AACR Annual Meeting 2023: Understanding Immune Suppression in Brain Metastases 

The 2015 approval of an immune checkpoint inhibitor for renal cell carcinoma (RCC) provided a much-needed new treatment option for patients with this disease, which causes almost 14,000 deaths each year, the majority of which occur in patients with distant metastases. 

Alarmingly, brain metastases, which are particularly dangerous, have become more frequent among patients with RCC in recent years, a trend that may be an unfortunate consequence of immunotherapy, according to Elshad Hasanov, MD, PhD, a medical oncology fellow at The University of Texas MD Anderson Cancer Center.  

“The brain is becoming a common metastatic site of disease progression in patients treated with immune checkpoint inhibition, which may be due to the fact that it is an immune-privileged organ and might, therefore, allow tumor cells to escape the activity of immunotherapy,” he explained, noting that recent studies estimate that over a quarter of patients with RCC treated with immune checkpoint inhibitors develop brain metastases, up from 8-15% of patients with RCC in the pre-immunotherapy era. Additional factors underlying the recent increase might include longer patient survival and technological advances that facilitate the detection of brain tumors. 

Patients with brain metastases are less likely to respond to available treatments, possibly due to the unique immune environment of the brain, Hasanov added.  

Elshad Hasanov
Elshad Hasanov,

“To develop better treatments for these patients, we first need to understand the biology of brain metastases and identify the factors that are driving immune suppression in the brain,” he said. Hasanov and colleagues hypothesized that the components of the tumor microenvironment (TME) surrounding brain metastases could have an important role. 

To better understand the role of the TME, he and colleagues used patient samples to compare the TME of brain metastases with those of primary RCC tumors and other sites of metastasis, including the lung, adrenal gland, stomach, and thigh muscle.  

Hasanov reported the findings of their analysis at the AACR Annual Meeting 2023, held April 14-19. 

During the presentation, he explained that he and colleagues employed a combination of single-nucleus RNA sequencing and spatial transcriptomics to characterize gene expression and cell-cell interactions, respectively. Unlike standard methods, single-cell or single-nucleus technologies allow researchers to examine features of cells that might otherwise be masked when examining them together with other cells. Because cells in different parts of a tumor might express different genes, examining them individually can help researchers identify features that might only exist in a few cells. By combining single-nucleus RNA sequencing with spatial transcriptomics, the researchers were also able to identify cell-cell interactions and examine where different cells were situated in relation to one another. 

Using these technologies, the researchers observed that the TME of brain metastases had fewer proliferating immune cells than primary tumors and other metastatic sites. Furthermore, T cells in brain metastases expressed higher levels of immune checkpoint proteins than T cells in other sites, and macrophages in the brain were more likely to express an immune-suppressing M2 gene signature.  

The analysis also revealed that brain metastases had greater infiltration of neuronal and glial cells compared with primary tumors or extracranial metastases. Hasanov explained that interactions between these brain cells and tumor cells were previously shown to enhance cancer growth and metastasis. They found that neuronal and glial cell interaction with immune cells potentially suppressed antitumor immune activity through known immunosuppressive ligand-receptor interactions, adding further evidence of an immunosuppressive TME in the brain. 

In addition, tumor cells in the brain were found to have greater expression of immune-suppressing and tumor-promoting genes. They also had greater expression of genes that allowed them to adapt to the brain environment, including MYC target genes and genes involved in mTORC1 signaling, epithelial-mesenchymal transition, fatty acid metabolism, oxidative phosphorylation, and the response to reactive oxygen species. 

Getty Images

These findings build on prior studies that characterized brain metastases in patients with other cancer types. Spatial and single-cell transcriptomics analyses of brain metastases in patients with melanoma and non-small cell lung cancer found distinct features compared with non-brain metastases. In patients with untreated melanoma, brain metastases were found to have more chromosomal abnormalities, a neuronal-like cell state, and variable expression of metabolic proteins across different regions of the tumor. In patients with non-small cell lung cancer, brain metastases were characterized by immune suppression, specifically reduced antigen presentation and immune cell function and more neutrophils, M2-expressing macrophages, and reactive astrocytes (which have been suggested to drive immune evasion).  

Another study examining brain metastases from patients with immunotherapy-naïve lung, breast, melanoma, and other tumor types, including RCC, found that different T cell populations were spatially separated in the brain TME. Exhausted T cells, such as those expressing high levels of the PD-1 immune checkpoint protein, were located closer to the tumor, whereas memory-like T cells with lower expression of PD-1 were found farther away.  

Together, these studies suggest that brain metastases are characterized by an immunosuppressive TME, regardless of the origin of the primary tumor. Consistent with these findings, a separate analysis found that several tumor-promoting features of brain metastases, including immune suppression, were conserved across tumor types.  

According to Hasanov, the immune-suppressive features he and others identified could serve as biomarkers or therapeutic targets. To explore this potential, he and colleagues plan to initiate a multicenter clinical trial to evaluate lenvatinib (Lenvima), a multitarget inhibitor of several receptors expressed in brain metastases, in combination with the PD-1 inhibitor pembrolizumab (Keytruda) for patients with RCC who have brain metastases. While this regimen is currently approved for the first-line treatment of patients with metastatic RCC, prior trials testing this combination did not include patients with active brain metastases, Hasanov noted.  

“Our study highlights the power of single-nucleus RNA sequencing and spatial transcriptomics to uncover facets of disease biology,” said Hasanov. “We hope that the findings of this study will contribute to the design of better therapies for these patients.”