Potential Immunotherapy Response Markers Emerge in Metastatic Breast Cancer

James M. Reuben, PhD, MBA, a professor in the Department of Hematopathology at The University of Texas MD Anderson Cancer Center

James M. Reuben, PhD, MBA

Data suggest that baseline T-cell clonality and exhaustion markers have significant translational relevance and can serve as prognostic factors of response to pembrolizumab (Keytruda) as maintenance therapy in metastatic breast cancer, explained James M. Reuben, PhD, MBA.

In a phase II trial that was presented at the 2019 San Antonio Breast Cancer Symposium, patients with metastatic inflammatory breast cancer (IBC) and non-IBC triple-negative breast cancer (TNBC) were given maintenance pembrolizumab after achieving a clinical response or stable disease to chemotherapy. At 5 months of follow-up, 7 patients had stable disease (SD) and 8 patients had progressive disease (PD).

CTLA-4 expression in CD4+ T cells at baseline, detected via minimally invasive blood-based analyses of T-cell repertoire and phenotype, was significantly higher in patients with PD than in patients with SD. Meanwhile, patients with a low percentage of CD4+ cells expressing exhaustion markers—including CTLA-4, Tim3, and 2B4—at baseline were more likely to have SD and a longer median progression-free survival (PFS) than patients with PD. Additionally, patients with high clonality and low CD4+ T-exhaustion markers were more likely to have SD and a longer median PFS.

Overall, results showed that T-cell reactivity at baseline as well as a lower percentage of CD4+ T cells expressing CTLA-4/Tim3/2B4 were favorable prognostic factors for maintenance pembrolizumab—demonstrating an intriguing potential use for liquid biopsies in oncology.

“We're still learning how to use [liquid biopsies],” said Reuben. “As we develop more and more of these data, we'll see what's possible and how to use them in the future.”

In an interview with OncLive, Reuben, a professor in the Department of Hematopathology at The University of Texas MD Anderson Cancer Center, discussed his research on using liquid biopsies to identify predictive biomarkers for immunotherapy in patients with metastatic breast cancer.

OncLive: Could you provide an overview of the study you presented?

Reuben: My lab is interested in looking for biomarkers in patients with cancer, specifically those with IBC and those with breast cancer in general. We try to define biomarkers that are relevant to the clinical trial that is being conducted, and we try to tailor [the biomarkers] to the target that we're looking at. In this particular study, we looked at patients with TNBC who have stable disease and were treated with an anti—PD-1 agent to see if we can prolong the responses in a maintenance trial.

We use blood samples because, in the metastatic situation, you're not able to get repeat sampling using tissue. We want to develop areas where we can look at biomarkers in the peripheral blood, which might be useful to us to monitor patients, stratify them, and see if they are benefiting from therapies.

In this trial, we took blood samples before patients received the drug, while they are on therapy at around 4 to 6 cycles in, and when they progress. Since the drug is targeting T cells, we looked at T-cell markers to see the phenotype of these T cells. Typically, you're looking for CD4+ T cells and CD8+ T cells, but there are other protein markers on the cells that are related to cell function. Some cells might have a particular biomarker that [allows them] to respond to an immune response challenge, or there may be cells that are present but are unable to respond. Those are called exhausted T cells.

In the T-cell pool, we looked at the genetic level. What is the repertoire or distribution of clones of T cells that can respond to any endogenous route? We want to look at the types of T cells, whether they are 1 clone or multiple clones, and if they are expanding. Patients with disease progression typically seem to have higher levels of PD-L1 on both CD4+ and CD8+ cells, and it didn't go down with treatment. Additionally, we found that patients who progressed had higher numbers of exhausted T cells. The T cells are present, but they're unable to respond to any kind of a stimulus.

When looking at the clonal level, we found that patients who were benefitting had T cells that are able to expand to a larger repertoire of T cells, meaning they can respond to any kind of a challenge. Most patients who did not respond and had disease progression had a very limited number of T cells.

This gives us an idea that we are able to use a biomarker now in peripheral blood to monitor patients on treatment; it might affect the tissue level responses. We need to validate this hypothesis, and the trial is ongoing. We're going to have an additional 15 patients that we are going to put on trial. We'll be monitoring these patients and hopefully we'll have a comprehensive report by the 2020 ASCO Annual Meeting.

How could these data impact the breast cancer field?

This biomarker gives you another idea of how you can use these tests to stratify patients upfront or to see whether they are actually benefitting from treatment. This is done in peripheral blood, so it's analogous to what you're looking at in circulating tumor DNA (ctDNA). It's all a blood-based assay, which is more appropriate because you're not going to be able to get tissue samples from a patient when they are metastatic. [Tissue samples] are not typically available, and [liquid biopsies] are also less stressful and minimally invasive for the patient.

How are liquid biopsies being used and what are your recommendations for them?

Liquid biopsies are still an exploratory field, but we are learning, every day, how to use them. If certain liquid biopsy assays show high numbers of circulating tumor cells (CTCs), that's not good—and patients usually don't do well with [treatment]. If they have a lower number of CTCs in peripheral blood, then they do well [with treatment]. Looking at the ctDNA, we can see the mutations that are present in the blood and can identify mutations based on that. It's a static response if you look at each time point, but if you look at the cumulative points of these patients, then you can see the dynamic changes that occur either on or off treatment.

What message would you like to convey from these preliminary data?

We need to monitor patients more carefully and know what's going on with them. If we see markers informing us that the patient is not benefitting from these therapies, perhaps we should consider alternative therapies rather than wait for the patient to continue on treatment and potentially [experience] more adverse events.

The trial is ongoing, and we are recruiting more patients. There are more positions available for recruitment; we will design additional studies in the future along these lines.

Gao H, Kida K, Cohen EN, et al. Peripheral T cell clonality and exhaustion as novel biomarkers for anti-PD-1 (pembrolizumab) maintenance therapy in patients with metastatic inflammatory breast cancer (mIBC) and non-IBC triple negative breast cancer (mTNBC). Presented at: 2019 San Antonio Breast Cancer Symposium; December 10-14; San Antonio, TX. Abstract P3-09-12.