Modern Pathology最新文献_第8页

Comparative Analysis Reveals Recurrent Molecular Alterations in Low-Risk Human Papillomavirus 6 and Human Papillomavirus 11-Associated Squamous Cell Carcinoma of the Uterine Cervix and Vulva 比较分析揭示低危HPV6和hpv11相关的宫颈和外阴鳞状细胞癌复发性分子改变。

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-10 DOI: 10.1016/j.modpat.2025.100909

Ying Sun , Colton Smith , Jason Murray , Jing Zhu , Aparna Pallavajjala , Sichen Liang , Melanie Klausner , Ya-Chea Tsai , Chien-Fu Hung , Tzyy-Choou Wu , Ying S. Zou , Deyin Xing

Although the association between human papillomavirus (HPV) 6 and HPV11 and squamous cell carcinomas (SCCs) has been well documented, the molecular alterations and mechanisms by which these low-risk HPVs contribute to carcinogenesis remain largely unknown. In this study, we comparatively elucidated the molecular landscape of HPV6/11-associated condylomas (9 cases) and SCCs (8 cases) of the uterine cervix and vulva. Sixteen of 17 cases were successfully analyzed using deep next-generation sequencing. Recurrent molecular alterations in the SCCs included mutations in the TERT promoter (pTERT, 71%), NOTCH1 (57%), NFE2L2 (57%), NOTCH3 (29%), and CDKN2A (29%). Mutations in NOTCH1 and NFE2L2 tended to be mutually exclusive. Less common somatic mutations were each detected in the following genes: AR, ARID1A, ASXL1, BCORL1, DAZAP1, FNDC1, HRAS, KMT2B, NOTCH2, NRAS, PIK3R1, and TP53. Etiologically similar to the SCCs, the vast majority of vulvar and cervical condylomas (7/9, 78%) were infected with HPV6 rather than HPV11. Unlike the SCCs, condylomas rarely harbored recurrent pathogenic somatic mutations. NOTCH1 and NOTCH2 were the only mutant genes detected in both SCCs and condylomas in this series, suggesting a critical role for the NOTCH pathway in the initiation and maintenance of HPV6/11-related early squamous lesions. Although pathogenic mutations in NOTCH1, NOTCH2, ERBB3, ATRX, FGFR2, EPHA5, CARD11, STAG2, and TSC2 were detected in condylomas, none were recurrent, indicating diverse genetic alterations involved in the development of these lesions. Case 8 illustrated a stepwise progression from condyloma to invasive SCC, in which pathogenic mutations in pTERT and NOTCH1 were detected in the SCC (age 58) and the condyloma at age 55, but not in the condyloma at age 44. Our study presents the first report on the molecular landscape of HPV6/11-associated SCCs of the uterine cervix and vulva. We provide evidence that SCCs associated with low-risk HPV are distinct entities, differing from those related to high-risk HPV and more closely resembling HPV-independent neoplasms. Given that low-risk HPV-associated SCCs of the cervix and vulva exhibit unique morphological and molecular features, they should either be described separately within existing classification systems or classified as a distinct new entity.

虽然HPV6和HPV11与鳞状细胞癌（SCCs）的关系已经有了很好的文献记载，但这些低风险hpv导致癌变的分子改变和机制在很大程度上仍然未知。在本研究中，我们比较阐明了hpv6 /11相关尖锐湿疣（9例）和宫颈及外阴SCCs（8例）的分子图谱。17例中有16例使用深度下一代测序成功分析。SCCs中复发性分子改变包括TERT启动子（pTERT, 71%）、NOTCH1（57%）、NFE2L2（57%）、NOTCH3（29%）和CDKN2A（29%）的突变。NOTCH1和NFE2L2的突变倾向于互斥。在以下基因中分别检测到不太常见的体细胞突变：AR、ARID1A、ASXL1、BCORL1、DAZAP1、FNDC1、HRAS、KMT2B、NOTCH2、NRAS、PIK3R1和TP53。病因学上与SCCs相似，绝大多数外阴和宫颈尖锐湿疣（97.78%）感染HPV6而不是HPV11。与SCCs不同，尖锐湿疣很少有复发性致病性体细胞突变。NOTCH1和NOTCH2是该系列中仅有的在SCCs和尖锐湿疣中检测到的突变基因，这表明NOTCH通路在hpv6 /11相关的早期鳞状病变的启动和维持中起关键作用。尽管在尖锐湿疣中检测到NOTCH1、NOTCH2、ERBB3、ATRX、FGFR2、EPHA5、CARD11、STAG2和TSC2的致病突变，但没有一个复发，表明这些病变的发展涉及多种遗传改变。病例8显示了从尖锐湿疣到侵袭性鳞状细胞癌的逐步进展，其中在58岁的鳞状细胞癌和55岁的尖锐湿疣中检测到pTERT和NOTCH1的致病突变，但在44岁的尖锐湿疣中未检测到。我们的研究首次报道了宫颈和外阴hpv6 /11相关SCCs的分子景观。我们提供的证据表明，与低风险HPV相关的SCCs是不同的实体，不同于与高危HPV相关的SCCs，更接近于与HPV无关的肿瘤。鉴于宫颈和外阴的低风险hpv相关SCCs表现出独特的形态学和分子特征，它们应该在现有的分类系统中单独描述或归类为一个独特的新实体。

{"title":"Comparative Analysis Reveals Recurrent Molecular Alterations in Low-Risk Human Papillomavirus 6 and Human Papillomavirus 11-Associated Squamous Cell Carcinoma of the Uterine Cervix and Vulva","authors":"Ying Sun , Colton Smith , Jason Murray , Jing Zhu , Aparna Pallavajjala , Sichen Liang , Melanie Klausner , Ya-Chea Tsai , Chien-Fu Hung , Tzyy-Choou Wu , Ying S. Zou , Deyin Xing","doi":"10.1016/j.modpat.2025.100909","DOIUrl":"10.1016/j.modpat.2025.100909","url":null,"abstract":"<div><div>Although the association between human papillomavirus (HPV) 6 and HPV11 and squamous cell carcinomas (SCCs) has been well documented, the molecular alterations and mechanisms by which these low-risk HPVs contribute to carcinogenesis remain largely unknown. In this study, we comparatively elucidated the molecular landscape of HPV6/11-associated condylomas (9 cases) and SCCs (8 cases) of the uterine cervix and vulva. Sixteen of 17 cases were successfully analyzed using deep next-generation sequencing. Recurrent molecular alterations in the SCCs included mutations in the <em>TERT</em> promoter (<em>pTERT</em>, 71%), <em>NOTCH1</em> (57%), <em>NFE2L2</em> (57%), <em>NOTCH3</em> (29%), and <em>CDKN2A</em> (29%). Mutations in <em>NOTCH1</em> and <em>NFE2L2</em> tended to be mutually exclusive. Less common somatic mutations were each detected in the following genes: <em>AR</em>, <em>ARID1A</em>, <em>ASXL1</em>, <em>BCORL1</em>, <em>DAZAP1</em>, <em>FNDC1</em>, <em>HRAS, KMT2B</em>, <em>NOTCH2</em>, <em>NRAS</em>, <em>PIK3R1</em>, and <em>TP53</em>. Etiologically similar to the SCCs, the vast majority of vulvar and cervical condylomas (7/9, 78%) were infected with HPV6 rather than HPV11. Unlike the SCCs, condylomas rarely harbored recurrent pathogenic somatic mutations. <em>NOTCH1</em> and <em>NOTCH2</em> were the only mutant genes detected in both SCCs and condylomas in this series, suggesting a critical role for the NOTCH pathway in the initiation and maintenance of HPV6/11-related early squamous lesions. Although pathogenic mutations in <em>NOTCH1</em>, <em>NOTCH2</em>, <em>ERBB3</em>, <em>ATRX</em>, <em>FGFR2</em>, <em>EPHA5</em>, <em>CARD11, STAG2,</em> and <em>TSC2</em> were detected in condylomas, none were recurrent, indicating diverse genetic alterations involved in the development of these lesions. Case 8 illustrated a stepwise progression from condyloma to invasive SCC, in which pathogenic mutations in <em>pTERT</em> and <em>NOTCH1</em> were detected in the SCC (age 58) and the condyloma at age 55, but not in the condyloma at age 44. Our study presents the first report on the molecular landscape of HPV6/11-associated SCCs of the uterine cervix and vulva. We provide evidence that SCCs associated with low-risk HPV are distinct entities, differing from those related to high-risk HPV and more closely resembling HPV-independent neoplasms. Given that low-risk HPV-associated SCCs of the cervix and vulva exhibit unique morphological and molecular features, they should either be described separately within existing classification systems or classified as a distinct new entity.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 12","pages":"Article 100909"},"PeriodicalIF":5.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145280708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep Learning–Based Segmentation of Lung Adenocarcinoma Whole-Slide Images for Objective Grading, Tumor Spread Through Air Spaces Identification, and Mutation Prediction 基于深度学习的肺腺癌切片图像分割，用于客观分级、STAS识别和突变预测。

