Genetically Determined Circulating Lactase/Phlorizin Hydrolase Concentrations and Risk of Colorectal Cancer: A Two-Sample Mendelian Randomization Study.

Previous research has found that milk is associated with a decreased risk of colorectal cancer (CRC). However, it is unclear whether the milk digestion by the enzyme lactase-phlorizin hydrolase (LPH) plays a role in CRC susceptibility. Our study aims to investigate the direct causal relationship of CRC risk with LPH levels by applying a two-sample Mendelian Randomization (MR) strategy. Genetic instruments for LPH were derived from the Fenland Study, and CRC-associated summary statistics for these instruments were extracted from the FinnGen Study, PLCO Atlas Project, and Pan-UK Biobank. Primary MR analyses focused on a cis-variant (rs4988235) for LPH levels, with results integrated via meta-analysis. MR analyses using all variants were also undertaken. This analytical approach was further extended to assess CRC subtypes (colon and rectal). Meta-analysis across the three datasets illustrated an inverse association between genetically predicted LPH levels and CRC risk (OR: 0.92 [95% CI, 0.89-0.95]). Subtype analyses revealed associations of elevated LPH levels with reduced risks for both colon (OR: 0.92 [95% CI, 0.89-0.96]) and rectal cancer (OR: 0.92 [95% CI, 0.87, 0.98]). Consistency was observed across varied analytical methods and datasets. Further exploration is warranted to unveil the underlying mechanisms and validate LPH's potential role in CRC prevention.

American Society of Cytopathology Telecytology validation recommendations for rapid on-site evaluation (ROSE).

Telecytology has multiple applications, including rapid onsite evaluation (ROSE) of fine-needle aspiration (FNA) specimens. It can enhance cytopathology practice by increasing productivity, reducing costs, and providing subspecialty expertise in areas with limited access to a cytopathologist. However, there are currently no specific validation guidelines to ensure safe practice and compliance with regulations. This initiative, promoted by the American Society of Cytopathology (ASC), intends to propose recommendations for telecytology implementation. These recommendations propose that the validation process should include testing of all hardware and software, both separately and as a whole; training of all individuals who will participate in telecytology with regular competency evaluations; a structured approach using retrospective slides with defined diagnoses for validation and prospective cases for verification and quality assurance. Telecytology processes must be integrated into the laboratory's quality management system and benchmarks for discrepancy rates between preliminary and final diagnoses should be established and monitored. Special attention should be paid to minimize discrepancies that downgrade malignant cases to benign (false positive on telecytology). Currently, billing and reimbursement codes for telecytology are not yet available. Once, they are, recommendation of the appropriate usage of these codes would be a part of the recommendations. These proposed guidelines are intended to be a resource for laboratories that are considering implementing telecytology. These recommendations can help to ensure the safe and effective use of telecytology and maximize its benefits for patients.

Association between ambient exposure to PM2.5 and upper aerodigestive tract cancer in Los Angeles.

Fine particulate matter (PM2.5 ) contains carcinogens similar to those generated by tobacco smoking, which may increase the risks of developing smoking-related cancers, such as upper aerodigestive track (UADT) cancers, for both smokers and never-smokers. Therefore, it is imperative to understand the relation between ambient PM2.5 exposure and risk of UADT cancers. A population-based case-control study involving 565 incident UADT cancer cases and 983 controls was conducted in Los Angeles County from 1999 to 2004. The average residential PM2.5 concentration 1 year before the diagnosis date for cases and the reference date for controls was assessed using a chemical transport model. The association between ambient PM2.5 and the UADT cancers was estimated by unconditional logistic regression, adjusting for confounders at the individual and block-group level. Stratified analyses were conducted by sex, tobacco smoking status and UADT subsites. We also assessed the interaction between PM2.5 and tobacco smoking on UADT cancers. PM2.5 concentrations were associated with an elevated odds of UADT cancers (adjusted odds ratio = 1.21 per interquartile range [4.5 µg/m3 ] increase; 95% confidence interval: 1.02, 1.44). The association between PM2.5 and UADT cancers was similar across UADT subsites, sex and tobacco smoking status. The interaction between PM2.5 and tobacco smoking on UADT cancers was approximately additive on the odds scale. The effect estimate for PM2.5 and UADT cancers was similar among never smokers. Our findings support the hypothesis that exposure to PM2.5 increases the risk of UADT cancers. Improvements in air quality may reduce the risk of UADT cancers.

