Clinical and pathological characteristics of bladder cancer in post brachytherapy patients

The long-term risk of secondary malignancy is a potential late effect of brachytherapy. However, the time interval, anatomic site and histopathology are not well studied. We sought to characterize the bladder cancers that developed following treatment of prostate cancer with brachytherapy. Between 1998 and 2014, 4570 patients were treated with brachytherapy at the BC Cancer Agency. Out of those, 69 patients subsequently developed bladder cancer, some of which could have been radiation induced. Histology slides were reviewed for all cases, and site and pathologic features were recorded. Cases were classified as luminal and basal subtypes based on GATA3 and CK5/6 immunohistochemistry. Bladder neck and trigone were among the common sites of involvement. Pathologic review of cases showed that 68 % were high-grade, 25 % were muscle-invasive, and 20 % showed variant histology, including small cell carcinoma, sarcomatoid carcinoma, squamous cell carcinoma, and adenocarcinoma. A subgroup of cases more likely to be radiation-induced, based on site and time interval, was associated with increased pathologic stage (pT1 or higher) compared to the other cases (70 % vs 34 %, p = 0.01). In conclusion, the majority of bladder cancers following brachytherapy in this cohort were of high grade and low stage at diagnosis, most of them demonstrating luminal immunophenotype. A significant number of variant histologies are seen, each demonstrating a specific immunophenotype.

Comparison of Artificial Intelligence Techniques to Evaluate Performance of a Classifier for Automatic Grading of Prostate Cancer From Digitized Histopathologic Images

IMPORTANCE

Proper evaluation of the performance of artificial intelligence techniques in the analysis of digitized medical images is paramount for the adoption of such techniques by the medical community and regulatory agencies.

OBJECTIVES

To compare several cross-validation (CV) approaches to evaluate the performance of a classifier for automatic grading of prostate cancer in digitized histopathologic images and compare the performance of the classifier when trained using data from 1 expert and multiple experts.

DESIGN, SETTING, AND PARTICIPANTS

This quality improvement study used tissue microarray data (333 cores) from 231 patients who underwent radical prostatectomy at the Vancouver General Hospital between June 27, 1997, and June 7, 2011. Digitized images of tissue cores were annotated by 6 pathologists for 4 classes (benign and Gleason grades 3, 4, and 5) between December 12, 2016, and October 5, 2017. Patches of 192 µm2 were extracted from these images. There was no overlap between patches. A deep learning classifier based on convolutional neural networks was trained to predict a class label from among the 4 classes (benign and Gleason grades 3, 4, and 5) for each image patch. The classification performance was evaluated in leave-patches-out CV, leave-cores-out CV, and leave-patients-out 20-fold CV. The analysis was performed between November 15, 2018, and January 1, 2019.

MAIN OUTCOMES AND MEASURES

The classifier performance was evaluated by its accuracy, sensitivity, and specificity in detection of cancer (benign vs cancer) and in low-grade vs high-grade differentiation (Gleason grade 3 vs grades 4-5). The statistical significance analysis was performed using the McNemar test. The agreement level between pathologists and the classifier was quantified using a quadratic-weighted κ statistic.

RESULTS

On 333 tissue microarray cores from 231 participants with prostate cancer (mean [SD] age, 63.2 [6.3] years), 20-fold leave-patches-out CV resulted in mean (SD) accuracy of 97.8% (1.2%), sensitivity of 98.5% (1.0%), and specificity of 97.5% (1.2%) for classifying benign patches vs cancerous patches. By contrast, 20-fold leave-patients-out CV resulted in mean (SD) accuracy of 85.8% (4.3%), sensitivity of 86.3% (4.1%), and specificity of 85.5% (7.2%). Similarly, 20-fold leave-cores-out CV resulted in mean (SD) accuracy of 86.7% (3.7%), sensitivity of 87.2% (4.0%), and specificity of 87.7% (5.5%). Results of McNemar tests showed that the leave-patches-out CV accuracy, sensitivity, and specificity were significantly higher than those for both leave-patients-out CV and leave-cores-out CV. Similar results were observed for classifying low-grade cancer vs high-grade cancer. When trained on a single expert, the overall agreement in grading between pathologists and the classifier ranged from 0.38 to 0.58; when trained using the majority vote among all experts, it was 0.60.

CONCLUSIONS AND RELEVANCE

Results of this study suggest that in prostate cancer classification from histopathologic images, patch-wise CV and single-expert training and evaluation may lead to a biased estimation of classifier's performance. To allow reproducibility and facilitate comparison between automatic classification methods, studies in the field should evaluate their performance using patient-based CV and multiexpert data. Some of these conclusions may be generalizable to other histopathologic applications and to other applications of machine learning in medicine.

