Evidence-Based Strategies for Selecting and Sequencing Systemic Therapies in RCC: Highlights From ASCO GU 2022 and Beyond - Episode 9

Is There a Role for Biomarkers in Advanced RCC?

Dr Braun explains how biomarkers help to stratify treatment decision-making for patients with RCC.

Tian Zhang, MD:I’d love to think a little more about not just subsets of patients but also the role for biomarkers. Dr Braun, is there a role for biomarkers in 2022, in terms of thinking about our patients and treatment decision-making? In particular, tell us about your own work on nivolumab-ipilimumab, lenvatinib-pembrolizumab, and other molecular alterations and what we heard about sites of metastasis from GU ASCO [American Society of Clinical Oncology Genitourinary Cancers Symposium].

David Braun, MD, PhD: I’m happy to go through a rundown of some of the molecular analyses that were prevented to GU ASCO. The overall answer is that I hope we get there someday. We’re making progress, but we’re definitely not there in terms of molecular biomarkers helping us choose the right drug for the right patient. What’s needed is more sophisticated analyses—high-dimensional analyses, single-cell RNA sequencing, and spatial technologies. [Also, we need] better approaches to incorporate these data. Not looking at 1 feature at a time, 1 gene, or 1 protein but intubating a lot of things, including clinical factors to create an integrative analysis. Those are going to be the key.

We’re heading in that direction. There are some interesting bullets from work that was presented to GU ASCO, so [I want to] run through some of the larger studies presented. Dr [Marc-Oliver] Grimm and colleagues presented biomarker analysis from the TITAN-RCC trial, which was a trial of up-front nivolumab for clear cell RCC [renal cell carcinoma]. Patients could get up to 2 to 4 doses of a nivolumab-ipilimumab boost if they didn’t have a response to that initial nivolumab. They had peripheral blood mononucleosis collected at various time points, and by conventional flow cytometry they looked at different immune populations circulating within the blood before and during treatment and asked the question whether any were associated with response.

The short answer is there wasn’t a huge signal. There were some hints here and there. There’s an activated CD4 population that might have an association. There were some myeloid populations that seemed to have an association, but these were modest in terms of their effect size. It highlights how difficult this work is and that looking at 1 thing at a time might be challenging. An important thing is the signals that came up. There was a signal within the myeloid compartment, not just the T-cell compartment. We think of these immune checkpoint drugs as T-cell–mediated immunity, but it’s highlighting the role of a myeloid cell, the myeloid component, and how we’re going to address that.

Another molecular analysis by Dr [Chung-Han] Lee at [Memorial] Sloan Kettering [Cancer Center] and colleagues looked at lenvatinib and pembrolizumab. KEYNOTE-146 was a phase 2 study of either treatment-naïve patients or previously treated patients. In this case they looked at whole exome sequencing, a genetic analysis, and transcriptome sequencing from patient tumors who were enrolled in that study. They looked at this preliminary analysis, with a few genes at the genetic-alteration level and a few known signatures for association with response or progression-free survival, and there wasn’t much that they were able to identify. This is a rich data set, so this is probably the first of many analyses to come. I know they’re working hard at it, but at least at this initial stage, this first pass, there wasn’t anything that could help people know if there would be a robust response or if a tumor would be resistant.

Moving from a predictive molecular analysis that wants to try to understand the underlying biology, there’s an interesting presentation from Dr Rana McKay from UCSD [University of California San Diego] that asked a very different question: is there any molecular underpinning for the organotropism of RCC metastases? RCC doesn’t distribute randomly throughout the body. There are sites of the body that it tends to metastasize to more than others, including some of the problematic spots we’ve been talking about—the bone, the brain. We use molecular technologies—whole exome sequencing, RNA sequencing, even immunohistochemistry—to understand the differences among these sites of metastases and understand how they work as a first step toward site-of-metastases selective treatments.

There were some initial signals. There were definitely differences in terms of the number of alterations in classic RCC genes like PBRM1 that may be more enriched in the lung vs other sites. When we look at transcriptional signatures, particularly these [Robert] Motzer–[Brian] Rini clusters that were a positive advance forward in terms of the molecular understanding of kidney cancer, there were different distributions in different sites of disease. It’s early work, but it’s a big step toward understanding the molecular underpinnings of different sites of metastases.

There was a nice study of an older trial, the CALGB 90206 trial, which was back in cytokine era: interferon vs interferon plus bevacizumab. Dr Themecron and colleagues looked back at those pretreatment nephrectomy specimens and targeted PCR [polymerase chain reaction]. Look at the expression of individual genes, in this case HIF pathway genes. It was a smaller analysis. They looked at 28 HIF pathway genes, but there were some associations between expression and prognosis. This is a nice demonstration that we have all these wonderful new technologies, and I’m a big advocate to utilize them. But at the end of the day, coming in with a focused hypothesis that can be tested with a classic technology like PCR can be really helpful. There’s a lot of exciting work being done. We’re not there in terms of molecular biomarker, but we’re certainly walking toward that direction.

Tian Zhang, MD: All helpful insights, David. That’s a super summary and concise for us to grapple our minds around each indicator and biomarker.

This transcript has been edited for clarity.