Impact of 18F-FDG PET Intensity Normalization on Radiomic Features of Oropharyngeal Squamous Cell Carcinomas and Machine Learning-Generated Biomarkers.

We aimed to investigate the effects of 18F-FDG PET voxel intensity normalization on radiomic features of oropharyngeal squamous cell carcinoma (OPSCC) and machine learning-generated radiomic biomarkers. Methods: We extracted 1,037 18F-FDG PET radiomic features quantifying the shape, intensity, and texture of 430 OPSCC primary tumors. The reproducibility of individual features across 3 intensity-normalized images (body-weight SUV, reference tissue activity ratio to lentiform nucleus of brain and cerebellum) and the raw PET data was assessed using an intraclass correlation coefficient (ICC). We investigated the effects of intensity normalization on the features' utility in predicting the human papillomavirus (HPV) status of OPSCCs in univariate logistic regression, receiver-operating-characteristic analysis, and extreme-gradient-boosting (XGBoost) machine-learning classifiers. Results: Of 1,037 features, a high (ICC ≥ 0.90), medium (0.90 > ICC ≥ 0.75), and low (ICC < 0.75) degree of reproducibility across normalization methods was attained in 356 (34.3%), 608 (58.6%), and 73 (7%) features, respectively. In univariate analysis, features from the PET normalized to the lentiform nucleus had the strongest association with HPV status, with 865 of 1,037 (83.4%) significant features after multiple testing corrections and a median area under the receiver-operating-characteristic curve (AUC) of 0.65 (interquartile range, 0.62-0.68). Similar tendencies were observed in XGBoost models, with the lentiform nucleus-normalized model achieving the numerically highest average AUC of 0.72 (SD, 0.07) in the cross validation within the training cohort. The model generalized well to the validation cohorts, attaining an AUC of 0.73 (95% CI, 0.60-0.85) in independent validation and 0.76 (95% CI, 0.58-0.95) in external validation. The AUCs of the XGBoost models were not significantly different. Conclusion: Only one third of the features demonstrated a high degree of reproducibility across intensity-normalization techniques, making uniform normalization a prerequisite for interindividual comparability of radiomic markers. The choice of normalization technique may affect the radiomic features' predictive value with respect to HPV. Our results show trends that normalization to the lentiform nucleus may improve model performance, although more evidence is needed to draw a firm conclusion.

Diagnosis of acinic cell carcinoma of the salivary gland on cytology specimens: Role of NOR-1 (NR4A3) immunohistochemistry.

OBJECTIVES: Acinic cell carcinoma (AcCC) is often a challenging diagnosis on cytology. Recently, NOR-1 (NR4A3) has been demonstrated as a sensitive and specific marker for AcCC. Therefore, we conducted this study to evaluate NOR-1 expression in AcCC cytology specimens and to compare its reactivity in other salivary gland tumours (non-AcCC). METHODS: We retrospectively reviewed our database and selected cytology cases with available cell blocks, including 10 AcCC and 24 non-AcCC tumours (12 benign tumours and 12 malignant tumours). NOR-1 (1:50 dilution; SC393902 [H-7]; Santa Cruz Biotech) immunohistochemistry (IHC) was performed on all cases. RESULTS: All AcCC cases except two (2/10, 80%) showed positive nuclear staining of variable intensity for NOR-1, with the majority of cases (75%) demonstrating at least moderately intense nuclear expression. All non-AcCC cases were negative for NOR-1, demonstrating a specificity of 100%. CONCLUSION: We conclude that NOR-1 IHC is sensitive and very specific on cytology specimens and is able to distinguish AcCC from its mimickers reliably and classify them appropriately for further management.

Copy number variations identified in thyroid FNA specimens are associated with Hürthle cell cytomorphology.

