Immunotherapy Combos Take Next Step in Gynecologic Cancers

Numerous investigations of PD-1 inhibitors in combination with PARP inhibitors or other targeted agents, as well as chemotherapy, could finally start the next chapter of immunotherapy in gynecologic malignancies, especially in ovarian cancer, said R. Wendel Naumann, MD.

“Immunotherapy has revolutionized the treatment of certain cancers, such as melanoma. However, when it comes to gynecologic malignancies, the response rates to single-agent PD-1 inhibitors have been in the 15% range, which is fairly low,” said Naumann. “The field is now anticipating additional data on novel immunotherapy combinations that are on the rise.”

In an interview with OncLive^® during a 2020 Institutional Perspectives in Cancer webinar on Ovarian Cancer, Naumann, a gynecologic oncologist at Levine Cancer Institute, Atrium Health, highlighted challenges to date with immunotherapy gynecologic cancers, but also pointed to several exciting combinations that are setting the stage for a new era. 

OncLive^®: What is most interesting to you about immunotherapy combinations in ovarian cancer?

Naumann: It’s currently unknown if immunotherapy is going to advance the frontline treatment paradigm for patients with ovarian cancer; however, I believe it shows promise in the recurrent setting. This is especially true in smaller subsets, such as those who express high levels of tumor mutational burden or who have mismatch repair deficient [dMMR] tumors, who tend to respond very well. In ovarian cancer, dMMR tumors are much more common, as well as in endometrial cancer. Meanwhile, they are practically nonexistent in other tumor types.

We've combined immunotherapies with PARP inhibitors, such as in the TOPACIO trial, or with TKIs, such as lenvatinib (Lenvima). We've seen some response rates at around 25%, which, for platinum-resistant ovarian cancer, is a reasonable response rate to get excited about. These patients have very limited treatment options and poor outcomes.

What were the most significant takeaways from the TOPACIO trial?

This trial is evaluating the combination of pembrolizumab [Keytruda] and niraparib [Zejula] in patients with advanced or metastatic triple-negative breast cancer or recurrent ovarian cancer.

Whether patients have homologous recombination deficiency, are BRCA wild type, or harbor a BRCAmutation, they demonstrated a 25% to 30% response rate, which was a bit surprising. This could suggest that there is some sort of synergistic activity between PD-1 and PARP inhibitors. If PARP is creating DNA damage that the immune system is able to recognize, then it would certainly make sense. However, these are pretty small groups of patients, as of now.

In the LEAP-005 trial, 90% of the patients in the ovarian cancer cohort had received a minimum of 3 prior lines of therapy. Therefore, these were heavily pretreated patients with limited therapeutic options. TKIs alter the tumor microenvironment, while overcoming resistance to immunotherapy. However, we've just started to scratch the surface, so we don't know what's happening on a molecular level. Looking forward, it is important for us to understand how to

overcome resistance to immunotherapy, whether it's primary resistance, T-cell exhaustion, immune tolerance, or whatever mechanism that is preventing the immunotherapy from working.

What is so interesting about the ATHENA trial? When can we expect mature data?

The ATHENA trial is evaluating various combinations of immunotherapy and PARP inhibitors. The trial is fully enrolled, although it will be about 2 years before we obtain mature data. I would expect us to see results around 2022.

The problem with these up-front trials is that, although you may have a patient population that responds to these combinations of immunotherapy, PARP inhibitors, and antiangiogenic agents, the question is: Are those populations represented enough to be able to show a difference in overall response rate [ORR]? The ORRs are pretty good [up front]; thus, it's been very difficult to improve upon.

For example, we would expect 80% of patients with ovarian cancer to respond to up-front treatment. This is a much steeper hill than recurrent cancer, where there are limited benefits from other therapies. You can separate the patients who are going to benefit very easily from the resistant patient population. It's much more difficult [to see a benefit] in the up-front setting, where the prognosis and response rates are [already] pretty good.

How does this compare with the benefit with immunotherapy in other gynecologic cancers?