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-10 DOI: 10.1016/j.modpat.2025.100907

David Joon Ho , Jason C. Chang , Rania G. Aly , Hai Cao Truong Nguyen , Prasad S. Adusumilli , Thomas J. Fuchs , William D. Travis , Chad M. Vanderbilt

Manual quantification of morphologic patterns in lung adenocarcinoma is subject to reproducibility issues due to interpathologist variability. In this study, we developed a deep learning–based multiclass segmentation model providing a modality for objective and quantitative grading of digitized lung adenocarcinoma images from resected specimens. Additionally, the model can detect tumor spread through air spaces and show enrichment of specific morphologic patterns in tumors with different genomic alterations. The study was based on 766 resected nonmucinous lung adenocarcinomas. Deep Multi-Magnification Network was trained to segment 14 tissue subtypes based on annotations of 108 internal whole-slide images at pixel level by thoracic pathologists (J.C.C. and W.D.T.). The trained model was validated on an external cohort of 130 cases for determining predominant patterns and on the remaining 528 internal cases for the 3 clinical tasks. The model graded nonmucinous lung adenocarcinomas based on the International Association for the Study of Lung Cancer Pathology Committee recommendation and successfully stratified patients into well, moderately, and poorly differentiated morphologies (P < 1 × 10⁻⁴). Pixels categorized as spread through air spaces significantly correlated with pathologists’ interpretations. For molecular analysis, solid pattern was enriched with TP53 mutations and depleted of EGFR kinase domain mutations. Lepidic pattern was inversely associated with TP53 mutations. Acinar was enriched with EGFR mutations, whereas papillary was associated with RET fusions. Our study demonstrated that deep learning–based segmentation can accurately quantify histologic patterns in lung adenocarcinoma and identify additional prognostic features. By simultaneously providing an objective assessment of various tasks, our comprehensive methodology in lung adenocarcinoma paves way for deep learning–assisted pathologic diagnosis and treatment guidance.

由于病理间的可变性，肺腺癌形态学模式的人工定量受到可重复性问题的影响。在这项研究中，我们开发了一种基于深度学习的多类分割模型，为从切除标本中提取的数字化肺腺癌图像提供了一种客观定量分级的模式。此外，该模型可以检测肿瘤通过空气空间（STAS）扩散，并显示不同基因组改变的肿瘤中特定形态模式的富集。该研究基于766例切除的非粘液肺腺癌。基于胸部病理学家对108张内部整张切片图像的像素级注释，训练深度多重放大网络对14个组织亚型进行分割。经过训练的模型在130例外部队列中进行验证，以确定主要模式，并在其余528例内部病例中进行验证，以完成三个临床任务。该模型根据国际肺癌病理研究协会的建议对非粘液肺腺癌进行分级，并成功地将患者分为良好、中度和低分化形态（p-4）。归类为STAS的像素与病理学家的解释显著相关。对于分子分析，固体模式富含TP53突变和EGFR激酶结构域突变。Lepidic模式与TP53突变呈负相关。腺泡富含EGFR突变，而乳头状细胞则与RET融合有关。我们的研究表明，基于深度学习的分割可以准确地量化肺腺癌的组织学模式，并确定其他预后特征。通过同时提供各种任务的客观评估，我们的肺腺癌综合方法为深度学习辅助病理诊断和治疗指导铺平了道路。

{"title":"Deep Learning–Based Segmentation of Lung Adenocarcinoma Whole-Slide Images for Objective Grading, Tumor Spread Through Air Spaces Identification, and Mutation Prediction","authors":"David Joon Ho , Jason C. Chang , Rania G. Aly , Hai Cao Truong Nguyen , Prasad S. Adusumilli , Thomas J. Fuchs , William D. Travis , Chad M. Vanderbilt","doi":"10.1016/j.modpat.2025.100907","DOIUrl":"10.1016/j.modpat.2025.100907","url":null,"abstract":"<div><div>Manual quantification of morphologic patterns in lung adenocarcinoma is subject to reproducibility issues due to interpathologist variability. In this study, we developed a deep learning–based multiclass segmentation model providing a modality for objective and quantitative grading of digitized lung adenocarcinoma images from resected specimens. Additionally, the model can detect tumor spread through air spaces and show enrichment of specific morphologic patterns in tumors with different genomic alterations. The study was based on 766 resected nonmucinous lung adenocarcinomas. Deep Multi-Magnification Network was trained to segment 14 tissue subtypes based on annotations of 108 internal whole-slide images at pixel level by thoracic pathologists (J.C.C. and W.D.T.). The trained model was validated on an external cohort of 130 cases for determining predominant patterns and on the remaining 528 internal cases for the 3 clinical tasks. The model graded nonmucinous lung adenocarcinomas based on the International Association for the Study of Lung Cancer Pathology Committee recommendation and successfully stratified patients into well, moderately, and poorly differentiated morphologies (<em>P</em> < 1 × 10<sup>−4</sup>). Pixels categorized as spread through air spaces significantly correlated with pathologists’ interpretations. For molecular analysis, solid pattern was enriched with <em>TP53</em> mutations and depleted of <em>EGFR</em> kinase domain mutations. Lepidic pattern was inversely associated with <em>TP53</em> mutations. Acinar was enriched with <em>EGFR</em> mutations, whereas papillary was associated with <em>RET</em> fusions. Our study demonstrated that deep learning–based segmentation can accurately quantify histologic patterns in lung adenocarcinoma and identify additional prognostic features. By simultaneously providing an objective assessment of various tasks, our comprehensive methodology in lung adenocarcinoma paves way for deep learning–assisted pathologic diagnosis and treatment guidance.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 12","pages":"Article 100907"},"PeriodicalIF":5.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145275333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Clinicopathologic and Molecular Genetic Features of Spindle Cell Rhabdomyosarcoma Harboring ZFP64::NCOA2/3 Fusions: A Series of 14 Cases 含有ZFP64::NCOA2/3融合体的梭形细胞横纹肌肉瘤14例临床病理及分子遗传学特征

