Oncologic Diagnosis Research Articles

Leveraging Large Language Models to Generate Clinical Histories for Oncologic Imaging Requisitions.

Background Clinical information improves imaging interpretation, but physician-provided histories on requisitions for oncologic imaging often lack key details. Purpose To evaluate large language models (LLMs) for automatically generating clinical histories for oncologic imaging requisitions from clinical notes and compare them with original requisition histories. Materials and Methods In total, 207 patients with CT performed at a cancer center from January to November 2023 and with an electronic health record clinical note coinciding with ordering date were randomly selected. A multidisciplinary team informed selection of 10 parameters important for oncologic imaging history, including primary oncologic diagnosis, treatment history, and acute symptoms. Clinical notes were independently reviewed to establish the reference standard regarding presence of each parameter. After prompt engineering with seven patients, GPT-4 (version 0613; OpenAI) was prompted on April 9, 2024, to automatically generate structured clinical histories for the 200 remaining patients. Using the reference standard, LLM extraction performance was calculated (recall, precision, F1 score). LLM-generated and original requisition histories were compared for completeness (proportion including each parameter), and 10 radiologists performed pairwise comparison for quality, preference, and subjective likelihood of harm. Results For the 200 LLM-generated histories, GPT-4 performed well, extracting oncologic parameters from clinical notes (F1 = 0.983). Compared with original requisition histories, LLM-generated histories more frequently included parameters critical for radiologist interpretation, including primary oncologic diagnosis (99.5% vs 89% [199 and 178 of 200 histories, respectively]; P < .001), acute or worsening symptoms (15% vs 4% [29 and seven of 200]; P < .001), and relevant surgery (61% vs 12% [122 and 23 of 200]; P < .001). Radiologists preferred LLM-generated histories for imaging interpretation (89% vs 5%, 7% equal; P < .001), indicating they would enable more complete interpretation (86% vs 0%, 15% equal; P < .001) and have a lower likelihood of harm (3% vs 55%, 42% neither; P < .001). Conclusion An LLM enabled accurate automated clinical histories for oncologic imaging from clinical notes. Compared with original requisition histories, LLM-generated histories were more complete and were preferred by radiologists for imaging interpretation and perceived safety. © RSNA, 2025 Supplemental material is available for this article. See also the editorial by Tavakoli and Kim in this issue.

P-2009. COVID-19 in Children with Hematological/Oncological Diagnoses – Exploring Reasons for Better Clinical Outcomes

Abstract Background The majority of COVID-19 infections in children, including in those with underlying hematological and oncological (Heme-Onc) diagnoses are mild with good outcome compared to adults. Postulated reasons include cross-protection from immunity against seasonal coronaviruses (HCoV’s), subdued inflammatory responses in children than in adults and differences in humoral and T-cell responses. The objectives of this study are to compare antibody responses against HCoV’s, anti-SARS-CoV-2 humoral, T-cell, and inflammatory profiles between Heme-Onc children with COVID-19 and control groups. Methods Binding IgG antibodies against Spike (S) of SARS-CoV-2, four HCoV’s (HKU1, OC43, NL63 and 229E), quantitative levels of 21 cytokines and 9 chemokines (multiplexed solid-phase electrochemiluminescence assay by Meso Scale Diagnostics/MSD) were compared among five cohorts: 1) children with Heme-Onc diagnoses and COVID-19 (n = 41) 2) healthy children with COVID-19 (n = 21) 3) adults with COVID-19 (n = 14) 4) children with multisystem inflammatory syndrome (MIS-C) [n = 23] and 5) healthy children without COVID-19 (n = 12). Additionally, T-cell responses against S and nucleocapsid (N) proteins of SARS-CoV-2 was assessed using ELISpot in Heme-Onc children a group of healthy children COVID-19. Results Anti-S antibodies and neutralizing antibodies against SARS-CoV-2 and HCoV were significantly lower in children in Heme-Onc and healthy children with COVID-19 compared to adults with COVID-19 (Figure 1). The concentration of eight pro-inflammatory cytokines and chemokines were significantly lower in Heme-Onc and healthy children compared to adults with COVID-19 and children with MIS-C. The responder T-cell frequency varied between 0 and 860/100,000 PBMC at study enrollment in children with Heme-Onc diagnoses. Conclusion In this single-center study, cross-protection from HCoV antibodies is an unlikely mechanism to explain mild COVID-19 in children, including in those with underlying hematological and oncological diagnoses. Subdued inflammatory responses to SARS-CoV-2 could be one of the factors for the mild COVID-19 disease phenotype in children. Disclosures Swetha Pinninti, MD, Moderna: Grant/Research Support|Pfizer: Grant/Research Support Suresh Boppana, MD, GSK: Advisor/Consultant|Merck: Grant/Research Support|Pfizer: Grant/Research Support