Diagnostic performance of molecular testing in indeterminate (Bethesda III and IV) thyroid nodules with Hürthle cell cytology.

BACKGROUND: Indeterminate thyroid nodules with Hürthle cell cytology remain a diagnostic challenge. The low benign call rate and positive predictive value of first-generation molecular tests precluded their use to rule out malignancy. We examined the diagnostic performance of current tests. METHOD: This subset analysis of our prospective randomized trial compared the benign call rate and positive predictive value of Afirma Gene Sequencing Classifier and Thyroseq v3 in Bethesda III and IV nodules with Hürthle cell cytology. Molecular test samples were obtained at initial fine-needle aspiration (8/2017-7/2022) and reflexively sent for processing. RESULTS: Molecular testing was performed on 140 Hürthle cell nodules. Of 79 nodules tested with the Afirma Gene Sequencing Classifier, the benign call rate was 84% (66/79). Nine of 66 nodules with benign results were resected, with no malignancies. Twelve of 13 nodules with suspicious results were resected, revealing 3 malignancies-2 papillary thyroid carcinomas and one Hürthle cell carcinoma (positive predictive value 25%). Of 61 nodules tested with Thyroseq v3, the benign call rate was 56% (34/61; (P < .01 versus Afirma Gene Sequencing Classifier). Five of 34 nodules with negative results were resected, with no malignancies. Nineteen of 27 nodules with positive results were resected, revealing 3 malignancies-2 papillary thyroid carcinomas and 1 Hürthle cell carcinoma (positive predictive value 16%). CONCLUSION: The high benign call rate of current molecular tests in Hürthle cell nodules strengthens their value in enabling patients to avoid surgery.

Pathologic response to neoadjuvant chemotherapy in ovarian cancer and its association with outcome: A surrogate marker of survival.

OBJECTIVE: We aimed to validate whether pathologic response (pR) to neoadjuvant chemotherapy (NACT) using a three-tier chemotherapy response score (CRS) is associated with clinical outcome in ovarian cancer (OC) and could be used as surrogate marker for survival. METHODS: We conducted a retrospective study of OC patients with FIGO stage III/IV disease who received NACT and graded response as no or minimal (CRS 1), partial (CRS 2), or complete/near-complete (CRS 3) pR using tissue specimens obtained from omentum. Uni- and multivariate survival analyses were performed accounting for age, FIGO stage, debulking and BRCA status as well as neoadjuvant use of bevacizumab. RESULTS: CRSs 1, 2 and 3 were found in 41(31%), 62 (47%) and 30 (22%) of the 133 examined cases. Response to NACT was associated with significantly improved progression-free (PFS, p < 0.001) and overall survival (OS, p = 0.011). Complete/ near-complete pathologic response (CRS3) was associated with improved PFS (median 24.8 vs. 12.5 months, unadjusted HR 0.28 [95%CI 0.15-0.54], p < 0.001; adjusted hazard ration (aHR) 0.31 [95% CI 0.14-0.72], p = 0.007) and OS (median 63.3 vs. 32.1 months, unadjusted HR 0.27 [95%CI 0.10-0.68], p = 0.006; aHR 0.32 [95% CI 0.09-1.11], p = 0.072) when compared to no or minimal response (CRS1). CONCLUSIONS: We validate a three-tier CRS for assessment of pathologic response to NACT in OC and demonstrate its prognostic independence of BRCA status or neoadjuvant bevacizumab use. Improving pR rates may be a useful goal of NACT in OC with the expectation of improved survival. The CRS may be a useful endpoint in clinical trials in OC.

