Ornstein Discusses the Use of Cabozantinib in Clear Cell Renal Cell Carcinoma

During a virtual Targeted Oncology Case-Based Roundtable event, Moshe Ornstein, MD, MA, a Genitourinary medical oncologist, Cleveland Clinic Main Campus, Taussig Cancer Center, discussed the case of a 59-year-old patient with clear cell renal cell carcinoma (ccRCC).

Targeted Oncology^TM: In which trial was this patient enrolled? Can you discuss the design and efficacy for this trial?

ORNSTEIN: This patient went on the CheckMate 9ER trial [NCT03141177] and got cabozantinib/nivolumab as part of the trial. [On January 22, 2021], the FDA approved the combination of cabozantinib/nivolumab based on the phase 3 CheckMate 9ER trial looking at previously untreated metastatic RCC.1 Patients had to have a clear-cell component. The trial looked at patients across IMDC [International Metastatic Renal Cell Carcinoma Database Consortium] risk groups…and compared cabozantinib/nivolumab in combination versus sunitinib [Sutent] as monotherapy. A couple of interesting things: nivolumab on trial was given at the every-2-weeks dose [240 mg] and…cabozantinib was given at the 40-mg dose,2 whereas in clinical practice, as monotherapy in RCC, we’ll start at 60 mg [of cabozantinib], based on clinical trial data. The dosing does come into play when we talk about clinical practice of wanting to give 480 [mg of nivolumab] every 4 weeks.

The primary end point was progression-free survival [PFS]. This is, in some ways, in contrast to other trials—axitinib [Inlyta]/pembrolizumab [Keytruda], axitinib/ avelumab [Bavencio], etc—in which overall survival [OS] was [also a primary end point]. The secondary end points were OS, objective response rate [ORR], and safety.

The Kaplan-Meier curves for the primary end point of PFS, you can drive a small car through, demonstrating sunitinib [Sutent] performed the way one would expect in all-comers [8.3 months].2 Usually PFS is in the 8- to 10-month range, when you include favorable risk all the way to poor-risk disease. Cabozantinib/nivolumab showed impressive activity, with a median PFS of 16.6 months, within the range of other IO/TKI [immunooncology/tyrosine kinase inhibitor] agents, maybe a bit higher in all-comers [HR, 0.51; 95% CI, 0.41-0.64].

PFS by subgroup…[demonstrated] no major surprises. One of the items that’s interesting is PD-L1 expression 1% or more [HR, 0.49; 95% CI, 0.32-0.73] and less than 1% [HR, 0.52; 95% CI, 0.40-0.67], where you saw activity regardless of PD-L1 expression. Maybe PD-L1 expression doesn’t have [much of] a role in RCC. There was improved PFS as you move down the IMDC risk groups [favorable: HR, 0.62; 95% CI, 0.38-1.01; intermediate: HR, 0.54; 95% CI, 0.40-0.72; poor: HR, 0.37; 95% CI, 0.23-0.58], not unlike other trials with similar agents and in similar settings.

The secondary end point was OS; for sunitinib, OS was not reached but the lower end of the confidence interval was met, at 22.6 months. For cabozantinib/nivolumab, OS was not reached, you can’t estimate the 95% CI, and you’re looking at an impressive hazard ratio of 0.60 [98.89% CI, 0.40-0.89],² somewhat in line with other IO-based combinations, which are deemed to hover between the low 0.5s and the mid 0.6s. The P value met the statistical boundaries initially set [P = .0010].

How do these data compare with other trials in this setting? How do you decide which regimen to use? If you look at the hazard ratio of 0.53 in the axitinib/pembrolizumab trial [KEYNOTE-426; NCT02853331], the hazard ratio in CheckMate 9ER is not going to deviate too much; it’s in the same ballpark.³ For a lot of patients, axitinib/pembrolizumab is our go-to regimen, or it’s become that.

The survival data are similar between these trials for that time frame and, if you’re not giving IO/IO, you’re probably giving axitinib/pembrolizumab. The response rate and best overall response in CheckMate 9ER were similar to the KEYNOTE-426 study of axitinib/pembrolizumab: ORR of close to 56%, complete response rate in the range you would expect for cabozantinib/nivolumab, at 8%.²

Progressive disease was seen in 5.6% of patients on cabozantinib/nivolumab. This is one of the challenges of these trials….Cabozantinib is different in some ways from axitinib; we know that based on the half-life, based on the targets, and based on toxicity. The survival data and the response data fall in line. There was nothing shocking that came out of this trial. The question is, what is it going to take for any IO/TKI regimen to knock out the current standard of care? I’m not expecting an IO/TKI regimen to pull people over who use ipilimumab [Yervoy]/nivolumab, but what’s it going to take for people who like to use an IO/TKI versus IO/IO? What do we need to see? Maybe we need to see complete response rates of 15% to 20% or ORRs of 80% or duration of response, and we’ll see that over 3 to 4 years.

