EB is a rare debilitating genetic dermatologic disorder characterized by skin fragility, leading to blistering and wounding; just wearing normal clothing can lead to wound formation. In some cases, EB also leads to the erosion of the epithelial lining of other organs. To give a sense of how severe the disease can be, there was a documentary about a child with EB entitled The boy whose skin fell off. There are several variations of EB (see Exhibit 6) but all share the problem of painful blistering and wounding at the slightest friction. Prevalence is estimated at 11.07 per million, which would indicate approximately 3,600 patients in the US and 5,700 in the EU.
Exhibit 6: EB variations and the genetic defects that cause them
EB type |
% of EB population |
Genetic defect |
Type of defect |
Defective protein |
EB simplex |
55% |
K5 |
Autosomal dominance |
Keratin-5 |
K14 |
Autosomal recessive, autosomal dominance |
Keratin-14 |
TGM5, DSP, PKP1, PLEC, DST, ITGA6, ITGB4, COL17A1 |
Autosomal recessive |
Transglutaminase 5, desmoplakin, plakophilin-1, plectin, α6β4 integrin, type XVII collagen |
JUP |
Autosomal recessive, autosomal dominance |
Plakoglobin |
EB junctional |
5% |
LAMA3 (9% of cases) |
Autosomal recessive |
Laminin-332, type XVII collagen, α6β4 integrin |
LAMB3 (70% of cases) |
LAMC2 (9% of cases) |
COL17A1 (10% of cases) |
ITGA6, ITGB4 |
EB dystrophic |
30% |
COL7A1 |
Autosomal recessive or autosomal dominance |
Type VII collagen |
EB Kindler type |
Rare |
FERMT1 |
Autosomal recessive |
Kindlin-1 |
As there are currently no therapies to treat the underlying causes of EB, treatment is based on promoting wound healing. EB patients often have large areas of their body in need of care and can take 30 minutes to 3–4 hours a day (seven hours in an extreme example) just changing their dressings. According to the Dystrophic Epidermolysis Bullosa Research Association, bandages and other necessary supplies can have a retail cost of over $10,000 per month.
Exhibit 7: Wound dressings typically used in EB patients
Type of dressing |
Comments |
Honey |
Honey is an antimicrobial agent but can be very sticky and may sting. |
Silicone |
Soft and adhere easily but may move or buckle and can be expensive. |
Foam |
Encourage healing by absorbing liquids oozing from the wound but tends to cause overheating. |
Alginates |
Long-lasting dressing that becomes a gel in the presence of liquid oozing from the wound but is difficult to remove, stings and can cause damage to surrounding skin. |
Hydrocolloid |
Provides a moist environment for healing but may be difficult to remove and can be problematic for open wounds. |
Charcoal |
Controls odour but can lose effectiveness. |
Eclipse |
Highly absorbent and thick but difficulties in conforming to body parts, causes blistering and sweatiness. |
Type of dressing |
Honey |
Silicone |
Foam |
Alginates |
Hydrocolloid |
Charcoal |
Eclipse |
Comments |
Honey is an antimicrobial agent but can be very sticky and may sting. |
Soft and adhere easily but may move or buckle and can be expensive. |
Encourage healing by absorbing liquids oozing from the wound but tends to cause overheating. |
Long-lasting dressing that becomes a gel in the presence of liquid oozing from the wound but is difficult to remove, stings and can cause damage to surrounding skin. |
Provides a moist environment for healing but may be difficult to remove and can be problematic for open wounds. |
Controls odour but can lose effectiveness. |
Highly absorbent and thick but difficulties in conforming to body parts, causes blistering and sweatiness. |
Source: Grocott et al., Living in dressings and bandages: findings from workshops with people with Epidermolysis bullosa. International Wound Journal. 2013; 10:274
The biopharmaceutical industry has found it difficult to develop therapies for EB. Most recently, in September 2017, SD-101 from Amicus Therapeutics failed in a 169-patient Phase III study. SD-101 was a topical 6% formulation of allantoin, a common ingredient in over-the-counter cosmetics at lower concentrations due to its moisturizing effect. Amicus acquired the drug through its acquisition of Scioderm in September 2015 for $229m in cash and stock, an additional $361m in clinical and regulatory milestones and a further $257m in sales milestones, for a total consideration of $847m. Additionally if a priority review voucher (PRV) was awarded for SD-101, 50% of the PRV’s value would have had to be transferred to Scioderm’s shareholders (a PRV had been sold in 2017 for $110–130m).
The acquisition was based on data from 45 patients in a Phase IIb trial in which SD-101 had a statistically significant benefit in the proportion of patients with complete target wound closure at the two-month time point (82% in the SD-101 6% concentration versus 41% placebo, p=0.04). However, at the one-month and three-month time points the results were not significant. In the 169-patient Phase III ESSENCE study there was practically no difference in one primary endpoint (time to target wound closure within three months, p=0.985) and a trend towards placebo in the other (percentage of target wound closure by month three, 49% SD-101 vs 54% placebo).
