Archives of pathology & laboratory medicine最新文献

{"title":"Age- and Sex-Dynamic Fluctuations and Reference Intervals for Alkaline Phosphatase Among the Spanish Population.","authors":"Laura Castells Vilella, Paula Sánchez-Pintos, José Félix Muñiz Llama, Matías Gámez Martínez, María Luz Couce, Jordi Antón","doi":"10.5858/arpa.2023-0335-OA","DOIUrl":"10.5858/arpa.2023-0335-OA","url":null,"abstract":"<p><strong>Context.—: </strong>Interpretation of alkaline phosphatase (ALP) activity is essential for the diagnosis of certain diseases. ALP changes during life and may vary between different populations.</p><p><strong>Objective.—: </strong>To establish reference intervals (RIs) and percentile charts for ALP activity in the Spanish population through a multicentric observational study and to compare the RIs to those defined in other countries.</p><p><strong>Design.—: </strong>A total of 662 350 ALP measurements from individuals ages 0 to 99 years from 9 Spanish tertiary care centers collected between 2020 and 2022 were analyzed. This study is the largest published on this topic in the literature to date.</p><p><strong>Results.—: </strong>Continuous percentile charts for ALP according to sex and age were established which can be used as RIs. Higher levels are reached during the first weeks of life. In puberty, a differential evolution is observed in both sexes, reaching a peak at 10 to 13 years of age in boys and remaining stable in girls at this age. Significant differences were also observed in adults, higher in men between ages 20 and 49 years and between ages 50 and 79 years in women, as reported in some countries.</p><p><strong>Conclusions.—: </strong>ALP activity follows an age- and sex-dependent fluctuation with geographic differences. It is important to have appropriate reference values for each population in order to allow for a correct diagnostic interpretation and early diagnosis of diseases related to ALP abnormalities.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":"e19-e25"},"PeriodicalIF":0.0,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141155940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Tissue Prior to the Initial Hematoxylin-Eosin Section Demonstrates Value as an Alternative Source of DNA for Molecular Testing.","authors":"Peter Sabatini, Robert Ta, Melanie Peralta, Melanie Anderson, Shehnaz Khan, Rosetta Belcastro, Andrea Arruda, Mark David Minden, Michael Cabanero, Anca Prica, Tong Zhang, Robert Kridel, Tracy Stockley, Daniel Xia","doi":"10.5858/arpa.2024-0222-OA","DOIUrl":"https://doi.org/10.5858/arpa.2024-0222-OA","url":null,"abstract":"<p><strong>Context.—: </strong>Small biopsies are used for histologic, immunophenotypic, cytogenetic, molecular genetic, and other ancillary studies. Occasionally, this diagnostic tissue is exhausted before molecular testing can be performed.</p><p><strong>Objective.—: </strong>To investigate a simple banking protocol for currently discarded tissues trimmed off prior to the initial hematoxylin-eosin section, as an alternative source of DNA for molecular studies.</p><p><strong>Design.—: </strong>Mock biopsies of lung adenocarcinomas, benign testes, and B-cell lymphomas were constructed from biobank blocks; these simulated biopsies were assessed via epidermal growth factor receptor (EGFR) p.L858R droplet digital polymerase chain reaction (PCR), Biomed B-cell clonality testing by PCR, or a custom next-generation sequencing panel for lymphomas. For each cancer mock biopsy, DNA amounts and molecular test results from the \"trimmings\" samples were compared to data from corresponding molecular samples acquired via a \"standard\" clinical protocol.</p><p><strong>Results.—: </strong>The data show that although trimmings samples usually contained less DNA than standard samples, both sample classes generally had sufficient DNA for testing and produced essentially identical molecular results. A single sample showed low-level carryover contamination on droplet digital PCR testing.</p><p><strong>Conclusions.—: </strong>Tissue trimmings banked by using the studied protocol demonstrated value as a potential alternative sample for molecular testing.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142960251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prognostic Implications of the Bethesda System in Fine-Needle Aspiration for Follicular Thyroid Carcinoma.","authors":"Hyunju Park, Young Lyun Oh, Myoung Kyoung Kim, Soo Yeon Hahn, Jun-Ho Choe, Man Ki Chung, Bogyeong Han, Sun Wook Kim, Jae Hoon Chung, Tae Hyuk Kim","doi":"10.5858/arpa.2024-0304-OA","DOIUrl":"https://doi.org/10.5858/arpa.2024-0304-OA","url":null,"abstract":"<p><strong>Context.