Original ArticlePrognostic and predictive effects of TP53 co-mutation in patients with EGFR-mutated non-small cell lung cancer (NSCLC)

•In EGFR-mutated NSCLC, TP53 co-mutation is not a strong prognostic marker.•TP53 missense mutations might be predictive of lower ORR and shorter PFS on TKIs.•Larger datasets are required to validate these observations.

IntroductionTP53 mutations are common in non-small cell lung cancer (NSCLC) and have been reported as prognostic of poor outcome. The impact of TP53 co-mutations in epidermal growth factor receptor (EGFR)-mutated NSCLC is unclear.Materials and methodsTissue from 105 patients with EGFR-mutated NSCLC at Princess Margaret Cancer Centre was analyzed by next-generation or Sanger sequencing to determine TP53 mutational status. Associations between TP53 status and baseline patient and tumor characteristics, treatments and outcomes (relapse-free survival [RFS] after surgical resection, overall survival [OS], overall response rate [ORR] and progression-free survival [PFS] on EGFR tyrosine kinase inhibitors [TKIs]), were investigated.ResultsDual TP53/EGFR mutations were found in 43/105 patients (41%). Among 76 patients who underwent surgical resection, neither RFS (HR 0.99, CI 0.56-1.75, p = 0.96) nor OS (HR 1.39, CI 0.70-2.77; p = 0.35) was associated with TP53 status. Sixty patients (24 TP53 MUT; 36 TP53 WT) received first-generation EGFR TKIs for advanced disease. ORR was not significantly different (TP53 MUT 54%, WT 66%, p = 0.42). There was a non-significant trend towards shorter PFS on EGFR TKIs with TP53 mutation (HR 1.74, CI 0.98-3.10, p = 0.06). When limited to TP53 missense mutations (n = 17), PFS was significantly shorter (HR 1.91, CI 1.01-3.60, p = 0.04). Among 11 evaluable patients treated with T790M inhibitors, ORR was not significantly different (TP53 MUT 3/3 [100%], WT 7/8 [88%]).ConclusionsPatients with dual TP53/EGFR mutations, especially missense mutations, had marginally lower response rates and shorter PFS when treated with EGFR TKI therapy. Larger datasets are required to validate these observations.