Meeting NewsPerspective

Pembrolizumab shows promise for high-risk, BCG-unresponsive nonmuscle-invasive bladder cancer

Arjun V. Balar, MD
Arjun V. Balar

SAN FRANCISCO — Pembrolizumab induced durable complete responses among patients with high-risk, nonmuscle-invasive, bacillus calmette-guérin-unresponsive bladder carcinoma in situ with or without papillary tumors, according to updated interim results from cohort A of the phase 2 KEYNOTE-057 trial presented at Genitourinary Cancers Symposium.

Patients with high-grade nonmuscle-invasive bladder cancer have high risk for developing muscle-invasive and metastatic disease. Bacillus calmette-guérin (BCG) has been used since the 1970s for these patients, inducing a complete response in about 70% of those treated.

“BCG is very effective at getting a response but, for many of these patients, that response is not durable,” Arjun V. Balar, MD, director of the genitourinary medical oncology program at NYU Langone’s Perlmutter Cancer Center, told HemOnc Today. “At least one-third or more will develop recurrent disease or progress straight through BCG.”

Options for these patients include intravesical mitomycin with interferon, or intravesical valrubicin (Valstar, Endo Pharmaceuticals), Balar said.

“These can induce another response in the bladder, but again, the durability for these agents is not very good, either,” he said. “For many patients, urologists will often just recommend a radical cystectomy. That is a very tough surgery for patients to undergo, and about two-thirds will suffer a major complication. It has a major impact on quality of life, and it takes about 3 months before patients have recovered from the surgery.”

Thus, Balar and colleagues evaluated the use of the PD-1 antibody pembrolizumab (Keytruda, Merck), which has shown significant activity for patients with metastatic bladder cancer.

“Now, the question is, can we use pembrolizumab in nonmuscle-invasive disease and can it help delay, or maybe prevent, needing a cystectomy?” Balar said.

The analysis included 102 patients (median age, 73 years) with histologically confirmed high-risk, BCG-unresponsive carcinoma in situ with or without (71.8%) papillary disease, who received adequate BCG therapy and could not or did not wish to undergo radical cystectomy.

Balar and colleagues defined adequate BCG therapy as receiving at least five of six doses of an initial induction course, followed by at least two of three doses of maintenance therapy or at least two of six doses of a second infusion course. Patients received a median of 12 prior BCG installations.

“There was really never a formal definition of BCG-unresponsive bladder cancer, and for regulatory and clinical trial purposes, we sometimes give one or two installations of BCG and then determine that patient to be BCG refractory or intolerant,” Balar told HemOnc Today. “It was never really codified in terms of what this progression means — did they ever receive adequate BCG? For instance, there are some patients who get a course of induction of BCG and they may have persistent disease and are taken to radical cystectomy. But earlier data that established BCG showed a course of maintenance can induce response in carcinoma in situ, so they just needed more BCG. This formal definition is really important because it lays a framework for other agents to be tested, and then the FDA can look at that for regulatory approval.”

Patients received 200 mg pembrolizumab every 3 weeks for 24 months or until recurrence, progression or unacceptable toxicity.

Complete response rate served as the study’s primary endpoint.

Overall, 40.2% (95% CI, 30.6-50.4) of the population achieved a complete response.

Median follow-up for patients in complete response was 16.7 months (range, 5.9-28.2).

Of these 41 responding patients, 24 (58.5%) had an ongoing response at the time of data cutoff. Fifteen (36.6%) responding patients subsequently experienced recurrent nonmuscle-invasive bladder cancer. However, none of these patients developed muscle-invasive or metastatic disease.

Median duration of complete response was 12.7 months (range, 0+ to 20.5+). Seventy-five percent of patients had a complete response that lasted 6 or more months, and 53% responded for 9 months or longer.

“The major challenge for this group is that we don’t get too excited about complete responses,” Balar said. “We’ve always been able to get complete responses to a second agent. The question is, does it last? To see a 40% complete response rate is very good, but now we must see how durable it is. About 50% of these lasted 9 months or longer, which also is encouraging. Our estimate for the median duration of those complete responses is just over a year. If PD-1 keeps patients with carcinoma in situ in complete response on average for a year or longer, that is pretty good, because in theory you’ve delayed cystectomy for at least a year. I think that is meaningful.”

Of the 57 nonresponders, 41 experienced persistent disease, six experienced recurrent disease, nine developed nonmuscle-invasive bladder cancer stage progression and one developed nonbladder malignancy. However, no patients progressed to T2 disease.

Treatment-related adverse events occurred in 64.7% of the population, and 12.7% experienced treatment-related grade 3 to grade 4 adverse events. Discontinuation due to a treatment-related adverse event occurred among 7.8% of the cohort.

The most frequent treatment-related adverse events included pruritus (10.7%), fatigue (9.7%), diarrhea (8.7%), hypothyroidism (7.8%) and maculopapular rash (5.8%).

