Revealing the Mechanisms of CAR T-Cell Therapy Resistance in Solid Tumors

Physician-scientists at Massachusetts General Hospital recently published groundbreaking results in Cancer Immunology Research revealing how some cancers acquire and maintain therapeutic resistance to chimeric antigen receptor (CAR) T-cell therapy. Their use of ex vivo models to simulate patient-specific tumor physiology provides a roadmap for understanding and addressing resistance to cancer immunotherapy.

Subscribe to the latest updates from Cancer Center Advances in Motion

Error: Please enter a valid email address.

Email Address Submit

reCAPTCHA helps prevent automated form spam.

The submit button will be disabled until you complete the CAPTCHA.

"CAR T-cell therapy has proven effective at treating various blood cancers and certain types of metastatic melanomas previously considered incurable," says Russell W. Jenkins, MD, PhD, investigator in the Krantz Family Center for Cancer Research. "We want to expand the success of CAR T-cell therapy to target solid tumors with a track record of being unresponsive to even the most advanced forms of therapy."

Finding the Devil in the Details

Cancer immunotherapy's effectiveness relies on harnessing a patient's immune system to selectively target cancer cells. Although success depends on multiple factors, two are particularly critical in the immune cells enlisted for this treatment:

• An ability to distinguish cancer cells from healthy cells
• The capacity for sustained activation and proliferation to maximize their effectiveness

Two immunotherapeutic approaches involve harvesting a patient's T cells, optimizing their ability to target and destroy cancer cells, and reinfusing them into the patient. These include:

• Tumor-infiltrating lymphocyte (TIL) therapy: Based on their previous recognition of tumor-specific markers, TILs navigate to the tumor and begin killing cancer cells.
• Chimeric antigen receptor (CAR) T-cell therapy: Cells are genetically engineered to present one or more surface markers that allow them to identify and kill cancer cells selectively.

Prior to reinfusion, the numbers of cells are exponentially expanded to levels that will maximize their therapeutic effect. Although both techniques have proven effective in multiple cancers, success in certain solid tumors, such as those associated with pancreatic, bile duct, or liver cancer, has been elusive. Evidence suggests that in solid tumors, cancer cells rapidly evolve to recognize the immune response initiated by these therapies and fight back.

"In many cases, immunotherapy triggers mechanisms within cancer cells that allow them to adapt, evade, and survive," explains Dr. Jenkins. These adaptions occur in both cancer and non-cancer cells throughout the tumor microenvironment (TME), resulting in an ecosystem biased toward tumor survival. "In many solid tumors, treatment drives cancer cells to dynamically alter their programming and shape the TME into a therapeutic barrier."

These changes also extend to TILs and CAR T-cells delivered to the TME.

"Soon after interacting with the tumor, these cells appear exhausted and dysfunctional, effectively eliminating their therapeutic efficacy," Dr. Jenkins explains.

This suggests the capability of these tumors to alter not only their own phenotype but also that of the cells sent to destroy them. Identifying the factors responsible for these changes would provide critical insight into how these tumors acquire and maintain therapeutic resistance.

Capitalizing on an Authentic Model of Human Cancer

The previous decade has seen increased interest in non-animal alternatives that allow a more physiologically relevant approach to human-targeted research. Among these alternatives, spheroids and organoids represent three-dimensional (3D) cultures of cell aggregates capable of replicating the complexity of biological systems.

In 2015, Dr. Jenkins and his colleagues began developing patient-derived organotypic tumor spheroids (PDOTS) using tumor biopsies. Tumor spheroids can be created using only cancer cells, but PDOTS also includes a patient's immune cells. As the PDOTS grow, the resulting 3D tumor spheroids accurately mimic the patient's TME and its specific cell populations.

"TMEs are both tumor- and patient-specific," he explains. Creating models that accurately mimic that physiology provides invaluable research platforms. "The ability to break down complex systems into key components is critical to learning how they work, including why a therapy might work in one patient and not another."

Research on TILs as part of a 2023 Krantz Family Center for Cancer Research Breakthrough Award provided both confidence in the models' effectiveness and insights into promising targets possibly involved in treatment resistance. Discussions with Mass General colleagues also introduced Dr. Jenkins to a surface marker called B7-H3, which is present almost exclusively on cancer cells within solid tumors.

Having the right model to test multiple hypotheses at once sets the stage for finding answers to a number of very difficult questions.

Breaking New Ground

For this study, Dr. Jenkins' collaborators engineered CAR T-cellscapable of identifying cancer cells presenting the B7-H3 antigen (B7-H3.CAR-T). After confirming their selectivity for and effectiveness against tumor spheroids, they evaluated them in PDOTS of multiple solid tumor types.

