Frontiers in Oncology最新文献

Purpose: In the context of lung cancer screening, the scarcity of well-labeled medical images poses a significant challenge to implement supervised learning-based deep learning methods. While data augmentation is an effective technique for countering the difficulties caused by insufficient data, it has not been fully explored in the context of lung cancer screening. In this research study, we analyzed the state-of-the-art (SOTA) data augmentation techniques for lung cancer binary prediction.

Methods: To comprehensively evaluate the efficiency of data augmentation approaches, we considered the nested case control National Lung Screening Trial (NLST) cohort comprising of 253 individuals who had the commonly used CT scans without contrast. The CT scans were pre-processed into three-dimensional volumes based on the lung nodule annotations. Subsequently, we evaluated five basic (online) and two generative model-based offline data augmentation methods with ten state-of-the-art (SOTA) 3D deep learning-based lung cancer prediction models.

Results: Our results demonstrated that the performance improvement by data augmentation was highly dependent on approach used. The Cutmix method resulted in the highest average performance improvement across all three metrics: 1.07%, 3.29%, 1.19% for accuracy, F1 score and AUC, respectively. MobileNetV2 with a simple data augmentation approach achieved the best AUC of 0.8719 among all lung cancer predictors, demonstrating a 7.62% improvement compared to baseline. Furthermore, the MED-DDPM data augmentation approach was able to improve prediction performance by rebalancing the training set and adding moderately synthetic data.

Conclusions: The effectiveness of online and offline data augmentation methods were highly sensitive to the prediction model, highlighting the importance of carefully selecting the optimal data augmentation method. Our findings suggest that certain traditional methods can provide more stable and higher performance compared to SOTA online data augmentation approaches. Overall, these results offer meaningful insights for the development and clinical integration of data augmented deep learning tools for lung cancer screening.

Introduction: We report for the first time a case of a postpubertal patient presenting with a metastatic prepubertal-type testicular teratoma.

Case discussion: A 29-year-old male with a history of corrected unilateral cryptorchidism presented with progressive bilateral lower extremity edema. Imaging revealed an inferior vena cava thrombus associated with a complex mass. A left testicular ultrasound identified a solid lesion suggestive of a germ cell tumor, leading to a left radical orchiectomy, which revealed a mature pure teratoma with no evidence of germ cell neoplasia in situ (GCNIS). Excision of the retroperitoneal mass confirmed the presence of mature teratomatous elements without evidence of non-teratomatous germ cell tumor elements or cytological atypia. Fluorescence in situ hybridization (FISH) showed no evidence of gain of 12p, and next-generation sequencing showed no alterations in genes known to be associated with GCT.

Conclusion: This case illustrates that pure mature teratomas lacking chromosome 12p abnormalities, GCNIS, and other dysgenetic features, occurring in postpubertal males, cannot invariably be classified into the benign prepubertal-type teratoma category. Contrary to current paradigm, in rare cases these may represent tumors with metastatic potential.

Mast cell leukemia (MCL) is an extremely rare and aggressive hematologic malignancy, characterized by a poor prognosis and short survival. Currently, no standardized treatment guidelines have been established. This study presents the clinical data of a patient with primary MCL accompanied by C-findings, and analyzes the clinical features, diagnostic challenges, and therapeutic approaches for this disease through a comprehensive review of the relevant literature. Furthermore, the study discusses current perspectives on research developments in MCL. By increasing clinical awareness of MCL, this work aims to provide valuable insights and references for clinicians in the diagnosis and management of this challenging condition.

Background: Cancer chemotherapy is a treatment that systematically kills cancer cells but causes expected side effects, known as chemotherapy-induced toxicities. These toxicities are managed with supportive care medications. This study aimed to determine the prescription patterns of supportive care medications in children receiving chemotherapy at a major referral hospital in Tanzania.

Methodology: A hospital-based descriptive cross-sectional study was conducted at Bugando Medical Centre (BMC). The study analyzed 104 prescription slips of pediatric cancer patients receiving chemotherapy and qualitatively assessed national guidelines and disease-specific protocols used in guiding treatment. Data were cleaned in Microsoft Excel, analyzed using STATA version 15, and presented as frequencies, percentages, and narrative summaries.

Results: Ondansetron (84.6%) and pre-hydration normal saline (20.2%) were the most prescribed pre-chemotherapy supportive care medications. Similarly, oral ondansetron (80.8%) and post-hydration normal saline (22.1%) were the most prescribed post-chemotherapy medications. Few prescriptions included a combination of antiemetics, fluids, and proton pump inhibitors for regimens with multiple chemotherapeutic agents. National cancer treatment guidelines lacked detailed sections on supportive care medications, leaving prescribing decisions to clinicians, while Burkitt's lymphoma and nephroblastoma protocols offered more detailed guidance.

