Recessive dystrophic epidermolysis bullosa (RDEB) is an incurable, devastating, inherited skin disease caused by mutations in the COL7A1 gene that encodes for type VII collagen (C7), the major component of anchoring fibrils (AFs), structures that mediate epidermal-dermal adherence. Thirty percent of RDEB patients have nonsense mutations. The investigators recently demonstrated in 5 such patients that intradermal and topical gentamicin induced "read-through" of their nonsense mutations and created robust and sustained new C7 and AFs at the dermal-epidermal junction (DEJ) of their skin and also stimulated wound closure and reduced new blister formation. No untoward side effects occurred. Herein, the investigators propose evaluating the safety and efficacy of intravenous gentamicin in these patients. In theory, this intravenous administration has the possibility of treating simultaneously all of the patients' skin wounds. The investigators also propose optimizing the concentration and manner of delivery of topical gentamicin. The unambiguous milestones will be increased C7 and AFs in the patients' DEJ, improved EB Disease Activity Scores, and absence of significant gentamicin side effects.

Aim 1. Evaluate the Safety and Efficacy of Short Term Intravenous Gentamicin for RDEB Patients.

Overview & Rationale: The rationale for short-term systemic intravenous gentamicin in these patients is outlined in the sections above, namely the drug could treat multiple skin sites simultaneously including the esophagus. This approach has proven safe and efficacious in other genetic diseases such as CF and DMD22-25 Our HYPOTHESIS is that intravenous gentamicin will, like topical and intradermal gentamicin, create new C7 and AFs at the patient's DEJ, improve multiple skin sites simultaneously, improve the patient's clinical disease and improve his or her quality of life. In accordance with the enclosed letters from two experienced pediatricians, we believe this therapy will be safe. Because C7 and AFs are incredibly stable, we further hypothesize that newly induced C7 and AFs will persist for many months in the patient's skin. Using dosing regimens of gentamicin that have proven to be safe, we will: 1) determine the percentage increase of C7 expression in the patients' skin before and after treatment, 2) determine how durable and sustainable the new C7 and AFs are in the patients' skin after treatment, 3) determine if this treatment is safe in this patient population, 4) determine if this treatment improves the patients clinically and improves their quality of life and 5) generate data for a multi-centered Phase III study using a larger number of patients.

Patients/Intervention: We have already lined up 3 well-characterized, adult RDEB patients with nonsense mutations who wish to participate in this study. These patients had positive responses to topical and intradermal gentamicin in our clinical trial (Table II). The patients will receive intravenous gentamicin (7.5 mgs per kilogram), in 100 ml of 5% dextrose in saline, for 14 consecutive days, a dose proven to not cause ototoxicity or neprhotoxicity28 and to be efficacious in inducing PTC read-through in DMD patients.22 The first few infusions will be done in the Infusion Center of the Keck Hospital of USC. We will measure gentamicin trough levels (24 hours after infusion) and gentamicin peak levels (30 minutes after the infusion) during the second day and, if needed, adjust the gentamicin doses such that the peak level is between 20 and 40 μg/ml and the trough level <2 μg/ml. If no untoward side effects occur, the patients will then have their infusions in their home via a commercial infusion service.

Safety Evaluation: Patients will be consented, enrolled and evaluated on Day 0 and then evaluated on post-treatment days 30, 90, and 180. On Day 0, the patients will have a complete medical history, review of systems (ROS), physical examination, vital signs, and weight. They will also have baseline urine analysis, complete blood count (CBC), electrolytes, glucose, comprehensive metabolic panel (CMP), and calculated creatinine clearance. Pure tone audiometry and a calculated creatinine clearance will be done on Days 0, 7, 14 and 30 post-treatment. Stop criteria will be a decline of >15 dB on pure tone audiometry at 2 consecutive frequencies and a creatinine clearance <60ml/min.28 Whenever there is new protein created in a patient, there is the possibility that the patient's immune system will raise autoantibodies against it. This did not occur in our clinical trial with topical and intradermal gentamicin, but it behooves us to evaluate this again with intravenous gentamicin and potentially more auto-antigenic protein created. Using indirect immunofluorescence and our anti-C7 ELISA, we will test the subject's serum for anti-C7 antibodies at each visit.

