What is the PD-L1 (Programmed Death-Ligand 1) expression cutoff for a patient with gastric cancer to be considered for treatment with Keytruda (pembrolizumab)?

Medical Advisory BoardAll articles are reviewed for accuracy by our Medical Advisory Board
Educational purpose only • Exercise caution as content is pending human review
Article Review Status
Submitted
Under Review
Approved

Last updated: January 11, 2026View editorial policy

Personalize

Help us tailor your experience

Which best describes you? Your choice helps us use language that's most understandable for you.

PD-L1 Cutoff for Pembrolizumab in Gastric Cancer

For gastric cancer, pembrolizumab requires a PD-L1 Combined Positive Score (CPS) ≥1 as the minimum threshold, with CPS ≥5 representing the strongest evidence for benefit and Category 1 recommendation status.

HER2-Negative Gastric Cancer

First-Line Treatment Thresholds

Pembrolizumab combined with fluoropyrimidine- and platinum-based chemotherapy is the standard approach, with treatment recommendations stratified by PD-L1 CPS level 1:

  • CPS ≥5: Category 1 preferred option (strongest recommendation) 1
  • CPS ≥1: Preferred option, though evidence is less robust than CPS ≥5 1
  • CPS 0: Pembrolizumab is NOT recommended; use chemotherapy alone 1

The KEYNOTE-859 trial demonstrated that in HER2-negative gastric/GEJ cancer with CPS ≥10, median OS was 15.7 months with pembrolizumab versus 11.8 months with placebo (HR 0.65, P <0.0001) 1. For CPS ≥1, median OS was 13 months versus 11.4 months (HR 0.74, P <0.0001) 1.

Evidence Quality by CPS Level

The survival benefit becomes progressively stronger at higher CPS thresholds 1, 2:

  • CPS ≥10: Objective response rate 23-25% with pembrolizumab monotherapy 2, 3
  • CPS ≥5: Objective response rate 20% 3
  • CPS ≥1: Objective response rate 13% 3
  • CPS <1: Objective response rate only 3% 3

HER2-Positive Gastric Cancer

For HER2-positive gastric or GEJ adenocarcinoma, pembrolizumab combined with trastuzumab and fluoropyrimidine/platinum chemotherapy requires CPS ≥1 1, 4. This is a Category 1 preferred first-line therapy option 1.

The FDA modified approval in 2023 specifically for patients with PD-L1 CPS ≥1 in this setting 1.

MSI-H/dMMR Gastric Cancer

Pembrolizumab monotherapy is approved regardless of PD-L1 expression for MSI-H or dMMR gastric cancer 1, 4. Among 42 gastric cancer patients with MSI-H tumors in KEYNOTE-158, the objective response rate was 31% 1. In KEYNOTE-059,4 of 7 MSI-H patients achieved objective responses (ORR 57.1%) 1.

PD-L1 Testing Requirements

Companion Diagnostic Specifications

PD-L1 testing must be performed using FDA-approved companion diagnostic assays in CLIA-approved laboratories 1:

  • Approved assays: 22C3 pharmDx or SP263 assay 1, 5
  • Scoring method: Combined Positive Score (CPS) 1
  • CPS calculation: (Number of PD-L1-stained cells [tumor cells, lymphocytes, macrophages] ÷ total viable tumor cells) × 100 1
  • Minimum requirement: At least 100 tumor cells must be present for adequate evaluation 1
  • Positive threshold: CPS ≥1 1

The two FDA-approved assays (22C3 and SP263) show high concordance, with overall percent agreement >90% at CPS ≥1 and 99.2% at CPS ≥10 5.

Critical Clinical Caveats

Tissue Sampling Considerations

Fresh tumor specimens are preferred if archival tissue shows PD-L1 negativity 6. The FDA approval for pembrolizumab in gastric cancer specifically recommends obtaining fresh biopsy if archival specimens are PD-L1-negative 6.

Geographic Variation in Approval

While the NCCN and FDA use CPS ≥1 as the minimum threshold 1, 4, ASCO guidelines provide more nuanced recommendations 1:

  • Strong recommendation: CPS ≥10 for esophageal/GEJ adenocarcinoma 1
  • Case-by-case consideration: CPS 1-10 1

Optimal Cutoff Uncertainty

The optimal PD-L1 cutoff remains uncertain, but efficacy clearly increases with higher PD-L1 expression levels 1. The interaction between treatment effect and PD-L1 expression was significant at CPS ≥5 in CheckMate 649 1, supporting the Category 1 designation at this threshold 1.

Interobserver Variability

Pathologist interpretation variability exceeds assay variability 5. Agreement between pathologists is higher at CPS ≥10 (99.2%) compared to CPS ≥1 (94.7%) 5, making higher cutoffs more reliable for clinical decision-making.

Treatment Algorithm Summary

  1. Test all advanced gastric cancer for PD-L1 CPS using FDA-approved assay 1
  2. If HER2-positive and CPS ≥1: Pembrolizumab + trastuzumab + chemotherapy 1
  3. If HER2-negative and CPS ≥5: Pembrolizumab + chemotherapy (Category 1) 1
  4. If HER2-negative and CPS 1-4: Consider pembrolizumab + chemotherapy on case-by-case basis 1
  5. If CPS 0: Chemotherapy alone without pembrolizumab 1
  6. If MSI-H/dMMR: Pembrolizumab monotherapy regardless of PD-L1 status 1, 4

Professional Medical Disclaimer

This information is intended for healthcare professionals. Any medical decision-making should rely on clinical judgment and independently verified information. The content provided herein does not replace professional discretion and should be considered supplementary to established clinical guidelines. Healthcare providers should verify all information against primary literature and current practice standards before application in patient care. Dr.Oracle assumes no liability for clinical decisions based on this content.

Have a follow-up question?

Our Medical A.I. is used by practicing medical doctors at top research institutions around the world. Ask any follow up question and get world-class guideline-backed answers instantly.