Understanding the tumor immune microenvironment, especially tumor-infiltrating lymphocytes (TILs), remains crucial in ovarian cancer. However, the distribution and prognostic significance of TILs across histological subtypes and genetic backgrounds remain unclear. As part of the JGOG3025-A1 study, diagnostic slides from 400 cases were collected. Two artificial intelligence-based cell classification models were used to evaluate the spatial distribution of TILs. The TIL score was calculated as the number of TILs divided by the analyzed area, and the immune-inflamed group was defined based on TIL scores. Among histological subtypes, high-grade serous carcinoma (HGSC) exhibited the highest TIL scores, clear cell carcinoma the lowest, and endometrioid carcinoma intermediate values. In HGSC, TIL scores did not significantly differ according to BRCA alteration or homologous recombination deficiency (HRD) status. The HRD/immune-inflamed group had the most favorable prognosis. In the homologous recombination-proficient population, the immune-inflamed group had better progression-free survival, whereas this trend did not appear for overall survival. In analyzes of the relationships between TIL levels and genomic structure, whole-genome doubling was associated with lower TIL infiltration in HRD tumors but showed no association in HRP tumors. In conclusion, HGSC showed the highest amount of TIL infiltration among subtypes, and prognostic stratification can be achieved by integrating pathology-based immunophenotypes and HRD status.

Transcriptomic classification methods have been proposed for ovarian clear cell carcinoma. However, their clinical significance and association with pathologically evaluated tumor-infiltrating lymphocytes (TILs) remain unclear. We established two large transcriptomic datasets and analyzed RNA-sequencing data from 189 (JGOG3025-TR1 cohort) and 38 (Kyoto cohort) ovarian clear cell carcinomas. Representative histopathological slides were also digitized (102 and 38, respectively). Cell types were classified by two state-of-the-art artificial-intelligence models, and TILs were quantified. The transcriptomically defined immune subtype was associated with significantly poor prognosis (hazard ratio, 2.54; 95

Background:The integration of traditional and complementary medicine (T&CM) into modern medical education remains a global challenge. Kampo medicine, a Japanese traditional pharmacotherapy, is recognized in the International Classification of Diseases, 11th Revision, and is widely used; however, no structured methodology exists for efficiently designing curricula within limited time and resources. To address this gap, this study proposes a methodological framework-illustrated through Kampo medicine but generalizable to other forms of T&CM-for organizing and visualizing curricular content to guide educational needs assessment and curriculum design. Objective:The objective of this study was to develop and illustrate a reproducible mapping framework that integrates (1) real-world clinical utilization frequency and (2) the accumulation of biomedical evidence to inform educational prioritization and stepwise curriculum design for T&CM, using Kampo medicine as an exemplar. Methods:A mapping approach was developed based on two perspectives: frequency of use in clinical training and level of biomedical evidence. Twelve years of outpatient prescription data from Kyoto University Hospital were analyzed to identify the most frequently prescribed Kampo formulas. For the 10 most common formulas, PubMed searches were conducted to determine the number of randomized controlled trials. Data were integrated using hierarchical clustering and plotted along frequency-evidence axes to produce an educational priority map, which informed a stepwise curriculum design grounded in adult learning theory. Results:Prescription heat maps revealed substantial interdepartmental variation, and clustering identified distinct groups of formulas based on usage patterns. No statistically significant correlation was observed between prescription frequency and level of evidence (Pearson r=0.228, 95% CI -0.469 to 0.750; P=.53; Spearman ρ=0.498, 95% CI -0.308 to 0.926; P=.14). Integrating these two perspectives enabled interpretation of real-world prescription patterns and supported a transparent educational prioritization framework. Conclusions:This study presents a feasible and adaptable framework that links real-world clinical data with biomedical evidence to inform curriculum design in T&CM. Rather than prescribing specific content, the framework offers visual decision-making tools that align educational priorities with institutional practice patterns and can be readily adapted to complementary, alternative, and integrative medicine programs internationally.

