New study offers hope for a rare and devastating eye cancer

After more than a decade of research into a rare eye cancer that causes some of the hardest-to-fight tumors, researchers at the University of Pittsburgh Medical Center have found a treatment that works in some patients and, more importantly, a tool that can predict when is likely to succeed.

The work, published in Nature Communications, is being validated in a clinical trial involving at least 30 patients. It could pave the way for similar methods designed to overcome one of the enduring frustrations of cancer care.

Because tumors differ not only between patients but even within the same patient, a treatment that works on one mass may fail on another, even if both are the same type of cancer.

The researchers in Pittsburgh tackled this problem in uveal melanoma, an eye cancer that affects only 5 in a million people but spreads to other parts of the body, often the liver, half the time. The median survival after uveal melanoma has spread is less than seven months, according to a 2018 study in the journal JAMA Ophthalmology.

“We chose this because it was one of the few cancers for which nothing had been approved 10 years ago when we started,” said Udai Kammula, who led the study and directs the Solid Tumor Cell Therapy Program at UPMC Hillman Cancer Center in Pittsburgh.

Scientists have long speculated that the reason uveal melanoma is so difficult to fight is that something helps the tumor keep out T cells, an important part of the body’s immune system that develops in the bone marrow. However, previous studies by Kammula and his colleagues have shown that uveal melanoma tumors actually contain T cells, and that they are switched on.

The problem? The cells lie dormant instead of multiplying and reaching numbers large enough to overwhelm the tumor.

The culprit appears to be somewhere in the tumor’s ecosystem of cells, molecules and blood vessels, formally known as the tumor’s ‘microenvironment’. Kammula compares this ecosystem to the infrastructure that supports a city. Something in that infrastructure helps protect uveal melanoma tumors by preventing critical T cells from multiplying.

“If we want to eradicate cancer, we ultimately have to get rid of this infrastructure,” Kammula said.

A tool to predict success

He and his colleagues have had some success with a treatment known as adoptive cell therapy, which was developed in the 1980s by Steven Rosenberg of the National Institutes of Health.

The treatment involves removing the T cells from the tumor, where they have not been able to multiply. Scientists then take those T cells and grow them outside the body in a laboratory dish. They treat patients with chemotherapy to kill the last of their old immune systems. Finally, they reintroduce the laboratory-grown T cells into the patient’s bloodstream and the cells, now in much larger numbers, attack the tumor.

In this treatment, the T cells are often called tumor-infiltrating leukocytes or TILs.

Kammula said his team has found that tumors shrink completely or partially in about 35 percent of patients who receive the treatment. But they wanted to know why it doesn’t work in most cases, and whether there might be a way to predict in advance when it will work.

To find out, the researchers analyzed samples from 100 different uveal melanoma tumors that had spread to different parts of the body in 84 patients, in an attempt to examine all the genetic material of the tumors.

“We basically put the tumor biopsy in a blender with the stroma (supporting tissue), the blood vessels, the immune cells and the tumor cells. It had everything,” Kammula said, explaining that they then analyzed all the genetic material from the tumor.

They found 2,394 genes that could have helped make the tumor susceptible to treatment, some of which experts would consider “the usual suspects” and others that were unexpected. Using this long list of genes, the scientists looked for characteristics that they had in common.

The genes were mainly involved in helping the body defend itself against viruses, bacteria and other foreign invaders by removing the invaders and helping the tissue to heal. Kammula and the study’s lead author, Shravan Leonard-Murali, a postdoctoral researcher in the laboratory, used the different activity levels of these genes to develop a clinical tool.

The tool, known as a biomarker, assigns a score to a uveal melanoma tumor based on the likelihood that it will respond well to treatment: removing T cells, growing them outside the body and reinserting them.

So far, Kammula says, the biomarker has been “extremely good” at predicting when the treatment will be effective, although he added, “these findings will need to be confirmed in the current ongoing clinical trial.”

“I thought it was a bit of a tour de force, to be honest,” says Eric Tran, associate member of the Earle A. Chiles Research Institute, a division of the Providence Cancer Institute in Portland, Oregon. Tran did not participate in the study.

He said that while it will be important to validate these results, “I was certainly encouraged by their studies. And from my perspective, I wonder if that kind of strategy can be used for other types of cancer.”

Ryan J. Sullivan, an oncologist at Massachusetts General Hospital and associate professor at Harvard Medical School, who was not involved in the study, called the team’s work “timely” and said, “It is even more important that they (tool) seem to have. that seems to predict which patients will benefit from it.”

The UPMC team is already exploring a possible broader application of both the treatment and the biomarker in a second clinical trial involving a dozen different types of cancer.