Results tagged “Neurontin” from Pathophilia
Here's something any perpetrator should catch hell for: changing the primary outcome of a trial.*
In studies of gabapentin (funded by Parke-Davis and later by Pfizer) for neuropathic pain, bipolar disorder, or migraine prevention (all off-label uses), the protocol-defined primary outcome was often changed in the published report. This is just one observation of Johns Hopkins investigators after their analysis of internal documents from 20 industry-sponsored gabapentin studies. Their full report is available in the latest issue of the NEJM.
Among the 20 trials assessed, only 12 (60%) were published (see Table); the results of 9 trials were reported as full-length articles. In 8 (67%) of the 12 trials, the protocol-defined primary outcome differed from the primary outcome in the final report. In 6 (50%) studies, a completely new primary outcome was introduced; and in 5 studies, the new primary outcome favored gabapentin treatment.
|
Article |
Statistical Significance Favoring Gabapentin? |
|
Primary outcomes agreed with protocol | |
|
Yes | |
|
Backonja M, Mutisya EM. Review of gabapentin dosing in five placebo-controlled clinical trials for neuropathic pain. Eur J Neurol. 2002;9(suppl 2):191. |
No |
|
Yes | |
|
Gomez-Perez FJ et al. Gabapentin for the treatment of painful diabetic neuropathy: dosing to achieve optimal clinical response. Br J Diabetes Vasc Dis. 2004;4:173-178. |
Yes |
|
Primary outcomes did not agree with protocol | |
|
Mathew NT et al. Efficacy of gabapentin in migraine prophylaxis. Headache. 2001;41:119-128. |
Yes |
|
Wang PW et al. Gabapentin augmentation therapy in bipolar depression. Bipolar Disord. 2002;4:296-301. |
Yes |
|
Yes | |
|
Yes | |
|
Gorson KC et al. Gabapentin in the treatment of painful diabetic neuropathy: a placebo-controlled, double-blind, cross-over trial. Neurology. 1998;50(suppl 4):A103. |
Yes |
|
Wessely P et al. Preliminary results of a double-blind study with the new migraine prophylactic drug gabapentin. Cephalalgia. 1987;7(suppl 6):477-478. |
No |
|
No | |
|
No | |
Other scientifically unsound behavior included the failure to differentiate between primary and secondary outcomes (2 trials), the relegation of primary outcomes to secondary outcomes (2 trials), and the failure to report 1 or more protocol-defined primary outcomes (5 trials). The Hopkins investigators found that trials with statistically insignificant primary outcomes were more likely to be partially reported or reported with a changed primary outcome.
The internal documents were made available to the Hopkins researchers as a result of ongoing litigation regarding the improper marketing of gabapentin. The anchor author of the analysis (Kay Dickersin, PhD) has served as a paid expert witness in litigation related to gabapentin clinical trials, and the lead author (S. Swaroop Vedula, MD) has received fees from plaintiffs' lawyers.
A study published earlier this week, in the Annals of Family Medicine, revealed that the primary outcomes of randomized studies published in 5 highly regarded journals (including the NEJM), were often changed from those defined in the trials' registry listings. The original primary outcomes of the gabapentin trials were only known through access to internal documents. Dickersin argued to MedPage Today that public registration of industry-sponsored trials should be mandatory to aid the transparency of these clinical studies.
N.B.--In 2004, Pfizer paid a $430-million fine to the government for the off-label marketing of gabapentin (ie, Neurontin).
* Without a darned-good reason.
