Newly-released ESMO recommendations critically assess specific germline pathogenic variants in terms of cancer-related mortality and testing implications for patients and their families
A “more is better” approach is increasingly emerging in testing germline pathogenic variants in breast cancer as a result of a widespread access to next-generation sequencing and expanded molecularly-guided indications for PARP inhibitors (JAMA Oncol. 2024;10(5):658-670), but it is not without some potential negative implications for patients, their families and healthcare systems. This was highlighted in a collaborative effort of the ESMO Precision Oncology Working Group and a subgroup of international experts who critically evaluated a series of breast cancer susceptibility genes for evidence of impact on cancer-related mortality, recommending on that basis only seven to be included for breast cancer multigene germline panel testing (Ann Oncol. 2025, June 15, Articles in press).
The new ESMO recommendations, which are published in the Annals of Oncology, provide guidance on how to optimise the identification of carriers of germline pathogenic variants (GPV) for the genes for which surveillance, prevention and early detection (SPED) interventions offer the highest mortality benefit. Currently, the mainstream oncology setting where breast cancer multigene panel tests are performed is highly heterogeneous. “There are wide discrepancies between countries and institutes in the set of genes tested”, notes Prof. Clare Turnbull, Division of Genetics and Epidemiology, The Institute of Cancer Research, London, UK, lead author of the paper. Unlike population-level cancer screening programmes which are only implemented following randomised trial evidence of cancer-related mortality benefit, SPED interventions for carriers of germline pathogenic variants often rely on expert opinion or modeling, due to limited direct clinical data. “In some contexts, just BRCA1 and BRCA2 genes are tested, while on the other hand some institutions offer a ‘pan-cancer’ germline panel, which may comprise up to 100 genes, most of which have no association with breast cancer”. She proposes that these large pan-cancer panels are typically offered to ‘future-proof’ against any further indications for genetic testing that may arise in relation to emergent family history of other cancers. Turnbull also highlights the great variation in eligibility criteria, meaning that the proportion of women diagnosed with breast cancer to whom germline genetic testing is offered varies between settings.
The expert group proposes only seven genes should be included on a breast cancer germline multi-gene panel test – BRCA1, BRCA2, PALB2, RAD51C, RAD51D, BRIP1 and TP53 (the latter being tested in women under 40 only) – and recommends against inclusion of CHEK2 and ATM, both being of moderate penetrance only and associated near-exclusively with ER-positive breast cancer. Also, it recommends against inclusion of ‘syndromic’ genes such as NF1, STK11, CDH1 and PTEN, advising that they should be tested for only in a very small proportion of breast cancer patients with relevant features or family history and only after review by clinical genetics.
The paper is a timely reminder to reflect on the complexities of cancer genetic testing in oncology. While the identification of a GPV (primarily, BRCA1/BRCA2) may inform initial oncological breast cancer management with accordant survival benefit, it also has lifelong implications for the proband as well as for other GPV-carriers in the family. “Risk-reducing salpingo-oopherectomy is well demonstrated to reduce ovarian-cancer specific mortality, and possibly all-cause mortality, meaning genes associated with ovarian cancer are some of those of highest clinical-public health utility. By contrast the data on risk-reducing mastectomy are more equivocal even for BRCA1/BRCA2,” explains Turnbull. High levels of surveillance and false-positive results, for instance, may increase the psychological burden for those being assigned as at high risk of cancer, however “whilst we offer GPV-carriers decades of surveillance for multiple different cancers, there have been no randomised trials, and often no prospective longitudinal studies, demonstrating benefit in terms of cancer-related mortality,” she concludes. “Furthermore, for many cancer susceptibility genes, the risks of the associated cancer remain highly uncertain when the GPV is found outside of a familial or syndromic context”.
Contrasting an appealing “the more, the better” approach, the expert working subgroup emphasises the importance of focusing on clinical and public health utility, specifically the evidence-based benefit in reducing cancer-related mortality, when identifying GPV carriers and planning their lifelong risk management.