Efficacy: A. Clinical Evaluations: At baseline Day 0 and at each follow-up visit, (days 30, 90 and 180) the patients will complete a Quality of Life Assessment tool, a Pruritus Score tool, and a validated epidermolysis bullosa disease activity index (EBDASI)29,30. Patients will keep a daily diary using a self-ROS questionnaire. Patients will be telephoned weekly by a USC study member inquiring about any new signs, symptoms or ROS changes since the advent of the study. At baseline Day 0, we will identify four Test Sites at least 25 cm2 in size to follow sequentially. Two will be from areas that at baseline have blisters or erosions and the other two from intact skin areas. The two Test Sites with active blisters and erosions will be traced using a transparency sheet and the areas of the lesions calculated on a digitizing tablet. Likewise, all of the Test Sites will have standardized photographs and the lesional areas calculated by computer-assisted morphometry, as previously published.29-31 Within the context of their daily diary, patients will note the number of new blisters and erosions that occur in the Test Sites. A USC study member will telephone the patient each week and inquire about the Test Sites and ask the patient to estimate the percentage area of the Test Site consumed by erosions and blisters. Also, at each follow-up visit, three dermatologists will observe the four Test Sites and count the number of bullae and erosions in each. The intact Test Site areas will be scored at each follow-up visit using the following 3-point scale: 0 = no lesions; +1 = any active lesions; +2 = active lesions consuming over 50% of the Test Site area. Tracings to evaluate the areas of blisters or open erosions will be done, and standardized photographs paired with computer-assisted morphometry to calculate the areas of any lesions will be performed. B. Evaluations of C7 and AF Expression: At each post-treatment time point, 8 mm punch biopsies from the four Test Sites will be divided into three parts for (i) H&E histology and quantitation of epidermal-dermal adherence, (ii) the expression of C7 at the DEJ relative to that of normal human skin using anti-C7 antibodies to the NC1 and NC2 domains of C7 and NIH J Image software. One major "milestone" of the study will be the expression of NC2 since this would clearly indicate restoration of full-length C7 and (III) evaluation of newly created AFs by immuno-EM and AF enumeration by computer-assisted morphometry. Performing these evaluations out to 6 months will determine how durable the newly created C7 and AFs are. In our recent clinical trial outlined above, it was surprising that gentamicin-induced C7 remained at a similar level at 3 months post-treatment as that of 1 month for Patient #3 (Please see Figure 1 and Table II).

Aim 2: Evaluate the Safety and Efficacy of Higher Dose Topical Gentamicin for RDEB Patients.

Overview and Rationale: In our gentamicin clinical trial outlined earlier, there was significant variation in the production of new C7 and AFs, as shown in Table II. Patients #2 and #5 only restored C7 to 20% of normal skin in response to gentamicin. We only used one concentration of gentamicin, 0.1% ointment, and our in vitro studies with RDEB fibroblasts from these same two patients showed that the C7 response to gentamicin was dose-dependent.26 In these patients, there could have been sub-optimal dosing or sub-optimal access of the gentamicin to the patient's dermal fibroblasts and basal keratinocytes. In this aim, we will use 0.5% gentamicin ointment rather than 0.1% and determine if the C7 response is more robust in these two patients. Based on our encouraging data shown in Figure 2, we will also treat intact skin and determine if higher dose gentamicin can prophylactically prevent frank skin blisters and erosions from forming.

