < 6 months) to most recent therapy compared to 66.4% of SADAL patients. Further demonstration of the efficacy and safety consistent with the novel mechanism of action and lack of cross-resistance is observed with combination treatment of selinexor with backbone chemotherapy: 100% ORR with 90% CR using R-CHOP + selinexor as frontline treatment for DLBCL and follicular lymphoma [8] and 78% ORR using R-ICE + selinexor for relapsed/refractory DLBCL [9]. These combination results are consistent with a significant anti-DLBCL contribution of selinexor to standard chemotherapy and are being further evaluated.

Single-agent oral selinexor treatment was associated with a longer OS than expected based on contemporary case series [10,11,12] despite patient treatment, response history, age, and comorbidities. Given the beneficial impact of selinexor as a single agent and the poor prognosis of many patients, randomized studies of selinexor in combination with a variety of other anti-DLBCL agents are planned. Taken together, selinexor represents a safe, orally available option for patients whose disease has relapsed or is refractory to ≥ 2 prior therapies, including patients > 70 years old or those with significant comorbidities.