Cutaneous melanocytic tumors with concomitant NRAS Q61 and IDH1 R132C mutations have been described as intermediate-grade melanocytomas with characteristic biphasic morphology, but the malignant end of this genotype-defined spectrum remains poorly characterized. We assessed histopathologic, immunohistochemical, molecular, and clinical features of 16 primary cutaneous melanocytic tumors harboring both mutations. Following integrated review, 7 tumors were classified as melanocytoma and 9 as melanoma. Melanocytomas showed reproducible biphasic architecture with congenital nevus-like features, a biphasic HMB-45 pattern, low Ki-67, PRAME negativity, retained p16, and minimal copy number variations (CNVs). Melanomas retained partial morphologic overlap in a subset but were distinguished by higher-grade cytology, immunohistochemical features supportive of malignancy, and progression-associated genomic alterations, including TERT promoter mutation (9/9), 9p21/CDKN2A loss (4/7), and higher CNV burden. NRAS and IDH1 variant allele frequencies were strongly concordant (r = 0.83, P < 0.001), supporting their presence in the same dominant clone. Clinically, two patients presented with stage IIIB disease, but no distant metastasis or melanoma-related death occurred during a median melanoma follow-up of 3.9 years (IQR, 2.5-5.1). In exploratory analyses, moderate-to-severe atypia (RR, 6.2; 95% CI, 1.0-38.8; P = .009), Ki-67 ≥10% (RR, 4.4; 95% CI, 1.1-18.4; P = .003), lymphocytic infiltrate (RR, 2.4; 95% CI, 1.1-5.3; P = .03), absence of the typical biphasic pattern (RR, 2.4; 95% CI, 1.1-5.3; P = .03), and complete p16 loss (RR, 2.4; 95% CI, 1.1-5.3; P = .03) were associated with molecular or clinical progression to melanoma, defined as the presence of at least one of the following: TERT promoter mutation, pathogenic CDKN2A mutation, 9p21/CDKN2A loss, ≥3 genome-wide segmental CNVs, or any metastasis. These findings support the existence of NRAS/IDH1 co-mutated melanoma as the malignant counterpart of NRAS/IDH1-mutated melanocytoma within a single genotype-defined spectrum.
We present the clinicopathologic, immunohistochemical, and molecular findings of 27 primary cutaneous apocrine carcinoma (PCACs). Most patients were male (67%) and the axilla was the most common site (52%). Regional lymph node metastasis was seen in 14 patients (56%) and both lymph node and distant metastasis in one patient (4%). Eight patients (32%) were deceased at last follow-up. PCACs exhibit various morphologies including papillary, tubular, cribriform, solid, and signet ring cell/histiocytoid. The signet ring cell/histiocytoid morphology was seen adjacent to a tubular component in six cases (22%). An in-situ component was seen in 14 cases (52%). Perineural invasion and lymphovascular invasion were seen in 9 (33%) and 6 (22%) cases, respectively. All PCACs showed diffuse and strong androgen receptor (AR) expression (23/23; 100%); estrogen receptor (ER) (6/25; 24%) and progesterone receptor (PR) expression (4/23; 17%) was seen in subset of cases. Six cases showed 3+ HER2 expression; three showed HER2 amplification. Gross cystic disease fluid protein-15 (GCDFP-15) expression was seen in 20/20 (100%). Adipophilin was seen in 1/15 (7%), keratin 5/6 in 13/17 (76.5%), and mammaglobin in 5/16 (31%). Targeted DNA sequencing and fusion assays were performed on 12 and 9 cases, respectively. Recurrently mutated genes include ERBB2 (5/11, 45%), KMT2C (5/11, 45%), PIK3CA (3/12, 25%), TP53 (2/12, 17%), and SETD2 (2/11, 17%). One case harbored both an ERBB2::PPP1R1B fusion and an AR-V7 splice variant. Another case harbored only an AR-V7 splice variant; a RARA intronic deletion was seen in one case. In summary, PCACs frequently occur in the axilla of males, demonstrate a variety of histomorphologies, and may have an in-situ component; they are AR+, ER +/-, PR +/- and can have HER2 amplification. ERBB2, KMT2C, and PIK3CA recurrent alterations were observed. These findings may allow for both distinction from histopathologic mimics and for the use of targeted therapy.
