A Computationally Virtual Histological Staining Method to Ovarian Cancer Tissue by Deep Generative Adversarial Networks.

Histological analysis to tissue samples is elemental for diagnosing the risk and severity of ovarian cancer. The commonly used Hematoxylin and Eosin (H&E) staining method involves complex steps and strict requirements, which would seriously impact the research of histological analysis of the ovarian cancer. Virtual histological staining by the Generative Adversarial Network (GAN) provides a feasible way for these problems, yet it is still a challenge of using deep learning technology since the amounts of data available are quite limited for training. Based on the idea of GAN, we propose a weakly supervised learning method to generate autofluorescence images of unstained ovarian tissue sections corresponding to H&E staining sections of ovarian tissue. Using the above method, we constructed the supervision conditions for the virtual staining process, which makes the image quality synthesized in the subsequent virtual staining stage more perfect. Through the doctors' evaluation of our results, the accuracy of ovarian cancer unstained fluorescence image generated by our method reached 93%. At the same time, we evaluated the image quality of the generated images, where the FID reached 175.969, the IS score reached 1.311, and the MS reached 0.717. Based on the image-to-image translation method, we use the data set constructed in the previous step to implement a virtual staining method that is accurate to tissue cells. The accuracy of staining through the doctor's assessment reached 97%. At the same time, the accuracy of visual evaluation based on deep learning reached 95%.

Polyclonal PAX8 expression in carcinomas of the biliary tract - Frequent non-specific staining represents a potential diagnostic pitfall.

Paired box protein 8 (PAX8) is a transcription factor that is considered a relatively specific marker of carcinomas of the thyroid, kidney, and Müllerian/Wolffian duct derivatives. Unexpected PAX8 immunoreactivity has occasionally been reported in other tumors. The frequency of PAX8 expression in carcinomas of the biliary tract is not well studied. We evaluated the immunohistochemical expression of PAX8 in 73 cases of biliary tract carcinoma. We found that 28 of 73 (38%) biliary tract carcinomas had variable immunoreactivity for PAX8, assessed by a widely used polyclonal antibody (ProteinTech Group, Chicago, IL). This included 3 (4%) of cases with strong diffuse, and 14 (19%) of cases with strong focal staining. Strong PAX8 expression was more frequent in distal bile duct carcinomas than other biliary sites (p = 0.015), and showed a weak association with advanced T stage (T3-T4 versus T1-T2; p = 0.09). No correlation was observed between PAX8 positivity and age at diagnosis, gender, or lymph node metastasis. The 28 polyclonal PAX8-positive cases were largely negative for monoclonal PAX8 and PAX6 immunostains, with only rare tumor cells with weak immunoreactivity being present in a subset of cases. We show that a substantial fraction of biliary tract carcinomas exhibit immunoreactivity with a widely used polyclonal PAX8 antibody. Pathologists should be aware of this potential pitfall during the diagnostic workup of hepatobiliary lesions to avoid misdiagnosis as a metastasis from a PAX8-positive tumor.

Malignancy is in the eye of the beholder: Pathologic diagnosis of challenging follicular neoplasms in the era of noninvasive follicular thyroid neoplasms with papillary-like nuclear features and immunohistochemical and molecular adjuncts.

