The discovery of synthetic lethality between BRCA mutations and PARP inhibitors ranks has led to major advances in the treatment of BRCA-mutated cancers.

Mutations in BRCA1 and BRCA2 can leave cells with a deficiency in homologous repair (HR). And that deficiency can make them vulnerable to PARP inhibitors, which block alternate DNA repair pathways, as well as platinum-based treatment, which induces DNA mutations that BRCA-deficient cells are unable to cope with.

PARP inhibitors can be useful not just for breast cancer, which gave the BRCA gene its name. It is now clear that BRCA mutations are very important in ovarian cancer as well. Subsets of prostate and pancreatic cancers are also driven by BRCA mutations, and patients with such tumors can benefit from PARP inhibitors.

Likewise, PARP inhibitors can be useful not just for mutations. There are other mechanisms, such as epigenetic silencing through methylation, that can silence BRCA.

"We need to understand how to use scars of HR deficiency as ways of detecting it in tumors where it is not a germline mutation," Andrew Tutt told his audience at the ESMO opening session, where he received the Breast Cancer Award and gave an award lecture on "Finding an Achilles Heel in hereditary breast cancer." Tutt is head of the division of breast cancer research and director of the Breast Cancer Now Toby Robins Research Centre at the Institute for Cancer Research and Guy's Hospital King's College London.

Finally, HR deficient status can change over time.

"HR DNA repair is known to return during tumor evolution after treatment exposure," Tutt said. This is especially likely when the original repair defect was due to epigenetic silencing of BRCA, but there are also what he called "fix-it" mutations that can reverse mutation-induced repair defects.

The diversity of mechanisms that can lead to DNA repair deficiency has spurred a search for biomarkers that could predict which patients could benefit from PARP inhibitors in the absence of a BRCA mutation - and which ones may not predict despite the presence of such a mutation.

At a proffered paper session, researchers from the Vall d'Hebron Institute of Oncology (VHIO) reported on the potential use of RAD51 as such a biomarker.

Violeta Serra, principal investigator in the VHIO's Experimental Therapeutics group, reported on data collected as part of the phase II RIO trial on "detection of [homologous recombination deficiency] via RAD51 foci and comparison with DNA-based tests."

And Alba Llop-Guevara, a postdoctoral scientist at VHIO, presented results regarding "Association of RAD51 with Homologous Recombination Deficiency (HRD) and clinical outcomes in untreated triple-negative breast cancer (TNBC): analysis of the GeparSixto randomized clinical trial," a retrospective analysis which looked at the ability of RAD51 to predict the benefit of adding neoadjuvant carboplatin chemotherapy to the treatment regimen of triple-negative breast cancer patients.

RAD51 is a DNA repair protein that plays a role in double-stranded break repair. As a biomarker, its strength is that it is "nodal point at the base of a complex DNA repair pathway" that can report out many things happening upstream, including but not limited to BRCA mutation, said Tutt, who was the discussant of for both presentations.

And Llop-Guevara noted that "genomic tests provide static information, but functional tests are dynamic and can identify changes... during tumor evolution."

The data presented in the two talks showed that RAD51 has potential as a functional biomarker.

"Being able to use a functional assay just before you start treatment in a treatment-exposed patient may be very valuable in the future, rather than using archival material," said discussant Tutt.

Mutation tests, he added, "Don't have this same ability to detect change of repair function... I'm really very worried about using a mutation signature in advanced disease, whether it's Myriad MyChoice or LOH" -- loss of homozygosity -- "analysis, because I think that's a historical measure and the tumor may not be repair-deficient anymore if it doesn't have a mutation, if it's epigenetic. In early disease it may be much more easy to use these tests to detect the noncarriers who are sensitive. But we don't know that yet."