SEARCH

SEARCH BY CITATION

We recently performed what to our knowledge is the largest survey published to date regarding clinical trials of biotherapies for solid tumors to review their content and quality. Data concerning general study characteristics, patient and disease factors, study methodology, and factors related to the completeness of reporting in 334 prospective clinical trials were collected from 5 distinguished journals between 1990 and 2002. We found a wide diversity with regard to the modalities of conducting and reporting such trials as well as the presence of some methodologic pitfalls, suggesting that the methodologic standards for the conduction and publishing of clinical trials of biotherapies should be improved.1

In an additional analysis, we found a statistically significant association between the high impact factor of the journal (> 5) and a positive conclusion for the study (P = 0.0003). It is interesting to note that, in a multiple regression model performed to analyze the correlation between impact factor and two explanatory variables of Phase II studies (high quality vs. poor quality and a positive study conclusion vs. a negative study conclusion), the only variable found to be predictive of publication in a high-impact factor journal was the positive conclusion of the study (P = 0.0024). These latter findings could indicate that a positive trial regarding biologic therapies is more likely to be published in a high-impact factor journal independent of its quality. Publication bias in other medical contexts already has been found and discussed previously.2, 3

We would point out that interest in biotherapies is increasing and that, in the near future, these new compounds will enrich the therapeutic possibilities of oncology. However, the considerable heterogeneity observed between trials and evidence of a publication bias might affect the reliability of the body of published data concerning biotherapies for cancer. This issue should prompt editors and reviewers to improve their strategies and methodologic tools, particularly when discussing and evaluating negative study results.

REFERENCES

  1. Top of page
  2. REFERENCES