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-10 DOI: 10.1016/j.modpat.2025.100906

Carina A. Dehner , Baptiste Ameline , Fernanda Amary , John M. Gross , Ying Zou , Michael Michal , Zdenek Kinkor , Jorge Torres-Mora , Faizan Malik , Erica Y. Kao , Robert W. Ricciotti , Nasir Ud Din , Ivy John , Brendan C. Dickson , Elizabeth G. Demicco , Abbas Agaimy , Konstantinos Linos , Meera R. Hameed , Andrew L. Folpe , Daniel Baumhoer

Spindle cell rhabdomyosarcomas (SCRMS), recognized by the 2020 World Health Organization Classification of Tumors of Soft Tissue and Bone as a distinct entity, comprise a family of malignant skeletal muscle tumors sharing spindle cell morphology. To date, members of this family include (1) MyoD1-mutated SCRMS/sclerosing rhabdomyosarcomas (RMS), (2) intraosseous SCRMS with FET::TFCP2 or MEIS1::NCOA2 fusions, and (3) infantile/congenital SCRMS harboring NCOA1/2 or VGLL3 rearrangements. A rare, emerging subtype of SCRMS has been reported to harbor recurrent ZFP64::NCOA3 fusions. We studied 14 cases of this rare SCRMS subtype. The tumors presented in 11 men and 3 women (median age, 39.5 years; range, 22-69 years) and involved the thigh (4), lower leg (2), gluteal soft tissues (2), abdominal wall (1), mediastinum (1), subperiosteal surface of third rib (1), glottis (1), prostate (1), and pelvis (1). Morphologically, 11 tumors showed uniform spindle cell morphology with a fascicular architecture, whereas the remaining 3 tumors demonstrated focal or predominant round cell morphology. Extensive chondro-osseous differentiation was seen in 2 cases. By immunohistochemistry, tumors were variably positive for both desmin and MyoD1 (6 tumors), desmin, MyoD1, and myogenin (1 tumor), desmin alone (3 tumors of which only 1 was also tested for MyoD1), or MyoD1 alone (3 tumors). Smooth muscle actin was noted in 6 of 10 tested cases, and 2 of 5 tested cases showed ALK expression. A ZFP64::NCOA3 fusion was detected in 8 tumors, and a ZFP64::NCOA2 fusion was detected in 6 tumors. Methylation studies showed all but 1 tested tumor to form a tight cluster, clearly separate from other RMS subtypes and non-RMS morphologic mimics. Clinical follow-up (10/14 cases; median, 35 months; range, 3-108 months) demonstrated local recurrence in 2 patients and distant metastases in 5 patients (median, 12 months; range, at presentation - 106 months). At the time of last follow-up, 5 patients were alive without evidence of disease, 3 patients were alive with disease, and 2 patients died of disease at 34 and 108 months. We conclude that SCRMS with ZFP64::NCOA2/3 fusions represents a distinct, clinically aggressive sarcoma characterized by fascicular and sometimes round cell morphology, occasional chondro-osseous differentiation, and variable skeletal muscle marker expression. Recognition of this emerging subtype of SCRMS may have prognostic and therapeutic implications.

纺锤形细胞横纹肌肉瘤（SCRMS）被2020年世界卫生组织软组织和骨骼肿瘤分类认定为一个独特的实体，由一个具有纺锤形细胞形态的恶性骨骼肌肿瘤家族组成。迄今为止，该家族的成员包括1)myod1突变的纺锤体细胞/硬化性RMS (SCRMS/SRMS), 2)伴有FET::TFCP2或MEIS1::NCOA2融合的骨内SCRMS，以及3)含有NCOA1/2或VGLL3重排的婴儿先天性SCRMS。据报道，一种罕见的新兴SCRMS亚型存在复发性ZFP64::NCOA3融合。我们研究了14例这种罕见的SCRMS亚型。11例男性和3例女性（年龄中位数：39.5岁，范围：22-69岁）出现肿瘤，累及大腿（4例）、小腿（2例）、臀软组织（2例）、腹壁（1例）、纵隔（1例）、第三肋骨膜下表面（1例）、声门（1例）、前列腺（1例）和骨盆（1例）。形态学上，11例肿瘤表现为均匀的梭形细胞形态和束状结构，其余3例肿瘤表现为局灶性或主要的圆形细胞形态。2例可见广泛的软骨骨分化。通过免疫组化，肿瘤中desmin和MyoD1（6个肿瘤）、desmin、MyoD1和myogenin（1个肿瘤）、desmin单独（3个肿瘤中只有1个也检测了MyoD1）或MyoD1单独（3个肿瘤）均呈阳性。10例中有6例检测到平滑肌肌动蛋白，5例中有2例检测到ALK表达。ZFP64::NCOA3融合8例，ZFP64::NCOA2融合6例。甲基化研究表明，除了一个被测试的肿瘤外，所有的肿瘤都形成了一个紧密的簇，与其他RMS亚型和非RMS形态模拟明显分开。临床随访（14例中的10例，中位35个月，范围3-108个月）显示2例局部复发，5例远处转移（中位12个月，范围：发病时- 106个月）。末次随访时，5例患者无疾病证据存活，3例患者有疾病存活，2例患者于34个月和108个月死于疾病。我们得出结论，具有ZFP64::NCOA2/3融合的SCRMS代表了一种独特的临床侵袭性肉瘤，其特征是束状细胞形态，有时是圆形细胞形态，偶尔出现软骨骨分化和骨骼肌标志物表达变化。认识到这种新出现的scms亚型可能具有预后和治疗意义。