BACKGROUND: Classification of thyroid follicular neoplasms can be challenging for pathologists. Introduction of noninvasive follicular thyroid neoplasms with papillary-like nuclear features, the utilization of immunohistochemistry, and molecular analysis are all thought to be valuable diagnostic adjuncts. Our aim was to determine whether interobserver variability for follicular neoplasms has improved since the application of these adjuncts. METHODS: One representative section from a cohort of follicular neoplasms previously proven difficult for pathologists were examined independently by 7 pathologists and assigned to 1 of 3 diagnostic categories (benign, neoplasms with papillary-like nuclear features, or malignant). This process was carried out separately 3 times: (1) after viewing hematoxylin and eosin stain slides, (2) hematoxylin and eosin stain in conjunction with immunohistochemistry, and (3) hematoxylin and eosin stain/immunohistochemistry in conjunction with molecular analysis. The interobserver variability and overall agreement were then calculated using the free-marginal kappa coefficient. RESULTS: Agreement on hematoxylin and eosin stain was 57%, with a kappa coefficient of 0.36 (minimal agreement). The agreement improved slightly with the application of immunohistochemistry (kappa coefficient = 0.49 [weak agreement] and a percentage agreement 67%). The level of agreement decreased slightly after the addition of molecular analysis (kappa coefficient = 0.43 [weak agreement] and percentage agreement 62%). CONCLUSION: Despite attempts to standardize the diagnostic criteria for neoplasms with papillary-like nuclear features and the utilization immunohistochemistry and molecular analysis, attaining pathologic consensus for difficult follicular neoplasms of the thyroid remains a challenge.

The importance of histological patterns on PD-L1 staining heterogeneity: Should we use pattern-based approach for selecting tumor samples for PD-L1 testing in lung adenocarcinomas?

Background/aim: Programmed death ligand-1 (PD-L1) is a predictive marker for immunotherapeutic agents. However, heterogeneous staining of PD-L1 can cause false-negative results. The aim of this study is to evaluate the importance of histological patterns on PD-L1 staining heterogeneity in lung adenocarcinomas (LAC). Materials and methods: PD-L1 immunohistochemistry (IHC) stain was performed to two different tissue cores of 128 LAC cases, and cut-off values are given for grouping the cases according to the percentage of staining (1%-10%, 11%-49%, 50%-100%). Staining rates between cores were compared and analyzed by their histological patterns. Also, the relation of the PD-L1 expression with the clinicopathological characteristics of the cases was analyzed. Results: Overall, PD-L1 expression was observed in 53 of 128 cases (41.4%, 1% cut-off), 23.5% of them were positive at 10% cut-off and 14.1% at 50% cut-off. PD-L1 expression was significantly related to the high grade micropapillary and solid patterns of adenocarcinomas (p:0.01). Staining cut-offs were mostly similar between cores (43/50, 86%) (k:0.843). However, 14% of them were positive only in one core (7 of 50). This false negativity was mostly related to the histological patterns. Conclusion: Our data reveal the heterogeneous staining of PD-L1 expression, also micropapillary and solid patterns show higher rates of PDL expression. Therewithal, these findings also highlight the importance of taking into consideration of histological patterns, when choosing a paraffin block for the PDL1.

Effects of various fixatives and temperature on the quality of glycogen demonstration in the brain and liver tissues.

The visualization of glycogen deposits in cells and tissues is important for studying glycogen metabolism as well as diagnosis of glycogen storage diseases. Evidence suggests that the demonstration of glycogen can better be enhanced by factors such the choice of fixative and temperature during fixation. Here, we assessed efficacy of neutral buffered formalin (NBF), alcoholic formalin (AF) and paraformaldehyde (PFA) at 4 °C, 37 °C and 40 °C using Periodic Acid Schiff's staining method. Each liver specimen was fixed in NBF and AF while the brain tissues were fixed in NBF, AF and PFA. We found that there was a better PAS staining intensity with the liver tissues fixed in AF compared with NBF. Also, there was no difference in the quality of the staining for tissues fixed in AF at 37 °C, 4 °C and 40 °C, but fixation with NBF at 4 °C gave the best staining quality when compared with 40 °C and 37 °C. Furthermore, hippocampal tissues fixed in AF showed better quality of PAS staining compared with NBF and PFA. A significant increase in staining intensity was observed for PFA when compared with NBF. Superior staining intensity for PAS was observed at 4 °C for hippocampal tissues fixed with NBF, AF and PFA. Taken together our results show that AF at a temperature of 4 °C gave the best result. Hence, glycogen demonstration can better be enhanced by the choice of fixative and temperature during fixation.

Immunohistochemical analysis of tumor budding as predictor of lymph node metastasis from superficial esophageal squamous cell carcinoma.

