A combination of intrinsic and extrinsic features improves prognostic prediction in malignant pleural mesothelioma
Menée à l'aide d'échantillons tumoraux de mésothéliome pleural malin, cette étude met en évidence la performance d'un système de score, basé sur l'association entre le statut mutationnel et le niveau d'expression de 5 gènes (BAP1, NF2, SETD2 et TP53) et sur les niveaux d'infiltration des principales cellules immunitaires dans le microenvironnement tumoral, pour prédire la survie des patients
Background : Malignant pleural mesothelioma (MPM) is a lung pleural cancer with very poor disease outcome. With limited curative MPM treatment available, it is vital to study prognostic biomarkers to categorise different patient risk groups.
Methods : We defined gene signatures to separately characterise intrinsic and extrinsic features, and investigated their interactions in MPM tumour samples. Specifically, we calculated gene signature scores to capture the downstream pathways of major mutated driver genes (BAP1, NF2, SETD2 and TP53) as tumour-intrinsic features. Similarly, we inferred the infiltration levels for major immune cells in the tumour microenvironment to characterise tumour-extrinsic features. Lastly, we integrated these features with clinical factors to predict prognosis in MPM.
Results : The gene signature scores were more prognostic than the corresponding genomic mutations, mRNA and protein expression. High immune infiltration levels were associated with prolonged survival. The integrative model indicated that tumour features provided independent prognostic values than clinical factors and were complementary with each other in survival prediction.
Conclusions : By using an integrative model that combines intrinsic and extrinsic features, we can more correctly predict the clinical outcomes of patients with MPM.
British Journal of Cancer , résumé, 2022