2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Ulka Nitin Vaishampayan, MBBS, discusses factors to consider for choosing between immuno-oncology/immuno-oncology and VEGF TKI/immuno-oncology combinations for the frontline treatment of patients with renal cell carcinoma, areas of ongoing research in this disease, and treatment updates across the prostate cancer spectrum.
Although multiple immuno-oncology (IO)–based combinations are approved for the frontline treatment of patients with metastatic renal cell carcinoma (RCC), the majority of these patients eventually experience disease progression, and novel agents and combination therapies are needed to improve outcomes in this setting, according to Ulka Nitin Vaishampayan, MBBS.
“In both kidney and prostate cancer, our work is far from done. We have a long way to go, and evaluating novel treatments and pathways is going to be of critical importance. For patients whose diseases are not responding to standard therapy, even in the second- or third-line setting, please consider clinical trials,” Vaishampayan said in an interview with OncLive® following a State of the Science Summit™ on prostate cancer and RCC, which she chaired.
In the interview, Vaishampayan, director of the Phase I Program at the Rogel Cancer Center and professor of internal medicine at the University of Michigan in Ann Arbor, discussed factors to consider for choosing between IO/IO and VEGF TKI/IO combinations for the frontline treatment of patients with RCC, areas of ongoing research in RCC, and treatment updates across the prostate cancer spectrum.
Vaishampayan: There are 5 different FDA-approved combinations in the frontline [treatment of patients with RCC]. One is an IO/IO combination with ipilimumab [Yervoy] and nivolumab [Opdivo]. The other 4 are VEGF TKI/IO combinations. There are some factors to consider, especially with current updates. Overall survival [OS] and response rates [between combinations] seem to be similar. However, the IO/IO combination [of nivolumab and ipilimumab] has a slightly less chance of immediate response.
Therefore, for patients with significantly symptomatic, high-volume disease, a VEGF TKI/IO [combination] would be preferred because you improve symptoms relatively quickly. The longer-term outcomes seem to be the same, although [treatment with] IO/IO has an edge because patients have much longer treatment-free intervals. You're able to stop treatment and the patient can continue in remission for a prolonged period.
In general, VEGF TKIs/IO combinations have about a two-thirds chance of response, but tumor shrinkage can happen in majority of the patients. Only about 5% to 10% of patients have clear progression on therapy. With the response rates and in terms of progression-free survival [PFS], all the VEGF TKI/IO combinations are showing promising numbers.
There are some nuances and differences in terms of toxicity. For toxicity, it is important to anticipate and educate [patients]. Some of the specific toxicities were discussed at the meeting, including hypertension, hand-foot syndrome, and mucositis, and ways to handle these. Shawna Kraft, PhD, of the University of Michigan Rogel Cancer Center, gave tips of how to address each of these toxicities and the best antihypertensives to use in each situation, [stressing the importance] of monitoring [for these toxicities].
Usually [with] VEGF TKIs, the first few weeks [of treatment] inform both patients and providers as to how toxicities are going to pan out. Diarrhea is another one that needs to be addressed as soon as possible, as well as hand-foot syndrome or skin rashes. Skin protection or using emoluments should be a critical part of supportive care.
There are a number of trials [examining] different ways of addressing how to tackle [eventual disease progression]. What we are seeing, even with long-term updates for each of these studies, is that we are not curing a majority of these patients. Many of these patients are going to continue to progress on this therapy, even if they respond initially, and [a portion] of patients are refractory up-front to frontline therapy. With that in mind, the development of novel agents is in progress.
The HIF-2α inhibitor belzutifan [Welireg] is an [agent] that’s currently [FDA] approved for patients with [cancers linked with] von Hippel-Lindau [VHL] disease.1 Triplet combinations in the frontline setting of HIF-2α inhibitors plus VEGF/IO combinations such as lenvatinib [Lenvima] and pembrolizumab [Keytruda], are in testing.