P-2161. Evaluating the Clinical Utility of Next-Generation Sequencing Cell-Free DNA Testing in Pediatric Patients: A Focus on Medical History

Abstract Background The Karius test is a noninvasive, cell-free next-generation sequencing (NGS) test that can detect over 1,000 pathogens from a single blood sample. Validating the Karius test's useful clinical applications in pediatric patients is necessary to elucidate its benefits and limitations compared to conventional diagnostic testing. This pilot study aims to expand on existing evidence for Karius indications by investigating its utility among pediatric patients with specific underlying diagnoses.Figure 1:Age Distribution of Patients in Sample Methods A retrospective chart review was conducted, consisting of 251 consecutive pediatric patients [03/2018-06/2022] (fig 1) for whom a Karius NGS test was performed at a free-standing 240-bed children’s hospital and affiliated clinic in Central Texas. Statistical analysis was conducted to evaluate whether a child’s underlying diagnosis was associated with clinically useful Karius results. The Karius result was considered clinically useful if it was the first or only diagnostic test used to make the diagnosis, had negative predictive value essential to the plan of care, or contributed to changes in the antimicrobial regimen. Underlying diagnoses were categorized into ten groups (fig 2). Statistical analysis was conducted using R software via Pearson’s Chi-Squared Test and Fischer’s Exact Test.Figure 2:Distribution of Underlying Diagnosis in Sample Results Evaluation of underlying diagnoses and the clinical usefulness of the Karius test revealed a significant association only for patients with an oncologic diagnosis (p=0.038) (Fig 3).Figure 3:Underlying Diagnosis and Reason Karius Result Deemed Clinically Useful Conclusion While Karius testing can be useful in specific patients regardless of the underlying diagnosis, this analysis found the most significant utility among patients with an oncology diagnosis. Further investigation through larger sample sizes and prospective designs is essential to validate and extend these findings, shedding light on cost-effectiveness and patient outcomes. Disclosures All Authors: No reported disclosures

Effectiveness of Radiofrequency in Treating Pain Associated With Trigeminal Neuralgia in Oncologic Patients.

Introduction In cancer patients, trigeminal neuralgia (TN) has been not only observed in its secondary form but also described as one of the common neuropathic pain syndromes related to cancer. Patients with TN experience a marked reduction in quality of life due to the nature and severity of the pain. Although the standard treatment is microvascular decompression, this approach is not always feasible in oncologic patients. Therefore, radiofrequency (RF) guided by tomography has been described as an ablative intervention that allows achieving immediate pain relief in TN. It is of interest to identify the therapeutic response to pain in oncologic patients treated by this procedure. Objective The study aimed to evaluate the effectiveness of percutaneously guided tomography-assisted RF in oncologic patients with TN. Materials and methods We conducted an observational, retrospective, longitudinal, and analytical study. Patients with cancer and TN treated at the Pain Clinic of the National Cancer Institute in Mexico, who underwent RF, were included. A non-probabilistic sampling of consecutive cases was used, and selection criteria were applied. Clinical records were reviewed to measure study variables, including pain by TN before and after the procedure, as well as patient satisfaction. Data were analyzed using descriptive and inferential statistics, employing Student's t-test for related samples, with a significant level set at p ≤ 0.05 to determine statistical significance. The recorded data were analyzed using IBM SPSS Statistics for Windows, V. 26.0 (IBM Corp., Armonk, NY, USA). Results In the sample of 76 subjects (50% women and 50% men) with a mean age of 60.6 years, a variety of oncologic diagnoses were presented, with skin and soft tissue tumors being the most frequent, mainly located at the level of the head and neck. The etiology of TN was secondary in most cases (89.5%), with similar anatomical involvement on both hemifaces (50% left, 50% right). The branches most frequently involved were the maxillary nerve (V2) and mandibular nerve (V3) (39.5%) and V3 alone (22.4%). The most common probable cause was tumor activity (53, 69.7%). Most patients were receiving anticonvulsant treatment (carbamazepine-oxcarbazepine) (58.9%), followed by gabapentinoids (31.5%). Of the 76 patients, only six received computed tomography (CT)-guided RF. No complications associated with RF application were reported. The mean baseline pre-procedure pain (3.8 ± 2.48) and mean baseline post-procedure pain (2.17 ± 1.94) were significantly different in the Wilcoxon signed-rank test (p = 0.043), as were the mean incidental pre-procedure pain (9.0 ± 1.09) and mean incidental post-procedure pain (5.8 ± 2.31) (p = 0.019). Conclusions Our research provides relevant data on the effectiveness of RF in managing TN in oncologic patients, demonstrating a clinically and statistically significant reduction in pain intensity. These findings need to be confirmed through prospective, longitudinal, long-term studies with control groups in a larger number of patients.