{"title":"Clinicopathologic and Molecular Genetic Features of Spindle Cell Rhabdomyosarcoma Harboring ZFP64::NCOA2/3 Fusions: A Series of 14 Cases","authors":"Carina A. Dehner , Baptiste Ameline , Fernanda Amary , John M. Gross , Ying Zou , Michael Michal , Zdenek Kinkor , Jorge Torres-Mora , Faizan Malik , Erica Y. Kao , Robert W. Ricciotti , Nasir Ud Din , Ivy John , Brendan C. Dickson , Elizabeth G. Demicco , Abbas Agaimy , Konstantinos Linos , Meera R. Hameed , Andrew L. Folpe , Daniel Baumhoer","doi":"10.1016/j.modpat.2025.100906","DOIUrl":"10.1016/j.modpat.2025.100906","url":null,"abstract":"<div><div>Spindle cell rhabdomyosarcomas (SCRMS), recognized by the 2020 <em>World Health Organization Classification of Tumors of Soft Tissue and Bone</em> as a distinct entity, comprise a family of malignant skeletal muscle tumors sharing spindle cell morphology. To date, members of this family include (1) MyoD1-mutated SCRMS/sclerosing rhabdomyosarcomas (RMS), (2) intraosseous SCRMS with FET::<em>TFCP2</em> or <em>MEIS1::NCOA2</em> fusions, and (3) infantile/congenital SCRMS harboring <em>NCOA1/2</em> or <em>VGLL3</em> rearrangements. A rare, emerging subtype of SCRMS has been reported to harbor recurrent <em>ZFP64::NCOA3</em> fusions. We studied 14 cases of this rare SCRMS subtype. The tumors presented in 11 men and 3 women (median age, 39.5 years; range, 22-69 years) and involved the thigh (4), lower leg (2), gluteal soft tissues (2), abdominal wall (1), mediastinum (1), subperiosteal surface of third rib (1), glottis (1), prostate (1), and pelvis (1). Morphologically, 11 tumors showed uniform spindle cell morphology with a fascicular architecture, whereas the remaining 3 tumors demonstrated focal or predominant round cell morphology. Extensive chondro-osseous differentiation was seen in 2 cases. By immunohistochemistry, tumors were variably positive for both desmin and MyoD1 (6 tumors), desmin, MyoD1, and myogenin (1 tumor), desmin alone (3 tumors of which only 1 was also tested for MyoD1), or MyoD1 alone (3 tumors). Smooth muscle actin was noted in 6 of 10 tested cases, and 2 of 5 tested cases showed ALK expression. A <em>ZFP64::NCOA3</em> fusion was detected in 8 tumors, and a <em>ZFP64::NCOA2</em> fusion was detected in 6 tumors. Methylation studies showed all but 1 tested tumor to form a tight cluster, clearly separate from other RMS subtypes and non-RMS morphologic mimics. Clinical follow-up (10/14 cases; median, 35 months; range, 3-108 months) demonstrated local recurrence in 2 patients and distant metastases in 5 patients (median, 12 months; range, at presentation - 106 months). At the time of last follow-up, 5 patients were alive without evidence of disease, 3 patients were alive with disease, and 2 patients died of disease at 34 and 108 months. We conclude that SCRMS with <em>ZFP64::NCOA2/3</em> fusions represents a distinct, clinically aggressive sarcoma characterized by fascicular and sometimes round cell morphology, occasional chondro-osseous differentiation, and variable skeletal muscle marker expression. Recognition of this emerging subtype of SCRMS may have prognostic and therapeutic implications.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 12","pages":"Article 100906"},"PeriodicalIF":5.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145275323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Head-to-Head Comparison of 2 Artificial Intelligence Tools for Detecting Lymph Node Metastases in Whole-Slide Pathology Images Within and Beyond Their Intended Use 两种人工智能工具在其预期使用范围内和超出其预期用途的全切片病理图像中检测淋巴结转移的正面比较。

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-08 DOI: 10.1016/j.modpat.2025.100905

Rachel N. Flach, Milan Samuels, Natalie D. ter Hoeve, Nikolas Stathonikos, Trudy G.N. Jonges, Jan E. Freund, Gerben E. Breimer, Willeke A.M. Blokx, Frans Schutgens, Tri Q. Nguyen, Paul J. van Diest, Carmen van Dooijeweert

The increasing diagnostic workload in pathology, driven by rising cancer incidences, highlights the need for scalable, cost effective solutions. Artificial intelligence (AI) has shown promise in supporting lymph node (LN) metastasis detection, a key prognostic factor in cancer staging. However, the current Conformité Européene In Vitro Diagnostics--certified AI tools are often limited to specific tumor types, reducing their cost efficiency and clinical use. This study evaluates the performance of 2 Conformité Européene In Vitro Diagnostics-certified AI tools—Visiopharm Metastasis Detection App (VMD) and DeepPath LYDIA (DPL)—for multipurpose LN metastasis detection across 6 tumor types, both within and beyond their intended use. We retrospectively analyzed whole-slide images from 455 patients with LN metastases from melanoma, colorectal, head and neck, lung, vulvar, and breast cancer. Both sentinel and nonsentinel LNs were included, with expert pathologists establishing the reference standard, according to clinical practice. Sensitivity was calculated per case and stratified based on metastasis size. False-positive alerts (FPAs) were assessed in 1012 tumor-negative slides. Both applications demonstrated excellent sensitivity for macrometastases across tumor types. DPL showed slightly higher sensitivity for micrometastases and isolated tumor cells compared with VMD, particularly in lung cancer and melanoma. FPA rates were substantial for both tools, with VMD generally producing more alerts, especially in lung and breast cancer. Our findings suggest that a single AI tool may be suitable for LN metastasis detection across multiple tumor types, even beyond its intended use. However, high FPA rates—particularly in lung cancer (inside intended use for DPL)—may limit practical use. Prospective studies are needed to confirm workflow efficiency gains and define optimal implementation strategies. These results support a broader, pragmatic approach to AI validation and regulatory approval, potentially improving the business case for AI adoption in pathology laboratories.

由于癌症发病率的上升，病理诊断工作量不断增加，这突出了对可扩展的、具有成本效益的解决方案的需求。人工智能（AI）在支持淋巴结（LN）转移检测方面显示出希望，这是癌症分期的关键预后因素。然而，目前CE-IVD认证的人工智能工具通常仅限于特定的肿瘤类型，降低了其成本效益和临床实用性。本研究评估了两种CE-IVD认证的人工智能工具——visiopharm Metastasis Detection App （VMD）和DeepPath LYDIA （DPL）——的性能，用于检测六种肿瘤类型的多用途LN转移，无论是在其预期用途内还是超出其预期用途。我们回顾性分析了455例黑色素瘤、结直肠癌、头颈癌、肺癌、外阴癌和乳腺癌淋巴结转移患者的整张幻灯片图像。包括前哨和非前哨淋巴结，由病理学专家根据临床实践建立参考标准。每个病例计算敏感性，并根据转移大小分层。在1012份肿瘤阴性的载玻片中评估假阳性警报（fpa）。这两种应用都显示出对不同肿瘤类型的大转移具有良好的敏感性。与VMD相比，DPL对微转移和分离肿瘤细胞（ITCs）的敏感性略高，特别是在肺癌和黑色素瘤中。两种工具的FPA率都很高，VMD通常产生更多的警报，特别是在肺癌和乳腺癌中。我们的研究结果表明，单一的人工智能工具可能适用于多种肿瘤类型的淋巴结转移检测，甚至超出了其预期用途。然而，高FPA率-特别是肺癌（DPL的预期使用范围内）-可能限制实际可用性。需要前瞻性研究来确认工作流效率的提高和确定最佳的实施策略。这些结果为人工智能验证和监管批准提供了更广泛、更务实的方法，有可能改善病理实验室采用人工智能的商业案例。

