What is the role of PD-L1 (Programmed Death-Ligand 1) inhibitors, such as pembrolizumab (pembrolizumab), in the management of metastatic castration-resistant prostate cancer (mCRPC)?

PD-L1 Inhibitors in Metastatic Castration-Resistant Prostate Cancer

Pembrolizumab (a PD-1 inhibitor, not PD-L1 inhibitor) has an extremely limited role in mCRPC and should only be used in highly selected molecular subgroups: MSI-H/dMMR tumors (2-5% of cases) or TMB-high tumors (≥10 mut/Mb), after progression on docetaxel and novel hormone therapy. 1

Critical Context: Why Immunotherapy Fails in Most Prostate Cancer

• Prostate cancer is an immunologically "cold" tumor, making checkpoint inhibitors largely ineffective in unselected populations 2
• Four major phase III trials (KEYNOTE-641, KEYNOTE-921, KEYNOTE-991, KEYLYNK-010) evaluating pembrolizumab combinations with chemotherapy, androgen receptor pathway inhibitors, or PARP inhibitors all failed to show benefit in unselected mCRPC patients 3
• PD-L1 expression does NOT predict response to pembrolizumab in prostate cancer, unlike other malignancies 3
• The KEYNOTE-199 trial demonstrated dismal overall response rates of only 5% in PD-L1-positive and 3% in PD-L1-negative mCRPC patients 1

FDA-Approved Indications and NCCN Recommendations

MSI-H/dMMR mCRPC (Category 2A)

• Patient Selection: MSI-H or dMMR status confirmed by DNA testing or immunohistochemistry, disease progression after docetaxel AND novel hormone therapy (abiraterone or enzalutamide) 1
• Prevalence: Only 2-5% of mCRPC cases have MMR deficiency 1
• FDA Approval Basis: Accelerated approval in May 2017 based on 149 patients across 5 studies (only 2 had mCRPC: 1 partial response, 1 stable disease >9 months) 1
• Efficacy Data: Objective response rate of 40% across all MSI-H/dMMR solid tumors, but prostate cancer data remains extremely limited 1
• Mandatory Next Step: Refer to genetic counseling for germline testing for Lynch syndrome if MSI-H/dMMR identified 1

TMB-High mCRPC (≥10 mut/Mb)

• Patient Selection: TMB ≥10 mutations/megabase, prior docetaxel and novel hormone therapy 1
• FDA Approval: June 2020 accelerated approval based on KEYNOTE-158 (no prostate cancer patients included in the TMB cohort) 1
• Evidence Quality: NCCN recommendation based entirely on extrapolation from other tumor types showing 29% response rate in TMB-high tumors 1

Emerging Biomarker: CDK12 Mutations

• Special Population: Patients with CDK12 mutations have aggressive disease unresponsive to hormonal therapy, PARP inhibitors, or taxanes 1
• Preliminary Data: Two large retrospective studies showed 11-33% of CDK12-mutated mCRPC patients responded to PD-1 inhibitors with some durable responses 1
• Current Status: NCCN panel awaits more data before formal recommendation 1

Real-World Efficacy in MSI-H mCRPC

• Recent institutional series of 13 MSI-H metastatic prostate cancer patients treated with pembrolizumab showed 75% achieved PSA50 response, with 58.3% achieving complete biochemical response (undetectable PSA) 4
• Among evaluable patients, 57.1% overall response rate with 2 complete and 2 partial radiographic responses 4
• Median PFS and OS not reached after 14.4 months median follow-up, suggesting durable benefit in this molecular subset 4
• Notably, 6 of 7 complete responders had concurrent HRR gene mutations, and all 3 patients with PTCH1 mutations achieved complete biochemical response 4

Treatment Algorithm for Appropriate Candidates

Step 1: Molecular Testing

• Perform MSI/MMR testing via DNA sequencing or immunohistochemistry on tumor tissue 1
• Consider TMB testing if tissue available (≥10 mut/Mb threshold) 1
• Test for CDK12 mutations in patients with aggressive disease features 1

Step 2: Prior Treatment Requirements

• Confirm progression on docetaxel chemotherapy 1
• Confirm progression on at least one novel hormone therapy (abiraterone or enzalutamide) 1
• Ensure no satisfactory alternative treatment options available 1

Step 3: Pembrolizumab Dosing

• 200 mg intravenously every 3 weeks 1
• Continue androgen deprivation therapy concurrently 5

Step 4: Response Monitoring

• Do NOT rely solely on PSA for monitoring—radiographic progression can occur without PSA elevation in up to 24.5% of patients 6
• Continue treatment until clinical progression or intolerable toxicity, not just PSA progression 6
• Responses can be durable (range 1.9 to >21.8 months when they occur) 1

Adverse Effects Profile

Common Toxicities (>10%): Fatigue, pruritus, diarrhea, anorexia, constipation, nausea, rash, fever, cough, dyspnea, musculoskeletal pain 1

Immune-Mediated Toxicities: Colitis, hepatitis, endocrinopathies (thyroid dysfunction, adrenal insufficiency), pneumonitis, nephritis 1

Grade 3-4 Events: Occurred in 17% of patients in KEYNOTE-028 (grade 4 lipase increase, grade 3 peripheral neuropathy, grade 3 asthenia, grade 3 fatigue) 1

Critical Pitfalls to Avoid

• Do not use pembrolizumab in unselected mCRPC patients—multiple phase III trials confirm no benefit without molecular selection 3
• Do not use PD-L1 expression as a biomarker for patient selection in prostate cancer—it does not predict response 3
• Do not delay testing for MSI-H/dMMR status in patients with mCRPC, as this represents the only validated biomarker for immunotherapy response 1
• Do not use pembrolizumab before progression on both docetaxel and novel hormone therapy unless MSI-H/dMMR status confirmed 1
• Do not discontinue therapy based on PSA progression alone—continue until radiographic or clinical progression 6

Preferred Alternative Therapies for Non-Selected mCRPC

For patients without MSI-H/dMMR or TMB-high status who have progressed on docetaxel and novel hormone therapy, ASCO and NCCN recommend 5:

• 177Lu-PSMA-617 radiopharmaceutical therapy (preferred option with OS benefit) 5
• Cabazitaxel chemotherapy (Category 1 evidence from CARD trial showing OS benefit over switching androgen receptor inhibitors) 1, 5
• Olaparib or rucaparib for patients with BRCA1/2 or other HRR gene mutations 5
• Radium-223 for symptomatic bone-only disease 5