I don’t see myself changing practice based on these data, because when you talk about clinical practice, you have to take the patient into consideration. You’re also taking the rest of your team into consideration. We’re focused on genitourinary but, in general, physicians are treating breast cancer, gastrointestinal [GI] malignancies, lung cancer. Why would you consider adding another combination? For the nurses, the advanced practitioners, the other staff, the pharmacy—how much more are you going to throw on when there doesn’t seem to be a massive benefit over what the current standard of care is?

These data don’t necessarily change the standard of care. There are people who believe that cabozantinib—and I don’t see data to support this—might be the superior TKI compared with, say, axitinib. Faced with cabozantinib/nivolumab and axitinib/pembrolizumab, where nivolumab and pembrolizumab are a wash, you’ll go with the one that has the “superior” TKI, because you want to give what you consider your “best shot” up front. That’s more practice specific and less data driven.

What were the toxicities on the CheckMate 9ER study?

We expect with all these IO/TKI regimens that there’s going to be some overlapping toxicity. In patients who had at least 1 dose reduction, in either the cabozantinib or the sunitinib arm, even though cabozantinib…was used at a dose of 40 mg/d, 56% of patients dose reduced [vs 71.3% of sunitinib patients]. This tells us a couple of things. Even though they used a lower dose, cabozantinib/nivolumab was still toxic enough to require a dose reduction yet produced very impressive efficacy.² There is a hypothesis that…what the TKI is doing is manipulating the tumor microenvironment. If you manipulate the tumor microenvironment just enough up front, you might not need to sustain the [higher] dosage of the TKI, so long as those changes are already locked into place.

The way we look at toxicity on clinical trials, just about every patient on any clinical trial has a “some cause” adverse event [AE], and because you don’t know how to attribute every AE, most of them tend to be treatment-related by default. For grade 3 and higher treatment-related AEs, there was 61% for patients on cabozantinib and nivolumab [versus 51% for patients on sunitinib]. The 2 main toxicities that have a significant degree of overlap between nivolumab and cabozantinib—and I’m not referring to hypothyroidism or hypertension, which you can manage with medications— I’m talking about AEs that really can drive a patient nuts and can drive us, as providers, berserk: diarrhea and hepatotoxicity.

Diarrhea grade 3 and above in the combination arm was only 6% [vs 4% with sunitinib] and hepatotoxicity, elevated AST [aspartate aminotransferase; 3% vs < 1% with sunitinib] and ALT [alanine aminotransferase; 5% vs < 1% with sunitinib], were low when you take into consideration that you had 2 hepatotoxic agents in combination. Compare these data—which I know you’re not supposed to do, but we have to—to axitinib and pembrolizumab, where the rates of grade 3 and higher hepatotoxicity were significantly higher compared with cabozantinib/nivolumab.³

How has cabozantinib been used in other ways?

Using cabozantinib as a backbone and a partner for immunotherapy has been demonstrated in other solid tumors as well. In the COSMIC-021 trial [NCT03170960], cabozantinib in combination with atezolizumab [Tecentriq] is being investigated in epithelial carcinoma, in prostate cancer, and in RCC. I think what it tells us more than anything else is that the role of cabozantinib is not just as a monotherapy but, in some ways, as an immune modulator in the tumor microenvironment, to create a more immune-permissive state. In this cohort, there were 10 patients with previously untreated clear-cell RCC who were enrolled in the doseescalation phase. The primary end point was ORR per the investigator.⁴

Compared with the cabozantinib/nivolumab study, CheckMate 9ER, in which the response rate was about 56%,² with cabozantinib/atezolizumab [40 mg and 1200 mg, respectively] it was 53%.⁴ I always look at the percentage of patients whose best response was disease progression. [In COSMIC-021 it was] just 6% of the patients.

These trials tell us the same thing: that this combination of IO/TKI has high activity. Cabozantinib/atezolizumab is being investigated in the CONTACT-03 trial [NCT04338269] in patients with refractory RCC following IO therapy. A patient who got axitinib and pembrolizumab up front can go on a trial of cabozantinib/atezolizumab versus cabozantinib, or patients who got ipilimumab/nivolumab followed by axitinib can also go on that sort of trial.