BACKGROUND: Tumor budding is known predictors of lymph node metastasis from esophageal squamous cell carcinoma. However, it is not easy to detect such small cell clusters on hematoxylin-eosin (HE) staining. Therefore, we evaluated tumor budding using immunohistochemistry (IHC) for epithelial cell markers. METHOD: We analyzed tumor budding in 50 cases of superficial esophageal squamous cell carcinoma. We evaluated the impact of clinicopathological factors and tumor budding to predict lymph node metastasis. A total of 565 tumor sections were assessed using HE staining and IHC for cytokeratin 5/6. RESULTS: Based on receiver operating characteristic curves, the cut-off values for high-grade tumor budding evaluated using HE staining or IHC were 2 and 11, respectively. High-grade tumor budding evaluated using HE staining (P = 0.007) and IHC (P ≤ 0.001) were significantly correlated with lymph node metastasis. For tumors with pT1a-MM to pT1b-SM1, high-grade tumor budding evaluated using IHC was correlated with lymph node metastasis (P = 0.050). CONCLUSIONS: Tumor budding was significantly associated with lymph node metastasis. The optimal cut-off values of tumor budding on HE staining and tumor budding on IHC were 2 and 11, respectively. Even though both tumor budding on HE staining and tumor budding on IHC were significantly associated with lymph node metastasis, tumor budding on IHC tend to be more associated with lymph node metastasis.

Presence of pink-color sign within 1 min after iodine staining has high diagnostic accordance rate for esophageal high-grade intraepithelial neoplasia/invasive cancer.

BACKGROUND/AIM: The dramatic color change after iodine staining (from white-yellow to pink after 2-3 min), designated as the "pink-color sign" (PCS), is indicative of esophageal high-grade intraepithelial neoplasia (HGIN) or an invasive lesion. However, no study has yet examined the association between the time of PCS appearance and histopathology. We investigated the association between the time of PCS appearance and esophageal histopathology in 456 lesions of 438 patients who were examined for suspected esophageal cancer. MATERIALS AND METHODS:: The records of 495 consecutive patients who had suspected esophageal cancer based on gastroscopy and who underwent Lugol's chromoendoscopy from January 2015 to March 2018 were retrospectively reviewed. The time of PCS appearance was recorded in all patients, and tissue specimens were examined. RESULTS: We examined 456 lesions in 438 patients. Use of PCS positivity at 2 min for the diagnosis of HGIN/invasive cancer had a sensitivity of 84.1%, a specificity of 72.7%, and an accuracy of 80.4%. We classified the PCS-positive patients in whom the time of PCS appearance was recorded (168 lesions) into 4 groups: 0-30, 31-60, 61-90, and 91-120 s. Based on a 60-s time for appearance of the PCS, the area under the receiver operating characteristic curve was 0.897, indicating good validity. At the optimal cutoff value of 60 s, the sensitivity was 90.2% and the specificity was 82.3%. The appearance of the PCS within 60 s had a diagnostic accordance rate of 88.6%, significantly higher than appearance of the PCS within 2 min (79.7%, P < 0.05). CONCLUSION: Appearance of the PCS within 1 min after iodine staining has a higher diagnostic accordance rate for esophageal HGIN/invasive cancer than appearance of the PCS at 2 min.

Stain Deconvolution Using Statistical Analysis of Multi-Resolution Stain Colour Representation.

Stain colour estimation is a prominent factor of the analysis pipeline in most of histology image processing algorithms. Providing a reliable and efficient stain colour deconvolution approach is fundamental for robust algorithm. In this paper, we propose a novel method for stain colour deconvolution of histology images. This approach statistically analyses the multi-resolutional representation of the image to separate the independent observations out of the correlated ones. We then estimate the stain mixing matrix using filtered uncorrelated data. We conducted an extensive set of experiments to compare the proposed method to the recent state of the art methods and demonstrate the robustness of this approach using three different datasets of scanned slides, prepared in different labs using different scanners.

Use of the term "rule out" in requisition forms may cause diagnostic delays in dermatopathology practice.