{"title":"Head-to-Head Comparison of 2 Artificial Intelligence Tools for Detecting Lymph Node Metastases in Whole-Slide Pathology Images Within and Beyond Their Intended Use","authors":"Rachel N. Flach, Milan Samuels, Natalie D. ter Hoeve, Nikolas Stathonikos, Trudy G.N. Jonges, Jan E. Freund, Gerben E. Breimer, Willeke A.M. Blokx, Frans Schutgens, Tri Q. Nguyen, Paul J. van Diest, Carmen van Dooijeweert","doi":"10.1016/j.modpat.2025.100905","DOIUrl":"10.1016/j.modpat.2025.100905","url":null,"abstract":"<div><div>The increasing diagnostic workload in pathology, driven by rising cancer incidences, highlights the need for scalable, cost effective solutions. Artificial intelligence (AI) has shown promise in supporting lymph node (LN) metastasis detection, a key prognostic factor in cancer staging. However, the current Conformité Européene In Vitro Diagnostics--certified AI tools are often limited to specific tumor types, reducing their cost efficiency and clinical use. This study evaluates the performance of 2 Conformité Européene In Vitro Diagnostics-certified AI tools—Visiopharm Metastasis Detection App (VMD) and DeepPath LYDIA (DPL)—for multipurpose LN metastasis detection across 6 tumor types, both within and beyond their intended use. We retrospectively analyzed whole-slide images from 455 patients with LN metastases from melanoma, colorectal, head and neck, lung, vulvar, and breast cancer. Both sentinel and nonsentinel LNs were included, with expert pathologists establishing the reference standard, according to clinical practice. Sensitivity was calculated per case and stratified based on metastasis size. False-positive alerts (FPAs) were assessed in 1012 tumor-negative slides. Both applications demonstrated excellent sensitivity for macrometastases across tumor types. DPL showed slightly higher sensitivity for micrometastases and isolated tumor cells compared with VMD, particularly in lung cancer and melanoma. FPA rates were substantial for both tools, with VMD generally producing more alerts, especially in lung and breast cancer. Our findings suggest that a single AI tool may be suitable for LN metastasis detection across multiple tumor types, even beyond its intended use. However, high FPA rates—particularly in lung cancer (inside intended use for DPL)—may limit practical use. Prospective studies are needed to confirm workflow efficiency gains and define optimal implementation strategies. These results support a broader, pragmatic approach to AI validation and regulatory approval, potentially improving the business case for AI adoption in pathology laboratories.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 12","pages":"Article 100905"},"PeriodicalIF":5.5,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145275388","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Genomic Profiling of Pediatric Mycosis Fungoides, Lymphomatoid Papulosis, and Primary Cutaneous Anaplastic Large Cell Lymphoma Identifies Recurrent Tyrosine Kinase Gene Fusions 儿童蕈样真菌病、淋巴瘤样丘疹病和原发性皮肤间变性大细胞淋巴瘤的基因组分析确定了复发性酪氨酸激酶基因融合。

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-08 DOI: 10.1016/j.modpat.2025.100908

Grant M. Fischer , Harrison K. Tsai , Jennifer Huang , Mark C. Mochel , Sam Sadigh , Zenggang Pan , Meagan Montesion , Erik A. Williams , Mark Sabbagh , Paolo Chetta , Va Lip , Judith A. Ferry , Julie D.R. Reimann , Lyn Duncan , Steven Chen , Thomas Kupper , Marilyn G. Liang , Lynda M. Vrooman , Jessica Pollard , Dimitra Pouli , Jacob R. Bledsoe

Pediatric cutaneous T-cell lymphoproliferative disorders encompass a diagnostically complex set of rare diseases of undefined pathogenesis, including mycosis fungoides (MF), lymphomatoid papulosis (LyP), and primary cutaneous anaplastic large cell lymphoma (pcALCL). In the pediatric population, these disorders are much less common than in adults, which has precluded systematic evaluation of their molecular pathogenesis. We report the clinicopathologic and molecular features of pediatric MF (n = 14, ages 5-17 years at diagnosis), LyP (n = 8, ages 4-17 years), and pcALCL (n = 2). Next-generation sequencing analysis (targeted 72-gene fusion panel, targeted 447-gene exome sequencing panel, and/or FoundationOneHeme) was performed. JAK2 fusions were detected in 64% (7/11) MF (PCM1::JAK2, ILF3::JAK2 [n = 2], ATXN2L::JAK2 [n = 2], and NUP214::JAK2 [n = 2]) by next-generation sequencing of available cases. TYK2 fusions were identified in 1 of 11 MF (novel LMNA::TYK2), 3 of 5 LyP (NPM1::TYK2 [n = 2], and novel RAN::TYK2), and 1 of 2 pcALCL (RAN::TYK2) cases tested. A novel NUP214::FRK fusion was observed in the other pcALCL. Fusions were in-frame and retained the kinase domain of the 3’ partner in all cases. MF cases demonstrated clonal TCR gene rearrangements (12/12 tested). Treatment was heterogeneous, although it usually included narrowband ultraviolet B phototherapy for MF and topical steroids for MF and LyP. We demonstrate that pediatric MF, LyP, and pcALCL harbor frequent tyrosine kinase gene fusions with enrichment of JAK2 and TYK2 fusions, genomic alterations that are diagnostically useful and may be amenable to targeted therapy.

儿童皮肤t细胞淋巴增生性疾病包括一组诊断复杂的发病机制不明的罕见疾病，包括蕈样真菌病（MF）、淋巴瘤样丘疹病（LyP）和原发性皮肤间变性大细胞淋巴瘤（pcALCL）。在儿童人群中，这些疾病比成人少得多，这妨碍了对其分子发病机制的系统评估。我们报告了小儿MF （n=14，诊断时年龄5-17岁）、LyP （n=8，年龄4-17岁）和pcALCL （n=2）的临床病理和分子特征。进行下一代测序分析（靶向72基因融合面板，靶向447基因外显子组测序面板，和/或FoundationOneHeme）。通过对现有病例的下一代测序，在64% (7/11)MF [PCM1::JAK2, ILF3::JAK2 (n=2), ATXN2L::JAK2 (n=2), NUP214::JAK2 (n=2)]中检测到JAK2融合。在1/11 MF[新型LMNA::TYK2]、3/5 LyP [NPM1::TYK2 （n=2）和新型RAN::TYK2]和1/2 pcALCL [RAN::TYK2]病例中鉴定出TYK2融合。在另一个pcALCL中观察到新的NUP214::FRK融合。在所有情况下，融合在框架内并保留了3'伴侣的激酶结构域。MF病例显示克隆性TCR基因重排（12/12检测）。治疗是异质性的，尽管通常包括窄带UVB光疗治疗MF，和局部类固醇治疗MF和LyP。我们证明，儿童MF、LyP和pcALCL携带频繁的酪氨酸激酶基因融合与JAK2和TYK2融合的富集，基因组改变是诊断有用的，可能适合靶向治疗。

{"title":"Genomic Profiling of Pediatric Mycosis Fungoides, Lymphomatoid Papulosis, and Primary Cutaneous Anaplastic Large Cell Lymphoma Identifies Recurrent Tyrosine Kinase Gene Fusions","authors":"Grant M. Fischer , Harrison K. Tsai , Jennifer Huang , Mark C. Mochel , Sam Sadigh , Zenggang Pan , Meagan Montesion , Erik A. Williams , Mark Sabbagh , Paolo Chetta , Va Lip , Judith A. Ferry , Julie D.R. Reimann , Lyn Duncan , Steven Chen , Thomas Kupper , Marilyn G. Liang , Lynda M. Vrooman , Jessica Pollard , Dimitra Pouli , Jacob R. Bledsoe","doi":"10.1016/j.modpat.2025.100908","DOIUrl":"10.1016/j.modpat.2025.100908","url":null,"abstract":"<div><div>Pediatric cutaneous T-cell lymphoproliferative disorders encompass a diagnostically complex set of rare diseases of undefined pathogenesis, including mycosis fungoides (MF), lymphomatoid papulosis (LyP), and primary cutaneous anaplastic large cell lymphoma (pcALCL). In the pediatric population, these disorders are much less common than in adults, which has precluded systematic evaluation of their molecular pathogenesis. We report the clinicopathologic and molecular features of pediatric MF (n = 14, ages 5-17 years at diagnosis), LyP (n = 8, ages 4-17 years), and pcALCL (n = 2). Next-generation sequencing analysis (targeted 72-gene fusion panel, targeted 447-gene exome sequencing panel, and/or FoundationOneHeme) was performed. <em>JAK2</em> fusions were detected in 64% (7/11) MF (<em>PCM1::JAK2</em>, <em>ILF3::JAK2</em> [n = 2], <em>ATXN2L::JAK2</em> [n = 2], and <em>NUP214::JAK2</em> [n = 2]) by next-generation sequencing of available cases. <em>TYK2</em> fusions were identified in 1 of 11 MF (novel <em>LMNA::TYK2</em>), 3 of 5 LyP (<em>NPM1::TYK2</em> [n = 2], and novel <em>RAN::TYK2</em>), and 1 of 2 pcALCL (<em>RAN::TYK2</em>) cases tested. A novel <em>NUP214::FRK</em> fusion was observed in the other pcALCL. Fusions were in-frame and retained the kinase domain of the 3’ partner in all cases. MF cases demonstrated clonal <em>TCR</em> gene rearrangements (12/12 tested). Treatment was heterogeneous, although it usually included narrowband ultraviolet B phototherapy for MF and topical steroids for MF and LyP. We demonstrate that pediatric MF, LyP, and pcALCL harbor frequent tyrosine kinase gene fusions with enrichment of <em>JAK2</em> and <em>TYK2</em> fusions, genomic alterations that are diagnostically useful and may be amenable to targeted therapy.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 12","pages":"Article 100908"},"PeriodicalIF":5.5,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145275383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Corrigendum to “Spatial Evaluation and Prognostic Significance of V-Domain Immunoglobulin Suppressor of T-Cell Activation (VISTA) in Human Resectable Cervical Carcinoma: Implications for Immune Activation and Suppression” [Modern Pathology 2025;38(12):100851] “t细胞活化的v域免疫球蛋白抑制因子（VISTA）在人类可切除宫颈癌中的空间评价和预后意义：免疫激活和抑制的含义”的更正[现代病理学2025；38(12):100851]。

