ESMO 2020 — Expert commentary: the "evolutionary" effect of treatments on metastatic cancer


  • Cristina Ferrario — Agenzia Zoe
  • Oncology Conference reports
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.

Joris Van De Haar is an MD and PhD student in the groups of Emile Voest and Lodewyk Wessels in the fields of bioinformatics and systems biology at the Netherlands Cancer Institute in Amsterdam. At ESMO 2020, he presented the results of a large whole genome sequencing (WGS) study  evaluating the genomic evolution of metastatic tumors under therapeutic pressure in paired biopsies (prior to and after various types of systemic treatments) from 239 patients with metastatic cancer with tumors from 21 different primary sites. Here, he discusses some of these findings.

  • It is very relevant for clinicians to know if genomic characterization should be repeated in patients with metastatic cancer after a new line of treatment.
  • Looking at paired biopsies of these patients (median interval 6.4 months), we do see a modest increase in the overall numbers of somatic variants. For some “outlier” samples, these changes can even be dramatic. However, if we zoom in on the “driver” variants (the somatic genomic variants likely involved in the malignant growth of these lesions), we already see much more homogeneity between paired biopsies.
  • When we zoomed in even further and focused on “clinically actionable” drivers (the somatic genomic drivers that determine treatment indications), we found that these are even more conserved over time.
  • Especially for the variants that determine standard-of-care (on-label) treatment indications, we hardly see any changes happening. For clinical trial (off-label) indications, we observe some minor changes over time.
  • It will be a question of cost-effectiveness to determine whether a second genomic analysis is worthwhile within this setting.
  • An interesting observation was that the landscape of clinically actionable somatic variants was most strongly impacted in patients treated with targeted therapies. This is in line with the fact that “second genomic hits” in genes involved in the pathway targeted by the drug are often well-known acquired resistance mechanisms. We also identified many on-target genomic changes in the drug targets themselves.
  • As these “second hits” within the same pathway, or within the target itself, occur in a very limited set of genes, more targeted sequencing approaches could be preferred over WGS for follow-up, if desired.
  • We conclude that a 1-time WGS analysis early on in the disease course of a patient with metastatic cancer is (1) sufficient to chart all standard-of-care treatment options, and (2) informative to identify investigational treatment opportunities that are being missed by regular diagnostics (the latter having a 33% clinical benefit rate in the DRUP study [van der Velden et al, Nature, 2019]).
  • With ever-decreasing sequencing costs and increasing numbers of genomic treatment indications in oncology, it seems inevitable that comprehensive sequencing techniques such as WGS are going to be used in regular diagnostics to comprehensively chart all genomic treatment opportunities of patients with metastatic cancer.
  • Broad implementation of WGS in regular diagnostics would be the beginning of a new era in precision oncology. It is exciting to imagine how quickly we would generate huge databases that could be mined to profoundly improve the management of patients with metastatic cancer.

Follow all ESMO 2020 coverage on   Univadis .