The thing that surprises me about these data is that, for the last 5 years, we’ve seen, with checkpoint inhibitors in RCC…that PD-L1 inhibitors perform worse than PD-1 inhibitors. I don’t know why that is. I haven’t seen or heard an acceptable explanation for that. The idea that the PD-L1 inhibitors might be effective as well, in combination with a TKI, is encouraging and, hopefully, can provide something beyond another IO/TKI combination.

What do the National Comprehensive Cancer Network (NCCN) guidelines recommend for second-line treatment in RCC?

In the NCCN guidelines, the second-line options for clear-cell RCC are cabozantinib and—because cabozantinib/nivolumab was just approved [and therefore not in the NCCN guidelines yet], let’s say treatment is post axitinib/pembrolizumab—TKI monotherapy, checkpoint inhibitor monotherapy, or combined checkpoint inhibition with ipilimumab and nivolumab.⁵

There are [several other NCCN-recommended] second-line therapy options, including axitinib, lenvatinib [Lenvima] plus everolimus [(Afinitor), axitinib (Inlyta)/pembrolizumab, and axitinib/avelumab]. There has been a lot of discussion about balancing in the second-line setting, especially efficacy versus toxicity.

I hesitate to use the word cure, but we have patients who are on the phase 1 and phase 2 trials of these [regimens] in the frontline setting, and 5, 6, or 7 years out are off of therapy, with minimal disease burden or on a monotherapy and doing great—so in some ways, a clinical cure, if not a radiographic one; a deeper durable remission. When it comes to second-line therapy, the odds of achieving a durable response…are much less likely, and [that’s the] reason toxicity plays an even greater role.

Please describe some of the data behind these second-line regimens.

[In the phase 3 AXIS trial (NCT00678392)], the ORR for axitinib in the second-line setting was 19%; not particularly high, but with an AE rate that led to discontinuation of only 7%, so slightly lower efficacy but higher safety standards.⁶

Conversely, for lenvatinib plus everolimus [Afinitor; Study 205 (NCT01136733)], response rates were incredibly impressive: ORR of 43% in a treatment-refractory setting; PFS of 14.6 months, which is almost in line with the frontline IO/TKI agents; OS in the second-line setting of over 2 years. To achieve that kind of efficacy, though, you have more toxicity; 57% of patients had grade 3 toxicity, and a higher number of patients, almost 1 in 4 patients, discontinued therapy secondary to the toxicity.⁷

What if this patient had received axitinib/pembrolizumab first? Would you have given lenvatinib/everolimus or cabozantinib as second line?

Our patient got cabozantinib and nivolumab, then went on lenvatinib and everolimus.

In patients on axitinib/pembrolizumab, I’d use cabozantinib, because even though lenvatinib/everolimus has a 43% response rate, the toxicity is quite evident as well. Grade 3 or 4 AEs are up to 50% to 60%. If they’re on cabozantinib/nivolumab, why not give them axitinib in second line? I thinkthat’s where cabozantinib as a superior TKI comes into play.

After axitinib/pembrolizumab, choosing cabozantinib certainly makes sense. You’re sacrificing a bit on the efficacy side, but the toxicity, in terms of discontinuation, is not as high as with lenvatinib and everolimus. You don’t have axitinib at that point, but after cabozantinib/nivolumab, you might have axitinib or lenvatinib/everolimus. Both are reasonable options.

There is this idea that hasn’t necessarily panned out but sounds nice scientifically, where you add a different mechanism of action. If somebody had a TKI and an IO, somebody had a VEGFR inhibitor and an IO, then you give them everolimus, which is an mTOR inhibitor, or you add lenvatinib, which has FGFR inhibition. You mix and match mechanisms of action as opposed to going from cabozantinib/nivolumab to axitinib, where all you’ve done is eliminated the checkpoint inhibitor and changed the cabozantinib, which has metalaxyl inhibition, for axitinib, which is a pure VEGFR inhibitor. You might say, “Why would I do that if I can add another mechanism of action?” You’re trying to find that balance of toxicity and efficacy.

What data led to everolimus/lenvatinib becoming approved?

Study 205 had 3 arms, lenvatinib [18 mg]/everolimus [5 mg] versus lenvatinib monotherapy at a higher dose [24 mg] and everolimus [10 mg] as monotherapy.⁷

It is important…to keep in mind that there are overlapping toxicities with lenvatinib and everolimus, much like there are overlapping toxicities with TKIs and IO therapy. In Study 205, there were gastrointestinal toxicities, fatigue, arthralgias, some decreased appetite. Not the easiest regimen in the world to give but fairly effective. One thing I would point out is, with the mTOR pathway, you see more metabolic changes, hypertriglyceridemia, increased cholesterol, etc, so keep that in mind as you’re giving patients mTOR inhibitors.