Research Progress of CRISPR Technology for Accurate Diagnosis and Treatment in Oncology

Research Progress of CRISPR Technology for Accurate Diagnosis and Treatment in Oncology

Optimizing time-of-flight of PET/CT image quality via penalty β value in Bayesian penalized likelihood reconstruction algorithm.

Optimizing the image quality of Positron Emission Tomography/Computed Tomography (PET/CT) systems is crucial for effective monitoring, diagnosis, and treatment planning in oncology. This study evaluates the impact of time-of-flight (TOF) on PET/CT performance, focusing on varying penalty β values within Q. Clear reconstruction algorithm. The study measured signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) using the Discovery MI PET/CT scanner and NEMA IQ phantom filled with the radiotracer fluorodeoxyglucose (18F-FDG). PET/CT scans were performed with and without TOF using β values of 100, 500, 1000, 1500, 2000, and 3000. Pixel intensity values were measured using ImageJ software, and SNR and CNR were calculated. Results indicated that increasing β values improved SNR and CNR for both non-TOF and TOF images. At a β value of 100, SNR and CNR increased across all sphere sizes (10mm, 13mm, 17mm, 22mm, 28mm, 37mm) when comparing non-TOF and TOF images. However, β values of 500 or higher led to decreased SNR and CNR, particularly in larger spheres (22mm, 28mm, 37mm), when TOF was utilized. These findings underscore the importance of optimizing β values and employing TOF reconstruction in PET/CT scans to achieve the highest possible image quality. In clinical practice, practitioners should adjust β values in accordance with routine protocols, considering the size of the target region and the use of TOF reconstruction.

Algorithm for Constructing the Hazard Function of the Extended Cox Model and its Application to the Prostate Cancer Patient Database

Introduction. In medicine and related industries, bioinspired approaches are used for the survival analysis, among which the Cox regression model holds a specific place. The practice of its application is described in the theoretical and applied literature. However, a significant drawback of this method requires careful study. The fact is that the features correlate with the hazard function linearly, and the model does not use more complex dependences. This causes some difficulties in studying survival analysis. The presented work is aimed at solving this problem. The object of study is the extended Cox model, in which the hazard function includes a nonlinear combination of features.Materials and Methods. A database of prostate cancer patients was used, since this is a common diagnosis in global oncology. A class of extended Cox models with an additive/multiplicative hazard function was defined. To solve the problem using the optimization method, a fitness function was constructed that evaluated the results of prognosis, the number of features, and the degree of overtraining of the model — the complexity and load of the compiled hazard function. An algorithm of pollinating ants has been developed to optimize the fitness function. It simulates the reproduction of flowering plants using pollinating insects and consists of three parts: an ant colony algorithm, a genetic algorithm, and an ant pollinator algorithm. The quality of training of the Cox model was assessed by C-index.Results. A metaheuristic algorithm for ant pollinator optimizing was proposed, providing for the construction of hazard functions of the extended Cox model. The set of parameters for training the standard Cox model was the entire set of features used: TNM, prostate-specific antigen doubling time (PSADT), Gleason score, serum PSA concentration at diagnosis, patient age and education, Rh factor. C-index value of the trained model was 0.853691. The extended Cox model with the found additive/multiplicative hazard function had a higher C-index value — 0.856241 with a smaller number of features used (TNM, PSADT, and Gleason score). In terms of quality, this approach is not inferior to or superior to the classical Cox model. Reducing the number of features involved should improve the efficiency of medical decisions and speed up the start of treatment.Discussion and Conclusion. The presented algorithm for constructing survival analysis models increased the accuracy of predicting the occurrence of a terminal event, and reduced the number of features used for this purpose. The difference in accuracy for the studied data set seemed insignificant — C-index increased from 0.853691 to 0.856241 (by 0.3%). At this, the number of features taken into account was reduced from 7 to 3 (by 57.1%). Consequently, the proposed method effectively solves the problem of feature selection, and can be applied to improve the quality of prognostication.