Accurate evaluations of stromal cellularity, stromal cell atypia, and tumor border are critical for the classification of fibroepithelial lesions. This study evaluated the agreement among 19 subspecialized academic breast pathologists for evaluation of these morphologic features. Nineteen pathologists evaluated 54 images for both stromal cellularity and stromal cell atypia and 23 images for tumor border. Virtual training sessions were provided to the participants. The training sessions included a separate set of 57 images for stromal cellularity and stromal cell atypia and 16 images for tumor border; these images were different from the images used in evaluations. After the training sessions and at least 3 weeks of washout period, the 19 pathologists re-evaluated the same images. We used Krippendorff's alpha (α) to assess agreement in ordinal evaluations (mild vs moderate vs marked stromal cellularity and stromal cell atypia); and Fleiss's kappa (κ) for binary evaluations (mild/moderate vs marked stromal cellularity and stromal cell atypia, and infiltrating vs well-circumscribed tumor border). Before training, agreement on stromal cellularity was moderate in both ordinal (α=0.727) and binary evaluations (κ=0.675); agreement on stromal cell atypia was poor in ordinal (α=0.570) but moderate in binary evaluation (κ=0.446); agreement on tumor border was moderate (κ=0.532). After training, agreement on stromal cellularity remained moderate in both ordinal (α=0.747) and binary evaluations (κ=0.675); agreement on stromal cell atypia remained poor in ordinal (α=0.554) and moderate in binary evaluation (κ=0.427); agreement on border status was improved but remained moderate (κ=0.615). After training sessions, 15 pathologists became more conservative in evaluating stromal cellularity (assigning lower cellularity to the same cases); and 13 became more conservative in evaluating stromal cell atypia. To our knowledge, this is the first systematic study specifically evaluating interobserver agreement in the assessment of key morphologic criteria used in the classification of fibroepithelial lesions. There was moderate agreement in evaluating stromal cellularity and stromal cell atypia using a two-tier system (mild/moderate vs. marked) and moderate agreement in evaluating tumor border. Structured training had limited impact on improving agreement among highly subspecialized breast pathologists but promoted more conservative interpretations. The agreement benchmarks and reference images established from this study may help build consensus on diagnostic criteria for fibroepithelial lesions.
Pheochromocytomas and paragangliomas (PGLs) are rare neural crest-derived neuroendocrine neoplasms with highly variable clinical behavior and limited therapeutic options in metastatic disease. From a pathology perspective, diagnosis and risk assessment have traditionally relied on histomorphology and immunohistochemistry. However, conventional parameters remain insufficient to reliably predict biological behavior. Over the past decade, genomic and transcriptomic profiling have refined molecular classification, yet genotype-based stratification alone does not fully explain heterogeneity in tumor progression, immune composition, or treatment response.This review examines the evolution of experimental and analytical approaches to PGLs through a pathology-centered lens. We first discuss how advances in developmental biology and molecular classification have reshaped understanding of tumor heterogeneity beyond morphology. We then highlight the growing recognition of the tumor microenvironment as a critical determinant of tumor behavior. In this context, spatially resolved protein profiling, including multiplex immunofluorescence (mIF), has emerged as an important extension of traditional immunohistochemistry, enabling simultaneous visualization of immune and tumor cell populations within preserved tissue architecture. These approaches provide functional insight that cannot be derived from bulk genomic analyses alone.Despite these advances, translation into therapeutic innovations has been hampered by the lack of physiologically relevant human model systems. Early two-dimensional cell cultures and rodent models provided foundational mechanistic insights but fail to recapitulate human tumor architecture and microenvironmental interactions. The recent establishment of patient-derived PGL organoids represents a significant step forward, preserving chromaffin lineage features and enabling ex vivo pharmacologic interrogation.Taken together, the progression from histopathological assessment to spatial immune profiling and advanced three-dimensional human modeling defines a conceptual framework in which functional therapeutic testing may become integrated into future pathology-driven translational research in PGLs.
