Differential clonal evolution in oesophageal cancers in response to neo-adjuvant chemotherapy

How chemotherapy affects carcinoma genomes is largely unknown. Here we report whole-exome and deep sequencing of 30 paired oesophageal adenocarcinomas sampled before and after neo-adjuvant chemotherapy. Most, but not all, good responders pass through genetic bottlenecks, a feature associated with higher mutation burden pre-treatment. Some poor responders pass through bottlenecks, but re-grow by the time of surgical resection, suggesting a missed therapeutic opportunity. Cancers often show major changes in driver mutation presence or frequency after treatment, owing to outgrowth persistence or loss of sub-clones, copy number changes, polyclonality and/or spatial genetic heterogeneity. Post-therapy mutation spectrum shifts are also common, particularly C>A and TT>CT changes in good responders or bottleneckers. Post-treatment samples may also acquire mutations in known cancer driver genes (for example, SF3B1 , TAF1 and CCND2 ) that are absent from the paired pre-treatment sample. Neo-adjuvant chemotherapy can rapidly and profoundly affect the oesophageal adenocarcinoma genome. Monitoring molecular changes during treatment may be clinically useful. Oesophageal adenocarcinoma is often treated with chemotherapy before surgery. Here, the authors sequence cancer samples before and after chemotherapy and examine how the genome changes, focusing on changes in driver gene mutations and differential clonal evolution between good and poor responders.