Evaluation of Emotional Distress in Oncology Palliative Patients: Edmonton Symptom Assessment System (ESAS) and Palliative Outcome Scale (POS)-A Portuguese Cross-Sectional Study.

Cancer patients in palliative care frequently experience psychological distress, encompassing cognitive, behavioral, and emotional dimensions. This distress can significantly affect their capacity to accept the inevitability of death. Commonly, such distress manifests as sadness, depression, anxiety, and fear, which may culminate in an existential crisis. The purpose of this study was to analyze the psychometric properties of the Edmonton Symptom Assessment System (ESAS) and the Palliative Outcome Scale (POS) among hospitalized cancer patients receiving palliative care in a Portuguese palliative care unit. Data were collected from psychological screening instruments (ESAS and POS) used in oncology patients admitted to an oncology palliative service. The study period spanned from January 2021 to March 2021. The study included a convenience sample of 60 patients diagnosed with advanced cancer who were sequentially admitted to a palliative care unit. The most frequently reported primary oncological diagnoses were gastrointestinal (41.7%) and head/neck (20%) tumors. The majority of the patients were male (n = 37; 61.7%), with a median age of 72 years (range: 43-94 years). Key findings from the analysis indicated the following associations: anxiety and self-reported symptoms (higher levels of patient anxiety related to illness or treatment were strongly correlated with increased self-reported anxiety (p < 0.01)); emotional sharing and perceived meaning of life (patients who shared their feelings less frequently with family and friends were significantly less likely to find meaning in life (p < 0.01) and were more likely to report a poorer self-perception (p < 0.05)); understanding the importance of life and self-perception (a diminished understanding of the importance of life was associated with worsened self-perception (p < 0.01)); anxiety and depressive symptoms (greater anxiety related to the disease or its treatment correlated with an increased severity of depressive symptoms (p < 0.01)). There are numerous validated instruments available for measuring emotional distress. Among these, the ESAS and POS complement each other as practical and easy-to-use tools for evaluating emotional distress in patients admitted to palliative care settings. Findings revealed that anxiety and concerns related to the disease and its treatment, as measured by the POS, served as predictors of the risk for self-reported anxiety and depression identified by the ESAS. Furthermore, the items within the POS that evaluate the emotional component demonstrated strong internal correlation, supporting their combined use for a comprehensive assessment of this dimension.

Artificial intelligence’s current involvement in urology and future implementation in clinical environments

Artificial intelligence (AI) is a set of computational methods that interprets the data given, uncovers the underlying patterns associated with the complexity of data, and provides the prediction of outcomes that have become increasingly relevant in urology. The current application of AI in urology predominately focuses on disease diagnosis and risk factor analysis in urologic oncology and male infertility. While many candidate models have been proposed in the literature, efforts to construct clinically meaningful data by incorporating patient-specific and multidisciplinary approaches should be carried out to improve the clinical applicability of AI in driving personalised treatment planning and disease prognosis. Looking forward, AI has the potential to drive targeted training in urology, from surgical techniques to patient-specific surgical procedure simulation, in combination with other technologies such as augmented reality. In order to achieve this, patient involvement should be considered in the model development stage, which also addresses issues surrounding the ethical deployment of AI in the clinical environment. It is possible to see AI playing a collaborative role with surgeons in improving clinical efficiency in the future. Level of evidence: Not Applicable

A multimodal machine learning model for the stratification of breast cancer risk.

Machine learning models for the diagnosis of breast cancer can facilitate the prediction of cancer risk and subsequent patient management among other clinical tasks. For the models to impact clinical practice, they ought to follow standard workflows, help interpret mammography and ultrasound data, evaluate clinical contextual information, handle incomplete data and be validated in prospective settings. Here we report the development and testing of a multimodal model leveraging mammography and ultrasound modules for the stratification of breast cancer risk based on clinical metadata, mammography and trimodal ultrasound (19,360 images of 5,216 breasts) from 5,025 patients with surgically confirmed pathology across medical centres and scanner manufacturers. Compared with the performance of experienced radiologists, the model performed similarly at classifying tumours as benign or malignant and was superior at pathology-level differential diagnosis. With a prospectively collected dataset of 191 breasts from 187 patients, the overall accuracies of the multimodal model and of preliminary pathologist-level assessments of biopsied breast specimens were similar (90.1% vs 92.7%, respectively). Multimodal models may assist diagnosis in oncology.

The epidemiology of pediatric oncology and hematopoietic cell transplant admissions to U.S. intensive care units from 2001-2019.

