A Phase II Trial of Selinexor, an Oral Selective Inhibitor of Nuclear Export Compound, in Abiraterone- and/or Enzalutamide-Refractory Metastatic Castration-Resistant Prostate Cancer.

LESSONS LEARNED: In abiraterone- and/or enzalutamide-refractory metastatic castration-resistant prostate cancer (mCRPC) patients, selinexor led to prostate-specific antigen and/or radiographic responses in a subset of patients, indicating clinical activity in this indication.Despite twice-a-week dosing and maximal symptomatic management, selinexor was associated with significant anorexia, nausea, and fatigue in mCRPC patients refractory to second-generation anti-androgen therapies, limiting further clinical development in this patient population.This study highlights the challenge of primary endpoint selection for phase II studies in the post-abiraterone and/or post-enzalutamide mCRPC space.
BACKGROUND: Selinexor is a first-in-class selective inhibitor of nuclear export compound that specifically inhibits the nuclear export protein Exportin-1 (XPO-1), leading to nuclear accumulation of tumor suppressor proteins.
METHODS: This phase II study evaluated the efficacy and tolerability of selinexor in patients with metastatic castration-resistant prostate cancer (mCRPC) refractory to abiraterone and/or enzalutamide.
RESULTS: Fourteen patients were enrolled. Selinexor was initially administered at 65 mg/m2 twice a week (days 1 and 3) and was subsequently reduced to 60 mg flat dose twice a week (days 1 and 3), 3 weeks on, 1 week off, to improve tolerability. The median treatment duration was 13 weeks. At a median follow-up of 4 months, two patients (14%) had ≥50% prostate-specific antigen (PSA) decline, and seven patients (50%) had any PSA decline. Of eight patients with measurable disease at baseline, two (25%) had a partial response and four (50%) had stable disease as their best radiographic response. Five patients (36%) experienced serious adverse events (SAEs; all unrelated to selinexor), and five patients (36%) experienced treatment-related grade 3-4 AEs. The most common drug-related adverse events (AEs) of any severity were anorexia, nausea, weight loss, fatigue, and thrombocytopenia. Three patients (21%) came off study for unacceptable tolerability.
CONCLUSION: Selinexor demonstrated clinical activity and poor tolerability in mCRPC patients refractory to second-line anti-androgenic agents. ©AlphaMed Press; the data published online to support this summary is the property of the authors.

Discussion

Exportin 1 (XPO‐1) is one of seven mammalian export proteins involved in the transportation of macromolecules from the nucleus to the cytoplasm, including tumor suppressor proteins p53, p73, BRCA1, p21, p27, IkB and FOXO transcription factors [1], [2], [3]. XPO‐1 is overexpressed in multiple malignancies, including prostate cancer. Selinexor is a potent and selective oral XPO‐1 inhibitor found to inhibit tumor growth and incidence of metastasis in preclinical models of prostate cancer [4], [5], [6], [7].

This single‐center phase II study enrolled patients with progressive mCRPC. Eligibility criteria included prior progression on abiraterone, enzalutamide, other second‐generation investigational anti‐androgen/androgen receptor (AR)‐targeted therapies (e.g., apalutamide), or their combination. Prior chemotherapy for mCRPC was not allowed. The primary endpoint was radiographic progression‐free survival (rPFS).

A total of 14 patients were enrolled between October 2014 and October 2016. The median age was 72 years (range: 56–79). The median baseline PSA was 73.8 ng/mL (range: 12.8–686.9; Table 1). The first two patients received selinexor 65 mg/m2 p.o. twice a week (days 1 and 3), one of whom came off study for a serious adverse event (grade 4 psychosis, related to dexamethasone prophylaxis for nausea and anorexia), and the other came off study for unacceptable treatment‐related adverse events (AEs; grade 3 anorexia, nausea, vomiting, and weight loss). The study protocol was subsequently revised to reduce the selinexor dose to 60 mg flat dose twice a week (days 1 and 3), 3 weeks on, 1 week off.

Table 1.

Demographic and baseline characteristics

Selinexor dose: 65 mg/m2 twice a week (days 1 and 3) for the initial two patients enrolled. Subsequently reduced to 60 mg flat dose twice a week (days 1 and 3), 3 weeks on, 1 week off, for the remaining 12 patients enrolled.

Three patients had received abiraterone in the context of a phase II study of increased‐dose abiraterone after progression on standard‐dose abiraterone (NCT01637402). One patient had received concurrent abiraterone and enzalutamide in the context of a phase III randomized clinical trial (NCT01949337). No patient had received other second‐generation investigational anti‐androgen/AR‐targeted therapies, including ARN‐509 or Apalutamide.

No patient had received prior chemotherapy for mCRPC prior to study enrollment. Two patients had received prior docetaxel for metastatic hormone‐sensitive prostate cancer.

