Comparative evaluation of machine learning models in predicting overall survival for nasopharyngeal carcinoma using 18F-FDG PET-CT parameters.

PURPOSE: The objective of this study is to assess the prognostic efficacy of 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET-CT) parameters in nasopharyngeal carcinoma (NPC) and identify the best machine learning (ML) prognostic model for NPC patients based on these 18F-FDG PET/CT parameters and clinical variables. METHOD: A cohort of 678 patients diagnosed with NPC between 2016 and 2020 was analyzed in this study. The model was constructed using four advanced ML algorithms, namely Random Forest (RF), Extreme Gradient Boosting (XGBoost), Least Absolute Shrinkage and Selection Operator (LASSO), and multifactor COX step-up regression. Statistical significance of the models was assessed using Kaplan-Meier (K-M) curves, with a significance level established at P < 0.05. The prognostic efficacy of the models was evaluated through the analysis of receiver operating characteristic (ROC) curves, with the area under the ROC curve (AUC) serving as a criterion for model selection. The decision curve analysis (DCA) and concordance index (C-index) were employed to assess the precision of the optimal model. RESULTS: Multivariate analysis revealed age, T stage, and metabolic tumor volume (MTV) for the primary nasopharyngeal tumor (MTVT) as significant independent prognostic factors for overall survival (OS) in NPC patients. Additionally, the LASSO model identified six key variables, including peak standardized uptake value (SUV-peak) for the primary nasopharyngeal tumor (SUV-peak(T)), MTVT, heterogeneity index for neck lymph nodes (HIN), age, pathological type, and T stage. Remarkably, the LASSO model demonstrated superior performance with a 5-year AUC of 0.849 compared to other models. Further assessment using the C-index and DCA confirmed the accuracy of the LASSO model. Subgroup analysis revealed notable risk factors, such as a high heterogeneity index (HI) for the primary nasopharyngeal tumor (HIT), MTV values for neck lymph nodes (MTVN), and HIN. CONCLUSIONS: We developed a novel prognostic machine learning model that integrates 18F-FDG PET-CT parameters and clinical characteristics, significantly enhancing prognosis prediction in NPC.

Can neoadjuvant systemic therapy provide additional benefits for T1 HER2+ breast cancer patients: a subgroup analysis based on different high-risk signatures.

INTRODUCTION: Neoadjuvant systemic therapy (NAST) is vital in the management of HER2-positive (HER2+) breast cancer. Nevertheless, the indications for NAST in tumors <2 cm remain controversial. METHOD: A total of 7961 patients were screened from the Surveillance, Epidemiology, and End Result database. Independent prognostic factors were identified using multivariate Cox analysis. Subgroup analyses and Kaplan-Meier analyses were used to simulate whether NAST would provide a survival benefit with different high-risk characteristics. Nomograms were constructed, and an internal validation cohort was employed. RESULTS: Of the 7961 included patients, 1137 (14.3%) underwent NAST. In the total population, NAST was associated with poorer overall survival (OS) and breast cancer-specific survival (BCSS) (OS: P = 0.00093; BCSS: P  <  0.0001). Multivariate Cox analysis confirmed that NAST markedly affected the prognosis of enrolled patients. Besides, a direct association between T, N, age, subtype, and prognosis was observed. Subgroup analyses yielded in these three subgroups, T1c, hormone receptor-negative, and 61-69 years of age, NAST and AST had comparable OS, while NAST possessed worse BCSS. Notably, even in the N3, we still did not observe any additional benefit of NAST. The calculated C-index of 0.72 and 0.73 confirmed the predictability of the nomograms. The AUCs exhibit consistency in the training and validation cohorts. CONCLUSION: Our findings suggest that NAST does not provide additional benefit to patients with T1 HER2+ breast cancer, even in the presence of lymph node metastasis, T1c, or hormone receptor negativity. This study facilitates the implementation of individualized management strategies.

Rivaroxaban in patients with symptomatic peripheral artery disease after lower extremity bypass surgery with venous and prosthetic conduits.