Immune checkpoint inhibitors show limited efficacy against immune-desert tumors, including ovarian cancer. We investigated triple therapy combining anti-programmed cell death-ligand 1 (PD-L1) antibody, anti-vascular endothelial growth factor (VEGF) antibody, and Poly ADP-ribose polymerase inhibitor (PARPi) on tumor microenvironment using spatial profiling. Two mouse models were employed: MC38 (immune-inflamed phenotype) and HM-1 (immune-desert phenotype). MC38 mice received anti-PD-L1 and anti-VEGF as monotherapy or dual combination. HM-1 mice received anti-PD-L1, anti-VEGF, and PARPi as monotherapy, dual combinations (anti-PD-L1 + anti-VEGF, anti-PD-L1 + PARPi, anti-VEGF + PARPi), or triple combination (anti-PD-L1 + anti-VEGF + PARPi). Spatial distribution of immune cells and the tumor microenvironment was analyzed using immunohistochemistry (CD8) and dual immunofluorescence (CD8/Granzyme B) with distance-based density quantification from tumor margins (0 to − 150, − 150 to − 300, − 300 to – 450 μm). High endothelial venule (HEV) formation was evaluated via CD31/MECA79 dual immunofluorescence. MC38 tumors responded to all treatments by day 10. Conversely, HM-1 tumors showed no response at day 10 but responded to two combination therapies by day 20: anti-PD-L1 + anti-VEGF (1.5-fold reduction, p = 0.04) and triple combination therapy (1.7-fold reduction, p = 0.03). In MC38, at − 150 to − 300 μm, anti-PD-L1 + anti-VEGF enhanced CD8 + Granzyme B + cells 1.9-fold versus Control (p = 0.01). In HM-1, at 0 to − 150 μm, triple therapy enhanced CD8 + Granzyme B + cells 2.8-fold (p = 0.02), while anti-PD-L1 + anti-VEGF increased CD8 + Granzyme B + cells 2.5-fold (p = 0.03). Both triple and anti-PD-L1 + anti-VEGF therapies induced CD31 + MECA79 + HEV formation (p < 0.01). Triple therapy may overcome immune-desert ovarian cancer through additive HEV formation, enhancing cytotoxic CD8 + T cell infiltration into the tumor.

In this study, we aimed to investigate the association between heart rate variability (HRV) and quality of life (QoL) in patients with gynecological cancer and examine whether continuous HRV monitoring may help identify QoL deterioration. This multicenter, prospective, observational study screened 215 patients with gynecological cancer undergoing chemotherapy at five designated cancer hospitals. Participants measured HRV with a smartphone application for 1 min daily and completed patient-reported outcomes every 1–2 weeks for 10 weeks. The primary outcome was overall QoL (Global Health Status [GHS]), and the discriminative performance of HRV parameters for lower GHS was evaluated using the area under the receiver operating characteristic curve. Changes in HRV were evaluated at the time of a 10-point decrease in GHS or a 7-point reduction in the FACT-G total score over 7 days. We analyzed 2880 data points from 85 patients. The root mean square of successive differences (rMSSD), a parasympathetic HRV parameter, and the low frequency/high frequency ratio, an index of autonomic balance, demonstrated moderate to high discriminative performance in classifying GHS level. In exploratory within-individual analyses, the rMSSD significantly declined from the preceding week to the week of GHS deterioration and FACT-G total score deterioration. Although these findings may not be generalizable to patients with limited digital literacy, smartphone-based HRV was associated with clinically meaningful QoL deterioration, suggesting that low-burden HRV monitoring may complement patient-reported QoL assessment in patients with cancer.

BackgroundPremenstrual disorder (PMD), which includes premenstrual syndrome and premenstrual dysphoric disorder, has a complex pathogenesis and may be closely related to emotional cognition and memory. However, the mechanisms underlying these associations remain unclear. Therefore, this study used machine learning to explore the roles of various factors that are not typically considered risk-factors for PMD.MethodsA predictive model for PMD was constructed using a dataset of questionnaire responses and heartrate variability data collected from 60 participants during their follicular and luteal phases. Based on the Japanese version of the Premenstrual Symptom Screening Tool, the binary objective variable (PMD status) was defined as "PMD" for moderate-to-severe premenstrual syndrome and premenstrual dysphoric disorder and other conditions as "non-PMD." The contribution of each feature to the predictive model was assessed using the Shapley Additive exPlanations (SHAP) model-interpretation framework.ResultsOf the 58 participants (providing 117 data points), 17 (34 data points) were in the PMD group and 41 (83 data points) were in the non-PMD group. The area under the receiver operating characteristic curve was 0.90 (95% confidence interval: 0.82-0.98). Among the top 20 features with the highest SHAP values, six were associated with maternal bonding. Four of the six mother-related characteristics were associated with overprotection.ConclusionsBased on these findings, parental bonding experiences, including maternal overprotection, may be associated with the presence of PMD.