Abbreviations: AR, androgen receptor; ECOG, Eastern Cooperative Oncology Group; LDH, lactate dehydrogenase; mCRPC, metastatic castration‐resistant prostate cancer; PSA, prostate‐specific antigen.

The median treatment duration was 13 weeks (range: 4–48). At a median follow‐up of 4 months, two patients (14%) had ≥50% PSA decline, and seven patients (50%) had any PSA decline (Fig. 1A). Only one patient (7%) had ≥50% PSA decline after 12 weeks of therapy. The median time to PSA progression was 12 weeks (range: 7–14). Of eight patients who had measurable disease at baseline; two (25%) had partial response and four (50%) had stable disease as best radiographic response. All fourteen patients had bone metastasis at baseline; four (29%) had confirmed radiographic progression on bone scan, with a median time to progression of 31 weeks (range: 7–47).

Figure 1.

Selinexor treatment results. (A): Waterfall plot of maximal PSA decline during selinexor treatment. Dotted line indicates the threshold for definition of PSA response (≥50% PSA decline from baseline). (B): Swimmer plot of individual patient experience on selinexor. Time to PSA progression determined by PCWG2. Time to radiographic progression determined by RECIST v1.1 for soft tissue lesions and PCWG2 for bone lesions.

Abbreviations: NLCB, no longer clinically benefiting; PCWG2, Prostate Cancer Working Group 2; PSA, prostate‐specific antigen; SAE, serious adverse event.

Five patients (36%) experienced SAEs (psychosis, diplopia, orthostatic hypotension, congestive heart failure, and angina), all of which were deemed unrelated to selinexor. Five patients (36%) experienced treatment‐related grade 3–4 AEs (anemia, n = 3; nausea, n = 2; vomiting, n = 2; anorexia, fatigue, thrombocytopenia, neutropenia, hypophosphatemia, n = 1 each). The most common drug‐related AEs of any severity were anorexia (86%), nausea (64%), weight loss (50%), fatigue (50%), and thrombocytopenia (50%). Three patients (21%) came off study for poor tolerability (Fig. 1B).

Although oral selinexor showed clinical activity, the study was terminated early due to the significant AE profile observed in the study population. The primary endpoint of rPFS is not reported because limited sample size and high rate of censoring precluded accurate estimation.

Trial Information

Prostate cancer

Metastatic/advanced

No designated number of regimens

Phase II

Single arm

Progression‐free survival

PSA decline (at least 50%) at 12 weeks

Time to PSA progression (PCWG2 criteria)

Time to confirmed radiographic progression on bone scan (PCWG2 criteria)

Safety

Pharmacodynamics

Pharmacokinetics

Clinical activity and significant toxicity profile

Drug Information for Phase II Selinexor

Selinexor

Karyopharm Therapeutics

Small molecule

60 mg/m2

p.o.

Patient Characteristics for Phase II Selinexor

14

0

Metastatic, castration resistant

Median (range): 72 years (56–79 years)

Median (range): 2 (1–4)

0 — 9

1 — 5

2 — 0

3 — 0

Unknown — 0

Cancer types or histologic subtypes: Prostate

Primary Assessment Method for Phase II Selinexor

PCWG2 for bone lesions

19

14

14

10

PCWG2 for bone lesions

n = 0 (0%)

n = 9 (64%)

n = 1 (7%)

Total patient population: RECIST v1.1 for soft tissue lesions

6

RECIST 1.1

n = 0 (0%)

n = 2 (25%)

n = 4 (50%)

n = 0 (0%)

Adverse Events

Serious Adverse Events

Assessment, Analysis, and Discussion

Study terminated before completion

Toxicity

Not collected

Clinical activity and significant toxicity profile

In this phase II trial, selinexor was found to have clinical activity with significant nausea, anorexia, fatigue, and weight loss in abiraterone‐ and/or enzalutamide‐refractory metastatic castration‐resistant prostate cancer (mCRPC) patients. Of the 14 patients treated, 3 patients withdrew from study for treatment‐unrelated SAEs, and 3 patients withdrew from study for treatment‐related AEs, translating to a withdrawal rate due to selinexor of 21%. Although tolerability was improved by protocol dose reduction of selinexor from 65 mg/m2 twice a week to 60 mg flat dose twice a week, 3 weeks on, 1 week off, 2 of the 12 patients receiving the latter dose (17%) withdrew from study for unacceptable adverse events.

The study protocol originally planned to accrue up to a total of 54 patients and specified early stopping if after the initial accrual of 18 patients, 10 or fewer patients (≤56%) were classified as progression free at 16 weeks, or if more than 6 patients (>33%) experience unacceptable toxicity. The study was terminated early based on the assessment of tolerability and efficacy data in the first 14 patients accrued. All patients received maximal symptomatic management, including nausea prophylaxis with ondansetron and olanzapine and, if needed, dexamethasone, transfusions for cytopenias, dose reductions, and drug holidays.

