Back

The Predictive and Prognostic Value of T- and B-cell Transcriptomic Signatures for Clinical Response to Immune Checkpoint Blockade in Pleural Mesothelioma

van Genugten, J.; Faulkner, D.; Hahne, J. C.; Disselhorst, M.; Wessels, L.; Fennell, D.; Baas, P.

2025-05-13 oncology
10.1101/2025.05.12.25326806 medRxiv
Show abstract

BackgroundMalignant pleural mesothelioma (PM) is an aggressive cancer with limited treatment options. Although immune checkpoint blockade (ICB) with nivolumab plus ipilimumab improves overall survival and has become standard-of-care, responses are variable and there are no predictive biomarkers for treatment outcome among PM patients with epithelioid histology. MethodsWe developed an eight-gene T-cell and six-gene B-cell transcriptomic signature. To quantify T- and B-cell infiltration, a geometric-mean T- and B-cell expression score was calculated for each patient. Immune profiles were generated in three ICB-naive PM cohorts (n = 448), and Kaplan-Meier analysis was used to evaluate prognostic value of the signatures. Predictive value was tested in five independent ICB-treated cohorts (n = 104). ResultsIn ICB-naive patients, high B-cell infiltration was associated with longer overall survival (hazard ratio (HR): 0.50, 95% confidence interval (CI): 0.31-0.82), whereas T-cell infiltration had no prognostic value. Among patients treated with 2nd line treatment with nivolumab plus ipilimumab, high T-cell infiltration predicted better objective response and improved overall survival (HR: 0.06, 95% CI: 0.01-0.36). This effect was absent in patients treated with 2nd line nivolumab alone or any other anti-PD1/PDL1 drug combined with non-immunotherapeutics. B-cell infiltration showed no predictive value in any ICB-treated group. ConclusionsThe eight-gene T-cell signature is a specific predictor of outcome after 2nd line treatment with nivolumab plus ipilimumab, while B-cell infiltration is prognostic in ICB-naive disease.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.