Science has just retracted [*] a paper purporting to show that even one-time use of MDMA (“ecstasy”) brings with it damage to the dopamine system that creates a risk of developing Parkinson’s Disease later in life. [*] The paper, one of a series published by George Ricaurte of Hopkins Medical School and colleagues, was based on an experiment that intended to give MDMA to a group of primates.
It now turns out, according to the experimenters, that the animals were actually administered methamphetamine instead.
The authors of the study insist that their earlier results showing that MDMA is toxic to serotonergic neurons remain valid. That work, combined with the now-retracted paper, formed part of the basis for such laws as the Ecstasy Anti-Proliferation Act and the RAVE Act, and have also led to an almost-complete blackout on human experimentation with MDMA in the US.
The account in the retraction letter seems plausible at a first reading. The laboratory ordered 10 grams each of MDMA and methamphetamine from the same supplier, and received two vials, one labeled “MDMA” and one methamphetamine, in a single shipment. Apparently the two labels were switched; at least, the vial labeled “methamphetamine” turns out to contain only MDMA. (The other was used up and has been discarded.) Moreover, analysis of frozen sections of the brains of two animals that died during the experiment show methamphetamine and its metabolites but not MDMA or its metabolites.
Accidents happen, and it would be churlish to cast too much blame on the scientists involved for not analyzing a chemical received from a reputable supplier to verify that it was what the label said it was. But the retraction letter leaves out or elides some important facts, which it seems to me should have troubled both the researchers and Science’s reviewers and editors.
The experiment was purportedly intended to represent in an animal model the consequences of human “recreational” MDMA use, and perhaps of therapeutic use of the drug were it ever approved for that purpose. In the experiment, two out of fifteen animals died. The death rate among human MDMA users is no more than one in a million.
Yet it appears that the researchers failed to investigate the causes of those deaths. Moreover, they went on to draw inferences about the effects of MDMA on humans from the observed damage to the brains of the remaining animals. That didn’t seem to trouble the reviewers for Science or the administrators at the National Institute on Drug Abuse who trumpeted the findings as evidence of the dangers of MDMA. (Science is published by the AAAS, whose president, Alan Leshner, was the Director of NIDA when the grant in question was awarded; he made MDMA his particular crusade.)
It is hard to escape the thought that many of the people involved were less cautious than they might have been because the results seemed to support their already strongly-held beliefs.
The now-retracted study was not the first by Ricaurte to attract severe scholarly criticism. The earlier work claiming large and irreversible losses in the serotonin system has also been contradicted by other findings, and brain-scan images that seemed to show large inactive areas in the brains of MDMA users — imagines heavily promoted by NIDA during the Leshner regime — have now been withdrawn from the NIDA website, and NIDA refuses to make them available to journalists.
Even before the retraction, then, Ricaurte’s work was under a cloud. One very senior figure in the field had said in an open scientific meeting “I will believe any result George Ricaurte comes up with as soon as it has been independently confirmed.”
The other detail not mentioned in the retraction letter is that Peter Jennings had gotten wind of the controversy, and had a special highly critical of Ricaurte’s work “in the can” and ready to show. (Ricaurte had refused to be interviewed.) That Ricaurte’s group had failed to replicate the now-retracted work is no doubt true; but that the paper would have been retracted had ABC not been ready to make an issue of it is much less clear. (But see correction in “second update” below.)
Given all the circumstances, and especially given the huge policy superstructure that has been erected on the increasingly shaky foundation of MDMA neurotoxicity research, it seems to me that it’s time for a complete review of the bidding, in the form probably of a National Academy review. The panel for such a review ought to be charged with evaluating the whole body of MDMA research, with a special eye on the neurotoxicity work, and ought to be asked to make a recommendation about whether human preclinical and clinical trials ought to be allowed to go forward.
If the panel were to find serious flaws not only in this paper but in other work by the Ricaurte group and other NIDA-funded work on MDMA, then more serious questions ought to be raised about how to protect the NIDA funding and evaluation process from being unduly influenced by the political necessities of the drug war. Moreover, if I were the editor of one of the relevant journals I would want to take a close look at whether there were subtle pressures making it easier to publish results consistent with the national crusade against illicit chemicals than results inconsistent with that crusade.
Only rarely does a first-rank scholarly journal has to retract a published research report because of flaws in the data or analysis. Most of us hope that means that the disincentives for either faking results or making crucial errors are so strong, and the journal editors and referees sufficiently alert, so that few papers actually need retraction.
A less cheerful view would be that the common interests of the authors and the journals in not retracting means that many papers that deserve retraction don’t receive it. Probably the chances of retraction, among papers so flawed that they ought to be retracted, is higher when the results help one side or another in a political or policy controversy. The temptations to deviate from objectivity are also probably greater in those instances.
A reader who has worked in research labs employing controlled substances makes the following observation. (Note that the Drug Enforcement Administration has regulatory authority over any place with a license to possess Schedule I drugs):
It is not clear to me WHEN or HOW the labels were switched.
The first thing that came to mind when this story was released was
that DEA inspectors of laboratories which use controlled
substances suggest that labels (of stock compounds in the safe as
well as working aliquots) be changed to mask the contents of the
drug. I’ve seen “WP” for “white powder” (actually morphine
sulfate) and single letter codes to distinguish between series of
compounds. I’ve heard stories of inspectors reacting
enthusiastically to sarcastic suggestions that the compounds be
falsely labeled radioactive (a spectacularly bad idea). These
things are supposed to decrease likelihood of diversion, a minor
concern except by people who would already have access and
knowledge of the coding.
Confusion caused by relabeling or diversion by a lab member
seem more likely than an error by the manufacturer.
Second update and correction
Someone in a position to know informs me that pressure from ABC News, if it played any role at all, was secondary to pressure from NIDA. Apparently Leshner’s enthusiasm for Ricaurte’s work was not shared by the career staff or by the subsequent Interim Director and Director. (For example, though NIDA had funded the research reported in Science, it put out no press release when the article appeared. As noted above the “hole-in-your brain” images disappeared from the NIDA website some time ago.) NIDA was pressing Ricaurte for details about how the work was done, and word that the results reported in Science could not be reproduced in subsequent oral-administration studies was beginning to leak from Ricaurte’s lab.
Other questions are being raised: What happened to the rest of the 10 grams of methamphetamine purportedly mislabeled “MDMA”? What other studies might be compromised by the same error? Did Ricaurte continue to make public claims about the risk of Parkinsonism after his lab failed to confirm the earlier results?
One knowledgeable observer doubts that a National Academy panel is necessary. “All you need is a careful review of the lab notebooks.” The question now is whether Hopkins will stick with its “accidents happen” stance.