Salivary duct carcinoma (SDC) is a rare and aggressive malignancy, with up to half of cases being HER2-positive. Trastuzumab combined with docetaxel (Tmab + DTX) is a key therapeutic regimen for unresectable HER2-positive SDC; however, robust predictors of the treatment response in this patient population remain undefined. Although HER2 intratumoral heterogeneity (ITH) and the degree of HER2 amplification are known to correlate with the response to HER2-targeted agents in breast and gastric cancers, their predictive relevance in SDC remains unclear. To address this issue, we retrospectively analyzed 98 patients with unresectable HER2-positive SDC treated with Tmab + DTX. In this study, HER2 ITH was defined based on the 2023 ASCO/CAP breast cancer guidelines as the presence of spatially distinct HER2-positive and HER2-negative regions, each comprising ≥10% of the tumor area, evaluated using both immunohistochemistry (IHC) and dual-color in situ hybridization (DISH). HER2 ITH was identified in 11 of the 98 cases (11.2%). The results of IHC and DISH demonstrated strong concordance, with the exception of 1 sarcomatoid case. HER2 ITH was identified only in resected primary tumor specimens. In HER2 ITH-positive cases, no HER2 ITH was observed in any available synchronous or metachronous metastatic lesions, and the HER2 status was concordant across metastatic lesions within each patient, with most lesions being HER2-positive. The cohort exhibited an objective response rate of 80.6%, with median progression-free survival and overall survival of 9.9 and 43.8 months, respectively. No significant associations were observed between HER2 ITH or other HER2 parameters (HER2 IHC score, HER2 copy number, or HER2/CEP17 ratio) and any clinical outcome measures. The predominance of HER2-positive metastatic lesions may explain the lack of association between HER2 ITH in primary tumors and treatment efficacy. In this largest cohort of HER2-positive SDC patients treated with Tmab + DTX, HER2 ITH was present in a clinically relevant subset of cases. In contrast to breast and gastric cancers, neither HER2 ITH nor HER2 amplification levels were associated with the therapeutic response. Our findings suggest that the presence of HER2 ITH may not preclude the use of HER2-targeted therapy in SDC.
In recent years, rare fibrosarcoma-like tumors have been identified in the gynecologic tract characterized by variable CD34 and S-100 positivity and recurrent tyrosine kinase gene fusions in many cases. We identified 17 of these tumors by searching for gynecologic sarcomas expressing S-100 and/or CD34 and supplementing these with cases contributed by collaborating institutions. Tumors involved the uterine cervix (n=11), uterine corpus (n=4), and vagina (n=2). Patients ranged from 28 to 64 (median: 50) years. Fifteen tumors showed overlapping morphologic features including haphazard or fascicular growth of relatively monotonous spindled to ovoid cells with minimal to moderate cytoplasm. Tyrosine kinase gene fusions were identified in 9 tumors, including NTRK fusions in 6 and COL1A1::PDGFB in 3 tumors. Oncogenic variants in ERBB2/ERBB3 were identified in 6 additional tumors. These two subgroups showed distinct immunohistochemical profiles, with all tyrosine kinase fusion-positive tumors being CD34 positive, variably S-100 positive, and SOX10 negative. In contrast, the ERBB-altered tumors were CD34 negative and diffusely positive for S-100 and SOX10. The remaining 2 tumors showed morphologic and immunohistochemical overlap with the tyrosine kinase fusion-positive sarcomas; one was negative for fusion or sequence variants, and sequencing failed in the other. In conclusion, most fibrosarcoma-like tumors of the gynecologic tract are defined by either tyrosine kinase fusions or ERBB2/3 mutations. Immunohistochemistry can be used to predict these subgroups to guide confirmatory genetic testing and targeted therapies.
