A collaborative workflow between pathologists and deep learning for the evaluation of tumour cellularity in lung adenocarcinoma.

AIMS: The reporting of tumour cellularity in cancer samples has become a mandatory task for pathologists. However, the estimation of tumour cellularity is often inaccurate. Therefore, we propose a collaborative workflow between pathologists and artificial intelligence (AI) models to evaluate tumour cellularity in lung cancer samples and propose a protocol to apply it to routine practice. METHODS AND RESULTS: We developed a quantitative model of lung adenocarcinoma that was validated and tested on 50 cases, and a collaborative workflow where pathologists could access the AI results and adjust their original tumour cellularity scores (adjusted-score) that we tested on 151 cases. The adjusted-score was validated by comparing them with a ground truth established by manual annotation of haematoxylin and eosin slides with reference to immunostains with thyroid transcription factor-1 and napsin A. For training, validation, testing the AI and testing the collaborative workflow, we used 40, 10, 50 and 151 whole slide images of lung adenocarcinoma, respectively. The sensitivity and specificity of tumour segmentation were 97 and 87%, respectively, and the accuracy of nuclei recognition was 99%. One pathologist's visually estimated scores were compared to the adjusted-score, and the pathologist's scores were altered in 87% of cases. Comparison with the ground truth revealed that the adjusted-score was more precise than the pathologists' scores (P < 0.05). CONCLUSION: We proposed a collaborative workflow between AI and pathologists as a model to improve daily practice and enhance the prediction of tumour cellularity for genetic tests.

A case of discrepancy between three ERA tests in a woman with repeated implantation failure complicated by chronic endometritis.

BACKGROUND: Endometrial receptivity array (ERA) is used to determine the timing of embryo transfer (ET) synchronized with the window of implantation (WOI). The effectiveness and evaluation of ERAs in women with recurrent implantation failure remain controversial. We report the case of a patient with recurrent implantation failure that raises the issue of reproducibility of ERA tests. CASE REPORT: A 36-year-old Japanese woman with secondary infertility who had previously given birth failed to conceive after three frozen-thawed embryo transfer (FET) cycles. An ERA test was conducted to confirm the WOI. The first ERA test was performed 125 h after progesterone exposure. The laboratory reported that the endometrium was in a non-receptive (post-receptive) phase, and recommended retesting 101 h after progesterone exposure. A simultaneous chronic endometritis (CE) test showed a score of 3. After the antibiotics administration to treat CE, the second ERA test was performed after 101 h of progesterone exposure. The laboratory reported that the endometrium had not reached the WOI and estimated the WOI to be 113 ± 3 h after progesterone exposure. The third ERA test was performed 113 h after progesterone exposure. The laboratory reported that the endometrium was in a non-receptive (pre-receptive) phase and estimated the WOI to be 137 ± 3 h after progesterone exposure. A CE test performed at the same time as the second and third ERA tests showed a score of 1 for the collected endometrium. According to the third ERA test results, the vitrified-warmed blastocyst was transferred at 137 h of progesterone exposure. Pregnancy was achieved and the patient had an uncomplicated vaginal delivery at 39 weeks. One year later, another pregnancy was achieved after FET at 137 h of progesterone exposure, and the patient delivered at 33 weeks due to an unexpected membrane rupture. CONCLUSION: Because the results of the ERA test may vary in the presence of CE, CE should be diagnosed simultaneously with or before conducting ERA tests. If CE is diagnosed, ERA testing should be performed after treatment with antimicrobials or other drugs.

Detection of Lung Cancer Lymph Node Metastases from Whole-Slide Histopathologic Images Using a Two-Step Deep Learning Approach.

The application of deep learning for the detection of lymph node metastases on histologic slides has attracted worldwide attention due to its potentially important role in patient treatment and prognosis. Despite this attention, false-positive predictions remain problematic, particularly in the case of reactive lymphoid follicles. In this study, a novel two-step deep learning algorithm was developed to address the issue of false-positive prediction while maintaining accurate cancer detection. Three-hundred and forty-nine whole-slide lung cancer lymph node images, including 233 slides for algorithm training, 10 slides for validation, and 106 slides for evaluation, were collected. In the first step, a deep learning algorithm was used to eliminate frequently misclassified noncancerous regions (lymphoid follicles). In the second step, a deep learning classifier was developed to detect cancer cells. Using this two-step approach, errors were reduced by 36.4% on average and up to 89% in slides with reactive lymphoid follicles. Furthermore, 100% sensitivity was reached in cases of macrometastases, micrometastases, and isolated tumor cells. To reduce the small number of remaining false positives, a receiver-operating characteristic curve was created using foci size thresholds of 0.6 mm and 0.7 mm, achieving sensitivity and specificity of 79.6% and 96.5%, and 75.5% and 98.2%, respectively. A two-step approach can be used to detect lung cancer metastases in lymph node tissue effectively and with few false positives.