Background: The prognosis for recurrent ovarian cancer is poor, but a small percentage of patients can be cured. The aim of this study was to clarify the criteria for being cured and the characteristics of cured cases. Methods: Ovarian cancer cases at 2 university hospitals and 8 community hospitals were analyzed to identify patients who were considered cured after complete remission (CR) following recurrence. Analyses of the tumors were performed and included BRCA1/2 mutation analysis. Results: Of the 157 cases of recurrence, 21 (13%) showed no evidence of disease (NED). NED cases had a lower rate of ascites at the initial diagnosis, longer disease-free survival, a higher rate of solitary lesions, and a higher rate of secondary debulking surgery. All CR cases except for one showed no further recurrence when DFS reached 4 years, which was considered a criterion for being cured. The case of relapse occurred after long-term treatment with bevacizumab. Furthermore, 19.4% of the CR cases achieved 4-year DFS, which represents 9.3% of the cases of recurrent ovarian cancer and 2.3% of all cases of ovarian cancer. BRCA mutation analysis of the tumor was possible in 17 of the 30 cases of recurrent ovarian cancer that achieved a 4-year DFS. Pathogenic variants of BRCA were found in 5 of the 11 cases of high-grade serous carcinoma. Conclusions: Approximately 10% of patients with recurrent ovarian cancer achieved a 4-year DFS and were mostly cured. The curing of cases not involving high-grade serous carcinoma (HGSC) was unrelated to the presence of pathogenic BRCA variants.

AIM:Kampo medicines, which are covered under national health insurance, are widely accessible and affordable in Japan; however, knowledge about their current prescription practices remains limited. This study investigated the outpatient prescription practices of Kampo formulations at Kyoto University Hospital from April 2011 to March 2023. RESULTS:Kampo medicine was prescribed in all medical departments. A total of 42,453 prescriptions were recorded during the study period, with an average of 3,538 prescriptions per year. The gynecology and obstetrics department had the highest number of Kampo prescriptions, followed by anesthesiology. The anesthesiology department prescribed various formulations for 12 years. The usage rate of Processed Aconite Root-containing formulations increased with increasing patient age, reaching 3.3% in individuals in their teens, 6.6% in their twenties, 13.3% in their thirties, and over 30% in their eighties and nineties. CONCLUSION:The substantial use of Kampo medicines across various departments indicates their critical role in specialized and general medical practice. The adaptability and personalized approach of treatments using Kampo medicines in the anesthesiology department are evident from the dynamic prescription patterns. Notably, the increasing use of Processed Aconite Root-containing formulations with advancing patient age suggests a potential age-related preference or therapeutic indication. These findings underscore the integral role of Kampo medicine in modern clinical practice and emphasize the necessity of further research into its age-specific applications and departmental prescribing trends.

Abstract With the incorporation of immune checkpoint inhibitors into the treatment of endometrial cancer (EC), a deeper understanding of the tumor immune microenvironment is critical. Tertiary lymphoid structures (TLSs) are considered favorable prognostic factors for EC, but the significance of their spatial distribution remains unclear. B cell receptor repertoire analysis performed using six TLS samples located at various distances from the tumor showed that TLSs in distal areas had more shared B cell clones with tumor-infiltrating lymphocytes. To comprehensively investigate the distribution of TLSs, we developed an artificial intelligence model to detect TLSs and determine their spatial locations in whole-slide images. Our model effectively quantified TLSs, and TLSs were detected in 69% of the patients with EC. We identified them as proximal or distal to the tumor margin and demonstrated that patients with distal TLSs (dTLSs) had significantly prolonged overall survival and progression-free survival (PFS) across multiple cohorts [hazard ratio (HR), 0.56; 95% confidence interval (CI), 0.36–0.88; p = 0.01 for overall survival; HR, 0.58; 95% CI, 0.40–0.84; p = 0.004 for PFS]. When analyzed by molecular subtype, patients with dTLSs in the copy-number-high EC subtype had significantly longer PFS (HR, 0.51; 95% CI, 0.29–0.91; p = 0.02). Moreover, patients with dTLSs had a higher response rate to immune checkpoint inhibitors (87.5 vs. 41.7%) and a trend toward improved PFS. Our findings indicate that the functions and prognostic implications of TLSs may vary with their locations, and dTLSs may serve as prognostic factors and predictors of treatment efficacy. This may facilitate personalized therapy for patients with EC.

