PowerPoint Presentation REFERENCES •Lesions on the head and neck in the ≥18 yo population were analyzed from the cohort published in Estrada et al.; samples are described in Table 1. •Clinically diagnosed melanomas tested with the 31-GEP (prognostic melanoma test available from Castle Biosciences Inc.) were included in this study. Benign samples were acquired from 7 centers. Benign samples were reviewed and included in the study if 2/3 or 3/3 diagnoses were concordant. •The 35-GEP utilizes dual algorithms based on neural networks to provide a result of benign, intermediate-risk or malignant.7 Table 2. Performance of the 35-GEP in different subtypes of nevi and melanoma​ of the head and neck RESULTS Table 3. 35-GEP accuracy metrics in a subset of lesions located on the head and neck METHODS SYNOPSIS Dysplastic nevi had different degrees of atypia: a – mild (n=1); b – mild (n=1), moderate (n=1).​ Table 1. Demographic information Performance of a 35-Gene Expression Profile Test in Suspicious Pigmented Lesions of the Head and Neck Clay Cockerell, M.D.1, Mathew S. Goldberg, M.D.2,3, Sarah I. Estrada, M.D.4, Gregory A. Hosler, M.D., Ph.D.5, Howard L. Martin, M.D.6, Brooke H. Russell, Ph.D.2, Olga Zolochevska, Ph.D.2, Natalie D. Depcik-Smith, M.D.7, Nathan J. Cleaver, D.O.8 1Cockerell Dermatopathology, Dallas, TX; 2Castle Biosciences, Inc., Friendswood, TX; 3Icahn School of Medicine at Mount Sinai, NY; 4Affiliated Dermatology, Scottsdale, AZ; 5ProPath, Dallas, TX; 6Sagis, Houston, TX; 7Aurora Diagnostics GPA Laboratories, Greensboro, NC; 8Cleaver Medical Group, Cumming, GA. •The accurate diagnosis of melanocytic neoplasms is a significant clinical challenge in dermatopathology; while histopathologic assessment is frequently sufficient, high rates of diagnostic discordance are reported.1-4 •Visual assessment of hematoxylin and eosin (H&E) stained lesions is inherently subjective and relies on expert interpretation and integration of a wide spectrum of architectural and cytologic features that are weighted differently based on the presumed subtype of melanocytic neoplasm and heavily influenced by the pathologists’ personal experience and training.5 •Difficult-to-diagnose lesions are commonly sent for second opinions to expert dermatopathologists who have more experience with challenging cases; however, the nature of many lesions remains ambiguous with discordant rates of diagnoses ranging from 25-43%.1,6 •The 35-gene expression profile (GEP) test has reported accuracy metrics of 99.1% sensitivity, 96.2% specificity, 96.1% positive predictive value (PPV) and 99.1% negative predictive value (NPV) within the clinically available ≥18-year-old (yo) population (n=474).7 •The 35-GEP test is intended to refine diagnoses of melanocytic neoplasms by providing clinicians with an objective ancillary tool with high accuracy. •The test provides a narrow intermediate-risk zone of 2.86% in lesions on the head and neck. •The most common melanoma subtype was lentigo maligna melanoma and the 35-GEP performed well in this subtype. The most common benign subtype was intradermal nevus and the 35-GEP also performed well in this subtype. •A test with these accuracy metrics has been shown to alleviate uncertainty in difficult-to-diagnose lesions leading to recommendations for decreased unnecessary procedures while appropriately identifying at-risk patients.9 CONCLUSIONS Funding: This study was sponsored by Castle Biosciences, Inc. (CBI), which provided funding to contributing centers for tissue and clinical data retrieval. SIE is a CBI advisor and shareholder. CC is a CBI advisor. BHR and OZ are employees and shareholders of CBI. MSG is an employee of CBI. FUNDING & DISCLOSURES 1. Elmore et al. BMJ. 2017;357:j2813 2. Shoo et al. J Am Acad Dermatol. 2010;62(5):751-756 3. Patrawala et al. J Am Acad Dermatol. 2016;74(1):75-80 4. Farmer et al. Hum Pathol. 1996;27(6):528-531 Samples that received the intermediate-risk result were excluded from the calculation. The PPV and NPV were calculated with an assumption that the cohort presented here is a random sample of the population. PPV – positive predictive value; NPV – negative predictive value; CI – confidence interval. 35-GEP Result ​≥18 yo population Benign, n​ Intermediate- risk, n​ Malignant, n​ Melanomas​ 1 1 48 Desmoplastic​ 0 0 4 Lentiginous​ 0 0 2 Lentigo maligna​ 0 0 15 In situ​ 0 0 7 Nevoid​ 0 0 3 Nodular​ 1 0 9 Spitzoid​ 0 1 0 Superficial spreading​ 0 0 8 Nevi​ 53 2 0 Blue​ 14 1 0 Common nevi​ Compound​ 3 0 0 Intradermal​ 17 0 0 Junctional​ 1 0 0 Not specified​ 10 0 0 Dysplastic​ Compound​ 4a 0 0 Junctional​ 2b 0 0 Spitz​ 2 1 0 5. Gonzalez et al. J Am Acad Dermatol. 2017;77(3):543- 548 6. Piepkorn et al. JAMA Netw Open. 2019;2(10):e1912597 7. Estrada et al. SKIN J Cutan Med. 2020;4(6):506-522 8. Kienstra et. al. Cancer Control. 2005;12(4):242-247 9. Farberg et. al. SKIN J Cutan Med. 2020;4(6):523-533 Melanoma​ Benign nevi​ N=50 N=55 Age, median (range)​ 73 (31-92)​ 51 (18-90)​ Sex, % male​ 84 36 Breslow thickness, mm (range)​ 1.1 (0.2-4.0)​ NA​ T stage, % (n)​ T0 14 (7) - T1a​ 22 (11)​ - T1b​ 26 (13)​ - T2a​ 12 (6)​ - T2b​ 8 (4)​ - T3a​ 12 (6)​ - T3b​ 4 (2)​ - Unknown 2 (1) - Ulceration % (n)​ Present​ 14 (7)​ - Absent​ 70 (35)​ - Not addressed​ 16 (8) 100 (55)​ Sub-location on head/neck, % (n) Cheek 24 (12) 24 (13) Ear 8 (4) 5.5 (3) Forehead 8 (4) 9 (5) Lip 0 (0) 4 (2) Neck 26 (13) 16 (9) Nose 8 (4) 5 (3) Scalp 20 (10) 31 (17) Other 6 (3) 5.5 (3) N=105 35-GEP​ 95% CI​ Sensitivity​ 97.96%​ 89-100% Specificity​ 100%​ 93-100% PPV​ 100%​ 93-100% NPV​ 98.15%​ 90-100% Intermediate- risk result​ 2.86%​ Melanoma in situ and invasive melanoma of the head and neck require special consideration in regard to excision, surgical staging, and treatment regimens, often making diagnostic timing and accuracy critical for this subset of lesions.8 Our objective is to demonstrate accuracy of the 35-GEP within lesions located on the head and neck. OBJECTIVE