Declared pharmaceutical industry involvement in a random sample of abstracts from the ASCO Annual Meetings

Abstract

e17525 Background: The pharmaceutical industry (PI) plays an important role in drug development. We examined the extent and correlates of declared PI involvement in abstracts presented at ASCO Meetings 2001–2003 and 2006–2008. Methods: After stratification of abstracts into three categories of presentation ([1] oral, including plenary and oral presentations, and integrated symposia; [2] poster, including poster discussion and general poster; and [3] publication only), we randomly selected 10% of the abstracts for each of the 3-year periods of interest. Involvement by the PI was defined as declared sponsorship or by co-authorship in abstracts. Multivariate analysis (MVA) was done using logistic regression. Results: In the 2,206 abstracts analyzed, there was declared sponsorship or co-authorship by the PI in 425 (19.3%) of cases. Declared sponsorship (N = 73) was less frequent than co-authorship (N = 364). Declared sponsorship or co-authorship was present in 160 of 332 (48.2%) multinational studies, 222 of 969 (22.9%) multicenter, uninational studies, and 43 of 905 (4.3%) of single-institution studies (p < 0.001). For the 30.6% of abstracts stating the study phase in the title or text, PI involvement was present in 116 of 239 (48.5%) phase I studies, 95 of 364 (26.1%) phase II studies, and 26 of 72 (36.1%) phase III trials (p < 0.001). PI involvement was present in 280 of 1,080 (25.9%) US studies and in 145 of 1,126 (12.9%) non-US studies (p < 0.001). In MVA, significant predictors of PI involvement were multinationality (odds ratio [OR] = 6.61), phase I/II/III (versus other, OR = 3.65), and US nationality (OR = 2.98). Conclusions: Declared PI involvement is seen in nearly 20% of abstracts and is more frequent in multinational studies, clinical trials and studies with US nationality. Additional work is needed to assess the correlation between PI involvement, abstract quality and likelihood of future publication. No significant financial relationships to disclose.