Opium, phencyclidine, and crack cocaine smoking associations with lung and upper aerodigestive tract cancers: exploratory findings from a case-control study in Los Angeles County.

Background: Illicit drug use has become a global epidemic, yet it is unclear if drug smoking increases the risk of tobacco-related cancers.Objectives: We aimed to evaluate hypothesized associations between smoking three drugs - opium, phencyclidine (PCP) and crack cocaine and lung and upper aerodigestive tract (UADT) cancers.Methods: A population-based case-control study with 611 lung cancer cases (50% male), 601 UADT cancers cases (76% male), and 1,040 controls (60% male) was conducted in Los Angeles County (1999-2004). Epidemiologic data including drug smoking histories were collected in face-to-face interviews. Associations were estimated with logistic regressions.Results: Adjusting for potential confounders, ever vs. never crack smoking was positively associated with UADT cancers (aOR = 1.56, 95% CI: 1.05, 2.33), and a dose-response relationship was observed for lifetime smoking frequency (p for trend = .024). Heavy (> median) vs. never crack smoking was associated with UADT cancers (aOR = 1.81, 95% CI: 1.07, 3.08) and lung cancer (aOR = 1.58, 95% CI: 0.88, 2.83). A positive association was also observed between heavy PCP smoking and UADT cancers (aOR = 2.29, 95% CI: 0.91, 5.79). Little or no associations were found between opium smoking and lung cancer or UADT cancers.Conclusion: The positive associations between illicit drug use and lung and/or UADT cancers suggest that smoking these drugs may increase the risk of tobacco-related cancers. Despite the low frequency of drug smoking and possible residual confounding, our findings may provide additional insights on the development of lung and UADT cancers.

Artificial Intelligence-Driven Morphology-Based Enrichment of Malignant Cells from Body Fluid.

Cell morphology is a fundamental feature used to evaluate patient specimens in pathologic analysis. However, traditional cytopathology analysis of patient effusion samples is limited by low tumor cell abundance coupled with the high background of nonmalignant cells, restricting the ability of downstream molecular and functional analyses to identify actionable therapeutic targets. We applied the Deepcell platform that combines microfluidic sorting, brightfield imaging, and real-time deep learning interpretations based on multidimensional morphology to enrich carcinoma cells from malignant effusions without cell staining or labels. Carcinoma cell enrichment was validated with whole genome sequencing and targeted mutation analysis, which showed a higher sensitivity for detection of tumor fractions and critical somatic variant mutations that were initially at low levels or undetectable in presort patient samples. Our study demonstrates the feasibility and added value of supplementing traditional morphology-based cytology with deep learning, multidimensional morphology analysis, and microfluidic sorting.

"SMART" cytology: The next generation cytology for precision diagnosis.

Cytology plays an important role in diagnosing and managing human diseases, especially cancer, as it is often a simple, low cost yet effective, and non-invasive or minimally invasive diagnostic tool. However, traditional morphology-based cytology practice has limitations, especially in the era of precision diagnosis. Recently there have been tremendous efforts devoted to apply computational tools and to perform molecular analysis on cytological samples for a variety of clinical purposes. Now is probably the appropriate juncture to integrate morphology, machine learning, and molecular analysis together and transform cytology from a morphology-driven practice to the next level - "SMART" Cytology. In this article we will provide a rather brief review of the relevant works for computational analysis on cytology samples, focusing on single-cell-based multiplex quantitative analysis of biomarkers, and introduce the conceptual framework of "SMART (Single cell, Multiplex, AI-driven, and Real Time)" Cytology.

Retrospective analysis of adjuvant treatment for localized, operable uterine leiomyosarcoma.

