Dusa Levulan Label Should Disclose Partial Responses, Advisory Panel Says

Dusa Levulan Kerastick labeling which includes a 75% partial lesion clearance response for the actinic keratoses treatment could benefit clinicians and patients, FDA's Dermatologic & Ophthalmic Drugs Advisory Committee decided Nov. 5.

"It gives someone a good confidence level of what the numbers are like when you do not have complete responses. I think that is a clinically meaningful outcome that 75% of all the lesions clear," committee consultant Philip Lavin, PhD, Boston Biostatistics Research Foundation, said.

Dusa received an "approvable" letter June 27 for Levulan (aminolevulinic acid) for use in combination with Blu-U blue light photodynamic therapy, pending agreement on labeling and reinspection of Dusa's drug manufacturer. The manufacturer has been rated "acceptable" after a Sept. 16-20 re-inspection.

Even though Levulan was already "approvable," under Sec. 404 of FDAMA an advisory committee review can be requested at any time should a scientific controversy arise "regarding an obligation." The Levulan meeting was held at FDA's request, Dusa said, adding that it was not held earlier in the approval process because of scheduling problems.

Levulan's "approvable" status was based on two vehicle-controlled pivotal trials with identical protocols that evaluated a total of 243 patients with four to 15 target AKs of the face or scalp. At the eight-week primary analysis, lesions that did not clear were re-treated, and follow-up was conducted at 12 weeks.

At week eight, 66% of patients had 100% complete response (total lesion clearance) and 77% of patients had a 75% or greater partial response. By week 12, 72% of patients had a complete response and 89% had a 75% or greater response, Dusa said. FDA's review focused on the complete response primary endpoint.

The committee agreed that the 75% response rate data was relevant, but questioned how much additional information would be provided given how difficult a compete response is to interpret.

Labeling should caution against use in AKs that are very thick and/or hyperkeratotic since they were excluded from clinical trials, the committee said.

In addition, efficacy differences in subgroups should be noted, the committee suggested. "Until there is good data to the contrary...there should be an exclusion or a warning about the lack of proven efficacy both in the scalp and for hyperkeratotic lesions," committee member Robert Stern, MD, Beth Israel Deaconness Medical Center, said. "That would perhaps encourage the sponsor to do studies that give us proven data that would allow us to modify the label in the future," he added.

While Levulan's proposed indication is for the treatment of AK of both the face and scalp, Phase III clinical data reveals greater efficacy for facial lesions. In the pivotal trials, 68% of facial lesions had 100% complete response rate by week eight, increasing to 75% by week 12. On the other hand, 55% of scalp lesions were 100% cleared at eight weeks, a figure that decreased to 48% by week 12, despite the fact that non-responders at week eight were re-treated.

Stern called the reduction in scalp efficacy between week eight and week 12, "prima facia evidence" that "if this stuff works on the scalp it doesn't appear in very many cases to work for very long."

A higher proportion of thinner lesions on the face may explain the greater efficacy for patients with face lesions, FDA medical officer Martin Okun, MD/PhD, explained. Thinner, grade one lesions responded to Levulan treatment more often (88%) than thicker, grade two lesions (78%). In the trials, 57% of facial lesions were grade one whereas 53% of scalp lesions were grade two. However, "adjusting to the difference of thicker lesions between the face and scalp can't explain these differences in efficacy," Stern concluded.

The committee repeatedly stressed the importance of long-term follow-up studies. It should be "emphasized that the data that we have is based on a three month study," committee consultant Joseph McGuire, MD, Stanford University, said.

"After eight weeks of therapy, of the 117 individuals who cleared, 14 have recurred by 12 weeks and one does not know if in another four weeks another 10 or 14 would recur....I do not see how one could come to the conclusion that one is achieving remissions with this therapy. Although that may very well be the case, but I think we need a longer window to look at the results," McGuire, the former committee chair, said.

Dusa has agreed to perform two Phase IV studies: a dermal allergenicity study and a trial to determine the long-term recurrence rate in at least 70 patients with a one-year follow-up.

Since Levulan may cause "oxidative damage to DNA," committee consultant John DiGiovanna, MD, National Institute of Arthritis and Muscoskeletal and Skin Diseases, said, "the adverse event that I would be interested in is the long-term development of malignancy in the areas that have been treated."

"Given the frequency of AK in skin cancer and the nature of this approach, I think that it is essential that a study be done [on] treated lesions, particularly those lesions that recurred in the area of treatment, to look for an increase in the incidence of development of skin cancer," DiGiovanna said.

There needs to be some histology on "unresponsive or quickly recurring lesions," Committee Chairman Lynn Drake, MD, University of Oklahoma, agreed.

The panel also discussed how labeling should be crafted in order to properly warn patients about exposure to solar or incandescent light.

Adverse events such as burning/stinging that occurred between drug application and device activation, were possibly the "result from inadvertent exposure of the target lesions to ambient light...thereby initiating a low grade photodynamic response," Okun explained.

During the baseline visit, patients were told to "avoid direct exposure of target sites to sunlight or other high intensity light sources, to wear light protective clothing and not to wash target lesions. The panel noted that a patient package insert explaining the risks would be helpful.

Consistent adverse responses were burning/stinging, (present in 100% of patients during and/or 24 hours after light treatment) and transient post-treatment lesion erythema and edema, Dusa said.

The committee expressed concern about Levulan on clinically "normal" skin, which Dusa did not test in trials. Although Levulan is the first product designed to spot-treat lesions, there is still "little selectivity" between AK and adjacent skin sites, FDA's Okun said.

With 4 mil. patient visits for AK per year, the U.S. AK market is $400 mil., Dusa reported. Initially, Dusa will target 5-fluorouracil cream patients and then go for nitrogen patients, Dusa CEO Geoffrey Shulman told analysts recently.