Robert L. Coleman, MD, FACOG, FACS, discusses the evolution of treatment for patients with ovarian cancer, the therapies with the most potential, and the importance of classification schemes in directing treatment in recurrent disease.
Robert Coleman, MD, FACOG, FACS
Although the prevalence of ovarian cancer has increased in the last 3 years, incidence and death rates have decreased as a result of novel therapies, including angiogenesis inhibitors, immunotherapies, PARP inhibitors, and refined patient classification schemes, explained Robert L. Coleman, MD, FACOG, FACS.
“We’re excited about where this field is going now that we have some active compounds,” said Coleman, a professor in the Department of Gynecologic Oncology and Reproductive Medicine at The University of Texas MD Anderson Cancer Center.
In an interview during the 2018 OncLive® State of the Science Summit™ on Ovarian Cancer, Coleman discussed the evolution of treatment for patients with ovarian cancer, the therapies with the most potential, and the importance of classification schemes in directing treatment in recurrent disease.Coleman: One of the things that I’ve mentioned many times, but [that] is becoming more apparent, is that the prevalence of ovarian cancer is increasing. A lot of that is due to the fact that women who have recurred are being treated multiple times with therapies that are extending the time of progression independently. Very few trials we have now can show an overall survival (OS) change in patients who enter the trial at one point as a cohort. We do see a lot of progression-free survival (PFS) advancements. On an individual level, when that is what a patient experiences, the life expectancy of that patient increases.
I pulled the Surveillance, Epidemiology, and End Results data to see what the prevalence of ovarian cancer is doing. Incidence and death rates are declining, but they’re declining at the same rate, by approximately 1% per year. The prevalence, or the number of women who have the disease at any one time, has increased by almost 20% in the last 3 years. That’s really remarkable. [The decline] is a tribute to a lot of the standards of care that we’ve now developed through multiple clinical trials.
I walked through the treatment regimens for the individual classification schemes. Traditionally, we’ve looked at the time the patient is finished with frontline treatment and the time they recur. This is the platinum-free or treatment-free interval. That number seems to be increasing slightly because we’re getting a little bit better at surgery and we’re doing a little bit better job with our frontline therapies. Now, bevacizumab (Avastin) is approved in the frontline setting.
These patients are classified as platinum-sensitive or -resistant. Now, we need to put more details into that. The molecular subtype for BRCA-positive and -negative patients must be considered as well as their histology and treatment-free interval. These factors have now been incorporated into how we classify patients and what we use for their treatments.
Platinum-resistant patients are generally approached with single agents. We have done some doublet chemotherapy, but the only regimen that seems to have improved the PFS again has been the addition of an anti-VEGF therapy like bevacizumab. The AURELIA trial led to an improvement in PFS and in response, which is important because many of these patients are symptomatic. In 2014, it [became] an additional treatment option for women for which that therapy was considered safe in. Chemotherapy plus bevacizumab is now an approved label.
Most of the therapy for the larger populations of patients that we see, who are initially platinum-sensitive, is dictated by platinum-based doublets. There was a trial that looked at whether or not we should use platinum-based chemotherapy first or following a nonplatinum agent. Results showed that patients did better when they got the most active therapy first; that was the platinum-based chemotherapy. That kind of settled the question of whether or not we should start with something else and see whether we could extend the platinum-free interval. Most of us are convinced that, in many cases, platinum-based chemotherapy with bevacizumab has improved PFS. In GOG-0213, which included paclitaxel and carboplatin, there was an improvement in OS.
The ground rule for patients with platinum-sensitive disease, whether it’s initially platinum-sensitive or considered platinum-sensitive down the road, is to use some kind of platinum doublet in combination with an anti—VEGF-based therapy.
The newest thing to enter this domain is PARP inhibitors. Initially, in 2014, there was olaparib (Lynparza) and then in 2016, there was rucaparib (Rubraca) in a select group of patients. These are patients who carry the BRCA mutation. Since then, there have been 3 major phase III trials showing that these PARP inhibitors are of merit in the recurrent platinum-sensitive population. The biggest “bang for our buck” is in the patients who carry this homologous recombination deficiency (HRD) signature, whether it be BRCA or non-BRCA HRD. That is starting to enter our treatment paradigm. We call that therapy switch maintenance, because it is an induction of platinum-based therapy with a response. Then, switching these patients onto an alternative drug that, at least in ARIEL3, led to responses.
