Cancer Investigation最新文献

As use of real-world evidence (RWE) in oncology continues to increase, guidance is needed to ensure the patient voice is captured when generating RWE. This paper proposes a practical methodological framework for patient-centered RWE (PCRWE) throughout oncology product development. The need for a novel framework was first established by a review of existing literature and RWE guidelines. This review indicated an unmet need for a clear definition of PCRWE and a framework to guide PCRWE research in oncology. We define PCRWE as RWE that incorporates patient-centered objectives, provides insights into patient-relevant questions, and may lead to assessments of the usage, benefits, or risks of a medical treatment reflecting the patient perspective. The review's findings were used to create a preliminary PCRWE framework, which was finalized following interviews with RWE stakeholders and oncologists. The final PCRWE framework, which is grounded in the existing regulatory and scientific landscape, is an interactive visual tool for generating, implementing, and disseminating PCRWE in oncology. It accommodates various levels of expertise among users and supports the alignment of terminology to describe PCRWE. The framework will enable stakeholders to identify unmet needs from the patient perspective and to more effectively demonstrate the value of new oncology products.

Introduction: Metastatic melanoma carries a poor prognosis. Immune checkpoint inhibitors (ICIs) have improved outcomes, but responses remain variable, highlighting the need for simple prognostic biomarkers. Inflammatory markers such as neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and lactate dehydrogenase (LDH) reflect tumour burden and inflammation, though their clinical utility is unstandardised.

Methods: We retrospectively analysed 103 metastatic melanoma patients treated with anti-PD-1 monotherapy at two centres in Western Australia (2014-2020). Baseline NLR, PLR, and LDH were assessed within 30 days pre-treatment. Outcomes included clinical benefit, progression-free survival (PFS), and overall survival (OS).

Results: Poor ECOG performance status (PS ≥2) (RR 2.39, 95% CI 1.48-3.86) and elevated LDH (≥250 U/L) (RR 1.68, 95% CI 1.21-2.31) were associated with no clinical benefit (p < 0.001). NLR ≥5 predicted significantly worse OS (9.1 vs 28.2 months; HR 8.54, 95% CI 2.58-28.32; p < 0.001). Elevated LDH predicted shorter OS (6.0 vs 50.3 months; HR 3.68, 95% CI 1.65-8.21; p = 0.002) and PFS (19.0 months vs not reached; HR 2.51, 95% CI 1.37-4.72; p = 0.004).

Conclusion: ECOG PS ≥2 and elevated NLR were associated with no clinical benefit, while elevated NLR and LDH independently predicted poorer survival. These markers may serve as practical prognostic tools in metastatic melanoma treated with ICIs.

Background: Parathyroid carcinoma, a rare endocrine malignancy, is a significant diagnostic and therapeutic challenge due to its overlapping features with benign parathyroid diseases and high recurrence rates.

Objectives: The study assessed the demographic, biochemical, surgical, and histopathological characteristics of parathyroid carcinoma patients at a high-volume endocrine surgery center and identified clinical predictors of postoperative outcomes.

Methods: A retrospective cohort study was conducted on 8 patients who underwent surgery for histopathologically confirmed PC between January 2018 and December 2023. Data on demographics, biochemical markers, imaging, operative approach, and postoperative outcomes were analyzed. Spearman's rank correlation and multivariate linear regression were applied to identify key predictors of postoperative parathyroid hormone levels.

Results: The cohort comprised 4 females and 4 males with a mean age of 57.8 ± 7.1 years (range: 50-71 years). The mean preoperative serum PTH and calcium levels were: 688.96 ± 196.17 pg/mL and 12.90 ± 1.38 mg/dL, respectively. Distant metastasis was observed in 25% of cases, and lymph node involvement in 12.5%. Multivariate analysis revealed that male sex, preoperative calcium, intraoperative PTH, presence of comorbidities, and adjuvant therapies significantly influenced postoperative PTH levels (p < 0.05). Imaging was universally performed but lacked specificity for malignancy.

Conclusion: Parathyroid carcinoma presents a diagnostic challenge due to its similarity to benign disease. Preoperative evaluation, comprehensive histopathology, and en bloc surgical resection are crucial for curative treatment.

The primary objective of this review is to provide a comprehensive analysis of salivary biomarkers in the context of oral cancer, with a particular focus on oral squamous cell carcinoma. Oral cancer is a serious global health concern, ranking as the sixth most common cancer worldwide with over 300,000 new cases annually, and as the third most prevalent cancer in India. Its high morbidity and mortality are largely attributed to late-stage diagnosis and limited access to timely care. The current diagnostic gold standard, tissue biopsy, is invasive, costly, and unsuitable for population-level screening, creating a need for alternative approaches. This review critically evaluates recent advancements in diagnostic methodologies, emphasizing saliva as a noninvasive diagnostic medium. It examines relevant clinical case studies to demonstrate the diagnostic efficacy of salivary biomarkers and explores key etiological factors associated with oral cancer. Public health strategies initiated by governmental agencies to improve early detection, screening, and awareness are also discussed. The findings highlight that salivary biomarkers hold significant promise for early detection and cancer diagnostics. Conclusions emphasize the translational gaps that persist in this area, underscoring the need for further research to enable their integration into diagnostic protocols, screening programs, and public health initiatives.

