Deep Learning Can Predict Bevacizumab Therapeutic Effect and Microsatellite Instability Directly from Histology in Epithelial Ovarian Cancer.

Epithelial ovarian cancer (EOC) remains a significant cause of mortality among gynecologic cancers, with the majority of cases being diagnosed at an advanced stage. Before targeted therapies were available, EOC treatment relied largely on debulking surgery and platinum-based chemotherapy. Vascular endothelial growth factors have been identified as inducing tumor angiogenesis. According to several clinical trials, anti-vascular endothelial growth factor-targeted therapy with bevacizumab was effective in all phases of EOC treatment. However, there are currently no biomarkers accessible for regular therapeutic use despite the importance of patient selection. Microsatellite instability (MSI), caused by a deficiency of the DNA mismatch repair system, is a molecular abnormality observed in EOC associated with Lynch syndrome. Recent evidence suggests that angiogenesis and MSI are interconnected. Developing predictive biomarkers, which enable the selection of patients who might benefit from bevacizumab-targeted therapy or immunotherapy, is critical for realizing personalized precision medicine. In this study, we developed 2 improved deep learning methods that eliminate the need for laborious detailed image-wise annotations by pathologists and compared them with 3 state-of-the-art methods to not only predict the efficacy of bevacizumab in patients with EOC using mismatch repair protein immunostained tissue microarrays but also predict MSI status directly from histopathologic images. In prediction of therapeutic outcomes, the 2 proposed methods achieved excellent performance by obtaining the highest mean sensitivity and specificity score using MSH2 or MSH6 markers and outperformed 3 state-of-the-art deep learning methods. Moreover, both statistical analysis results, using Cox proportional hazards model analysis and Kaplan-Meier progression-free survival analysis, confirm that the 2 proposed methods successfully differentiate patients with positive therapeutic effects and lower cancer recurrence rates from patients experiencing disease progression after treatment (P < .01). In prediction of MSI status directly from histopathology images, our proposed method also achieved a decent performance in terms of mean sensitivity and specificity score even for imbalanced data sets for both internal validation using tissue microarrays from the local hospital and external validation using whole section slides from The Cancer Genome Atlas archive.

Annotation-Free Deep Learning-Based Prediction of Thyroid Molecular Cancer Biomarker BRAF (V600E) from Cytological Slides.

Thyroid cancer is the most common endocrine cancer. Papillary thyroid cancer (PTC) is the most prevalent form of malignancy among all thyroid cancers arising from follicular cells. Fine needle aspiration cytology (FNAC) is a non-invasive method regarded as the most cost-effective and accurate diagnostic method of choice in diagnosing PTC. Identification of BRAF (V600E) mutation in thyroid neoplasia may be beneficial because it is specific for malignancy, implies a worse prognosis, and is the target for selective BRAF inhibitors. To the authors' best knowledge, this is the first automated precision oncology framework effectively predict BRAF (V600E) immunostaining result in thyroidectomy specimen directly from Papanicolaou-stained thyroid fine-needle aspiration cytology and ThinPrep cytological slides, which is helpful for novel targeted therapies and prognosis prediction. The proposed deep learning (DL) framework is evaluated on a dataset of 118 whole slide images. The results show that the proposed DL-based technique achieves an accuracy of 87%, a precision of 94%, a sensitivity of 91%, a specificity of 71% and a mean of sensitivity and specificity at 81% and outperformed three state-of-the-art deep learning approaches. This study demonstrates the feasibility of DL-based prediction of critical molecular features in cytological slides, which not only aid in accurate diagnosis but also provide useful information in guiding clinical decision-making in patients with thyroid cancer. With the accumulation of data and the continuous advancement of technology, the performance of DL systems is expected to be improved in the near future. Therefore, we expect that DL can provide a cost-effective and time-effective alternative tool for patients in the era of precision oncology.

Ovarian steroid cell tumor causing isosexual pseudoprecocious puberty in a young girl: an instructive case and literature review.

