Westin Highlights the Evolving Role of PARP Inhibitors in Ovarian Cancer

Shannon Westin, MD

PARP inhibitors have exploded into the ovarian cancer space, demonstrating reductions in the risk of disease progression and garnering FDA indications in the relapsed, maintenance, and recently, the upfront maintenance settings, said Shannon Westin, MD. As even more options enter the armamentarium, molecular tumor testing is becoming even more important for selecting which therapy is best for an individual patient.

“We have been very lucky to discover these drugs and to be able to utilize them in ovarian cancer,” said Westin, an associate professor at The University of Texas MD Anderson Center.

The initial indications were in the recurrent space. In 2014, the FDA granted an accelerated approval to olaparib (Lynparza) for patients with advanced ovarian cancer who have had multiple lines of prior therapy, had either a germline or somatic BRCA mutation. Rucaparib (Rubraca) received an accelerated approval December 2016 for those who had been treated with ≥2 chemotherapies and whose tumors had deleterious BRCA.

Several indications in the maintenance setting soon followed: olaparib, rucaparib, and niraparib (Zejula).

In December 2018, the FDA granted olaparib approval in the frontline maintenance setting in patients who have germline or somatic BRCA mutations who are in complete or partial response to first-line platinum-based chemotherapy. The decision was based on positive data from the SOLO-I trial, which was a randomized, multicenter trial that compared the efficacy of olaparib with placebo in patients with BRCA¬-mutated advanced ovarian, fallopian tube, or primary peritoneal cancer following first-line platinum-based chemotherapy.

For the trial, patients were randomized to either receive oral olaparib tablets 300 mg twice daily (n = 260) or placebo (n = 131). Results showed a statistically significant improvement in investigator-assessed progression-free survival (PFS) in the olaparib arm versus the placebo arm. The estimated median PFS had not been reached in those who received olaparib versus 13.8 months in those who were given placebo (HR, 0.30; 95% CI, 0.23-0.41; P <.0001).

Other PARP inhibitors are also being investigated in the space. Veliparib (ABT-888) is being explored in combination with chemotherapy, while talazoparib (Talzenna) is being assessed in the frontline maintenance setting for ovarian cancer.

Regarding genetic testing, providers have been utilizing them in the recurrent setting to help select a PARP inhibitor for treatment, said Westin. However, as more information is reported, it will be the responsibility of healthcare professionals and institutions to determine what the best approach may be for performing these tests.

OncLive: What are some recent advances made with PARP inhibitors in ovarian cancer?

In an interview with OncLive during the 2019 SGO Annual Winter Meeting, Westin discussed the evolving role of PARP inhibitors in ovarian cancer, novel investigational agents, and the importance of molecular testing.Westin: The initial indications were in the recurrent space with measurable disease. These are patients who have had multiple lines of prior therapy and have a mutation in BRCA, whether that be germline or somatic. [These patients] could be treated either with olaparib or rucaparib, which are the 2 drugs that have indications. [This is] still a very nice option for our patients who are past the upfront and early recurrent settings.

Then, we received a series of indications in the recurrent setting, so platinum-sensitive recurrence for a maintenance strategy. [This is] different from treatment in that we're dealing with measurable disease that we're trying to [reduce]. These are patients who had a response to a platinum-based therapy in that setting and then are getting transitioned—a switch maintenance—to one of the PARP inhibitors. There are 3 PARP inhibitors that are available there: olaparib, rucaparib, and niraparib.

Could you highlight the 3 approved PARP inhibitors and their roles in this paradigm?

Most recently, we got an approval for the upfront setting, for patients who have a germline or somatic BRCA mutation, received upfront therapy, and have had a nice response. They can be transitioned to maintenance PARP; specifically, olaparib received the indication in that space. That’s exciting and it will be interesting to see how [this plays out]. With patients now starting to get PARP in the upfront setting, how might that affect what happens in later lines of therapy?The initial data that led to the approval [of olaparib] in the recurrent space were early phase I/II trials where [investigators] selected out a population of germline-mutant ovarian cancer who had received 3 or more prior lines of therapy. They had multiple lines of prior therapy, but they still saw responses in that group. This was one of the first times for us that the FDA approved [an agent] based on response rates alone. [They] didn't look at PFS or OS; [the decision was] just based on pure response. Frankly, that's because there are so few agents that have any response rate over 5%, so to see a 25% to 30% response rate was exciting.