—: </strong>Fine-needle aspiration is an effective tool for sampling thyroid nodules; its results are classified according to the Bethesda System for Reporting Thyroid Cytopathology (BSRTC), whose categories define malignancy risks.</p><p><strong>Objective.—: </strong>To compare the histologic outcomes and disease-free survival (DFS) with the preceding BSRTC categories, we hypothesized that the initial cytologic categories may reflect long-term outcomes in follicular thyroid carcinoma (FTC), similar to those observed in papillary thyroid carcinoma.</p><p><strong>Design.—: </strong>This retrospective study enrolled 134 patients with FTC who underwent preoperative cytology between April 2011 and December 2020. Results were classified into 6 categories according to the BSRTC: nondiagnostic, benign, atypia of uncertain significance (AUS), follicular neoplasm (FN), suspicious for malignancy, or malignant.</p><p><strong>Results.—: </strong>Overall, 8 of 134 patients (6.0%) were categorized as having a nondiagnostic FTC, 35 of 134 (26.1%) as benign, 51 of 134 (38.1%) as AUS, and 40 of 134 (29.9%) as FN. No lesions were classified as suspicious for malignancy or malignant. The nondiagnostic, AUS, and FN categories were associated with a progressively higher risk of vascular invasion, disease recurrence, and high-risk FTC, based on the 2022 World Health Organization classification (P for trend = .01, .01, and .01, respectively). Disease-free survival was lower in the FN group (log-rank P = .01).</p><p><strong>Conclusions.—: </strong>The initial BSRTC results may reflect not only the risk of malignancy but also the presence of vascular invasion and poor prognosis when the thyroid nodule is diagnosed as FTC. These results may provide prognostic information for therapeutic decision-making and clinical management of FTC.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142904308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical Molecular Testing for Clonal Relatedness of Second Melanoma Tumors.","authors":"Jaclyn M Plotzke, David Manthei, Douglas R Fullen, May P Chan, Scott C Bresler, Hong Xiao, Aleodor A Andea, Paul W Harms","doi":"10.5858/arpa.2024-0267-OA","DOIUrl":"https://doi.org/10.5858/arpa.2024-0267-OA","url":null,"abstract":"<p><strong>Context.—: </strong>Patients with melanoma can develop second tumors representing either metastases or new primary melanoma. This distinction has profound implications for management. Although clinicopathologic features are often sufficient, molecular assays can support the presence or absence of clonal relatedness in challenging cases. However, the potential for false-positive and false-negative results in this context is not well described.</p><p><strong>Objective.—: </strong>To evaluate clinical molecular assays used to determine whether melanoma tumors represent primary-metastasis pairs or unrelated tumors.</p><p><strong>Design.—: </strong>We identified clinical cases at our institution in which paired melanocytic tumors were analyzed for clonal relatedness by molecular assays. Results were compared against data sets and/or controls to establish the likelihood that paired tumors were clonally related.</p><p><strong>Results.—: </strong>In total, 12 pairs were evaluated by single-nucleotide polymorphism (SNP) array, targeted next-generation sequencing (NGS), or both. SNP array predicted relatedness in 5 of 9 cases and unrelatedness in 4 cases. In SNP comparisons, whole-chromosomal and arm-level changes were often nonspecific (coincidentally similar between unrelated tumors). Targeted NGS predicted relatedness in 2 of 4 cases and unrelatedness in 1 case, and was equivocal/noncontributory in 1 case. For targeted NGS, nonspecific (coincidentally similar) results were related to recurrent oncogenic drivers or pairs lacking detected oncogene mutations.</p><p><strong>Conclusions.—: </strong>The genome-wide analysis provided by SNP array was optimal for assessment of clonality. Targeted NGS can be informative but may be equivocal in some cases. The choice of assay may rely upon considerations including the amount of DNA, likelihood of distinctive mutations, and need for therapeutic target identification.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142901096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Online Portal Use of Pathology Reports in Patients With Solid Tumors.","authors":"Amber Y Bo, Yee Chung Cheng, Ben George, Deepak Kilari, Jonathan R Thompson, Julie M Jorns","doi":"10.5858/arpa.2024-0327-OA","DOIUrl":"https://doi.org/10.5858/arpa.2024-0327-OA","url":null,"abstract":"<p><strong>Context.