Intrahepatic cholangiocarcinoma is a rare liver cancer that is difficult to diagnose due to its morphologic and immunophenotypic overlap with metastatic tumors from other primary sites, a challenge further complicated by the liver's role as one of the most common destinations for metastatic spread. FGFR fusions and IDH1/2 mutations are characteristic of intrahepatic cholangiocarcinoma, and recent studies have shown that inhibitors targeting these alterations significantly improve survival. However, diagnosis remains challenging and often relies on molecular techniques that are not universally available, limiting the potential impact of these therapies. Albumin mRNA in situ hybridization (albumin ISH) is a well-established test that aids in the diagnosis of cholangiocarcinoma, yet its expression level has not been systematically evaluated in relation to tumor genetic background. Using a cohort of 63 cholangiocarcinomas and 157 tumors of other origins, we demonstrate that nearly all (93%) cholangiocarcinomas with FGFR fusions and IDH1/2 mutations (8 and 20 cases, respectively) showed strong albumin ISH positivity, whereas tumors with other characteristic cholangiocarcinoma alterations in BAP1, ARID1A, or PBRM1 were less likely to be positive and showed comparatively weaker expression. These findings suggest that albumin ISH may serve as a valuable screening tool for identifying tumors with actionable targets, while also posing an intriguing link between albumin expression and specific carcinogenesis pathways.
Myxoid pleomorphic liposarcoma (MPLPS), a rare, aggressive liposarcoma subtype, is thought to occur chiefly in adolescents with a mediastinal predilection. MPLPS lack DDIT3 rearrangements or MDM2 amplification; a small number have recently been shown to harbor widespread copy-neutral loss of heterozygosity (cnLOH). We studied a large series of MPLPS and compared them to other sarcomas, particularly other LPS. Available slides/blocks for cases coded as "MPLPS" were retrieved (2008-2026). We also searched our single nucleotide polymorphism (SNP) assay records for tumors demonstrating 1) high cnLOH frequency (>30%), and 2) "liposarcoma" or "pleomorphic sarcoma" diagnosis. Non-mesenchymal tumors were excluded. Twenty-four MPLPS cases were identified, occurring in 10 females (42%) and 14 males (58%), ranging from 5-85 years of age (median 46 years). Involved anatomical locations included the mediastinum/thorax (n=12), trunk (n=3), head and neck (n=3), various intra-abdominal sites (n=3), retroperitoneum/pelvis (n=2) and extremities (n=1). Two patients had clinical features of Li-Fraumeni syndrome or a germline TP53 mutation. All tumors displayed characteristic features of MPLPS, including areas resembling conventional myxoid liposarcoma and hypercellular, pleomorphic liposarcoma-like foci. SNP testing (n=22) demonstrated cnLOH of >50% in 19 cases (86%), often with pseudohyperdiploidy. One tumor showed cnLOH of 38%; two otherwise typical tumors did not have widespread cnLOH. Other recurrent alterations included gains of chromosomes 1/1q, 6- 8, and 18-21 and loss of chromosome 14. Specific losses/gains involving the RB1 and TP53 loci were seen in 12 and 2 cases, respectively. Clinical follow-up (n=22; median 16 months; range 1-75 months) showed 9 patients dead of disease (median survival 12 months), 6 patients alive with disease, and 6 patients alive without disease. Local recurrences and distant metastases were seen in 6 and 8 patients, respectively. We conclude that the age range and anatomical distribution of MPLPS are considerably wider than has been previously appreciated, including tumors arising in the elderly and various non-mediastinal locations. Although most MPLPS harbor widespread cnLOH, rare otherwise-typical tumors do not. MPLPS are aggressive sarcomas with poor prognosis.
