Nivolumab shows promise for advanced cervical cancer

CHICAGO — Nivolumab conferred clinical activity in patients with recurrent or metastatic cervical cancer and demonstrated a manageable safety profile across HPV–associated gynecologic cancers, according to results from CheckMate 358 presented at the ASCO Annual Meeting.

Perspective from Lainie P. Martin, MD

“Recurrent or metastatic cervical, vaginal and vulvar cancers have a poor prognosis, around 12 to 18 months, and a high unmet need,” Antoine Hollebecque, MD, senior medical physician at Gustave Roussy Cancer Institute in Villejuif, France, said during his presentation. “Second-line options for recurrent or metastatic cervical cancers results in a median PFS of around 2 to 4 months and an objective response rate of 0% to 14%.”

Antoine Hollebecque

HPV infection causes more than 90% of cervix cancers and 40% to 70% of vulva/vaginal cancers.

“HPV can evade host immune surveillance through increased expression of PD-L1, enabling viral persistence and the development of malignant lesions,” Hollebecque said. “PD-L1 has been shown to be a biomarker of HPV infection of the cervix and is significantly upregulated in cervical cancer.”

Nivolumab (Opdivo, Bristol-Myers Squibb) — which disrupts PD-1–mediated signaling — can restore antitumor immunity.

Hollebecque and colleagues evaluated data from 24 patients (median age, 51 years) with cervical (n = 19) or vaginal or vulvar cancer (n = 5) treated in CheckMate 358, an ongoing multicohort study of five virus-associated cancers. The study included PD-L1–unselected adults with relapsed and metastatic gynecologic cancers with an ECOG performance score of 0 to 1 who had received two or fewer prior systemic therapies.

All patients who underwent HPV testing were positive, with 41.7% of patients having an unreported HPV status. Ten patients had PD-L1 expression above 1%.

Patients received 240 mg nivolumab every 2 weeks until progression or unacceptable toxicity.

ORR and safety served as primary endpoints; secondary endpoints included duration of response, PFS and OS.

Median follow-up was 31 weeks (range, 6-38).

Researchers reported an ORR of 20.8%, which included one complete response (4.2%) and four partial responses (16.7%). Twelve patients (50%) achieved stable disease, for a disease control rate of 70.8%.

All response occurred in patients with cervical cancer (ORR, 26.3%) and occurred regardless of PD-L1 or HPV status.

“Exploratory analyses have evaluated the role of PD-L1 expression and HPV status on efficacy,” Hollebecque said. “Objective responses were observed among PD-L1–positive and –negative patients. The disease control rate was quite similar between the two groups and irrespective of PD-L1 expression. However, the limited number of patients in the two groups precludes any significant comparison.”

PAGE BREAK

Because all tested tumors were HPV positive, researchers could not compare these data with HPV–negative tumors, Hollebecque added.

ORR was 28.6% in patients who had received no previous systemic therapies and 17.6% in patients who had received one or more.

Median duration of response was 5.3 months (range, 1.9-7.1). At the time of the data cutoff, the five responding patients remained in response and 24 patients remained free from progression.

Median PFS was 5.5 months (95% CI, 3.5 to not reached), and median OS had not been reached.

Adverse events of any grade occurred in 70.8% of patients, including 12.5% that were grade 3 to grade 4.

“These CheckMate 358 results demonstrate the value of studying the potential of an immuno-oncology agent to address the significant challenge of treating patients with advanced cervical, vaginal and vulvar cancers,” Hollebecque said in a company-issued press release. “As a clinical investigator, I am encouraged by these findings in the women with advanced cervical cancer, and look forward to the anticipated data from the planned longer-term analyses.” – by Alexandra Todak

Reference:

Hollebecque A, et al. Abstract 5504. Presented at: ASCO Annual Meeting; June 2-6, 2017; Chicago.

Disclosure: Hollebecque reports consultant/advisory roles with Amgen, Merck and Serono. Please see the abstract for a list of all other researchers’ relevant financial disclosures.

Perspective

Lainie P. Martin, MD

These patients have a very high unmet need. For first metastatic recurrence of cervical cancer, we use cisplatin or carboplatin–paclitaxel with bevacizumab (Avastin, Genentech), and that has demonstrated a survival benefit based on data from Tewari and colleagues published in The New England Journal of Medicine. Beyond that, the response rates are very low. There has not been anything that shows the possibility for durable responses that some patients seem to enjoy when they respond to checkpoint inhibitors.

As a single agent, immunotherapies are helping a small percentage of patients across a broad range of diseases. However, these data in gynecologic cancers are important and foundational. We are trying to find that small fraction of patients who enjoy a durable response. Although only patients with cervical cancer demonstrated a response in this study, I would not discount that patients with vaginal and vulvar cancers could benefit, but there were too few in this study to determine that. With the 20% response rate in this study, there may not have been enough patients with vaginal and vulvar cancers to identify patients who would benefit. This is something that should be explored, and we need to enroll more patients into trials with rare tumors.

There is enough similarity between cervical, vaginal and vulvar cancers that it is fair to study them together, because researchers will look at the subsets of basket trials that include different disease sites separately. This is a reasonable approach to capture these rare patients, because if you run a trial separately in a rare disease, it may suffer from lack of accrual, the study may close and it will take much longer to assess whether a treatment may have benefit in a rare disease. Allowing a broader eligibility to the trials than we’ve done so far might be the best approach to getting data.

The mechanism of action of these immune checkpoint inhibitors is somewhat global; we see activity across a broad range of diseases, but only in a fraction of patients within a disease site. Early data for checkpoint inhibitors in head and neck cancer and lung cancer showed response rates in the 20% range — similar to what we saw in this study — but more importantly there is a smaller fraction of patients that had durable responses, and we need to develop this further to increase the proportion of patients with these significant benefits.

Data on duration of response in the current study is limited to date; because the study completed accrual in early 2016, there are a few patients out at 36 weeks on the swimmers plot. Researchers will continue to follow these patients to update us. Although responses are important, we need to build on this, as is really these durable responses that we’re aiming for with immunotherapy.

Reference:

Tewari KS, et al. N Engl J Med. 2014;doi:10.1056/NEJMoa1309748.

Lainie P. Martin, MD

Fox Chase Cancer Center

Disclosures: Martin reports no relevant financial disclosures.