The primary endpoint selected for this study was radiographic progression‐free survival (rPFS), one of the coprimary endpoints in the COU‐AA‐302 study [8]. Of note, all seven patients with reductions in prostate‐specific antigen (PSA) had mCRPC refractory to both abiraterone and enzalutamide. Two patients (of eight) with measurable disease at baseline had partial responses by RECIST criteria. Two patients (14%) remained on study for close to 48 weeks and eventually came off study for radiographic progression. It is worthwhile to note that both patients experienced a long interval of PSA progression (9–10 months) before confirmed radiographic progression on bone scan by PCWG2 criteria (Fig. 1B). One patient had an increase in PSA from 41 ng/mL at baseline to 342 ng/mL at the time of study discontinuation, without any PSA decline. The other patient had a maximal PSA decline of −32%. These observations raise the question of optimal primary endpoint selection for phase II studies in post‐abiraterone and/or post‐enzalutamide patients. Perhaps a PSA‐based endpoint could improve efficacy and safety signal detection in novel drug development in this disease space; however, the best approach remains to be elucidated.

Abiraterone acetate and enzalutamide are important treatment options for men with mCRPC. However, approximately 30% of chemotherapy‐naïve and 60% of postchemotherapy patients do not achieve PSA decline (i.e., primary resistance), and essentially all patients who initially benefit from treatment eventually develop progressive disease (i.e., acquired resistance) [8], [9], [10], [11]. Furthermore, the emergence of resistant disease is often accompanied by an aggressive clinical course and poor prognosis, with a median overall survival of 12–18 months. The development of novel therapeutics for mCRPC patients after progression on abiraterone and enzalutamide should continue to be a research priority.

It is recognized that there is a significant degree of cross‐resistance between abiraterone and enzalutamide, and mechanisms of resistance include constitutively active androgen receptor (AR) splice variants that effectively bypass the AR signaling pathway [12], [13], [14]. Whether targeting nuclear export with XPO‐1 counteracts this and other mechanisms of resistance would be interesting to explore but is beyond the scope of this study. However, the current data do demonstrate clinical activity of XPO‐1 inhibition in patients with mCRPC resistant to second‐generation anti‐androgen therapies, supporting further research in this area.

XPO‐1 inhibition with selinexor is being actively investigated in multiple solid and hematologic malignancies and has demonstrated promising clinical activity in advanced refractory hematologic malignancies in early phase trials, with response rates ranging from 30% to 50% [15], [16], [17]. Although it is possible that differences in tissue distribution, tumor penetration, and biology across and within tumor types (i.e., differential dependence on XPO‐mediated nuclear export for survival) contribute to variations in clinical activity, pharmacodynamic studies evaluating XPO‐1 expression and clinical response were not performed in this study given early trial termination. It is possible that baseline fatigue and other comorbidities, bony metastases (present in all patients in this study), prior therapies, and/or other medical conditions in mCRPC patients contributed to the reduced tolerability in this population relative to other patients with solid tumors.

The adverse event profile observed in this study hinders further clinical development of twice‐weekly selinexor in patients with progressive mCRPC. In contrast to this study, selinexor was found to be tolerable in patients with advanced refractory bone or soft tissue sarcoma when administered at different twice‐weekly doses (30 mg/m2, 50 mg/m2, or 60 mg flat dose) on a continuous schedule (no break) [18]. The sarcoma patients treated were younger, with a median age of 55 years (range: 18–86), and likely had fewer comorbidities and better bone marrow reserve, which could have contributed to the differences in tolerability observed in the studies. Indeed, the overall frequency of weight loss and anorexia was higher in the current mCRPC study. Selinexor at doses of 50 mg/m2 twice weekly, 35 mg/m2 twice weekly, and 50 mg/m2 weekly without break were also found to be relatively well tolerated in patients with advanced treatment‐refractory gynecological cancers [19].

Second‐generation selective inhibitor of nuclear export (SINE) compounds (e.g., KPT‐8602) are currently being developed, which have reduced brain penetration and thereby reduced central nervous system‐mediated side effects, including anorexia, fatigue, and weight loss [20], [21]. The improved adverse event profile of KPT‐8602 may allow for daily administration compared with twice‐a‐week dosing for selinexor, which may increase therapeutic effect. Consistent with this, early results from a phase I/II study of KPT‐8602 in patients with relapsed/refractory multiple myeloma in which KPT‐8602 was dosed daily (QDx5) in 28‐day cycles was well tolerated [22]. The investigation of second‐generation SINE compounds in advanced prostate cancer, including incorporating such agents into combination strategies, appears to be worthwhile given the clinical activity of selinexor observed in this study.

All adverse events are at least possibly related to study therapy, graded according to National Cancer Institute Common Terminology Criteria for Adverse Events v4.0.

Abbreviation: NC/NA, no change from baseline/no adverse event.

Psychosis was attributed to prophylactic dexamethasone, and diplopia was attributed to disease progression.