GEM-3: Phase 3 Safety and Immunogenicity Results of Beremagene Geperpavec (B-VEC), an Investigational, Topical Gene Therapy for Dystrophic Epidermolysis Bullosa (DEB) Presented at the Society for Investigative Dermatology (SID) 2022 Annual Meeting, May 18-21, 2022, Portland, Oregon • Dystrophic epidermolysis bullosa (DEB) is a serious, ultra-rare genetic blistering disease caused by mutations in the COL7A1 gene that lead to skin fragility and wounds1-3 – DEB affects ≈9000 people globally, including ≈3000 people in the United States and ≈3000 people in Europe4 – Patients with DEB require proactive management and care due to an increased risk of aggressive squamous cell carcinoma (SCC) and a wide range of other serious secondary complications, regardless of wound size or chronicity5-7 – Current management of DEB is limited to supportive care, such as ameliorating symptoms, palliative wound care, and managing secondary complications8,9 • Beremagene geperpavec (B-VEC) is an investigational herpes simplex virus type 1 (HSV-1)–based, topical, redosable gene therapy designed to restore type VII collagen (COL7) protein by delivering the COL7A1 gene10 (Figure 1) – B-VEC utilizes a differentiated HSV-1 vector platform that allows for episomal delivery, high payload capacity, tropism for the skin, and evasion of the immune system, enabling repeat delivery10 Introduction • GEM-3 (NCT04491604) is a phase 3, multicenter, double-blind, placebo-controlled intra-patient–randomized study evaluating the efficacy and safety of B-VEC in patients with DEB (Figure 2) • Serum samples before (screening or Week 1) and after (Week 26) B-VEC treatment were collected from patients and evaluated for anti-drug antibodies – Anti–HSV-1 specific antibodies were evaluated using a validated plaque reduction neutralization test (PRNT), which determines the percent reduction in B-VEC–mediated plaque formation in the presence of serially diluted patient sera (1:80 to 1:5120) and is reported as PRNT50 (defined as the serum dilution at which a ≈50% reduction in plaques is observed) – Immunoglobulin G antibodies against human COL7 were evaluated using a commercially available anti-COL7 enzyme-linked immunosorbent assay (EA 1947-4801 G, EUROIMMUN, Lübeck, Germany), which qualitatively determines anti-COL7 serostatus Methods M. Peter Marinkovich,1 Mercedes E. Gonzalez,2 Shireen V. Guide,3 I. Sinem Bagci,1 Surya Chitra,4 Brittani Agostini,5 Hubert Chen,5 Trevor Parry,5 Suma Krishnan5 1Stanford University, Redwood City, CA; 2University of Miami, Miami, FL; 3Mission Dermatology Center, Children’s Hospital of Orange County, University of California, Irvine, Department of Dermatology, Rancho Santa Margarita, CA; 4Savio Group Analytics & Statistics, Hockessin, DE; 5Krystal Biotech, Inc., Pittsburgh, PA Efficacy • The proportion of primary wounds with complete wound healing was significantly greater with B-VEC than placebo at both the 3- and 6-month timepoints (P<0.005; Figure 3) – In the patient with DDEB, the primary endpoint of complete wound healing at 6 months was achieved by the B-VEC–treated wound, but not by the placebo-treated wound • Pain and health-related quality of life assessments demonstrated improvement consistent with a wound healing response Table 1. Baseline demographics and clinical characteristics *In a small number of patients, the predefined threshold values for wound area by size category fell in between the size of the 2 wounds. B-VEC, beremagene geperpavec; DEB, dystrophic epidermolysis bullosa; SD, standard deviation. Figure 3. Proportion of primary wounds with complete healing at 6 and 3 months of B-VEC or placebo treatment Results Patient Disposition • The intent-to-treat (ITT) population, used for all primary and secondary efficacy analyses, included 31 randomized patients, each with a primary wound pair (Table 1) • The safety population, used for all safety analyses, was the same as the ITT population • Of 31 randomized patients, 3 withdrew from the study for nondrug-related reasons • B-VEC treatment demonstrated a durable and statistically significant improvement in complete wound healing at 3 and 6 months compared with placebo • B-VEC was generally well tolerated, with no treatment-related discontinuations • No clinically significant immunologic reactions were reported during the study • Treatment response to B-VEC was not associated with anti–HSV-1 serostatus at baseline or with anti-COL7 seroconversion • An ongoing open-label extension study (NCT04917874) is investigating the long-term efficacy and safety of B-VEC in patients with DEB, regardless of prior enrollment in GEM-3 Conclusions This study was funded by Krystal Biotech, Inc. Acknowledgments Disclosures 1. Harari S. Eur Respir Rev. 2016;25(140):101-103. 