Children with cancer or hematopoietic cell transplant (HCT) frequently require ICU care. We conducted a retrospective cohort study using Healthcare Cost and Utilization Project's State Inpatient Databases from 21 U.S. states from 2001-2019. We included children <18 years with oncologic or HCT diagnosis and used ICD-9-CM and ICD-10-CM codes to identify diagnoses, comorbidities, and organ failures. We used generalized linear Poisson regression and Cuzick's test of trend to evaluate changes from 2001-2019. Among 2,157,991 total pediatric inpatient admissions, 3.9% (n=82,988) were among oncology patients and 0.3% (n=7,381) were among HCT patients. ICU admission prevalence rose from 13.6% in 2001 to 14.4% in 2019 for oncology admissions and declined from 23.9% to 19.5%, for HCT admissions. Between 2001-2019, the prevalence of chronic non-oncologic comorbidities among ICU patients rose from 44.3% to 69.1% for oncology patients (RR 1.60 [95% CI 1.46-1.66]) and from 41.4% to 81.5% (RR 1.94 [95% CI 1.61-2.34]) for HCT patients. The risk of Multiple Organ Dysfunction Syndrome more than tripled for oncology (9.5% to 33.3%; RR 3.52 [95% CI 2.97-4.18]) and HCT (12.4% to 39.7%; RR 3.20 [95% CI 2.09-4.89]) patients. Mortality decreased most for ICU patients with acute myeloid leukemia (AML) (14.6% to 8.5%) and oncology-related HCTs (15.5% to 9.2%). Critically ill pediatric oncology and HCT patients are increasingly medically complex with greater prevalence of chronic comorbidities and organ failure, but mortality did not increase. Pediatric ICUs may require increased financial and staffing support to care for these patients in the future.

Long term survival and outcomes in patients with paraneoplastic neurologic syndromes.

It is unknown whether delay in diagnosis affects morbidity reportedly in paraneoplastic syndromes (PNS). We aimed to explore various aspects of PNS, including prevalence, clinical characteristics, diagnostic criteria, and treatment outcomes. We studied n-PNS diagnosis between 2016 to 2023, and included only patients with positive onconeural antibodies, who developed cancer, and exhibited a recognizable PNS phenotype. We identified 12 patients with positive Abs and co-occurring cancer, most prevalent PNS antibodies included anti-GAD65, anti-Recoverin and anti-Yo. The most common phenotypes were limbic encephalitis (n=5, 42%) and encephalomyelitis (n=4,33%). Cancer preceded neurological presentation in 6 cases. Among the 6 patients who initially presented with n-PNS, median time from neurological presentation to oncologic diagnosis was 73 days, as five of them (83%) were diagnosed with cancer during oncological evaluation prompted by the PNS diagnosis or suspicion. Lymphoma was the most frequent cancer (n=3, 25%), followed by lung cancer (n=2, 17%), and ovarian cancer (n=2, 17%). Among patients who received immunotherapy as n-PNS treatment (n=9, 75%), steroids were a part of the management at 78% (n=7). Another immunotherapy used included plasmapheresis (n=5, 55%) and steroid sparing immunosuppressant (n=2, 29%). Four (33%) patients had short term therapeutic benefit with improvement or stabilization at mRS ≤ 4. Median Disability-adjusted life years (DALYs), as disease burden value, was 13 years. Death occurred in 9 of the 12 patients, with most cases deaths attributed to cancer progression. Compering to the expected median survival by type and stage of tumor, from 9 deceased patients, 56% (n=5) died younger than expected. Median survival was 410 days (range 29-2738 days), and 152 days since the appearance of n-PNS (range 8-1434 days). There were no differences in survival between patients who initially presented with n-PNS versus cancer (p=0.39). In up to 8 years of follow up, there was no difference in mortality among patients who presented initially n-PNS. There was a significant decline in the quality of life, most face substantial disability and functional impairment long term.