Then, rucaparib came not long after; that was an indication in both germline and somatic BRCA patients who had received 2 or more prior lines of therapy, had some measurable disease, and were treated.

In the maintenance setting, it was definitely a staggered approach. The first drug to get approval in the maintenance setting was niraparib; it was in an all-comers setting. The NOVA study took all comers, high-grade serous, or if they were another histology but had some type of BRCA mutation, they could go on niraparib. They had platinum-sensitive disease, had a response to chemotherapy, and then were randomized to niraparib or placebo. They demonstrated a clear improvement in PFS, and there was a lovely relative risk reduction across all settings. It was best in the BRCA group—a little less good but still good in the more homologous recombination deficient group, and still had an impact in the population that was completely biomarker-negative. Therefore, that yielded them an approval across all comers.

Not long after that, olaparib maintenance data were presented; there were 2 studies that led to their approval. The first was Study 19, which was in an all-comers setting; [patients were] platinum-sensitive, had a response to therapy, and were randomized to olaparib or placebo.

[They looked] at PFS and demonstrated the difference in all comers. There was an additional study called SOLO-2, which was in that same population but only [those who were] BRCA mutant. There was a clear difference between the patients who received olaparib and those who received placebo. They took the combination of all those data to yield an indication in all comers, so anyone with high-grade serous ovarian cancer can get that maintenance strategy.

Last but not least, there is rucaparib. They had the ARIEL3 study, which had a very similar study design; patients had a response in the platinum-sensitive recurrent setting, and [it was in] all comers, although they did tease out and stratify by germline and somatic mutation in BRCA, versus some other abnormality in homologous recombination gene pathway, than just completely BRCA wild-type. They looked at each one of those groups, and again, similar to what we saw with NOVA, similar to what we saw with Study 19, diminishing returns but still a very significant difference and very significant improvement in PFS amongst those patients who were treated with rucaparib. [Those results] yielded them an indication in all of those settings.

Looking at olaparib, rucaparib, and niraparib as maintenance therapy, do any of these agents provide more benefit than the others? How do you choose between them?

Most recently, Kathleen Moore, MD, of the Stephenson Cancer Center, presented the SOLO-1 data. Those data were only in patients with mutations—predominantly germline BRCA-mutated patients—but also there were some patients with somatic mutations. They had upfront therapy, had a complete response to it, and then were randomized to receive either olaparib or placebo. They received it for only 2 years, so it wasn't continued until progression. They demonstrated a clear benefit in PFS even past when the patients received it. That PFS benefit exceeded 2 years; it went on and was demonstrated in 3 years. Therefore, these was very exciting data [showing a] very impactful reduction in the risk of progression.There have been no data to indicate one is more efficacious than the other. There are some studies looking at relative potency and things like that, but we don't know if potency yields better efficacy. Certainly, when you look at the curves, with any one of these drugs—whether it be niraparib, olaparib, or rucaparib—compared with placebo, there's a clear benefit; there's a clear difference in PFS. Then [you need to consider] patients' existing toxicities and side effects. [For example,] a patient who has hypertension may not want to use niraparib, and in a patient with liver issues, rucaparib may not be the best choice. You can kind of select and choose based on their existing issues.

Are there other PARP agents under investigation that you're excited about?

The schedule of treatment is also a factor. Both rucaparib and olaparib are required twice daily; niraparib is only once daily, so that might help you decide if you have a patient who doesn't want to take a bunch of pills. These are the types of factors that we consider in order to make those decisions, but efficacy isn't one of them.Veliparib has been looked at; that's exciting because we have been able to potentially combine that with chemotherapy. All the studies we spoke about are either evaluating PARP alone in the treatment setting or PARP on its own in the maintenance setting. It has been a little tricky to combine some of those PARP inhibitors with our standard of care chemotherapy of paclitaxel and carboplatin. Veliparib is a little bit easier to combine. There was a recent study that completed that combined veliparib with chemotherapy followed by [PARP] maintenance, and we're all awaiting the data. It will be very exciting to see if that's impactful, [if it will be] similar to what we've seen with bevacizumab (Avastin) in combination with chemotherapy.