—: </strong>Patients can now immediately review pathology reports via online health portals.</p><p><strong>Objective.—: </strong>To better characterize patient perceptions of pathology report helpfulness and preferences for access of pathology reports via a patient portal.</p><p><strong>Design.—: </strong>Semistructured interviews were conducted with oncology patients with breast, endocrine, gastrointestinal, genitourinary, and thoracic malignancies. Patient demographic information, cancer type, question responses, and thematically grouped comments were statistically analyzed.</p><p><strong>Results.—: </strong>Among 230 patients, there was equal sex distribution (116 of 230, 50.4% female; 114 of 230, 49.6% male). Patients who viewed or had a support member view their reports in the portal (172 of 230; 74.8%) differed from those who did not (58 of 230; 25.2%) only in perception of helpfulness (P < .001) of the report. Difficulty understanding medical terminology was the most frequently cited challenge among both those who found the reports helpful (30 of 160; 18.75%) and not helpful (31 of 46; 67.4%). Most patients (196 of 230; 85.2%) preferred immediate release of results, even if the news was bad, whereas some (34 of 230; 14.7%) would opt out of immediate release for fear of misunderstanding (11 of 34; 32.4%) or receiving distressing information from reading the report (23 of 34; 67.6%).</p><p><strong>Conclusions.—: </strong>Options for portal flexibility (ie, patient choice of opting for immediate release of some, but not all, results), patient-centered pathology reports, educational materials, clinician preparation of patients, and tailored patient support are strategies that can help more patients benefit from reviewing pathology report information.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-12-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142901097","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Technical Competency Assessment of Peripheral Blood Smears: Tools and Trends Learned From 2 College of American Pathologists Q-Probes Studies.","authors":"Jeffrey A Vos, Girish Venkataraman, Liuyan Jennifer Jiang, Barbara J Blond, Suzanne Coulter, Rhona J Souers","doi":"10.5858/arpa.2024-0259-CP","DOIUrl":"https://doi.org/10.5858/arpa.2024-0259-CP","url":null,"abstract":"<p><strong>Context.—: </strong>Morphologic evaluation of peripheral blood smears provides valuable information to diagnose and manage a variety of hematologic disorders.</p><p><strong>Objective.—: </strong>To measure the competency of the technical staff in the morphologic evaluation of peripheral blood smears and provide performance trends.</p><p><strong>Design.—: </strong>Participating technologists accessed 10 whole slide-imaged peripheral blood smears through a web-based imaging tool in 2 separate studies. Participants performed a 100-cell differential and morphologic evaluation for each slide image. Grading criteria, determined by 3 hematopathologists, were weighted according to their clinical significance (score range, 0-100 for each case). Each institution and participant answered a questionnaire to assess the impact of current practices and educational programs on competency scores.</p><p><strong>Results.—: </strong>A total of 776 technologists from 92 institutions participated in study 1 and 1495 technologists from 179 institutions participated in study 2. Median performance scores for institutions were 78.9 and 87.6 for studies 1 and 2, respectively, encompassing a range of hematologic disorders. Based on results of the questionnaire for study 1, higher performance scores were seen when institutions required a specific number of continuing education credits per year through an agency (P = .005). In study 2, institutions with remediation procedures following a failed competency demonstrated higher performance scores (P = .03).</p><p><strong>Conclusions.—: </strong>Medical technologist competency of peripheral blood smears improves with level of experience and is positively impacted through attending educational programs. Whole slide images offer a convenient means of assessing technical competence and provide data to allow institutions to appropriately focus their procedures and educational efforts.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142883862","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Significance of Detecting an Unusual Myeloblast Immunophenotype in a Presumptive Clinical Diagnosis of Myelodysplastic Syndromes.","authors":"Fnu Sameeta, Wei Wang, Fatima Zahra Jelloul, Okechukwu V Nwogbo, Beenu Thakral, Jie Xu, Shaoying Li, Chi Young Ok, Guilin Tang, Fuli Jia, L Jeffrey Medeiros, Sanam Loghavi, Jeffrey L Jorgensen, Sa A Wang","doi":"10.5858/arpa.2024-0228-OA","DOIUrl":"https://doi.org/10.5858/arpa.2024-0228-OA","url":null,"abstract":"<p><strong>Context.