| NEW YORK
NEW YORK (Reuters Health) - Studies that are funded by pharmaceutical companies or involve industry-backed scientists tend to be more prominent at cancer meetings than independent studies, a new report suggests.
"Figuring out the reasons behind these findings is critical," said Dr. Beverly Moy, who led the analysis at Massachusetts General Hospital Cancer Center in Boston.
She and her colleagues also found the proportion of presentations with a financial conflict of interest increased between the 2006 and 2011 American Society of Clinical Oncology (ASCO) annual meetings.
"As long as the studies are done well, I don't think there's any objection to them becoming more prominent," Moy told Reuters Health.
However, she added, past evidence suggests industry-funded research is more likely to be published if it's positive - in favor of a product or pill - than if it's negative, a phenomenon known as publication bias.
So it's important to make sure relationships between scientists and companies stay "productive," she said, rather than become untrustworthy.
The researchers analyzed conflicts of interest, ratings and conference prominence for more than 20,700 scientific abstracts, or research summaries, presented at ASCO meetings in 2006, 2008, 2009, 2010 and 2011.
Over those years, the proportion of studies reporting a financial conflict of interest rose from 33 percent to 38 percent. Their abstracts were rated as slightly better quality by peer reviewers, on average: 2.76 on a scale from 1 to 5, where lower is better, versus 3.01 for studies without a link to industry.
Abstracts tied to pharmaceutical companies were also more likely to have their own session at a conference, or to be presented during a talk or poster discussion. For example, the ratio of industry-supported to non-supported studies was twice as high among oral presentations as among general posters, which receive less attention.
The findings appear in the Journal of Clinical Oncology, which is published by ASCO.
"With the increase in these relationships, we really need to figure out how to manage them," Moy said. It's possible, she said, that such academic-industry teams "yield an alliance that produces better research."
Alternatively, companies might seek out the most prominent researchers for partnerships, or vice-versa.
"I think a lot of the work that gets a high profile is, of necessity, work that has a relationship with industry. They are, after all, providers of most of the ‘breakthrough' drugs," said Dr. David Johnson, chair of internal medicine at the UT Southwestern School of Medicine in Dallas.
But the study does serve as a caution as well, said Johnson, who has studied conflicts of interest in medical research but wasn't involved in this analysis.
"It again points out that there is this potential for greater and greater influence in the relationship of industry and the direction that biomedical research takes, and we have to constantly be on guard," he told Reuters Health.
"I think there is a pretty clear influence that industry has on research that's not always bad, but it's not always good either … We just have to be really, really cautious."
SOURCE: bit.ly/128Yg4U Journal of Clinical Oncology, online June 17, 2013.