Current products in later stages include AP101 from Amryt and diacerein from Castle Creek (see Exhibit 8). AP101 is in a 164-patient Phase III trial, which is expected to read out by the end of this year. It is a mixture of birch bark extract and sunflower oil that works by stimulating keratinocyte migration and differentiation into mature epithelial cells, promoting wound healing. Previous data are limited as the prior Phase II was conducted in 10 patients where wounds in essence were their own controls in that one half of the wound was treated with AP101 and the other half was not. Improvement in wound epithelialization at days seven and 14 was trending in the right direction but not significant. The primary endpoint is the proportion of patients with first complete closure of the target wound within 45 days of treatment, which is an endpoint we do not have data on for this drug.
Exhibit 8: Competitive landscape in EB
Company |
Drug |
Composition |
Mechanism |
Phase |
Comments |
Amryt |
AP101 (Oleogel-S10) |
10% birch bark extract in 90% sunflower oil |
Causes keratinocytes (cells that regenerate outer skin layer) to migrate and differentiate into mature epithelial skin cells, promoting wound healing. |
Phase III |
In a 164-pt Phase III trial, which is expected to complete in Q318. Primary endpoint is proportion of patients with first complete closure of the EB target wound within 45 days of treatment. Prior Phase II data in 10 patients showed 69.7% wound epithelialization at day seven versus 57.4% placebo (p=0.21) and 87.7% wound epithelialization at day 14 versus 79.2% placebo (p=0.33). |
Castle Creek |
Diacerein 1% |
A prodrug of the IL-1 converting enzyme inhibitor rhein which is approved for the systemic treatment of osteoarthritis |
Suppresses interleukin-1 beta, which is believed to reduce keratin 14 and stabilize the intermediate filament network of basal keratinocytes. |
Phase II |
In an 80-pt Phase II trial that is expected to complete in Q418. Primary endpoint is proportion of subjects who achieve a greater than 40% reduction in body surface area of lesions from baseline to week 16. In a pilot study in five patients the number of blisters was reduced significantly by 78% in the left armpit and 66% in the right in the Phase I portion. In the Phase II portion there was no loss of efficacy in those patients who had previously received diacerein, so there was no significant difference between the two arms compared to the end of the Phase I portion. |
InMed |
INM-750 |
Proprietary formulation of a combination of two cannabinoids |
Upregulates K15. |
Preclinical |
Expected to file an IND in H219. |
Source: InMed, Amryt, Castle Creek, Clinicaltrials.gov, Wally et al. Topical diacerein for epidermolysis bullosa: a randomized controlled pilot study. Orphanet Journal of Rare Diseases 2013, 8:69
Castle Creek is developing a topical 1% formulation of diacerein, currently approved as an oral version in certain EU and Asian companies for the treatment of osteoarthritis. It is supposed to suppress interleukin-1 beta, which may then reduce keratin 14 and stabilize the intermediate filament network of basal keratinocytes. Again, data so far have been limited, as we only have results from a pilot study of five patients. In the Phase I portion where all patients received diacerein, blister counts in armpits were reduced by 78% in the right armpit and then 66% in the left. In the Phase II portion, left armpits were given placebo but there was no little or no loss of efficacy so the study was not able to discern a statistically significant benefit for diacerein over placebo.
INM-750 is a combination of two undisclosed cannabinoids and may be able to help EB patients in several ways. According to management, based on preclinical research findings, INM-750 may significantly upregulate the keratin K15, which may be able to compensate for a malfunctioning K14 and combine with K5 to form the necessary adhesion between the epidermis and dermis, potentially reversing the underlying cause of EB simplex in some patients. There is also evidence of increase in the level of E-cadherin, a major component of epithelium integrity and MCP-1, which plays a key role in wound healing. Also, not surprisingly as cannabinoids are known to have a mild impact on pain, INM-750 has demonstrated a positive impact in Nerve growth factor (NGF) induced pain models in rats. In addition, there is evidence of antibacterial activity among cannabinoids, which could reduce the infection risk. In sum, INM-750 may be able to treat the many troublesome symptoms as well as the underlying cause of the disease in patients with K14 defects.
The company expects to complete its IND-enabling toxicology studies next year and file the IND around H219, with the initiation of a Phase I in healthy volunteers to proceed thereafter. Recognizing its early stage, we project a 2026 launch and a 5% probability of success, our standard probability of success for a preclinical product (which would increase to 10–20% as INM-750 advances into the clinic). As EB is an orphan indication, we would expect INM-750 to be priced at a premium: $100,000 per year in the US and $50,000 per year in the EU, based on, but still allowing for, a discount to typical orphan pricing for indications that are of similar size to EB of $300,000–500,000 per year. The market share will ultimately be determined by the quality of the data and benefit to patients but we estimate 16.5% market share for EB, as the underlying disease may be treated in the subset of EB simplex patients with K14 defects while the rest would receive only symptomatic relief. Also, as EB is an especially difficult condition to treat, positive data in EB may indicate applications in other wound healing-related indications, although we do not model any of these currently. Using these preliminary estimates, peak sales could achieve C$345m per year. As this is an orphan indication that would not require a large sales force, the company expects to market INM-750 for EB itself. We model out to 2037 although that will likely be extended as the company files to patent the formulation once it is finalized (the company has indicated multiple patent filings will occur later in 2018).