Potential biases resulting from the use of patients as their own controls within Q-Med's Restylane pivotal trial drew criticism during FDA panel review of the hyaluronic acid-based dermal filler Nov. 21.

In the 138-subject study supporting Medicis/Q-Med's premarket approval application, patients were injected with Restylane in the nasolabial fold on one side of their face, and with the control (Inamed's Zyplast collagen filler) on the other side.

General & Plastic Surgery Devices Panel members agreed that blinding in the study was compromised. Roughly 40% of subjects had undergone previous Zyplast injections while evaluators were dermatologists who had injected the collagen; both were familiar with the product's degradation relative to Restylane's.

Panelist Amy Newburger, MD, Dermatology Consultants of Westchester, Scarsdale, N.Y., stressed that the informed guessing of experienced evaluators biased the already subjective Wrinkle Severity Rating Scale assessment on which the trial is based. "These individuals would recognize characteristics," she said.

Responding to FDA's question about Restylane efficacy, panelist Brent Blumenstein, PhD, TriArc Consulting, maintained that while "there is evidence that there is some effectiveness," it is undermined because the "SRS score is not highly reliable...and the integrity of masking is not good."

In contrast to the Restylane trial, Genzyme Biosurgery's 261-patient Hylaform study included one cohort injected with the animal-based HA wrinkle treatment and a second with Zyplast. Genzyme notes in a summary document that an ideal control would have been a placebo, but that non-biased assessments would be impossible with this approach.

Panelists generally agreed that unlike the Restylane trial, Genzyme's study largely avoided biased evaluations. Blumenstein, however, chided his peers for recommending Hylaform approval in a 6-3 vote. Hylaform is non-inferior to Zyplast, he acknowledged, but he also contended that Zyplast's efficacy was unproven. Trial data did not venture to show effectiveness, he charged.

"What we're doing here today is stunning to me," said Blumenstein, who dissented on both the Restylane and Hylaform approval votes. Other dissenters for both products were surgeon Joseph LoCicero, MD, University of Alabama, and Joseph Boykin, MD, Wound Healing Center, Medical College of Virginia.

Study design problems also identified with Hylaform were exclusion of avian protein-allergic patients and insufficient data on non-white populations, a concern also raised with Restylane. Both flaws were addressed in conditions (¹ Also see "Gen/Surg Panel Okays Dermal Fillers 6-3; Minority Patient Studies Proposed" - Medtech Insight, 24 November, 2003., p. 6).

Panelist John Halsey, PhD, president and clinical laboratory director of IBT Reference Lab/ProGene Biomedical, offered further critiques of the Hylaform study design. Limited to a 12-week follow-up, the trial was too short in duration, he maintained. Members also questioned why for both trials the sole injection site was the nasolabial fold. Trial investigators maintain it is the most difficult wrinkle to treat.

Facial Implant Panel Endorsement Masks Study Bias Concerns

More from Archive

More from Medtech Insight