What is the next step? This conference provided the opportunity to look at what those next steps are. There are some very interesting ideas. One of the next steps that we’re looking at in the recurrent setting, especially in patients who carry the BRCA mutation and have the highest probability of responding, is to use a nonchemotherapy [regimen] like a PARP inhibitor alone or in combination. We’re also looking at this in the frontline population. The angiogenesis inhibitors, the immuno-oncology agents, and combinations with PARP inhibitors as doublets and triplets are already being evaluated. We’re excited about where this field is going now that we have some active compounds.The impact of PARP [inhibitors] in the recurrent setting has been pretty profound. [Although] the maintenance indication has been around for a while, I’m surprised it’s not taking up more. It makes me think that either people don’t believe the data or they’re selecting certain patients for which this therapy might be considered most efficacious. A patient identified early to have a BRCA mutation is high on their list to receive the therapy. Now we’re routinely doing germline assessment and some centers are doing somatic assessment for BRCA mutations.
If a patient has a BRCA mutation, the question is whether they have had a PARP inhibitor yet, because we know it has single-agent activity, which rivals chemotherapy. We’ve already started doing trials of chemotherapy versus PARP inhibitors in the treatment setting. The labels are indicated for patients with somatic and germline BRCA mutations. The drugs have a big impact in those patients.
The maintenance indications probably still need a little more education. We’re grateful for opportunities like this conference to be able to talk about what the role of these drugs are in the maintenance setting. As many know, we’ll have data in the frontline maintenance setting very soon for olaparib and, in the next year, data for niraparib (Zejula).We have learned a lot about where immunotherapy plays a role in ovarian cancer. It’s clear that we can’t just adopt what we’ve seen in other tumor types and expect the same kinds of outcomes. The work that has been done over the last 3 to 4 years has shown that the tumor mutational burden (TMB) of high-grade serous ovarian cancer is relatively low. It’s different than the other tumor types, such as lung cancer and melanoma, that we see high activity in. We’ve shown in single-arm trials that the response rates are not that great. Every single one those trials has somebody who had an unusual response who lived a long time. We know that from our initial assessment of those patients, it’s not due to a recognizable biomarker. We’re still hunting for that.
There are subgroups of patients for which these drugs might work. We’ve said that patients with a BRCA mutation have endogenously higher levels of TMB and tend to have higher levels of predicting the tumor neoantigens. We expect that they may work better in that scenario. The data that we have seen recently from the MEDIOLA trial looking at a PD-L1 inhibitor [durvalumab (Imfinzi)] in combination with olaparib showed some very high response rates in that subgroup of patients.
Interestingly, [we may be able to] use the same idea in a population of patients who do not have a BRCA mutation. [This was seen] in the TOPACIO trial, which looked at niraparib and pembrolizumab (Keytruda). By doing this, we may be able to overcome this kind of barrier that we’re seeing with single-agent immunotherapy. We know that single-agent PD-L1 [inhibitors show] a 10% to 15% response rate. The response to a PARP inhibitor in a BRCA wild-type patient is also relatively low. By putting the 2 together, we are starting to see response rates that are better than we would expect from either one of them [alone].
Identifying patients is a strategy. Identifying combinations that would make sense is another strategy. Those strategies will be enhanced by a number of different agents. There is the use of stereotactic radiation, and there is an expansion of using angiogenesis inhibitors that seems to potentially work. Learning more about the microenvironment, what kind of immune cells are present, their morphology, and whether they are exhausted offer opportunities to add agents that would specifically alter the dynamic that limits the response to these drugs.
We’re also doing cellular-based therapies for ovarian cancer. Tumor-infiltrating lymphocyte-based therapies and chimeric antigen receptor T-cell products are expanding. We know that we have a little bit higher hill to climb than other tumor types.In frontline disease, the GOG-0218 and ICON7 trials led to nearly universal approval of bevacizumab. In certain patient subgroups in the United States, bevacizumab gained approval for a supplemental application in June 2018. For many places, for instance in Texas, we were unable to get that kind of therapy approved for frontline use, so it’s already having an immediate impact. Of course, we believe that the data are of value. We want to give the best therapy we can at that time rather than waiting. It’s entering into our practice guidelines, and it will continue to expand in the frontline setting now that it’s more available.
It also raises a good question. In platinum-sensitive and resistant patients, bevacizumab is already approved. The question is, does exposure to bevacizumab at one time negate or have an impact on bevacizumab later? Fortunately, that question has been asked and has now been formally addressed in a phase III trial where patients had to have bevacizumab as part of their primary therapy. They were randomized to a chemotherapy regimen with or without bevacizumab. We saw that the benefit with bevacizumab was recapitulated in that audience. It was not negated by their prior exposure. We saw the same effect that we saw in all of our platinum-sensitive patients, many of whom were previously untreated or unexposed to bevacizumab in the frontline setting.
To me, it had always made sense because the individual target of anti-VEGF therapy is not just on cancer cells. The tumor microenvironment, endothelial cells, and pericytes already express those receptors. We would expect that even if the tumor were mutated away from de-repressing the VEGF receptor, this therapy would still have an effect on OS.