Article Text

Download PDFPDF
Original research
Genetic characterisation of sarcomatoid carcinomas reveals multiple novel actionable mutations and identifies KRAS mutation as a biomarker of poor prognosis

Abstract

Background Sarcomatoid component occurs in various epithelial malignancies and is associated with an aggressive disease course and poor clinical outcome. As it is largely rare, the molecular events underlying sarcomatoid carcinomas (SCs) remain poorly characterised. Here, we performed targeted next-generation sequencing (NGS) on patients with surgically resected SCs comprising distinct tissues of origin.

Methods A total of 71 patients with pathological diagnosis of sarcomatoid carcinomas and underwent surgery were retrospectively enrolled in this study. Overall survival (OS) was defined as the time from surgery to death from any cause. Patients alive or lost to follow-up were censored. Genomic DNA from formalin-fixed paraffin-embedded samples was extracted for NGS and tumour mutation burden (TMB) analysis.

Results In general, SCs occurred more commonly in males, except those of the gallbladder. SCs of the lung and the larynx were associated with a higher proportion of smokers (p=0.0015). Alterations in TP53, RB1, TERT and KRAS were highly frequent, with KRAS mutations being a biomarker of poor prognosis (median OS=8 vs 16 months, p=0.03). Multiple alterations in potentially actionable genes, including ROS1 and NTRK1 fusions and ERBB2 amplification, were detected in the extra-pulmonary cohort. A relatively high proportion (30%) of patients with extra-pulmonary SC had high TMB, with a median of 5.39 mutations per Mb. Lastly, copy number variations were common in SCs, and were non-overlapping between the primary and metastatic tumours.

Conclusion Taken together, our results suggest that comprehensive genetic testing may be necessary to inform treatment options and identify prognostic biomarkers.

  • gene expression profiling
  • medical oncology
  • pathology

Data availability statement

Data are available on reasonable request. Some or all data, models, or code generated or used during the study are available from the corresponding author (zhangzh@njmu.edu.cn) by request.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.