Results
Patient Sample
A total of 25 ambiguous melanocytic neoplasms from 25 patients met the study inclusion criteria of a benign gene expression result with subsequent treatment as a benign neoplasm during follow-up. The patient sample statistics are summarized in Table 1. Most patients were younger than 65 years, with an average age at the time of biopsy of 48.4 years. All 25 neoplasms produced negative (benign) gene expression signature scores, all were diagnosed as benign nevi posttest by the submitting dermatopathologist, and all patients were initially treated in accordance with the benign diagnosis by the dermatologist(s) involved in clinical follow-up care. Prior to testing with the gene expression signature, most of these histopathologically indeterminate lesions received differential diagnoses, the most common of which were dysplastic nevus (84%), melanoma arising from a nevus (72%), and superficial spreading melanoma (64%; eTable). After testing with the gene expression signature and receiving a benign score, most lesions received a single differential diagnosis of dysplastic nevus (88%).
Follow-up and Survival
Clinical follow-up time ranged from 0.6 to 53.3 months, with a mean duration (SD) of 38.5 (16.6) months, and patients attended an average of 4 postbiopsy dermatology appointments (mean [SD], 4.6 [3.6]). According to the participating dermatology care providers, none of the 25 patients developed any indication during follow-up that the diagnosis of benign nevus was inaccurate. No patient had evidence of locally recurrent or metastatic melanoma, and none died during the study period.
Treatment/Interventions
The treatment recorded in the CRF was examination only for 21 of 25 patients, excision for 3, and other for 1 (Table 2). Because the explanation for the selection of other in this case described an excision performed at the same anatomic location as the biopsy, this treatment also was considered an excision for purposes of the study analyses. The 3 excisions all occurred at the first postbiopsy dermatology encounter. Across all follow-up visits, no additional surgical interventions occurred (Table 2).
The first excision (case 1) involved a 67-year-old woman with a lesion on the mid pubic region described clinically as an atypical nevus that generated a pretest histopathologic differential diagnosis including dysplastic nevus, superficial spreading melanoma, and melanoma arising within a nevus (Table 3; Figure, A and B). The gene expression test result was benign (score, −5.4), and the final pathology report diagnosis was nevus with junctional dysplasia, moderate. Surgical excision was performed at the patient’s first return visit, 505 days after initial diagnosis, with moderately dysplastic nevus as the recorded indication for removal. No repigmentation or other evidence of local recurrence or progression was detected, and the treating dermatologist indicated no suspicion that the original diagnosis of benign nevus was incorrect during the 23-month follow-up period.
Histopathologic features of the 3 melanocytic neoplasms that were excised during the study follow-up period. The histopathologic differential diagnosis for each case included dysplastic nevus, superficial spreading melanoma, and melanoma arising within a preexisting nevus. All were classified as benign by the gene expression signature. For each, the indication for excision was patient or physician preference. A and B, Case 1 (a 67-year-old woman with a lesion on the mid pubic region)(H&E, original magnifications ×10 and ×100). C and D, Case 2 (a 27-year-old woman with a lesion on the back)(H&E, original magnifications ×10 and ×100). E and F, Case 3 (a 32-year-old woman with a lesion on the abdomen)(H&E, original magnifications ×10 and ×40).
The second excision (case 2) involved a 27-year-old woman with a pigmented neoplasm on the mid upper back (Figure, C and D) biopsied to rule out dysplastic nevus that resulted in a pretest histopathologic differential diagnosis of dysplastic nevus vs superficial spreading melanoma or melanoma arising within a nevus. The gene expression test result classified the lesion as benign (score, −2.9), and the final pathology diagnosis was nevus, compound, with moderate dysplasia. Despite the benign diagnosis, residual neoplasm (or pigmentation) at the biopsy site prompted the patient to request excision at her first postbiopsy visit, 22 days after testing (Table 3). The CRF completed by the dermatologist reported no indication that the benign diagnosis was inaccurate, but the patient was subsequently lost to follow-up.
The third excision (case 3) involved a 32-year-old woman with a pigmented lesion on the abdomen (Table 3; Figure, E and F). The clinical description was irregular-appearing black papule, nevus with atypia, and the histopathologic differential diagnosis again included dysplastic nevus, superficial spreading melanoma, and melanoma arising within a preexisting nevus. The gene expression signature result was benign (score, −7.2), and the final diagnosis issued within the accompanying pathology report was nevus with moderate junctional dysplasia. Despite the benign diagnosis, excision was performed 89 days after test result availability, with apparent residual pigmentation as the specified indication. As with the other 2 cases, the treating dermatologist confirmed that neither clinical features nor follow-up events suggested malignancy.
Comment
This study followed a cohort of 25 patients with histopathologically ambiguous melanocytic neoplasms that were classified as benign by a diagnostic gene expression test with the intent of determining the outcomes of patients whose treatment aligned with their benign test result. All patients initially were managed according to their test result. During an average posttest clinical follow-up time of more than 3 years (38.5 months), the 25 biopsied lesions, most of which received a differential diagnosis of dysplastic nevus, were regarded as benign nevi by their dermatologists, and the vast majority (88%) received no further surgical intervention. Three patients underwent subsequent excision of the biopsied lesion, with patient or physician preference as the indication in each instance. None of the 25 patients developed evidence of local recurrence, metastasis, or other findings that prompted doubt of the benign diagnosis. The absence of adverse events during clinical follow-up, particularly given that most lesions were not subjected to further intervention, supports use of the gene expression test as a safe and effective adjunct to the diagnosis and treatment of ambiguous melanocytic neoplasms by dermatologists and dermatopathologists.