What would your approach be at this point?

A patient who’s on any combination—lenvatinib with pembrolizumab, should that get approved in RCC, or lenvatinib/everolimus, cabozantinib/nivolumab, axitinib/pembrolizumab, etc; drugs with different mechanisms of action—can have a shared AE. This patient had one of those, grade 1 diarrhea.

For grade 1 diarrhea, on a combination, continue therapy and add antidiarrheal prophylaxis. The idea, when there’s overlapping toxicity, if it’s mild, is you can push through it with symptomatic control. If it’s more toxic, you can stop, provide symptomatic relief, then continue at the same dose or reduce the dose, depending on what the repeated toxicities are.

How do you prevent, mitigate, and manage TKI-associated toxicity?

These combinations are not new, although lenvatinib/everolimus was much less commonly used than it is now. The general approach is to see if you can tease out which drug is causing the toxicity. The half-lives of lenvatinib and everolimus are 28 and 30 hours, respectively,8,9 so you’re going to stop therapy. You’re going to be stopping therapy for a couple of days before you have a sense as to which drug is causing it. If it’s mild, you can work through it; if it’s more drastic, it’s not as clear as with IO/TKI, where you stop the TKI and see if it resolves. Here the management can be a little dicier.

These combinations are not new, although lenvatinib/everolimus was much less commonly used than it is now. The general approach is to see if you can tease out which drug is causing the toxicity. The half-lives of lenvatinib and everolimus are 28 and 30 hours, respectively,8,9 so you’re going to stop therapy. You’re going to be stopping therapy for a couple of days before you have a sense as to which drug is causing it. If it’s mild, you can work through it; if it’s more drastic, it’s not as clear as with IO/TKI, where you stop the TKI and see if it resolves. Here the management can be a little dicier.

We start patients at 18 mg daily of lenvatinib and 5 mg of everolimus. The first dosage reduction of lenvatinib, when they have recurrent toxicities and can no longer tolerate 18 mg even with supportive care, is 14 mg, then 10 mg, then 8 mg.⁸

_References:

_{1. FDA approves nivolumab plus cabozantinib for advanced renal cell carcinoma. FDA. January 22, 2021. Accessed March 18, 2021 https://bit.ly/3lEbuIX}

_{2. Choueiri TK, Powles T, Burotto M, et al. 696O_PR Nivolumab + cabozantinib vs sunitinib in first-line treatment for advanced renal cell carcinoma: first results from the randomized phase III CheckMate 9ER trial. Ann Oncol. 2020;31(suppl 4): S1142-S1215. doi:10.1016/j.annonc.2020.08.2257}

_{3. Rini BI, Plimack ER, Stus V, et al; KEYNOTE-426 Investigators. Pembrolizumab plus axitinib versus sunitinib for advanced renal-cell carcinoma. N Engl J Med. 2019;380(12):1116-1127. doi:10.1056/NEJMoa1816714}

_{4. Pal S, Tsao CK, Suarez C, et al. 702O Cabozantinib (C) in combination with atezolizumab (A) as first-line therapy for advanced clear cell renal cell carcinoma (ccRCC): results from the COSMIC-021 study. Ann Oncol. 2020;31(suppl 4):S554. doi:10.1016/j.annonc.2020.08.774}

_{5. NCCN. Clinical Practice Guidelines in Oncology. Kidney cancer, version 2.2021. Accessed March 18, 2021. https://bit.ly/3lvkDDu}

_{6. Rini BI, Escudier B, Tomczak P, et al. Comparative effectiveness of axitinib versus sorafenib in advanced renal cell carcinoma (AXIS): a randomised phase 3 trial. Lancet. 2011;378(9807):1931-1939. doi:10.1016/S0140-6736(11)61613-9}

_{7. Motzer RJ, Hutson TE, Glen H, et al. Lenvatinib, everolimus, and the combination in patients with metastatic renal cell carcinoma: a randomised, phase 2, openlabel, multicentre trial. Lancet Oncol. 2015;16(15):1473-1482. doi:10.1016/S1470-2045(15)00290-9}

_{8. Lenvima. Prescribing information. Eisai; 2020. Accessed March 18, 2021. https://bit.ly/3cRblOs}

_{9. Afinitor. Prescribing information. Novartis; 2010. Accessed March 28, 2021. https://bit.ly/30VE3rU}