SURG-11. COMPARATIVE RISK ANALYSIS OF DEVELOPMENT OF ANEMIA IN RADIOTHERAPY-TREATED VS. NON-TREATED MULTIPLE SCLEROSIS PATIENTS: A MULTICENTER RETROSPECTIVE COHORT STUDY

Abstract INTRODUCTION The use of RT in patients who are diagnosed with multiple sclerosis (MS) is controversial due to adverse effects such as inflammation. Simultaneously, the relationship between oncologic RT and worsening anemia is well established. However, there is not a clear understanding of the risk associated with undergoing RT due to their oncologic diagnosis in MS patients for developing anemia. OBJECTIVE This study aims to analyze the risk of developing anemia for patients with MS who have been treated with RT versus who have not. METHODS This retrospective study consists of analysis of the data acquired from TriNetX, a database that houses clinical data from 92 healthcare organizations. Cohort 1 included MS patients with history of RT, while cohort 2 included MS patients without any history of RT. The cohorts were propensity score matched for demographics such as age, sex, race, and ethnicity. All neoplasms were also matched to eliminate cancer as a confounding variable. The analysis included outcomes of interest, ICD-10 code D64 (Other anemia), up to 6 months after the first occurrence of the index event. RESULTS 4,844 patients were identified for each cohort post-propensity score matching. The mean age was 65.4 and 67.3 years old, and most patients were white at 76.3% and 77.0% for the MS with RT and without RT cohorts, respectively. Cohort 1 had 628 patients who developed anemia, while cohort 2 only had 511 patients. The odds ratio was 1.263 with 95% confidence interval of 1.116 to 1.430 showing the statistically significant difference between the cohorts. CONCLUSION This study demonstrates that there is a significant increase in the likelihood of developing broadly categorized anemia for multiple sclerosis patients with or without radiation therapy. Further study is warranted to understand other potential factors leading to an increased risk of developing anemia in the irradiated cohort.

DISP-10. RACIAL AND SOCIOECONOMIC DIFFERENCES IN CLINICAL AND ACCESS TO CARE OUTCOMES IN NEUROSURGICAL-ONCOLOGY PATIENTS

Abstract BACKGROUND Large database studies have identified racial and socioeconomic differences in neurosurgical-oncologic outcomes. For clearer insights and understanding, more detailed and prospective evaluation is needed. METHODS We retrospectively analyzed a database of 1475 patients undergoing surgery for CNS tumors over 5 years. Multivariate analysis, specific to various pathologies, was conducted to identify demographic and socioeconomic factors (including age, sex, racial identity, primary language, insurance status, and income estimate) associated with access to care and clinical outcomes. Additionally, prospective patient-centered interviews were conducted on neuro- oncology patients to qualitatively explore reasons for disparities in healthcare. RESULTS Significant associations were found between racial identities (White n=911; Hispanic n=285; Asian n=206; Black n=74) and sex (ANOVA p 0.0043), age (p&amp;lt;0.0001), income estimate (p&amp;lt;0.0001), primary language (p&amp;lt;0.0001), and insurance status (p&amp;lt;0.0001). Among different pathologies, presentation to the ED, rather than outpatient setting, was consistently associated with a non- English primary language (p=0.0312) and Medi-Cal insurance (p=0.0029). Time from imaging diagnosis to neurosurgical consultation was associated with non- English primary language (p=0.0262) and Medi-Cal (p=0.0030). Delay between recommendation for surgery and date of surgery was significant in Black patients (p=0.0112). A non- English primary language was negatively associated with having a gross total resection (p=0.0087). Notably, adjuvant treatments were less likely for those with non-English primary language (p=0.0203), Black race (p=0.0049), and Medi-Cal coverage (p=0.0014). Cox regression analysis identified only age (p&amp;lt;0.0001) as a predictor of poorer overall survival in oncologic cases. Prospective interviews identified cultural, economic (work- related), and healthcare navigation issues as reasons for differences in access to care. CONCLUSIONS Across the spectrum of neurosurgical oncology diagnoses, racial and socioeconomic factors independently associate with access to care but not clinical outcome variables. Further investigation is needed, particularly in patients with non- English primary languages and cultural attitudes towards chemotherapy and radiation therapy.

Endodontic Treatment in Patients Undergoing Hematopoietic Stem Cell Transplantation: A Retrospective Cohort Study

IntroductionThis study investigated the implications of root canal treatment (RCT) of infected teeth with apical periodontitis (AP) in patients undergoing hematopoietic stem cell transplantation (HSCT). MethodsOne hundred patients were selected. Clinical data, oncologic and endodontic diagnosis and treatment, and type of HSCT were recorded. Data on the exacerbation of AP were investigated. Leukocyte/neutrophil and C-reactive protein levels were evaluated before and after RCT. ResultsOne hundred patients underwent HSCT (55% autologous). The need for RCT was identified in 41% of the participants. AP was diagnosed in 75.6% of these patients and large lesions (≥5 mm) occurred in 45.2%. A higher prevalence of AP was observed in males (P = .012). RCT was performed in 54.8% of the individuals and no AP exacerbation, pain, swelling, bacteremia, and need for additional medications was observed. Blood cultures were negative for bacteria and fungi. C-reactive protein levels were ≥1 mg/dL in 54.9% of the individuals 1 month before RCT and <1 mg/dL in 64.6% 6 months after treatment (P > .05). ConclusionsRCT was effective in controlling infection with no complications in patients undergoing HSCT. Given the potential for exacerbation of endodontic infections, RCT and patient monitoring are essential and highly recommended.

