European Journal of Cancer Care最新文献

Objective

This study aimed to construct interpretable machine learning models to predict family intimacy in cancer patients and identify the most influential predictors through SHAP-based analysis.

Methods

A total of 259 cancer patients were surveyed. The data cleaning process involved handling missing values, normalizing continuous variables, and applying one-hot encoding to categorical variables. Statistically significant sociodemographic variables (age, marital status, education, and income) and psychosocial attributes (self-esteem and three resilience subdimensions: tenacity, strength, and optimism) were selected using LASSO regression. Four regression models—gradient boosting (GB), random forest (RF), XGBoost (XGB), and decision tree (DT)—were trained and evaluated using R², mean-squared error (MSE), and mean absolute percentage error (MAPE).SHapley Additive exPlanations (SHAP) was used to interpret the GB model.

Results

The GB model achieved the best predictive performance (R² = 0.6985, MSE = 0.2405), followed by XGB (R² = 0.6794), RF (R² = 0.6653), and DT (R² = 0.5912). SHAP analysis revealed that psychological variables—tenacity, strength, and self-esteem—were the most influential predictors, all exerting strong positive effects. Age group and education showed moderate impact, while income, gender, and marital status contributed minimally.

Conclusion

Gradient boosting offers a robust and interpretable framework for predicting family intimacy in cancer patients. Positive psychological resources—especially resilience and self-esteem—outperform traditional demographics as a predictive foundation, highlighting their clinical significance in survivorship care planning.

Background and Objectives

Breast cancer patients are at high risk of malnourishment. However, data concerning the nutritional status of breast cancer patients are very limited. We aimed to assess the nutritional status and diet of women diagnosed with breast cancer.

Methods

A total of 32 women diagnosed with breast cancer were included in this cross-sectional study. Patients were recruited via an online survey that was distributed in multiple social media applications as well as shared in oncology clinics. Demographic data were collected via the online survey, while dietary data were collected during a phone interview.

Results

Prevalence of overweight and obesity among study samples was 18.8% and 56.3%, respectively. Only 18.8% of the patients reported visiting a dietitian, while 59.4% of the patients met the recommendation of free sugar intake of < 5%. Most of the patients (81%) did not meet their energy requirements, whereas none of the patients met their protein requirement. Diet quality score and dietary intake were not associated with dietitian visit or stages of breast cancer. Stepwise regression analysis indicated significant association between iron intake and weight status (B = −1.56, SE = 0.69, and [95% confidence interval (CI): −2.97–−0.15]). Results also indicated that age and percentage of free sugar from total energy predicted BMI of patients (age: B = 0.20, SE = 0.08, and [95% CI: 0.06–035]; percent of free sugar: B = 0.66, SE = 0.24, and [95% CI: 0.18–1.15]). Weight status was significantly associated with adequacy of energy intake.

Conclusions

High prevalence of overweight and obesity, high intake of free sugar, and limited intake of protein were observed among women diagnosed with breast cancer. There is an urgent need to provide nutritional care for breast cancer patients in Saudi Arabia that is delivered by experienced dietitian to improve the nutritional status of these patients.

Objective

Lung cancer surgery can lead to postoperative complications such as pleural effusion and chylothorax. This study aims to investigate the impact of a medium-chain triglyceride (MCT) diet on recovery and complications in lung cancer patients.

Methods

This randomized controlled trial enrolled lung cancer patients at the Ningbo University First Affiliated Hospital between February 2022 and December 2023. Participants were randomly assigned to either the MCT group or the routine diet group. The MCT group received an MCT diet from the first to the third day postoperatively and resumed a normal diet thereafter. The RD group received a staple regular diet with no restrictions. The primary outcome was chylothorax incidence within 72 h. The secondary outcomes included the pleural effusion volume, protein, triglycerides, and mononuclear cells in pleural effusion at 24 h and 48 h postoperatively.

Results

A total of 127 patients (46 males) were enrolled, with 57 (mean age 54.91 ± 11.21 years, 21 males) receiving MCT and 70 (mean age 56.54 ± 11.84 years, 25 males) receiving RD. No chylothorax cases were observed in either group at 72 h. At 48 h, the MCT group had significantly lower pleural effusion volume (123.07 vs. 166.79 mL, p = 0.006), protein concentration (33.64 vs. 36.34 g/L, p = 0.007), and triglyceride levels (0.43 vs. 0.63 mmol/L, p = 0.001) compared to the RD group. No significant differences were observed in mononuclear cell counts between the two groups, but significant differences in mononuclear cell counts (p = 0.04) were observed at 24 h and 48 h in the MCT group, while no significant differences were observed in the RD group.

Conclusion

The MCT diet reduced protein and triglyceride levels in pleural fluid and decreased pleural effusion volume after 48 h. While no cases of chylothorax were noted, MCT dietary modifications may enhance postoperative recovery in lung cancer patients.

Trial Registration: Chinese Clinical Trial Registry: ChiCTR2400089381

Background

Personalised interventions are recommended to optimise participation in prehabilitation programmes, but this requires insight into what motivates patients to participate in, and complete, such programmes. We investigated the experiences of patients with oesophageal cancer who had participated in a prehabilitation programme, with a view to identifying factors that contribute to programme participation.

