2005D-0340 Guidance for Industry on Acne Vulgaris: Developing Drugs for Treatment
FDA Comment Number : EC5
Submitter : Dr. Todd Plott Date & Time: 12/21/2005 11:12:25
Organization : Medicis Pharmaceutical
Category : Private Industry
Issue Areas/Comments
The following comments are being submitted electronically in four parts. The section below is part 4 of 4.

13. The IGA scale is inaccurate and insensitive in skin of color.

The IGA will be confounded in dark-skinned patients in two ways: a) post-inflammatory pigmentary changes (PIH) can be confused with ongoing disease although the underlying lesions have successfully resolved on test therapy; and b) lesions cannot be readily detected in patients with Fitzpatrick Skin Type VI because the erythema of inflammatory lesions may not be evident visually. In any event, validation of the successful use of any such IGA scale in this patient subset should be accomplished before any form of such a scale is made mandatory.

14. The IGA scale is confounded by baseline severity

Ideally, an efficacy scale should provide near linear grades of improvement so that improvement from lower or worse disease is measured equally to improvements from milder disease. The dichotomous nature of the scale does not allow for accurate measurements of clinical improvement of severe disease as compared to measurement of improvements in milder disease. In a presentation by Dr. Alosh at the Advisory Panel it was stated that severe patients had a much lower chance of achieving success than moderate or mild patients. First, it is unclear whether requiring 2-grades of improvement as the definition of success for mild patients is helpful, since we do not know if improvement from severe to mild is similar to improvement from moderate to almost clear or, similarly, from mild to clear. This requires prospective assessment with a validated scale and use of an agent known to be effective as a test agent. If the IGA scale, as is believed, is non-linear, then use of `clear or almost clear? or `2-grades of improvement? may not correct the bias introduced by overpopulating a study with moderate or mild patients (as compared to a randomly picked population). The ability to overpopulate a study with one type of disease severity and, thereby affect rate of success means that the scale is susceptible to gaming by investigators or study sponsors. For instance, an investigator can acquire a reputation for selectively including more mild patients and thus attract more sponsor study contracts.

15. The Agency?s expressed view that acne is a cosmetic disease is not supported by the clinical or patient community.

One rationale for the scale is that if a patient is not visually improved, then the drug product is ineffective because the disease is cosmetic only. However, even non-inflammatory lesions can result much later in pitting scars or keloid formation. Thus, improvement in lesion counts provides a clinical benefit that does not correspond to what might be missed in a visual IGA examination.