Conclusion: Antiemetics and hydration fluids dominated supportive care prescriptions. Significant gaps were identified in the inclusion of supportive care in national guidelines, with reliance on disease-specific protocols. These findings highlight the need for standardized, evidence-based supportive care guidelines tailored to resource-limited settings.

Anaplastic lymphoma kinase (ALK) inhibitors have shown efficacy in treating ALK-positive advanced non-small cell lung cancer (NSCLC) patients. However, the effectiveness of ensartinib neoadjuvant therapy remains ambiguous. Herein, we reported that preoperative systemic treatment with the ALK inhibitor ensartinib can be beneficial for treating initially inoperable tumors. In this study, we present a case of a 60-year-old female patient who was diagnosed with stage IIIA (cT2aN2aM0, ninth TNM stage) lower left lung adenocarcinoma harboring an EML4-ALK fusion. After three months of therapy, the neoadjuvant treatment with ensartinib provided a partial response, with significant tumor and lymph node shrinkage. Preoperative ensartinib neoadjuvant therapy for NSCLC is safe and effective. Nevertheless, clinical trials can be conducted in the future to validate our results. Moreover, we performed multiple immunofluorescence staining analyses on samples before and after neoadjuvant therapy, observed and compared the changes in the expression of relevant immune cells (CD8+ T cells, macrophages, PD-1, and PD-L1), and performed a simple analysis.

Objective: Prostate cancer is a major threat to global male health. This study uses bibliometric methods to analyze the dynamics and trends in prostate cancer screening research, with the aim of optimizing screening strategies and informing policy decisions.

Methods: Utilizing the Web of Science Core Collection database, this study retrieved prostate cancer screening-related literature published between 2014 and 2024, totaling 5,409 articles. Data processing and analysis were conducted using CiteSpace and the Bibliometrix R package, including citation network analysis, co-word analysis, cluster analysis, and trend analysis.

Results: The analysis revealed the following key findings: (1) Global literature on prostate cancer screening has grown annually, with the United States, Europe, and China leading research activity; (2) Research hotspots include the risks and benefits of prostate-specific antigen (PSA) testing, MRI-based screening technologies, and the use of molecular and genetic biomarkers; (3) Emerging technologies, such as machine learning and nanodiagnostic techniques, are enhancing diagnostic precision and reducing overdiagnosis; (4) Network analysis of collaborations reveals a shift toward transnational and interdisciplinary research, particularly in integrating biomedical and computer science to drive rapid advancements in screening technologies.

Conclusion: This study confirms the ongoing vibrancy and technological advancement in the global field of prostate cancer screening research, emphasizing the trend towards precision medicine. Future development of prostate cancer screening strategies should focus on risk-adapted screening and the application of novel biomarkers to optimize screening outcomes and reduce unnecessary medical interventions.

Purpose: This study aims to investigate the diagnostic accuracy of various deep learning methods on DCE-MRI, in order to provide a simple and accessible tool for predicting pathologic response of NAC in breast cancer patients.

Methods: In this study, we enrolled 313 breast cancer patients who had complete DCE-MRI data and underwent NAC followed by breast surgery. According to Miller-Payne criteria, the efficacy of NAC was categorized into two groups: the patients achieved grade 1-3 of Miller-Payne criteria were classified as the non-responders, while patients achieved grade 4-5 of Miller-Payne criteria were classified as responders. Multiple deep learning frameworks, including ViT, VGG16, ShuffleNet_v2, ResNet18, MobileNet_v2, MnasNet-0.5, GoogleNet, DenseNet121, and AlexNet, were used for transfer learning of the classification model. The deep learning features were obtained from the final fully connected layer of the deep learning models, with 256 features extracted based on DCE-MRI data for each patient of each deep learning model. Various machine-learning techniques, including support vector machine (SVM), K-nearest neighbor (KNN), RandomForest, ExtraTrees, XGBoost, LightGBM, and multiple-layer perceptron (MLP), were employed to construct classification models.

Results: We utilized various deep learning models to extract features and subsequently constructed machine learning models. Based on the performance of different machine learning models' AUC values, we selected the classifiers with the best performance. ResNet18 exhibited superior performance, with an AUC of 0.87 (95% CI: 0.82 - 0.91) and 0.87 (95% CI: 0.78 - 0.96) in the train and test cohorts, respectively.

