Association of BAP1 alterations with malignant pleural mesothelioma treated with trimodality therapy.

Abstract

8552 Background: Trimodality therapy with pleurectomy/decortication, cytotoxic chemotherapy, and adjuvant pleural intensity modulated radiation therapy (IMPRINT) is an emerging standard of care for locally advanced epithelioid mesothelioma (Rimner, Zauderer et al. JCO 2016). Some patients, however, progress rapidly and we therefore sought to identify potential predictive markers of response to this treatment. Given the putative role of BAP1 in DNA damage repair, we hypothesized that alteration in BAP1 would be associated with improved local control after radiation therapy. Methods: We identified patients previously treated at our institution with IMPRINT to a median dose of 4680cGy in 26 fractions. Targeted next generation sequencing was performed with MSK-IMPACT on archival tissue samples. Chart review was undertaken for clinicopathologic features and outcome data. Results: MSK-IMPACT testing was successfully performed on 58 patients who completed IMPRINT. The majority were male with a median age of 70 years. Ninety-seven percent had epithelioid subtype while 3% were biphasic with predominantly epithelioid histology. Median overall survival was 30.2 months with a median follow-up of 45.3 months, consistent with prior reports. Somatic BAP1 mutations were identified in 34% of the specimens. Those with BAP1 mutant tumors had a median time to local failure of 22.4 months (IQR 10.9 – 36.9 months) while those with BAP1 wild type tumors only had a median of 12.1 months (IQR 8.7-15.85 months) to local failure (p = 0.057). We identified a trend towards improved overall survival among those with BAP1 altered tumors compared to those with BAP1 wild type (HR = 0.61, p = 0.14). Conclusions: BAP1 alteration may be associated with improved duration of local control and improved overall survival after IMPRINT therapy. Further analysis and validation in a large data set is needed and a platform for identifying and validating predictive biomarkers should be included in the planned NRG randomized trial of IMPRINT.