Incubation with B7-H3.CAR T-cells reduced PDOTS viability in 60% of samples. They then performed combined treatment with B7-H3.CAR T-cells and an anti-PD-1 antibody to determine whether the PDOTS had used an immune checkpoint to limit treatment efficacy.

The observed 100% response rate suggested two things:

• The PDOTS had somehow induced surface expression of an immune checkpoint receptor (PD-1) on the B7-H3.CAR T-cells.
• The presence of this receptor resulted in B7-H3.CAR T-cellexhaustion and dysfunction.

Numerous studies confirm that cancer cells alter their transcription profile and activate signaling pathways to escape the immune response. Members of Dr. Jenkins' lab previously described TANK-binding kinase 1 (TBK1) involvement in immune evasion by cancer cells.

The present study targeted TBK1 to determine its capacity to enhance CAR T-celleffectiveness against solid tumors. Their findings were:

• 100% response rate: Co-treatment of PDOTS with B7-H3.CAR T and a TBK1 inhibitor (TBK1i) led to all PDOTS showing significantly increased response and decreased viability.
• Decreased CAR T-cell exhaustion: Combined B7-H3.CAR T/TBK1i treatment prevented surface expression of the PD-1 receptor, as well as other markers of T-cell dysfunction.
• Increased CAR T-cell activity: Either TBK1 deletion in B7-H3.CAR T-cells or co-treatment with TBK1i increased B7-H3.CAR T-cell proliferation and cytokine production and release.
• Reduced capacity for tumor adaption: Either TBK1 deletion in cancer cells or co-treatment with TBK1i increased tumor susceptibility to B7-H3.CAR T-mediated killing.

Overall, the results implicated TBK1 in multiple pathways related to the resistance of solid tumors to CAR T-celltherapy. Using single-cell RNA sequencing (scRNA-seq), they also revealed changes in the transcription profiles of tumor cells within 24 hours of CAR T-cellinteraction. This implied that TME reprogramming for tumor survival occurs very early in response to treatment.

Dr. Jenkins admits that the findings' significance prompted them to continually increase the proof threshold. "We wanted to remove any bias toward TBK1 by setting higher standards for reproducibility of the results in each experiment," he says. For each iteration, the results consistently met or exceeded their expectations, ultimately making it difficult to argue with 100% response rates.

Although the road to approving TBK1 as a possible target for clinical applications is long, these data provide a strong first step in that direction. "It was gratifying to realize that this pathway we were chasing appeared to be as promising as we believed it might be."

The Synergy of Great Minds

Groundbreaking discoveries are often simply a matter of proximity and timing. Dr. Jenkins says that despite Mass General's substantial size and opportunities for collaboration, it's interesting that the study's genesis and completion occurred in one relatively small area on one floor in one building.

"The origin story for this effort began with my office being a few steps away from that of Soldano Ferrone, MD, PhD, resulting in our sharing daily coffee-making rituals." The late Dr. Ferrone, a translational tumor immunologist aware of B7-H3 as a potential therapeutic target, would offer gentle hints about collaborating to evaluate hisCAR T-cells in Dr. Jenkins' PDOTS.

Dr. Ferrone's daughter, Cristina Ferrone, MD, who was a hepatobiliary surgeon at Mass General at the time, engaged in helping translate CAR T-celltherapy's success to untreatable solid tumors. Now the chair of the Department of Surgery at Cedars-Sinai, her continued support was instrumental in the study's completion. "This study would have had neither a beginning nor a successful ending without their inspiration and collaboration," says Dr. Jenkins.

Another of Dr. Jenkins' neighbors, Moshe Sade-Feldman, PhD, investigator in the Krantz Family Center for Cancer Research, is a co-recipient of the 2023 Krantz Breakthrough Award with Dr. Jenkins and Genevieve M. Boland, MD, PhD, vice chair of Research for Mass General's Department of Surgery. In this study, Dr. Sade-Feldman was instrumental in performing the scRNA-seq experiments that revealed changes in the TME driven by the tumor's response to therapy.

Dr. Jenkins says that the excitement of the study's results also highlights the amount of effort still required.

"It's clear that fighting cancer requires us to have a solid understanding of how it fights back," he says. "Our findings emphasize that the models we use to learn these things matter and will be critical to our creating faster, stronger and more effective therapies."

Learn more about CAR T-cell therapy

Learn more about the Krantz Family Center for Cancer Research