Accurate embryo assessment on embryonic day 3 of assisted reproductive technology (ART) is crucial for deciding whether to continue the culture until day 5 (blastocyst stage) or opt for earlier transfer or cryopreservation. Prolonged culture often improves pregnancy outcomes in patients with multiple high-quality embryos, but may offer limited benefits for older patients or those with few available embryos. In Japan, where donor eggs are rarely used, cleavage-stage vitrification remains common in poor-prognosis cases, making early embryo assessment clinically relevant. To address this clinical challenge, analyzing embryo quality in early stages by artificial intelligence (AI) can be useful. We retrospectively analyzed images of 7111 two-pronuclear embryos (Veeck grade ≤3) using four different time-lapse incubators. We fine-tuned ImageNet-1k-pretrained NASNet-A Large to automatically classify each time-lapse image into 17 morphological categories, including cell stages and Veeck grades 1-3. This model achieved 95 % cell-stage accuracy on the test set. We combined these annotations with age at egg retrieval in a gradient boosting framework (XGBoost) to predict blastocyst formation, good blastocysts, and poor blastocyst + arrested embryos (PBAE). The ROC AUCs were 0.87, 0.88, and 0.87 for blastocyst formation, good blastocysts, and PBAE, respectively, indicating good predictive performance for day 3 embryo assessment. Notably, the PBAE model reached a precision-recall AUC of 0.90, accurately identifying embryos unlikely to benefit from extended culture. This novel AI prediction model could ensure transparency and addresses the "black box" limitation often associated with AI. By integrating a high-accuracy auto-annotation pipeline with interpretable AI (via SHapley Additive exPlanations), our device-independent approach supports appropriate embryo-specific decisions, potentially reducing unnecessary culture, optimizing workflows, and improving clinical outcomes in ART.

Abstract Background Recent studies have identified premenstrual disorders (PMDs) as a risk factor for postpartum depression. However, routine screening for preconception PMDs is not yet common in Japan. This study investigated the association between preconception PMDs and perinatal depression in a single tertiary care setting. Methods We analyzed data from pregnant women who gave birth at Kyoto University Hospital between April 2020 and October 2023. The Premenstrual Symptoms Screening Tool was administered at the first postconception visit to retrospectively assess PMD status before the current pregnancy. The Edinburgh Postnatal Depression Scale (EPDS) was administered during pregnancy and one month postpartum as a prospective measure of perinatal depression. EPDS cutoff values were set at 12/13 during pregnancy and 8/9 at one month postpartum. Results Of the 781 women analyzed, 53 had preconception PMD. Univariate and multivariate logistic regression analyses revealed that preconception PMD was associated with an EPDS score of ≥ 13 during pregnancy, with a crude odds ratio (OR) of 5.78 (95% confidence interval [CI]: 2.70–11.75) and an adjusted OR of 3.71 (95% CI: 1.54–8.35). For an EPDS score of ≥ 9 at 1 month postpartum, the crude OR was 3.36 (95% CI: 1.79–6.12) and the adjusted OR was 2.16 (95% CI: 1.04–4.35). Conclusions Our findings indicate that preconception PMDs are a significant risk factor for both depression during pregnancy and postpartum depression. These results support the implementation of preconception PMD screening during antenatal checkups as a preventive measure and to identify women in need of early mental health care.

