Advanced Ovarian Cancer: New Perspectives on Systemic Therapy - Episode 2

Predictive Biomarkers in Ovarian Cancer


Bradley J. Monk, MD, FACS, FACOG: Katie, let’s transition to frontline PARP inhibitors. We’ve now had 4 New England Journal of Medicine papers in 3 years that changed treatment in ovarian cancer. Yours was last year. We’re about at the 1-year anniversary. Teach us about frontline PARP inhibitors in newly diagnosed advanced ovarian cancer in your wonderful study SOLO-1.

Kathleen Moore, MD: SOLO-1 was actually launched in 2013 and just reported last year because of how well the patients did. It was really the first time we incorporated PARP inhibitors into the frontline treatment armamentarium. It was the first opportunity to do that, and it selected patients who were at the highest likelihood to benefit, which were those women with BRCA-associated cancers. It enrolled women with advanced-stage disease that was BRCA related. They had to have an attempt at a surgery. At the end of their induction platinum-based chemotherapy, which they had had at least 4 cycles of, they had to be either in complete or partial response.

This is an important distinction, because this really ushered in an era of switch maintenance. That is different from what we talked about earlier with bevacizumab, which is basically continuance maintenance. This is not the very first, but 1 of the first, opportunities to look at that strategy.

The primary endpoint for this study was progression-free survival. Secondary endpoints were overall survival and a number of others, including quality of life. We presented the progression-free survival at ESMO [European Society for Medical Oncology Congress] in 2018. There was a median 42 months of follow-up, which is double what we saw yesterday because that’s how long it took to get to the endpoint.

We still have not reached the median progression-free survival for the women who were treated on olaparib. The women who were not treated on olaparib had a median progression-free survival of 13.2 months from the end of chemotherapy. The average time on chemotherapy was 7 months. It’s about 20 months of progression-free survival for the control group and about 49 months for the treatment arm with a hazard ratio of 0.3.

Bradley J. Monk, MD, FACS, FACOG: That’s almost 36 months.

Kathleen Moore, MD: Right. We think we improved progression-free survival by about 3 years.

Bradley J. Monk, MD, FACS, FACOG: That’s unbelievable. Put that in perspective to what you said about bevacizumab. Bevacizumab was a hazard ratio of 0.62—all comers, I get it—with about 6.2 months. You have half the hazard ratio at 0.3, and 30 months and about 5 times the median. That was really exciting.

Michael J. Birrer, MD, PhD: If you look at those curves when BEV [bevacizumab] is stopped, the curves come back together.

Kathleen Moore, MD: All come down. That’s right.

Michael J. Birrer, MD, PhD: However, the effects from PARP inhibitors in SOLO-1 seemed to be lasting.

Bradley J. Monk, MD, FACS, FACOG: It was oral treatment for 2 years. Rob, let’s set the stage here. Tell us how PARP inhibitors work, so we can understand that.

Robert L. Coleman, MD, FACOG, FACS: The easiest way of thinking about the PARPs, and the 1 that we talk about frequently, is that PARPs are responsible for basic excision repair. When we inhibit that, it makes it statistically more likely that the cell will have a double-strand break. It relies on homologous recombination to fix the double-strand break because of the high-fidelity system. Of the many systems that can fix DNA, that homologous recombination response is the highest fidelity through the BRCA-driven mechanisms.

With tumors that are vulnerable because they don’t have a homologous recombination compliancy, which we call the HRD, or homologous recombination deficiency, those cells are sick and they die. They’re unable to fix the DNA damage that we’ve induced. We specifically promote the opportunity for the double-strand break to happen and then take advantage of the fact that the cancer cell can’t fix itself. You have 2 nonlethal events that lead to a lethal event.

Bradley J. Monk, MD, FACS, FACOG: PARP repair single-strand breaks. If you inhibit that, they form a double-stranded break, and BRCA repairs double-stranded breaks.

Robert L. Coleman, MD, FACOG, FACS: Correct.

Bradley J. Monk, MD, FACS, FACOG: With an unrepaired double-stranded break, they apoptosis. You talked about this HRD, so it’s not just BRCA—the sister chromatids or homologs could be lined up. The double-stranded breaks can be repaired through BRCA light chain, and you called that homologous recombination. Let’s spend a few more minutes here and then transition to what we saw at this meeting about how to test for HRD. I get it—we test for BRCA, but we really don’t test in this company for homologous recombination repair deficiency. However, we can. Tell me, Michael, how we can test for HRD.

Michael J. Birrer, MD, PhD: The historic perspective here is that we’ve always been interested in sequencing. We sequenced BRCA1 and BRCA2. Then we learned that there are probably some other genes that are important and sequenced them. It became clear that that’s a lot of work and a lot of money, and it was questioned whether there would be functional measures of this homologous recombination deficiency. It was observed once these big genomic databases were available and you could look at the BRCA patients, who we knew had mutations compared with everybody else. They had different patterns of what we call loss of heterozygosity [LOH].

When you have a DNA repair abnormality like homologous recombination deficiency, the cell loses pieces of DNA, and you can map that. LOH is the basis of these HRD assays. Foundation Medicine, Inc has a relatively pure LOH assay. Myriad Genetics, Inc has an LOH assay, and then they add on to it something called telomeric imbalance or allelic imbalance because the telomeres are so far out at the end of the chromosomes that it’s hard to map them. They add that in specifically.

Then they also have these large-scale transitions, which are so big that you can’t map them with LOH. The 2 assays are very similar. It’s a little more complex with Myriad, and they’re looking at a functional state of homologous recombination deficiencies.

Bradley J. Monk, MD, FACS, FACOG: Both the LOH-high and the Myriad HRD assays are commercially available.

Michael J. Birrer, MD, PhD: That’s correct.

Bradley J. Monk, MD, FACS, FACOG: Shannon, I get that we’re supposed to test for BRCA, right?

Shannon N. Westin, MD, MPH: Yes.

Bradley J. Monk, MD, FACS, FACOG: Do you think the community is doing HRD in newly diagnosed advanced ovarian cancer?

Shannon N. Westin, MD, MPH: Generally, no. I think that with the data that we had previously, there wasn’t really a reason to get HRD up front, right? Since we know from Dr Moore’s SOLO-1 data, we wanted to get that BRCA information right away—not just germline but also somatic. There’s a small proportion of that trial in whom the thought would be that you’re still going to get benefit. It’s really small. It’s still good information, but we didn’t have a reason to look for that HRD right away up front.

Bradley J. Monk, MD, FACS, FACOG: Yes, I think that’s a point.

Robert L. Coleman, MD, FACOG, FACS: That’s a point to emphasize with Katie’s talk. That then drives home the fact that we need to know this information. That testing, as you just mentioned, is now essential.

Bradley J. Monk, MD, FACS, FACOG: All BRCA patients are HRD. Is that right?

Michael J. Birrer, MD, PhD: Based on what we know scientifically, all BRCA patients are going to be HRD-positive. It gets a little bit complicated because of the nature of the mutation, but we’re not there yet.

Bradley J. Monk, MD, FACS, FACOG: Maybe when we say HRD, we should take the BRCA patients out because all BRCA patients are HRD-positive, and we’ll get into that. This is 1 of the key take-home messages of today. We’re going to talk about data from frontline PARP inhibition, and there is an opportunity in this HRD beyond BRCA.

Shannon N. Westin, MD, MPH: Absolutely.

Bradley J. Monk, MD, FACS, FACOG: That’s why, at this meeting, the first 3 presidential session papers were all ovarian cancer based, 2 of which have already been published in the New England Journal of Medicine.

Transcript Edited for Clarity