Assessing Suicidality in Clinical Trials: Paxil's Wakeup Call

The recent news regarding Le Noury and colleagues’ re-analysis of a 14-year old study demonstrating the safety of paroxetine (Paxil) in children and teenagers (2015; British Medical Journal) made national headlines and caused a stir about how drug manufacturers evaluate and document the safety and efficacy of new drugs. To sum up, the authors concluded that its secondary analysis of existing data from Study 329 (Keller et al., 2001) demonstrated clinically significant increases in harms, including suicidal ideation and behavior (SIB) and other serious adverse events in individuals taking paroxetine. Le Noury and colleagues’ (2015) results contrast with both the published conclusions of the original publication by Keller and colleagues (2001) and the way that the conclusions were reported and interpreted.

Although there are questions on suicidal ideation within validated depression measures (e.g., K-SADS-L; HAM-D), the original Keller et al., 2001 paper did not use a formal assessment tool to identify and mitigate risk of SIB. The new findings of Le Noury and colleagues (2015) provide evidence for a utilizing a methodical, anchored approach to categorizing suicidality, which can provide an accurate and comprehensive identification of suicidal events during and after participation in a clinical trial

It is coincidental that the questions surrounding paroxetine are occurring this month, as September marks the five-year anniversary of the U.S. FDA’s Guidance to Industry, “Suicidality:  Prospective Assessment of Occurrence in Clinical Trials.” Now more than ever, the pharmaceutical industry needs to take a closer look at their drug development pipelines to confirm they are taking appropriate, effective steps to ensure patient safety by identifying the risk of treatment-emergent SIB during clinical trials.

FDA’s Stance re: Suicidality Assessment

Prospective and proactive assessment of the risks of SIB during drug development stages and in clinical trials has been strongly recommended by the FDA. First released in September 2010, the FDA’s draft guidance cited the Agency’s current thinking on the assessment of treatment-emergent suicidality. Revisions issued in August 2012 reinforced the application principles of the Guidance and brought more clarity to the expectations of when, where and how to assess SIB.

In the Guidance, FDA suggests that, “…prospective suicidal ideation and behavior assessments should be carried out in all clinical trials involving any drug being developed for any psychiatric indication, as well as for all antiepileptic drugs and other neurologic drugs with central nervous system (CNS) activity, both inpatient and outpatient, including multiple-dose phase 1 trials involving healthy volunteers.”

The FDA’s revised Guidance recommends the specific activities that researchers should include when conducting clinical trials with products that have high-potential SIB risk, and includes expectations of what types of data are to be delivered to evaluate SIB. It should be noted that some populations may be difficult to assess for suicidality, predominantly if they suffer from dementia, Alzheimer's disease or cognitive impairments. Sponsors are informed to obtain permission with FDA's review division before excluding these patients from SIB review.

Challenges of Traditional Suicide Risk Assessments

Procedural variances in the way SIB risk and all clinical assessments are performed by human raters are associated with a number of shortcomings negatively impacting the reliability of results. Even with extensive training, rater skills deteriorate over time and clinicians are often influenced by prior experiences with patients. Further, performing these risk assessments in a traditional manner – on paper and rated by the clinician – represents a significant burden for the investigative site, requiring additional staff training to ensure proper SIB evaluation. As a result, researchers are actively looking for ways to overcome these limitations.

Electronic SIB assessment tools are increasingly being used to overcome the shortcomings associated with traditional, paper/clinician-based assessments and to capture more reliable data.  Electronic, self-rated tools developed from existing psychometrically reliable and valid instruments provide sponsors with a fast and reliable means of collecting valuable information relating to the welfare and safety of patients and ultimately allow swift and effective interventions for at risk patients. Additionally, given the confidential nature of questions in suicidality assessments, patients are more likely to reveal true answers on an electronic instrument rather than to a clinician¹, further improving the quality of data. This approach is expected to ensure patient safety and provide more consistent and reliable data.

Protecting Patients and Compounds

Drug developers should consider using proactive and efficient approaches to incorporate SIB assessment into clinical trials consistent with the assessment of vital signs. Additionally, for drugs with a long half-life, assessment of SIB in patients should be maintained even after the clinical trial has ended. By doing so, drug developers can ensure patient safety, efficiently adhere to the requirements of the U.S. FDA Guidance, and recognize additional benefits, including the likelihood of protecting their compounds from false positive findings.

Anzalee Khan, PhD, is Clinical Science Advisor at ERT.

References 

1) Gao K, Wu R, Wang Z, Ren M, Kemp DE, Chan PK, Conroy CM, Serrano MB, Ganocy SJ, Calabrese JR. Disagreement between self-reported and clinician-ascertained suicidal ideation and its correlation with depression and anxiety severity in patients with major depressive disorder or bipolar disorder. J Psychiatr Res. 2015 Jan; 60:117-24. doi: 10.1016/j.jpsychires.2014.09.011.