BACKGROUND: The fine-needle aspiration (FNA) diagnosis of thyroid Hürthle cell neoplasms (HCNs) remains challenging. This study explored a possible association of copy number variations (CNVs) with Hürthle cell lesions of the thyroid. METHODS: Thyroid FNA cases that were diagnosed as follicular lesion of undetermined significance (FLUS) or follicular neoplasm (FN)/HCN for which the ThyroSeq version 3 genomic classifier test was performed were retrieved. RESULTS: A total of 324 thyroid FNA cases (228 FLUS cases, 46 HCN cases, and 50 FN cases) were included in the study. FLUS cases were further classified as Hürthle cell type (follicular lesion of undetermined significance-Hürthle cell type [FLUS-HCT]; 20 cases) or non-Hürthle cell type (follicular lesion of undetermined significance-non-Hürthle cell type [FLUS-NHCT]; 208 cases). HCN and FLUS-HCT cases showed a higher prevalence of CNVs (23 of 66 [35%]) in comparison with those classified as FN or FLUS-NHCT (14 of 258 [5%]; P < .001). A total of 105 patients had histopathologic follow-up. Cases with CNVs were more likely to be neoplastic (18 of 26 [69%]) and associated with Hürthle cell changes (14 of 26 [54%]) in comparison with cases without any molecular alterations (neoplastic, 8 of 24 [33%]; Hürthle cell changes, 2 of 24 [8%]; P < .05). In HCN/FLUS-HCT cases with CNVs (n = 14), Hürthle cell changes (13 of 14 [93%]) and neoplasms (9 of 14 [64%]) were more likely to be seen on surgical follow-up in comparison with the 17 cases without CNVs (Hürthle cell changes, 6 of 17 [35%]; neoplastic, 3 of 17 [18%]; P < .05). CONCLUSIONS: CNVs identified in thyroid FNA cases are associated with Hürthle cell morphology and are suggestive of a neoplasm with Hürthle cell features in thyroid FNAs classified as FLUS-HCT/HCN. This finding may be helpful in triaging patients who would benefit from surgical management.

Anaplastic Thyroid Carcinoma: Cytomorphologic Features on Fine-Needle Aspiration and Associated Diagnostic Challenges.

OBJECTIVES: Anaplastic thyroid carcinoma (ATC) is an aggressive malignancy, and early diagnosis, often aided by fine-needle aspiration (FNA), is key to improving patient prognosis. While the current literature describes some of the cytologic features (CFs) of this entity, a comprehensive examination of the CFs has not yet been performed. METHODS: We retrospectively searched our electronic database for ATC cases with available slides between January 2008 and December 2019. Cases were examined for 22 CFs and compared with a control group of differentiated thyroid carcinoma. RESULTS: A total of 18 ATC cases meeting our inclusion criteria were identified. Most cases showed moderate to high cellularity (83%) and epithelioid cytomorphology (83%). Architecture included either predominantly groups/clusters of tumor cells (56%) or single tumor cells (44%). The other CFs were as follows: nuclear enlargement (100%), nuclear crowding (89%), nuclear membrane irregularities (100%), multinucleated tumor cells (33%), and background acute inflammatory cells (50%). Of the CFs examined, statistically significant differences between ATC and the control groups were found in the following: nuclear pleomorphism, coarse/clumped chromatin, macronucleoli, apoptosis, and necrosis. CONCLUSIONS: Identification of key CFs in FNA coupled with the clinical history aids in the diagnosis of ATC and helps distinguish it from other mimickers.

Impact of Transoral Endoscopic Vestibular Approach Thyroidectomy on Pathologic Assessment.