—: </strong>Blasts in myelodysplastic syndromes (MDSs) typically have a primitive myeloid immunophenotype (CD34+CD117+CD13+CD33+HLA-DR+). On rare occasions, blasts were found to be CD34 negative or minimally expressed in a presumptive MDS.</p><p><strong>Objective.—: </strong>To investigate the occurrence of these cases, and to examine any unique molecular genetic features, and clinical relevance.</p><p><strong>Design.—: </strong>Over 2000 flow cytometry immunophenotyping tests for MDS performed during a 5-year period were retrospectively reviewed. Chronic myelomonocytic leukemia and overt acute myeloid leukemia (AML) (≥20% blasts) were excluded.</p><p><strong>Results.—: </strong>Approximately 800 cases had abnormal myeloblasts consistent with myeloid neoplasms; 96% of cases showed a typical primitive phenotype, but 31 patients (4%) had unusual blasts that were either completely or partially negative for CD34. Of the latter, recurrent genetic abnormalities were identified in 13 (42%) including 10 with nucleophosmin 1 (NPM1) mutation, 1 with lysine methyltransferase 2A (KMT2A) rearrangement, and 2 with t(3;5)(q25.3;q35.1)/NPM1::myeloid leukemia factor 1 (MLF1). These cases were classified as MDS prior to the 2022 classifications, but 9 of 13 (69%) and 7 of 13 (52%) cases would be reclassified as AML according to the 5th edition of the World Health Organization classification and the International Consensus Classification, respectively. Eight cases (26%) had multihit tumor protein p53 (TP53) mutation, and 6 of them were ultimately diagnosed as or quickly evolved to pure erythroid leukemia. Of the remaining 10 cases, 4 uncharacteristically had no detectable molecular genetic abnormalities.</p><p><strong>Conclusions.—: </strong>Our data show that, if a presumptive MDS shows a nonprimitive blast phenotype, caution is needed to rule out AML with recurrent genetic abnormality with an oligoblastic presentation, high-risk myeloid neoplasms with double-hit TP53 mutation with abnormal erythroid proliferation, and MDS with molecular-genetic and clinical features more akin to AML.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142878844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deployment of a Machine Learning Algorithm in a Real-World Cohort for Quality Control Monitoring of Human Epidermal Growth Factor-2-Stained Clinical Specimens in Breast Cancer.","authors":"Benjamin Glass, Michel E Vandenberghe, Surya Teja Chavali, Syed Ashar Javed, Murray Resnick, Harsha Pokkalla, Hunter Elliott, Sudha Rao, Shamira Sridharan, Jacqueline A Brosnan-Cashman, Ilan Wapinski, Michael Montalto, Andrew H Beck, Craig Barker","doi":"10.5858/arpa.2024-0111-OA","DOIUrl":"https://doi.org/10.5858/arpa.2024-0111-OA","url":null,"abstract":"<p><strong>Context.—: </strong>Precise determination of biomarker status is necessary for clinical trial enrollment and endpoint analyses, as well as for optimal treatment determination in real-world practice. However, variabilities may be introduced into this process due to the processing of clinical specimens by different laboratories and assessment by distinct pathologists. Machine learning tools have the potential to minimize inconsistencies, although their use is not presently widespread.</p><p><strong>Objective.—: </strong>To assess the applicability of machine learning to the quality control process for biomarker scoring in oncology, we developed and validated an automated machine learning model to be applied as a quality control tool for monitoring the assessment of human epidermal growth factor-2 (HER2).</p><p><strong>Design.—: </strong>The model was trained using whole slide images from multiple sources to quantify HER2 expression and measure immunohistochemistry stain intensity, tumor area, and the presence of artifacts or ductal carcinoma in situ across breast cancer phenotypes. The quality control tool was deployed in a real-world cohort of HER2-stained breast cancer sample images collected from routine diagnostic practice to evaluate trends in HER2 testing quality indicators and between pathology laboratories.</p><p><strong>Results.—: </strong>Automated image analysis for HER2 scoring is consistent and reliable using this algorithm. Deployment of the HER2 quality control tool across 3 clinical laboratories revealed interlaboratory variability in HER2 scoring and inconsistencies in data reporting.