Immunotherapy in cervical cancer is interesting. Cervical cancer is a virally induced disease; thus, you would think that the immune response would be excellent. Besides for the combination of chemotherapy and bevacizumab (Avastin), with paclitaxel and carboplatin, which is a standard in cervical cancer, we don't have any satisfactory second-line treatment options right now. Pembrolizumab received regulatory approval for the treatment of patients with PD-L1–expressing tumors, with an ORR around 14%. This tells us that the response rate is fairly low, but dramatic, in the patients who do respond. The FDA felt as though this was an unmet need. Notably, when those responses do occur, they are durable.

There have been several trials, such as CheckMate-358, which evaluated the combination of nivolumab (Opdivo) plus ipilimumab (Yervoy) and demonstrated a response rate that's up to 46% in previously untreated patients. Interestingly, the toxicities increase as you start combining therapies, which is certainly going to be a challenge. Agenus [had a] study, which looks at the combination of PD-1 and CTLA-4 inhibition, demonstrated that the combination did have a better response rate than PD-1 inhibition alone. Again, there was a bit of increased toxicity. There were differing doses between those 2 trials, with respect to the CTLA-4 inhibitor dose. This could explain the different response rates that were observed, but cross-trial comparisons are difficult.

As cervical cancer progresses, we're going to see movement to the frontline setting, whether it's the combination of PD-1 with chemotherapy, or even with radiation therapy. I’m very excited about those types of trials because I believe they can boost the cure rates up front, or they will help produce durable responses in patients with recurrent disease.

The endometrial cancer is exploding. Notably, this is the first real excitement that we've seen in endometrial cancer, especially since about 20% to 30% of patients have MSI or dMMR tumors. Those patients respond extremely well to PD-1 inhibition, with response rates around 50% to 60%; these response rates are durable. At 2 years, about 50% of people who respond will actually continue to respond, which is exciting, especially considering that

the prior median survival in recurrent endometrial cancer was only about 8 months. For the MMR-proficient tumors, the combination of PD-1 and a TKI, specifically pembrolizumab and lenvatinib, have produced very durable responses in certain patients with endometrial cancer. The response rates have been around 35%.

What investigational therapies in ovarian cancer space are you most excited about? 

The novel therapies in ovarian cancer are interesting. There are a number of targets specifically used for ovarian care, which allows us to spare the normal tissues. These were initially conjugated drugs, but now, with antibody-drug conjugates [ADCs] we can put 3 or 4 of the chemotherapy molecules into the antibody. This is then directed to various targets. The most advanced target that we have studied is the folate receptor target, which is present in about 80% of patients with ovarian cancer.

The idea is that the chemotherapy is not normally active while it is bound to the antibody; however, when it attaches to the cell and undergoes [internalization], the cell releases the chemotherapy. You can use very potent chemotherapies that that have a limited adverse effect profile. The problem is that we have some dissociation of the chemotherapy—just naturally from the antibodies. Second-generation antibodies bind differently. For example, Sutro’s antibody has artificial amino acids, which helps chemotherapy moiety bind better to the antibody.

In addition, there are other targets. For example, there is an ADC that has a polymer, which has 10 to 12 molecules. This is a different target called NaPi2b, which is present in ovarian and lung cancer, but not in normal tissue. Ultimately, the idea is to see targets that are differentially expressed in ovarian cancer so that we can deliver very specific and potent chemotherapy—just to those cells.

What challenges would you like to see addressed with future research efforts?

Platinum-resistant ovarian cancer represents a very high unmet need. These patients typically live less than 1 year and have extremely low response rates. We need to understand how to have a meaningful impact on response rates and survival in the patient population. We need to figure out how to overcome resistance and find new potent chemotherapies that we can deliver in very specific way, while also minimizing the toxicities.

In cervical cancer, finding optimal combinations and ways to make tumors respond to chemotherapy has been a challenge. We must determine why some patients have resistance to immunotherapy, while others respond very well, especially since these are virally induced tumors.

For endometrial cancer, a challenge is moving immunotherapy to the up-front setting. We're currently evaluating the addition of frontline PD-1 inhibition to chemotherapy. We are hoping to produce better and more durable response rates to chemotherapy, while also overcoming resistance. We must determine if there is anything we can do for patients with dMMR tumors after they stop responding to PD-1 inhibition. How can we overcome the T-cell exhaustion or immune tolerance in those patients? For the MMR-proficient tumors we're beginning to do that, but we need to figure out how to make those tumors more immunologically recognizable, so we can improve the response rates in that patient population.