Cancer Cells Go Incognito to Cause Therapy Relapse

Clinicians struggle to treat triple-negative breast cancer (TNBC), one of the most aggressive forms of the disease. While a combination of immunotherapy and chemotherapy is approved to combat metastatic TNBC, this therapy only helps a small portion of patients. Often patients initially see promising results, but then their cancer comes back, possibly due to resistant cells that survived the treatment to regrow tumors.

Despite these challenges, Judith Agudo, an assistant professor at Harvard Medical School, holds hope for cancer immunotherapy’s future. “That's really my overarching dream: to use the immune system to cure cancer. I really believe that we have in our bodies the power to cure cancer if we find a way.”

In a recent study published in Cell, Agudo described her team’s efforts to understand the factors that contribute to breast cancer therapy relapse.¹ The scientists hunted for immunotherapy-resistant cells and sought to understand how they avoided immune cell attack by harnessing the power of Jedi T cells,² which specifically target and destroy cells expressing green fluorescent protein (GFP). By injecting the Jedi cells into mice that express GFP, Agudo visualized and isolated cancer cells that survived the T cell attack.

Using a mouse model of mammary carcinoma, Agudo found that the Jedi cells killed most GFP cancer cells, but some remained clustered within tumors. After dissociating the tumors and performing single cell RNA-sequencing (scRNA-seq) on the GFP positive cells, she found that they had upregulated genes related to chemotherapy resistance, tumor initiation, and hypoxia and downregulated genes involved in proliferation and DNA replication. These quiescent cancer cells (QCCs) had a cellular program that allowed them to remain dormant, avoid the detrimental effects of cancer therapy, and endure in the tumor’s oxygen-poor environment.

To figure out how the QCCs survived the Jedi attack, Agudo and her team developed PADME-seq, a new spatial biology technique that allowed them to study the entire cellular neighborhood in which the QCCs resided. Using mice that express a protein that turns from green to red after being hit by a laser, the researchers precisely marked the QCC cellular niches and cut them out of tumor tissue sections. They isolated the photoconverted cells, which included cancer cells, immune cells, and supportive fibroblasts, and performed scRNA-seq again.

“Tumor resistance happens in specific neighborhoods. So, if you take a tumor and you just dissociate and study all the cells, you lose that resolution. You no longer are able to see that neighborhood. So, for us, the spatial resolution made a whole world of difference,” Agudo said.

“I really believe that we have in our bodies the power to cure cancer if we find a way.”
— Judith Agudo, Harvard Medical School

Agudo found that immune defense pathways were downregulated and collagen-deposition was upregulated in the fibroblasts, which suggested that the QCC neighborhood was not welcoming to immune cells. Additionally, the few T cells that made it into the niche were dysfunctional, which appeared to result from the reduced ability of dendritic cells to recruit and activate them.

“A clear message that comes out of this paper is that if we want to understand why T cells fail—why T cells give up in solid tumors that are tricky—we have to understand what's going on with the rest of the immune system,” said Neville Sanjana, a core faculty member at the New York Genome Center and an assistant professor at New York University, who was not involved in this study. “We like to think of things in isolation—T cells attack tumors; B cells make antibodies—but they really are a coordinated system working together.”

After injecting isolated QCCs into new mice, the research team found that the once dormant cells regrew tumors better than their proliferating counterparts. Agudo doesn’t yet know why QCCs spring into action, but she has a few ideas. “We see that these quiescent cells only exist when there is a big tumor mass,” she said. “When there is no surrounding tumor mass around them, they automatically…start the growth program. It could be that they no longer have the hypoxia signature; it could be other molecular signals that they receive.”

Agudo also assessed cancer cell quiescence in patients receiving TNBC immunotherapy. scRNA-seq results showed that those who responded favorably to the treatment had proliferative rather than quiescent cells in biopsied tissue, which suggested that exploring a patient’s cancer cell status could help predict treatment success or failure.

“Finding that a subpopulation of quiescent cells in tumors is resistant to immunotherapy, now we believe that these are the same cells that also escaped from chemotherapy,” said Agudo. “We have to invest a lot of effort to study them because we need to find some way to kill them.”