Prostatic ductal adenocarcinoma with cribriform architecture has worse prognostic features than non-cribriform-type

Prostatic ductal adenocarcinoma (PDA) is a rare histologic subtype of prostate cancer characterized by large glands lined with tall columnar pseudostratified epithelium. PDA has several architectural patterns, with papillary and cribriform being the most common. The cribriform pattern of acinar carcinoma has shown to be associated with a worse prognosis in terms of disease progression and disease-specific mortality. However, the significance of cribriform pattern in PDA is unknown. In this study, we sought to compare the adverse pathologic features between cribriform-type and non-cribriform-type PDA, and between PDA and acinar carcinoma with Gleason scores 8-10. We identified PDA cases diagnosed between 2008 and 2018 and 428 radical prostatectomy (RP) specimens containing Gleason 8-10 acinar carcinoma. The slides of all PDA cases were reviewed, and pathologic features were recorded. We found that the vast majority of PDA contained admixed acinar carcinoma, with a median percentage of the ductal component of 50% (range 5-100). 29% of PDA was graded as Grade Group 4 and 35.5% as Grade Group 5. At the time of RP, 45.2% of cases presented as pathologic stage T3a and 29% as T3b. Cribriform-type PDA demonstrated a significantly higher likelihood of extraprostatic extension (84% vs 33.3%, p = 0.01), seminal vesical invasion (36% vs 0%, p = 0.04), lymphovascular invasion (40% vs 0%, p = 0.03) and advanced pathologic stage (84% vs 33.3%, p = 0.01) compared to PDA without cribriform architecture. The proportion of stage ≥pT3 tumors in PDA was similar compared to that in Gleason 8-10 acinar carcinoma (74.2% vs 70.8%, p = 0.68).

Fine-needle aspiration of soft tissue myoepithelioma

Soft tissue myoepithelioma is a rare neoplasm composed of myoepithelial cells. We describe the cytologic features of a soft tissue myoepithelioma arising in the right lower chest wall in a 65-year-old woman. The fine-needle aspiration (FNA) smears showed round to oval, spindle, epithelioid, and plasmacytoid cells in the myxoid background. The nuclei were uniform, round to ovoid, with finely distributed chromatin and eosinophilic or pale cytoplasm, and resembled lobular carcinoma of breast. Ultrasound guided core biopsy showed the tumor cells had bland cytologic features, arranged in small cords, nests, and dissociated single cells, with no glandular differentiation or breast tissue seen. The tumor cells demonstrated immunoreactivity for cytokeratin (AE1/AE3) and glial fibrillary acidic protein, but were negative for estrogen receptor. Fluorescence in situ hybridization demonstrated the EWSR1 rearrangement, confirming the diagnosis of myoepithelioma.

Amyloid goiter with parathyroid involvement: a case report and review of the literature

Although involvement of the thyroid gland by amyloid is a relatively common phenomenon, diffuse enlargement of the gland secondary to infiltration by amyloid is infrequent. We report the pathologic findings in a case of primary amyloid goiter with involvement of a parathyroid gland in an euthyroid patient who presented to his physician with an enlarged thyroid, upper airway obstructive symptoms, and dysphagia. The clinical features of amyloid goiter, including difficulties in preoperative diagnosis and management, are discussed.

Rapid detection of HSV from cytologic specimens collected into ThinPrep fixative

OBJECTIVE

Herpes simplex virus (HSV) infection is associated with substantial morbidity and mortality in neonates. A diagnosis of HSV on cervical cytologic studies could lead to a cesarean section, with an increase in the risk of maternal morbidity. The identification of viral lesions in sexually active women has medical and social implications. There have been reports of false positive diagnoses of HSV in patients with altered endocervical cells and with cervical intraepithelial neoplasia 3. We evaluated a polymerase chain reaction (PCR)-based assay to detect HSV-1 and HSV-2 in routinely collected cervical cytology specimens in ThinPrep fixative (Cytyc Corp., Marlborough, Massachussets, U.S.A.).

STUDY DESIGN

DNA was extracted from five cases that demonstrated cytologic changes suggestive of an HSV infection. PCR amplification with subsequent gel electrophoresis was performed to detect the presence of HSV.

RESULTS

HSV DNA was detected in three of five cases that were cytologically diagnosed as suspicious or strongly suspicious for HSV infection.

CONCLUSION

The combination of the ThinPrep liquid-based method for cervical cytology with PCR allows prompt confirmation of the diagnosis of HSV without sacrificing the diagnostic morphology on the slide.