Prieto, Víctor G.

Importance Tumor-infiltrating lymphocytes (TILs) are a provocative biomarker in melanoma, influencing diagnosis, prognosis, and immunotherapy outcomes; however, traditional pathologist-read TIL assessment on hematoxylin and eosin–stained slides is prone to interobserver variability, leading to inconsistent clinical decisions. Therefore, development of newer TIL scoring approaches that produce more reliable and consistent readouts is important. Objective To evaluate the analytical and clinical validity of a machine learning algorithm for TIL quantification in melanoma compared with traditional pathologist-read methods. Design, Setting, and Participants This multioperator, global, multi-institutional prognostic study compared TIL scoring reproducibility between traditional pathologist-read methods and an artificial intelligence (AI)-driven approach. The study was conducted using retrospective cohorts of patients with melanoma between January 2022 and June 2023 across 45 institutions, with tissue evaluated by participants from academic, clinical, and research institutions. Participants were selected to ensure diverse expertise and professional backgrounds. Main Outcomes and Measures Intraclass correlation coefficient (ICC) values were calculated for the manual and AI-assisted arms using log-transformed data. Kendall W values were calculated for Clark scores (brisk = 3, nonbrisk = 2, and sparse = 1). Reliabilities of ICC and W values were classified as moderate (0.40-0.60), good (0.61-0.80), or excellent (>0.80). AI TIL measurements were dichotomized using the 16.6 and median cutoffs. Univariable and multivariable Cox regression analyses assessed the prognostic value of TIL scores adjusted for clinicopathologic variables. Results There were 111 patients with melanoma in the independent testing cohort (median [range] age at diagnosis, 61.0 [25.0-87.0] years; 56 [50.5%] male) who contributed melanoma whole tissue sections. A total of 98 participants evaluated TILs on 60 hematoxylin and eosin–stained melanoma tissue sections. All 40 participants in the manual arm were pathologists, while the AI-assisted arm included 11 pathologists and 47 nonpathologists (scientists). The AI algorithm demonstrated superior reproducibility, with ICCs higher than 0.90 for all machine learning TIL variables, significantly outperforming manual assessments (ICC, 0.61 for AI-derived stromal TILs vs Kendall W , 0.44 for manual Clark TIL scoring). AI-based TIL scores showed prognostic associations with patient outcomes (n = 111) using the median cutoff approach with a hazard ratio (HR) of 0.45 (95% CI, 0.26-0.80; P = .005), and using the cutoff of 16.6, with an HR of 0.56 (95% CI, 0.32-0.98; P = .04). Conclusions and Relevance In this prognostic study of TIL quantification in melanoma, the AI algorithm demonstrated superior reproducibility and prognostic associations compared with traditional methods. Although the retrospective nature of the cohorts limits demonstration of clinical utility, the publicly available dataset and open-source AI tool offer a foundation for future validation and integration into melanoma management.

Abstract: Breast cancer1-associated protein 1 (BAP-1)-inactivated melanocytic tumors are a group of familial or sporadic lesions with distinctive histology and molecular features. Inherited germline inactivating mutations in BAP1 have been associated with the development of multiple epithelioid melanocytic neoplasms resembling Spitz nevi and increased susceptibility for developing several malignancies, including uveal melanoma, cutaneous melanoma, renal cell carcinoma, mesothelioma, and other tumors. Cutaneous melanoma with loss of BAP1 expression is rare. We present a unique case of BAP1 -inactivated melanoma with anaplastic lymphoma kinase ( ALK ) fusion arising in a pre-existing BAP1 -inactivated nevus in a 47-year-old female patient who presented with a dome-shaped red papule on the superior crus of the right antihelix. Histology revealed intradermal melanocytic proliferation with biphenotypic morphology. There was a proliferation of atypical melanocytes showing epithelioid features in the background of nevus. Mitotic figures were identified in the cytologically atypical component of the lesion. Mart-1/Ki67 dual stain demonstrated a higher proliferation index in the larger epithelioid atypical cells, supporting the diagnosis of melanoma. Nuclear BAP-1 expression was lost in the larger atypical cells and associated nevoid cells. Preferentially expressed antigen in melanoma stain demonstrated focal positive staining in 20%–30% of the melanocytes. Immunostaining for B-Raf proto-oncogene, serine/threonine kinase V600E was diffusely positive and ALK demonstrated patchy immunoreactivity in the melanocytic proliferation. Interphase fluorescence in situ hybridization studies showed gains at chromosome 6p25 (Ras responsive element binding protein 1) in the tumor cells. The comprehensive next-generation sequencing revealed B-Raf proto-oncogene, serine/threonine kinase V600E mutation, TP53 mutation, ALK fusion, BAP1 loss (copy number variation = 0.0, potentially germline), and loss of MAP2K7, Von Hippel-Lindau tumor suppressor , FGFR3, CDKN2A , 19q, and telomerase reverse transcriptase . The tumor was microsatellite stable with a low tumor mutational burden (5.76 mutations/Mb). The tumor was completely excised with negative margins. The patient is doing well at 17 months follow-up with no signs of recurrence.

