Response to FDA Commissioner’s Statement on Dsuvia Approval

In October 2018, the Food and Drug Administration (FDA) approved the opioid analgesic Dsuvia, which is a sublingual formulation of Sufentanil, for use in “medically-supervised health care settings.”Dsuvia is up to 10 times more potent than fentanyl, which is one major driver of the opioid overdose increase (1). In concert with swift media attention and skepticism, the FDA Commissioner issued a statement about Dsuvia approval (2). One specific point warrants further attention. Dr. Gottlieb stated “As we look at the public health implications of each new approval [for opioid drugs], we should evaluate whether we need to take additional steps to systematically consider new opioids relative to the comparative benefit and risk of other opioids already on the market.” In our opinion, evaluating the efficacy and safety of medications compared to existing drugs is essential when drugs have potentially deadly side-effects. We are concerned about the relative lack of this research with Dsuvia.

In their review of studies examining the efficacy of sublingual Sufentanil, Bantel and Laycock (3) report that only one study (4) compared sublingual Sufentanil (15 mcg/dose; 20-minute lockout) with an active comparator (1 mg/dose morphine; 6-minute lockout). While this study did show that sublingual Sufentanil was not inferior (and perhaps even superior) to morphine, results should be interpreted in light of certain limitations. Most importantly, this was an open-label study, so the efficacy of Dsuvia has seemingly not been compared to another active treatment in the context of a double-blind RCT. Second, Melson and colleagues (4) provided supplemental intravenous morphine to all participants in the first 30 minutes of the study that could be self-administered as needed for pain. The group receiving Sufentanil took more than 2.5 times more of this morphine as those in the active comparator group (2.6 mg v. 1.0 mg).While they report this difference was “not clinically meaningful” we would argue the group difference in supplemental morphine intake is a potential confound that compromises the interpretation of pain outcomes by experimental condition. To be clear, these results could be biased in the direction of overestimating Sufentanil’s efficacy. Finally, on an absolute scale, the level of pain-relief from Sufentanil peaked at approximately a 1.7 point reduction compared with baseline using pain ratings on an 11-point Numerical Rating Scale. While this does meet the threshold for clinical significance by some standards, it is not clinically significant according to the more conservative 2-point reduction standard (5). Given the potentially serious harm of Sufentanil, applying a more liberal standard does not seem appropriate in this context. Even if the study conclusions are unbiased and generalizable to an acute pain population, we argue that non-inferiority in a single open-label study with the aforementioned weaknesses should not be enough to warrant use of a potent opioid analgesic.

Thus in our view new analgesic treatments are warranted, particularly in certain settings (e.g., military combat), but the approval of sublingual Sufentanil was premature. Before a drug of this potency becomes used in medical settings, rigorous research is needed to show that it is superior to (and potentially safer than) existing alternatives.