The [phase 3] COSMIC-313 trial [NCT03937219] looked at the triplet of cabozantinib [Cabometyx], ipilimumab, and nivolumab, and [this study] met its primary end point of improvement in PFS [vs nivolumab plus ipilimumab alone].2 Unfortunately, without adequate and mature OS follow-up, it is hard to justify using this triplet. [The triplet] also needs to be compared to a sequential regimen of ipilimumab plus nivolumab followed by cabozantinib, which is what is typically used right now when you use ipilimumab plus nivolumab in the up-front setting.
Those kinds of combinations and novel pathways are [being investigated]. Novel pathways include the cytokine pathways and LAG-3 inhibition, which are gradually starting to begin clinical trials. There are also novel CTLA-4 inhibitors, such as ONC-392, that we are investigating. Those are things on our radar that will hopefully lead to bigger randomized trials, assuming that they show adequate efficacy in the phase 1/2 trials.
The top toxicities with VEGF TKIs include fatigue, hypotension, diarrhea, and skin rash. Those are the main things that you need to discuss with patients, anticipate them, and educate [patients] and their caregivers. Typically, we'll find that these toxicities start developing within the first few weeks of therapy, and dose adjustment is an effective way to reduce the toxicity impact. However, sometimes excellent supportive care [alone] can be adequate to relieve the toxicity and continue with the same dose.
Usually, if it's a severe toxicity, you should hold the therapy. When there is an overlap between toxicities when treating a patient with a VEGF TKI and IO [agent], it is important to hold both. If the toxicity improves on its own, then it is [likely] a VEGF TKI–related toxicity, but if it's not improving, even after holding the therapy, then it's more likely an IO-based toxicity. [In the case of a toxicity] such as immune colitis, the way to [determine if it is related to a VEGF TKI or IO] is to wait it out. If they're not improving, then you need to put the patient on steroids and immunosuppression to try to reverse the immune colitis.
For metastatic disease, the role of cytoreductive nephrectomy has been changing continuously over the decades. When we had suboptimal immune therapy, the role of cytoreductive nephrectomy had been shown to improve OS. Now, the pendulum shifted the other way in the last few years. With a sunitinib [Sutent]- or VEGF TKI–based regimen, there was no role and no improvement [in OS] with cytoreductive nephrectomy. What we do know is that up-front cytoreductive nephrectomy is not a good idea in advanced kidney cancer, unless you have a solitary metastasis that you can resect to put the patient in complete remission.
A majority of synchronous metastatic disease, which is when a patient presented with a primary renal tumor and metastases, needs to be managed carefully up-front, starting with systemic immune-based regimens, followed by consideration of cytoreductive nephrectomy at a later date, assuming the patient is benefiting from the frontline therapy. That is what the ongoing phase 3 SWOG 1931 trial [NCT04510597] is designed around. This trial is randomly assigning patients after starting [standard-of-care], systemic IO-based regimens. At about 10 to 12 weeks, responses are assessed, and then assuming that the patient has at least stable disease, they are eligible for randomization to get nephrectomy or not.
Part of [the rationale for] deferred nephrectomy comes from [the phase 3] SURTIME trial [NCT01099423], which compared [immediate and deferred] cytoreductive nephrectomy. The deferred arm had a median OS [of 32.4 months] vs [15.0 months] for the up-front cytoreductive nephrectomy arm.3 Additionally, there is a lot of preclinical information that tells us that immune checkpoint therapy has a better chance of working if you leave the primary [tumor] in place while you're treating.
In melanoma, there was a randomized [phase 2] trial [NCT03698019] that showed that there was better [event-free survival] when giving immune checkpoint therapy before resection of the melanoma.4
Deferred cytoreductive nephrectomy is the way to go, but in kidney cancer, it remains to be proven, hence the [SWOG 1931] trial.
There are three FDA-approved regimens: apalutamide [Erleada], enzalutamide [Xtandi], and darolutamide [Nubeqa]. All 3 clinical trials had similar designs. They were all placebo controlled, double-blind, randomized trials, with a primary end point of metastasis-free survival. Standard scans were used for imaging in that study to determine metastases.
