Katherine Nathanson, MD
Researchers in the Perelman School of Medicine at the University of Pennsylvania determined that there is a relationship between the genetics of BRCA1/2 mutations and the risk of resistance to platinum-based chemotherapy for breast and ovarian cancers, according to the study’s senior author Katherine Nathanson, MD.
The investigators evaluated the genetic profiles of 160 breast and ovarian cancers associated with germline mutations in BRCA1
. Historically, it had been thought that all BRCA1
- and BRCA2
-associated tumors lose the second genetic allele; however, the findings showed that this was not true.
In an interview with OncLive
, Nathanson, a deputy director of the Abramson Cancer Center, and director of Genetics at Basser Center for BRCA, University of Pennsylvania, discussed how the genetics of BRCA1/2
-associated tumors impacts tumor development and response to treatment.
OncLive: Can you please provide an overview of this research?
: We looked at breast and ovarian cancers, first using the data from The Cancer Genome Atlas (TCGA) and then using our local data from the University of Pennsylvania. These were all breast and ovarian cancers, specifically from patients who carried BRCA1
The goal of the research was to characterize tumors from women who have BRCA1
mutations. Although that had been done on the small scale, it hadn’t yet been done on a very large scale. We ended up investigating a total of 160 tumors associated with BRCA1
mutations, with 100 from TCGA and 60 that were looked at locally at the University of Pennsylvania. Of those, we looked at 94 ovarian cancers and 76 breast cancers.
The dogma had been that all BRCA1
- and BRCA2
-associated breast and ovarian cancers lost the second allele or the wild-type allele. We were somewhat surprised to find that they didn’t all lose the second allele. It was much more striking for BRCA2-
associated tumors than for tumors associated with BRCA1
mutations, most of them lost the second allele, but some did not. However, for BRCA2
, a substantial proportion did not lose the second allele or the wild-type allele.
We then went on to see if there were any differences between those tumors that retain the second allele and those that lose the second allele. We found differences that were quite significant in terms of whether they had homologous recombination deficiency or whether they had a BRCA
-associated mutational signature.
We also looked at mutations associated with those that either did or did not retain the second allele and, surprisingly, found some significant differences in mutation patterns, particularly around p53
. Those patients who retained the second allele were less likely to have p53
Finally, we looked at the outcomes of whether they retained the second allele. Ovarian cancer is easier to investigate in this context because most ovarian cancer diagnoses are at relatively late stages; therefore, there is a consistent time of diagnosis. There is also consistent treatment since most patients with ovarian cancer get platinum-based treatment.
We found that if patients retain the wild-type allele, they are equally likely to respond to platinum-based therapy as if they did not carry a BRCA
-mutation at all. However, those patients who had BRCA1
mutations and lost the second wild-type allele, were much more likely to respond to platinum-based regimens.
However, we did not see the same thing with breast cancer. In fact, we saw better outcomes with breast cancer. The problem with breast cancer is that patients are treated at different stages with different treatments making it harder to generalize based on that. However, we are able to generalize with ovarian cancer and say that if you retained a second allele of BRCA1
, you are unlikely to respond to platinum therapy. This also led us to believe that it may be unlikely to respond to PARP inhibitors, as well.
Can you elaborate on how this research affects determining treatments?
The findings affect determining treatments in 2 ways. The first is that if you retain the second allele, you are unlikely to respond to DNA-damaging agents. We particularly looked at platinum-based regimens because that was the data we had, but this could be generalized.