The first two rounds of treatment went off without a hitch. But last November, after receiving a third dose of potent immunotherapy for his skin cancer, Rich Lenihan started to feel tired and weak.

He was urinating constantly, and no amount of water could abate his thirst. A blood test revealed glucose levels that were through the roof.

Lenihan, at age 62, had developed a disease akin to type 1 diabetes — formerly called “juvenile” diabetes — a rare complication of drugs known as checkpoint inhibitors that rev up the body’s immune assault on tumor tissue.

advertisement

“The good news is we know what’s wrong with you,” Lenihan recalled his doctors telling him — as they whisked him off to the ICU.

The bad news: He would need to control his glucose with insulin shots for the rest of his life.

Roughly 1% of patients receiving immunotherapy drugs experience the same irreversible side effect. Making matters worse, oncologists have little clue why.

“There are a lot of case reports,” said Dr. Kevan Herold, a clinical immunologist at the Yale School of Medicine who last year published a case series of his own. “But there isn’t a good understanding yet of who’s most likely to develop these adverse events from checkpoint inhibitors.”

On Wednesday, the Parker Institute for Cancer Immunotherapy, JDRF (formerly called the Juvenile Diabetes Research Foundation), and the Helmsley Charitable Trust announced they are joining forces to launch a $10 million, three-year research initiative designed to identify the root causes of drug-induced diabetes among cancer patients.

Eventually, researchers involved in the project hope to discover new therapeutic strategies for preventing diabetic complications without sacrificing any of the cancer-destroying potential of livesaving immunotherapy drugs.

“But the first step,” said Dr. Zoe Quandt, an endocrinologist at the University of California, San Francisco, “is understanding the immune responses of patients so we can consider which drugs would be good ones to try.”

‘A lot to handle’

Autoimmune-related toxicity problems have dogged checkpoint inhibitors since their earliest days of clinical testing — although generally patients see them as a small price to pay for a treatment that can beat metastatic cancer back into remission.

Participants in some of the first trials for Yervoy, a drug that blocks an immune-cell brake called CTLA-4, developed inflammatory diseases of the pituitary gland, thyroid, and other organs. More cases soon followed among patients given the next generation of agents, drugs such as Opdivo and Keytruda that target PD-1, another brake pedal found on T cells.

Those complications, if caught quickly, rarely prove debilitating for all that long; once- or twice-daily hormone replacement therapy can usually make up for the lost organ function.

Insulin-dependent diabetes is a different story. It’s more of a daily slog, with patients having to carefully and constantly track their blood sugar levels and self-inject different types of insulin at different times of the day — or else risk falling into a diabetic coma or having one’s blood turn into a toxic, acidic stew.

“Type 1 diabetes is like being in hell, man. It’s horrible,” said Jaime Vidal, 79, a retired mailman from San Bruno, Calif., who developed the disease two months ago after receiving Opdivo through a trial protocol for people with operable cancer of the esophagus. “If I knew then, when I opted for the clinical trial, what type 1 diabetes entailed, I would never have gone for the immunotherapy. Never. I would have taken the chance of the cancer coming back.”

The condition can also come on quickly. Blood sugar levels spike precipitously. Many patients land in intensive care. Once discharged, they face a particularly hard-to-control form of the disease, known as “brittle” diabetes, in which glucose levels can quickly swing from too high to too low, or vice versa. Even the smallest insulin injection can throw the balance out of whack.

“Type 1 diabetes is like being in hell, man. It’s horrible.”

Jaime Vidal

“It can be a lot to handle,” said Dr. Monica Girotra, an endocrinologist at Memorial Sloan Kettering Cancer Center.

The only potential upside: A self-directed attack on beta cells may be a sign of amped-up immunity more generally. And according to case reports, patients who develop drug-induced diabetes tend to have stronger and more long-lasting anti-cancer responses to checkpoint inhibitor treatments. “So that’s kind of the silver lining,” Quandt said.

The new grant money will go toward launching a massive prospective study involving 1,500 to 2,000 patients undergoing checkpoint inhibitor therapy at more than 30 community hospitals and clinics across the United States. Only a few dozen may eventually develop diabetes — but researchers won’t know whom in advance. So to catch that fulminant disease process in action, they plan to collect blood samples from every patient along each step of their therapeutic odysseys.

For those ultimately diagnosed with diabetes (and some who aren’t), scientists will then throw the kitchen sink of single-cell “omics” technologies on their samples before, during, and after disease onset, in an exploratory effort to uncover biomarkers underpinning the phenomenon.