OBJECTIVE: Currently, there is no standard adjuvant treatment protocol for localized uterine leiomyosarcoma (uLMS) as clinical trials to address this question have been retrospective, underpowered, or undermined by slow accrual rates. The aim of this study is to determine the benefit of adjuvant chemotherapy for uLMS. METHODS: We reviewed the medical records of localized uLMS patients who had underwent adjuvant therapy after upfront surgery between 2000 and 2020. The cases were blinded for review. We evaluated the influence of various clinical characteristics and different types of adjuvant therapies on specific outcomes. RESULTS: Sixty-eight patients (median age: 50 years) were included for analysis. Forty of 68 (58.8%) patients received adjuvant chemotherapy +/- radiation therapy and 25 patients (38.6%) did not receive any adjuvant therapy. At a median follow-up time of 43.3 months, 45 patients (66.1%) had relapsed disease. The median disease-free survival (mDFS) for all patients was 23.1 months. Patients who received any adjuvant treatment (chemotherapy and/or radiation) trended toward a longer mDFS compared with those who did not receive any adjuvant therapy (29.7 vs. 14.1 months, p = 0.26). Patients who received adjuvant chemotherapy alone had a longer, but nonstatistically significant mDFS compared with those who did not receive any adjuvant treatment (22.2 vs. 14.1 months, p = 0.18). Additionally, univariate analysis found that tumor size large than 10 cm, and a mitotic rate >10/10hpf were independent prognostic factors for worse DFS. CONCLUSIONS: Though DFS was more favorable among those who received adjuvant therapy, it was not statistically significant, and thus based on this data adjuvant therapy for resected uLMS is still in question.

Familial Duodenal Somatostatinomatosis Not Associated With a Known Genetic Syndrome.

ABSTRACT: We report a father and his daughter who both had multiple somatostatinomas in the duodenal bulb without a known syndrome. The father, at age 68 years, was incidentally found to harbor 4 approximately 1.5-cm well-differentiated neuroendocrine tumors in the duodenal bulb. His preoperative somatostatin level was elevated. He underwent partial duodenectomy and regional lymph node dissection; one lymph node was positive for metastasis. One year postoperatively, a recurrence was found in the surgical bed; he was treated with octreotide for 2 years, which stabilized the recurrent tumor. Ten years postoperatively, the mucosa of his remaining duodenum was normal. His daughter, at age 53 years, was found to harbor multiple small neuroendocrine tumors in the duodenal bulb. Immunostaining of available specimens showed that the neuroendocrine tumors from the father and daughter both were strongly positive for somatostatin. Micronodules of somatostatin-expressing neuroendocrine cells were found in the parts of the specimens uninvolved with the tumors. Both patients exhibited no evidence of known syndromes associated with somatostatinoma. The daughter did not harbor mutations in 93 genes commonly found in genetic tumor syndromes. The 2 cases thus suggest a novel, autosomal dominant, genetic syndrome of familial duodenal somatostatinomatosis.

Cytokeratin 13 Is a New Biomarker for the Diagnosis of Limbal Stem Cell Deficiency.

PURPOSE: The purpose of this study was to evaluate the expression of cytokeratin (K) 13 on the corneal surface and to validate its application in the diagnosis of limbal stem cell deficiency (LSCD). METHODS: This prospective comparative study included 26 corneal impression cytology (IC) specimens from patients diagnosed with LSCD. Twenty-three IC specimens from normal donors served as controls. K12 and K13 expression were detected on the IC specimens by immunohistochemistry study. The number of K12 + or K13 + cells in all areas of the IC was quantified using ImageJ software. RESULTS: The epithelial cells harvested from IC specimens from control corneas were all K12 + . In eyes with LSCD, K13 + and K12 + /K13 + cells accounted for 93.8% and 2.6%, respectively, in the cornea. In eyes with sectoral LSCD, the median number of K13 + cells in the clinically affected area was higher than that in the unaffected area (810.0 vs. 115.0 cells/mm 2 ; P < 0.001). No significant correlation was found between the LSCD severity and the number of K12 + cells (r = -0.284, P = 0.16) or K13 + cells (r = -0.011, P = 0.95). The presence of at least 16 K13 + cells/mm 2 was suggestive of LSCD. CONCLUSIONS: Identification of K13 + cells on IC specimens provides a simple and reliable method to detect conjunctival epithelial cells on the cornea. K13 is a marker for diagnosing LSCD and localizing the involved area in sectoral LSCD.

