Association of Conflicts of Interest for Public Speakers for the Peripheral and Central Nervous System Drugs Advisory Committee of the US Food and Drug Administration With Their Statements

The Peripheral and Central Nervous System Drug Advisory Committee (PCNSDAC) of the US Food and Drug Administration (FDA) makes recommendations for the approval or denial of drugs that treat nervous system diseases. Before issuing recommendations, the PCNSDAC hears public testimony about these drugs in an open forum.¹ For example, eteplirsen, a treatment for Duchenne muscular dystrophy, was approved after 52 public speakers provided testimony at the PCNSDAC meeting.² After approval of eteplirsen, critics spoke out against its approval, citing weak evidence from a 12-patient trial based on a surrogate end point.³ Because public speakers overwhelmingly spoke in favor of eteplirsen,³ critics raised concerns regarding the influence of public testimony on FDA drug approval.⁴

In the interest of all patients, conflicts of interest (COI) among public speakers must be mitigated to ensure the ethical approval of drugs. Without addressing public speaker COIs, the integrity of new drug approvals may be compromised. Recent evaluations of public speaker COIs reported that pharmaceutical companies often pay traveling and lodging expenses for them.^5,6 The aim of this study was to evaluate public PCNSDAC speakers to determine the proportion with an industry COI and the association of those COIs with the speaker’s statements about the drug.

We based our methods on previous studies.^5,6 A total of 19 publicly available meeting transcripts for PCNSDAC meetings from January 7, 2009, to September 18, 2017, were analyzed. The following data were extracted in duplicate by masked investigators (W.A., J.A.): the public speaker’s name, the organization that he or she represented, the drug being discussed, whether the speaker had the disease for which the drug was indicated, whether the speaker took the drug, the speaker’s disclosed COIs, and whether his or her statements about the drug were positive, neutral, or negative. This study used publicly available data and was therefore exempt from institutional review board approval. Logistic regression was used to examine the association between declaring a COI and having the disease for which the drug is indicated and positive statements about the drug. We attempted to include a third variable (whether the speaker took the drug) but were not able to because no patients who took the drug spoke negatively. Stata statistical software, version 15.1 (StataCorp) was used for data analysis.

We identified 226 public speakers from 17 meetings (Table 1 and Table 2). Of the 226 speakers, 129 (57.1%) disclosed a COI, 42 (18.6%) denied COIs, and 55 (22.3%) did not mention COIs. A total of 89 of the 129 public speakers (69.0%) were reimbursed for their travel. A total of 125 (96.9%) spoke positively, 2 (1.6%) spoke negatively, and 2 (1.6%) were neutral. Speakers with a COI were more likely to give favorable remarks about a drug compared with speakers with no COIs (adjusted odds ratio, 4.80; 95% CI, 1.48-15.50). Having the disease was not associated with positive remarks (adjusted odds ratio, 2.60; 95% CI, 0.56-12.08). All patients who took the drug spoke positively.

Our results suggest that public speakers at PCNSDAC meetings are more likely to provide positive statements about a drug if they have an industry COI or take the drug in question but not more likely if they have the disease for which the drug is indicated. These findings are consistent with previous studies.^5,6 This study is limited by the use of only publicly available meeting transcripts. Thus, our analysis may exclude potential confounding variables.

Given the proportion of patients who disclosed a COI and the association between COIs and positive statements, the acceptability of industry influence in gathering public speakers should be questioned. Of note, a significant portion of speakers did not mention COIs at all. Given the possibility that public speakers are capable of swaying approval committee opinion about a drug, declaring or denying COIs should be required.

Accepted for Publication: June 22, 2018.

Corresponding Author: Wade Arthur, BS, Oklahoma State Center for Health Sciences, 1111 W 17th St, Tulsa, OK 74017 (wade.arthur@okstate.edu).

Published Online: December 17, 2018. doi:10.1001/jamaneurol.2018.3997

Author Contributions: Mr Aruther and Ms Austin had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Wayant, Vassar.

Acquisition, analysis, or interpretation of data: Arthur, Austin, Vassar.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: Arthur, Wayant, Vassar.

Statistical analysis: Arthur, Wayant.

Supervision: Vassar.

Conflict of Interest Disclosures: None reported.