Zyprexa, Eli Lilly’s blockbuster “atypical” antipsychotic, causes weight gain. So — why not use it to treat anorexia?

For years, drug companies denied that the new antipsychotic medications, such as Seroquel and Risperdal, cause undue weight gain and even diabetes. Slowly, after private lawsuits and then disputes and negotiations with the Food and Drug Administration, Pharma came to acknowledge the side effect.

Of the medications, the worst offender for metabolic changes was Zyprexa. In 2003, all drugs in the class received an FDA label concerning the risk of diabetes. Late last year, after large-scale outcome trials confirmed the extent of the problem, Lilly agreed to an additional warning indicating that the increases in blood glucose levels appeared to be greater with Zyprexa than with similar medications.

Researchers who gave credence to the side effect had repeatedly tried using Zyprexa in the treatment of the eating disorder, anorexia nervosa, with promising but mixed results. There were two reasons to test the medication: it might lessen psychological elements of anorexia, like compulsivity, and it might simply cause anorexics to eat, via the metabolic changes which here become a main effect in treatment.

The on-line edition of the American Journal of Psychiatry has just published the results of a double-blind study, sponsored by Lilly, of Zyprexa as an aid in the treatment of anorexia in a day hospital setting. The trial, conducted by Hany Bissada and others at the University of Ottawa, was of modest size and duration: 34 patients, of whom 28 completed the study, treated for ten weeks and followed three weeks longer.

As often happens with small samples, the treatment and placebo groups were not fully equivalent. The patients on placebo were older; the report does not say whether they had lived more years with anorexia, but likely they had. The subjects given medication had more obsessive symptoms than did those in the control group.

Zyprexa did seem to facilitate or at least speed weight gain. The patients on medication were more likely to achieve their “target body mass index,” that is, a safer weight, in the ten weeks of the study; but by the thirteenth week, the day hospital program had worked for almost all the patients, on or off medication.

The psychological symptom changes were similar for the two groups as well. Over the course of hospitalization, both sets of subjects became less depressed, less anxious, less compulsive, and less obsessive. The report notes one difference, a statistically significant change in obsessionality in the treated group, compared to the controls. But recall that the patients on Zyprexa had more obsessive symptoms to start with; they also had more at the end of the study. And often it is easier to change symptom ratings when you start at a high baseline, especially if the statistic gives credit for patients who are better but not well. The finding about effects on obsessionality has to be taken with a grain of salt.

None of the study’s results is impressive, but patients in the Zyprexa group did get to an acceptable weight faster. It may be that Zyprexa simply makes it hard to keep off weight, through changes in appetite and metabolism. Past studies of Zyprexa have found relatively more weight gain in women and in people who are lighter to begin with. There is even a report of an anorexic on Zyprexa developing high blood sugar, at the level of diabetes.

The current on-line article, about the Lilly-sponsored study, indicates that “participants were recruited from patients referred to the eating disorders program . . . between September 2000 and April 2006.” Even regarding pharmaceutical houses, it is possible to be too cynical. After all, a previous small trial (with its own severe problems in randomization) had suggested that Zyprexa might decrease obsessive ruminations in anorexia. The disease is dangerous and hard to treat; certainly new approaches would be welcome. Still — thinking of the years in which Lilly sparred with the FDA over the level of warning that should apply to the drug — it is hard not to speculate about who believed what regarding Zyprexa and weight gain.