The Relieving Effects of a Polyherb-Based Dietary Supplement ColonVita on Gastrointestinal Quality of Life Index (GIQLI) in Older Adults with Chronic Gastrointestinal Symptoms Are Influenced by Age and Cardiovascular Disease: A 12-Week Randomized Placebo-Controlled Trial.

Chronic gastrointestinal symptoms (CGS) negatively affect the quality of life in about 15-30% of the population without effective drugs. Recent studies suggest that dietary supplement may improve CGS, but inconsistent results exist. The goal of this study is to evaluate the effect of a polyherbal-based supplement ColonVita on the gastrointestinal quality of life index (GIQLI) in 100 old adults with CGS (63.1 ± 9.6 years) who were randomly assigned to daily ColonVita or placebo tablets (n = 50/group) for 12 weeks in a double-blind, randomized controlled trial design. No significant fibrdifferences were found between ColonVita and placebo in the baseline total GIQLI score (101.12 ± 16.87 vs. 101.80 ± 16.48) (P > 0.05) or postintervention total GIQLI score (114.78 ± 9.62 vs. 111.74 ± 13.01) (P > 0.05). However, ColonVita significantly improved 16 scores of the 19 core GI symptoms compared with 10 items improved by placebo. The ColonVita group significantly improved the remission rate of 5 core GI symptoms compared to placebo and significantly improved the total GIQLI scores (118.09 ± 7.88 vs. 109.50 ± 16.71) (P < 0.05) and core GI symptom scores (64.61 ± 3.99 vs. 60.00 ± 8.65) (P < 0.05) in people ≥60 years of age (n = 49) but not in those under 60 y (n = 51). ColonVita significantly improved the total GIQLI scores and core GI symptom scores in people without cardiovascular diseases (CVD) (n = 56) (116.74 ± 9.38 vs. 110.10 ± 14.28) (P < 0.05) and (63.11 ± 4.53 vs. 59.93 ± 8.03) (P=0.07), respectively, but not in those with CVD (n = 44). Thus, ColonVita was beneficial for old adults with CGS, especially those ≥60 years of age and without CVD. Because a heterogenous pathogenesis of CGS-like irritable bowel syndrome (IBS) and inflammatory bowel disease (ISD) is differentially associated with CVD, different comorbidities may have influenced the outcomes of different trials that should be controlled in further studies.

Interactions Between Adiponectin-Pathway Polymorphisms and Obesity on Postmenopausal Breast Cancer Risk Among African American Women: The WHI SHARe Study.

BACKGROUND: A decreased level of serum adiponectin is associated with obesity and an increased risk of breast cancer among postmenopausal women. Yet, the interplay between genetic variants associated with adiponectin phenotype, obesity, and breast cancer risk is unclear in African American (AA) women. METHODS: We examined 32 single-nucleotide polymorphisms (SNPs) previously identified in genome-wide association and replication studies of serum adiponectin levels using data from 7,991 AA postmenopausal women in the Women's Health Initiative SNP Health Association Resource. RESULTS: Stratifying by obesity status, we identified 18 adiponectin-related SNPs that were associated with breast cancer risk. Among women with BMI ≥ 30 kg/m2, the minor TT genotype of FER rs10447248 had an elevated breast cancer risk. Interaction was observed between obesity and the CT genotype of ADIPOQ rs6773957 on the additive scale for breast cancer risk (relative excess risk due to interaction, 0.62; 95% CI, 0.32-0.92). The joint effect of BMI ≥ 30 kg/m2 and the TC genotype of OR8S1 rs11168618 was larger than the sum of the independent effects on breast cancer risk. CONCLUSIONS: We demonstrated that obesity plays a significant role as an effect modifier in an increased effect of the SNPs on breast cancer risk using one of the most extensive data on postmenopausal AA women. IMPACT: The results suggest the potential use of adiponectin genetic variants as obesity-associated biomarkers for informing AA women who are at greater risk for breast cancer and also for promoting behavioral interventions, such as weight control, to those with risk genotypes.