Adaptive Transformer-Based Multi-Modal Image Fusion for Real-Time Medical Diagnosis and Object Detection

In recent years, medical diagnosis and object detection have been significantly enhanced by the integration of multi-modal image fusion techniques. This study proposes an Adaptive Transformer-Based Multi-Modal Image Fusion (AT-MMIF) framework designed for real-time medical diagnosis and object detection. The framework employs a Transformer architecture to capture both global and local feature correlations across multiple imaging modalities, including MRI, CT, PET, and X-ray, for more accurate diagnostic results and faster object detection in medical imagery. The fusion process incorporates spatial and frequency-domain information to improve the clarity and detail of the output images, enhancing diagnostic accuracy. The adaptive attention mechanism within the Transformer dynamically adjusts to the relevant features of different image types, optimizing fusion in real time. This leads to an improved sensitivity (98.5%) and specificity (96.7%) in medical diagnosis. Additionally, the model significantly reduces false positives and negatives, with an F1 score of 97.2% in object detection tasks. The AT-MMIF framework is further optimized for real-time processing with an average inference time of 120 ms per image and a model size reduction of 35% compared to existing multi-modal fusion models. By leveraging the strengths of Transformer architectures and adaptive learning, the proposed framework offers a highly efficient and scalable solution for real-time medical diagnosis and object detection in various clinical settings, including radiology, oncology, and pathology.

Do Fellowship-educated Military Orthopaedic Oncologists Who Practice in Military Settings Treat a Sufficient Volume of Patients to Maintain Their Oncologic Expertise?

Fellowship-trained orthopaedic oncologists in the US military provide routine clinical care and also must maintain readiness to provide combat casualty care. However, low oncologic procedure volume may hinder the ability of these surgeons to maintain relevant surgical expertise. Other low-volume specialties within the Military Health System (MHS) have established partnerships with neighboring civilian centers to increase procedure volume, but the need for similar partnerships for orthopaedic oncologists has not been examined. The purpose of this study was to characterize the practice patterns of US military fellowship-trained orthopaedic oncologists. We asked the following questions: (1) What are the diagnoses treated by US military fellowship-trained orthopaedic oncologists? (2) What are the procedures performed by US military fellowship-trained orthopaedic oncologists? We queried the Military Data Repository, a centralized repository for healthcare data for all healthcare beneficiaries (active duty, dependents, and retirees) within the Defense Health Agency using the MHS's Management and Reporting Tool for all international common procedure taxonomy (CPT) codes and ICD-9 and ICD-10 codes associated with National Provider Identifier (NPI) numbers of active duty, military fellowship-trained orthopaedic oncologists. Fellowship-trained orthopaedic oncologists were identified by military specialty leaders. Then, we identified all procedures performed by the orthopaedic oncologist based on NPI numbers for fiscal years 2013 to 2022. We stratified the CPT codes by top orthopaedic procedure categories (such as amputation [performed for oncologic and nononcologic reasons], fracture, arthroplasty, oncologic) based on associated ICD codes. These were then tabulated by the most common diagnoses treated. Thirteen percent (796 of 5996) of the diagnoses were oncologic, of which 45% (357 of 796) were malignant. Forty-four percent (158 of 357) of the malignancies were primary and 56% (199 of 357) were secondary; this translates to an average of 2 patients with primary and 2.5 patients with secondary malignancies treated per surgeon per year. During the study period, nine orthopaedic oncologists performed 5996 orthopaedic procedures, or 74 procedures per surgeon per year. Twenty-one percent (1252 of 5996) of the procedures were oncologic; the remaining procedures included 897 arthroplasties, 502 fracture-related, 275 amputations for a nononcologic indication, 204 infections, 142 arthroscopic, and 2724 other procedures. Although military orthopaedic oncologists possess expert skills that are directly translatable to combat casualty care and operational readiness, within MHS hospitals they treat relatively few patients with oncologic diagnoses, and less than one-half of those involve malignancies. Despite postgraduation procedure volume raining stable over the last decade, it is unknown how many new patient visits for oncologic diagnoses and how many corresponding tumor procedures are necessary to maintain competence or build confidence after musculoskeletal oncology fellowship training. It is important to note that there are no military orthopaedic oncology fellowships, and all active duty orthopaedic oncologists undergo training at civilian institutions. Military-civilian partnerships with high-volume cancer centers may enable military orthopaedic oncologists to work at civilian cancer centers to increase their oncologic volume to ensure sustainment of operationally relevant knowledge, skills, and abilities and improve patient care and outcomes.