Castleman disease (CD) represents a heterogeneous group of lymphoproliferative disorders, with the idiopathic multicentric CD subtype associated with TAFRO syndrome (iMCD-TAFRO) posing significant research challenges due to its severity and tissue scarcity. This study employed integrated bulk and spatial proteomics to characterize protein expression profiles in lymph node samples across clinical and pathological subtypes. Bulk proteomics revealed subtype-specific molecular signatures, including distinct protein profiles for hyaline vascular versus plasmacytic variants, systemic complement activation distinguishing iMCD from unicentric CD (UCD), and molecular evidence supporting the classification of iMCD-IPL as a distinct entity. Specifically, iMCD-TAFRO exhibited upregulation of angiogenesis drivers (PDGFRβ, NOTCH3), interferon signaling proteins (STAT1/ISG15), and fibrosis markers (COL3A1/LOXL1). Spatial proteomics, through laser capture microdissection of follicular and vascular niches, delineated compartmentalized pathogenesis in iMCD-TAFRO. Intrafollicular vessels showed enrichment of myofibroblast markers (ACTA2) and programmed cell death regulators (GSDMD, TFRC), while interfollicular regions displayed TGF-β/SMAD3-mediated upregulation of LOXL1 linked to fibrosis. Follicular zones demonstrated complement activation (CFHR1/CFHR2) and M2 macrophage infiltration. Our integrated approach delineates the molecular heterogeneity of CD, reveals spatially resolved pathogenic drivers in iMCD-TAFRO, and provides insights into its unique vascular-fibroinflammatory pathology, offering a foundation for advancing precision diagnostics and targeted therapies.
Invasive lobular carcinoma (ILC) is a distinct histological subtype of breast cancer. However, its molecular basis remains unclear. Although loss-of-function alterations in CDH1 are considered a defining genetic event, substantial biological heterogeneity persists, which cannot be explained by mutation status or histology alone. To address this limitation, we developed continuous transcriptional framework called the ILCness score to quantitatively capture lobular-like tumor biology. Using a rigorously constrained principal component-based approach trained in the METABRIC cohort and fixed without retraining, we enabled the reproducible application of the model across independent datasets and expression platforms. Global transcriptomic analyses demonstrated that invasive ductal carcinoma (IDC) and ILC did not form discrete expression clusters but were distributed along a continuous molecular spectrum. ILCness faithfully captured this continuum, with tumors of mixed histology occupying intermediate positions. Importantly, lobular-like transcriptional programs quantified by ILCness were not redundant with CDH1 mutation status. Across cohorts, higher ILCness was associated with coordinated biological programs, including reduced DNA repair and E2F target activity, as well as context-dependent modulation of epithelial-mesenchymal transition. Within histologically defined IDC, ILCness delineated biologically and clinically distinct subgroups independently of CDH1 mutation status, with divergent survival outcomes and therapeutic response patterns. These findings establish lobular-like breast cancer as a continuous transcriptional state rather than as a discrete entity defined by histology or single-gene alterations. The ILCness score provides a scalable RNA-based framework for integrating molecular heterogeneity with tumor biology and clinical behavior.
Claudin 18 isoform 2 (CLDN18.2) is a novel therapeutic target for advanced, HER2-negative gastric/gastroesophageal junction (GEJ) adenocarcinoma positive for CLDN18.2 immunohistochemical (IHC) expression, defined as ≥75% tumor cells with moderate-to-strong membranous staining. Clinical studies for emerging CLDN18.2-targeted therapeutics have used less stringent enrollment criteria to test the efficacy of such therapies in patients with moderate-to-low CLDN18.2 IHC expression. Anticipating the advent of such treatments, this study aimed to provide a comprehensive survey of CLDN18.2 expression using a clinical trial-validated CLDN18.2 monoclonal antibody in surgically resected gastric/GEJ adenocarcinomas and characterize CLDN18.2-positive carcinomas using both high- and low-expression thresholds. Crisp membranous CLDN18.2 staining was detected in 59% (335/564) carcinomas, including 57% (164/286) of gastric and 62% (171/278) of GEJ adenocarcinomas. Most gastric (147/164, 90%) and GEJ (166/171, 97%) adenocarcinomas with staining demonstrated moderate or strong staining intensity. Using the high-expression threshold (≥75% tumor cells with moderate-to-strong staining), positive CLDN18.2 expression was observed in 20% (57/286) of gastric and 27% (75/278) of GEJ adenocarcinomas and was associated with Epstein-Barr virus status (P < .001) and stage I gastric adenocarcinomas (P = .02) but stage IV GEJ adenocarcinomas (P = .03). Using a low-expression threshold (≥10% tumor cells with membranous staining), positive CLDN18.2 remained significantly associated with stage I (P = .004) and showed an unadjusted association with improved disease-specific survival in gastric adenocarcinomas (P = .045), which was not retained in multivariable analysis. Whole transcriptomic analysis showed concordance between IHC and CLDN18 messenger RNA expression. Transcriptomic alterations in CLDN18.2 IHC-positive gastric adenocarcinomas included pathways in drug resistance and tumor invasion. In summary, our study presents a detailed characterization of the prevalence and distribution of CLDN18.2 IHC expression patterns. Our results showed that 59% surgically resected gastric/GEJ adenocarcinomas exhibited CLDN18.2 staining. CLDN18.2 IHC positivity defined by both high- and low-expression thresholds may be associated with early-stage gastric adenocarcinoma. These findings expand our recognition of patients who may benefit from CLDN18.2-targeted therapy.
