Stan Collender takes a work call while receiving an infusion of the immunotherapy drug pembroluzimab at the Seattle Cancer Care Alliance in April. Collender is taking part in a national trial of the drug for patients with a rare skin cancer. (Fred Hutchinson Cancer Research Center)

A new type of cancer drug designed to unleash the immune system is revolutionizing treatment for advanced melanoma, lung cancer and other malignancies. But some patients who initially respond to the therapy relapse, and researchers are anxious to figure out how and why the delayed resistance occurs.

"Does the immune system stop working, or does the cancer change so that it's no longer responding to the immune system?" said Antoni Ribas, director of the Jonsson Comprehensive Cancer Center Tumor Immunology Program at the University of California at Los Angeles. 

New research by Ribas and others, published online Wednesday in the New England Journal of Medicine, provides some answers. The study, which outlines key mechanisms in how melanoma becomes resistant to immunotherapy, found that genetic changes in tumors allowed them to avoid recognition by the immune system or become less sensitive to its attacks.

"We are convinced that when late relapse occurs, it is the the tumor that has evolved and changed," Ribas said. "Now we have to figure out: What do we do about it?"

The UCLA study involved 78 patients treated with pembrolizumab, the drug given to former president Jimmy Carter. Forty-two of the patients responded but 15 relapsed. Of those, four met the criteria for inclusion in the study.

The researchers, led by Ribas and Jesse Zaretsky, a doctoral student in Ribas's lab, analyzed pairs of tissue samples taken from the individuals before they started treatment and after they relapsed. While the genomes of the post-relapse tumors were very similar to those of their initial tumors, researchers found some significant changes that allowed the cancer to fight back.

In one case, a tumor had lost a gene called B2M, making it harder for T cells, the workhorses of the immune system, to recognize the cancer. In two other cases, genetic mutations in the tumors interfered with genes called JAK1 and JAK2, which limited the immune system's effectiveness in killing cancer cells.

The fourth patient's tissue didn't have those genetic alterations, which indicates other mechanisms of resistance may be discovered in the future, Zaretsky said.

Cancer's ability to mutate and become resistant to treatments isn't new; the disease has long frustrated patients and physicians by rendering chemotherapy and targeted drug therapies impotent. But researchers are now intensely interested in how such resistance develops against the new class of immunotherapy agents called checkpoint inhibitors. These treatments don't attack the tumors themselves but instead spur the body's own immune systems to go after the cancer.

The UCLA researchers acknowledged that the study was small and that tumor resistance requires much more study. By learning more about the mechanisms involved, they hope to figure out how to combine treatments to delay resistance or block it entirely.

"This is part of an evolving story to examine why some patients do not respond to [immunotherapy] drugs, or why some patients respond at the beginning and later relapse," said Suzanne Topalian, director of the melanoma program at Johns Hopkins Kimmel Cancer Center. "Treatment resistance is a key issue now. We really need to understand it to improve the impact of these drugs."

About 30 to 50 percent of patients with advanced melanoma respond to checkpoint inhibitors, depending on whether they previously have been treated for their disease. 

"We are at the start of a new field of study," said Ribas, who was involved in the original trial for pembrolizumab, also known as Keytruda. The drug drew worldwide attention last December when Carter announced that all signs of his advanced melanoma had disappeared. He had been treated with surgery and radiation as well as the drug.  

Keytruda, made by Merck, is similar to another immunotherapy drug, Opdivo, which is manufactured by Bristol-Myers Squibb. Both block a protein called PD-1, which in turn allows the immune system to attack the tumors cancer.  

The UCLA research was partly funded by the National Cancer Institute, Stand Up To Cancer and private philanthropy.

Read more:

Researchers recommend change in prostate-cancer treatment

Sean Parker wants to turn cancer into a chronic illness

Silicon Valley bad boy tries to tame cancer