Testing times for epidermolysis bullosa topical therapies

ESSENCE and allantoin

So what went wrong in the phase 3 ESSENCE trial with allantoin, which was halted early in September 2017? The trial had included 169 patients with any type of EB – simplex, recessive dystrophic, and junctional non-Herlitz – who were randomized to treatment with the allantoin-containing cream SD-101 or a placebo cream containing only the vehicle. The creams were applied daily to the entire body for 3 months, with the primary endpoint being total wound closure at the end of the treatment period. Total wound closure was a requirement of the Food and Drug Administration, Dr. Murrell said, but it is now known that 100% closure is not always likely, which the agency itself now concedes.

“Most disappointingly, no significant difference was found [between the study drug and placebo], therefore it didn’t meet the primary endpoint, and you’re not even allowed to consider secondary endpoints – those are the rules of the game,” she said. As a result, the trial was stopped in 2017.

For inclusion in the study, patients had to have at least one target wound that had been present for at least 3 weeks, but there was no stratification on the duration of wounds in the randomization process. That meant that some individuals with wounds of shorter duration had unintentionally ended up in the placebo arm – favoring healing – and those with more chronic wounds had been in the allantoin arm. So, because the study arms might not have been equally balanced at baseline, it would have been harder for the actual treatment to demonstrate a benefit, Dr. Murrell suggested.

Another problem with the trial was that the vehicle cream contained elements, such as lanolin, already associated with wound healing. That would have given patients in the placebo arm an advantage because anyone applying the cream every day would probably get better or improve to some degree.

The patients were also required to have daily dressing changes and baths and, “if you give any patient that advice and they comply with it for a period of time, they are going to improve,” whether or not they are applying the study drug. Dr. Murrell said that the researchers likely should have done a run-in period first and then established a new baseline to randomize the patients.

“Lastly, no one had ever done a study of what we essentially tell eczema patients to do every day … to moisturize, because that will provide extra protection and barrier to their skin. So, if anything, the ESSENCE study shows that moisturizing has a protective effect of the vehicle for patients with EB,” she said.
 

DELIVERS and diacerein

Another trial that was stopped prematurely was the phase 2 DELIVERS study, which was set up to assess the benefits of topical diacerein in people with EB simplex. Diacerein, an extract of rhubarb root, was tested in 54 patients, who were randomized to apply either diacerein or vehicle ointment for 8 weeks.

Initially, the results “looked very promising,” Dr. Murrell said, because there was a trend toward improved EB simplex lesions, with the primary endpoint of at least a 60% reduction in lesions met by 57.1% of diacerein-treated and 53.8% of vehicle-treated patients.

However, the trial included use of the Investigator’s Global Assessment Scale at the FDA’s behest, but the tool had not been validated in previous EB trials, and which didn’t seem to show any benefit of the active over the placebo ointment. (The Investigator’s Global Assessment is a 5-point scale used for overall clinical assessment of severity of disease, ranging from 0 to 4, where a higher score denotes worse outcome.)In a poster presented separately at the meeting, the DELIVERS researchers noted that “the lack of statistical significance in the primary endpoint could be explained in part by milder disease in the diacerein group.” The mean body surface area of EB simplex lesions within the assessment area at baseline was 5.76% in the diacerein group and 7.13% in the vehicle group. The researchers proposed that perhaps a higher concentration of diacerein than the 1% used in the trial might have been needed.