BACKGROUND: Timely pathologic diagnosis relies on the communication of specific clinical information in the requisition form (RF). Clinical information may use nonspecific terms such as "rule out", the use of which may result in diagnostic delays and the unnecessary application of pathology stains and sections. OBJECTIVE: This study was designed to evaluate the relationship between use of the term "rule out" and time to diagnosis, and the use of additional pathology stains and sections in integrated and non-integrated dermatopathology practices. METHODS: A retrospective double-cohort study of 475 RFs from the integrated practice (of which 182 used the term "rule out" [RO] and 293 did not [NRO]) and 412 RFs from the non-integrated practice (RO, n = 126; NRO, n = 286) was performed. RESULTS: No significant differences emerged between groups of patients with, respectively, RO and NRO RFs in the integrated practice with respect to time to diagnosis, and numbers of additional tissue sections or stains applied. By contrast, the use of RFs containing the term "rule out" was associated with significantly longer times to diagnosis and higher rates of use of pathology stains and sections in comparison with NRO RFs in the non-integrated practice. However, the study is limited by its status as a retrospective review of data sourced from a single institution. CONCLUSIONS: Use of the term "rule out" in RFs may not significantly impact key care delivery outcomes in an integrated practice. However, it may cause diagnostic delays and the use of unnecessary pathology services in a non-integrated practice.

X-FISH: Analysis of cellular RNA expression patterns using flow cytometry.

Fluorescent in situ hybridization (FISH) is a powerful technique for the detection of RNA or DNA within cells and tissues, which provides a unique link between molecular and cell biology. This technique is broadly applicable across a range of biological systems. While FISH has been previously adapted to flow-based platforms, their use remains limited because of procedural challenges and costs associated with commercial kits. Herein we present a protocol that modifies existing techniques to sensitively and specifically detect and examine RNA expression patterns in primary cells and cell lines using flow cytometry (expression-FISH; X-FISH). As relevant examples, we show how this technique can be used to monitor changes in mRNA expression following activation, how it can be combined with antibody staining to study RNA and protein in the same sample, and how it can help distinguish among subsets in a mixed cell population. X-FISH can integrate multiple probes and can be performed in conjunction with other assays, allowing for informative multiparametric analyses and increased statistical robustness. For non-classical comparative animal models this procedure provides a time saving alternative to de novo production of antibody-based markers. Finally, X-FISH provides an economical solution that is applicable to conventional as well as multi-spectral imaging flow cytometry platforms.

Variability in measuring the Ki-67 labeling index in patients with breast cancer.

BACKGROUND: Luminal-type breast cancer is divided into types A and B, depending on the Ki-67 labeling index (LI). However, the area at which Ki-67 is measured and the choice of specimen greatly affects the results. The aim of the present study was to evaluate the Ki-67 LI variability using different measurement methods and specimens. We also evaluated how the chemotherapy indication changed for luminal-type breast cancer using the different measurements. MATERIALS AND METHODS: The Ki-67 levels in 87 patients with breast cancer were assessed, and the Ki-67 LI was calculated. Five measurement sites were randomly selected, including the most densely labeled areas (hot spots) in both core needle biopsy (CNB) and surgical specimens. RESULTS: The intraclass correlation coefficient of the CNB and surgical specimens was 0.91 and 0.95, respectively. If the hot spot was used, the correlation coefficient (CC) between the CNB and surgical specimens was 0.635. If the average score was used, the CC was 0.730. If the average score was used, the CNB specimens indicated that 49 patients had a high Ki-67 LI, and 48 patients had a high Ki-67 LI using surgical specimens. If the hot spot was used, 60 patients using the CNB specimens and 58 patients using the surgical specimens had a high Ki-67 LI. If the average score was used, 17 patients were identified as being in different groups, and if the hot spot was used, 16 patients were identified as being in different groups, depending on the specimens that were used. CONCLUSION: The results differed according to the method and specimen type that was used.

Utilization and utility of immunohistochemistry in dermatopathology.

