AICAR

AICAR is not a peptide. What comes in the vial is a small molecule called acadesine, a sugar joined to a nitrogen containing ring, and it is a close relative of something your own cells already build and carry around. It is sold as a powder in a small glass bottle, mixed with liquid and injected under the skin, and it is marketed as exercise in a bottle. That phrase did not come from a seller. It came out of the scientific press around a 2008 mouse study, in which mice that never trained ran about 23 percent longer and about 44 percent further than mice given a dummy injection. Unlike almost everything else in this reference it has a large human record. Roughly 7,100 people were randomised in heart surgery trials, about 3,550 of them to the drug itself, given by drip into a vein over about seven hours. It was never developed for exercise, and none of that work measured endurance.

Once it is inside a cell, the cell sticks a phosphate onto it and turns it into a molecule called ZMP. ZMP resembles AMP, the low fuel signal inside a cell, closely enough to switch on AMPK, the enzyme that reads that fuel gauge and tells a cell to stop storing energy and start burning it. That is the whole sales story, and it is where the human measurements stop agreeing with it. Three separate studies infused it into people and measured that enzyme in muscle, and in all three the enzyme itself did not move, while in one of them riding a bicycle moved it in the same men on the same day. Note also that the rationale the 1990s programme was actually built on is a different one: acadesine raises adenosine, a signal that widens blood vessels and calms inflammation, in tissue short of oxygen. Two mechanisms, two eras, one molecule, and they do not support each other.

People buy it for endurance, for body composition, and for the general idea of getting the effect of training without doing the training. That set of claims traces back to a single mouse study, Narkar 2008 in Cell, 500 milligrams per kilogram a day injected into the abdominal cavity for four weeks. I could find no human study of AICAR for endurance, body composition or athletic performance in healthy adults, by any route. The only people I found put on a treadmill with this drug in them were twelve patients with chronic stable angina, given it into a vein across four doses spanning an eightfold range, and the placebo adjusted change in exercise time wandered up and then back down again.

The costly mistake is reading the trial programme as evidence for the thing being sold. Studied in thousands of patients is true, and it describes a drip into a vein during heart surgery, on a rationale about blood vessels that predates the exercise story entirely, and the trial built to settle whether it worked was stopped early for futility with the bad outcome landing on 5.0 percent of the dummy group and 5.1 percent of the drug group. The second mistake is inside the mouse study everyone cites. The famous 44 percent is distance and the time figure was about 23 percent, and the mice were injected into the abdominal cavity, though the paper's own abstract calls the compound orally active. How much survives being swallowed had already been measured in people in 1991, at below 5 percent.

The safety file here is unusually large and all of it is by the wrong route. In the trial with posted results, of 1,442 people given a single infusion into a vein, 327 had a serious adverse event, against 329 of 1,457 people given salt water. The consistent signal is a rise in uric acid, because the drug is broken down into it, and uric acid at high levels causes gout and stresses the kidney. In the leukaemia trial a patient at the lowest dose tested had a rise counted as a dose limiting toxicity, after which everyone in the higher dose groups was given allopurinol to hold it down and no further cases appeared. Four of those 24 patients had kidney impairment, two of them at the top dose, and all recovered. A French trial in bone marrow disorders closed in June 2015 with five people enrolled, and the registry gives its reason in two words: renal toxicity, meaning harm to the kidneys. What stops that being a clean bill is that the exposure studied is not the exposure sold. One infusion, or five doses over 15 days, into a vein, under supervision. I found no human repeat dose study at daily dosing for weeks, and none for injection under the skin.