Background: Atypical fibroxanthoma (AFX) is a dermal-based, low-grade neoplasm with no specific lineage of differentiation. The occurrence of AFX with osteoclast-like giant cells is exceptionally rare. Less than 20 cases have been reported in the literature. Case Presentation: A 77-year-old man with a medical history of multiple basal and squamous cell carcinomas of the skin, presented with a progressively growing erythematous nodule on the sun-damaged right central parietal scalp. A shave biopsy showed a dermal spindle cell proliferation accompanied by numerous osteoclast-like multinucleated giant cells and predominant atypical mitotic figures. The immunohistochemical staining showed a diffuse positive staining for CD68 and SMA, patchy staining for CD10, and negative staining for SOX-10, pan-cytokeratin, CK5/6, S100, CD34, and desmin. The tumor was completely excised with negative margins. A subsequent follow-up over a period of 13 months showed no recurrence. Conclusion: Distinguishing AFX with osteoclast-like giant cells from both malignant and benign skin lesions with osteoclast-like giant cells is crucial. Although AFX tumors display worrisome malignant histologic features, most cases have a favorable prognosis with a local recurrence rate below 5% and exceedingly rare metastasis.

AbstractBackgroundAppropriate use criteria (AUC) provide patient‐centered physician guidance in test selection. An initial set of AUC was reported by the American Society of Dermatopathology (ASDP) in 2018. AUC reflect evidence collected at single timepoints and may be affected by evolving evidence and experience. The objective of this study was to update and expand AUC for selected tests.MethodsRAND/UCLA (RAND Corporation [Santa Monica, CA]/University of California Los Angeles) methodology used includes the following: (a) literature review; (b) review of previously rated tests and previously employed clinical scenarios; (c) selection of previously rated tests for new ratings; (d) development of new clinical scenarios; (e) selection of additional tests; (f) three rating rounds with feedback and group discussion after rounds 1 and 2.ResultsFor 220 clinical scenarios comprising lymphoproliferative (light chain clonality), melanocytic (comparative genomic hybridization, fluorescence in situ hybridization, reverse transcription polymerase chain reaction, telomerase reverse transcriptase promoter), vascular disorders (MYC), and inflammatory dermatoses (periodic acid‐Schiff, Gömöri methenamine silver), consensus by panel raters was reached in 172 of 220 (78%) scenarios, with 103 of 148 (70%) rated “usually appropriate” or “rarely appropriate” and 45 of 148 (30%), “appropriateness uncertain.” Limitations: The study design only measures appropriateness. Cost, availability, test comparison, and additional clinical considerations are not measured. The possibility that the findings of this study may be influenced by the inherent biases of the dermatopathologists involved in the study cannot be excluded.ConclusionsAUC are reported for selected diagnostic tests in clinical scenarios that occur in dermatopathology practice. Adhering to AUC may reduce inappropriate test utilization and improve healthcare delivery.

Abstract Context. —Primary esophageal melanoma (PEM) is a rare disease and is difficult to distinguish from other esophageal malignancies and from metastatic melanoma. Objective. —To develop diagnostic criteria for PEM, we compared the clinicopathologic features of 5 PEMs and 5 metastatic melanomas to esophagus. Design. —Ten cases of esophageal melanoma, including 4 surgically resected specimens, 2 autopsy cases, and 4 cases reported on mucosal biopsies, were reviewed. The histologic parameters used in this study were well-characterized features for cutaneous melanoma, including junctional component (in situ melanoma), radial growth phase, modified Breslow thickness, depth of invasion, lymphovascular invasion, satellitosis, predominant type of cytology, and regional lymph node metastasis. Clinical and follow-up information was obtained by reviewing patients' medical records. Results. —Previous history of cutaneous melanoma was present in all 5 cases of metastatic esophageal melanoma but was not present in the 5 patients with PEMs. In situ melanoma and/or radial growth phase were identified in all 5 PEMs but were not present in any of the metastatic cases. Among the 4 resected and 2 autopsy cases, melanocytosis and mixed epithelioid and spindle cell morphology was present in 2 (50%) of 4 PEMs but was not present in 2 (40%) of the metastatic melanomas. Melanin pigment was detectable in all cases. Patients with PEM had better survival than those who had metastatic melanoma to esophagus ( P = .03). Conclusions. —The presence of in situ melanoma, radial growth phase, melanocytosis, and mixed epithelioid and spindle cell morphology, in the context of no history of melanoma, distinguishes PEM from metastatic melanoma.