BACKGROUND: Steroid cell tumors (SCTs) are very rare sex cord-stromal tumors and account only for less than 0.1% of ovarian neoplasms. SCTs might comprise diverse steroid-secreting cells; hence, the characteristic clinical features were affected by their propensity to secrete a variety of hormones rather than mass effect resulting in compression symptoms and signs. To date, ovarian SCTs have seldom been reported in children, particularly very young children; and pseudoprecocious puberty (PPP) as its unique principal manifestation should be reiterated. CASE PRESENTATION: We reported a 1-year-8-month-old girl presenting with rapid bilateral breast and pubic hair development within a 2-month period. Undetectable levels of LH and FSH along with excessively high estradiol after stimulation with gonadotropin-releasing hormone (GnRH), as well as a heterogeneous mass inside left ovary shown in pelvic sonography indicate isosexual PPP. Her gonadal hormones returned remarkably to the prepubertal range the day after surgery, and histology of the ovary mass demonstrated SCTs containing abundant luteinized stromal cells. CONCLUSION: The case highlighted that SCTs causing isosexual PPP should be taken into consideration in any young children coexistent with rapidly progressive puberty given a remarkable secretion of sex hormones. This article also reviewed thoroughly relevant reported cases to enrich the clinical experience of SCTs in the pediatric group.

The MTNR1A mRNA is stabilized by the cytoplasmic hnRNPL in renal tubular cells.

The downregulation of melatonin receptor 1A (MTNR1A) is associated with a range of pathological conditions, including membranous nephropathy. Knowledge of the mechanism underlying MTNR1A expression has been limited to the transcriptional regulation level. Here, RNA interference screening in human kidney cells revealed that heterogeneous nuclear ribonucleoprotein L (hnRNPL) upregulated MTNR1A RNA post-transcriptionally. hnRNPL knockdown or overexpression led to increased or decreased levels of cyclic adenosine monophosphate-responsive element-binding protein phosphorylation, respectively. Molecular studies showed that cytoplasmic hnRNPL exerts a stabilizing effect on the MTNR1A transcript through CA-repeat elements in its coding region. Further studies revealed that the interaction between hnRNPL and MTNR1A serves to protect MNTR1A RNA degradation by the exosome component 10 protein. MTNR1A, but not hnRNPL, displays a diurnal rhythm in mouse kidneys. Enhanced levels of MTNR1A recorded at midnight correlated with robust binding activity between cytoplasmic hnRNPL and the MTNR1A transcript. Both hnRNPL and MTNR1A were decreased in the cytoplasm of tubular epithelial cells from experimental membranous nephropathy kidneys, supporting their clinical relevance. Collectively, our data identified cytoplasmic hnRNPL as a novel player in the upregulation of MTNR1A expression in renal tubular epithelial cells, and as a potential therapeutic target.

TET1 reprograms the epithelial ovarian cancer epigenome and reveals casein kinase 2α as a therapeutic target.

Ten-eleven translocation methylcytosine dioxygenase-1, TET1, takes part in active DNA demethylation. However, our understanding of DNA demethylation in cancer biology and its clinical significance remain limited. This study showed that TET1 expression correlated with poor survival in advanced-stage epithelial ovarian carcinoma (EOC), and with cell migration, anchorage-independent growth, cancer stemness, and tumorigenicity. In particular, TET1 was highly expressed in serous tubal intraepithelial carcinoma (STIC), a currently accepted type II EOC precursor, and inversely correlated with TP53 mutations. Moreover, TET1 could demethylate the epigenome and activate multiple oncogenic pathways, including an immunomodulation network having casein kinase II subunit alpha (CK2α) as a hub. Patients with TET1high CK2αhigh EOCs had the worst outcomes, and TET1-expressing EOCs were more sensitive to a CK2 inhibitor, both in vitro and in vivo. Our findings uncover the oncogenic and poor prognostic roles of TET1 in EOC and suggest an unexplored role of epigenetic reprogramming in early ovarian carcinogenesis. Moreover, the immunomodulator CK2α represents a promising new therapeutic target, warranting clinical trials of the tolerable CK2 inhibitor, CX4945, for precision medicine against EOC. Copyright © 2019 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

O-Glycosylation-mediated signaling circuit drives metastatic castration-resistant prostate cancer.