The effect of tolvaptan on renal excretion of electrolytes and urea nitrogen in patients undergoing coronary artery bypass surgery.

BACKGROUND: Adequate fluid management is an important component of patient care following cardiac surgery. Our aim in this study was to determine the benefits of tolvaptan, an oral selective vasopressin-2 receptor antagonist that causes electrolyte-free water diuresis, in postoperative fluid management. We prospectively examined the effect of tolvaptan on renal excretion of electrolytes and urea nitrogen in cardiac surgery patients. METHODS: Patients undergoing coronary artery bypass surgery were randomized to receive conventional loop diuretics (Group C, n = 30) or conventional loop diuretic therapy plus tolvaptan (Group T, n = 27). Fractional excretions of sodium (FENA), potassium (FEK) and urea nitrogen (FEUN) were measured in both groups during post-surgical hospitalization. RESULTS: Urine output was greater with tolvaptan (Group T) than without it (Group C), and some patients in Group C required intravenous as well as oral loop diuretics. Serum sodium concentrations decreased after surgery in Group C, but were unchanged in Group T (postoperative day [POD] 3, 139.8 ± 3.5 vs. 142.3 ± 2.6 mEq/L, p = 0.006). However, postoperative FENA values in Group C did not decrease, and the values were similar in both groups. Serum potassium levels remained lower and FEK values remained higher than the preoperative values, but only in Group C (all p < 0.05). BUN increased postoperatively in both groups, but it remained higher than its preoperative value only in Group C (all p < 0.01). Group T showed an initial increase in BUN, which peaked and then returned to its preoperative value within a week. The FEUN increased postoperatively in both groups, but the change was more pronounced in Group T (POD7, 52.7 ± 9.3 vs. 58.2 ± 6.5 %, p = 0.025). CONCLUSIONS: Renal excretion of sodium and potassium reflects the changes in serum concentration in patients treated with tolvaptan. Patients treated only with loop diuretics showed a continuous excretion of sodium and potassium that led to electrolyte imbalance, whereas the combination of loop diuretics and tolvaptan increased renal urea nitrogen elimination. Tolvaptan therefore appears to be an effective diuretic that minimally affects serum electrolytes while adequately promoting the elimination of urea nitrogen from the kidneys in patients undergoing coronary artery bypass surgery. TRIAL REGISTRATION: The present study is registered with the UMIN Clinical Trials Registry (ID: UMIN000011039 ).

Substantial improvement of histopathological diagnosis by whole-slide image-based remote consultation.

Consultation by subspecialty experts is the most common mode of rendering diagnosis in challenging cases in pathological practice. Our study aimed to highlight the diagnostic benefits of whole-slide image (WSI)-based remote consultation. We obtained diagnostically challenging cases from two institutions from the years 2010 and 2013, with histological diagnoses that contained keywords "probable," "suggestive," "suspicious," "inconclusive," and "uncertain." A total of 270 cases were selected for remote consultation using WSIs scanned at 40 × . The consultation process consisted of three rounds: the first and second rounds each with 12 subspecialty experts and the third round with six multi-expertise senior pathologists. The first consultation yielded 44% concordance, and a change in diagnosis occurred in 56% of cases. The most frequent change was from inconclusive to definite diagnosis (30%), followed by minor discordance (14%), and major discordance (12%). Out of the 70 cases which reached the second round, 31 cases showed discrepancy between the two consultants. For these 31 cases, a consensus diagnosis was provided by six multi-expertise senior pathologists. Combining all WSI-based consultation rounds, the original inconclusive diagnosis was changed in 140 (52%) out of 266 cases. Among these cases, 80 cases (30%) upgraded the inconclusive diagnosis to a definite diagnosis, and 60 cases (22%) changed the diagnosis with major or minor discordance, accounting for 28 cases (10%) and 32 cases (12%), respectively. We observed significant improvement in the pathological diagnosis of difficult cases by remote consultation using WSIs, which can further assist in patient healthcare. A post-study survey highlighted various benefits of WSI-based consults.