2. Fine J-D. JAMA Dermatol. 2016;152(11):1231-1238. 3. Has C, et al. Br J Dermatol. 2020;182(3):574-592. 4. Data on file, Krystal Biotech. 5. Has C, et al. Br J Dermatol. 2020;183(4):614-627. 6. Fine J-D, et al. J Am Acad Dermatol. 2009;60(2):203-211. 7. Rogers H, et al. J Am Acad Dermatol. 2021;85(3):741-743. 8. Bardhan A, et al. Nat Dis Rev Primers. 2020;6(1):78. 9. Bruckner AL, et al. Orphanet J Rare Dis. 2020;15(1):1. 10. Gurevich I, et al. Nat Med. 2022;28(4):780-788. 11. Xu F, et al. J Infect Dis. 2002;185(8):1019-1024. References MPM: Krystal Biotech (investigator), Abeona Therapeutics (investigator), CastleCreek (investigator), Phoenix Tissue Repair (investigator), WINGS Therapeutics (investigator). MEG: Dove/Unilever (consultant), Gerson Lehrman Group (consultant), Galderma USA (speaker), Krystal Biotech (investigator), Sanofi Genzyme (speaker), AbbVie (investigator), Primus Pharmaceuticals (speaker), Verrica Pharmaceuticals Inc (consultant), Regeneron Pharmaceuticals Inc (investigator), National Eczema Association (consultant), Dermira (investigator), Accuitis Inc (investigator), Innovaderm Research Inc (investigator), Pierre Fabre Dermatologie (consultant). SVG: Krystal Biotech (investigator), Innovaderm Research Inc (investigator), Arcutis Inc (investigator), Premier Research (investigator), Paidion (investigator), Castle Biosciences (investigator). ISB: Krystal Biotech (investigator), Castle Creek Biosciences (investigator). SC: Krystal Biotech (consultant), Paratek Pharmaceuticals (consultant). BA, HC, TP, SK: Krystal Biotech (employees and stockholders). †Maximum weekly dose (PFU/week) was based on patient's age and unit dose (PFU/wound) was determined based on wound area at baseline. *Complete wound healing defined as 100% wound closure from exact wound area at baseline, specified as skin re-epithelialization without drainage. B-VEC, beremagene geperpavec; COL7, type VII collagen; DEB, dystrophic epidermolysis bullosa; HSV-1, herpes simplex virus type 1; PFU, plaque-forming unit. Primary Efficacy Endpoint Secondary Efficacy Endpoints • Complete wound healing* at Weeks 8 & 10 or Weeks 10 & 12 (3 months) • Mean change in pain severity associated with wound dressing changes • Complete wound healing* at Weeks 22 & 24 or Weeks 24 & 26 (6 months) Safety Endpoints • Adverse events, vital signs, physical/skin examination, and laboratory evaluations • Pre- and post-treatment anti–HSV-1 and anti-COL7 antibodies 31 patients enrolled • Age ≥6 months • Genetically confirmed DEB • Two cutaneous primary wounds similar in size, appearance, and anatomical region • B-VEC dose based on patient age and wound size† Secondary Wounds Primary Wound Pair Randomized 1:1 Open-Label Extension Period Dosing continues after safety period 30-Day Safety Period B-VEC Once-weekly treatment† until wound closure; treatment resumed if wound reopened Placebo Once-weekly treatment† until wound closure; treatment resumed if wound reopened Open-Label B-VEC Remaining weekly dose† used to treat secondary wounds Figure 2. GEM-3 study design Patient demographics/characteristics Total patients (N=31) Age, years Mean (SD) 17.2 (10.7) Range 1-44 Age category, n (%) ≤12 years 10 (32.3) >12-≤18 years 9 (29.0) >18 years 12 (38.7) Male sex, n (%) 20 (64.5) Race, n (%) White 20 (64.5) Asian 6 (19.4) American Indian or Alaska native 5 (16.1) Genotype, n (%) Dominant DEB (DDEB) 1 (3.2) Recessive DEB (RDEB) 30 (96.8) Primary wound B-VEC (N=31) Placebo (N=31) Wound area by size, cm2 Mean (SD) 14.4 (12.7) 15.6 (12.1) Range 2.3-57.3 2.3-51.5 Wound area by size category*, n (%) <20 cm2 23 (74.2) 22 (71.0) 20-<40 cm2 6 (19.4) 8 (25.8) 40-60 cm2 2 (6.5) 1 (3.2) 0% 10% 20% 30% 40% 50% 60% 70% 80% 67.4% 21.6% 70.6% 19.7% 45.8% absolute difference (95% CI: 23.6-68.0) P<0.005 51.0% absolute difference (95% CI: 29.3-72.6) P<0.005 B-VEC Placebo Pr op or tio n of P rim ar y W ou nd s W ith C om pl et e W ou nd H ea lin g in th e IT T Po pu la tio n (N =3 1) 6 Months (Primary Endpoint) 3 Months (Key Secondary Endpoint) Data as of database lock on November 19, 2021; data in figure based on ITT population (imputed); P values and CIs are based on exact McNemar’s test. B-VEC, beremagene geperpavec; CI, confidence interval; ITT, intent to treat. Figure 1. B-VEC mechanism of action B-VEC, beremagene geperpavec; COL7, type VII collagen; COL7A1, collagen type VII alpha 1 chain gene; DEB, dystrophic epidermolysis bullosa. 