Disseminated castration-resistant prostate cancer (CRPC) is a common disease in men that is characterized by limited survival and resistance to androgen-deprivation therapy. The increase in human epidermal growth factor receptor 2 (HER2) signaling contributes to androgen receptor activity in a subset of patients with CRPC; however, enigmatically, HER2-targeted therapies have demonstrated a lack of efficacy in patients with CRPC. Aberrant glycosylation is a hallmark of cancer and involves key processes that support cancer progression. Using transcriptomic analysis of prostate cancer data sets, histopathologic examination of clinical specimens, and in vivo experiments of xenograft models, we reveal in this study a coordinated increase in glycan-binding protein, galectin-4, specific glycosyltransferases of core 1 synthase, glycoprotein- N-acetylgalactosamine 3-ß-galactosyltransferase 1 (C1GALT1) and ST3 beta-galactoside α-2,3-sialyltransferase 1 (ST3GAL1), and resulting mucin-type O-glycans during the progression of CRPC. Furthermore, galectin-4 engaged with C1GALT1-dependent O-glycans to promote castration resistance and metastasis by activating receptor tyrosine kinase signaling and cancer cell stemness properties mediated by SRY-box 9 (SOX9). This galectin-glycan interaction up-regulated the MYC-dependent expression of C1GALT1 and ST3GAL1, which altered cellular mucin-type O-glycosylation to allow for galectin-4 binding. In clinical prostate cancer, high-level expression of C1GALT1 and galectin-4 together predict poor overall survival compared with low-level expression of C1GALT1 and galectin-4. In summary, MYC regulates abnormal O-glycosylation, thus priming cells for binding to galectin-4 and downstream signaling, which promotes castration resistance and metastasis.-Tzeng, S.-F., Tsai, C.-H., Chao, T.-K., Chou, Y.-C., Yang, Y.-C., Tsai, M.-H., Cha, T.-L., Hsiao, P.-W. O-Glycosylation-mediated signaling circuit drives metastatic castration-resistant prostate cancer.

Epigenetic loss of heparan sulfate 3-O-sulfation sensitizes ovarian carcinoma to oncogenic signals and predicts prognosis.

Precision medicine requires markers for therapeutic guidance. The purpose of this study was to determine whether epithelial ovarian cancer (EOC) epigenetics can lead to the identification of biomarkers for precision medicine. Through integrative methylomics, we discovered and validated the epigenetic signature of NEFH and HS3ST2 as an independent prognostic factor for type II EOC in our dataset (n = 84), and two independent methylomics datasets (total n = 467). Integrated transcriptomics dataset (n = 1147) and tissue microarrays (n = 54) of HS3ST2 also related to high-methylation statuses and the EOC prognosis. Mechanistic explorations of HS3ST2 have assessed responses to oncogenic stimulations such as IL-6, EGF, and FGF2 in cancer cells. The combination of HS3ST2 and various oncogenic ligands also confers the worse outcome. 3-O-sulfation of heparan sulfate by HS3ST2 makes ovarian cancer cells intrinsically sensitive to oncogenic signals, which sheds new light on the application of HS3ST2 as a companion diagnostic for targeted therapy using kinase inhibitors or therapeutic antibodies.

Role of melatonin receptor 1A and pituitary homeobox-1 coexpression in protecting tubular epithelial cells in membranous nephropathy.

