Spin and Bias in Breast Cancer Studies
Authors call for guidelines for reporting efficacy and toxicity in published trials (Jan. 9)
Spin and bias exist in a high proportion of published studies of the outcomes and adverse side effects of phase III clinical studies of breast cancer treatments, according to new research published in the Annals of Oncology.
In the first study to investigate how accurately outcomes and side effects are reported in breast cancer trials, researchers in Toronto, Canada, found that in a third of all trials that failed to show a statistically significant benefit for the treatment under investigation, the reports focused on other, less important outcomes in order to influence positively the interpretation of the results.
Moreover, two-thirds of the reports showed bias in the way adverse effects of the treatment were reported, with more serious side effects (those with grade III of IV toxicities) poorly reported. Only 32% of articles gave details of the frequency of grade III or IV toxicities in their abstracts.
The authors analyzed published reports for 164 phase III randomized controlled trials — 148 for systemic therapy, 11 for radiation therapy, and five for surgical therapy. They found that 54 (33%) of these studies were reported as positive, despite not finding a significant difference in the primary endpoint. In addition, 110 papers (67%) met the researchers’ definition of biased reporting of toxicity. Such bias was significantly associated with the use of overall survival as the primary endpoint (P = 0.028).
Funding from industry partners was reported in 103 studies (63%); 32 studies (20%) were supported by academic or government grants; and 29 studies (18%) did not indicate the source of funding. The authors found no association between industry sponsorship and biased reporting of either efficacy or toxicity.
In order for clinical trials to be published in most academic journals, it is now mandatory that they be registered before they start. Studies in the U.S., for example, are registered with ClinicalTrials.gov. Some of the trials analyzed in the new report began before registration became compulsory. However, for those that were registered, the researchers found that some changed the primary endpoint between registration and the publication of the study’s outcomes. “Among these trials, there was a trend towards change of the primary endpoint being associated with positive results, suggesting that it may be a strategy to make a negative trial appear positive,” the authors reported. “Trial registration does not necessarily remove bias in reporting outcomes, although it does make it easier to detect.”
The authors recommend that clinicians, reviewers, journal editors, and regulators apply a “critical eye” to trial reports. In their opinion, guidelines are necessary to improve the reporting of both efficacy and toxicity.
Source: Annals of Oncology; January 9, 2013.