CONTEXT.­: Since 2016, transoral endoscopic thyroid resection with vestibular approach (TOETVA) has been increasingly performed in the United States. Although guidelines for the procedure are evolving, indeterminate and malignant preoperative cytopathologic diagnoses are not a contraindication. There are limited data related to the pathologic examination of these specimens. OBJECTIVE.­: To examine the clinicopathologic features of TOETVA specimens with particular attention to limitations of interpretation of pathologic parameters and final diagnosis. DESIGN.­: We reviewed age, sex, preoperative imaging and cytologic diagnoses, surgical pathology, and clinical follow-up data in TOETVA resections from our institution for procedures performed between March 2016 and December 2019. RESULTS.­: Fifty cases of TOETVA were identified, comprising 48 women and 2 men with a mean age of 47 years. Preoperative cytologic diagnoses were available in 47 cases and included 19 nondiagnostic/benign (Bethesda I/II), 24 follicular lesion of undetermined significance/suspicious for follicular neoplasm (Bethesda III/IV), and 4 suspicious/malignant diagnoses (Bethesda V/VI). Thirty-four cases (68%) among the surgical resection specimens showed disruption and/or fragmentation. Thirty-nine cases were negative for carcinoma, including hyperplasias and benign/indolent neoplasms. Eleven cases exhibited papillary thyroid carcinoma. Final diagnoses were reached in all disrupted/fragmented cases. In 2 cases of papillary thyroid carcinoma, tumor size, microscopic extrathyroidal extension, and margin status could not be determined. CONCLUSIONS.­: A significant proportion of TOETVA specimens are disrupted/fragmented, which can compromise information about tumors, including size, number, margin status, and microscopic extrathyroidal extension. Given that these parameters inform treatment and follow-up, this should be considered when selecting patients for TOETVA.

RAS mutation and associated risk of malignancy in the thyroid gland: An FNA study with cytology-histology correlation.

BACKGROUND: Activating point mutations of the RAS gene (NRAS, HRAS, and KRAS) can be seen in benign and malignant thyroid tumors; among these, NRAS mutations are more commonly seen. This study was conducted to evaluate the thyroid risk of malignancy (ROM) associated with RAS mutations in thyroid fine-needle aspiration (FNA) at the authors' institution. METHODS: The authors searched their electronic database system between January 2015 and May 2021 for thyroid FNA cases with any type of RAS mutation. Molecular alterations were identified with the ThyroSeq Genomic Classifier, ThyGeNEXT (thyroid oncogene panel)/ThyraMIR (miRNA classifier), or ThyroSure gene panel. RESULTS: A total of 127 cases (age, 51 ± 14 years; 100 females and 27 males) were identified, and 72 had histologic follow-up. The overall ROM associated with RAS mutations (with or without any other molecular alterations) was 29%, whereas the ROM was lower (18%) with RAS mutations only. Isolated NRAS, HRAS, and KRAS mutation-associated ROMs were 15%, 27%, and 14%, respectively. Among these RAS-mutated cases, the cases with a Bethesda category IV cytologic diagnosis had a higher ROM than the cases with a category III diagnosis (38% vs 17%). Twenty-one histologically confirmed malignant cases were mostly classified on cytology as category IV lesions (14 of 34; 41%), and the remainder were either category III (6 of 35; 17%) or V lesions (1 of 1; 100%). CONCLUSIONS: This study demonstrated that the overall RAS mutation-associated ROM in thyroid FNA was intermediate (29%), and isolated HRAS mutations appeared to have a higher ROM (27%) than NRAS and KRAS mutations (15% and 14%, respectively).

Proceedings of the North American Society of Head and Neck Pathology, Baltimore, MD, March 17, 2021: The Mistakes I Made When I Stepped Out of My Neck of the Woods.

Rapidly growing, symptomatic, non-hematological, malignant neck masses are unusual in young adults. We report a case of a 34-year-old African American male with sickle cell trait who presented with a large left supraclavicular/cervical mass comprising of poorly differentiated malignant epithelial cells consistent with metastatic carcinoma of unknown origin. Upon immunohistochemistry, the tumor showed loss of INI1 (BAF47) and retained PAX-8 expression. After extensive clinical and radiological work-up the primary tumor was found to be a 2.6 cm renal medullary carcinoma. This case highlights the role of multidisciplinary approach to the diagnosis of a neck mass and to understanding that certain genetically-defined tumors can occur at and metastasize to any site.