Cervical cancer is the fourth most frequent cancer in females, with a high mortality rate globally. Persistent infection with high-risk, oncogenic human papillomavirus (HPV) types is a critical etiologic factor in the progression of the disease. Unfortunately, cervical cancer often remains undiagnosed until advanced stages, hence limiting treatment effectiveness. Therefore, identifying precise and significant biomarkers is crucial. High-throughput sequencing technologies have revolutionized targeted cancer therapy research by generating extensive data for analysis. This study employed bioinformatics and machine learning (ML) approaches to identify dysregulated genes with significant diagnostic value in cervical cancer, utilizing transcriptomics datasets. Seven potential diagnostic biomarker genes (APOD, SPARCL1, AR, MCM2, NUSAP1, PLK1, and STIL) were validated by a real-time polymerase chain reaction (RT-PCR) experiment. The ML models were developed using significantly differentially expressed genes (DEGs) involved in important pathways for cervical cancer. ML prediction models are available at https://github.com/PGlab-NIPER/CC_Pred.

Mammary Paget's disease (MPD) is a rare breast malignancy often associated with ductal carcinoma in situ or invasive carcinoma. However, its diagnosis remains challenging owing to the subtlety or absence of findings on conventional imaging. In this study, we retrospectively analyzed 12 Japanese patients with MPD. All patients showed uniform overexpression of the human epidermal growth factor receptor 2 (HER2; immunohistochemistry score = 3+), with 92% exhibiting associated ductal carcinoma in situ. Magnetic resonance imaging (MRI) revealed skin and nipple enhancement in 78% of patients, along with non-mass enhancement and nipple thickening that correlated with the pathological findings. Moreover, Ki-67 proliferation index was high in most cases (median, 67%), indicating the presence of biologically active tumors. No recurrence or death was observed during the median follow-up period of 96 months. Overall, our findings suggest that HER2-positive MPD exhibits aggressive biological behaviors despite a subtle clinical presentation and highlight the importance of MRI in its detection. Furthermore, integration of imaging with pathological and molecular marker assessment is essential for accurate MPD diagnosis and treatment. This study on a Japanese cohort provides valuable insights and highlights the diagnostic utility of MRI for MPD, especially HER2-driven MPD.

Lung cancer (LC) is one of the major causes of death worldwide. Early diagnosis helps to improve the patient survival outcome. The surgeon makes use of Computed Tomography (CT) for detecting LC using the aid of a Computer-Aided Diagnosis (CAD) system to identify LC effectively, but it has issues related to processing time and diagnostic precision that continue to pose significant challenges. To address this, a Deep Residual Xception Network (DRX-Net) approach has been introduced for identifying the LC. Initially, the CT image is obtained and then denoising is performed using a Wiener filter. Subsequently, the segmentation of lung nodule is conducted using Pyramidal Attention-based Y Net (PAY-Net), which uses a hybrid loss function combining Binary Cross Entropy, Tanimoto Similarity, and Dice Loss. The segmented image undergoes data augmentation followed by feature extraction. For LC detection, the selected features are processed using DRX-Net, which merges the Xception with a Deep Residual Network (DRN). Furthermore, the results show that the proposed DRX-Net achieved an accuracy of 93.988%, a True Positive Rate (TPR) of 95.567%, and a True Negative Rate (TNR) of 91.432% when evaluated using a K Group of 8.

Breast cancer primarily affects women, caused due to the excess growth of malignant breast tissues. The segmentation and early detection process suffered due to the complex and varied nature of breast tissue. To address this challenge, this research proposes a Convolutional Neural Network model with Contextual Relationship Embedding to accurately segment pathological mass regions in mammogram images. In this research work, the mammogram images are collected from datasets and are preprocessed to enhance image quality, noise reduction and contrast enhancement. By using a Deep Convolutional Neural Network, the edges in the highly contrasted regions, complex structure and spatial relationships of the images are gathered by using different operators. The extracted features are concatenated through the Fully Connected-Convolutional Block Attention Module. The contextual relationship embedded features are integrated with the original features, guided by the cross-entropy loss function with contextual relationship constraints. This enables the model to generate more precise decisions for segmentation and boundary identification. The proposed method's efficiency is validated and the proposed model achieves superior performance with an accuracy of 99.59% and an error rate of 0.405%. Overall, this research article concludes that the proposed model is more efficient for breast cancer detection than other existing models.

Colorectal cancer is a leading cause of morbidity and mortality worldwide. This study investigates the association between GLP-1 receptor agonists (GLP-1 RAs) and five-year mortality in patients with primary colon cancer, considering BMI. Using data from the University of California Health Data Warehouse, 6,871 patients were analyzed. Five-year mortality was 15.5% for GLP-1 RA users compared to 37.1% for non-users. Analyses showed significantly lower odds of five-year mortality with GLP-1 RA use (OR = 0.38, 95% CI: 0.21-0.64). This benefit persisted after adjusting for confounders, including disease severity, but was found to only extend to high obese patients (BMI > 35) in stratified modeling.

Deep learning (DL) has transformed medical imaging, particularly in the realm of Oral Cancer (OC) diagnosis using histopathological images. Timely detection of OC is essential for enhancing precision medicine and saving lives. However, incorrect diagnosis may impede effective treatment. In this study, we have proposed a DL model for OC classification, enhanced diagnosis decision-making, and interpretability. We achieve this by starting with color normalization of histopathology images using the Vahadane Three-Stain Parameter Normalization and watershed segmentation method, followed by tiling and augmentation. Key features are selected using the Weighted Fisher Score (WFS) to address class imbalance. The U-Net classifier has been improved by using feature-based inputs instead of full images, reducing computational complexity and training time. The integration of Vahadane normalization for consistent preprocessing across samples, WFS, and Explainable Artificial Intelligence (XAI) addresses critical challenges in histopathological image analysis. The proposed model surpasses existing approaches with a classification accuracy of 99.54% and outperforms DenseNet201 and VGG10 in precision and reliability. The efficiency in handling imbalanced datasets and explainability features make it suitable for early precise OC detection, which can reduce diagnostic errors and enhance treatment outcomes.​.