Immunotherapy 'May Help Advanced Prostate Cancer Patients'

Liam Davenport

June 07, 2018

CHICAGO — A small proportion of men with advanced prostate cancer for whom other treatments have failed may experience a significant and ongoing benefit with the programmed death ligand 1 (PD-L1) inhibitor pembrolizumab (Keytruda, Merck), say UK researchers who are now working on how to reliably identify those men.

Intriguingly, the immunotherapy, which continued to show a benefit after a year in 11% of the 258 men with metastatic castration-resistant prostate cancer (mCRPC), appeared to have a benefit even in men who did not express PD-L1.

Presenting the new data here at the American Society of Clinical Oncology (ASCO) 2018 Annual Meeting, Professor Johann De Bono, from the Royal Marsden Hospital, London, suggested that genetic biomarkers may point to those who are at risk, although the data were preliminary.

He said: "In a small population of patients with advance prostate cancer, pembrolizumab clearly has some anti-tumour activity, and activity was observed in PD-L1 positive and negative groups."

De Bono added: "Biomarker work is ongoing and we have clearly seen of antitumour activity in mismatch repair defective cancers, in at least one definite BRAC2 mutated cancer, as well as some patients that we still don't fully understand why they responded.

"Further evaluation of the subset that respond is ongoing, as are combination trials in this population of patients."

'Smarter, Kinder Treatment'

Professor Paul Workman, Chief Executive of The Institute of Cancer Research, London, whose researchers also took part in the trial, said in a press release: "Immunotherapy has proven to be a smarter, kinder treatment for many types of cancer, but it still only works for a minority of patients.

"The challenges we now face are how to predict in advance who will benefit, and how to make immunotherapy work for more people."

Workman added that, if if gene mutations can be shown to identify the patients who will respond, "it should be possible to provide some men with advanced prostate cancer with an exciting new treatment option."

Latest Research

Although PD-1 and PD-L1 inhibitors have shown activity in numerous cancers, there have been few signs of activity in prostate cancer.

However, two recent studies, including KEYNOTE-028, suggested that pembrolizumab may achieve objective responses in some men with previously treated PD-L1-positive mCRPC.

De Bono told the audience that KEYNOTE-199 was therefore launched to try to confirm this activity and determine which patents might benefit from PD-L1 blockade.

Of the five cohorts included in the study, he focused on three, which included mCRPC patients who had received ≥1 prior targeted endocrine therapy and who had undergone 1–2 prior chemotherapy regimens, including docetaxel (Taxotere, Sanofi-Aventis). All patients also had an ECOG performance status of 0–2.

In cohort one (n=131), all of the men had PD-L1-positive mCPRC, while those in cohort two (n=67) had PD-L1-negative disease. In both groups, the patients had measurable disease as per the Response Evaluation Criteria In Solid Tumours (RECIST) v1.1 criteria.

In cohort three (n=60), the patients had bone metastases but no measurable disease on RECIST v1.1, and had any PD-L1 status.

Patients in all three cohorts were treated with pembrolizumab 200 mg for 35 weeks or until confirmed progressive disease, intolerable toxicity, investigator decision or patient withdrawal.

The patients, who fitted the typical post-chemotherapy patient profile, underwent imaging every 9 weeks for 1 year and then every 12 weeks. The primary endpoint was the objective response rate as per RECIST v1.1 in cohorts one and two.

After a median follow-up of 8.1 months in cohort one, 11% of patients were still being treated at the data cut-off for the study. In cohort two, 9% were still receiving treatment after a median of 7.9 months, while, in cohort three, 12% were still on treatment after 11.8 months of follow-up.

In cohorts one and two, 10% of patients had a change in the sum of the target lesions from baseline of ≥30% with pembrolizumab, while 11% of patients in all three cohorts had a reduction in prostate-specific antigen levels from baseline of ≥50%.

In both cases, the responders to pembrolizumab therapy included those who did not have any measurable PD-L1.

Central review of the best responses indicated that, in cohort one, 2% of patients had a complete response, 4% had a partial response, and 17% had stable disease, with 4% having stable disease for ≥6 months.

In cohort two, 3% of patients had a partial response and 21% had stable disease, which lasted for ≥6 months in 3%. There were no responses to treatments in cohort three on central review.

The researchers calculated that the disease control rate lasting ≥6 months, which included the best response of complete or partial response of stable disease, across all three cohorts was 11%.

"So we have clearly evidence of a small population that is benefiting," said De Bono.

After 12 months, 39% of patients in cohort one were still alive, while 38% of cohort two patients were still alive, alongside 61% of cohort three patients, although De Bono emphasised that the non-randomised nature of the trial means that it is "hard to draw conclusions" from the data.

Interestingly, genomic analysis of responders to pembrolizumab suggested that there may be some gene mutations that could be associated with response, such as in the BRCA2 gene.

Indeed, analysis of response by the presence of somatic aberrations in DNA repair genes indicated that patients with BRCA1/2 or ATM gene mutations had a disease control rate of 22%.

De Bono noted that 59% of patients across the trial cohorts had any treatment-related adverse events, and 14% had grade 3–5 adverse events.

The most common treatment-related events were fatigue (15%) and diarrhoea (10%), while the immune-mediated events included hyperthyroidism and hypothyroidism.


Study Discussant Dr Douglas McNeel, from the University of Wisconsin School of Medicine and Public Health, Madison, in the US, said following its presentation, that the objective response rates in cohorts one and two are "fairly small, and it's curious that there is no difference between the patients who are PD-L1 positive and PD-L1 negative".

He added: "While I think there's activity there, it's smaller and definitely different from what we've seen with PD-L1 blockade in other diseases."

McNeel agreed with the conclusion that DNA repair defects may be associated with the antitumour activity seen in the study, saying that this "could be important, since the rate of homologous recombination repair mutations is likely higher than microsatellite instability."

He said that the data suggests that this, nevertheless, "may be a premature conclusion but certainly worth more attention".

The study was funded by Merck Sharp & Dohme.

Johann De Bono: Honoraria - Astellas Pharma; AstraZeneca; Genentech/Roche; Pfizer; Sanofi; Consulting or Advisory Role - Astellas Pharma; AstraZeneca; Bayer; Boehringer Ingelheim; Genentech/Roche; Merck Serono; Merck Sharp & Dohme; Pfizer; Sanofi; Research Funding - AstraZeneca (Inst); Genentech (Inst); Sanofi (Inst); Patents, Royalties, Other Intellectual Property - Abiraterone Rewards to Inventors (Inst); PARP inhibitors and DNA repair defects (Inst); Travel, Accommodations, Expenses - Astellas Pharma; AstraZeneca; Genmab; GlaxoSmithKline; Orion Pharma GmbH; Qiagen; Sanofi; Taiho Pharmaceutical; Vertex.

Douglas McNeel: Leadership - Madison Vaccines, Inc.; Stock and Other Ownership Interests  - Madison Vaccines, Inc.; Consulting or Advisory Role - Dendreon; Genocea Biosciences; Madison Vaccines, Inc.; Research Funding - Bristol-Myers Squibb (Inst); Dendreon (Inst); Janssen (Inst); Madison Vaccines, Inc.; Madison Vaccines, Inc. (Inst); Medivation (Inst); Patents, Royalties, Other Intellectual Property - Patents that have been licensed by UW to Madison Vaccines, Inc; Travel, Accommodations, Expenses - Genocea Biosciences; Madison Vaccines, Inc.

American Society of Clinical Oncology (ASCO) 2018 Annual Meeting. Presented June 4, 2018. Abstract 5007.

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.