KEYNOTE-057 Results Add New Treatment Option for Certain Patients with Bladder Cancer

Petros Grivas, MD, PhD

New data from cohort B of the phase 2 KEYNOTE-057 trial (NCT02625961) showed that pembrolizumab (Keytruda) can be an effective treatment for patients with bacillus Calmette-Guérin (BCG)-unresponsive, papillary high-risk non–muscle-invasive bladder cancer (NMIBC).¹

After about 45 months of follow up, researchers found that the use of the checkpoint inhibitor allowed 43.5% of patients to remain disease free for a year after treatment initiation on pembrolizumab. Moreover, among 132 patients enrolled on the study with high-grade Ta or any T1 stage disease had a durable disease control rate on pembrolizumab.

In a discussion with Targeted Oncology^TM, Petros Grivas, MD, PhD, clinical director of the genitourinary cancers program at the University of Washington Medicine and professor in the clinical research division of the Fred Hutch Cancer Center, explained how the KEYNOTE-057 trial results for these patients have provided a new treatment option for physicians to consider.

Please describe the design and endpoints of the study.

Here we were focusing specifically on cohort B [of the study]. [These were] patients with papillary high-risk NMIBC without CIS. Patients should have had a TURBT within 3 months prior to the first pembrolizumab dose on the trial, and they were continued on treatment with pembrolizumab on the standard dose of 200 milligrams [given intravenously] every 3 weeks for up to 2 years. The first assessment was at 3 months, and the second assessment at 6 months, and in that particular [part of the] study, when it was designed, there were no mandatory biopsies.

These assessments were done with urine cytology, cystoscopy every 3 months for the first couple of years, and then every 6 months through year 5, and of course, CAT scan imaging [was done] every 6 months for the first 2 years, and then yearly thereafter. The other relevant point to make is that, as I mentioned, pembrolizumab was continued up to 2 years and if someone did not achieve a complete response regarding progression of high-risk disease, at any time, they were discontinued from treatment.

The efficacy analysis [looked at] assigned patients who received at least 1 dose for pembrolizumab, and the same was the case for safety. The primary endpoint was the disease-free survival rate for higher risk non muscle disease in 1 year, and safety [along with] many secondary endpoints.

What were the patient characteristics of the study?

Overall, the patient population is exactly what you expect in this treatment setting. Most patients had an EGOG performance score of 0-1 and there was some representation from different racial groups, but two thirds of patients we're white, and 17% Asian population, but some of them had missing race [data]. Overall, we make every effort to improve an increase accrual of a different racial and ethnic groups in clinical trials to of course, support diversity, inclusion, and equity.

The median number of priority installations with PCG was 10, and about half of the patients had T-1 [disease] and the other half had TA-high-grade. About a quarter of patients were from the US and three quarters were from outside the US. There were different entry status for baseline Harrison muscles invasive disease, with 6% had recurrent disease after adequate visited fitment. About a quarter had persistent high risk NMIBC, and about 14.14% had progressive disease after adequate treatment.

What were some of the key highlights of the study?

The disease-free survival [DFS] rate in one year was 43.5%. So, a little bit less than half of the patients had a DFS for high risk disease specifically, at 1 year, and at the median DFS for high risk and it may be six or 7.7 months. Now, if you look at the longer follow up at 3 years, 35% of patients about 1/3 were disease free for high-risk disease. And if you now take in disease, high risk and low risk, the survival rate was 42% in 1 year and about 33% in 3 years.

The survival rate for high-risk disease seems consistent across different subsets of patients and across the different age groups, gender, race, region, performance status, and T1 and TA different tumor patterns, as well as the baseline high risk on muscle wasting disease status. I would say few patients had progressive disease baseline, but most of them had persistent or recurrent disease.

We also looked at the progression free survival [PFS] towards an ingrained stage of death and PFS to invasive or metastatic disease or death, and those numbers looked high in terms of a PFS rate in one year, this was at 88%. So almost 9 out of 10 patients had no worsening of grade stage or death, or no progression to invasive or metastatic disease or death in 1 year. [Regarding] overall survival, 96% of patients were alive in 1 year.

What was the safety data of this study?

In terms of toxicity, there were no new safety signals. What you expect from pembrolizumab is what we saw, the grade 3 or 4 treatment related AEs [TRAEs] was about 14%, and we had 13% serious TRAEs. [Moreover,] about 1 out of 10 patients discontinued pembrolizumab because of a TRAE and there were no treatment related deaths in the study because of toxicity. The most common immune related adverse events [irAEs], included puritis, hyperthyroidism, and with hypothyroidism there was fatigue that sometimes we see with checkpoint inhibition.

Less than 10% of patients had diarrhea, ischemia, joint pains, rash, and some other rarer AEs. Overall, the treatment was well tolerated and the quality-of-life data showing that the quality of life of those patients seems to be preserved over time, again, going together with the toxicity profile of a checkpoint inhibitor in this setting, and the baseline good performance status of the patients.

What are your takeaways from this study?

This dataset is probably one of the most robust, I would say, in terms of sample size, follow up duration, that in some cases approach no more than 3 or 4 years. I think this relevant in terms of robustness of the follow up time. So, the question here is whether pembrolizumab can be used in this NBCIM papillary tumors without CIS, and I think that's a lot of great discussion that happened at [ASCO GU 2023]. It's exciting to see the improvements in the field, excited to have more options for our patients, and I’m excited to present this data.

^Reference

^{Necchi A, Roumiguié M, Esen AA, et al. Pembrolizumab (pembro) monotherapy for patients (pts) with high-risk non–muscle-invasive bladder cancer (HR NMIBC) unresponsive to bacillus Calmette–Guérin (BCG): Results from cohort B of the phase 2 KEYNOTE-057 trial. J Clin Oncol. 2023;41(suppl 60):LBA442. doi:10.1200/JCO.2023.41.6_suppl.LBA442}