Prognostic Assessment of BRAF Mutation in Preoperative Thyroid Fine-Needle Aspiration Specimens.

OBJECTIVES: We investigate the potential role of BRAF testing in guiding surgical intervention in papillary thyroid carcinoma (PTC). METHODS: Thyroid fine-needle aspiration (FNA) cases with available BRAF result and follow-up thyroidectomy for PTC were included in the study. Cytology and surgical diagnoses were correlated with BRAF status. RESULTS: There were 151 cases of thyroid FNA specimens with BRAF testing (70 mutant and 81 wild-type BRAF) and histologically confirmed unilateral, unifocal PTCs. There were no differences in age, sex, tumor size, or lymphovascular invasion on thyroidectomy specimens between mutant and wild-type BRAF cases. BRAF mutation was significantly associated with cytology diagnosis (P < .001), PTC subtype (P < .001), extrathyroidal extension (ETE) (P = .006), and higher tumor (T) stage (P = .04). However, an analysis within the histologic subtypes of PTC revealed no significant association between BRAF mutation and ETE or higher T stage. There was also no difference in central (P = .847) or lateral (p = 1) neck lymph node (LN) metastasis. CONCLUSIONS: BRAF mutation identified in thyroid FNA specimens correlates with histologic subtypes but is not an independent factor for predicting PTC biological behavior and should not be used to guide the extent of LN dissection.

Molecular alterations in Hürthle cell neoplasms of thyroid: A fine needle aspiration cytology study with cytology-histology correlation.

BACKGROUND: Hürthle cell features are frequently observed on the fine-needle aspiration (FNA) cytology of thyroid nodules and often pose a diagnostic challenge because of a significant overlap between cytomorphologic features seen in benign and malignant lesions. Molecular alterations (MAs) associated with these lesions are not well described. The objective of the current study was to evaluate the molecular profile of Hürthle cell lesions classified as Hürthle cell neoplasm (HCN) on cytologic evaluation. METHODS: The authors retrospectively reviewed their electronic database for cytologic diagnoses of HCN from January 1, 2017 to March 31, 2020. RESULTS: In total, 279 cases from 275 patients who had a diagnosis of HCN were included in the study. Molecular testing results were available in 85 cases (51 with MAs and 34 without MAs) and, of those, 42 had histologic follow-up available. Eight of 10 malignant cases had MAs, whereas the remaining 2 cases were negative for MAs. The most frequently encountered predominant genetic alterations or classifier findings were chromosome copy number alterations (n = 15), followed by NRAS (n = 8), KRAS (n = 7), suspicious (n = 6), EIF1AX (n = 4), TSHR (n = 3), gene overexpression (n = 3), positive microRNA classifier (n = 2), and 1 each of BRAF K601E, TERT, and HRAS mutations. One hundred thirty-seven cases had histologic follow-up available; of those, 28 were classified as malignant, and 109 were classified as benign (neoplastic and nonneoplastic). The overall risk of malignancy associated with HCN was 20%, and the risk of HCN with MAs was 25%. CONCLUSIONS: The cytologic diagnosis of HCN includes various MAs without any obvious trend, and most malignant cases (80%) have some type of MA.

Prediction of post-radiotherapy locoregional progression in HPV-associated oropharyngeal squamous cell carcinoma using machine-learning analysis of baseline PET/CT radiomics.

