KEYNOTE-826 data in cervical cancer and next steps for research

Pembrolizumab (Keytruda) plus chemotherapy with or without bevacizumab (Avastin) resulted in a statistically significant and clinically meaningful improvement in overall survival (OS) and progression-free survival (PFS) in patients with persistent, recurrent, or metastatic cervical cancer, according to Bradley J. Monk, MD, FACS, FACOG, who added that next steps will examine the addition of checkpoint inhibitors to chemotherapy and radiation.

Results from the phase 3 KEYNOTE-826 (NCT04221945) showed that pembrolizumab plus chemotherapy with or without bevacizumab resulted in a median PFS of 10.4 months (95% CI, 9.1-12.1) vs 8.2 months (95% CI, 6.4-8.4) with chemotherapy with or without bevacizumab in the all-comer population (HR, 0.65; 95% CI, 0.53-0.79; P < .001).¹ The median OS was 24.4 months (95% CI, 19.2–not reached) in the investigative arm vs 16.5 mmonths (95% CI, 14.5-19.4) in the control arm (HR, 0.67; 95% CI, 0.54-0.84; P < .001).

“[Other efforts are now adding] checkpoint inhibitors, such as pembrolizumab, to chemotherapy and radiation, [such as the] phase 3 study called KEYNOTE-818 [NCT04221945]. Hopefully, [with this approach, we will] cure more patients, which is really what we're trying to do,” Monk said. “I like helping [patients] live longer and I like helping them feel better, but really, the goal is to cure patients which is what the other sponsor is doing with durvalumab [Imfinzi]. Two studies are now examining frontline chemotherapy and radiation.”

In an interview with OncLive®, Monk, professor in the Division of Gynecologic Oncology at Arizona Oncology, University of Arizona College of Medicine, Creighton University School of Medicine at St. Joseph’s Hospital; medical director of the Gynecologic Program of the US Oncology Research Network; and co-director of GOG Partners, discussed the practice-changing data from KEYNOTE-826 examining pembrolizumab plus chemotherapy with or without bevacizumab and next steps for research in recurrent or metastatic cervical cancer.

OncLive®: What was the rationale for the phase 3 KEYNOTE-826 study?

Monk: The first pivotal trial in recurrent or metastatic cervical cancer [was done] in 2009, [and was] published in the Journal of Clinical Oncology by the Gynecologic Oncology Group, and it was called Protocol 204. That study showed that the best cytotoxic chemotherapy doublet was cisplatin and paclitaxel; that changed the world. Even to this day, it continues to be the global standard of the chemotherapy backbone. It's true that in 2015, the Japan Clinical Oncology Group, also in the Journal of Clinical Oncology, published a noninferiority trial that we could use carboplatin rather than cisplatin and then paclitaxel could be delivered safely and effectively at 175 mg/m².

The second concept was the addition of an antiangiogenic, which we all know as bevacizumab. In 2014, that [agent] became the first targeted therapy ever, in a gynecologic cancer, and it was added to this platinum/taxane chemotherapy backbone and was approved [for use] all over the world. [The regimen was] reimbursed, evolved, and transformed the standard of care from chemotherapy alone to chemotherapy plus bevacizumab in the appropriate patient; there are some contraindications. About 60% of patients in the world get bevacizumab added to their chemotherapy.

We have evolved the treatment paradigm again with KEYNOTE-826, which added pembrolizumab [to] chemotherapy with or without bevacizumab in the first-line setting. At the first interim analysis of this study, which enrolled 617 patients, it basically met all the therapeutic end points. It wasn't underperformance of the control arm. The bevacizumab [arm] showed a response rate of 48% [vs 50.8%] in the control arm; 48% vs 50% are [very similar]. However, when pembrolizumab was added, the response rate increased to 66%.

What really matters is tumor control, and the HR for progression-free survival was 0.65. Also, what also matters is survival, [that] patients live longer, and tolerance. No new safety signals [were observed] in the patient experience, and patient-reported outcomes look like they're better [with the addition of pembrolizumab]. Patients are having more tumor shrinkage, longer tumor control, living longer, and they're feeling better—that transforms the standard of care in a meaningful way.

The median in the intervention arm has not been met. At 2 years, 33% more of the patients, or a HR of 0.67, are alive. Think of it. At 2 years, half of the patients are still alive, which means half will live longer. We changed the standard of care. I'm so happy to see these [data] for patients because that's why we do [what we do]; I'm certain that this [approach will acquire] global regulatory approval and broad reimbursement.

What is the role of biomarkers in informing treatment with pembrolizumab?

Everyone knows that the accelerated approval of pembrolizumab in the second-line [setting] was restricted to PD-L1–positive patients with a CPS score of greater than 1. Interestingly, KEYNOTE-826 was positive in all-comers and the reason [for that] is because 89% of the patients were PD-L1 positive. There's no question that PD-L1 positivity is important and we should check it to help inform [treatment] decisions. However, if [a patient is not PD-L1 positive], you should still use pembrolizumab.

Eleven percent of patients were PD-L1 negative, and I presume most of [them had] adenocarcinomas. I suspect that many of those [patients] also have a high mutational burden or are mismatch repair deficient, [subsets in which] pembrolizumab is also an opportunity. Biomarkers are important, but we need better biomarkers moving forward.

Where will future research directions be focused?

With the resurgence of checkpoint inhibitors, the first thing [that will arise is] a sequencing question. Should we use checkpoint inhibitors in second-line [treatment]? Remember, [we saw that] cemiplimab is better than physician’s choice of chemotherapy [per data from the phase 3] EMPOWER-Cervical 1 [NCT03257267]. Or should we utilize all the agents up front in first line, chemotherapy with or without bevacizumab and pembrolizumab? Next will be tisotumab vedotin-tftv, an antibody-drug conjugate against tissue factor, which doesn't need a biomarker because tissue factor is expressed in virtually all cervical cancers.

Finally, it's all about prevention and screening. Like COVID-19, it's all about vaccination. The prevention of cervical cancer is about [administering] a vaccine, which…has been licensed since 2006; it’s broadly reimbursed, and incidentally, it's made by the same company as pembrolizumab.

The other evolution is that all these cancers are basically caused by the human papillomavirus [HPV] and we have evolved the pap smear to test for the virus on the cervix. Just like with COVID-19, you see if [a patient] has the virus, and if they don't, [then] they're good. We can do that in a woman's cervix, and it's now recommended as the primary screening technique [in women] over the age of 25 years. Women [should] have [this testing] as their primary modality of seeing whether they are at risk to testing for the HPV right on the cervix. Every woman [should] protect [themselves].

Is there anything else you would like to add?

All innovation to help patients is dependent on clinical trials. Clinical trials need 4 components: a patient, a doctor and investigator, a sponsor, and a network, which we have through the GOG or European Network. Then, they need an institution [or] a framework.

I need all of us to work together to take our patients and investigators and the trials that we have and teach our institutions, our hospitals, and universities. If we don't have clinical trials in our clinic, we cannot help patients. It is only through clinical trials that good things happen.

This article was initially published by our sister publication Onc Live.

Reference:

1. Colombo N, Dubot C, Lorusso D, et al. Pembrolizumab plus chemotherapy versus placebo plus chemotherapy for persistent, recurrent, or metastatic cervical cancer: randomized, double-blind, phase 3 KEYNOTE-826 study. Presented at: 2021 ESMO Congress; September 16-21, 2021; virtual. Abstract LBA2_PR.