Typically, I look for patients who have an adequate life expectancy who are healthy enough to tolerate the therapy long-term, despite their comorbidities. In terms of disease characteristics, it is important to look at rapid doubling time. A prostate-specific antigen [PSA] doubling time of less than 10 months was used in these studies. At a minimum, [patients considered for these treatments] should have that. Other risk characteristics may be important to select patients. Patients with small lymph nodes that are noted on imaging were also eligible for these studies, and these patients should be considered for therapy.
Within the 3 therapies, all of them have a similar mechanism of action with androgen receptor [AR] inhibition. I would slightly lean toward darolutamide because it does not have as much blood-brain barrier penetration, based on preclinical studies. Patients with risk factors, such as seizures or prior strokes, were allowed and eligible on this study, whereas on the other 2 [studies], those patients were excluded. Although I may lean more toward darolutamide, all 3 have similar outcomes in terms of metastasis-free survival.Darolutamide has also now shown OS benefit with further follow-up.
Patients who [would be candidates for this regimen] are those who have not received a previous AR inhibitor. However, that is not so easy in today's world, because now for patients with metastatic hormone-sensitive prostate cancer, it is becoming a growing standard to intensify therapy. On top of androgen deprivation therapy [ADT], if you've already used abiraterone acetate [Zytiga], enzalutamide, or apalutamide, or if you’ve triplet therapy such as ADT plus darolutamide and docetaxel, it becomes hard to justify using [talazoparib plus enzalutamide] in the patients with mCRPC because most of the studies were done with patients who had never been treated with a AR inhibitors.
The phase 3 TALAPRO-2 trial [NCT03395197] did include patients who had prior treatment with docetaxel in the metastatic hormone-sensitive setting, although they were a small proportion of the population enrolled on the study.
I envision [talazoparib plus enzalutamide] being used for the most part in the patients who have progressed with PSA relapse, were treated with ADT, and now have mCRPC. In those patients who also have HRR mutations, I would consider using the combination of talazoparib and enzalutamide. The median PFS was prolonged and statistically significantly improved as compared with enzalutamide alone.
What we don't quite know is if giving enzalutamide followed by talazoparib would have had an equal impact. Do we absolutely need to give them in combination? However, the reason I would favor the combination in a patient like this is that there is preclinical data that shows that the AR inhibitors can sensitize and enhance the efficacy of PARP inhibitors.
Another combination that's approved for BRCA2-mutated disease is abiraterone and olaparib [Lynparza]. However, initially a lot of patients have been pretreated with abiraterone, making it hard to put [this combination] into practical application. What I'm not quite clear on is in patients who are progressing on abiraterone or enzalutamide, should keep that going and add the PARP inhibitor for that synergy? There is no evidence to do that. As of now, in those patients, I would use them in sequence.
In kidney cancer, we are investigating novel agents, and some of them are HIF-2α inhibitors. There are some novel combinations of HIF-2α inhibitors with CDK4/6 inhibitors that are in testing. We are also proposing to investigate neoadjuvant HIF-2α inhibitor therapy [in patients] with high-risk kidney masses. The other big study that's likely to start is the evaluation of a novel CTLA-4 inhibitor that seems to have less toxicity and better efficacy so far in preliminary testing.
In prostate cancer, there is a big push to develop novel pathways. A PI5K pathway has been described by our [research] group. Other novel combinations that will be explored include MYC inhibitors. MYC is a pathway that creates an aggressive clinical prostate cancer course. Thus far MYC has been undruggable, but we are working with a company that is producing a MYC inhibitor that we will have in clinical trial testing.
Moreover, a lot of patients with neuroendocrine features of the prostate are not responding to therapy. We have an ongoing study [the phase 2 PLANE-PC trial (NCT04848337)] with lenvatinib and pembrolizumab, because immune checkpoint therapy is likely to have a big role in efficacy against neuroendocrine. We are evaluating this combination in neuroendocrine prostate cancer. This study is being conducted in multiple sites across the country and the University of Michigan. We are about halfway through our accrual, and we're beginning to see promising efficacy.