Invokana, a diabetes drug made by Johnson & Johnson, decreases the risk of heart attacks and strokes, but also increases the risk of amputation, particularly of toes.
The result is a blow to J&J, and a boost to rivals Eli Lilly and Boehringer Ingelheim, who make a similar drug called Jardiance, which has been shown to reduce the risk of cardiovascular death but does not appear to carry the amputation risk.
The study also deepens the mystery about such drugs, known as sodium-glucose co-transporter 2 (SGLT2) inhibitors, which prevent the kidney from absorbing sugar from the blood. This lowers blood sugar, is good for the kidney, but leads to more urinary tract infections and weakened bones. But scientists are not sure why the drugs would prevent cardiovascular disease, and it’s entirely unclear why one of them would lead to amputations. "It justifies the need to test each medicine," says Harlan Krumholz, of Yale University.
For every three heart attacks, strokes, or cardiovascular deaths prevented by Invokana, there were two amputations, 71% of them of toes or the lower foot, according to the results of the 10,142-patient study, which was funded by Johnson & Johnson. It is being presented at the annual meeting of the American Diabetes Association tonight, and published by the New England Journal of Medicine.
Heart attacks were far more common than amputations. For every year patients took the drug, there were 31.5 heart attacks, strokes, or cardiovascular deaths per 1,000 patients in the placebo group and 26.9 in the group that took Invokana, a 14% decrease. But, there would be 3.4 amputations per 1,000 patients in the placebo group and 6.3 amputations per 1,000 patients among those who received Invokana – a doubling of risk that held true for both minor amputations (think toes) and major ones (think legs).
Bruce Neal, the senior director of the George Institute for Global Health in Sydney, Australia, and the lead investigator of the study, argued that the results were positive, saying in a press release they showed a “clear benefit” over current diabetes treatments. But other researchers struggled with the data.
“Personally, I would much rather have a small heart attack than lose a toe,” says John Buse, the chief of endocrinology at UNC-Chapel Hill. “And I think I would much rather have a big heart attack than lose a leg.”
The Food and Drug Administration has already added a warning –a ‘black box,’ the strongest kind – to the product labeling for Invokana. But the study shows the magnitude of the tradeoff physicians and patients will be making with the drug, essentially exchanging heart attacks and strokes for amputations.
It also may shows something else: that, overall, Invokana may have a similar effect on heart disease risk as Jardiance. But it’s going to lead to a lot of debate about what the results for the studies of both drugs actually mean.
In a 7,000 patient study published in the New England Journal in 2015, Jardiance reduced the risk of heart attack, strokes, and cardiovascular deaths by 14%, just as with Invokana.
Scientists pool these very different problems in order to get more statistical power. Even though heart attacks are the most common cause of death, they don’t happen that often. In the 10,142-patient Invokana study, there were fewer than 700 heart attacks, strokes, or deaths from cardiovascular disease. But Jardiance and Invokana look very different when these pooled results are broken apart. Jardiance showed a 38% decrease in cardiovascular death, a slight decrease in heart attacks, and an increase in strokes. For Invokana, the results are about equivalent for strokes, heart attacks, and cardiovascular deaths.
Buse predicts “a tremendous debate,” saying that he suspects the benefits and harms are “over- and under-estimated in each study.”
Some key questions: are the drugs really different when it comes to heart disease? The FDA has granted Lilly and Boehringer the right to claim that Jardiance reduces the risk of cardiovascular death. J&J may not get the same, because none of the individual parts of its combined endpoint is statistically significant.
Another question will be whether Jardiance, or Farxiga, another SGLT2 from AstraZeneca, causes amputations. In an interview, James List, the Global Therapeutic Head for cardiovascular and metabolic drugs at J&J’s Janssen unit, said that noticing the amputation problem required a heroic effort by the data safety monitoring committee charged with protecting patients in the study. “They may just not have looked for it adequately in the other trials,” says Steven Nissen, chair of cardiology at the Cleveland Clinic. “You can bet the FDA is going to make them go back and look.” Buse counters that it would be tough to miss a side effect that involves major surgery.
It was Nissen, during a battle a decade ago over whether GlaxoSmithKline’s Avandia causes heart attacks, who pushed the FDA to adopt a policy of demanding larger studies to test whether diabetes medicines were safe for the heart. He calls it “one of the things I’m proudest about.” It led to Lilly and Boehringer finding a heart benefit for Jardiance; it also has led to finding the amputation risk for Invokana. But why either is happening remains a puzzle. The drugs decrease blood pressure and blood sugar, both good for the heart, but that can’t explain the whole benefit. Perhaps the SGLT2 thickens the blood, leading it to pool in the foot.
One possibility, Nissen says, is that the Lilly trial overestimated the benefit on cardiovascular death. A lot of questions, he says, may be answered by the next big clinical trial, being funded by AstraZeneca for its drug, Farxiga. That study involves 17,150 patients, and its size will act as a buffer against chance findings.
One group that may not be that surprised by today’s results: investors. Many had already digested the importance of the FDA’s warning on amputations. In a note to investors on June 8, Vamil Divan, an analyst at Credit Suisse, had predicted a positive result to Invokana, but wrote that the Jardiance – and maker Eli Lilly – would be the biggest beneficiary. Merck, which makes the diabetes drug Januvia, could also be hurt.
