Scientists at The Institute of Cancer Research and the Royal Marsden NHS Foundation Trust in the U.K. have found that colorectal cancer (CRC) tumors can switch off expression of a cell surface molecule that is a key target of immunotherapy, which stops the treatment from working. Using miniature laboratory-grown tumors known as patient-derived organoids (PDOs) as models of CRC, the researchers also showed how expression of the immunotherapy target could be reactivated by blocking a particular molecular pathway, and so potenitally resensitize the tumors to treatment.
“Cancer is very good at hiding from the body’s immune system,” commented research head Marco Gerlinger, PhD, team leader in translational oncogenomics at The Institute for Cancer Research. “The latest successful immunotherapies work by acting as a matchmaker to bring the immune system together with cancer, so that it can see it and attack it. Our study has found that bowel cancers have a way of dodging even the newest of immunotherapies—changing their spots by altering the levels of a key molecule on the surface of cells, so that they become harder to recognize.”
Gerlinger and colleagues reported on their studies in the Journal for Immunotherapy of Cancer, in a paper titled, “CEA expression heterogeneity and plasticity confer resistance to the CEA-targeting bispecific immunotherapy antibody cibisatamab (CEA-TCB) in patient-derived colorectal cancer organoids.”
Colorectal cancer is the third most common cause of cancer-related mortality worldwide, the researchers stated, and while anticancer treatments can extend patient lifespan, new treatments are needed to improve outcomes further. Therapy using checkpoint-inhibitor immunotherapies can increase survival in some CRC patients with hypermutated microsatellite instability (MSI) tumors, but are ineffective against the majority of cases, including microsatellite stable (MSS) CRC. “Thus, benefit from checkpoint inhibitors is currently restricted to a small subgroup of CRC patients,” the authors noted.
Cibisatamab (CEA-TCB) is a novel bispecific antibody immunotherapy that effectively brings the cancer cells and cancer-killing T cells of the immune system together. One arm of the antibody binds to carcinoembryonic antigen (CEA), a glycoprotein that is expressed on the cell surface of a wide variety of tumor types. The other antibody arm binds to T cells that kill the cancer cells. Cibisatamab has shown promising results in early clinical trials, but some tumors are resistant to therapy, and studies in cancer cell lines have suggested that the effectiveness of cibisatamab is dependent upon expression of CEA on the tumor cells. “They identified CEA expression as a major determinant of cibisatamab sensitivity as only cell lines expressing moderate to high CEA levels were susceptible to T cell mediated killing,” the researchers commented.
To investigate CRC resistance to cibisatamab sensitivity in a more relevant model than cell lines, the researchers generated mini tumor PDOs from biopsy samples from eight CRC patients, including seven with treatment-resistant metastatic CRC, and another with treatment naïve primary CRC, and co-cultured the organoids together with the CD8 T cells. “We used a new technique for growing miniature replicas of tumors to develop a way of testing whether patients will respond to immunotherapy,” Gerlinger stated.
Initial tests with the new system first confirmed that, “as anticipated,” while organoids expressing high levels of CEA (CEAhi) were highly sensitive to treatment with CD8 T cells and cibisatamab, CEAlo PDOs that expressed only low levels of the glycoprotein were resistant to therapy. They also found that, unlike cancer cell lines, PDOs were capable of exhibiting mixed CEA expression, and these showed partial response to cibisatamab. “Taken together, CEA expression was frequently heterogeneous, showing a bimodal pattern in 50% of PDOs which has not been observed in cancer cell lines, and this phenomenon was associated with poor susceptibility to cibisatamab treatment,” the researchers stated.
Interestingly, when single CEAhi and CEAlo cells were cultured separately, the resulting cell populations again exhibited mixed CEA expression, demonstrating that the cancer cells could switch CEA expression on and off. “This experiment shows that CEA expression is highly plastic in many PDOs which promotes population heterogeneity and cibisatamab resistance … These results are likely highly relevant for CEA targeting immunotherapies as all 4/8 PDOs with mixed CEA expression profiles were at least partially resistant to cibisatamab treatment in vitro.”
Examining RNA expression in individual CEAhi and CEAlo cells indicated that CEA expression was dependent upon WNT/β-catenin signaling, which is genetically activated in the majority of CRCs, the scientists noted. Encouragingly, treating PDO lines that exhibited mixed CEA expression using two different WNT signaling inhibitors led to increased CEA expression and CEAhi cell populations. “We hope that this could in future help immunotherapies work in more patients, by making cancer cells more visible to immune cells,” Gerlinger said.
The authors suggest that the observed CEA plasticity may be susceptible to combination cibisatamab-based therapies that either increase CEA expression in the CEAlo cell population or which co-target another vulnerability of the CEAlo subpopulations. The reported studies also demonstrated that WNT inhibitors might feasibly be used to increase CEA expression and improve the effectiveness of immunotherapies such as cibisatamab.
“Our results provided proof of principle that CEA expression can be pharmacologically perturbed and indicate WNT pathway inhibitors, which are in clinical development, as a potential strategy to boost CEA expression and increase therapeutic benefit from bispecific CEA-targeting antibodies,” the investigators wrote. Determining tumor CEA expression levels may in addition point to predictive biomarkers for CEA-targeting immunotherapies, and so indicate which therapeutic strategies will be effective.
“As we move away from one-size-fits-all therapy for cancer, it’s so important that we are able to identify which patients are most likely to respond to a drug, and do everything we can to avoid resistance to treatment for as long as possible,” noted Paul Workman, FMedSci, FRS, CEO of The Institute of Cancer Research. “This research reveals a way in which cancers are able to hide from a promising new type of immunotherapy. Although the work is still in its early stages, it could be used to develop a test for who is most likely to respond to the drug, and points to possible drug combinations that might prevent or delay resistance.”
The reported studies separately demonstrated the utility of their PDO technology to model CRC tumors and immunotherapy treatment response, the authors stated. “… this study shows the potential of PDO and allogeneic T cell co-cultures as novel tools to dissect mechanisms of immunotherapy resistance in vitro …This addresses a major need in translational immunotherapy research, particularly as studies in immunocompetent animal models are often slow and expensive and sometimes even impossible as animal models are not available for all tumor types, or are incompatible with human-specific reagents, including many monoclonal antibodies.”
“Mini lab-grown tumors have the potential to transform the way we test drugs before clinical trials, commented Andrew Beggs, PhD, FRCS, a Cancer Research UK expert on bowel cancer. “From a tiny biopsy, we can recreate the tumor in the lab to better reflect a patient’s cancer than with traditional ways of growing cells … we can use organoids to learn more about how cancers might respond to drugs, testing many treatments simultaneously to find potential vulnerabilities we might target. Organoid models are increasingly being used in this way to help researchers study, develop and refine possible treatments to test in clinical trials.”