Conclusions: Using pre-treatment DCE-MRI images, our study trained multiple deep models and developed the best-performing DLR model for predicting pathologic response of NAC in breast cancer patients. This prognostic tool provides a dependable and impartial basis for effectively identifying breast cancer patients who are most likely to benefit from NAC before its initiation. At the same time, it can also identify those patients who are insensitive to NAC, allowing them to proceed directly to surgical treatment and prevent the risk of losing the opportunity for surgery due to disease progression after NAC.

Background: Penile metastasis is an uncommon condition, with most primary malignancies originating in the abdominal cavity and pelvis. There have been very few reported cases originating from lung cancer, most of squamous cell carcinoma without small cell lung cancer.

Methods: We presented a case of penile metastasis secondary to small cell lung cancer, along with a review of relevant literature from the CNKI database.

Results: A 73-year-old male presented with a one-month history of palpable swelling in the penis without any chest symptoms. Beside penile lesion, PET/CT imaging also revealed a lesion in the left lobe of the lung, as well as multiple enlarged lymph nodes in the left hilum, mediastinum, and left supraclavicular fossa. Fiberoptic biopsy confirmed small cell lung cancer for the pulmonary mass, while biopsies of the penile mass confirmed metastatic small cell carcinoma. The patient received first-line treatment of 6 cycles of PD-1 inhibitor (Toripalimab) combined with etoposide and cisplatin, achieving a partial response (PR). Subsequently, second-line therapy of etoposide and cisplatin regimen and later-line therapies of Irinotecan followed by Anlotinib were administered. The overall survival was approximately 2 years.

Conclusion: Penile metastasis from small cell lung cancer is extremely rare. Treatment strategies based on guidelines for small cell lung cancer had been proven effective approaches.

Colorectal cancer (CRC) is a prevalent malignant tumor in the digestive system. As reported in the 2020 global cancer statistics, CRC accounted for more than 1.9 million new cases and 935,000 deaths, making it the third most common cancer worldwide in terms of incidence and the second leading cause of cancer-related deaths globally. This poses a significant threat to global public health. Early screening methods, such as fecal occult blood tests, colonoscopies, and imaging techniques, are crucial for detecting early lesions and enabling timely intervention before cancer becomes invasive. Early detection greatly enhances treatment possibilities, such as surgery, radiation therapy, and chemotherapy, with surgery being the main approach for treating early-stage CRC. In this context, artificial intelligence (AI) has shown immense potential in revolutionizing CRC management, serving as one of the most effective screening tools. AI, utilizing machine learning (ML) and deep learning (DL) algorithms, improves early detection, diagnosis, and treatment by processing large volumes of medical data, uncovering hidden patterns, and forecasting disease development. DL, a more advanced form of ML, simulates the brain's processing power, enhancing the accuracy of tumor detection, differentiation, and prognosis predictions. These innovations offer the potential to revolutionize cancer care by boosting diagnostic accuracy, refining treatment approaches, and ultimately enhancing patient outcomes.

Ectopic adrenocorticotropic hormone secretion (EAS) is responsible for approximately 10%-18% of Cushing's syndrome cases. Thymic neuroendocrine tumors (NETs) comprise 5%-16% of EAS; therefore, they are very rare and the data about this particular tumors is scarce. We present a case of a 34-year-old woman with a rapid onset of severe hypercortisolism in April 2016. After initial treatment with a steroid inhibitor (ketoconazole) and diagnostics including ⁶⁸Ga DOTA-TATE PET/CT, it was shown to be caused by a small thymic NET. After a successful surgery and the resolution of all symptoms, there was a recurrence after 5 years of observation caused by a metastasis to the breast, shown in the ⁶⁸Ga DOTA-TATE PET/CT result and confirmed with a breast biopsy. Treatment with a steroid inhibitor (metyrapone) and tumor resection were again curative. The last disease relapse appeared 7 years after the initial treatment, with severe hypercortisolism treated with osilodrostat. There was a local recurrence in the mediastinum, and a thoracoscopic surgery was performed with good clinical and biochemical effect. The patient remains under careful follow-up. Our case stays in accordance with recent literature data, showing that patients with thymic NETs are younger than previously considered and that the severity of hypercortisolism does not correlate with the tumor size. The symptoms of EAS associated with thymic NET may develop rapidly and may be severe as in our case. Nuclear medicine improves the effectiveness of the tumor search, which is crucial in successful EAS therapy. Our case also underlines the need for lifelong monitoring of patients with thymic NETs and EAS.