Patient-Reported Outcomes Data Earned Kybella A Claim For Emotional Impacts

FDA concluded results from a patient-reported outcomes measure could support a labeling claim for Kythera Biopharmaceuticals Inc.'s Kybella (deoxycholic acid) despite initial doubts about the instrument's ability to measure the emotional impacts of submental fat.

The Patient-Reported Submental Fat Impact Scale (PR-SMFIS) is a six-item questionnaire designed to evaluate the visual and emotional impacts of submental fullness (see box).

In the pivotal trials for Kybella, patients were given a copy of the instrument and a standard mirror was made available to assist in the self-rating. Patients were asked to rate each item in the questionnaire on an 11-point, horizontal scale ranging from zero (no impact/"not at all") to 10 (extreme impact/"extremely").

The PR-SMFIS was incorporated as a key secondary endpoint in the Phase III trials even though its inclusion was not agreed to by FDA as part of the Special Protocol Assessment (see related story, (Also see "Kybella Approval Shows How To Work With FDA On Clinical Outcome Assessments" - Pink Sheet, 21 Sep, 2015.)).

The instrument was developed based upon literature reviews, interviews of clinicians and patients with submental fat, and patient characterizations of the impacts of submental fat convexity or fullness on how it made them look or feel. The proposed PR-SMFIS items were evaluated for readability and meaning through cognitive debriefing interviews with 15 patients.

Kythera asserted that a three-point change from baseline on the PR-SMFIS total score constituted a meaningful treatment benefit.

During an April 2011 end-of-Phase II meeting, FDA questioned whether the items in the instrument "are adequately defined to allow a description of the scale’s results," according to meeting minutes.

"For example, when responding to the item, 'How much older do you look because of your chin fat?' patients may be evaluating different concepts related to looking older depending on the patient’s age and other factors such as body image," FDA said. "Therefore, we view the PR-SMFIS as an exploratory outcome assessment."

The agency also raised concerns about the contrary direction of the first question compared to the subsequent five.

Kythera proposed "the analysis be carried out by averaging the scores from the six questions," the meeting minutes states. "Please note that the first question is expressed in a different direction from the rest of the questions and may cause difficulties for interpretation."

FDA's concerns appear to have led to a change in how the PR-SMFIS total score is calculated.

In a Jan. 8 memo, study endpoints reviewer Sarrit Kovacs said that "because the first item’s response options are in the opposite direction from the other scores (i.e., for PR-SMFIS Item 1 (happy), a higher score characterizes a 'better' outcome), it is calculated as the sum of 10 minus the first item score, plus the scores of the other five items, divided by six. A higher PR-SMFIS Total Scale score represents a greater overall negative impact on patient self-perceptions due to SMF."

Total Score Not Misleading

Kovacs concluded the PR-SMFIS total score "is appropriate to support a claim in labeling as an assessment of the impact of treatment on how patients feel about their chin fat appearance."

The instrument was developed in alignment with the agency's 2009 final guidance on patient-reported outcome instruments to support labeling claims, and its measurement properties are adequate, Kovacs said.

There is concern that some of the instrument's individual questions, such as those about looking old or overweight, may be misleading if taken out of context, the reviewer said. "However, the total PR-SMFIS score does not appear to be misleading given that all six items' change scores from baseline showed improvement with a similar magnitude of change."

Detailed results on the PR-SMFIS secondary endpoint are not included in labeling.

Rather, labeling reflects the results with the following statement: "The overall patient-reported satisfaction and self-perceived visual attributes showed greater improvement in the Kybella group than in the placebo group."