Pembrolizumab Update Solidifies Survival Benefit in Bladder Cancer

Joaquim Bellmunt, MD, PhD, discusses the updated pembrolizumab findings, as well as the future of immunotherapy in bladder cancer.

Joaquim Bellmunt MD, PhD

In patients with locally advanced or recurrent urothelial cancer, pembrolizumab (Keytruda) sustained its improvement in overall survival (OS) compared with chemotherapy in a 2-year update of the phase III KEYNOTE-045 trial presented at the 2018 Genitourinary Cancers Symposium.

The KEYNOTE-045 update showed a median OS of 10.3 months after a median follow-up of 27.7 months for patients treated with pembrolizumab.1 This was in comparison with 7.3 months for those treated with chemotherapy (HR, 0.70; P =.00017). Previously reported results for the study showed that median OS was 10.3 months with pembrolizumab versus 7.4 months for chemotherapy (HR, 0.73, P = .0022).2

OncLive: Can you give some background on KEYNOTE-045?

In an interview with OncLive, lead KEYNOTE-045 investigator Joaquim Bellmunt, MD, PhD, director of the Bladder Cancer Center, Dana-Farber Cancer Institute, discussed the updated pembrolizumab findings, as well as the future of immunotherapy in bladder cancer.Bellmunt: Presently, we have 5 drugs that are approved for treating second-line patients with bladder cancer. Pembrolizumab was shown to be effective in a phase I trial, and then we decided to jump to a phase III study because of the positive results observed in the phase I trial. We started to accrue patients in 2014 and, after 1 year, the trial was completed. A total of 550 patients who failed on first-line platinum-based chemotherapy in the setting of metastatic disease were randomized to receive pembrolizumab or chemotherapy.

What were the updated results?

Chemotherapy was the standard of care being used at that time; it could have been docetaxel, paclitaxel, or vinflunine in Europe. The main endpoint of this trial was OS and PFS in the overall patient population, as well as in patients who had PD-L1 staining of more than 10% based on composite positive score that was defined by Merck for pembrolizumab. After 1 year of accrual, the first readout of this trial was in September 2016 and was presented at the 2016 SITC Annual Meeting. It was published in the New England Journal of Medicine in March 2017. At the time that the trial was published, the median follow-up was 14 months. What we presented at the 2018 Genitourinary Cancers Symposium are the updated results after a median follow-up of 27 months. We believe that the trial is now mature, and we can see if the benefits observed at the time of publishing are maintained, and whether there are any concerns in terms of toxicity.

Importantly in this update, we see that 27% of patients who received pembrolizumab are alive, compared with only 14% in patients assigned to receive chemotherapy. Compared with the previously published study, the hazard ratio decreased from 0.73 to 0.70, meaning there is a 30% reduction in the risk of death in patients receiving pembrolizumab. In terms of response rate, this trial has shown that pembrolizumab had a superior response rate to chemotherapy, with 21% compared with 11%.

We have some additional complete responses with subsequent follow-up, and one important thing is that the responses that have been observed are durable. At 2 years, the median duration of response in the pembrolizumab arm has not been reached—but it’s close to 50%—meaning responses are durable. Meanwhile, the median duration of response for the chemotherapy arm is 4.5 months. The main benefit is that the responses achieved are durable.

Additional data that we have observed are that the duration of response may be even better in patients who are PD-L1 positive based on the composite positive score. Also, an important thing is that the toxicity favors the use of pembrolizumab compared with chemotherapy, and with this long-term follow-up, we have not seen any unexpected toxicities appearing in the long term. What we can conclude is that the results seen with pembrolizumab in the second-line treatment of patients with metastatic bladder cancer who have failed platinum-based chemotherapy are more or less similar to what has been seen in tumors like melanoma, where we see a survival curve that is stable after 2 years. This percentage of patients who responded might have a long-term benefit with pembrolizumab.

Can you expand on the PD-L1 data with pembrolizumab?

What do these findings mean for the use of immunotherapy in bladder cancer?

Pembrolizumab has shown a statistically significant survival benefit when compared with chemotherapy. This is the level 1 evidence, and this is the first option for patients with advanced bladder cancer who are willing to receive or need to receive second-line treatment.The way that PD-L1 is being scored with pembrolizumab in these patients is completely different than scores being used for other drugs, such as atezolizumab (Tecentriq). For pembrolizumab, we are using the staining in tumor cells and immune cells divided by the number of tumor cells. In the initial report of the study, we established that more than [a score of] 10% might derive potential benefit, and this was coming from preliminary data. However, we did not see a difference in patients who had a PD-L1—positivity score of more than 10%. The benefit was there and the hazard ratio was much better, but we didn't see much change in terms of median survival or in terms of substantial improvement. Therefore, PD-L1 in the pembrolizumab-treated patients does not seem to impact therapeutic decision.Only about 15% to 20% of patients in this population receive a benefit from immunotherapy. Obviously, we need to build on that and provide benefit to a higher percentage of patients. The field is now moving to focus on patient selection; we are trying to identify the patients who are going to benefit. A way to improve results with the present use of immunotherapy is using combinations. There are several trials ongoing with combinations that are putting together 2 different checkpoint inhibitors.

For example, one of the trials that is open for patient accrual is a trial that will compare monotherapy with pembrolizumab to pembrolizumab plus an IDO inhibitor. We know that if we inhibit IDO, the immunosuppressive effect of kynurenine—the metabolite of tryptophan—is no longer there. Some preliminary phase II data have shown that there is an increase in response rates in these patients. This combination is moving to a randomized phase III study in second-line patients, as well as in the first-line setting for patients who are unfit for chemotherapy.

There are several other trials combining other targeted agents in the second-line setting. These trials are mainly phase II studies, but if we see a signal then perhaps we can jump to a phase III in the second-line setting.

How do you see immunotherapy progressing in the next 5 to 10 years in bladder cancer?

[Immunotherapy] drugs are moving to different settings, such as frontline metastatic disease, where there are 4 trials combining immunotherapy agents compared with standard-of-care chemotherapy. Perhaps with these trials we can change the way that we are treating our first-line patients with bladder cancer. The challenge is to improve on the 20% of patients who respond to immunotherapy. We are doing as much as we can, such as looking for biomarkers in addition to tumor mutational burden. There are RNA-specific profiles who may enrich the patients who are going to benefit from receiving immunotherapy. We are looking forward to seeing some combinations with chemotherapy, targeted therapy, and other immunotherapy approaches, such as vaccines—which may improve outcomes.

We know that these treatments are high-cost treatments and 80% of patients don't derive benefit, so we want to not only improve the results, but also better select patients for which immunotherapy is their best option. The biomarker space and combination therapy are where we are investing a lot of our efforts.


  1. Bellmunt J, De Wit R, Vaughn DJ, et al. Two-year follow-up from the phase 3 KEYNOTE-045 trial of pembrolizumab (pembro) vs investigator’s choice (paclitaxel, docetaxel, or vinflunine) in recurrent, advanced urothelial cancer (UC). J Clin Oncol. 2018;36 (suppl 6S; abstr 410).
  2. Bellmunt J, De Wit R, Vaughn DJ, et al. Pembrolizumab as second-line therapy for advanced urothelial carcinoma. N Engl J Med. 2017;376:1015-1026. doi: 10.1056/NEJMoa1613683.