OBJECTIVE: Artificial intelligence (AI)-based chatbots have demonstrated accuracy in a variety of fields, including medicine, but research has yet to substantiate their accuracy and clinical relevance. We evaluated an AI chatbot\'s answers to questions posed during a treatment planning conference.
METHODS: Pathology residents, pathology faculty, and an AI chatbot (OpenAI ChatGPT [January 30, 2023, release]) answered a questionnaire curated from a genitourinary subspecialty treatment planning conference. Results were evaluated by 2 blinded adjudicators: a clinician expert and a pathology expert. Scores were based on accuracy and clinical relevance.
RESULTS: Overall, faculty scored highest (4.75), followed by the AI chatbot (4.10), research-prepared residents (3.50), and unprepared residents (2.87). The AI chatbot scored statistically significantly better than unprepared residents (P = .03) but not statistically significantly different from research-prepared residents (P = .33) or faculty (P = .30). Residents did not statistically significantly improve after research (P = .39), and faculty performed statistically significantly better than both resident categories (unprepared, P < .01; research prepared, P = .01).
CONCLUSIONS: The AI chatbot gave answers to medical questions that were comparable in accuracy and clinical relevance to pathology faculty, suggesting promise for further development. Serious concerns remain, however, that without the ability to provide support with references, AI will face legitimate scrutiny as to how it can be integrated into medical decision-making.

OBJECTIVE: To characterize the role of pathology explanation clinics (PECs) in prostate cancer care and determine their impact on patients, urologic oncologists, and quality of care.
METHODS: Semistructured interviews with 10 patients with newly diagnosed prostate cancer were conducted before and after a PEC pilot and at the 1- and 6-month follow-up visits. Information about participants\' cancer knowledge and anxiety were collected quantitatively. Documented pathologist communications and proper review of outside biopsy slides were collected. Semistructured interviews were also completed with participating urologic oncologists following the pilot.
RESULTS: Pathology explanation clinics improved participants\' understanding of their diagnosis, cognitively and emotionally supporting them first in their urologic oncology visit and later in making an informed treatment decision. Mean knowledge scores were high, and a minority of participants had prostate cancer anxiety. Urologic oncologists noted improved understanding and reduced anxiety among participants, enabling nuanced conversations about prognosis and management during the visit. By ensuring review of outside biopsy slides and communication of clinically significant or unexpected diagnoses, PECs supported high-quality care and patient safety.
CONCLUSIONS: In this small pilot, PECs positively affected patients with prostate cancer, their clinicians, and the overall care system. Additional studies in larger populations and diverse settings will be useful.

Machine learning has been leveraged for image analysis applications throughout a multitude of subspecialties. This position paper provides a perspective on the evolutionary trajectory of practical deep learning tools for genitourinary pathology through evaluating the most recent iterations of such algorithmic devices. Deep learning tools for genitourinary pathology demonstrate potential to enhance prognostic and predictive capacity for tumor assessment including grading, staging, and subtype identification, yet limitations in data availability, regulation, and standardization have stymied their implementation.

UNASSIGNED: Kidney cancer accounted for 1. 8% of global cancer deaths according to Globocan 2020 estimates, with most of these being renal cell carcinomas. Lower rates of renal cell carcinoma are reported for Africa and these are expected to change for a combination of reasons. The clinical and morphologic characteristics of renal cell carcinoma seen within Kenya have not been described before. This study aims to partially fill this gap.
UNASSIGNED: This was a cross-sectional descriptive study examining electronic histopathology reports from the Aga Khan University Hospital Nairobi Laboratory for the period January 2016 to May 2022.
UNASSIGNED: Sixty cases of renal cell carcinoma were identified. The mean age at diagnosis was 55.3 years. The most common histologic subtype diagnosed was clear cell renal cell carcinoma (41.7%), followed by papillary renal cell carcinoma and renal cell carcinoma not further specified (both 21.7%), and chromophobe renal cell carcinoma (11.7%). The most frequent specimen type was resection, followed by cores of renal masses. The mean tumor size was 8.5 cm. Sixty-seven percent of patients presented with Stage III and above.
UNASSIGNED: Renal masses were the commonest clinical indication for biopsy among the records reviewed. The male to female ratio, as well as the mean age at presentation were comparable to what is described in literature for other regions of the world. The proportions of the commonest histologic subtypes matched what is described in other parts of the world. Challenges in the identification of histologic subtypes included having a limited panel of antibodies for diagnosis and the lack of genetic molecular tests for histotyping.
UNASSIGNED: The spectrum of histologic subtypes of renal cell carcinoma seen at a tertiary referral hospital in Nairobi, Kenya was similar to that described in other parts of Africa and the globe. The age at presentation with renal cell carcinoma was consistent with what has been described in literature. Challenges were identified in the accurate histotyping of renal cell carcinoma due to constrained resources. Majority of cases diagnosed presented at advanced stage.