Advances in Imaging Modalities, Artificial Intelligence, and Single Cell Biomarker Analysis, and Their Applications in Cytopathology.

Several advances in recent decades in digital imaging, artificial intelligence, and multiplex modalities have improved our ability to automatically analyze and interpret imaging data. Imaging technologies such as optical coherence tomography, optical projection tomography, and quantitative phase microscopy allow analysis of tissues and cells in 3-dimensions and with subcellular granularity. Improvements in computer vision and machine learning have made algorithms more successful in automatically identifying important features to diagnose disease. Many new automated multiplex modalities such as antibody barcoding with cleavable DNA (ABCD), single cell analysis for tumor phenotyping (SCANT), fast analytical screening technique fine needle aspiration (FAST-FNA), and portable fluorescence-based image cytometry analyzer (CytoPAN) are under investigation. These have shown great promise in their ability to automatically analyze several biomarkers concurrently with high sensitivity, even in paucicellular samples, lending themselves well as tools in FNA. Not yet widely adopted for clinical use, many have successfully been applied to human samples. Once clinically validated, some of these technologies are poised to change the routine practice of cytopathology.

Long-Term Exposure to PM2.5, Facemask Mandates, Stay Home Orders and COVID-19 Incidence in the United States.

Long-term PM2.5 exposure might predispose populations to SARS-CoV-2 infection and intervention policies might interrupt SARS-CoV-2 transmission and reduce the risk of COVID-19. We conducted an ecologic study across the United States, using county-level COVID-19 incidence up to 12 September 2020, to represent the first two surges in the U.S., annual average of PM2.5 between 2000 and 2016 and state-level facemask mandates and stay home orders. We fit negative binomial models to assess COVID-19 incidence in association with PM2.5 and policies. Stratified analyses by facemask policy and stay home policy were also performed. Each 1-µg/m3 increase in annual average concentration of PM2.5 exposure was associated with 7.56% (95% CI: 3.76%, 11.49%) increase in COVID-19 risk. Facemask mandates and stay home policies were inversely associated with COVID-19 with adjusted RRs of 0.8466 (95% CI: 0.7598, 0.9432) and 0.9193 (95% CI: 0.8021, 1.0537), respectively. The associations between PM2.5 and COVID-19 were consistent among counties with or without preventive policies. Our study added evidence that long-term PM2.5 exposure increased the risk of COVID-19 during each surge and cumulatively as of 12 September 2020, in the United States. Although both state-level implementation of facemask mandates and stay home orders were effective in preventing the spread of COVID-19, no clear effect modification was observed regarding long-term exposure to PM2.5 on the risk of COVID-19.

A Preliminary Study of Deep-Learning Algorithm for Analyzing Multiplex Immunofluorescence Biomarkers in Body Fluid Cytology Specimens.

INTRODUCTION: Multiplex biomarker analysis of cytological body fluid specimens is often used to assist cytologists in distiguishing metastatic cancer cells from reactive mesothelial cells. However, evaluating biomarker expression visually may be challenging, especially when the cells of interest are scant. Deep-learning algorithms (DLAs) may be able to assist cytologists in analyzing multiple biomarker expression at the single cell level in the multiplex fluorescence imaging (MFI) setting. This preliminary study was performed to test the feasibility of using DLAs to identify immunofluorescence-stained metastatic adenocarcinoma cells in body fluid cytology samples. METHODS: A DLA was developed to analyze MFI-stained cells in body fluid cytological samples. A total of 41 pleural fluid samples, comprising of 20 positives and 21 negatives, were retrospectively collected. Multiplex immunofluorescence labeling for MOC31, BerEP4, and calretinin, were performed on cell block sections, and results were analyzed by manual analysis (manual MFI) and DLA analysis (MFI-DLA) independently. RESULTS: All cases with positive original cytological diagnoses showed positive results either by manual MFI or MFI-DLA, but 2 of the 14 (14.3%) original cytologically negative cases had rare cells with positive MOC31 and/or BerEP4 staining in addition to calretinin. Manual MFI analysis and MFI-DLA showed 100% concordance. CONCLUSION: MFI combined with DLA provides a potential tool to assist in cytological diagnosis of metastatic malignancy in body fluid samples. Larger studies are warranted to test the clinical validity of the approach.