Immunohistochemistry (IHC) is considered a valuable ancillary tool for dermatopathology diagnosis, but few studies have measured IHC utilization by dermatopathologists or assessed its diagnostic utility. In a regionalized, community-based dermatopathology practice, we measured IHC utilization (total requests, specific antibodies requested, and final diagnosis) over a 12-month period. Next, we assessed diagnostic utility by comparing a preliminary "pre-IHC" diagnosis based on routine histochemical staining with the final diagnosis rendered after consideration of IHC results. The dermatopathology IHC utilization rate was 1.2%, averaging 3.6 stains requested per case. Melanocytic, hematolymphoid, and fibrohistiocytic lesions made up 23%, 18%, and 16%, respectively, of the total cases requiring IHC. S100 and Melan A were the most frequently requested stains, ordered on 50% and 34% of IHC cases, respectively. The utility study revealed that IHC changed the diagnosis in 11%, confirmed a diagnosis, or excluded a differential diagnosis in 77%, and was noncontributory in 4% of cases. Where IHC results prompted a change in diagnosis, 14% were a change from a benign to malignant lesion, whereas 32% changed from one malignant entity to another. IHC is most commonly used in cutaneous melanocytic and hematolymphoid lesions. In 11% of dermatopathology cases in which IHC is used, information is provided that changes the H&E diagnosis. Such changes may have significant treatment implications. IHC is noncontributory in only a small percentage of cases.

Analysis of immunohistochemical stain usage in different pathology practice settings.

This study compares the use of immunohistochemistry (IHC) for diagnosing carcinoma in private practice and commercial settings with use in a single academic center. H&E-stained slides and IHC stains, when present, of recently diagnosed carcinomas (n = 200) from patients referred to our institution for treatment were reviewed by a resident and mid-and senior-level pathologists. Diagnostic agreement between academic and referral pathologists was 98%; the former group used IHC stains in 11% and the latter in 26% of cases (P < .0001). Pathologists from commercial laboratories (12% of referrals) used IHC in 38% of cases, whereas private/hospital-based community laboratories (86% of referrals) used them in 24%. The average number of stains ordered per case was similar among all groups. We suggest that the use of IHC may reflect both the degree of experience of the pathologist and the pathology practice setting.

Evaluation of multitype mathematical models for CFSE-labeling experiment data.

Carboxy-fluorescein diacetate succinimidyl ester (CFSE) labeling is an important experimental tool for measuring cell responses to extracellular signals in biomedical research. However, changes of the cell cycle (e.g., time to division) corresponding to different stimulations cannot be directly characterized from data collected in CFSE-labeling experiments. A number of independent studies have developed mathematical models as well as parameter estimation methods to better understand cell cycle kinetics based on CFSE data. However, when applying different models to the same data set, notable discrepancies in parameter estimates based on different models has become an issue of great concern. It is therefore important to compare existing models and make recommendations for practical use. For this purpose, we derived the analytic form of an age-dependent multitype branching process model. We then compared the performance of different models, namely branching process, cyton, Smith-Martin, and a linear birth-death ordinary differential equation (ODE) model via simulation studies. For fairness of model comparison, simulated data sets were generated using an agent-based simulation tool which is independent of the four models that are compared. The simulation study results suggest that the branching process model significantly outperforms the other three models over a wide range of parameter values. This model was then employed to understand the proliferation pattern of CD4+ and CD8+ T cells under polyclonal stimulation.

Utility of formalin-fixed, paraffin-embedded liver biopsy specimens for global proteomic analysis in nonalcoholic steatohepatitis.

