Expert insights on selecting the appropriate patients and dose for treatment with the lenvatinib plus pembrolizumab regimen in patients with renal cell carcinoma.
Robert J. Motzer, MD: Now with the different options that you mentioned, David, in terms of your choice for a treatment option for a patient like this favorable-risk patient, what factors go into your decision? Do you have a go-to regimen that you prefer over the others?
David H. Aggen, MD: It’s a wealth of riches that we have 3 therapies where the response rate is between 55% and 70%, and it gets a bit nuanced in selecting these different therapeutics. One way that I think about choosing a TKI/IO [tyrosine kinase inhibitor/immunotherapy] regimen is related to the half-life of the medication. Axitinib has a half-life that’s on the range of about 4 to 6 hours. Lenvatinib has a half-life that’s about a day, and cabozantinib has a half-life that’s about 55 hours. So, if I have a patient who I think has labile hypertension, where I’m going to anticipate I need a dose-hold, having a medication with a shorter half-life with more flexibility and dosing is preferred. I also think that for patients where I anticipate they’re going to have difficulty tolerating the highest dose of the TKI, having flexibility and dosing to go down is beneficial. Particularly with pembrolizumab and axitinib, and pembrolizumab and lenvatinib, you have that. With lenvatinib we generally start at 20 mg, but you can titrate all the way down to 4 mg. Similarly with axitinib, a starting dose is usually 5 mg, and there’s flexibility to go down to 2 mg twice a day or up to 10 mg.
With cabozantinib, the half-life is substantially longer, and for patients who are maybe frailer, I may lean away from cabozantinib for that reason. There are also particular subgroups I think about. Based on data with cabozantinib and monotherapy, we know there’s strong activity with cabozantinib and bone metastases. So for patients with predominant bone metastases, I tend to favor nivolumab and cabozantinib. Although the phase 3 trials weren’t powered to look at those subgroups, it does seem that there’s a benefit to nivolumab and cabozantinib in patients with bone metastases. For patients with liver metastases, I actually tend to favor pembrolizumab and axitinib or pembrolizumab and lenva [lenvatinib] in part because of the half-life issue. If patients develop changes in LFTs [liver function test levels], it’s often difficult to tease out if is this an IO toxicity or is this a TKI toxicity, and having the ability to hold that TKI is very important. I don’t think about these therapies differently in terms of risk groups, and I would feel comfortable offering any of these regimens to people with favorable-, intermediate-, or poor-risk disease. I think with any of these regimens, the median time to response is on the order of 6 to 8 weeks, so if you have a patient who’s symptomatic, all of these are appropriate options.
Robert J. Motzer, MD: That’s great insight, David, in terms of the individualization of the treatment regimen based on the patient’s comorbid conditions and the half-life of the drug, and availability to reverse toxicity based on the half-life. I think that’s really great insight. Bob, one of the questions that comes up is, is there a biomarker that you use to select treatment options for RCC [renal cell carcinoma]? PD-L1 is very important in lung cancer and in some other malignancies. Is PD-L1 expression a factor, or any other biomarkers that you have your sights on, as being particularly promising for selecting treatment for our patients?
Robert S. Alter, MD: Thank you, Bob. This has been a topic that we’ve been discussing at meetings for a long time, and the meetings get longer and longer, and these topics get shorter and shorter. I think that the CheckMate 025 clinical trial did represent to us that patients who are PD-L1 positive or PD-L1 negative still respond to immunotherapy. I think it has proven to us that you can be PD-L1 negative and still receive immunotherapy, so I don’t think that is going to be a good biomarker. Going back several years, Brian has shown that you can use blood pressure as a marker for axitinib, but again, it seems to be isolated to that, yet we still do utilize that in our clinics. I think we’re utilizing next-generation sequencing often, and we do get information like the tumor mutation burden, which I think is still controversial in regard to a prognostic marker for renal cell cancer. MSI, MMR [microsatellite instability and mismatch repair deficiency] are also controversial, and I don’t think we’re using them very well. The loss of PBRM1 [gene expression], it’s negative in localized disease, or it could be positive in advanced disease. I don’t think we can use it clinically yet. We’ve been trying to use the neutrophil-to-lymphocyte ratio, which I think was negative as a biomarker. There happened to have been an article in Translational Oncology looking at TIM3, essentially a T-cell immunoglobulin, and to find the immune checkpoint is helpful in the tumor microenvironment. I think it may be potentially beneficial in regard to looking at efficacy for immunotherapies. But I think right now we don’t really have a good biomarker to gauge our success in how to predict response in our patients with RCC.
Robert J. Motzer, MD: Thanks very much. Mehmet, you chose lenvatinib and pembrolizumab for this patient. Can you tell us what helped you to make that choice over some of the others for this particular patient?
Mehmet A. Bilen, MD: Yes, Dr Motzer. It’s a very important question because all of us ask the same question when we see patients in our clinic on a weekly basis. As my colleagues mentioned earlier, there is no prospective head-to-head comparison nor biomarker. Because of this, we make our selection based on clinical features. In this case, a patient has pancreatic metastases in addition to lung metastases, and we have multiple data suggesting that pancreatic metastasis is more VEGF driven. In this patient, we initially observed, and then all of those lesions were growing and increasing in size, so we decided to start treatment. In addition to that, this patient is IMDC [International Metastatic RCC Database Consortium] favorable risk, and for this reason, an IO/TKI combination I think will be ideal. But we have 3 very good IO/TKIs. The reason we chose pembrolizumab and Lenvima [lenvatinib] is really based on patient preference. For example, this patient is traveling a long distance, and pembrolizumab can be given q6 [every 6] weeks instead of nivolumab, which is q4 [every 4 weeks]. I think this will be easier for the patient in terms of fewer infusion center visits. In addition, this patient preferred daily medication, and you always have a hard time remembering to take medication twice a day or 3 times a day. Lenvima is once a day and so will be easier. Also, Dr Rini mentioned the outcome data, including response rate, the CR [complete response] rate. I think overall, pembrolizumab and lenvatinib is a very appealing combination for this particular patient.
But again, we have other very good options, and there are times I choose nivolumab and cabozantinib. As Dr Aggen mentioned, for a patient with bone metastases, I also favor nivolumab and cabozantinib. If I have a patient with clear cell plus a variant component, like a non-clear cell papillary, I also prefer nivolumab and cabozantinib, since we have multiple data on cabozantinib working in the non-clear cell population in addition to clear cell. Also, there are new data for brain metastases with cabozantinib. So if I have a patient with brain metastases, I think nivolumab and cabozantinib is an appealing choice because immunotherapy tends to penetrate to the brain, so in addition to cabozantinib, to me it might be a helpful combination. But pembrolizumab and axitinib is also very effective if I am worried about the TKI tolerance; I think pembrolizumab and axitinib can be ideal because of its short half-life. I think overall, all 3 combinations are very effective and all of them are approved for any IMDC risk group. Because of this, either one should be a good option. Based on the patient-physician discussion, we finalize the treatment choice and move forward.
Robert J. Motzer, MD: Thanks very much for that discussion. Obviously different factors, including as you mentioned, feasibility of administration, are important in terms of making a choice for patients.
Transcript edited for clarity.