At the 2019 American Society of Clinical Oncology Annual Meeting, Gianpietro Dotti, MD, Professor, Immunology, University of North Carolina Lineberger Cancer Center, Chapel Hill, discussed ongoing efforts to extend the application of CAR T-cell therapy to solid tumors, such as ovarian cancer.
“The challenge for the next 5 to 10 years is to improve existing CAR T-cell therapy and to transfer it to solid tumors,” said Dr Dotti. “There are multiple problems that need to be overcome, however, including selection of the right antigen and T-cell migration/infiltration to the solid tumor, which is completely different than in hematologic diseases. The tumor environment in solid tumors is also highly immunosuppressive.”
The good news is that the identification of antigens in solid tumors is currently a very popular field, with numerous phase 1 and phase 2 clinical trials exploring a significant number of targets. According to Dr Dotti, investigators are searching for an antigen that is not expressed (or has limited expression) in normal tissues to avoid organ toxicity.
“We are looking for targets that are both highly expressed in tumors and homogeneously expressed in all tumor cells in the tumor,” he explained. “Ideally, the antigen should be critical for the tumor growth and metastasis to prevent antigen escape.”
B7-H3: The Next Antigen
Over the past 2 to 3 years, Dr Dotti and colleagues have been focused on B7-H3 in solid tumors. Like cluster of differentiation (CD)80, CD86, inducible costimulator ligand, and PD-L1, B7-H3 is a member of the B7 family, but the ligand is unknown although it is assumed to be an inhibitory receptor.
“B7-H3 has limited expression in normal tissues and is very highly expressed in many solid tumors, including pancreatic cancer, ovarian cancer, colon, prostate, and many, many others,” said Dr Dotti, who noted that the density of the antigen’s expression on target cells is critical to dictate the antitumor activity of the CAR T-cells.
“Based on the expression of this target on the tumor cells, it seems that we have a therapeutic window to distinguish between tumor cells and normal tissues,” he said.
So far, Dr Dotti and colleagues have observed antitumor activity in a pancreatic model without associated toxicity in immunocompetent mice, but the true test will be in human patients. The investigators are hoping to open a clinical study in ovarian cancer by the end of 2019.
Natural Killer T-Cells
Infiltration of CAR T-cells in solid tumors is another issue concerning antitumor activity. As Dr Dotti reported, natural killer (NK) T-cells represent an alternative platform for CAR T-cells for CAR engraftment and may result in better trafficking to the tumor site. In fact, infiltration of NK T-cells has already been shown to correlate with better survival in pediatric patients with stage IV neuroblastoma. Dr Dotti and colleagues have engineered NK T-cells with CARs to be tumor-specific by retaining the native capacity of these cells to recognize CD1 deposit targets, which are usually macrophages associated with the tumor.
“Although we are still at dose level 1, this approach has demonstrated remarkable antitumor activity in one of the first patients treated, completely eliminating a lesion in the sternum,” said Dr Dotti.
Overcoming Tumor Immunosuppression
Finally, for CAR T-cell therapy to be successful in solid tumors, immunosuppression in the tumor microenvironment must be overcome. Although researchers are exploring combination therapy with checkpoint inhibitors or agents that are targeting tumor-associated macrophages and myeloid-derived suppressive cells, they are also designing CARs to directly overcome immunosuppression themselves. As Dr Dotti reported, interleukin (IL)-15 expressed by CAR T-cells may improve persistence and antitumor activity of CAR T-cells in several subsets of solid tumors.
“We focused for many years on IL-15 as a way to improve the activity of CAR T-cells in a suppressive environment,” he said. “In a mouse model of very aggressive neuroblastoma, CARs engineered with IL-15 demonstrated a dramatic antitumor response.”
“Interestingly, expression of PD-1 of these CAR T-cells was also reduced within the tumor microenvironment, which suggests that these cells can be less susceptible to PD-L1 inhibition in the context of in immunosuppressive environment,” Dr Dotti added.
Based on these data, Dr Dotti and colleagues are currently enrolling pediatric patients with neuroblastoma for a clinical study at the University of North Carolina.