Dr. Arjun Balar discusses pembrolizumab in BCG-unresponsive NMIBC

December 5, 2020
Benjamin P. Saylor

Benjamin P. Saylor is associate editor of Urology Times, an Advanstar Communications publication.

The in-depth interview covers the data that led to the NMIBC approval, outlines the appropriate patient population for pembrolizumab, and reviews side-effect management with the immune checkpoint inhibitor.

The immune checkpoint inhibitor pembrolizumab (Keytruda) has indications in both bladder and kidney cancers, among other malignancies. Most recently, in January of 2020, the FDA approved pembrolizumab for the treatment of BCG-unresponsive non–muscle invasive bladder cancer (NMIBC).

In this interview, Arjun V. Balar, MD, discusses the data that led to the NMIBC approval, outlines characteristics to consider when evaluating use of pembrolizumab in this setting, and reviews side effects urologists must watch for.Balar is associate professor of medicine at the NYU Grossman School of Medicine and genitourinary medical oncology program director at NYU Langone’s Perlmutter Cancer Center, New York.

Pembrolizumab was approved this year for BCG-unresponsive high-risk NMIBC. Can you discuss the specific patient population for which the treatment is approved, as well as the pivotal trial and data that led to the approval?

The KEYNOTE-057 study (NCT02625961) was a single-arm, phase 2 trial of pembrolizumab, which is a systemic anti-PD-1 antibody, in patients specifically with BCG-unresponsive, non–muscle invasive bladder cancer. The trial focused on 2 patient populations. Cohort A comprised patients with carcinoma in situ with or without papillary tumors, and cohort B comprised patients with papillary tumors only, but no carcinoma in situ. The reason those 2 patient populations have to be evaluated separately is because carcinoma in situ cannot be fully resected, and therefore patients go into treatment with disease. Therefore, the primary end point you're really measuring is complete response. Patients who have papillary tumors only can actually have all of their disease fully resected. For those patients, the primary outcome measure was disease-free or recurrence-free survival.

In the primary analysis of the roughly 100 patients who were first treated as part of cohort A, we looked at complete response rate, which was measured at 3 months after the start of study treatment. We found that in the roughly 100 patients who were efficacy-valuable for the primary end point, the complete response rate was 40%.

The next question was about durability, because we do have agents that are also similarly active and can eradicate tumors within the bladder. The real question is how durable these responses are. We found that about half of those complete responses were durable beyond 1 year, which tells us there's clearly a subset, perhaps 1 in 5 patients in total out of an intent-to-treat analysis, that can actually have durable complete responses in the bladder and perhaps avoid a radical cystectomy when they receive pembrolizumab systemic therapy.

What should a urologist look for or consider regarding tumor and patient characteristics when deciding on the use of pembrolizumab in this setting?

To start, urologists need to make sure that the patient is eligible for this line of therapy. Number 1, they have to have carcinoma in situ, and number 2, they should have BCG-unresponsive carcinoma in situ. The more recent FDA guidance documentation that came out in 2018 really helped us define this patient population. This includes patients who have had at least 2 courses of induction BCG and subsequent persistence of carcinoma in situ, or patients who have, let's say, received a course of induction therapy, perhaps have already achieved a complete response with that, and then are going on to receive maintenance and have received at least 2 or 3 instillations of maintenance BCG, and then within a specific period of time they develop a relapse or recurrence of non–muscle invasive bladder cancer. Those are some of the descriptions of the true "BCG-unresponsive" disease state, and it really helps us make sure that we identify the patients who are getting adequate BCG and then have developed recurrence or persistence of carcinoma in situ despite adequate BCG.

Once you determine that a patient is eligible, it comes down to counseling. First and foremost, patients need to understand that radical cystectomy is the most curative option. Radical cystectomy is curative in approximately 90%+ of patients if you look at it from an all-comers standpoint. That's a very important consideration, whereas we cannot quote that with any of the other subsequent salvage regimens that we use in BCG-unresponsive cancer. But patients are also cognizant of the morbidity associated with surgery; two thirds of patients could have a major complication, and the perioperative mortality rate is approximately 3% to 5%.

You have to have a good, balanced discussion: "Cystectomy is highly curative, but you need to understand the impact it's going to have on you in the short and long term. Here's another option, but let's be smart about it. If you're eligible for this therapy, let's pursue treatment." However, once pembrolizumab is initiated, it's also critically important to say, "As you're receiving treatment, we need to be looking inside your bladder. Every 6 months or so, we're doing scans, including CAT scans, to monitor your cancer systemically." When this is done in the right way, patients can safely receive pembrolizumab and see if it achieves a good outcome. Patients who don't achieve a complete response to pembrolizumab need to be counseled about other options, including radical cystectomy.

What side effects should urologists be particularly mindful of when a patient receives pembrolizumab for NMIBC, and how are these side effects optimally managed?

Monitoring for side effects with systemic checkpoint blockade is a bit tricky, because they're not that common. About a third of patients will report nothing. About 40% to 45% of patients may have a rash and itchy skin, which is usually nothing more than a nuisance to the patient. You can prescribe oral antihistamines and topical corticosteroids.

For the most part, 80% to 85% of patients will go through treatment without any substantial toxicity. We have seen that about 15% to 20% of patients will have an adverse event in the grade 3 or 4 range. If you look at the metastatic trials, 5% to 10% of patients will need systemic corticosteroids to manage their toxicity. So in general, they're not that common, but when they do occur, they have to be addressed very quickly. You have to watch out for immune infiltration of the lungs, which leads to pneumonitis and can manifest as shortness of breath and cough and can be quite progressive.