T-cell receptor beta-chain constant region 1 (TRBC1) immunohistochemistry (IHC) identifies clonal αβ T-cell populations on tissue sections, but its real-world performance across cutaneous T-cell lymphoma (CTCL) and related infiltrates is uncharacterized. The analytic cohort comprised 665 biopsies (566 patients) with paired T-cell receptor (TCR) clonality testing, classified clinicopathologically as mycosis fungoides (MF-Patch, MF-Plaque, MF-Tumor, MF-Folliculotropic), MF/Sézary syndrome, primary cutaneous small/medium T-cell lymphoproliferative disorder, other CTCL-cutaneous lymphoproliferative disorder, or reactive. At the primary <15%/>85% threshold, TRBC1 IHC achieved 85.8% sensitivity (337/393), 79.8% specificity (217/272), 86.0% positive and 79.5% negative predictive value, and 83.3% accuracy, lowest in MF-Patch (84.2%). Three-reader agreement (Fleiss' κ = 0.943) fell to κ = 0.776 in 176 reflexed biopsies, with disagreement concentrated in MF-Patch and CD30-positive lymphoproliferative disorders. Monotypic TRBC1 predicted neoplasia, with odds rising with infiltrate density: MF-Patch (odds ratio 5.41), MF-Plaque (10.40), MF/Sézary syndrome with MF-Tumor (15.19), and CTCL-cutaneous lymphoproliferative disorder (18.16). Polytypic TRBC1 was associated with reactive disease (odds ratio 53.65), effectively excluded clonality (negative likelihood ratio 0.18); and was uniformly observed in an independent 270-biopsy reactive cohort. For observer-independent validation, digital image analysis-derived TRBC1 quantification (QuPath) was applied to a stratified random subset of 250 biopsies representative of the cohort's tumor-burden distribution. The digital read tracked molecular clonality (85.1% sensitivity, 80.1% specificity against TCR; area under the curve 0.842) and agreed with the dermatopathologist manual read in 87.6% of cases (κ = 0.752), with the data-derived cutoff matching the pre-specified <15%/>85% threshold. Discordance was directional for both manual scoring and digital quantification: in MF-Patch, 25/38 (65.8%) and 12/17 (70.6%) cases were polytypic with monoclonal TCR (false-negative-dominant); in reactive biopsies, 34/45 (75.6%) and 19/20 (95%) were monotypic with polyclonal TCR (false-positive-dominant). These findings support a TRBC1-first approach, reserving reflex TCR testing for borderline expression or clinicopathologic discordance, preserving diagnostic accuracy while reducing molecular testing and reimbursement-based costs.