“I’m pretty open-minded about what mechanistically can be happening,” Parker Institute president and CEO Jeff Bluestone told STAT. “And I’m really optimistic that we’re going to learn a lot quickly.”

Later this year, the National Cancer Institute will also start building a repository of tissue and blood specimens collected from patients who experience immune-related side effects after undergoing checkpoint inhibitor therapy. “We’ll then backtrack to learn more about what might have happened along the way,” said NCI senior investigator Dr. Elad Sharon.

But whereas Bluestone and his colleagues — including Herold and other physician-scientists from Yale, UCSF, the Benaroya Research Institute in Seattle, and the Dana-Farber Cancer Institute in Boston — plan to prospectively stockpile biospecimens from a single registry of patients treated as part of routine clinical care, the NCI team will work after the fact to centralize samples amassed through dozens of government-backed immunotherapy trials. “It’s really a complementary effort,” Sharon said.

Dr. Juan Jaume, an endocrinologist at the University of Toledo Medical Center, welcomes both initiatives. “We should not withhold the treatment of cancer just because of this side effect,” he said, “but we need to learn more about it so we can provide safeguards.”

Two of a kind

The research could also have spillover benefits beyond the cancer clinic.

There are already hints of a genetic predisposition shared by people with the two forms of insulin-dependent diabetes. And now by studying cancer patients whose bodies begin destroying insulin-producing cells in the pancreas after drug treatment, scientists hope to discover why this same process occurs spontaneously in children and young adults with type 1 diabetes.

“If the lessons that we learn from this study are generalizable,” said Ben Williams, a program officer for the Helmsley Trust’s T1D program, “then this could help a lot of people with the spontaneous diabetes that an otherwise healthy kid might get.”

For his part, Lenihan, a software quality assurance engineer from Griswold, Conn., who blogs about his struggle with melanoma, is getting used to living without a working pancreas — monitoring blood sugar levels with the device he wears on his abdomen, and pricking himself with insulin several times each day.

His latest scans this month came back clean, with no signs of active disease growth in his skin, lungs, brain, or anywhere else where metastatic lesions had cropped up before. But he had to stop taking immunotherapy after developing colitis, another autoimmune-related side effect, and now the prospect of the cancer returning weighs heavily on his mind.

His doctor, Yale oncologist Dr. Mario Sznol, is less concerned, though.

“He’s got a pretty good chance of remaining in remission for a very long time, maybe forever,” Sznol said.

As for living with diabetes? “That’s the unfortunate trade-off.”

Leave a Comment

Please enter your name.
Please enter a comment.

  • Checkpoint inhibitors seem to unleash auto-immunity. This is not hard to understand. T1D is an auto immune disorder, and recent evidence is that 40% of patients in the UK present with T1D after age 30. So it is not rare to see spontaneous T1D in older adults. But it is usually misdiagnosed as T2D, and may progress much more slowly than one typically sees in younger people.
    Many patients with checkpoint inhibitors who developed DKA do not have GAD 65 or Zn transporter antibodies. This finding may indicate that those antibodies are not causative but rather are markers of beta cell directed auto immunity.
    Such patients need aggressive insulin treatment with basal bolus insulin. Those developing hypothyroidism are often rapidly progressive as well. Starting low and going slow is a bad idea in that setting- monthly checks on thyroid levels seem appropriate when one sees it.

  • It’s probably a crazy idea, but could patients suffering from immunotherapy-induced diabetes perhaps avoid daily insulin shots by pre-harvesting for later “rescue” some of their own beta cells prior to immunotherapy?

    • Why is that a crazy idea? It probably would be prohibitive…more adverse effects than benefits…at the current risk for diabetes. Not practical to harvest 100 patients for one who needs treatment. But if some marker can identify who’s at risk, then it could be a nice approach. Assuming the longer term effects of the immunotherapy don’t kill the transplanted beta cells too.

    • Probably not, once the therapy has caused the beta cells to be recognized as foreign (broken self-tolerance), it would be similar to a vaccine effect where the patient has, in essence, been vaccinated against their own beta cells. They would have to restore self-tolerance first, which is another area of active research.

    • That’s a patch fix, but companies like Viacyte are looking to tackle T1d exactly that way. The real cure will probably involve injection of an RNA virus that converts the alpha cellls (which are almost identical to beta calls but release glucogon in response to lows) into beta cellls. This would remove the protein marker that causes the initial immune system confusion in the first place. Has already been successfuly solved in rats and we are probably just a few years from a human trail.

Sign up for our Daily Recap newsletter

A roundup of STAT’s top stories of the day in science and medicine

Privacy Policy