Locoregional failure remains a therapeutic challenge in oropharyngeal squamous cell carcinoma (OPSCC). We aimed to devise novel objective imaging biomarkers for prediction of locoregional progression in HPV-associated OPSCC. Following manual lesion delineation, 1037 PET and 1037 CT radiomic features were extracted from each primary tumor and metastatic cervical lymph node on baseline PET/CT scans. Applying random forest machine-learning algorithms, we generated radiomic models for censoring-aware locoregional progression prognostication (evaluated by Harrell's C-index) and risk stratification (evaluated in Kaplan-Meier analysis). A total of 190 patients were included; an optimized model yielded a median (interquartile range) C-index of 0.76 (0.66-0.81; p = 0.01) in prognostication of locoregional progression, using combined PET/CT radiomic features from primary tumors. Radiomics-based risk stratification reliably identified patients at risk for locoregional progression within 2-, 3-, 4-, and 5-year follow-up intervals, with log-rank p-values of p = 0.003, p = 0.001, p = 0.02, p = 0.006 in Kaplan-Meier analysis, respectively. Our results suggest PET/CT radiomic biomarkers can predict post-radiotherapy locoregional progression in HPV-associated OPSCC. Pending validation in large, independent cohorts, such objective biomarkers may improve patient selection for treatment de-intensification trials in this prognostically favorable OPSCC entity, and eventually facilitate personalized therapy.

Quantitative assessment of p16 expression in FNA specimens from head and neck squamous cell carcinoma and correlation with HPV status.

BACKGROUND: This study investigated p16 by immunohistochemistry (IHC) on cellblocks (CBs) and human papillomavirus (HPV) by polymerase chain reaction (PCR) in fine-needle aspiration (FNA) of head and neck squamous cell carcinoma (HNSCC). METHODS: Receiver operating characteristic (ROC) curve analysis was used to assess test performance in CBs compared with p16 IHC in 42 surgical specimens from patients with HNSCC and in correlation with HPV by PCR in cytology specimens. The study assessed HPV by PCR in FNA specimens as a substitute for p16 IHC in surgical specimens. RESULTS: Of 42 cases, 38 CBs showed malignant cells as cohesive clusters of viable cells with or without single tumor cells, whereas 4 specimens were composed exclusively of single tumor cells and degenerated cells. All p16-negative surgical specimens showed an absence of p16 staining in the corresponding CBs (n = 16). In the p16-positive surgical cases (n = 26), corresponding CBs with tumor clusters (n = 23) showed heterogeneous p16 expression ranging from 40% to 100%; however, scoring single cells was challenging and unreliable because of cellular degradation. ROC curve inspection showed the optimal threshold to be at least 40% p16 staining in tumor clusters with 100% sensitivity and specificity. In cases with inadequate CBs, HPV by PCR on needle rinse showed 88% sensitivity and 100% specificity for p16 expression in surgical specimens. CONCLUSIONS: A cutoff of at least 40% p16 expression in tumor clusters may be appropriate for p16 positivity in cytology CB specimens. A positive HPV finding by PCR on needle rinse can be used as a substitute for p16 expression in surgical specimens.

Thyroid tissue outside the thyroid gland: Differential diagnosis and associated diagnostic challenges.

The presence of thyroid tissue outside of the thyroid gland may occur in various clinical settings and anatomic locations and includes both benign and malignant differential diagnoses. Some of these entities include thyroglossal duct cyst, lingual thyroid, parasitic nodule, thyroid tissue within a lymph node and struma ovarii. In routine daily practice, these entities do pose diagnostic challenges for the pathologists. Differential diagnostic considerations depend largely on the location of lesion and the histologic features. A definitive diagnosis may remain unclear in some cases while knowledge is still evolving in others i.e., incidentally detected bland appearing thyroid follicles in a lateral neck lymph node. This article aims to elaborate on the various entities characterized by thyroid tissue outside of the thyroid gland, both benign and malignant, and the relevant differential diagnostic considerations.

Fine-needle aspiration cytomorphology of papillary thyroid carcinoma with NTRK gene rearrangement from a case series with predominantly indeterminate cytology.

