New research from the University of Virginia Cancer Center (UVA) could save once promising immunotherapies for the treatment of solid cancer tumors such as ovarian, colon and triple negative breast cancers that ultimately failed in human clinical trials.
Research, led by Jogender Tushir-Singh, PhD at UVA, explains why the antibody approaches were effective in killing cancer tumors in laboratory tests but were ineffective in humans. He found that the approaches had an unintended effect on the human immune system, potentially compromising the immune response they were trying to improve.
The new findings enabled Tushir-Singh of the Department of Biochemistry and Molecular Genetics at the UVA School of Medicine to significantly increase the effectiveness of the approaches in laboratory models, reduce tumor size and improve overall survival. The promising results indicate the renewed potential of the strategies in human patients, he and his team report. The results also have significant potential to further improve the clinical efficacy of PD-L1 antibodies, which have already been approved by the U.S. Food and Drug Administration for use in solid tumors, particularly those approved for fatal triple negative breast cancer .
Immunotherapy aims to use the body’s immune system to recognize and destroy cancer cells. Lab-made antibodies remain the primary mediator of immunotherapies and CAR-T-cell therapies, which have caused a tremendous stir over the past decade. However, these therapies have been shown to be less effective against solid tumors than against melanoma (skin cancer) and leukemia (cancer of the blood). One major obstacle: it is difficult for immune cells to get into the core of solid tumors efficiently.
To solve this problem, scientists developed an approach that selectively uses antibodies to target a receptor on the surface of cancer cells called death receptor-5 (DR5). This approach essentially instructs cancer cells to die and improves the permeation of the body’s immune cells into a solid tumor without the toxicity associated with chemotherapy. However, when tested in human Phase II clinical trials, these antibodies were consistently unable to improve patient survival.
Tushir-Singh, an antibody engineer, and his co-workers found that the anti-DR5 antibody was inadvertently approaching biological processes that suppress the body’s immune response. This enabled the cancer tumors to evade the immune system and continue to grow.
Tushir-Singh and his team found that it was possible to restore the effectiveness of the DR5-based antibody approach in human cancer cells and immune-safe mouse models by working together to address the negative biological processes with improved, immune-activating therapy. The new combination therapy “significantly” increased the effectiveness of cancer killer immune cells known as T cells, shrunk tumors, and improved survival in laboratory mice, they reported in the journal EMBO Molecular Medicine.
“We would like to see these strategies in clinical trials that we strongly believe have great potential in solid tumors,” said Tushir-Singh.
Meanwhile, researchers reported elsewhere in Nature communication that they had found another possible explanation for why many cancer drugs that kill tumor cells in mouse models don’t work in human studies.
In a study at the University of Texas Health Science Center at the Houston School of Biomedical Informatics and McGovern Medical School, researchers found that mouse viruses are widespread in patient-derived xenografts (PDX). PDX models are developed by implanting human tumor tissue into immunodeficient mice and are widely used for testing and developing cancer drugs.
“We found that when a human tumor was placed in a mouse, that tumor was not the same as the cancer patient’s tumor,” said Dr. W. Jim Zheng, professor at the School of Biomedical Informatics and senior author of the study. “Most of the tumors we tested were infected by mouse viruses. … It makes the results of a cancer drug look promising if you think the drug will kill the tumor – but in reality it won’t work the way it does in human studies as it does in mice.
Edited by Gary Cramer