Membranous nephropathy (MN), a type of glomerular nephritis, is one of the most common causes of nephrotic syndrome in adults. Although it is known that melatonin plays a protective role in MN, the role of melatonin receptors in the pathophysiology of MN is unclear. Using an experimental MN model and clinical MN specimens, we studied melatonin receptor expression and found that melatonin receptor 1A (MTNR1A) expression was significantly downregulated in renal tubular epithelial cells. Molecular studies showed that the transcription factor pituitary homeobox-1 (PITX1) promoted MTNR1A expression via direct binding to its promoter. Treatment of a human tubular cell line with albumin to induce injury resulted in the stable reduction in MTNR1A and PITX1 expression. PITX1 levels were significantly downregulated in tubular epithelial cells from mice MN kidneys and MN renal specimens. Knockdown of MTNR1A, PITX1, or cyclic adenosine monophosphate-responsive element-binding protein (CREB) decreased E-cadherin (CDH1) expression, but upregulated Per2 and α-smooth muscle actin (αSMA) expression. Blockade of the MTNR1A receptor with luzindole in MN mice further impaired renal function; this was accompanied by CDH1 downregulation and Per2 and αSMA upregulation. Together, our results suggest that in injured tissue, decreased PITX1 expression at the MTNR1A promoter regions leads to decreased levels of MTNR1A in renal tubular epithelial cells, which increases the future risk of MN.

Paired boxed gene 1 expression: A single potential biomarker for differentiating endometrial lesions associated with favorable outcomes in patients with endometrial carcinoma.

AIM: Paired boxed gene 1 (PAX1) has previously been reported to be a methylation-silenced gene in cases of cervical and ovarian cancers. We evaluated the expression of PAX1 in normal endometrium, endometrial hyperplasia and endometrial carcinoma (EC), and investigated the prognostic value of PAX1 expression in patients with EC. METHODS: We conducted a hospital-based retrospective review of PAX1 distribution immunohistochemically in 201 samples of endometrium from biopsy or hysterectomy. PAX1 immunoreactivity was classified into low and high score groups based upon the extent and intensity of staining. RESULTS: There was intense intranuclear staining for PAX1 in premalignant endometrial lesions. A high PAX1 score was observed in a high percentage of samples of normal endometrium (93.3%), in endometrial hyperplasia without atypia (97.2%) and in endometrial atypical hyperplasia (87.5%), but this level was found in only one-third of the EC samples (30.1%). The PAX1 protein score was significantly higher in samples of premalignant endometrial lesions compared with those of EC (P < 0.001). Importantly, a higher PAX1 score in EC cases was correlated with good overall survival, with a hazard ratio of 0.22 for death (95% confidence interval, 0.05-0.96). CONCLUSIONS: PAX1 protein expression is a potential histopathology biomarker for the differential diagnosis of malignant and premalignant endometrial lesions. PAX1 is also a potential prognostic marker in cases of EC.

Interpositional arthroplasty with palmaris longus tendon graft for osteonecrosis of the second metatarsal head: a case report.

Osteonecrosis of the second metatarsal head is often attributed to Freiberg's disease. We describe the case of a 27-year-old Taiwanese male soldier with persistent painful disability of the right forefoot of 9 months' duration, but no history of trauma. A series of radiographs suggested the diagnosis of late-stage Freiberg's disease. The lesion was treated with interpositional arthroplasty using a palmaris longus tendon graft, in a modification of the traditional interpositional arthroplastic technique for treating Freiberg's disease. After 2 years of follow-up examinations, the patient was satisfied with the clinical outcome, despite having a limited range of motion of the right second metatarsophalangeal joint relative to the adjacent toes. The patient returned to his army group with functional activity that was better than he had experienced before surgery. We believe this modified interpositional arthroplastic treatment strategy will provide more symptom relief and satisfactory functionality for the treatment of late-stage Freiberg's disease.

DNA methylation as a biomarker for the detection of hidden carcinoma in endometrial atypical hyperplasia.