Histologic factors associated with nintedanib efficacy in patients with idiopathic pulmonary fibrosis.

BACKGROUND: Histopathologic factors predictive of nintedanib efficacy in idiopathic pulmonary fibrosis have not been studied. We aimed to describe the characteristics, focusing on histopathology, of idiopathic pulmonary fibrosis patients who did and did not respond to nintedanib. METHODS: This study retrospectively examined the clinicoradiopathologic features of 40 consecutive patients with surgical lung biopsy-confirmed idiopathic pulmonary fibrosis treated with nintedanib. Additionally, we compared the histopathologic scoring of 21 microscopic features between patients with functional or radiological progression and those with non-progression during 12 months of treatment. RESULTS: The histopathologic evaluation showed edematous changes in the interlobular septum as the only histologic finding observed more frequently in patients with both functional and radiological progression than in those without (58% vs. 14%, P = 0.007 and 50% vs. 0%, P = 0.003, respectively). Regarding per-year change, patients with edematous changes in the interlobular septum showed greater progression in median changes in spared area (-12%, interquartile range: [-25%--5%], vs. -3% [-7%-0%], P = 0.004) and reticular shadow (7% [3%-13%], vs. 0% [0%-5%], P = 0.041) on computed tomography. Functional and radiological progression-free survival were shorter in patients with edematous changes in the interlobular septum than in those without (6.6 months, 95% confidence interval: [5.9-25.3], vs. event <50%, [12.1-Not available], P = 0.0009, and 6.1 months, [5.2-6.6] vs. 14.5 months [7.8-not available], P<0.0001). CONCLUSIONS: Edematous changes in the interlobular septum may indicate poor nintedanib efficacy in idiopathic pulmonary fibrosis. Further studies are needed to validate these findings and address the mechanism behind ECIS.

A narrative review of digital pathology and artificial intelligence: focusing on lung cancer.

The emergence of whole slide imaging technology allows for pathology diagnosis on a computer screen. The applications of digital pathology are expanding, from supporting remote institutes suffering from a shortage of pathologists to routine use in daily diagnosis including that of lung cancer. Through practice and research large archival databases of digital pathology images have been developed that will facilitate the development of artificial intelligence (AI) methods for image analysis. Currently, several AI applications have been reported in the field of lung cancer; these include the segmentation of carcinoma foci, detection of lymph node metastasis, counting of tumor cells, and prediction of gene mutations. Although the integration of AI algorithms into clinical practice remains a significant challenge, we have implemented tumor cell count for genetic analysis, a helpful application for routine use. Our experience suggests that pathologists often overestimate the contents of tumor cells, and the use of AI-based analysis increases the accuracy and makes the tasks less tedious. However, there are several difficulties encountered in the practical use of AI in clinical diagnosis. These include the lack of sufficient annotated data for the development and validation of AI systems, the explainability of black box AI models, such as those based on deep learning that offer the most promising performance, and the difficulty in defining the ground truth data for training and validation owing to inherent ambiguity in most applications. All of these together present significant challenges in the development and clinical translation of AI methods in the practice of pathology. Additional research on these problems will help in resolving the barriers to the clinical use of AI. Helping pathologists in developing knowledge of the working and limitations of AI will benefit the use of AI in both diagnostics and research.

Concordance between sequential transbronchial lung cryobiopsy and surgical lung biopsy in patients with diffuse interstitial lung disease.

BACKGROUND: Increasing evidence indicates the utility of transbronchial lung cryobiopsy (TBLC) for the diagnosis of interstitial lung disease (ILD). However, only one study has compared TBLC and surgical lung biopsy (SLB) performed on the same patients. METHODS: We identified seven patients with ILD with TBLC and SLB. We evaluated the clinical characteristics and made a pathological diagnosis based on the official ATS/ERS/JRS/ALAT clinical practice guideline of idiopathic pulmonary fibrosis with both TBLC and SLB. RESULTS: Six cases were diagnosed as Usual interstitial pneumonia (UIP) in both TBLC and SLB. One case was diagnosed as indeterminate for UIP with TBLC and probable UIP with SLB. Etiological diagnosis with TBLC and SLB were concordant in 2 cases of idiopathic pulmonary fibrosis (IPF) but discordant for other diagnoses. Major histological findings of UIP including dense fibrosis, peripheral distribution, and fibroblastic foci showed high concordance between TBLC and SLB, which implies that TBLC can reliably detect these features. In contrast, loose fibrosis, cellular infiltration, and airway disease showed poor concordance between the two methods. CONCLUSION: Our study showed that TBLC is useful for UIP diagnosis but not for other ILD. With a multidisciplinary approach, diagnosis of IPF may be determined by TBLC, whereas ILD other than IPF may require SLB.

