All in the Levels-Programmed Death-Ligand 1 Expression as a Biomarker for Immune Checkpoint Inhibitor Response in Patients with Gastrointestinal Cancer

Autor: Sarah K Cimino, Kristen K. Ciombor, Satya Das, Shemeka Davis
Rok vydání: 2020
Předmět:
Zdroj: Oncologist
ISSN: 1549-490X
Popis: Immune checkpoint inhibitors (ICIs) benefit rare subsets of gastrointestinal (GI) cancer patients. Significant interest exists to identify predictive biomarkers which may increase the applicability of ICI therapy for these patients. Programmed-death ligand 1 (PD-L1) is one such candidate; however, this biomarker has well-chronicled limitations. Combined positive score (CPS) ≥ 1 is the minimum PD-L1 expression threshold necessary for gastric/gastroesophageal junction (GEJ) cancer patients to qualify for treatment with pembrolizumab; however, studies suggest that patients with higher CPS scores may derive greater benefit. We present the cases of two patients, both with low tumor mutational burden, microsatellite stable and CPS ≥ 70 GI tumors (cholangiocarcinoma and GEJ cancer), who have achieved excellent tumor control with pembrolizumab. We postulate that by testing for CPS in all GI cancer patients, and identifying a CPS threshold predictive of ICI response, PD-L1 expression could identify the GI cancer patients, in tissue agnostic fashion, who could benefit from ICI therapy. IMPLICATIONS FOR PRACTICE: CPS measures PD-L1 expression on both tumor and immune cells and has been used to select advanced cancer patients for pembrolizumab therapy. A CPS ≥ 1, however, carries limited predictive capacity to identify the patients who may derive benefit from the immune checkpoint inhibitor (ICI). The two patient cases presented suggest that that perhaps by testing CPS in all gastrointestinal (GI) cancer patients, beyond those with gastric/gastroesophageal cancer, and defining an appropriate CPS threshold, PD-L1 expression may better predict ICI benefit for GI cancer patients irrespective of tumor type. Further studies are needed to verify this hypothesis.
Databáze: OpenAIRE