PURPOSE: To compare the proteomic profiles of formalin-fixed, paraffin-embedded (FFPE) liver biopsy material and matched frozen liver tissue from patients with nonalcoholic steatohepatitis (NASH). EXPERIMENTAL DESIGN: A label-free mass spectrometry-based approach was used to profile global protein expression in FFPE and frozen liver biopsy specimens from five patients with NASH. RESULTS: Eight hundred and sixty proteins were identified with >75% confidence: 225 common proteins were identified in both the FFPE and frozen tissues, and an additional 142 and 493 proteins were identified in the FFPE and frozen tissues, respectively. Functional analyses revealed a general, nonspecific reduction in the number of proteins identified in FFPE tissue compared with frozen tissue. No bias toward proteins located in any specific subcellular compartments or implicated in any particular biological functions was observed. The relative abundance of several proteins with functions relating to the pathogenesis of NASH (peroxiredoxin-1, fatty acid binding protein 1, fatty acid synthase, vimentin, catalase, and glutathione S-transferase A1) was similar in FFPE and frozen liver tissues. CONCLUSIONS AND CLINICAL RELEVANCE: FFPE liver biopsy material from NASH patients can be used for global proteomic analysis and biomarker identification studies, although a universal reduction in the number of identified proteins compared with frozen tissue is likely.

C4d staining in renal allograft biopsies with early acute rejection and subsequent clinical outcome.

BACKGROUND AND OBJECTIVES: Diffuse C4d staining in peritubular capillaries (PTCs) during an acute rejection episode (ARE) is the footprint of antibody-mediated rejection. In current clinical practice, diffuse C4d+ staining during acute rejection is regarded as an inferior prognostic sign. This case-control study investigated the prognostic role of mere C4d staining for graft outcome during an ARE in a well defined cohort of similarly ARE-treated patients. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: All kidney transplant recipients in the authors' center from January 1, 1995 to December 31, 2005 were reviewed. From these patients, 151 had a clinical ARE. Paraffin and/or frozen material was available for 128 patients showing a histologically proven ARE within the first 6 months after transplantation. All ARE patients were treated similarly with high-dose pulse steroids and in the case of steroid unresponsiveness with anti-thymocyte globulin. Biopsies were scored according to Banff criteria. Frozen and paraffin sections were stained by immunofluorescence (IF) and immunohistochemistry (IHC) for C4d, respectively, and scored for PTC positivity. RESULTS: Diffuse C4d+ staining in PTCs was found in 12.5% and 4.2% sections stained by IF or by IHC, respectively. Four patients showed diffuse positive staining with both methods but showed no different risk profile from other patients. No relation between C4d staining and clinical parameters at baseline was found. C4d staining was not associated with steroid responsiveness, graft, or patient survival. CONCLUSIONS: This study shows that C4d staining is not related to clinical outcome in this cohort of histologically proven early AREs.

Validation of lysyl oxidase as a prognostic marker for metastasis and survival in head and neck squamous cell carcinoma: Radiation Therapy Oncology Group trial 90-03.

PURPOSE: To validate lysyl oxidase (LOX), a hypoxia-related protein, as a marker for metastasis in an independent head and neck cancer (HNC) patient group enrolled onto a prospective trial. PATIENTS AND METHODS: We performed traditional immunohistochemical (IHC) staining and automated quantitative analysis (AQUA) for LOX expression in 66 HNC patients from one institution. We also performed AQUA staining for LOX in 306 of 1,113 patients treated on a phase III trial comparing four radiation fractionation schedules in locally advanced HNC (RTOG 90-03). Pretreatment characteristics and outcome were similar between patients with and without LOX assessment. We correlated AQUA LOX expression with time to metastasis (TTM), time to progression (TTP), and overall survival (OS). RESULTS: LOX expression from both staining methods predicted for TTM in the first 66 patients. Multivariate analysis, controlling for significant parameters including nodal stage and performance status, revealed tumor LOX expression, as a continuous variable, was an independent predictor for TTM (hazard ratio [HR], 1.21; 95% CI, 1.10 to 1.33; P = .0001), TTP (HR, 1.06; 95% CI, 1.02 to 1.10; P = .0069), and OS (HR, 1.04; 95% CI, 1.00 to 1.07; P = .0311) in RTOG 90-03 patients. This translates into a 259% increase in metastatic risk for a patient at the 75th percentile of LOX compared with one at the 25th percentile. CONCLUSION: AQUA LOX expression was strongly associated with increased metastasis, progression, and death in RTOG 90-03 patients. This study validates that LOX is a marker for metastasis and survival in HNC.