Atypical Familial Mediterranean Fever Complicated with Gastrointestinal Amyloidosis Diagnosed due to Paroxysmal Arthralgia and Intractable Diarrhea, Successfully Treated with Tocilizumab.

A 53-year-old man with recurrent episodes of large joint pain and a low-grade fever at irregular intervals for 16 years developed right knee and ankle arthralgia, watery diarrhea, and abdominal pain. Following an ileum and colon biopsy, he was diagnosed with gastrointestinal amyloidosis. We suspected familial Mediterranean fever (FMF) based on his history and administered colchicine; his symptoms subsequently improved. Thus, he was diagnosed with atypical FMF. After tocilizumab administration, the amyloid deposits disappeared. This case suggests that physicians should consider FMF even in cases with atypical symptoms in order to prevent the progression of amyloidosis and that amyloid deposits can be eliminated by interleukin (IL)-6 inhibition.

Multi-region exome sequencing reveals genomic evolution from preneoplasia to lung adenocarcinoma.

There has been a dramatic increase in the detection of lung nodules, many of which are preneoplasia atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA) or invasive adenocarcinoma (ADC). The molecular landscape and the evolutionary trajectory of lung preneoplasia have not been well defined. Here, we perform multi-region exome sequencing of 116 resected lung nodules including AAH (n = 22), AIS (n = 27), MIA (n = 54) and synchronous ADC (n = 13). Comparing AAH to AIS, MIA and ADC, we observe progressive genomic evolution at the single nucleotide level and demarcated evolution at the chromosomal level supporting the early lung carcinogenesis model from AAH to AIS, MIA and ADC. Subclonal analyses reveal a higher proportion of clonal mutations in AIS/MIA/ADC than AAH suggesting neoplastic transformation of lung preneoplasia is predominantly associated with a selective sweep of unfit subclones. Analysis of multifocal pulmonary nodules from the same patients reveal evidence of convergent evolution.

Localized malignant mesenteric mesothelioma causing small bowel obstruction.

Malignant mesothelioma is an uncommon lethal neoplasm in the serous membrane in which peritoneal mesothelioma is a rarer form. Herein is reported a case of malignant mesothelioma presenting as a localized mass inside the mesentery causing focal luminal obstruction of the small intestine. The diagnosis of malignant mesothelioma was obtained on repeat double balloon endoscopic biopsy. Partial resection of the small intestine along with the mesentery was performed, followed by a course of chemotherapy. No relapse of the disease has been found in the 8 months' follow up radiologically. To the best of the authors' knowledge this is the first reported case of localized malignant mesothelioma arising inside the mesentery. Mesothelioma should be considered as the differential diagnosis when small bowel obstruction occurs with unknown primary neoplasm.

Early Postoperative Outcome of Off-Pump Coronary Artery Bypass Grafting: A Report from the Highest-Volume Center in Japan.

BACKGROUND: Off-pump coronary artery bypass grafting (OPCAB) has evolved into a standard technique in coronary artery bypass grafting. However, a detailed investigation in Japanese population undergoing OPCAB has not yet been fully elucidated. METHODS: A total of 1109 consecutive patients undergoing isolated OPCAB between 2006 and 2013 at Juntendo University were reviewed. The data was evaluated in the light of previously published OPCAB-associated reports. RESULTS: There were 904 male (81.5%) and a mean was 67.5 ± 9.8 years. Eight patients (0.5%) died within 30 days postoperatively or before discharge, which was equivalent to or rather better than the previously reported mortality rates, including the European System for Cardiac Operation Risk Evaluation II (2.1 ± 2.1) data. A morbidity analysis revealed that prolonged intubation (>24 h) occurred in 43 patients (3.6%), surgical site infections in 18 (1.9%), neurological complications in 13 (1.3%). A reduced preoperative left ventricular ejection fraction (less than 40%) was found to be a risk factor for early postoperative death (odds ratio 10.58, respectively, p <0.05) in our cohort. CONCLUSIONS: Early postoperative mortality and morbidity rates in Japanese population after OPCAB were rather satisfactory and similar to those reported by other countries.