IOANNINA, Greece -- Clinicians may have to reassess the ABCs of the gender differences in disease manifestations specifically linked to X or Y chromosomes.
IOANNINA, Greece, Aug. 22 -- Clinicians may have to reassess the ABCs of the gender differences in disease manifestations specifically linked to X or Y chromosomes.
Of 432 claims reviewed in a database study, most prominently claimed sex-related differences in genetic diseases were insufficiently validated or spurious, while those with good documentation were rare, researchers here reported.
Only 12.7% were properly documented with good internal and external validity, John P.A. Ioannidis, M.D., of the University of Ioannina, and colleagues, reported in the Aug. 22 issue of the Journal of the American Medical Association.
For many common diseases, published research has often considered that some common gene variants may have different effects in men and women, and many studies have tried to determine differences in these risks, the researchers wrote.
However, empirical data from randomized trials suggest that many claimed subgroup differences on the basis of sex have not been valid and have led to serious misconceptions, they added.
For example, they said, aspirin was believed to be ineffective in secondary stroke prevention in women for more than 10 years on the basis of an underpowered subgroup analysis.
Claims covered a variety of genes and outcomes. Examples of articles making prominent claims for sex differences included studies of HDLs, total cholesterol, coronary artery calcification, conscientiousness and neuroticism, Alzheimer's disease, hypertension, BMI, acute stroke apolipoprotein levels, ischemic heart disease, and age at myasthenia gravis onset,.
To evaluate the validity of these claims, the researchers retrieved 215 genetic-effect studies claiming sex-related differences from a PubMed database through July 6, 2007.
They considered all studies that claimed different genetic effects across sexes of one or more gene variants for any human disease or phenotype. These included both biallelic and multiallelic markers (including haplotypes) and both binary and continuous phenotypes and traits.
Two evaluators independently extracted data with a third evaluator arbitrating their discrepancies. Data evaluation included whether analyses were specified a priori, whether sex effects were evaluated in the whole study or in subgroups, and whether the claims were appropriately documented, insufficiently documented, or spurious.
Finally, the investigators compared those sex-difference claims with the best internal validity against the results of other studies addressing the same interaction.
In an analysis of 432 sex-difference claims in 77 eligible articles, sex comparisons were decided a priori for 286 claims (66.2%), while 68 (15.7%) were analyzed after the study. In the remaining 78 (18.1%), the analysis plan was unclear. The entire sample size was used in 210 (48.6%) claims, the researchers said.
Appropriate documentation of gene-sex interaction was recorded for only 55 claims (12.7%), while documentation was insufficient for 303 claims and not valid for 74.
Data for reanalysis of claims were available for 188 comparisons. Of these, 83 (44.1%) were only nominally statistically significant at a P=0.05 threshold, and most of these had modest P values between 0.01 and 0.05.
Furthermore, of 60 claims with seemingly the best internal validity, only one was consistently replicated in at least two other studies, Dr. Ioannidis said.
Examples of insufficient documentation included finding a statistically significant effect in one sex but not in the other, or with no information on significance in the other sex, and unclear, opposite, or lacking point directions for the findings.
Discussing the study's limitations, the researchers wrote that the sampling strategy was heavily driven by convenience, tending to select more prominently claimed sex differences.
The researchers also acknowledged that for some claims, especially those first made most recently, corroboration may not yet have been done but may be done in time.
The issues addressed in this study, the researchers wrote, focus on gene-sex interaction, but their implication probably extends to any kind of subgroup analysis in genetics. Thus caution may be needed in the analysis and interpretation of all these postulated effect modifiers, such as age, racial-ethnic descent, even diet, lifestyle, and other exposures.
"We hope that our empirical evaluation will help sensitize clinicians, geneticists, epidemiologists, and statisticians who are pursuing subgroup analyses by sex or other subgroups on genetic associations. The pursuit of gene-sex interactions should not be necessarily abandoned," the investigators wrote.
Dr. Ioannidis and his colleagues offered this advice:
Ideally, sex differences should be based on a priori, clearly defined, and adequately powered subgroups.
Post hoc, discovery-based analyses are also of interest, but their post hoc character should be clearly stated in the manuscript.
Both a priori and post hoc claims should be documented by interaction tests and proper consideration of the multiplicity of comparisons involved.
Even then, results should be explained with caution and should be replicated by several other studies before being accepted as likely modifications of genetic or other risks.
Nikilaos A. Patsopoulos, M.D., a co-author, was supported in part by a PENED grant sponsored by the EU European Social Fund and the Greek Ministry of Development-General Secretariat for Research and Technology.