OBJECTIVE: Women with atypical hyperplasia (AH) are often found to have endometrial carcinoma (EC) at hysterectomy. The purpose of this study was to evaluate whether the hypermethylation of specific genes found by methylomic approaches to the study of gynecologic cancers is a biomarker for EC in women with AH. METHODS: We evaluated the methylation of AJAP1, HS3ST2, SOX1, and PTGDR from 61 AH patients undergoing hysterectomy. Endometrial biopsy samples were analyzed by bisulfite conversion and quantitative methylation-specific polymerase chain reaction. A methylation index was used to predict the presence of cancer. To confirm the silencing effects of DNA methylation, immunohistochemical analysis of AJAP1, HS3ST2, and SOX1 was performed using tissue microarray. RESULTS: Fourteen (23%) patients had EC at hysterectomy. AJAP1, HS3ST2, and SOX1 were highly methylated in the EC patients' biopsy samples (p≤0.023). AJAP1, HS3ST2, and SOX1 protein expression was significantly higher in patients with AH only (p≤0.038). The predictive value of AJAP1, HS3ST2, and SOX1 methylation for EC was 0.81, 0.72, and 0.70, respectively. Combined testing of both AJAP1 and HS3ST2 methylation had a positive predictive value of 56%, methylation of any one of AJAP1, SOX1, or HS3ST2 had a 100% negative predictive value. CONCLUSIONS: Hypermethylation of AJAP1, HS3ST2, and SOX1 is predictive of EC in AH patients. Testing for methylation of these genes in endometrial biopsy samples may be a hysterectomy-sparing diagnostic tool. Validation of these new genes as biomarkers for AH screening in a larger population-based study is warranted.

R1933X mutation in the MYH9 gene in May-Hegglin anomaly mimicking idiopathic thrombocytopenic purpura.

May-Hegglin anomaly (MHA) is a rare autosomal dominant disorder characterized by the triad of thrombocytopenia, giant platelets, and inclusion bodies in leukocytes. Recent evidence links MHA to mutations in the MYH9 gene. MHA has not been reported in Taiwan before. We report a 25-year-old Taiwanese man who presented with prolonged bleeding after dental extraction. Examination of peripheral blood smear revealed thrombocytopenia (platelet = 35,000/µL), giant platelets, and Döhle-like cytoplasmic inclusions in neutrophils. A strong family history of thrombocytopenia favored hereditary macrothrombocytopenia over idiopathic thrombocytopenic purpura (ITP). Electron microscopy revealed a spindle shape and parallel order of filaments in the inclusions, consistent with the diagnosis of MHA. We performed mutational analysis using polymerase chain reaction followed by direct sequence of the MYH9 gene for the patient, his maternal uncle and cousin, and all showed the same heterozygous R1933X mutation in exon 40. MHA should be considered when a young patient has thrombocytopenia, frequently misdiagnosed as ITP. Morphological examination of peripheral blood smear, family history tracing and genetic studies are required to make an accurate diagnosis and avoid unnecessary and even harmful therapies such as corticosteroids and splenectomy.

Chronic infective arthritis with osteomyelitis of the ankle due to Pseudomonas aeruginosa infection in a middle-age woman: A rare causative pathogen requiring vigilance.

RATIONALE: Pseudomonas aeruginosa-induced septic arthritis is a relatively uncommon phenomenon. It has been documented in children with traumatic wounds, young adults with a history of intravenous drug use, and elderly patients with recent urinary tract infections or surgical procedures. PATIENT CONCERNS: Fifty-nine year-old female had no reported risk factors. The patient sought medical attention due to a 6-month history of persistent pain and swelling in her right ankle. DIAGNOSES: Magnetic resonance imaging and a 3-phase bone scan revealed findings suggestive of infectious arthritis with concurrent osteomyelitis. Histopathological examination of the synovium suggested chronic synovitis, and synovial tissue culture confirmed the presence of P aeruginosa. INTERVENTION: Arthroscopic synovectomy and debridement, followed by 6 weeks of targeted antibiotic therapy for P aeruginosa. OUTCOMES: Following treatment, the patient experienced successful recovery with no symptom recurrence, although she retained a mild limitation in the range of motion of her ankle. LESSONS: To our knowledge, this is the first reported case of chronic arthritis and osteomyelitis caused by P aeruginosa in a patient without conventional risk factors. This serves as a crucial reminder for clinicians to consider rare causative organisms in patients with chronic arthritis. Targeted therapy is imperative for preventing further irreversible bone damage and long-term morbidity.

Automated Quantification of HER2 Amplification Levels Using Deep Learning.

