Why is the development of vitiligo a predictor of good response in patients taking immunotherapy for melanoma, and what can scientists learn from this?
Todd Greenlee first noticed a lump in his left armpit on Labor Day weekend in 2014. He didn’t think too much of it. Then, a couple of weeks later, he pressed on the lump and a bruise blossomed all the way down his left side. It hurt. Greenlee went to the doctor, and after several tests and appointments, he got the bad news: The lump was stage 4 melanoma that had spread through his lymph nodes and abdomen.
His oncologist, Joseph Clark of Loyola University Medical Center in Chicago, tried immunotherapy treatments: first a monoclonal antibody, then a clinical trial that multiplied Greenlee’s own immune T cells and trained them to fight his particular cancer. Each time, though, Greenlee’s disease would respond for a few months, then relapse. Finally, Clark tried a Hail Mary pass: very high doses of a 20-year-old treatment, one of the first immunotherapies, interleukin-2 (IL2).
Not long after beginning IL2, Greenlee started to notice white patches on his hands. Within a few months, Greenlee, a 43-year-old farm worker, had lost skin pigment over most of his body, a condition known as “vitiligo.”
That didn’t bother Greenlee too much, though, because soon after the vitiligo appeared, the spots of cancer that peppered his abdomen disappeared. More than a year later, his medical team still can find no evidence of disease.
“With regard to immune therapy, if people develop vitiligo while being treated for melanoma, that’s a good sign that treatment is working,” Clark explains.
TRACKING A PATTERN
Doctors had noticed a link between vitiligo and better melanoma outcomes as early as the 1940s, writing case studies about such patients. But no one had been able to determine whether vitiligo just happened to co-occur with good responses to treatment, or if the vitiligo actually contributed to those good responses.
In the last few years, however, a series of studies show that developing vitiligo during or after immunotherapy for melanoma, and sometimes for other skin cancers, does indeed mean that the immune system is in high gear, and is associated with a higher chance of remission or even cure.
While the exact mechanism of this phenomenon has yet to be well understood, experts say that figuring it out may lead to new immunotherapy treatments that can more effectively enlist a patient’s own immune system in fighting their cancer.
To understand how vitiligo might be related to better immune treatments, it’s good to review how the immune system works: All cells have proteins on their surfaces that act like biological bar codes, sort of like identity cards. Immune system cells — T cells, natural killer cells, white blood cells and so on — patrol the body, looking for cells that don’t have the right combination of proteins on their surface. When they find surface proteins, antigens, on cells that don’t pass the screen — that say “not-self” — the immune cells attack and kill the offenders.
An “auto-immune” disease or condition occurs when immune system cells start attacking cells that really aren’t invaders at all, but normal body cells. In the case of vitiligo, the immune cells attack “melanocytes,” the cells responsible for pigment in our eyes, hair and skin. An analysis of studies conducted over many years showed that this happens in only 3.4 percent of patients who receive immunotherapy for melanoma, but other researchers have found a much higher incidence; a 2016 study conducted in France found that 25 percent of 67 participating patients experienced vitiligo when taking the immunotherapy Keytruda (pembrolizumab) to treat their melanoma. This may be due to the fact that melanocytes and melanoma cells may share antigens that are targeted by the immune system when powerful immune-stimulating drugs are used.
When the immune system mistakenly attacks enough pigment cells, their destruction leaves behind white patches which may appear small and limited or may spread all over the body.
Outside the context of cancer, vitiligo tends to appear in those between the ages of 10 and 30. And it appears to help prevent melanoma and other skin cancers in those who develop it spontaneously.
“There are a number of recent genome-wide association studies that reveal that the versions of genes that put you at risk for vitiligo are the same ones that protect you from melanoma,” explains John Harris, director of the Vitiligo Clinic and Research Center at the University of Massachusetts Medical School.
And if the vitiligo develops in response to immunotherapy for melanoma, “Studies have recently shown that those who get vitiligo have a 70 to 80 percent chance of a clinical response” to the treatment, says David Fisher, director of the Melanoma Center at Harvard’s Massachusetts General Hospital. “If they don’t get vitiligo, they have a lower chance of a major response, 20 to 30 percent.” Researchers say that stories of amazing outcomes in melanoma patients who have vitiligo, or develop it during treatment, circulate around clinics. For instance, Harris recently treated a patient who had a softball-sized melanoma in her armpit, but no metastases anywhere. “That’s incredibly unusual,” Harris says. “It’s probably the case that her vitiligo helped to protect her.”