BackgroundAppropriate use criteria (AUC) provide physicians guidance in test selection, and can affect health care delivery, reimbursement policy and physician decision‐making.ObjectivesThe American Society of Dermatopathology, with input from the American Academy of Dermatology and the College of American Pathologists, sought to develop AUC in dermatopathology.MethodsThe RAND/UCLA appropriateness methodology, which combines evidence‐based medicine, clinical experience and expert judgment, was used to develop AUC in dermatopathology.ResultsWith the number of ratings predetermined at 3, AUC were developed for 211 clinical scenarios involving 12 ancillary studies. Consensus was reached for 188 (89%) clinical scenarios, with 93 (44%) considered “usually appropriate,” 52 (25%) “rarely appropriate” and 43 (20%) “uncertain appropriateness.”LimitationsThe methodology requires a focus on appropriateness without comparison between tests and irrespective of cost.ConclusionsThe ultimate decision of when to order specific test rests with the physician and is one where the expected benefit exceeds the negative consequences. This publication outlines the recommendations of appropriateness—AUC for 12 tests used in dermatopathology. Importantly, these recommendations may change considering new evidence. Results deemed “uncertain appropriateness” and where consensus was not reached may benefit from further research.

Abstract: Cellular blue nevomelanocytic lesions (CBNLs) frequently pose diagnostic problems to pathologists, and their biological potential may be difficult to establish. In this study, the authors have analyzed the clinical, histological, and outcome data of 37 cellular blue nevomelanocytic lesions and the molecular characteristics of 4 lesions. The cohort of cases comprised 8 cellular blue nevi (CBNs), 17 atypical cellular blue nevi (ACBNs), and 12 blue-nevus–like melanomas (BNLMs) with a mean follow-up of 5 years. The average age at diagnosis was 25.9 years for patients with ACBN, versus 30.4 years for CBN, and 44.6 years for BNLM. Both CBN and ACBN occurred most frequently on the trunk or extremities, whereas BNLM primarily involved the scalp. Histologically, CBN and ACBN were characterized by a mean diameter of <1 cm, absence of necrosis, low mitotic rate (mean: 1–2 mitotic figures/mm2), little or no infiltrative properties, and usually low-grade cytologic atypia. In contrast, BNLM had a mean diameter of 1.6 cm, necrosis, tissue infiltration, greater mitotic activity (mean: 6 mitotic figures/mm2), and high-grade cytologic atypia. ACBNs often were larger, more densely cellular, exhibited higher mitotic counts, and were cytologically more atypical than CBN. Seven CBN cases with follow-up had a benign clinical course (average follow-up of 4.7 years). Among 6 patients with ACBN who underwent sentinel lymph node (SLN) biopsy, 3 were positive, and a single additional case had 1 positive non-SLN (this patient did not have a SLN biopsy performed). All 14 cases of ACBN with follow-up were alive and without recurrence with mean follow-up of 5 years. Of the 9 melanoma cases with follow-up, 3 patients with SLN and non-SLN involvement died from their disease (average follow-up of 4.8 years). Array comparative genomic hybridization was performed on 2 ACBNs and 1 BNLM: One of the 2 ACBNs showed chromosomal aberrations and 1 BNLM showed multiple chromosomal gains and losses. Multiplex polymerase chain reaction was performed on 1 ACBN, and no mutations were found. From these results, the authors conclude that ACBN occupy an intermediate position within the spectrum of CBN and BNLM, yet many lesions cannot be reliably distinguished from either CBN or BNLM because of overlapping histologic features. However, in general, ACBNs seem to aggregate more closely with CBN in terms of clinical, histological, molecular profile (limited data), and biological behavior.