Topical Gene Therapy Heals Blistering Skin Disease Lesions

A novel topical gene therapy healed most lesions among patients with dystrophic epidermolysis bullosa, a rare genetic blistering skin disease, the GEM-3 trial showed.

Topical administration of beremagene geperpavec (B-VEC) led to complete healing of 67% of treated wounds as compared with 22% of those exposed to placebo at 6 months (P=0.002), reported M. Peter Marinkovich, MD, of Stanford University in California, and colleagues in the New England Journal of Medicine.

That's a "dramatic result" for a painful and disfiguring condition with no approved therapies, noted David V. Schaffer, PhD, of the University of California Berkeley, in an accompanying editorial.

Oleogel-S10, another novel topical therapy, which is derived from birch tree bark, had previously shown promise in patients with epidermolysis bullosa, with complete closure of target wounds in 41% within 45 days versus 29% given placebo in a phase III trial.

GEM-3's "positive results bode well not only for dystrophic epidermolysis bullosa, a devastating disorder for patients, but potentially for other forms of epidermolysis bullosa and other genetic diseases of the skin," Schaffer wrote.

The Science

Gene therapy with adeno-associated virus (AAV) vectors to deliver a working copy of at-fault genes has worked in a number of conditions. However, like a number of other skin disorders, the gene responsible for dystrophic epidermolysis -- COL7A1 -- is just too large for an AAV to carry.

B-VEC uses a herpes simplex virus type 1 (HSV-1) vector modified to be unable to replicate. Not only is this kind of vector able to carry much larger gene inserts, it doesn't insert into the chromosome in the host cell, avoiding risk of mutagenesis that could lead to cancer.

"Once inside the cell, viral DNA is transcribed and translated into C7, which is secreted into the extracellular space to assemble into anchoring fibrils," explained Aimee S. Payne, MD, PhD, of the University of Pennsylvania in Philadelphia, in a "science behind the study" editorial.

That payload of functional type VII collagen protein helps the skin keep the epidermis attached to the dermis instead of sheering into blisters that may get infected and create fibrosis that ends up fusing together fingers and toes, for example.

That the viral vector doesn't permanently integrate into host DNA, while potentially making the approach safer, also means it gets diluted with each cell division and thus needs repeated administration, Payne noted.

Durability

The big question is durability, Schaffer suggested. That 50% of treated wounds were completely closed at both 3 months and 6 months versus 7% of same-patient wounds of similar size and location administered placebo might "offer the potential for reasonable durability," he wrote.

And less chronic blistering and scarring might mean lower risk of the squamous cell carcinomas that these lead to in patients with dystrophic epidermolysis bullosa. "Although this has not been formally shown for this treatment, none of the three cases of squamous cell carcinoma that were reported during the trial occurred at wound sites that were exposed to the active drug or placebo," Schaffer pointed out.

"Furthermore, a greater reduction in pain intensity during wound-dressing change occurred with the active drug than with placebo, an effect that could become more pronounced as a greater number of wounds per patient are treated," he added.

The phase III trial included 31 patients -- children ages 6 months and older and adults (cohort median age 16 years) -- with genetically confirmed dystrophic epidermolysis bullosa who were seen at three sites in the U.S.

They were randomized to double-blind application of either B-VEC or placebo each week for 26 weeks on pairs of wounds matched according to size, region, and appearance. The sites selected couldn't have current evidence or history of squamous cell carcinoma or active infection. The treatment was given on an outpatient basis, as B-VEC consists of a cryopreserved thawed drug product mixed with methylcellulose gel. Placebo comprised the excipient gel mixed with normal saline.

Among secondary outcomes, B-VEC reduced pain during wound-dressing changes (-0.61 vs placebo from baseline to week 22 on a visual analog scale, 95% CI -1.10 to -0.13). Results were similar at weeks 24 and 26.

While adverse events were common overall, none of the five serious adverse events that occurred in three patients were considered to be related to study medication, although one case of mild erythema was considered to be related to B-VEC. The most common adverse events were three cases each (10%) of pruritus, chills, and squamous cell carcinoma of the skin.

There are theoretical concerns about immunogenicity against HSV-1 and C7 leading to immune rejection of vector or vector-transduced cells and less effective subsequent treatment, Payne noted. "In addition, antibodies against C7 might indicate the development of a secondary autoimmune blistering disorder known as epidermolysis bullosa acquisita, which typically warrants long-term immunosuppressive therapy."

But while almost all seropositive patients had an increase in antibody titers against HSV-1 and C7 and about 75% of the seronegative patients developed these antibodies on B-VEC therapy, there wasn't any association between immunogenicity and treatment failure.

"Longer and larger trials are warranted to determine the durability and side effects of B-VEC for this disease," Marinkovich and colleagues acknowledged.