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-07 DOI: 10.1016/j.modpat.2025.100892

Qingsheng Xie , Jinqing Li , Jieyao Li , Chaoqun Liu , Yangyang Li , Hong Zeng , Jingwei Yu , Yingchen Wu , Kaiqian Chen , Zhaonan Zhang , Bo Wang

引用次数: 0

Corrigendum to “Magnetic Resonance Imaging (MRI)–Adapted Prostate Cancer Risk Tool Incorporating Cribriform and Intraductal Carcinoma” (Modern Pathology 2025 Dec;38(12):100852) “磁共振成像(MRI)-纳入筛状癌和导管内癌的前列腺癌风险工具”的勘误表（现代病理学2025年12月；38(12):100852）。

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-07 DOI: 10.1016/j.modpat.2025.100893

Ngoc-Nhu Jennifer Nguyen , Kristen Liu , Katherine Lajkosz , Rui Bernardino , Leyi Bellinda Yin , Eva Hollemans , Lisa J. Kroon , Neil Fleshner , Geert J.L.H. van Leenders , Kenneth A. Iczkowski , Theodorus H. van der Kwast , Michelle R. Downes

引用次数: 0

A Rational Approach to the Diagnosis of Liver Metastases 一种合理的肝转移诊断方法。

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-02 DOI: 10.1016/j.modpat.2025.100903

Shaomin Hu, Daniela S. Allende

The diagnosis of liver metastases encompasses a broad spectrum of entities and relies on clinicopathological correlation, with some cases being diagnosable using hematoxylin and eosin staining alone. For more challenging cases, ancillary testing is often required, with a focus on maximizing diagnostic yield while preserving tissue. This review draws on years of experience at a high-volume, tertiary referral center and incorporates an in-depth analysis of the existing literature in the field. When evaluating a liver lesion, the first step is to determine whether it represents a primary hepatic tumor or a metastasis. This review discusses a practical panel of immunohistochemical stains valuable in making this distinction. Next, it outlines a morphologic pattern-based approach to metastatic liver tumors, categorized by epithelioid, spindle, undifferentiated, and small round blue cell morphology. For each category, readers will find a proposed list of differentials based on the morphologic pattern, a suggested screening immunohistochemical panel, a focused discussion of potential pitfalls and practical tips for ancillary testing (“important aspects of ancillary testing”), and additional insights on specific entities within each group (“special diagnostic considerations”). This review provides a comprehensive overview of the diagnosis of liver metastases, along with a current guide to immunohistochemical and molecular testing for the classification of these tumors. It underscores the importance of careful consideration and prioritizing site-agnostic biomarkers and tumor lineage classification (eg, carcinoma, sarcoma, lymphoma, or melanoma) when choosing ancillary stains in challenging cases with limited material.

肝转移的诊断包括广泛的实体，依赖于临床病理相关性，有些病例仅使用苏木精和伊红染色即可诊断。对于更具挑战性的病例，通常需要辅助检测，重点是在保留组织的同时最大限度地提高诊断率。这篇综述借鉴了多年来在一个高容量的三级转诊中心的经验，并结合了对该领域现有文献的深入分析。当评估肝脏病变时，第一步是确定它是否代表原发性肝脏肿瘤或转移。这篇综述讨论了一组实用的免疫组织化学染色，对进行这种区分很有价值。接下来，它概述了一种基于形态学模式的转移性肝肿瘤的方法，根据上皮样、纺锤形、未分化和小圆蓝色细胞形态进行分类。对于每个类别，读者将发现基于形态学模式的建议差异列表，建议筛选免疫组织化学小组，集中讨论潜在的陷阱和辅助测试的实用技巧（“辅助测试的重要方面”），以及对每组中特定实体的额外见解（“特殊诊断考虑”）。这篇综述提供了肝转移的诊断的一个全面的概述，以及当前的指南免疫组织化学和分子检测的分类这些肿瘤。它强调了在材料有限的挑战性病例中选择辅助染色时，仔细考虑和优先考虑与部位无关的生物标志物和肿瘤谱系分类（例如，癌、肉瘤、淋巴瘤或黑色素瘤）的重要性。

{"title":"A Rational Approach to the Diagnosis of Liver Metastases","authors":"Shaomin Hu, Daniela S. Allende","doi":"10.1016/j.modpat.2025.100903","DOIUrl":"10.1016/j.modpat.2025.100903","url":null,"abstract":"<div><div>The diagnosis of liver metastases encompasses a broad spectrum of entities and relies on clinicopathological correlation, with some cases being diagnosable using hematoxylin and eosin staining alone. For more challenging cases, ancillary testing is often required, with a focus on maximizing diagnostic yield while preserving tissue. This review draws on years of experience at a high-volume, tertiary referral center and incorporates an in-depth analysis of the existing literature in the field. When evaluating a liver lesion, the first step is to determine whether it represents a primary hepatic tumor or a metastasis. This review discusses a practical panel of immunohistochemical stains valuable in making this distinction. Next, it outlines a morphologic pattern-based approach to metastatic liver tumors, categorized by epithelioid, spindle, undifferentiated, and small round blue cell morphology. For each category, readers will find a proposed list of differentials based on the morphologic pattern, a suggested screening immunohistochemical panel, a focused discussion of potential pitfalls and practical tips for ancillary testing (“important aspects of ancillary testing”), and additional insights on specific entities within each group (“special diagnostic considerations”). This review provides a comprehensive overview of the diagnosis of liver metastases, along with a current guide to immunohistochemical and molecular testing for the classification of these tumors. It underscores the importance of careful consideration and prioritizing site-agnostic biomarkers and tumor lineage classification (eg, carcinoma, sarcoma, lymphoma, or melanoma) when choosing ancillary stains in challenging cases with limited material.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 12","pages":"Article 100903"},"PeriodicalIF":5.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145228671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Artificial Intelligence (AI)-Driven Screening of Equivocal Prostate Immunohistochemistry (IHC) Cases: Development and Validation of a Screening and Cancer Detection Framework 人工智能对模棱两可的前列腺IHC病例的筛查：筛查和癌症检测框架的开发和验证。