Pathologic diagnosis of flat urothelial lesions is subject to high interobserver variability. We expected that deep learning could improve the accuracy and consistency of such pathologic diagnosis, although the learning process is a black box. We therefore propose a new approach for pathologic image classification incorporating the diagnostic process of the pathologist into a deep learning method.
A total of 267 H&E-stained slides of normal urothelium and urothelial lesions from 127 cases were examined. Six independent convolutional neural networks were trained to classify pathologic images according to six pathologic criteria. We then used these networks in the main training for the final diagnosis.
Compared with conventional manual analysis, our method significantly improved the classification accuracy of images of flat urothelial lesions. The automated classification showed almost perfect agreement (weighted κ = 0.98) with the consensus reading. In addition, our approach provides the advantages of reliable diagnosis corresponding to histologic interpretation.
We used deep learning to establish an automated subtype classifier for flat urothelial lesions that successfully combines traditional morphologic approaches and complex deep learning to achieve a learning mechanism that seems plausible to the pathologist.

OBJECTIVE: To evaluate the frequencies and types of disagreements in a contemporary urological second-opinion consult service in order to improve pathologist awareness.
METHODS: For 7 years ending 30 October 2021, records were kept of our department\'s total urologic outside consultation and disagreed-upon cases. Disagreements were categorized according to specimen type and nature of conflict. All grading and staging assignments used International Society of Urological Pathology (ISUP) criteria. Statistical analyses for each specimen type included the percent disagreement. Cohen\'s kappa analysis was done to measure interrater reliability on the prostate biopsies, prostatectomies, and the bladder biopsies/resections. In addition, for the prostate biopsies, the potential for change in treatment candidacy calculation (CTC), was assessed as sum of changes from cancer to non-malignant tissue or the reverse, plus changes from Gleason Grade group (GG)1 to GG ≥ 2 (3 +4 =7) or the reverse.
RESULTS: Overall mean disagreement rate for all specimens was 15.2%. The highest rate was among 1545 prostate biopsy cases, where 410 contained disagreements (26.5%). 118 (7.6%) met criteria for CTC: 10 cases were altered from cancer to non-cancer, 38 cases downgraded from GG≥ 2 to GG1, and 70 upgraded from GG1 to GG≥ 2. Second opinion downgraded the overall highest GG more often than it upgraded it, with downgrade:upgrade ratios of 64:37 for the GG1/GG2 threshold, 79:67 for the GG2/GG3, and 14:0 for the GG3/GG4. 146 specimen parts had disagreements as to cancer vs. suspicious vs. benign, with 85 undercalled and 61 overcalled. Other rates of disagreement included: prostatectomy 34/198 (17.2%); bladder resection or biopsy 68/591 (11.5%); kidney 27/175 (15.4%); and orchiectomy 9/82 (11.0%). In bladder specimens, overgrading was 6X more frequent than undergrading; and overstaging muscularis propria invasion was 6X more frequent than understaging.
CONCLUSIONS: The review of uropathologic materials before definitive therapy can lead to changes that impact clinical decisions significantly. As an example, for prostate biopsies, candidacy for active surveillance versus definitive treatment hinges on GG1 versus 2 and this distinction constituted most CTC cases. The above findings highlight aspects of urological pathology to be emphasized to residents in training, and pathologists in practice.

