132 Precision diagnostics for immunotherapy: detecting PD1-PDL1 interactions in the in situ microenvironment of cancer tissues

Le résumé fourni par la source

Background Immune checkpoint inhibitors (ICIs) have revolutionized cancer treatment. However, their efficacy varies widely among different cancer types and individual patients. Current stratification methods based on PDL1 expression have limited predictive value. Since many ICIs target the PD1-PDL1 pathway, directly detecting PD1-PDL1 interactions within diagnostic samples may offer more precise prediction on immune responses and treatment outcomes. Methods We applied a second-generation in situ proximity ligation assay (Naveni PD1/PD-L1 AP, Navinci Diagnostics AB) to detect PD1-PDL1 interactions in diagnostic tissue samples using a digital analysis pipeline in QuPath. This assay was applied to tissues from 16 different cancer types, as well as a tissue microarray from 352 surgically resected non-small cell lung cancer (NSCLC) patients, and finally diagnostic biopsies from 142 advanced NSCLC patients, both with and without anti-PD1/PDL1 ICI treatment. Results Analysis of the ICT-naïve, surgically treated NSCLC cohort revealed variable protein expression level for both, PDL1 and PD1, for 200 cases, but only 108 (54%) of them demonstrated detectable PD1-PDL1 interactions with the isPLA. Notably, tumors with EGFR mutations exhibited significantly lower PD1-PDL1 interaction scores (p=0.012). Variability in PD1-PDL1 interaction levels was further observed across different cancer types, with liver cancer showing the lowest levels and testicular cancer the highest. These interaction levels correlated with literature-reported objective ICI-response rates. In ICI-treated NSCLC patients, higher PD1-PDL1 interaction levels were significantly associated with complete response (p=0.028) and longer overall survival (median survival 31 vs. 14 months; p=0.010, LogRank). Multivariable Cox regression identified PD1-PDL1 interaction as the only significant variable linked to longer survival post-ICI therapy (p=0.022). Conversely, standard diagnostic PDL1 expression (10% and 50% tumor proportion scores) showed no survival association (p=0.564 and p=0.325, LogRank). In a control cohort of ICI-naive NSCLC patients, no association between PD1-PDL1 interaction and survival was detected. An interaction term between PD1-PDL1 interaction status (high/low) and ICI treatment (no/yes) was significant (p=0.018). Despite high PD1-PDL1 interaction, some NSCLC cases did not respond to ICI therapy (progressive or stable disease). Corresponding differential gene expression analysis revealed that non-responders were associated with gene expression including chemokines (CCL18, CCL22), immune regulators (EOMES), and other checkpoint proteins (HAVCR1, JAML, FCRL1). Conclusions The in situ PD1-PDL1 interaction status offers functional insights and can identify NSCLC patient subsets likely to benefit from immunotherapy. The second-generation isPLA technique, suitable for use with small diagnostic lung biopsies and thus can enhance the predictive accuracy for patient selection in clinical practice. Ethics Approval The study was approved by the Ethics Board Uppsala (Dnr#2017/076).

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.