BACKGROUND: Neurotrophic tyrosine kinase receptor (NTRK) rearrangement has been reported in a subset of papillary thyroid carcinoma (PTC) cases. Little is known about the cytomorphologic features of NTRK-rearranged PTC. METHODS: We report an institutional series of 13 fine-needle aspiration (FNA) specimens of NTRK-rearranged PTC with a predominantly indeterminate cytology diagnosis. NTRK3 or NTRK1 rearrangements were detected on FNA or surgical specimens by next-generation sequencing. RESULTS: The 13 patients had a median age of 18 years; 10 patients were female and 3 patients were male. In 10 (77%) cases, cytology was indeterminate, and histopathologic follow-up was predominantly the follicular variant of PTC (n = 8 [62%]), mostly infiltrative subtype. Of 12 FNA specimens available for review, a predominant loosely cohesive group pattern was the most commonly encountered architectural pattern (n = 5 [41%]), followed by single cell (n = 3 [25%]), thick cord (n = 2 [17%]), and microfollicular pattern (n = 2 [17%]). Background lymphocytic thyroiditis was observed in 9 cases. At the cellular level, the cytoplasm was moderate and granular, occasionally vacuolated. Classic PTC nuclear features (eg, nuclear enlargement, elongation, grooves, and nuclear membrane irregularity) were present but were often focal and subtle. Chromatin was often granular. Intranuclear pseudoinclusions were absent or rare. CONCLUSION: Our study demonstrates that most cases of NTRK rearrangement lack classic PTC cytomorphologic characteristics. Loosely cohesive groups and single cells with granular, sometimes vacuolated cytoplasm and subtle nuclear features are often seen on FNA specimens. Recognizing these characteristics may be helpful to preoperatively prompt molecular testing, including NTRK rearrangement analysis.

Genetic basis of SMARCB1 protein loss in 22 sinonasal carcinomas.

SMARCB1-deficient sinonasal carcinoma (SNC) is an aggressive malignancy characterized by INI1 loss mostly owing to homozygous SMARCB1 deletion. With the exception of a few reported cases, these tumors have not been thoroughly studied by massive parallel sequencing (MPS). A retrospective cohort of 22 SMARCB1-deficient SNCs were studied by light microscopy, immunohistochemistry, fluorescence in situ hybridization (n = 9), targeted exome MPS (n = 12), and Fraction and Allele-Specific Copy Number Estimates from Tumor Sequencing (FACETS) (n = 10), a bioinformatics pipeline for copy number/zygosity assessment. SMARCB1-deficient SNC was found in 13 (59%) men and 9 (41%) women. Most common growth patterns were the basaloid pattern (59%), occurring mostly in men (77%), and plasmacytoid/eosinophilic/rhabdoid pattern (23%), arising mostly in women (80%). The former group was significantly younger (median age = 46 years, range = 24-54, vs 79 years, range = 66-95, p < 0.0001). Clear cell, pseudoglandular, glandular, spindle cell, and sarcomatoid features were variably present. SMARCB1-deficient SNC expressed cytokeratin (100%), p63 (72%), neuroendocrine markers (52%), CDX-2 (44%), S-100 (25%), CEA (4/4 cases), Hepatocyte (2/2 cases), and aberrant nuclear ß-catenin (1/1 case). SMARCB1 showed homozygous deletion (68%), hemizygous deletion (16%), or truncating mutations associated with copy neutral loss of heterozygosity (11%). Coexisting genetic alterations were 22q loss including loss of NF2 and CHEK2 (50%), chromosome 7 gain (25%), and TP53 V157F, CDKN2A W110∗, and CTNNB1 S45F mutations. At 2 years and 5 years, the disease-specific survival and disease-free survival were 70% and 35% and 13% and 0%, respectively. SMARCB1-deficient SNCs are phenotypically and genetically diverse, and these distinctions warrant further investigation for their biological and clinical significance.

Potential Added Value of PET/CT Radiomics for Survival Prognostication beyond AJCC 8th Edition Staging in Oropharyngeal Squamous Cell Carcinoma.