</p><p><strong>Conclusions.—: </strong>These results support the future incorporation of quality control algorithms for real-time monitoring of clinical laboratories contributing to clinical trials in oncology and in the real-world setting of HER2 immunohistochemistry testing in local clinical laboratories and hospitals.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142848452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pulmonary Adenocarcinoma Updates: Histology, Cytology, and Grading.","authors":"Jake Sharma, Fang Zhou, Andre L Moreira","doi":"10.5858/arpa.2023-0540-RA","DOIUrl":"https://doi.org/10.5858/arpa.2023-0540-RA","url":null,"abstract":"<p><strong>Context.—: </strong>Adenocarcinomas are the most common histologic subtype of lung cancer, and exist within a widely divergent clinical, radiologic, molecular, and histologic spectrum. There is a strong association between histologic patterns and prognosis that served as the basis for a recently described grading system. As the study of molecular pathology rapidly evolves, all targetable mutations so far have been found in adenocarcinomas, thus requiring accurate diagnosis and classification for triage of molecular alterations and adequate therapy.</p><p><strong>Objective.—: </strong>To discuss the rationale for adenocarcinoma classifications within the 2021 5th edition of the World Health Organization, with a focus on nonmucinous tumors, including tumor grading and biopsy/cytology diagnosis.</p><p><strong>Data sources.—: </strong>PubMed search.</p><p><strong>Conclusions.—: </strong>A grading system for adenocarcinoma has improved prognostic impact of the classification of pulmonary adenocarcinoma. An accurate diagnosis of adenocarcinoma in small biopsy material is important for tissue triage for molecular studies and ultimately for patient management and treatment.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142820246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Development and Evaluation of a Convolutional Neural Network for Cutaneous Melanoma Detection in Whole Slide Images.","authors":"Emily L Clarke, Derek Magee, Julia Newton-Bishop, William Merchant, Robert Insall, Nigel G Maher, Richard A Scolyer, Grace Farnworth, Anisah Ali, Sally O'Shea, Darren Treanor","doi":"10.5858/arpa.2024-0094-OA","DOIUrl":"https://doi.org/10.5858/arpa.2024-0094-OA","url":null,"abstract":"<p><strong>Context.—: </strong>The current melanoma staging system does not account for 26% of the variance seen in melanoma-specific survival, therefore our ability to predict patient outcome is not fully elucidated. Morphology may be of greater significance than in other solid tumors, with Breslow thickness remaining the strongest prognostic indicator despite being subject to high levels of interobserver variation. The application of convolutional neural networks to whole slide images affords objective morphologic metrics, which may reveal new insights into patient prognosis.</p><p><strong>Objective.—: </strong>To develop and evaluate a convolutional neural network for invasive cutaneous melanoma detection in whole slide images for the generation of objective prognostic biomarkers based on tumor morphology.</p><p><strong>Design.—: </strong>One thousand sixty-eight whole slide images containing cutaneous melanoma from 5 data sets have been used in the initial development and evaluation of the convolutional neural network. A 2-class tumor segmentation network with a fully convolutional architecture was trained using sparse annotations. The network was evaluated at per-pixel and per-tumor levels as compared to manual annotation, as well as variation across 3 scanning platforms.</p><p><strong>Results.—: </strong>The convolutional neural network located conventional cutaneous invasive melanoma tissue with an average per-pixel sensitivity and specificity of 97.59% and 99.86%, respectively, across the 5 test sets. There were high levels of concordance between the tumor dimensions generated by the model as compared to manual annotation, and between the tumor dimensions generated by the model across 3 scanning platforms.</p><p><strong>Conclusions.—: </strong>We have developed a convolutional neural network that accurately detects invasive cutaneous conventional melanoma in whole slide images from multiple data sources. Future work should assess the use of this network to generate metrics for survival prediction.</p>","PeriodicalId":93883,"journal":{"name":"Archives of pathology & laboratory medicine","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142815243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}