Selinexor demonstrates antitumor activity in heavily pretreated ovarian, endometrial cancers

Issue: November 25, 2016

COPENHAGEN, Denmark — Single-agent selinexor demonstrated antitumor activity in heavily pretreated patients with ovarian or endometrial cancers, according to phase 2 study results presented at the European Society for Medical Oncology Congress.

Perspective from Charlie Gourley, BSC, MB ChB, MRCP, PhD, FRCP

However, patients with cervical cancer did not demonstrate as much benefit as the other cohorts, results showed.

Ignace B. Vergote

Selinexor (KPT-330, Karyopharm Therapeutics) is a first-in-class inhibitor of XPO1, a nuclear export protein that is expressed in aggressive ovarian cancers and is linked to poor outcomes. Further, data from The Cancer Genome Atlas indicates XPO1 is common in patients with endometrial cancer.

Ignace B. Vergote, MD, PhD, head of the department of obstetrics and gynecology and of gynecologic oncology at Catholic University of Leuven in Belgium, and colleagues evaluated the safety and efficacy of selinexor in 66 patients (median age, 62 years) with ovarian cancer, 23 patients (median age, 67 years) with endometrial cancer and 25 patients (median age 53 years) with cervical cancer. All patients with ovarian cancer were refractory to platinum chemotherapy, and all patients with endometrial and cervical cancers had received at least one prior line of chemotherapy.

The median number of prior treatment regimens were six (range, 1-11) in patients with ovarian cancer, two (range, 1-5) in patients with endometrial cancer and three (range, 1-8) in patients with cervical cancer.

Researchers randomly assigned patients with ovarian cancer to one of three treatment schedules: 50 mg/m² twice weekly, 35 mg/m² twice weekly or 50 mg/m² every week in 4-week cycles. All other patients received selinexor 50 mg/m² twice weekly.

Disease control rate — or the incidence of complete response, partial response and stable disease at week 12 — served as the study’s primary endpoint. Secondary endpoints included overall response rate, PFS, duration of response, safety and tolerability.

The 12-week disease control rate was 49% in patients with ovarian cancer and appeared comparable across dosing schedules. Patients with endometrial cancer demonstrated a disease control rate of 45%, and patients with cervical cancer had a 6% disease control rate.

“We were quite surprised we saw the worst disease control rate in patients with cervical cancer, despite the fact that this is an HPV–induced tumor, where we expected more responses than in the two other cohorts,” Vergote said during his presentation.

The overall response rate was 14% among both patients with ovarian cancer, 15% among patients with endometrial cancer, and 4% among patients with cervical cancer.

Median PFS was 3 months among patients with ovarian and endometrial cancers, and 1 month among patients with cervical cancer. Median OS was 7 months in patients with ovarian cancer, 8 months in patients with endometrial cancer and 5 months in patients with cervical cancer.

Fifteen patients received treatment for more than 6 months, including four patients who have been treated for more than 1 year.

“The most important outcomes were the disease control rate, and that so many patients could tolerate treatment for longer periods despite having been heavily pretreated,” Vergote said.

Researchers also analyzed patients’ blood samples for circulating tumor cells (CTCs). Results showed that patients with CTCs prior to treatment tended to have shorter PFS.

Common grade 1 and grade 2 drug-related adverse events that occurred in all patients included nausea (56%), anorexia (47%), weight loss (44%) and fatigue (42%). Grade 3 drug-related events included thrombocytopenia (11%), fatigue (10%), anemia (9%) and nausea (8%).

One patient each experienced grade 4 cataract and hyponatremia.

Grade 3 to grade 4 toxicities were reduced in the once-weekly 50 mg/m² dosing compared with twice-weekly dosing, Vergote said.

Based on these results, phase 3 trials are planned to evaluate selinexor in patients with ovarian and endometrial cancers.

“Single-agent selinexor has interesting antitumor activity in heavily pretreated ovarian and endometrial cancers,” Vergote said. “With this data, cervical cancer will not be a priority for future research, but endometrial cancer will be.” – by Alexandra Todak

Reference:

Vergote I, et al. Abstract 854O. Presented at: European Society for Medical Oncology Congress; Oct. 7-11, 2016; Copenhagen, Denmark.

Disclosure: The study was funded by Karyopharm Therapeutics. Vergot reports no relevant financial disclosures. Other researchers report employment or board of directors roles with and stock ownership in Karyopharm Therapeutics.

Perspective

Charlie Gourley, BSC, MB ChB, MRCP, PhD, FRCP

This is a very exciting new agent, a first in its class. The mechanism of its action fits with biology of high-grade serous ovarian cancer, and when you think about cervical cancer and the interaction between some of the vital oncoproteins with some of the proteins in the cells, such as p53 and retinoblastoma, you would think it would have efficacy in that setting as well. It seems reasonably well tolerated, particularly in the once-weekly regimen. Clinical benefit was demonstrated in very heavily pretreated patients, which was really important, and there was evidence of a cohort who do exceptionally well.

However, one limitation is the lack of correlation of response with tumor biology, whether that be in the patients who responded to the drug, or in the exceptional responders. However, this work could be going on in the background.

What really intrigues me about this study is asking what is this drug actually doing? The idea of it blocking a plethora of tumor-suppressor genes may be happening, but it must be blocking many other things as well. If we think about high-grade serous ovarian cancer, which of the subgroups would benefit?

We know there are p53 mutations in almost everyone, but some of those are activating mutations. So, blocking them from coming out of the cell nucleus may not be a good thing to do, although it might be, depending on the exact mutation.

We can’t imagine that keeping p53 null, where they’ve lost the expression, is a mechanism of efficacy, and the same for retinoblastoma, where a lot of the mutations cause loss of expression completely. It could be that keeping some of the missense mutations in retinoblastoma in the nucleus might be beneficial, and the same might be true for BRCA1/2.

There is still a bit of work to do to work out which patients benefit and what the mechanism is. Nonetheless, the response in these heavily pretreated patients is really quite interesting.

Charlie Gourley, BSC, MB ChB, MRCP, PhD, FRCP

Edinburgh Cancer Research Centre

Disclosures: Gourley reports no relevant financial disclosures.