Researchers plan to conduct additional follow-up for this cohort, and for the entire 140 patients treated in KEYNOTE-057.

“I agree with those who say this is early,” Balar said. “We need longer-term follow-up. We will probably need a solid 24 months of follow-up to see how durable this is, but so far I think this is encouraging.”

An early look at PD-L1 expression did not relevel a correlation with outcomes, but researchers will continue to look at this biomarker in addition to immune infiltrates, gene expression profiling and molecular subtyping, Balar said.

Additional studies are evaluating atezolizumab (Tecentriq, Genentech) and durvalumab (Imfinzi, AstraZeneca) for this patient population.

Also, the confirmatory phase 3 KEYNOTE-676 trial is ongoing among a population of patients who experienced persistent disease after six installations of BCG induction, who are then randomly assigned to receive BCG reinduction alone or with pembrolizumab.

“The idea is that, with the addition of pembrolizumab to BCG reinduction, will we see an improved complete response rate in the bladder?” Balar said. – by Alexandra Todak

Reference:

Balar AV, et al. Abstract 350. Presented at: Genitourinary Cancers Symposium; Feb. 14-16, 2019; San Francisco.

Disclosures: Balar reports honoraria from AstraZeneca/MedImmune, Genentech/Roche and Merck; consultant/advisory roles with AstraZeneca/MedImmune, Cerulean Pharma, Genentech/Roche, Incyte, Merck, Pfizer/EMD Serono and Seattle Genetics/Astellas; and research funding from AstraZeneca/MedImmune, Genentech/Roche, Merck and Seattle Genetics. Please see the abstract for all other authors’ relevant financial disclosures.

Arjun V. Balar, MD
Arjun V. Balar

SAN FRANCISCO — Pembrolizumab induced durable complete responses among patients with high-risk, nonmuscle-invasive, bacillus calmette-guérin-unresponsive bladder carcinoma in situ with or without papillary tumors, according to updated interim results from cohort A of the phase 2 KEYNOTE-057 trial presented at Genitourinary Cancers Symposium.

Patients with high-grade nonmuscle-invasive bladder cancer have high risk for developing muscle-invasive and metastatic disease. Bacillus calmette-guérin (BCG) has been used since the 1970s for these patients, inducing a complete response in about 70% of those treated.

“BCG is very effective at getting a response but, for many of these patients, that response is not durable,” Arjun V. Balar, MD, director of the genitourinary medical oncology program at NYU Langone’s Perlmutter Cancer Center, told HemOnc Today. “At least one-third or more will develop recurrent disease or progress straight through BCG.”

Options for these patients include intravesical mitomycin with interferon, or intravesical valrubicin (Valstar, Endo Pharmaceuticals), Balar said.

“These can induce another response in the bladder, but again, the durability for these agents is not very good, either,” he said. “For many patients, urologists will often just recommend a radical cystectomy. That is a very tough surgery for patients to undergo, and about two-thirds will suffer a major complication. It has a major impact on quality of life, and it takes about 3 months before patients have recovered from the surgery.”

Thus, Balar and colleagues evaluated the use of the PD-1 antibody pembrolizumab (Keytruda, Merck), which has shown significant activity for patients with metastatic bladder cancer.

“Now, the question is, can we use pembrolizumab in nonmuscle-invasive disease and can it help delay, or maybe prevent, needing a cystectomy?” Balar said.

The analysis included 102 patients (median age, 73 years) with histologically confirmed high-risk, BCG-unresponsive carcinoma in situ with or without (71.8%) papillary disease, who received adequate BCG therapy and could not or did not wish to undergo radical cystectomy.

Balar and colleagues defined adequate BCG therapy as receiving at least five of six doses of an initial induction course, followed by at least two of three doses of maintenance therapy or at least two of six doses of a second infusion course. Patients received a median of 12 prior BCG installations.

“There was really never a formal definition of BCG-unresponsive bladder cancer, and for regulatory and clinical trial purposes, we sometimes give one or two installations of BCG and then determine that patient to be BCG refractory or intolerant,” Balar told HemOnc Today. “It was never really codified in terms of what this progression means — did they ever receive adequate BCG? For instance, there are some patients who get a course of induction of BCG and they may have persistent disease and are taken to radical cystectomy. But earlier data that established BCG showed a course of maintenance can induce response in carcinoma in situ, so they just needed more BCG. This formal definition is really important because it lays a framework for other agents to be tested, and then the FDA can look at that for regulatory approval.”

PAGE BREAK

Patients received 200 mg pembrolizumab every 3 weeks for 24 months or until recurrence, progression or unacceptable toxicity.

Complete response rate served as the study’s primary endpoint.

Overall, 40.2% (95% CI, 30.6-50.4) of the population achieved a complete response.