Patients/Intervention: We will study two RDEB children tested in our previous trial for this study. In addition, we also have 2 other children and 2 other adult patients who have contacted us and wish to participate in this study (Table III). Two of these patients share the same mutations (R578X and R2814) as our current patients. We will select a total of four Test Sites (two with open wounds of at least 3 cm by 3 cm and two areas of intact skin nearby the Test Sites with blisters or erosions) and apply 0.5% gentamicin ointment two times a day under an occlusive dressing of the patients' choice for two weeks. This dose was used in combination with 15% paromomycin to ulcerative cutaneous leishmaniasis lesions once daily for 20 days in two publications without renal or ototoxicity detected.32,33 Safety and Efficacy: We will essentially assess in these patients the safety and efficacy parameters outlined in Aim #1 at months 1, 3 and 6 after topical application. Potential renal toxicity and ototoxicity will be monitored by creatinine clearance and pure tone audiometry at day 0 and at each follow up visit.

Aim 3: Evaluate the Safety and Efficacy of Intradermal Administration of Gentamicin with a MR2 Micro-needle Roller Device.

Overview and Rationale: In our current clinical trial outlined above, there were excellent C7 responses to intradermal injections of gentamicin solution in several patients (Table II). Nevertheless, these injections were very localized with diffusion of the drug probably less than an inch from the single injection site. Recent evidence has shown that a MR2 microneedle roller device can readily deliver lidocaine into the lower epidermis and upper papillary dermis over a wide area of skin with good efficacy, safety and excellent patient tolerance.34 This may be an opportunity to deliver intradermal gentamicin solution (40 mg/ml) to wide areas of skin in a simple manner that can be done in an outpatient office with minimal logistical issues and avoidance of systemic exposure to the drug. Advantages include the possibility of delivering in a "pulse " fashion significant levels of gentamicin to the papillary dermis over wide areas of skin with minimal discomfort and minimal systemic exposure of gentamicin. Avoiding systemic exposure would be valuable for RDEB patients who have ototoxicity or renal compromise, poor venous access or cannot tolerate intravenous infusions.

Patients & Interventions: We will enroll two new RDEB patients (either adults or children) shown in Table III in this "proof-of-principle" study. Skin Test Sites will be areas measuring at least 40 cm2, and we will test one site of intact unwounded skin and one site of skin with lesions (blisters or erosions). The areas will be cleansed with 4% Hibiclens and then the MR2 Microneedle Roller Device passed over the area in two directions at 90 degrees. This device induces tiny micro-wounds in the skin to the depth of the papillary dermis. Gentamicin sterile solution (40 mg/ml) will then be liberally applied over the area and the area immediately occluded with a polyrurethane membrane (Opsite). Another equal sized Test Site of intact skin will be tested without prior intervention with the Microneedle Roller. The same gentamicin solution will be applied to the site and immediately occluded with Opsite. This will determine if the gentamicin solution can penetrate intact skin.

Safety and Efficacy: We will measure gentamicin in the serum of these patient 30 minutes after the microneedle delivery to determine if there is any systemic exposure when gentamicin is administered in this manner. Efficacy of the gentamicin administration will be assessed identically to those in Aim#1 by IF and immuno-EM analysis in conjunction with the patients keeping a daily diary of the treatment sites, weekly photographs of Test Sites, weekly staff interviews of the patients, and physician assessments at patient visits at months 1, 3 and 6 after administration of gentamicin. Potential renal toxicity and ototoxicity will be monitored by creatinine clearance and pure tone audiometry at day 0 and each follow up visit.

Inclusion Criteria:

(i) RDEB patients with a nonsense mutation in COL7A1 in either one or two alleles (ii) An absence or decrease in C7 expression at their DEJ when compared to that of normal human skin.

Exclusion Criteria:

(i) Pre-existing renal or auditory impairment (ii) Allergies to aminoglycosides or sulfate compounds (iii) Pregnancy (iv) Exposure to gentamicin within the past 6 weeks.

6 months

Small number of subjects analyzed due to rarity of the disease and experimental plan.