Atypical de Quervain's thyroiditis diagnosed as atypia of undetermined significance by cytology and suspicious for cancer by Afirma Genomic Sequencing Classifier.

We report a case of atypical de Quervain's thyroiditis diagnosed as atypia of undetermined significance by cytology and suspicious for cancer by Afirma Genomic Sequencing Classifier. A 71-year-old male underwent thyroid ultrasound for goiter and was found to have two American Thyroid Association (ATA) 2015 high-suspicion nodules. The larger, 2.2-cm nodule was biopsied and the cytology showed atypical follicular cells and histiocytes. The Afirma Genomic Sequencing Classifier (detecting mRNA expression profile) result was ''suspicious'' (risk of malignancy ~50%) but Afirma Xpression Atlas (detecting specific mutations) did not find mutations in BRAF V600E, RET/PTC1, or RET/PTC3. The patient saw two endocrine surgeons and two endocrinologists who each recommended hemithyroidectomy. The patient chose to monitor the nodules. A new diagnostic ultrasound performed 3 months after the first one showed that the thyroid was significantly smaller and the previously seen nodules were no longer found. Re-examination of the cellular smears confirmed that the cytological findings were also compatible with de Quervain's thyroiditis. This case illustrates that atypical de Quervain's thyroiditis should be in the differential diagnosis of thyroid nodules for cytologists, radiologists, and clinicians. Furthermore, this case demonstrates that atypical de Quervain's thyroiditis can generate false positive results of molecular tests for indeterminate thyroid nodules.

Developing a Machine Learning Algorithm for Identifying Abnormal Urothelial Cells: A Feasibility Study.

INTRODUCTION: Urine cytology plays an important role in diagnosing urothelial carcinoma (UC). However, urine cytology interpretation is subjective and difficult. Morphogo (ALAB, Boston, MA, USA), equipped with automatic acquisition and scanning, optical focusing, and automatic classification with convolutional neural network has been developed for bone marrow aspirate smear analysis of hematopoietic diseases. The goal of this preliminary study was to determine the feasibility of developing a machine learning algorithm on Morphogo for identifying abnormal urothelial cells in urine cytology slides. METHODS: Thirty-seven achieved abnormal urine cytology slides from cases with the diagnosis of atypical urothelial cells and above (suspicions or positive for UC) were obtained from 1 hospital. A pathologist (J.R.) reviewed the slides and manually selected and annotated representative cells to feed into Morphogo with following categories: benign (urothelial cells, squamous cells, degenerated cells, and inflammatory cells), atypical cells, and suspicious cells. Initial validation of the algorithm was performed on a subset of the original 37 cases. Urine samples from additional 12 unknown cases with various histological diagnoses (6 cases of high-grade urothelial carcinoma (HGUC), 1 case of low-grade urothelial carcinoma (LGUC), 1 case of prostate adenocarcinoma, 1 case of renal cell carcinoma, and 4 cases of non-neoplastic conditions) were collected from another hospital for initial blind testing. RESULTS: A total of 1,910 benign and 1,978 abnormal (atypical and suspicious) cells from 37 slides were annotated for developing and training of the algorithm. This algorithm was validated on 27 slides that resulted in identification of at least 1 abnormal cell per slide, with a total of 200 abnormal cells, and an average of 7.4 cells per slide. Of the 12 unknown cases tested, the original cytology was positive for tumor cells in 2 HGUC samples. Morphogo was abnormal (atypical or suspicious) for 6 samples from patients with UC, including one with LGUC and one with prostate adenocarcinoma. CONCLUSION: Morphogo machine learning algorithm is capable of identifying abnormal urothelial cells. Further validation studies with a larger number of urine samples will be needed to determine if it can be used to assist the cytological diagnosis of UC.