HER2 assessment is necessary for patient selection in anti-HER2 targeted treatment. However, manual assessment of HER2 amplification is time-costly, labor-intensive, highly subjective and error-prone. Challenges in HER2 analysis in fluorescence in situ hybridization (FISH) and dual in situ hybridization (DISH) images include unclear and blurry cell boundaries, large variations in cell shapes and signals, overlapping and clustered cells and sparse label issues with manual annotations only on cells with high confidences, producing subjective assessment scores according to the individual choices on cell selection. To address the above-mentioned issues, we have developed a soft-sampling cascade deep learning model and a signal detection model in quantifying CEN17 and HER2 of cells to assist assessment of HER2 amplification status for patient selection of HER2 targeting therapy to breast cancer. In evaluation with two different kinds of clinical datasets, including a FISH data set and a DISH data set, the proposed method achieves high accuracy, recall and F1-score for both datasets in instance segmentation of HER2 related cells that must contain both CEN17 and HER2 signals. Moreover, the proposed method is demonstrated to significantly outperform seven state of the art recently published deep learning methods, including contour proposal network (CPN), soft label-based FCN (SL-FCN), modified fully convolutional network (M-FCN), bilayer convolutional network (BCNet), SOLOv2, Cascade R-CNN and DeepLabv3+ with three different backbones (p ≤ 0.01). Clinically, anti-HER2 therapy can also be applied to gastric cancer patients. We applied the developed model to assist in HER2 DISH amplification assessment for gastric cancer patients, and it also showed promising predictive results (accuracy 97.67 ±1.46%, precision 96.15 ±5.82%, respectively).

Deep learning to assess microsatellite instability directly from histopathological whole slide images in endometrial cancer.

Molecular classification, particularly microsatellite instability-high (MSI-H), has gained attention for immunotherapy in endometrial cancer (EC). MSI-H is associated with DNA mismatch repair defects and is a crucial treatment predictor. The NCCN guidelines recommend pembrolizumab and nivolumab for advanced or recurrent MSI-H/mismatch repair deficient (dMMR) EC. However, evaluating MSI in all cases is impractical due to time and cost constraints. To overcome this challenge, we present an effective and efficient deep learning-based model designed to accurately and rapidly assess MSI status of EC using H&E-stained whole slide images. Our framework was evaluated on a comprehensive dataset of gigapixel histopathology images of 529 patients from the Cancer Genome Atlas (TCGA). The experimental results have shown that the proposed method achieved excellent performances in assessing MSI status, obtaining remarkably high results with 96%, 94%, 93% and 100% for endometrioid carcinoma G1G2, respectively, and 87%, 84%, 81% and 94% for endometrioid carcinoma G3, in terms of F-measure, accuracy, precision and sensitivity, respectively. Furthermore, the proposed deep learning framework outperforms four state-of-the-art benchmarked methods by a significant margin (p < 0.001) in terms of accuracy, precision, sensitivity and F-measure, respectively. Additionally, a run time analysis demonstrates that the proposed method achieves excellent quantitative results with high efficiency in AI inference time (1.03 seconds per slide), making the proposed framework viable for practical clinical usage. These results highlight the efficacy and efficiency of the proposed model to assess MSI status of EC directly from histopathological slides.

Ensemble transformer-based multiple instance learning to predict pathological subtypes and tumor mutational burden from histopathological whole slide images of endometrial and colorectal cancer.

