New in Renal Cell Carcinoma: Data From ESMO Virtual Congress 2020

New in Renal Cell Carcinoma: Data From ESMO Virtual Congress 2020

Question

In CheckMate-9ER, nivolumab, a checkpoint inhibitor, plus cabozantinib, a tyrosine kinase inhibitor (TKI), performed better than comparator sunitinib in the first-line setting. Do you think this is because the 2 inhibitors work better together against metastatic renal cell carcinoma (RCC), or to you, does it say more about sunitinib not being a good standard of care?

Answer

It could be both. Previous studies have shown that combination immunotherapy — either immunotherapy plus immunotherapy or immunotherapy plus a TKI — has shown to be superior to sunitinib in the past. So far, though, all the trials showing significant improvement were in intermediate- and poor-risk patients. At the recent [ASCO20 Virtual Scientific Program], KEYNOTE-426 showed that sunitinib was better than axitinib plus pembrolizumab in favorable-risk patients.¹

CheckMate-9ER was testing sunitinib vs cabozantinib plus nivolumab.² Cabozantinib is thought to be a very strong TKI and it has significant activity in the second-line setting after failing on a TKI. In combination with nivolumab, it has been shown to be very strong in earlier studies. The different targets that cabozantinib has in addition to traditional TKI targets makes it probably a more powerful combination with a checkpoint inhibitor.

I don’t know if the overall response rate has anything to do with patient selection or the percentage of patients with favorable-risk disease. We will wait for the breakdown analyses and make more sense out of it with coming data.

Question

It is hypothesized that circulating tumor DNA (ctDNA) analysis might be useful for revealing potentially actionable mutations in metastatic RCC. Do you think the best way to interrogate for mutations would be the use of a targeted next-generation sequencing assay, or through whole-genome sequencing instead?

Answer

We are still in the discovery phase with most of these cancers. There are certain identified mutations from other cancers that may or may not be applicable to renal cell carcinoma.

If I am trying to discover and have an open mind, I would use whole-genome sequencing approaches and then refine it from there rather than focusing on a narrow targeted analyses. Kidney cancer is very heterogeneous in terms of mutations. If you analyze different regions of the same tumor you may find that the genetic makeup is different. I would cast a wider net.

Question

In patients with metastatic clear cell renal cell carcinoma (ccRCC) receiving nivolumab, certain immune gene-expression signatures in tumors were associated with responses to programmed cell death 1 (PD-1) inhibition and good outcomes.³ Do you think this is evidence that immune-high, angiogenesis-low, and stromal-low signatures will be emerging biomarkers within the field of kidney cancer?

Answer

These studies have shown that the [vascular endothelial growth factor; VEGF]-based or immune-high or angiogenesis-high profiles are shown from one time point samples. We know that these tumors are dynamic and change and mutate. Thus, the tumor profile and phenotype will change.

It is probably important to continue prospective studies of these tumors to see if they are evolving. At some time point, is the tumor reversing back to more immune signature or more VEGF signature?

We know that in the clinic some patients respond to rechallenge with the same type of treatment that they have had in the past. At ASCO this year there were 2 trials, which demonstrated that rechallenge with ipilimumab plus nivolumab can produce up to 15% to 20% response rates in patients who received nivolumab in the past.^4,5 This means that they had formed resistance, but later they responded.

We are not measuring any of those. That is probably a good start but we still have a long way to go in identifying what is the best treatment at a given time rather than using a one–time point sample to predict all future treatment options. If someone has an immune-high signature in the beginning and goes on to have a complete response to immunotherapy, that is good. But if they did not have a complete response, they may still have the opportunity to do that in the future. The disease keeps changing. The tumor is very smart and we need to plan ahead. These studies are great and provide some clues as to how we should think for the future.

Question

Similarly, in the BIONIKK trial, certain signatures were found to be linked to good outcomes in naive metastatic clear cell renal cell carcinoma (m-ccRCC), and revealed the best candidates to receive a TKI, nivolumab alone, or nivolumab-ipilimumab.⁶ Do you think these kinds of transcriptomic analyses might have the same utility in other kidney cancer subtypes?

Answer

It is possible. Generally speaking most histologies — other than clear cell — do not respond to these treatments, including immunotherapy and VEGF-targeted treatment.

It is important to apply this knowledge and find out if non–clear cell histologies can benefit from these treatments. It is hard to do a large clinical trial in these less-common histologies. Rather than doing a 500-patient trial for papillary kidney cancer, which will take 10 years or more to do, we should apply these biomarker-based approaches and identify who might benefit.

Question

Results from COSMIC-021 in metastatic or advanced ccRCC found that higher levels of CD8-positive T cells in the tumor were significantly associated with overall response.⁷ This suggests that immune cell population may contribute to tumor susceptibility to a study treatment. How do you think studies of immune cell populations could drive treatment selection?

Answer

For a tumor to respond to immunotherapy the tumor should have certain factors or triggers for the immune system to identify and target. In addition, the immune system should be primed in order to work against the cancer, such as through inhibition of CTLA-4 and the PD-1/PD-L1 pathway. These treatments release the T cells from the anergic state and activate the T cells. Of course there should be sufficient T cells, and immune milieu should be primed to have an immune response. The combination of tumor factors, immune milieu, and activation of the immune system itself — all these put together should lead to a response.

We talked about genomic analyses, mutations, PD-L1 status, [microsatellite instability]-high, and other factors that may help make these treatments be more effective. Of course, the T cells have to be primed to do [their] job. Some of the drugs can do some of those activations. Then both (tumor factors and T-cell activation) come together and produce a response. If any of these are lacking, there won’t be a response.