番茄社区

At the annual meeting of the American Urological Association (AUA), a half-day event titled "Research Forum: Early-Career Investigators Showcase" was dedicated to highlighting the work of early-career physician-scientists and researchers.

In this exclusive 番茄社区 video, , of Massachusetts General Hospital (MGH) and Harvard Medical School in Boston, discusses his presentation on a study dissecting the role of homologous recombination deficiency in early-stage prostate cancer.

Following is a transcript of his remarks:

So the study that we presented at the AUA Research Forum was focused on studying the role of homologous recombination deficiency in early-stage prostate cancer. We've learned a lot about the role of this DNA repair pathway in the advanced disease setting, where there's more of an established prognostic and therapeutic role for DNA repair pathways and defects in those pathways.

We know that tumors with those defects have a more aggressive biology and also have preferential response to PARP inhibitors, for instance. But not as much is known about the role of this pathway in early stage disease. There's been limited studies looking at germline mutations in favorable-risk candidates for active surveillance, for instance. But there hasn't been more of a systematic kind of analysis of this type of DNA repair deficiency in early stage disease.

So that's what we set out to do in a cohort of about a thousand patients who had undergone whole exome sequencing of their tumors. About two thirds of them were primary tumors and about a third of them were metastatic tumors. And we kind of focused mostly on the patients that would be categorized as NCCN [National Comprehensive Cancer Network] low-risk or favorable intermediate-risk tumor -- so patients who would be essentially eligible for active surveillance.

And we used a variety of computational methods to evaluate these tumor exomes for evidence of homologous recombination deficiency. And what we found in general is that as the risk category went up -- so from low risk to intermediate risk, to high risk, etc. -- we found higher and higher rates of evidence of homologous recombination deficiency.

But the first interesting observation was that in patients who had low-risk or intermediate-risk disease, there were a subset of patients who had HRD scores -- evidence of homologous recombination deficiency, similar to patients with higher-risk disease -- which made us curious about whether these were a subset of kind of bad-acting tumors among what we are calling clinically lower-risk or favorable-risk disease.

And it turned out that that subset of patients with favorable-risk prostate tumors had a higher rate of adverse pathologic features at radical prostatectomy, as well as a higher rate of biochemical recurrence after prostatectomy.

And in fact, when you compared the HRD score, or you put it in the same model as well-validated risk predictors like the CAPRA-S [Cancer of the Prostate Risk Assessment, Post-Surgical] score, which is essentially a risk-predicting nomogram for biochemical recurrence after surgery, the extent of HRD, the extent of homologous recombination deficiency, actually had a similar predictive power and was a statistically independent predictor of biochemical recurrence compared to the CAPRA-S score. So it kind of added additional information above and beyond that, which we found to be an interesting finding.

And then we kind of moved after that to trying to study the timing or clonality of homologous recombination deficiency. So what we mean by that is trying to study when in the tumor's evolution these signatures become active -- so when does this mutational process become active? Is this something that's present from the very beginning of the tumor's formation, or is it something that evolves later during tumor progression?

And we found that most of the time when the HRD signature was present, it was clonal -- meaning that it was present from the earliest kind of stages of tumor development. And then we went on to kind of confirm that empirically in a small number of patients who actually were on active surveillance in our institution, so among the MGH active-surveillance cohort, we took patients who had either done really well on surveillance and remained with low-grade, low-risk disease for a number of years compared to some patients who had fallen off surveillance due to progression of their disease.

And it turned out, we were able to detect evidence of HRD even in the earliest biopsy samples years before the patient had clinical progression on biopsy and subsequent radical prostatectomy.

And so it's a small number of patients, but definitely needs validation in additional cohorts. But it's an encouraging finding that suggests that we may be able to detect evidence of HRD early on in a tumor's evolution. And that may be able to help inform precision cancer medicine strategies around active surveillance, decision making, focal therapy, or adjuvant treatment strategies after surgery.

Comment