In endometrial cancer (EC) and colorectal cancer (CRC), in addition to microsatellite instability, tumor mutational burden (TMB) has gradually gained attention as a genomic biomarker that can be used clinically to determine which patients may benefit from immune checkpoint inhibitors. High TMB is characterized by a large number of mutated genes, which encode aberrant tumor neoantigens, and implies a better response to immunotherapy. Hence, a part of EC and CRC patients associated with high TMB may have higher chances to receive immunotherapy. TMB measurement was mainly evaluated by whole-exome sequencing or next-generation sequencing, which was costly and difficult to be widely applied in all clinical cases. Therefore, an effective, efficient, low-cost and easily accessible tool is urgently needed to distinguish the TMB status of EC and CRC patients. In this study, we present a deep learning framework, namely Ensemble Transformer-based Multiple Instance Learning with Self-Supervised Learning Vision Transformer feature encoder (ETMIL-SSLViT), to predict pathological subtype and TMB status directly from the H&E stained whole slide images (WSIs) in EC and CRC patients, which is helpful for both pathological classification and cancer treatment planning. Our framework was evaluated on two different cancer cohorts, including an EC cohort with 918 histopathology WSIs from 529 patients and a CRC cohort with 1495 WSIs from 594 patients from The Cancer Genome Atlas. The experimental results show that the proposed methods achieved excellent performance and outperforming seven state-of-the-art (SOTA) methods in cancer subtype classification and TMB prediction on both cancer datasets. Fisher's exact test further validated that the associations between the predictions of the proposed models and the actual cancer subtype or TMB status are both extremely strong (p<0.001). These promising findings show the potential of our proposed methods to guide personalized treatment decisions by accurately predicting the EC and CRC subtype and the TMB status for effective immunotherapy planning for EC and CRC patients.

ATEC23 Challenge: Automated prediction of treatment effectiveness in ovarian cancer using histopathological images.

Ovarian cancer, predominantly epithelial ovarian cancer (EOC), is a global health concern due to its high mortality rate. Despite the progress made during the last two decades in the surgery and chemotherapy of ovarian cancer, more than 70% of advanced patients are with recurrent cancer and disease. Bevacizumab is a humanized monoclonal antibody, which blocks VEGF signaling in cancer, inhibits angiogenesis and causes tumor shrinkage, and has been recently approved by the FDA as a monotherapy for advanced ovarian cancer in combination with chemotherapy. Unfortunately, Bevacizumab may also induce harmful adverse effects, such as hypertension, bleeding, arterial thromboembolism, poor wound healing and gastrointestinal perforation. Given the expensive cost and unwanted toxicities, there is an urgent need for predictive methods to identify who could benefit from bevacizumab. Of the 18 (approved) requests from 5 countries, 6 teams using 284 whole section WSIs for training to develop fully automated systems submitted their predictions on a test set of 180 tissue core images, with the corresponding ground truth labels kept private. This paper summarizes the 5 qualified methods successfully submitted to the international challenge of automated prediction of treatment effectiveness in ovarian cancer using the histopathologic images (ATEC23) held at the 26th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) in 2023 and evaluates the methods in comparison with 5 state of the art deep learning approaches. This study further assesses the effectiveness of the presented prediction models as indicators for patient selection utilizing both Cox proportional hazards analysis and Kaplan-Meier survival analysis. A robust and cost-effective deep learning pipeline for digital histopathology tasks has become a necessity within the context of the medical community. This challenge highlights the limitations of current MIL methods, particularly within the context of prognosis-based classification tasks, and the importance of DCNNs like inception that has nonlinear convolutional modules at various resolutions to facilitate processing the data in multiple resolutions, which is a key feature required for pathology related prediction tasks. This further suggests the use of feature reuse at various scales to improve models for future research directions. In particular, this paper releases the labels of the testing set and provides applications for future research directions in precision oncology to predict ovarian cancer treatment effectiveness and facilitate patient selection via histopathological images.

ALDH1-bright epithelial ovarian cancer cells are associated with CD44 expression, drug resistance, and poor clinical outcome.

