BackgroundThe discovery of driver genes such as EGFR, KRAS, and ALK, has dramatically shifted treatment patterns in patients harboring these oncogenes. However, dissemination into the central nervous system (CNS) is a severe complication. In addition, the particular anatomical structure of the CNS has made it difficult to obtain tissue specimens from brain metastases (BM) to generate a gene map, as such, potential predictive markers for survival in patients with non-small cell lung cancer (NSCLC) and BM (NSCLC-BM) remain unclear. MethodsData from 28 patients diagnosed with NSCLC-BM between June 2019 and May 2021 at Guangdong Sanjiu Brain Hospital (Guangzhou, China), were reviewed. Targeted next-generation sequencing (NGS) of a 168 cancer-related gene panel was available for surgically resected brain tissues from all patients. In addition, molecular characteristics and overall survival (OS) were analyzed to determine potential predictive markers. ResultsAmong patients with NSCLC-BM, NGS revealed that TP53 was the most frequent mutation (61 %), with a detection rate of 39 %, closely by EGFR amplification. Additionally, CDKN2A, MYC, LRP1B, and RNF43 were frequently observed (18 %). The median OS was significantly shorter in the TP53 mutation group than in the wildtype group (14 versus undefined months, p = 0.014). Similar results were also found in the genetic alteration of EGFR amplification, suggesting that EGFR amplification was associated with worse OS (14 vs. 24 months, p = 0.039). Interestingly, NGS revealed that gene alternations such as TP53, EGFR amplification, and CDKN2A, tended to coexist and such a co-alteration panel indicated worse clinical outcomes (median OS, 5 months). In addition, the detection rate of negative survival genes, including TP53 or EGFR amplification, was much higher in tumor tissues than in plasma samples, indicating the limited predictive value of matched PLA samples. ConclusionsGene signatures, such as TP53 or EGFR amplification, were associated with worse survival in patients diagnosed with NSCLC-BM. These valuable findings may shed light on new strategies for the prognostic assessment of specific patient groups.
Read full abstract