Ultrasound-Guided Intranodal Lipiodol Lymphangiography for the Assessment and Treatment of Chylous Leaks: A Retrospective Case Series from a Single Center in Switzerland and a Systematic Review of the Literature.

Background: Lymphatic leaks are well-known complications of major thoracic or abdominal surgeries, which significantly heighten morbidity and mortality rates. While the existing literature provides insights into managing these post-operative leaks, with a step-up approach from conservative measures (CMs) to surgical intervention, there are no standardized treatment guidelines. The purpose of this paper is to offer a management algorithm of post-operative lymphatic leaks based on a systematic literature review (SLR) of the therapeutic effect of Lipiodol lymphangiography (LL), completed by a case series of five patients who underwent LL in our department. Methods: In this IRB-approved study, we conducted an SLR following the PRISMA guidelines, using a PICOS. A quality assessment was performed for each study. The case series consisted of consecutive patients who underwent LL for diagnostic and therapeutic purposes at our institution between September 2018 and December 2020. Results: A total of 39 observational studies were included in the SLR comprising 11 retrospective case reviews (Group 1), and 3 case series as well as 25 case reports (Group 2). In total, these studies report cases of 557 patients (51.52% presenting oncological diagnoses; 43.98% having benefited from lymphadenectomy). Lymphatic or chylous fistulas were the most encountered complication, followed by chylothorax. The median volume of Lipiodol injected during lymphography was 11.7 mL (range: 9.8-75 mL). Overall, LL was technically successful in 77.7% (366/471) of patients. The clinical success of all technically successful LLs was 80.6% (295/366). Time-to-leak resolution after lymphography varied between 1 and 31 days. The factors associated with treatment failure were a high leak output (>500 mL/day) and Lipiodol extravasation on post-LL imaging. Our case series consisted of five patients (mean age: 62 ± 9.24 years; 20% female; 100% oncological diagnoses; 60% having beneficiated from lymphadenectomy). Technical and clinical successes were 80% (4/5) and 75% (3/4), respectively. Time-to-leak resolution varied between 1 and 4 days. The volume and technique of LL was not different from that identified in the SLR. Conclusions: LL is a safe procedure with high technical and clinical success rates that could be proposed as both a diagnostic and therapeutic solution for patients with post-operative central lymphatic lesions.

The Effects of Metformin on Cisplatin-Induced Ototoxicity in Diabetic Patients.

Ototoxicity is an important side effect of cisplatin. Recent animal and in vitro studies suggest metformin may protect hearing, though human studies are lacking. We report the first retrospective clinical study exploring the effects of metformin on cisplatin-induced ototoxicity. Retrospective case-control. Tertiary-care center. We reviewed all diabetic patients treated with cisplatin who were enrolled in an auditory monitoring program between January 1, 2000 and December 31, 2018. Patients were grouped according to their use of metformin at the time of chemotherapy. Demographics, oncologic diagnosis and treatment, and prior noise exposure were recorded. Audiometric thresholds were compared before and after chemotherapy. The primary outcome measure was a change in hearing as defined by the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE). Secondary measures of ototoxicity included changes in hearing using the American Speech-Language-Hearing Association (ASHA) and TUNE criteria. Seventy-two diabetic patients met the inclusion criteria. Forty-six patients had taken metformin while on chemotherapy and 26 had not. The proportion of hearing loss as defined by both, CTCAE and ASHA criteria, was more than double in the metformin group relative to the nonmetformin group (CTCAE: 54.4% vs 19.2%, P = .004; ASHA: 62.2% vs 28.0%, P = .003). None of the potential covariates believed to be confounders were significantly associated with the outcome of multivariable analysis. Contrary to expectations from preclinical data, metformin did not reduce the incidence of hearing loss in patients receiving cisplatin and may, in fact, be associated with an increased risk.

Cancer, Consents, and Chaos: Navigating Health Literacy at the Time of a New Oncology Diagnosis.

Cancer, Consents, and Chaos: Navigating Health Literacy at the Time of a New Oncology Diagnosis.