The role of aldehyde dehydrogenase 1 (ALDH1) as an ovarian cancer stem cell marker and its clinical significance have rarely been explored. We used an Aldefluor assay to isolate ALDH1-bright (ALDH1(br)) cells from epithelial ovarian cancer cell lines and characterized the properties of the stem cells. ALDH1(br) cells were enriched in ES-2 (1.3%), TOV-21G (1.0%), and CP70 (1.2%) cells. Both ALDH1(br) and ALDH1(low) cells repopulated stem cell heterogeneity, formed spheroids, and grew into tumors in immunocompromised mice, although these processes were more efficient in ALDH1(br) cells. In the ES-2 and CP70 cells, ALDH1(br) cells conferred more chemoresistance, and were more enriched in CD44 (by 1.74-fold and 5.18-fold, respectively) than in CD133 (by 1.39-fold and 1.17-fold, respectively), compared with ALDH1(low) cells. Immunohistochemical staining for ALDH1 on a tissue microarray containing 84 epithelial ovarian cancer samples revealed that patients with higher ALDH1 expression (>50%) had poor overall survival, compared with those with lower ALDH1 (P = 0.004) and yielded an odds ratio of death of 2.43 (95% CI = 1.12 to 5.28) by multivariate analysis. The results did not support ALDH1 alone as an ovarian cancer stem cell marker, but demonstrated that ALDH1 is associated with CD44 expression, chemoresistance, and poor clinical outcome. The use of a combination of ALDH1 with other stem cell markers may help define ovarian cancer stem cells more stringently.

LIM-homeobox transcription factor 1, alpha (LMX1A) inhibits tumourigenesis, epithelial-mesenchymal transition and stem-like properties of epithelial ovarian cancer.

OBJECTIVE: We reported recently the hypermethylation of LMX1A, a LIM-homeobox gene, as a prognostic biomarker in ovarian cancer; however, the function of LMX1A in ovarian cancer remains unknown. The present study aimed to evaluate the hypothesized tumour-suppressor functions of LMX1A in ovarian cancer. METHODS: We analysed the function of LMX1A by examining cell lines, animal models and human ovarian cancer tissues. Overexpression of LMX1A in relation to chemotherapy was also analysed. RESULTS: The expression of LMX1A inhibited cell proliferation, migration, invasion and colony formation in vitro, as well as tumourigenicity in a xenotransplantation mouse model. LMX1A also sensitized ovarian cancer cell lines to chemotherapeutics, and affected epithelial-mesenchymal transition (EMT). The restoration of LMX1A down-regulated stem cell markers and inhibited tumour spheroid formation in SKOV3 cells. Univariate analysis of immunohistochemical staining of tissue arrays (n=83) revealed that low LMX1A expression was significantly associated with advanced stages (p=0.001), poor differentiation (p<0.001), early recurrence (p=0.023) and poor overall survival (p=0.042) in ovarian cancer. CONCLUSIONS: The present study demonstrated, for the first time, that LMX1A is a bona fide tumour suppressor of ovarian cancer. The prognostic values of LMX1A may provide a biomarker for personalized treatments of ovarian cancer patients. The mechanisms of LMX1A in EMT and stem-like properties in ovarian cancer warrant further investigation.

Efficient Convolution Network to Assist Breast Cancer Diagnosis and Target Therapy.

Breast cancer is the leading cause of cancer-related deaths among women worldwide, and early detection and treatment has been shown to significantly reduce fatality rates from severe illness. Moreover, determination of the human epidermal growth factor receptor-2 (HER2) gene amplification by Fluorescence in situ hybridization (FISH) and Dual in situ hybridization (DISH) is critical for the selection of appropriate breast cancer patients for HER2-targeted therapy. However, visual examination of microscopy is time-consuming, subjective and poorly reproducible due to high inter-observer variability among pathologists and cytopathologists. The lack of consistency in identifying carcinoma-like nuclei has led to divergences in the calculation of sensitivity and specificity. This manuscript introduces a highly efficient deep learning method with low computing cost. The experimental results demonstrate that the proposed framework achieves high precision and recall on three essential clinical applications, including breast cancer diagnosis and human epidermal receptor factor 2 (HER2) amplification detection on FISH and DISH slides for HER2 target therapy. Furthermore, the proposed method outperforms the majority of the benchmark methods in terms of IoU by a significant margin (p<0.001) on three essential clinical applications. Importantly, run time analysis shows that the proposed method obtains excellent segmentation results with notably reduced time for Artificial intelligence (AI) training (16.93%), AI inference (17.25%) and memory usage (18.52%), making the proposed framework feasible for practical clinical usage.