The primary sites-of-origin of mucinous adenocarcinomas are often difficult to determine, as there are no reliable immunohistochemical markers. The aim of our project is to use a panel of 4 methylation-based droplet digital PCRs (methyl-ddPCRs) to distinguish pulmonary mucinous adenocarcinomas from gastrointestinal (GI) metastases to the lung. The 4 PCRs developed use methylation-independent primers for amplification and methylation-dependent probes to measure signals at 4 individual CpG sites, predicted to be differentially methylated across lung (LUAD), gastroesophageal (GEAD), pancreatic (PAAD), and/or colorectal (CORE) adenocarcinomas based on The Cancer Genome Atlas methylation array data. The PCRs were tested on an internal cohort of 35 formalin-fixed paraffin-embedded lung adenocarcinomas tumors, which were mostly mucinous tumors, and 22 COREs, GEADs, and PAADs that metastasized to the lung. Decision trees using PCR data were applied to predict the diagnoses of these 57 references and 12 additional challenging cases. Decision tree predictions were found to be concordant with the primary versus metastatic assignments for 83-90% of reference cases across multiple analyses. For difficult/challenging cases, decision tree predictions were concordant with the favored impressions for 5/10 cases that underwent subsequent clinicopathologic review, and suggested diagnoses for 2 remaining cases with insufficient data. Preliminary assessments also support the application of these PCRs to smaller specimens. Our experiences suggest that methyl-ddPCRs can be viable solutions for many difficult differentials in surgical pathology.
Multiparameter flow cytometry is essential for diagnosing mature B-cell lymphomas, yet analysis remains largely manual, time-consuming, and subject to inter-operator variability. We developed an automated system for B-cell neoplasm detection using a three-stage deep learning architecture that produces interpretable intermediate outputs. The Light-chain module classifies each cell's surface immunoglobulin expression. The Cell-level module combines original flow parameters with engineered features capturing light-chain neighborhood and spatial position to identify abnormal cells. The Sample-level module renders case-level diagnoses from cells ranked by abnormality score. We evaluated this system on 3,070 clinical specimens from Memorial Sloan Kettering Cancer Center, comprising peripheral blood (n=1,369), bone marrow (n=316), and tissue (n=1,385) samples. The system achieved 97.5% case-level accuracy with 95.6% sensitivity and 98.9% specificity. Predicted abnormal cell counts correlated with expert manual counts (R2= 0.962 [95% CI: 0.937-0.988], slope=0.998 [0.983-1.014], intercept=-0.011 [-0.095 to 0.073]). Ablation experiments demonstrated that engineered features provide complementary information: light-chain neighborhood grounds spatial features that otherwise introduce noise, enabling the full model to outperform any feature subset. The system generates visualization outputs including marker expression comparisons and dimensionality-reduced embeddings colored by abnormality score, allowing pathologists to verify that flagged populations form coherent clusters with phenotypes consistent with disease. Features extracted from abnormal populations also supported lymphoma subtypes prediction (AUROC=0.925). This work demonstrates that automated flow cytometry interpretation can achieve high diagnostic accuracy while providing transparent, verifiable outputs suitable for clinical integration.
DNA methylation and copy number variation (CNV) profiling has emerged as a promising tool for the classification of bone and soft tissue tumors. We evaluated its utility in cartilage tumors, where distinguishing low-grade from high-grade conventional central chondrosarcomas (CSs) and atypical cartilaginous tumors (ACTs) from enchondromas (ECs) is a frequent diagnostic challenge, particularly on biopsy material. We analyzed 214 chondrogenic tumors, including ECs, ACTs, conventional CSs, dedifferentiated chondrosarcomas (DDCSs), and clear cell CSs, and determined their IDH1/2 mutation status. Unsupervised dimensionality reduction of genome-wide DNA methylation patterns revealed 4 clusters among IDH-mutant (MUT) tumors (IDH-MUT-1: mostly ECs and ACTs and some high-grade CSs; IDH-MUT-2: predominantly high-grade CSs; IDH-MUT-3: largely DDCSs; and IDH-MUT-SB: distinct skull base group with a markedly different methylation pattern) and 2 clusters among IDH-wild-type (WT) tumors (IDH-WT-1 and IDH-WT-2: both primarily high-grade CSs, with IDH-WT-2 showing higher tumor grade and more extensive CNVs). Clear cell CSs formed a separate cluster. The amount of CNVs, including loss of CDKN2A, increased with tumor grade, reflecting increased genomic instability during chondrosarcoma progression. Supervised classifiers trained separately, both on methylation and CNV data, and distinguished low-grade and high-grade cartilaginous tumors with area under the curve values of 0.87 to 0.97 and 85% to 90% accuracy. Furthermore, we tested whether DDCSs can be distinguished from metastatic carcinomas and other high-grade sarcomas of the bone. Across 246 reference samples, a supervised classifier achieved 97.2% accuracy (area under the curve, 99.8%) and correctly identified 30 of 32 DDCSs (93.8%). These results indicate that DNA methylation and CNV data analysis provide a valuable tool for distinguishing most low- and high-grade CSs, with additional utility also in differentiating DDCS from morphologic mimics.