Accurate risk-stratification can facilitate precision therapy in oropharyngeal squamous cell carcinoma (OPSCC). We explored the potential added value of baseline positron emission tomography (PET)/computed tomography (CT) radiomic features for prognostication and risk stratification of OPSCC beyond the American Joint Committee on Cancer (AJCC) 8th edition staging scheme. Using institutional and publicly available datasets, we included OPSCC patients with known human papillomavirus (HPV) status, without baseline distant metastasis and treated with curative intent. We extracted 1037 PET and 1037 CT radiomic features quantifying lesion shape, imaging intensity, and texture patterns from primary tumors and metastatic cervical lymph nodes. Utilizing random forest algorithms, we devised novel machine-learning models for OPSCC progression-free survival (PFS) and overall survival (OS) using "radiomics" features, "AJCC" variables, and the "combined" set as input. We designed both single- (PET or CT) and combined-modality (PET/CT) models. Harrell's C-index quantified survival model performance; risk stratification was evaluated in Kaplan-Meier analysis. A total of 311 patients were included. In HPV-associated OPSCC, the best "radiomics" model achieved an average C-index ± standard deviation of 0.62 ± 0.05 (p = 0.02) for PFS prediction, compared to 0.54 ± 0.06 (p = 0.32) utilizing "AJCC" variables. Radiomics-based risk-stratification of HPV-associated OPSCC was significant for PFS and OS. Similar trends were observed in HPV-negative OPSCC. In conclusion, radiomics imaging features extracted from pre-treatment PET/CT may provide complimentary information to the current AJCC staging scheme for survival prognostication and risk-stratification of HPV-associated OPSCC.

PET/CT radiomics signature of human papilloma virus association in oropharyngeal squamous cell carcinoma.

PURPOSE: To devise, validate, and externally test PET/CT radiomics signatures for human papillomavirus (HPV) association in primary tumors and metastatic cervical lymph nodes of oropharyngeal squamous cell carcinoma (OPSCC). METHODS: We analyzed 435 primary tumors (326 for training, 109 for validation) and 741 metastatic cervical lymph nodes (518 for training, 223 for validation) using FDG-PET and non-contrast CT from a multi-institutional and multi-national cohort. Utilizing 1037 radiomics features per imaging modality and per lesion, we trained, optimized, and independently validated machine-learning classifiers for prediction of HPV association in primary tumors, lymph nodes, and combined "virtual" volumes of interest (VOI). PET-based models were additionally validated in an external cohort. RESULTS: Single-modality PET and CT final models yielded similar classification performance without significant difference in independent validation; however, models combining PET and CT features outperformed single-modality PET- or CT-based models, with receiver operating characteristic area under the curve (AUC) of 0.78, and 0.77 for prediction of HPV association using primary tumor lesion features, in cross-validation and independent validation, respectively. In the external PET-only validation dataset, final models achieved an AUC of 0.83 for a virtual VOI combining primary tumor and lymph nodes, and an AUC of 0.73 for a virtual VOI combining all lymph nodes. CONCLUSION: We found that PET-based radiomics signatures yielded similar classification performance to CT-based models, with potential added value from combining PET- and CT-based radiomics for prediction of HPV status. While our results are promising, radiomics signatures may not yet substitute tissue sampling for clinical decision-making.

Histologic Classification and Molecular Signature of Polymorphous Adenocarcinoma (PAC) and Cribriform Adenocarcinoma of Salivary Gland (CASG): An International Interobserver Study.