Once it appears, whether spontaneously or because of immunotherapy, vitiligo usually does not go away. Doctors treat mild cases with topical creams such as systemic corticosteroids or topical immunomodulators, which can help with repigmentation. More widespread vitiligo may be treated with phototherapy, several sessions a week in a booth that emits ultraviolet B light, although this is considered a bad idea for people with melanoma, who are much better off avoiding any kind of tanning. If a patient stops these treatments, the vitiligo usually comes back.
Some patients, who are more bothered by the white patches, may choose to graft patches of their normal skin, or blister patches, onto small patches of vitiligo. More experimental treatments involve transplanting melanocytes from normal skin to the patches with vitiligo. These may not be options during treatment, though, because techniques such as micrografting only work on vitiligo that is inactive, points out Caroline Le Poole, who studies vitiligo and melanoma at Loyola University. In any case, she recommends not cosmetically treating vitiligo until there is absolutely no remaining evidence of melanoma tumor.
“Vitiligo is still kind of lumped into the ‘adverse events’ category,” says Mary Jo Turk, an associate professor of microbiology and immunology at Dartmouth College’s Geisel School of Medicine. “But now, people are starting to think, ‘Let’s look at vitiligo as a separate thing in this era of immunotherapy.’”
UNDERSTANDING THE CONNECTION
Generally, the strategies that have driven the field of immunotherapy over the last two decades have been to find therapeutic agents that target specific protein markers, antigens, on the surface of cancer cells. Or, scientists have tried to develop agents that somehow enhance the patient’s own immune system, enlisting it to help fight the cancer. For instance, if there’s normally a biological brake to regulate the immune response, an immunotherapy releases that brake so the immune system can move into motion.
For a long time, immunotherapy developers focused on “self-antigens,” protein markers that occurred on both the cancer cells and their normal counterparts, for instance cancerous melanoma cells and normal melanocytes. By appearing on cancer cells, these antigens make the aberrant cells appear normal, allowing them to hide from the immune system. More recently, the field has been exploring “tumor-specific” antigens, that is, protein markers that occur only in cancer cells.
The study that brought vitiligo into this discussion rose out of happenstance. Turk, the Dartmouth researcher, had a staffer come to her asking what to do about two cages of mice that had been on a shelf for about a year. The mice in one cage had vitiligo; the mice in the other did not. Turk decided to analyze the two populations and was shocked by what she found: The mice with vitiligo had enhanced immune T cells. The other mice did not.
Turk’s group published a 2011 paper using the data from the mice to show that, somehow, the destruction of melanocytes helped not only to keep T cells active, but to remind them to kill particular cells.
“Melanocyte destruction keeps the T cells alive for hundreds of days,” Turk says. “We were shocked to find this. This finding gave rise to other studies seeking to find out exactly what’s going on with melanoma patients who get vitiligo, and why they tend to have better outcomes. For instance, Le Poole has a grant to study autoimmune side effects of anti-tumor therapy and test measures to reverse them.
Le Poole also has a T cell receptor cloned from vitiligo skin that she’s studying to see if these cells can be used to devise new immunotherapies.
But as with so many things in cancer research, it’s not simple. “It’s not a one-to-one relationship,” Le Poole explains. For instance, not all patients who have a good response to immunotherapy will get vitiligo. Likewise, not all those who get vitiligo will go into remission. The relationship is there, but it’s not absolute.
“If you generically stimulate immune responses, one person will be very different from the other,” Le Poole says. Even in animal models, you may think they’re genetically the same, but one will develop vitiligo and one will not. So there must be something environmental, as well.”
Fisher, of Harvard, thinks that, somehow, melanoma may be changing immune cells. “These immune drugs, the immune checkpoint inhibitors, are given to lots of cancer patients. But only the patients with melanoma sometimes develop vitiligo. It’s not like the drugs are injected into the melanoma; they’re systemic. So we suspect that somehow the melanoma is educating the immune systems. It’s not just neo-antigens [produced by the tumor], but also normal melanocyte proteins at work.”
Moving forward, researchers will try to detail the mechanisms that drive auto-immune and tumor cells in vitiligo. In cases where vitiligo signals a good response to immunotherapy, which cells are helping, how and how much? Also, do other immune biomarkers, such as the protein PD-L1, make cancer cells even more susceptible to treatment if the patient has vitiligo? How much does each kind of T cell contribute to the immune response against the cancer?
“If you find T cells in a patient, that doesn’t tell you which T cells are doing the job,” explains Turk. “You could find 20 kinds of T cells. Some recognize melanocytes; some don’t. Was it all 20 [that helped the patient respond to immunotherapy]? Or was it just three? Which ones? You can answer some of these questions in mice, but it’s much more difficult in humans.” In coming years, vitiligo and melanoma researchers say they will do their best to find those answers.