Also, if the immune system is infiltrating the GI system, it can manifest as diarrhea. In terms of counseling, ask patients about changes in breathing status and bowel habits, as well as sudden development of diarrhea. These should prompt urgent attention by the treating physician to address whether this is an immune-related adverse event. If it is, the patient needs to be started on systemic steroids right away. Usually, if it's done promptly, with the appropriate dosing and monitoring, patients have resolution of symptoms within a couple of days.

Less than 1% of patients will have very high-grade catastrophic events, some that can permanently impact quality of life. The important conditions to watch out for include endocrinopathies, immune-related diabetes, immune-related adrenal insufficiency, and hypophysitis. These are conditions that lead to lifelong hormone replacement. Immune-related hypothyroidism is extremely common; between 5% and 10% of patients will develop it, but that doesn't tend to impact quality of life as much because it can be simply treated with an oral levothyroxine or other thyroid replacement. It’s really important to recognize that although toxicities are not common, and certainly severe toxicities are not common, when they do happen, they have to be addressed very quickly, and some can be permanent and life altering.

Could you please discuss the KEYNOTE-676 trial (NCT03711032) of pembrolizumab plus BCG?

KEYNOTE-676 is probably the most important study as it relates to non–muscle invasive bladder cancer. This is a randomized phase 3 trial specifically focusing on patients with high-risk non–muscle invasive bladder cancer who have received at least 1 induction course of intravesical BCG and have had persistent disease despite that initial induction course. Those patients are randomized either to BCG re-induction alone, which is currently a standard of care, or systemic pembrolizumab plus re-induction BCG.

The primary end point is complete response rate by blinded independent central review. Key secondary end points include event-free survival and time to cystectomy. The purpose of this study is to, in a randomized setting, judge the value of pembrolizumab when in the non–muscle invasive disease state. It's a slightly different patient population than the BCG-unresponsive population in KEYNOTE-057; this is slightly earlier in the course of treatment for patients. It's a very important study that's looking at systemic checkpoint blockade with pembrolizumab in non–muscle invasive bladder cancer.

Pembrolizumab has other approved indications in the first- and second-line settings for locally advanced or metastatic urothelial carcinoma. Could you please discuss these indications, including which patients coming into the urology clinic would be ideal for these indications?

Systemic checkpoint blockade was first tested in a patient population of metastatic bladder cancer patients, beginning with the randomized phase 3 trial KEYNOTE-045 (NCT02256436), which looked at the use of pembrolizumab versus single-agent chemotherapy after progression on prior platinum-based chemotherapy. That showed a survival benefit with a hazard ratio of 0.70; it's actually the only level 1 evidence that we have for systemic immunotherapy in the second-line setting. Systemic immunotherapy, either with pembrolizumab or another PD-1 or PD-L1 antibody—there are 5 total—is a reasonable treatment option for platinum-refractory disease.

In the first-line setting, which is the other approved indication, the first decision we have to make is whether patients are cisplatin eligible or ineligible. Patients who are eligible for cisplatin-containing chemotherapy in the first-line setting should receive cisplatin-based treatment. It's associated with a survival benefit, and in some patients can be highly curative.

For patients who are ineligible for cisplatin, carboplatin is often used as an option. There's also the use of pembrolizumab and also atezolizumab (Tecentriq), which is also approved in this setting. There have been some changes in the label. At first, in 2017, there was a fairly broad label for both of these agents in first-line cisplatin-ineligible patients. It's become a little bit more restricted since we found that in randomized trials, PD-L1–low patients actually had worse survival compared to chemotherapy. Therefore, in June of 2018, the labels for pembrolizumab and atezolizumab were restricted to either patients who are PD-L1 positive or for patients who are not eligible for any platinum-containing chemotherapy, even carboplatin.1 So there is clearly a subset of patients with metastatic bladder cancer in the first-line setting who are cisplatin-ineligible, PD-L1 positive, or entirely chemotherapy ineligible who should still have the option to receive checkpoint inhibitors.

Pembrolizumab is also approved for renal cell carcinoma (RCC). Could you discuss this indication and any anecdotal evidence that you may have of how the drug has fared in the real-world setting for RCC?

Pembrolizumab's use in metastatic RCC is really in the first-line setting in combination with the oral VEGFR-TKI axitinib (Inlyta). In a randomized phase 3 trial, it was tested against sunitinib (Sutent) and improved survival in all IMDC risk categories, so it's a very effective therapy up front. Looking back at the original phase 1 study that evaluated this combination, response rates were in the 60% to 70% range. It was evident to me that there's at least some clinical evidence of immune synergy between VEGFR-TKIs and checkpoint blockade.

What's particularly useful about this combination is that it gets you a response relatively quickly. For patients in the first-line setting, our choices are either ipilimumab (Yervoy)/nivolumab (Opdivo) which is another approved indication in the first-line setting, or VEGFR-TKI plus a PD-1 antibody. I've tended to lean toward the VEGFR-TKI plus PD-1 combinations, whether it's axitinib plus pembrolizumab or axitinib plus avelumab (Bavencio) or others. For patients who have a high disease burden and need a response right away, I've generally leaned toward a VEGFR-TKI combination, which is much more effective and much faster in terms of activity as compared to ipilimumab/nivolumab. The ipilimumab/nivolumab combination also has a fair bit of activity, including complete responses, but tends to take a little bit longer to demonstrate clinical efficacy.


1. FDA limits the use of Tecentriq and Keytruda for some urothelial cancer patients. Press release. FDA. July 5, 2018. Accessed December 4, 2020. https://bit.ly/2JLjJEn