Question · from the living review of Predictive Biomarkers Immunotherapy →

Is it true that an IFN-γ gene expression profile score is associated with immune checkpoint inhibitor response across multiple tumour types independent of PD-L1 expression?

Likely updated weekly · as of

Priors rates this Likely — 70 out of 100, updated weekly. Probably — but it is not fully settled. On the claim that an IFN-γ gene expression profile score is associated with immune checkpoint inhibitor response across multiple tumour types independent of PD-L1 expression, its four-agent AI review panel weighs 10 primary peer-reviewed studies.

RefutedDoubtfulUncertainLikelyEstablished
where this sits on Priors’ scale of how settled the evidence is

How we got this answer. Priors runs each claim through a panel of four AI agents, each acting as a specialist expert reviewer. They read the published, peer-reviewed studies behind the question, judge how strong, consistent and reliable the evidence is, and turn that judgment into a single rating from 0 to 100 — refreshed every week as new studies appear, so it reflects where the evidence stands today, not a one-off verdict.

The traceable studies behind this rating — and the panel’s single strongest counter-argument to it — are in Priors’ full Predictive Biomarkers Immunotherapy review.

Related Oncology — solid tumours questions

Is it true that MSI-H/dMMR status is the most validated pan-tumour predictive biomarker for immune checkpoint inhibitor response, with consistent high response rates across all tested solid tumour types? → Is it true that dMMR/MSI-H endometrial carcinoma shows robust immune checkpoint inhibitor response both in monotherapy and in combination with targeted agents? → Is it true that different PD-L1 assay clones and scoring methods are not interchangeable and produce clinically significant discordant results across cancer types? → Is it true that PD-L1 tumour proportion score (TPS) ≥50% is a validated predictive biomarker for pembrolizumab monotherapy benefit in previously untreated advanced NSCLC? →
Reflects the peer-reviewed evidence as of 17 July 2026 and updates as new studies land. AI can make mistakes. Not medical advice.