The pituitary is an important gland that secretes hormones which are critical for metabolism, growth and reproduction. Our knowledge of the diseases that can affect this small gland has expanded significantly, allowing us to better understand the various inflammatory and neoplastic lesions that result in altered hormone production and/or mass effects in the sella turcica at the base of the skull. In this review, we discuss recent advances in the classification of hypophysitis and of tumors that arise from pituitary adenohypophysial cells, now recognized as pituitary neuroendocrine tumors (PitNETs), as well as unusual tumors that arise in the posterior pituitary and hypothalamus. We emphasize the importance of immunohistochemistry in the accurate diagnosis of sellar lesions, and we show how proper classification of these tumors leads to a better understanding of their clinical features, pathogenesis and therapy.
The BCL6 gene is a transcriptional regulator essential for germinal center B-cell function, modulating their activation, survival, cell cycle, and differentiation into plasma or memory B-cells. In indolent B-cell lymphomas, particularly follicular lymphoma and a subset of marginal zone lymphomas, BCL6 rearrangements are increasingly recognized as recurrent but heterogeneous events. In follicular lymphoma, their clinicopathologic significance differs according to the BCL2 status. While BCL6 rearrangements have limited impact in conventional BCL2-rearranged follicular lymphoma, BCL2-negative/BCL6-rearranged follicular lymphoma frequently display atypical morphologic, immunophenotypic, and molecular features, including partial overlap with nodal marginal zone lymphoma. These cases may show reduced expression of germinal center markers, plasmacytic or marginal zone differentiation, and mutational profiles overlapping with MZL, making their classification challenging.Altogether, current data suggest that BCL6-rearranged indolent B-cell lymphomas do not represent a single entity but rather a heterogeneous biological spectrum spanning germinal center and post-germinal center features. Understanding these overlaps is essential for improving diagnostic accuracy and refining the classification of indolent B-cell lymphomas.
PDGFB or, more rarely, PDGFD rearrangements are well-established oncogenic drivers of dermatofibrosarcoma protuberans (DFSP). Recently, a TNC::PDGFD fusion has been identified in a superficial spindle cell tumor distinct from DFSP but similar to the tumor entity described as ossifying plexiform tumors of the skin. Herein, we report 7 additional cases of cutaneous ossifying plexiform tumors harboring a TNC::PDGFD fusion transcript. Four patients were female. Tumors were located on the nose (n = 2), hand (n = 2), forearm (n = 1), flank (n = 1), and foot (n = 1), with a median size of 6 (range, 4-9) mm. Microscopically, these neoplasms were located in the dermis (n = 6) and subcutaneous tissues (n = 1). All specimens exhibited a lobulated architecture and a biphasic appearance characterized by the association of cellular areas composed of bland spindle cells with central regions of bone formation. Immunohistochemistry showed SATB2 positivity in all tested tumors (n = 4). RNA-sequencing analysis confirmed the presence of the TNC::PDGFD fusion transcript in all cases and revealed a transcriptomic profile that was distinct from other soft tissue tumors, including DFSP. Our findings support that TNC::PDGFD is the oncogenic driver of ossifying plexiform tumor of the skin, a rare tumor distinct from DFSP.