Background Acute kidney injury (AKI) is a critical complication of immune checkpoint inhibitor therapy. Since the etiology of AKI in patients undergoing cancer therapy varies, clarifying underlying causes in individual cases is critical for optimal cancer treatment. Although it is essential to individually analyze immune checkpoint inhibitor-treated patients for underlying pathologies for each AKI episode, these analyses have not been realized. Herein, we aimed to individually clarify the underlying causes of AKI in immune checkpoint inhibitor-treated patients using a new clustering approach with Shapley Additive exPlanations (SHAP). Methods We developed a gradient-boosting decision tree-based machine learning model continuously predicting AKI within 7 days, using the medical records of 616 immune checkpoint inhibitor-treated patients. The temporal changes in individual predictive reasoning in AKI prediction models represented the key features contributing to each AKI prediction and clustered AKI patients based on the features with high predictive contribution quantified in time series by SHAP. We searched for common clinical backgrounds of AKI patients in each cluster, compared with annotation by three nephrologists. Results One hundred and twelve patients (18.2%) had at least one AKI episode. They were clustered per the key feature, and their SHAP value patterns, and the nephrologists assessed the clusters’ clinical relevance. Receiver operating characteristic analysis revealed that the area under the curve was 0.880. Patients with AKI were categorized into four clusters with significant prognostic differences (p = 0.010). The leading causes of AKI for each cluster, such as hypovolemia, drug-related, and cancer cachexia, were all clinically interpretable, which conventional approaches cannot obtain. Conclusion Our results suggest that the clustering method of individual predictive reasoning in machine learning models can be applied to infer clinically critical factors for developing each episode of AKI among patients with multiple AKI risk factors, such as immune checkpoint inhibitor-treated patients.

161 Background: Although depression in patients with cancer can lead to a deterioration of treatment adherence and quality of life (QOL), diagnosis of depression in patients with cancer is difficult due to symptoms overlapping with side effects of cancer treatment such as fatigue. Traditional screening tools like the PHQ-9 questionnaire are underutilized in clinical practice. Recently, interest has grown in utilizing patients' voices to assess their mental well-being. This study aims to evaluate depression in patients with gynecological cancer and explore the usefulness of voice analysis as a depression detection tool. Methods: PHQ-9 scores were collected from 197 patients with gynecological cancer, with 28 cases undergoing longitudinal assessments at diagnosis, postoperatively, and during chemotherapy. Voice recordings were obtained from 70 patients at diagnosis, alongside serum samples from 22 patients. The PHQ-9 scores were compared across different treatment periods, with a specific focus on patients exhibiting a PHQ score of ≧10, who can be diagnosed with Major Depressive Disorder (MDD). The provision of supportive care for MDD patients was also verified. A random forest model was developed from voice features obtained from 70 patients, with hyperparameter tuning and cross-validation to predict mild depression. Serum metabolites were comprehensively analyzed between depression prediction group and normal prediction group by the depression prediction model. Results: Mean PHQ-9 scores decreased over time since diagnosis; initially 6.50 ± 4.99 (mean ±SD), post-operatively 6.18 ± 4.25, and during chemotherapy 4.88 ± 4.29. The frequencies of MDD also decreased over the clinical course; at diagnosis 24.3%, post-operatively 15.5%, and during chemotherapy 15.4%. Out of 17 cases of MDD during chemotherapy, only one (6.25%) received psychiatric intervention. All MDD cases during chemotherapy had PHQ-9 scores of 5 or higher, representing mild depression at diagnosis. Mild depression at diagnosis could be predicted with an AUC of 0.89 using voice features. Metabolites identified by the voice-based depression prediction model were associated with depression, such as xanthine, methionine, and taurocholic acid. Conclusions: Depression during chemotherapy is often not intervened. Patients with mild depression at diagnosis tend to develop MDD during chemotherapy. Voice-based mental health assessment at diagnosis is possibly a promising screening tool for depression during subsequent cancer treatment. Clinical trial information: UMIN000044266 .