Polymorphous adenocarcinoma (PAC) shows histologic diversity with streaming and targetoid features whereas cribriform adenocarcinoma of salivary gland (CASG) demonstrates predominantly cribriform and solid patterns with glomeruloid structures and optically clear nuclei. Opinions diverge on whether CASG represents a separate entity or a variant of PAC. We aimed to assess the level of agreement among 25 expert Head and Neck pathologists in classifying these tumors. Digital slides of 48 cases were reviewed and classified as: PAC, CASG, tumors with ≥50% of papillary architecture (PAP), and tumors with indeterminate features (IND). The consensus diagnoses were correlated with a previously reported molecular alteration. The consensus diagnoses were PAC in 18/48, CASG in16/48, PAP in 3/48, and IND in 11/48. There was a fair interobserver agreement in classifying the tumors (κ=0.370). The full consensus was achieved in 3 (6%) cases, all of which were classified as PAC. A moderate agreement was reached for PAC (κ=0.504) and PAP (κ=0.561), and a fair agreement was reached for CASG (κ=0.390). IND had only slight diagnostic concordance (κ=0.091). PAC predominantly harbored PRKD1 hotspot mutation, whereas CASG was associated with fusion involving PRKD1, PRKD2, or PRKD3. However, such molecular events were not exclusive as 7% of PAC had fusion and 13% of CASG had mutation. In conclusion, a fair to moderate interobserver agreement can be achieved in classifying PAC and CASG. However, a subset (23%) showed indeterminate features and was difficult to place along the morphologic spectrum of PAC/CASG among expert pathologists. This may explain the controversy in classifying these tumors.

Composite Paraganglioma: Pioneering in the Head and Neck.

Composite paragangliomas are rare with less than 20 cases documented in the abdomen, retroperitoneum, and urinary bladder. In this article, we report the first case of composite paraganglioma in the head and neck presenting as a soft tissue mass in the neck adjacent to the carotid artery in a 50-year-old woman. We discuss the clinicopathologic findings and genetic implications, and we review the literature of this rare entity.

Positive surgical margins in parotid malignancies: Institutional variation and survival association.

OBJECTIVES/HYPOTHESIS: To establish benchmarks for positive margin incidence in parotid cancers, associate positive margin status with patient, tumor, and institutional factors and overall survival, and characterize institutional variation in positive margin incidence. STUDY DESIGN: Retrospective database analysis. METHODS: We identified patients surgically treated for a parotid malignancy 2004 to 2013 in the National Cancer Database. We associated positive margins with patient, tumor, and treatment factors by multivariable logistic regression and with overall survival by Cox proportional hazards regression. We characterized institutional variation in positive margin rates by facility type and volume. RESULTS: A total of 5,639 patients were identified. The overall positive margin rate was 31%. By institution, positive margin rates varied from 0% to 100%, with a median rate of 31% (interquartile range = 18%-43%). Adenoid cystic carcinoma was associated with increased, and low-grade mucoepidermoid carcinoma with decreased, odds of positive margins, (odds ratio [OR] = 1.91 [95% confidence interval {CI}: 1.54-2.38], P < .001 and OR = 0.68 [95% CI: 0.53-0.87], P = .002, respectively). Treatment at academic or high-volume facilities was associated with decreased odds of positive margins (OR = 0.79 [95% CI: 0.68-0.91], P = .001 and OR = 0.76 [95% CI: 0.63-0.91], P = .004). Positive surgical margins were associated with decreased overall survival (62% vs. 79% 5-year overall survival, hazard ratio [HR] = 1.34 [95% CI: 1.20-1.49], P < .001). Upon stratification by histology, this association was maintained for high-grade, but not low-grade, mucoepidermoid carcinoma (52% vs. 74%, HR = 1.56 [95% CI: 1.31-1.86], P < .001 and 89% versus 91%, HR = 1.05 [95% CI: 0.60-1.83], P = .874, respectively). CONCLUSIONS: Tumor and institutional factors are associated with positive surgical margins. Positive margin rates vary substantially between facilities and are less likely at academic and high-volume facilities. Positive surgical margins are associated with decreased overall survival and could be considered quality indicators in parotid malignancies. LEVEL OF EVIDENCE: 2c Laryngoscope, 129:129-137, 2019.