Eosinophilic solid and cystic renal cell carcinoma (ESC-RCC) is an emerging entity in renal neoplasia with distinctive histopathological findings and a generally favorable prognosis. The presence of melanin pigment in a renal tumor typically prompts the observer to consider the microphthalmia-associated transcription family translocation renal cell carcinomas. We present a renal tumor occurring in a 19-year-old male patient which had the typical morphology of ESC-RCC but showed the additional finding of focal melanin pigment. This tumor showed strong and diffuse positive immunolabeling with paired box gene 8 and cytokeratin 20, and was negative with epithelial membrane antigen, carbonic anhydrase 9, CD117, cytokeratin 7, and transcription factor E3. Human melanoma black-45 showed focal positivity, but Melan-A was negative. Next-generation sequencing revealed a mutation in the TSC2 gene (c.4490C > G, p.[Pro1497Arg] and c.1257 + 1del) and break apart fluorescence in-situ hybridization with TFE3 and TFEB probes was negative. In this case report, we present the novel finding of melanin pigment occurring in a genetically proven and otherwise typical ESC-RCC, and briefly discuss the differential diagnostic considerations.

Transcription factor E3-rearranged renal cell carcinoma (TFE3-RCC) has heterogenous morphologic and immunohistochemical (IHC) features.131 pathologists with genitourinary expertise were invited in an online survey containing 23 questions assessing their experience on TFE3-RCC diagnostic work-up.Fifty (38%) participants completed the survey. 46 of 50 participants reported multiple patterns, most commonly papillary pattern (almost always 9/46, 19.5%; frequently 29/46, 63%). Large epithelioid cells with abundant cytoplasm were the most encountered cytologic feature, with either clear (almost always 10/50, 20%; frequently 34/50, 68%) or eosinophilic (almost always 4/49, 8%; frequently 28/49, 57%) cytology. Strong (3+) or diffuse (>75% of tumour cells) nuclear TFE3 IHC expression was considered diagnostic by 13/46 (28%) and 12/47 (26%) participants, respectively. Main TFE3 IHC issues were the low specificity (16/42, 38%), unreliable staining performance (15/42, 36%) and background staining (12/42, 29%). Most preferred IHC assays other than TFE3, cathepsin K and pancytokeratin were melan A (44/50, 88%), HMB45 (43/50, 86%), carbonic anhydrase IX (41/50, 82%) and CK7 (32/50, 64%). Cut-off for positive TFE3 fluorescent in situ hybridisation (FISH) was preferably 10% (9/50, 18%), although significant variation in cut-off values was present. 23/48 (48%) participants required TFE3 FISH testing to confirm TFE3-RCC regardless of the histomorphologic and IHC assessment. 28/50 (56%) participants would request additional molecular studies other than FISH assay in selected cases, whereas 3/50 participants use additional molecular cases in all cases when TFE3-RCC is in the differential.Optimal diagnostic approach on TFE3-RCC is impacted by IHC and/or FISH assay preferences as well as their conflicting interpretation methods.

Mismatch repair-deficient (d-MMR) tumours have been reported to show susceptibility to immune checkpoint inhibitors targeting programmed death-1/PD ligand-1 (PD-1/PD-L1). In this study, we sought to correlate the association of d-MMR, PD-L1 and CD8 expression in muscle invasive, high-grade urothelial carcinoma (HGUC) of bladder. A tissue microarray (TMA) was constructed from 201 cases and sequentially stained with PD-L1, CD8, MSH2, MSH6, MLH1 and PMS2. PD-L1 was assessed in tumour and immune cells. CD8 was assessed in a hotspot fashion with results averaged across cores. Loss of nuclear MMR expression on TMA sections was further assessed using corresponding whole tissue sections. d-MMR was identified in four cases (2%). The mean CD8 count was significantly higher in d-MMR tumours (10 vs 35, p=0.007) as was the proportion of PD-L1 positivity (75% vs 20%, p=0.031). d-MMR is uncommon in HGUC of bladder but shows strong correlation with cytotoxic T lymphocyte infiltration and PD-L1 tissue expression.

暂无摘要。