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-01 DOI: 10.1016/j.modpat.2025.100904

Ramin Nateghi , Ruoji Zhou , Madeline Saft , Marina Schnauss , Clayton Neill , Ridwan Alam , Nicole Handa , Mitchell Huang , Eric V. Li , Jeffery A. Goldstein , Hiten D. Patel , Edward M. Schaeffer , Menatalla Nadim , Fattaneh Pourakpour , Bogdan Isaila , Christopher Felicelli , Vikas Mehta , Behtash G. Nezami , Ashley E. Ross , Ximing J. Yang , Lee A.D. Cooper

Artificial intelligence (AI) has been proposed as a solution to meet increasing demand for the diagnostic services of pathologists (B.I., C.F., V.M., B.G.N., X.J.Y.). Prostate biopsies are a significant source of this demand, and a substantial fraction of these biopsies require immunohistochemistry (IHC) staining, which adds work, time, and cost to the diagnostic process. Equivocal cases, which often prompt the use of IHC, not only present the greatest challenge to AI cancer detection tools but also represent the area where properly trained systems could offer the most clinical value by reducing the need for ancillary testing for cancer confirmation. From August 2021 to April 2023, we scanned 25,570 slides from 1641 patients. We investigated the performance of institutionally developed models for prostate cancer detection using digital pathology images, with an aim for reducing work, turnaround time, and costs related to the ordering of IHC equivocal cases. We advanced complementary sensitive and specific models to screen these challenging cases, aiming to identify slides that could be confidently diagnosed without any ancillary IHC tests. Additionally, we compared the performance of a prostate-specific model to a general-purpose foundation model for screening and cancer detection. Our screening models correctly classified 55% of challenging equivocal blocks where IHC was ordered with a 1.4% error rate. We found that the foundation model achieved higher screening rates (ie, percentage of cases where IHC could be avoided), but this came at the cost of uniformly higher error rates and significantly greater computational demands. When trained as a standalone prostate cancer detection system, our model demonstrated high concordance with pathologist ground truth, achieving an area under the curve of 98.5%, sensitivity of 95.0%, and specificity of 97.8%. Computational models can aid in the diagnosis of prostate cancer and can effectively screen challenging prostate biopsy cases, reducing unnecessary IHC utilization and helping to optimize pathology workflows.

Plain language summary

Traditional pathologic diagnosis of prostate cancer can be labor intensive. Equivocal cases in particular often require special testing (immunohistochemistry [IHC]) to establish a diagnosis, which introduces further delay and cost. This study develops an artificial intelligence system specifically designed to screen equivocal cases and reduce the need for IHC. Our model serves as a second-read tool to help optimize pathology workflow and reduce turnaround time and costs by flagging cases where IHC can be safely avoided.

人工智能已被提出作为一种解决方案，以满足病理学家对诊断服务日益增长的需求。前列腺活检是这种需求的重要来源，这些活检的很大一部分需要免疫组织化学染色，这增加了诊断过程的工作量、时间和成本。模棱两可的病例往往促使使用免疫组织化学（IHC），这对人工智能癌症检测工具提出了最大的挑战，但也代表了经过适当训练的系统可以通过减少对癌症确认辅助测试的需求来提供最大临床价值的领域。从2021年8月到2023年4月，我们扫描了1641名患者的25,570张幻灯片。我们研究了使用数字病理图像的机构开发的前列腺癌检测模型的性能，目的是减少与免疫组织化学模糊病例排序相关的工作，周转时间和成本。我们提出了互补的敏感性和特异性模型来筛选这些具有挑战性的病例，旨在确定无需任何辅助免疫组织化学测试即可自信诊断的载玻片。此外，我们比较了前列腺特异性模型与用于筛查和癌症检测的通用基础模型的性能。我们的筛选模型正确分类了55%的具有挑战性的模糊块，其中免疫组织化学排序为1.4%的错误率。我们发现基础模型实现了更高的筛查率（即可以避免IHC的病例百分比），但这是以更高的错误率和更大的计算需求为代价的。当作为一个独立的前列腺癌检测系统进行训练时，我们的模型显示出与病理学家基础事实的高度一致性，曲线下面积为98.5%，灵敏度为95.0%，特异性为97.8%。计算模型可以帮助前列腺癌的诊断，并可以有效地筛选具有挑战性的前列腺活检病例，减少不必要的免疫组织结构利用，并有助于优化病理工作流程。

{"title":"Artificial Intelligence (AI)-Driven Screening of Equivocal Prostate Immunohistochemistry (IHC) Cases: Development and Validation of a Screening and Cancer Detection Framework","authors":"Ramin Nateghi , Ruoji Zhou , Madeline Saft , Marina Schnauss , Clayton Neill , Ridwan Alam , Nicole Handa , Mitchell Huang , Eric V. Li , Jeffery A. Goldstein , Hiten D. Patel , Edward M. Schaeffer , Menatalla Nadim , Fattaneh Pourakpour , Bogdan Isaila , Christopher Felicelli , Vikas Mehta , Behtash G. Nezami , Ashley E. Ross , Ximing J. Yang , Lee A.D. Cooper","doi":"10.1016/j.modpat.2025.100904","DOIUrl":"10.1016/j.modpat.2025.100904","url":null,"abstract":"<div><div>Artificial intelligence (AI) has been proposed as a solution to meet increasing demand for the diagnostic services of pathologists (B.I., C.F., V.M., B.G.N., X.J.Y.). Prostate biopsies are a significant source of this demand, and a substantial fraction of these biopsies require immunohistochemistry (IHC) staining, which adds work, time, and cost to the diagnostic process. Equivocal cases, which often prompt the use of IHC, not only present the greatest challenge to AI cancer detection tools but also represent the area where properly trained systems could offer the most clinical value by reducing the need for ancillary testing for cancer confirmation. From August 2021 to April 2023, we scanned 25,570 slides from 1641 patients. We investigated the performance of institutionally developed models for prostate cancer detection using digital pathology images, with an aim for reducing work, turnaround time, and costs related to the ordering of IHC equivocal cases. We advanced complementary sensitive and specific models to screen these challenging cases, aiming to identify slides that could be confidently diagnosed without any ancillary IHC tests. Additionally, we compared the performance of a prostate-specific model to a general-purpose foundation model for screening and cancer detection. Our screening models correctly classified 55% of challenging equivocal blocks where IHC was ordered with a 1.4% error rate. We found that the foundation model achieved higher screening rates (ie, percentage of cases where IHC could be avoided), but this came at the cost of uniformly higher error rates and significantly greater computational demands. When trained as a standalone prostate cancer detection system, our model demonstrated high concordance with pathologist ground truth, achieving an area under the curve of 98.5%, sensitivity of 95.0%, and specificity of 97.8%. Computational models can aid in the diagnosis of prostate cancer and can effectively screen challenging prostate biopsy cases, reducing unnecessary IHC utilization and helping to optimize pathology workflows.</div><div><strong>Plain language summary</strong></div><div>Traditional pathologic diagnosis of prostate cancer can be labor intensive. Equivocal cases in particular often require special testing (immunohistochemistry [IHC]) to establish a diagnosis, which introduces further delay and cost. This study develops an artificial intelligence system specifically designed to screen equivocal cases and reduce the need for IHC. Our model serves as a second-read tool to help optimize pathology workflow and reduce turnaround time and costs by flagging cases where IHC can be safely avoided.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 12","pages":"Article 100904"},"PeriodicalIF":5.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145225603","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Predicting the Probability of Residual Axillary Nodal Metastases in Patients With Breast Cancer Treated With Neoadjuvant Chemotherapy 预测乳腺癌新辅助化疗患者腋窝淋巴结残留转移的可能性。