BACKGROUND: Patients with peripheral artery disease (PAD) requiring lower extremity revascularization (LER) have a high risk of adverse limb and cardiovascular events. The results from the VOYAGER PAD (efficacy and safety of rivaroxaban in reducing the risk of major thrombotic vascular events in subjects with symptomatic peripheral artery disease undergoing peripheral revascularization procedures of the lower extremities) trial have demonstrated that rivaroxaban significantly reduced this risk with an overall favorable net benefit for patients undergoing surgical revascularization. However, the efficacy and safety for those treated by surgical bypass, including stratification by bypass conduit (venous or prosthetic), has not yet been described. METHODS: In the VOYAGER PAD trial, patients who had undergone surgical and endovascular infrainguinal LER to treat PAD were randomized to rivaroxaban 2.5 mg twice daily or placebo on top of background antiplatelet therapy (aspirin 100 mg to be used in all and clopidogrel in some at the treating physician's discretion) and followed up for a median of 28 months. The primary end point was a composite of acute limb ischemia, major amputation of vascular etiology, myocardial infarction, ischemic stroke, and cardiovascular death. The principal safety outcome was major bleeding using the TIMI (thrombolysis in myocardial infarction) scale. The index procedure details, including conduit type (venous vs prosthetic), were collected at baseline. RESULTS: Among 6564 randomized patients, 2185 (33%) had undergone surgical LER. Of these 2185 patients, surgical bypass had been performed for 1448 (66%), using a prosthetic conduit for 773 patients (53%) and venous conduit for 646 patients (45%). Adjusting for the baseline differences and anatomic factors, the risk of unplanned limb revascularization in the placebo arm was 2.5-fold higher for those receiving a prosthetic conduit vs a venous conduit (adjusted hazard ratio [HR], 2.53; 95% confidence interval [CI], 1.65-3.90; P < .001), and the risk of acute limb ischemia was three times greater (adjusted HR, 3.07; 95% CI, 1.84-5.11; P < .001). The use of rivaroxaban reduced the primary outcome for the patients treated with bypass surgery (HR, 0.78; 95% CI, 0.62-0.98), with consistent benefits for those receiving venous (HR, 0.66; 95% CI, 0.49-0.96) and prosthetic (HR, 0.87; 95% CI, 0.66-1.15) conduits (Pinteraction = .254). In the overall trial, major bleeding using the TIMI scale was increased with rivaroxaban. However, the numbers for those treated with bypass surgery were low (five with rivaroxaban vs nine with placebo; HR, 0.55; 95% CI, 0.18-1.65) and not powered to show statistical significance. CONCLUSIONS: Surgical bypass with a prosthetic conduit was associated with significantly higher rates of major adverse limb events relative to venous conduits even after adjustment for patient and anatomic characteristics. Adding rivaroxaban 2.5 mg twice daily to aspirin or dual antiplatelet therapy significantly reduced this risk, with an increase in the bleeding risk, but had a favorable benefit risk for patients treated with bypass surgery, regardless of conduit type. Rivaroxaban should be considered after lower extremity bypass for symptomatic PAD to reduce ischemic complications of the heart, limb, and brain.

Two years of school-based intervention program could improve the physical fitness among Ecuadorian adolescents at health risk: subgroups analysis from a cluster-randomized trial.

BACKGROUND: Adolescents with overweight and poor physical fitness have an increased likelihood of developing cardiovascular diseases during adulthood. In Ecuador, a health promotion program improved the muscular strength and speed-agility, and reduced the decline of the moderate-to-vigorous physical activity of adolescents after 28 months. We performed a sub-group analysis to assess the differential effect of this intervention in overweight and low-fit adolescents. METHODS: We performed a cluster-randomized pair matched trial in schools located in Cuenca-Ecuador. In total 20 schools (clusters) were pair matched, and 1440 adolescents of grade 8 and 9 (mean age of 12.3 and 13.3 years respectively) participated in the trial. For the purposes of the subgroup analysis, the adolescents were classified into groups according to their weight status (body mass index) and aerobic capacity (scores in the 20 m shuttle run and FITNESSGRAM standards) at baseline. Primary outcomes included physical fitness (vertical jump, speed shuttle run) and physical activity (proportion of students achieving over 60 min of moderate-to-vigorous physical activity/day). For these primary outcomes, we stratified analysis by weight (underweight, normal BMI and overweight/obese) and fitness (fit and low fitness) groups. Mixed linear regression models were used to assess the intervention effect. RESULTS: The prevalence of overweight/obesity, underweight and poor physical fitness was 20.3 %, 5.8 % and 84.8 % respectively. A higher intervention effect was observed for speed shuttle run in overweight (ß = -1.85 s, P = 0.04) adolescents compared to underweight (ß = -1.66 s, P = 0.5) or normal weight (ß = -0.35 s, P = 0.6) peers. The intervention effect on vertical jump was higher in adolescents with poor physical fitness (ß = 3.71 cm, P = 0.005) compared to their fit peers (ß = 1.28 cm, P = 0.4). The proportion of students achieving over 60 min of moderate-to-vigorous physical activity/day was not significantly different according to weight or fitness status. CONCLUSION: Comprehensive school-based interventions that aim to improve diet and physical activity could improve speed and strength aspects of physical fitness in low-fit and overweight/obese adolescents. TRIAL REGISTRATION: Clinicaltrials.gov identifier NCT01004367 . Registered October 28, 2009.