Median follow-up for patients in complete response was 16.7 months (range, 5.9-28.2).

Of these 41 responding patients, 24 (58.5%) had an ongoing response at the time of data cutoff. Fifteen (36.6%) responding patients subsequently experienced recurrent nonmuscle-invasive bladder cancer. However, none of these patients developed muscle-invasive or metastatic disease.

Median duration of complete response was 12.7 months (range, 0+ to 20.5+). Seventy-five percent of patients had a complete response that lasted 6 or more months, and 53% responded for 9 months or longer.

“The major challenge for this group is that we don’t get too excited about complete responses,” Balar said. “We’ve always been able to get complete responses to a second agent. The question is, does it last? To see a 40% complete response rate is very good, but now we must see how durable it is. About 50% of these lasted 9 months or longer, which also is encouraging. Our estimate for the median duration of those complete responses is just over a year. If PD-1 keeps patients with carcinoma in situ in complete response on average for a year or longer, that is pretty good, because in theory you’ve delayed cystectomy for at least a year. I think that is meaningful.”

Of the 57 nonresponders, 41 experienced persistent disease, six experienced recurrent disease, nine developed nonmuscle-invasive bladder cancer stage progression and one developed nonbladder malignancy. However, no patients progressed to T2 disease.

Treatment-related adverse events occurred in 64.7% of the population, and 12.7% experienced treatment-related grade 3 to grade 4 adverse events. Discontinuation due to a treatment-related adverse event occurred among 7.8% of the cohort.

The most frequent treatment-related adverse events included pruritus (10.7%), fatigue (9.7%), diarrhea (8.7%), hypothyroidism (7.8%) and maculopapular rash (5.8%).

Researchers plan to conduct additional follow-up for this cohort, and for the entire 140 patients treated in KEYNOTE-057.

“I agree with those who say this is early,” Balar said. “We need longer-term follow-up. We will probably need a solid 24 months of follow-up to see how durable this is, but so far I think this is encouraging.”

An early look at PD-L1 expression did not relevel a correlation with outcomes, but researchers will continue to look at this biomarker in addition to immune infiltrates, gene expression profiling and molecular subtyping, Balar said.

PAGE BREAK

Additional studies are evaluating atezolizumab (Tecentriq, Genentech) and durvalumab (Imfinzi, AstraZeneca) for this patient population.

Also, the confirmatory phase 3 KEYNOTE-676 trial is ongoing among a population of patients who experienced persistent disease after six installations of BCG induction, who are then randomly assigned to receive BCG reinduction alone or with pembrolizumab.

“The idea is that, with the addition of pembrolizumab to BCG reinduction, will we see an improved complete response rate in the bladder?” Balar said. – by Alexandra Todak

Reference:

Balar AV, et al. Abstract 350. Presented at: Genitourinary Cancers Symposium; Feb. 14-16, 2019; San Francisco.

Disclosures: Balar reports honoraria from AstraZeneca/MedImmune, Genentech/Roche and Merck; consultant/advisory roles with AstraZeneca/MedImmune, Cerulean Pharma, Genentech/Roche, Incyte, Merck, Pfizer/EMD Serono and Seattle Genetics/Astellas; and research funding from AstraZeneca/MedImmune, Genentech/Roche, Merck and Seattle Genetics. Please see the abstract for all other authors’ relevant financial disclosures.

    Perspective
    Elizabeth Plimack

    Elizabeth Plimack

    Carcinoma in situ is a relatively benign state; it doesn’t frequently parlay into advanced muscle-invasive or metastatic disease. This is reflected in this cohort with carcinoma in situ— we didn’t see anyone progress to metastatic disease. But, as such, the bar is pretty high. Cancer needs to be eradicated in the bladder of at least a subset of patients for a meaningful amount of time for it to be something that will enter the clinic. I was a little disappointed to see the number of patients on the swimmer plot who had a response and then recurred on a subsequent transurethral resection.

    Still, these are early data cuts for an endpoint that, by definition, takes at least 18 to 24 months to hit. Even if it works to permanently eradicate the cancer in a small group of patients, especially if we can define who those are, it could be meaningfully useful. But, I wish there had been more durability to the responses presented here.

    • Elizabeth Plimack, MD, MS
    • Fox Chase Cancer Center

    Disclosures: Plimack reports consultant/advisory roles with AstraZeneca/MedImmune, Bristol-Myers Squibb, Clovis Oncology, Exelixis, Genentech/Roche, Horizon Pharma, Incyte, Inovio Pharmaceuticals, Janssen, Merck and Novartis, and research funding to her institution from Astellas Pharma, AstraZeneca, Bristol-Myers Squibb, Genentech/Roche, Merck Sharp & Dohme, Peloton Therapeutics and Pfizer.

    See more from Genitourinary Cancers Symposium