IF 5.5 1区医学 Q1 PATHOLOGY

Modern Pathology

Pub Date : 2025-10-01 DOI: 10.1016/j.modpat.2025.100902

Thaer Khoury , Haiying Zhan , Xiao Huang , Farhad Ghasemi , Fareed Rajack , Guangwei Yuan , Han Yu , Janie Theriot , Ryan Bragiel , Kanako Okamoto , Muhammad Ali

In patients with breast cancer treated with neoadjuvant chemotherapy (NACT), a positive sentinel lymph node (SLN) usually requires completion axillary lymph node dissection (ALND). To enable de-escalation of this traumatic surgery, we aimed to develop a model to accurately estimate the likelihood of axillary disease after a positive SLN biopsy in the NACT setting. We retrospectively analyzed clinicopathological data from 237 patients composed of a training set of 150 patients from a single institution and a validation set of 87 patients from 2 other institutions. Variables that were collected included the histologic type, lymphovascular invasion, the number of lymph nodes (LNs) (SLN and non-SLN), positive and negative, with and without treatment effect, extranodal extension, and the calculated residual cancer burden of the largest SLN metastasis. Residual axillary disease was defined as ≥1 positive LNs in the completion ALND specimen. Univariable and multivariable statistical analyses were performed. Then, a formula for the risk of predicted probability of residual axillary disease was created using a stepwise feature selection based on the Akaike Information Criterion to select variables in the model. Residual axillary disease was identified in 120 out of 237 (50.6%) cases (73 [48.7%] in the training set and 47 [54%] in the validation set). Independent predictors of residual axillary disease in the multivariable model included the greatest dimension of the largest SLN metastasis, lymphovascular invasion, greater number of positive LNs with no treatment effect, greater number of positive LNs with treatment effect, greater number of negative LNs with treatment effect, and fewer number of negative LNs. These variables along with residual cancer burden of the largest SLN metastasis and histologic type were incorporated into the final model by stepwise feature selection. The predictive formula achieved an area under the curve of 77.6% for the training set and 69.7% for the validation set. A predicted probability value of ≤20% yielded a negative predictive value of 86.5% in the training set and 64.7% in the validation set. This corresponds to 37 (25.3%) patients who could be spared ALND in the training set and 17 (19.5%) in the validation set. Using the formula, a subset of patients treated with NACT could be spared unnecessary ALND.

在接受新辅助化疗（NACT）的乳腺癌（BC）患者中，前哨淋巴结（SLN）阳性通常需要完成腋窝淋巴结清扫（ALND）。为了降低这种创伤性手术的风险，我们旨在建立一种模型，以准确估计NACT环境下SLN活检阳性患者发生腋窝疾病的可能性。我们回顾性分析了237例患者的临床病理资料，其中包括来自单一机构的150例患者的训练集和来自其他两个机构的87例患者的验证集。收集的变量包括组织学类型、淋巴血管侵袭（LVI）、ln [SLN和非SLN，阳性和阴性，有无治疗效果，结外延伸（ENE）以及最大SLN转移（RCBsln）的计算残余癌负担。残余腋窝病变定义为完成性ALND标本中≥1个LN阳性。进行单变量和多变量统计分析。然后，利用基于Akaike信息准则（AIC）的逐步特征选择，建立了预测残余腋窝疾病风险概率（PP）的公式，并对模型中的变量进行了选择。237例（50.6%）病例中有120例（训练集73例（48.7%），验证集47例（54%））发现腋窝残留病变。在多变量模型中，残余腋窝疾病的独立预测因子包括最大SLN转移的最大维度、LVI、无治疗效果的阳性LN数量较多、有治疗效果的阳性LN数量较多、有治疗效果的阴性LN数量较多、阴性LN数量较少。通过逐步特征选择将这些变量与RCBsln和组织学类型一起纳入最终模型。该预测公式的训练集和验证集的曲线下面积（AUC）分别为77.6和69.7。当PP≤2%时，训练集的负预测值（NPV）为86.5%，验证集的负预测值为64.7%。这相当于训练集中有37例（25.3%）患者可以避免ALND，验证集中有17例（19.5%）患者可以避免ALND。通过使用该公式，一部分接受NACT治疗的患者可以避免不必要的ALND。

{"title":"Predicting the Probability of Residual Axillary Nodal Metastases in Patients With Breast Cancer Treated With Neoadjuvant Chemotherapy","authors":"Thaer Khoury , Haiying Zhan , Xiao Huang , Farhad Ghasemi , Fareed Rajack , Guangwei Yuan , Han Yu , Janie Theriot , Ryan Bragiel , Kanako Okamoto , Muhammad Ali","doi":"10.1016/j.modpat.2025.100902","DOIUrl":"10.1016/j.modpat.2025.100902","url":null,"abstract":"<div><div>In patients with breast cancer treated with neoadjuvant chemotherapy (NACT), a positive sentinel lymph node (SLN) usually requires completion axillary lymph node dissection (ALND). To enable de-escalation of this traumatic surgery, we aimed to develop a model to accurately estimate the likelihood of axillary disease after a positive SLN biopsy in the NACT setting. We retrospectively analyzed clinicopathological data from 237 patients composed of a training set of 150 patients from a single institution and a validation set of 87 patients from 2 other institutions. Variables that were collected included the histologic type, lymphovascular invasion, the number of lymph nodes (LNs) (SLN and non-SLN), positive and negative, with and without treatment effect, extranodal extension, and the calculated residual cancer burden of the largest SLN metastasis. Residual axillary disease was defined as ≥1 positive LNs in the completion ALND specimen. Univariable and multivariable statistical analyses were performed. Then, a formula for the risk of predicted probability of residual axillary disease was created using a stepwise feature selection based on the Akaike Information Criterion to select variables in the model. Residual axillary disease was identified in 120 out of 237 (50.6%) cases (73 [48.7%] in the training set and 47 [54%] in the validation set). Independent predictors of residual axillary disease in the multivariable model included the greatest dimension of the largest SLN metastasis, lymphovascular invasion, greater number of positive LNs with no treatment effect, greater number of positive LNs with treatment effect, greater number of negative LNs with treatment effect, and fewer number of negative LNs. These variables along with residual cancer burden of the largest SLN metastasis and histologic type were incorporated into the final model by stepwise feature selection. The predictive formula achieved an area under the curve of 77.6% for the training set and 69.7% for the validation set. A predicted probability value of ≤20% yielded a negative predictive value of 86.5% in the training set and 64.7% in the validation set. This corresponds to 37 (25.3%) patients who could be spared ALND in the training set and 17 (19.5%) in the validation set. Using the formula, a subset of patients treated with NACT could be spared unnecessary ALND.</div></div>","PeriodicalId":18706,"journal":{"name":"Modern Pathology","volume":"38 12","pages":"Article 100902"},"PeriodicalIF":5.5,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145225609","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0