Where does molecular tumor testing factor in here? When should it be used?

Talazoparib is another PARP inhibitor that has had a lot of activity in breast cancer. Investigators are starting to explore it now in upfront maintenance for ovarian cancer, so it'll be interesting to see if there's any difference or any improvement in the effects with that drug.This is a kind of work in progress. I gave a similar talk about genomic testing about 6 months ago and had a completely different answer [than I do now]. I would say, in general, we've been using this in the more recurrent setting. Previously, we didn't have any indications for any molecularly targeted agent upfront, and so, we were using it in the recurrent setting to help us select PARP inhibitors for treatment and then also look for novel therapies.

There a lot of the new drugs that are getting tissue agnostic indications. It doesn't matter where the cancer started; if you have this aberration, you can get this drug. A recent example is pembrolizumab (Keytruda) for microsatellite instability-high (MSI-H) tumors. It doesn't matter if you got it for endometrial cancer, ovarian cancer, or bladder cancer; if you have an MSI-H tumor, you potentially can benefit from pembrolizumab.

Larotrectinib (Vitrakvi) just received approval in NTRK fusions. Again, this is a rare aberration found in 1% to 3% of solid tumors, but [there is an] 80% response rate in those tumors—that is unheard of. You potentially have a definitive treatment for a patient population that you previously had nothing for. That is where I was most excited; yes, there are costs associated with it, but as we get more of these targeted agents, we're going to have more opportunities to give these tiny select patient subgroups options.

Now, however, with our SOLO-1 indication—the indication for olaparib in the upfront setting—you must know upfront if that patient has a somatic mutation in BRCA. We have been generally testing any high-grade patient for germline BRCA mutations, as well as other potentially homologous recombination genes; however, it hasn't been universal. We're slowly improving; we have been identifying those patients, but we haven't been doing a lot of somatic testing upfront.

What are the challenges associated with this testing that we still need to overcome?

Now, it will be on the providers and the institutions to determine the best way to do that testing. Are we just going to do a quick somatic test just for BRCA, or are we going to take that opportunity and do a whole panel of genes—300 or 400 genes—at one time? That way we have that information and we won’t have to retest that patient down the line. Still, it's hard to know where things are going to fall with that. Our institution is actually working on that right now and trying to determine that the best opportunity is there.Cost is, of course, an issue. However, the costs for this testing have gone down markedly. Now that we have some FDA approvals for companion diagnostic tests, insurance companies have gotten a little bit more open to that testing—especially in ovarian cancer—so that has definitely opened up the floodgates a little bit.

However, if you're retesting someone, there is probably going to be some pushback there. Should we just do the whole panel at that time? However, that leads to another challenge. If we do the whole panel testing in the upfront setting, is that going to be informative to someone who needs treatment 3 years later? We don't know that answer yet. We don't know how much that tumor is going to change with other therapies that are received.

That has been our challenge, and generally, we've been doing a lot of our somatic testing down the line, so it represents exactly what's going on right now. However, who knows what that will mean for us if we do that testing 3 years prior. We do see a lot of tissue heterogeneity, so if I biopsy the lung compared with the primary ovarian cancer, are those mutations consistent? Is that molecular milieu consistent? We don't know the answer to that.

What is your take-home message?

In fact, some of the small studies have suggested that [the answer is], “Maybe not.” That has been a real challenge, making sure that the tissue you're selecting represents what's going on in the patient and that the drug you might be selecting is actually going to work.Just to make sure that people stay on top of this. It is truly an evolving science; there are a lot of exciting things coming down the pike. For example, liquid biopsies have been getting a lot of attention in some of the other cancer types for diagnosing, but there could also be potential in us identifying who might benefit from a particular treatment. Also, you can actually test as patients are getting treatment to determine if a patient is responding before you see anything in the imaging. It is on all of us to keep up with the research as it comes out. We need to keep up with those data and do our best to stay on top of it all, and modify our practice based on this rapidly evolving field.

Moore K, Colombo N, Scambia G, et al. Maintenance olaparib in patients with newly diagnosed advanced ovarian cancer. N Eng J Med. 2018; 379:2495-2505 doi: 10.1056/NEJMoa1810858

View more from the 2019 SGO Winter Meeting