1 B-VEC 3 COL7 transcripts COL7 protein Keratinocyte (or Fibroblast) Cell 4 B-VEC enters the compromised skin of patients with DEB and transduces both keratinocytes and fibroblasts Once in the nucleus of transduced cells, the vector genome is deposited (episomally) As a result, COL7A1 transcripts are generated, allowing the cell to produce and secrete functional COL7 protein The secreted COL7 protein assembles into anchoring fibrils that hold the epidermis and dermis together Anchoring fibrils 2 Nucleus Safety • The majority of adverse events (AEs) were mild or moderate; no AEs led to treatment discontinuation or death (Table 2) • One AE, mild erythema, was considered possibly related to study drug as assessed by the investigator • Three patients experienced a total of 5 serious AEs during the study: anemia (2 events) cellulitis, diarrhea, and positive blood culture (1 event each) – None were considered related to study drug • The most frequently reported AEs were pruritus, chills, and SCC (3 patients each) – All 3 reports of SCC occurred at sites that were not directly exposed to B-VEC or placebo and were deemed not related to study drug Total patients (N=31) Total number of AEs 45 Patients with ≥1 AE, n (%) 18 (58.1) Mild AE 15 (48.4) Moderate AE 3 (9.7) Severe AE 2 (6.5) Serious AE 3 (9.7) Drug-related AE 1 (3.2) AE leading to treatment discontinuation 0 (0) Death 0 (0) AEs reported in ≥5% of patients by System Organ Class and Preferred Term*, n (%) Skin and subcutaneous disorders Pruritus 3 (9.7) Erythema 2 (6.5) Rash 2 (6.5) General disorders and site conditions Chills 3 (9.7) Neoplasms benign, malignant, and unspecified Squamous cell carcinoma of the skin 3 (9.7) Respiratory, thoracic, and mediastinal disorders Cough 2 (6.5) Rhinorrhea 2 (6.5) Table 2. Safety summary *AEs were coded using MedDRA version 24.1. At each level of summarization, a patient was counted once if the patient reported ≥1 event. AE, adverse event; MedDRA, Medical Dictionary for Regulatory Activities. Antibody Responses • 22 of 31 patients (71.0%) provided a serum sample at baseline due to the difficulty of blood draws owing to skin fragility; 19 of the 22 patients (86.4%) also had matched serum samples at 6 months • At baseline, 14 of the 22 patients (63.6%) were anti–HSV-1 antibody seropositive and 8 were seronegative, in agreement with seropositivity rates of the general US population11; 6 of 8 (75.0%) baseline seronegative patients seroconverted by 6 months • For baseline seropositive patients, where quantitative differences at study completion could be calculated, antibody responses were variable (Figure 4), and none were determined to be meaningful, as defined by a >4-fold sustained increase in anti–HSV-1 antibody titer Figure 4. Anti–HSV-1 antibodies in pre- and post-treatment patient serum samples *Below limit of detection. Seroconversion defined as seronegative at baseline but seropositive at 6 months. HSV-1, herpes simplex virus type 1; PRNT, plaque reduction neutralization test; PRNT50, the serum dilution at which a ≈50% reduction in B-VEC–mediated plaques is observed. 0 1000 2000 3000 4000 5000 6000 1 12 26 1 26 1 26 1 26 1 26 1 26 1 26 1 26 1 14 26 1 26 1 26 1 26 1 26 1 26 1 26 1 26 1 26 1 5 26 1 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 PR N T5 0 (1 :X ) Patients who were seropositive at baseline Patients who seroconverted by 6 months Week Patient number * * * * * * * * • At baseline, 1 of 22 patients (4.5%) was positive for anti-COL7 antibodies • 13 of 18 patients (72.2%) with matched serum samples seroconverted by 6 months; no clinically significant immunologic reactions or differences in treatment response were seen • Post hoc analysis of response rates in primary wound pairs at 6 months suggested equivalent efficacy regardless of baseline anti–HSV-1 antibody status (Figure 5a) • At 6 months, post hoc analysis of treatment response to B-VEC was consistent regardless of anti-COL7 seroconversion (Figure 5b) – A responder was defined as meeting the primary endpoint of complete wound healing at 6 months Figure 5. B-VEC response rate at 6 months according to (a) baseline anti–HSV-1 serostatus and (b) anti-COL7 seroconversion *Seroconversion defined as seronegative at baseline but seropositive when tested at 6 months. Data in figure based on post hoc analysis using imputation. B-VEC, beremagene geperpavec; COL7, type VII collagen; HSV-1, herpes simplex virus type 1. (a) 67.9% 57.5% 29.3% 16.3% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% R es po ns e R at e (% P at ie nt s) Baseline Anti–HSV-1 Seropositive (n=14) Baseline Anti–HSV-1 Seronegative (n=8) (b) 66.2% 60.0% 25.4% 20.0% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% R es po ns e R at e (% P at ie nt s) Anti-COL7 Seroconversion* (n=13) No Anti-COL7 Seroconversion (n=5) Results (cont.) Results (cont.) B-VEC Placebo B-VEC Placebo 6 Months (Week 26) 38.6% absolute difference 41.3% absolute difference 40.8% absolute difference 40.0% absolute difference Exhibit 99.1