The more genetic mutations in a tumor, the more effective the new immunotherapy will be. New immune system-activating anticancer drugs used in clinical trials of human colon cancer cells have revived many patients with seemingly incurable melanoma or lung cancer, but these drugs do not appear to be effective against colon cancer. One exception, though – a male patient whose metastatic colon cancer tumor disappeared after receiving the drug in 2007 – piqued the researchers’ interest. They suspected that the patient’s recovery might be related to the presence of a large number of mutations in the tumor. Now, a small clinical trial has shown that even certain types of tumors that generally don’t work with new drugs can be treated with such drugs if they happen to have a large number of mutations, and that 3 to 4 percent of patients with these tumors will benefit from them. The drug tested in the above experiment is an antibody that blocks the PD-1 receptor on the surface of T cells of the immune system. Tumor cells evade T-cell attack by activating the PD-1 receptor. However, when the PD-1 inhibitor blocks this immune system “checkpoint,” the T cells are able to recognize and attack the tumor cells. This and other new cancer therapies that use the immune system are promising because they can suppress tumors for long periods of time in some patients with advanced cancers. Melanoma and lung cancer respond best to PD-1 inhibitors, for which the speculation is that these two cancers produce more mutations compared to other cancers. Some of these mutations alter genes to encode abnormal proteins (antigens) that can be recognized as foreign by the immune system. The more mutations, the more such tumor antigens, which can stimulate aggressive behavior in T cells released by PD-1 inhibitors. Johns Hopkins University researchers Hopkins researchers examined the tumor tissue of the first male colon cancer patient to respond to a PD-1 inhibitor and found a clue that his tumor had mutations in genes that “mismatch repair,” which encode proteins that function to repair DNA base errors during cellular gene duplication. If these genes were not expressed properly, they would be replaced by a protein. If these genes are not expressed properly, they could allow cancer-causing mutations to appear and eventually lead to colon cancer, which would contain at least 1,000 mutations in the tumor, 10 to 100 times more than the average tissue. The Hopkins team wanted to know if patients with a variety of other cancers that had mutations in mismatch repair genes in their tumors also responded to PD-1 inhibitors. To explore this idea, Hopkins oncologists Dung Le, Luis Diaz and others looked for tumor samples taken from patients with advanced cancers for whom other therapies had been ineffective. The researchers divided the 48 patients into 2 groups according to whether they contained mismatch repair gene mutations. The PD-1 inhibitor pembrolizumab (Keytruda) was given to all patients at 2-week intervals. The results were very different in the 2 groups. Individuals with mismatch repair mutations were more likely to respond – of the 13 colon cancer patients, 8 had shrinking tumor tissue, 4 remained stable, and only 1 had worsening disease. In contrast, none of the 25 colon cancer patients without the mismatch repair mutation responded to the drug. Some patients who did respond survived for a year or more, while those who did not respond lived an average of 7.6 months. Of the 10 patients with other types of tumors (including pancreatic, prostate, and uterine cancers) that contained the mismatch repair mutation, seven patients had improved or stable disease, and three others had progressed. The study was published in the New England Journal of Medicine, and Le presented the latest results at the recent annual meeting of the American Society of Clinical Oncology. According to Diaz, the results show that cancer patients with mismatch repair gene defects make up 3 to 4 percent of all cancer patients (the PD-1 inhibitor effective population). “That’s a small percentage and not appropriate for all cancers.” Still, it could extend the life of 30,000 to 40,000 advanced cancer patients in the United States by one year, he said. This study also supports another claim that the more genetic mutations in a tumor – whether due to problems with mismatch repair genes or other reasons – the more likely a PD-1 inhibitor or similar drug will be effective against it, says Memorial Sloan-Kettering Cancer Center ( says Jedd Wolchok, a cancer immunotherapy researcher at Memorial Sloan-Kettering Cancer Center. His team recently reported that the more mutations encoding new tumor antigens, the more likely patients with melanoma and lung cancer were to respond to immune “checkpoint” blockers. The Hopkins study “gave us a lot of confidence that our previous findings were important,” he said. One corollary to the finding is that cancer patients with fewer mutations in their tumors respond better to PD-1 inhibitors if they receive radiation or chemotherapy first, because radiation and chemotherapy can cause new mutations in the tumor. Although some patients have tried this approach in clinical trials with PD-1 inhibitors, no investigators have yet designed a clinical trial specifically to study the issue, “so it’s hard to draw conclusions on that,” Diaz said.