Background and purpose Glioblastoma is a highly aggressive brain tumor with limited survival that poses challenges in predicting patient outcomes. The Karnofsky Performance Status (KPS) score is a valuable tool for assessing patient functionality and contributes to the stratification of patients with poor prognoses. This study aimed to develop a 6-month postoperative KPS prediction model by combining clinical data with deep learning-based image features from pre- and postoperative MRI scans, offering enhanced personalized care for glioblastoma patients. Materials and methods Using 1,476 MRI datasets from the Brain Tumor Segmentation Challenge 2020 public database, we pretrained two variational autoencoders (VAEs). Imaging features from the latent spaces of the VAEs were used for KPS prediction. Neural network-based KPS prediction models were developed to predict scores below 70 at 6 months postoperatively. In this retrospective single-center analysis, we incorporated clinical parameters and pre- and postoperative MRI images from 150 newly diagnosed IDH wild-type glioblastoma, divided into training (100 patients) and test (50 patients) sets. In training set, the performance of these models was evaluated using the area under the curve (AUC), calculated through fivefold cross-validation repeated 10 times. The final evaluation of the developed models assessed in the test set. Results Among the 150 patients, 61 had 6-month postoperative KPS scores below 70 and 89 scored 70 or higher. We developed three models: a clinical-based model, an MRI-based model, and a multimodal model that incorporated both clinical parameters and MRI features. In the training set, the mean AUC was 0.785±0.051 for the multimodal model, which was significantly higher than the AUCs of the clinical-based model (0.716±0.059, P = 0.038) using only clinical parameters and the MRI-based model (0.651±0.028, P<0.001) using only MRI features. In the test set, the multimodal model achieved an AUC of 0.810, outperforming the clinical-based (0.670) and MRI-based (0.650) models. Conclusion The integration of MRI features extracted from VAEs with clinical parameters in the multimodal model substantially enhanced KPS prediction performance. This approach has the potential to improve prognostic prediction, paving the way for more personalized and effective treatments for patients with glioblastoma.

Tumor-infiltrating lymphocytes (TILs) are associated with improved survival in patients with epithelial ovarian cancer. However, TIL evaluation has not been used in routine clinical practice because of reproducibility issues. The current study developed two convolutional neural network models to detect TILs and to determine their spatial location in whole slide images, and established a spatial assessment pipeline to objectively quantify intraepithelial and stromal TILs in patients with high-grade serous ovarian carcinoma. The predictions of the established models showed a significant positive correlation with the number of CD8+ T cells and immune gene expressions. Patients with a higher density of intraepithelial TILs had a significantly prolonged overall survival and progression-free survival in multiple cohorts. On the basis of the density of intraepithelial and stromal TILs, patients were classified into three immunophenotypes: immune inflamed, excluded, and desert. The immune-desert subgroup showed the worst prognosis. Gene expression analysis showed that the immune-desert subgroup had lower immune cytolytic activity and T-cell-inflamed gene-expression profile scores, whereas the immune-excluded subgroup had higher expression of interferon-g and programmed death 1 receptor signaling pathway. The established evaluation method provided detailed and comprehensive quantification of intraepithelial and stromal TILs throughout hematoxylin and eosin-stained slides. It has potential for clinical application for personalized treatment of patients with ovarian cancer.

BACKGROUND:The purpose of this study was to reconstruct 3-dimensional (3D) computed tomography (CT) images from single anteroposterior (AP) postoperative total hip arthroplasty (THA) X-ray images using a deep learning algorithm known as generative adversarial networks (GANs) and to validate the accuracy of cup angle measurement on GAN-generated CT. METHODS:We used 2 GAN-based models, CycleGAN and X2CT-GAN, to generate 3D CT images from X-ray images of 386 patients who underwent primary THAs using a cementless cup. The training dataset consisted of 522 CT images and 2,282 X-ray images. The image quality was validated using the peak signal-to-noise ratio and the structural similarity index measure. The cup anteversion and inclination measurements on the GAN-generated CT images were compared with the actual CT measurements. Statistical analyses of absolute measurement errors were performed using Mann-Whitney U tests and nonlinear regression analyses. RESULTS:The study successfully achieved 3D reconstruction from single AP postoperative THA X-ray images using GANs, exhibiting excellent peak signal-to-noise ratio (37.40) and structural similarity index measure (0.74). The median absolute difference in radiographic anteversion was 3.45° and the median absolute difference in radiographic inclination was 3.25°, respectively. Absolute measurement errors tended to be larger in cases with cup malposition than in those with optimal cup orientation. CONCLUSIONS:This study demonstrates the potential of GANs for 3D reconstruction from single AP postoperative THA X-ray images to evaluate cup orientation. Further investigation and refinement of this model are required to improve its performance.