Methods

This multicentre, qualitative study included patients who had completed a curative trajectory of neoadjuvant chemoradiotherapy and surgery for oesophageal cancer and who had been offered a multimodal prehabilitation programme in the period between chemoradiotherapy and surgery. Prehabilitation consisted of nutritional support and a supervised physical training programme. Data were collected after surgery using semi-structured interviews. Data were analysed using an inductive thematic analysis.

Results

Twelve patients were interviewed. Main themes contributing to programme participation were ‘development of internal motivation’, ‘external motivation through relational support’ and ‘programme feasibility and perceived benefits’. Factors related to internal motivation were ‘a clear purpose (to get through surgery)’, ‘patients’ sense of responsibility for treatment success’ and ‘a positive change of mind during the programme’. Factors related to external motivation included ‘support and supervision from healthcare providers’ and ‘support from family members’. The programme was feasible, but patients differed in their preferred level of dietary support and training supervision. Overall, the patients were positive about the programme, feeling that it prepared them for surgery.

Conclusions

This study shows that patients with oesophageal cancer are motivated by internal and external factors to participate in a multimodal prehabilitation programme. Clear information on the purpose of the programme, tailored supervision by healthcare providers and active involvement of family members support this feasible form of prehabilitation care.

Background

Identifying patients’ decisional conflict and its predictors is essential since it is a critical step before developing a patient decision aid for gastric cancer patients to promote decision-making in perioperative oral nutritional supplementation.

Aims

To investigate the current status of decisional conflict and its relationship with decision preparation and decision participation regarding perioperative oral nutritional supplementation in gastric cancer patients and to explore the predictors of decisional conflict.

Methods

A convenience sampling method was employed to recruit inpatients undergoing radical gastrectomy for gastric cancer, and self-reported questionnaires were used to collect data. The Random Forest model, least absolute shrinkage and selection operator regression, and multivariate stepwise linear regression analysis were employed for data analysis.

Results

A total of 181 patients with gastric cancer were investigated. 42.5% of them had no decisional conflict, and 57.5% of them had decisional conflict. 71.3% of them expressed a desire to engage in shared decision-making, but only 29.3% of them actually participated in it. The actual type of decision-making and knowledge about oral nutritional supplementation were predictors of decisional conflict.

Conclusion

Over half of the gastric cancer patients experienced decisional conflict regarding oral nutritional supplementation during the perioperative period. The actual type of decision-making and knowledge about oral nutritional supplementation among gastric cancer patients are two predictors for decisional conflict. Healthcare professionals should consider the two predictors when developing a patient decision aid to provide decision support.

Purpose

There is a lack of comprehensive analysis on the associations between malnutrition, body composition, and clinical, anthropometric, biochemical, nutrition-inflammation markers, mineral and trace element levels, cytokine concentrations, and antioxidant activity in patients with cancer. Differences regarding sex at birth, ethnicity, and cancer types and stages further contribute to variations in body composition and mineral and trace element levels. This study examines factors associated with malnutrition and body composition, adjusting for relevant clinical, biochemical, nutritional data, and lifestyle confounders.

Methods

In total, 78 male patients with recurrent or metastatic head and neck cancer (HNC) or nasopharyngeal carcinoma (NPC) were recruited. Clinical characteristics (e.g., age, tumor status, and performance status), lifestyle factors, anthropometric data, nutrition-inflammation markers, plasma levels of minerals, trace elements, and cytokines as well as antioxidant activity were determined and analyzed using multivariable logistic regression and analysis of covariance models.

Results

Half of the patients (50.3%) exhibited advanced grades of BMI-adjusted body weight loss (BWL), and nearly 90% had C-reactive protein levels exceeding 5 mg/L, reflecting poor nutritional status and elevated inflammation, despite being medically fit for chemotherapy. Plasma copper levels were independently associated with BMI-adjusted BWL as well as with lean body mass and total fat mass. Similarly, plasma zinc levels showed an independent association with total fat mass.

Conclusions

Plasma levels of trace elements, including copper and zinc, are strongly correlated with malnutrition and body composition in male patients with recurrent or metastatic HNC or NPC.

Objective

Prevalence rates of depression and anxiety in patients with a serious physical illness are high. Psychodynamic psychotherapies show potential in mitigating such psychological distress because they address hindering relational experiences and unconscious conflicts that may impair patients’ capacity to cope with illness-related distress. We investigated the efficacy of psychodynamic psychotherapy for symptoms of depression and anxiety in patients with serious physical illness.

Methods

We conducted a systematic literature search to identify randomized controlled trials and quasi-experiments of psychodynamic psychotherapies in adults with serious physical illness. We conducted random-effects meta-analyses and narrative summaries.

Results

We identified 15,112 records and included 5 RCTs (n = 648, mean age: 52.4 years, 69% female). Meta-analyses showed no significant differences between psychodynamic and standard treatment for the reduction of depressive symptoms in serious physical illness (SMD = −0.24, 95% CI: −0.87 to 0.39). No meta-analysis was calculated for anxiety symptoms due to the small number of studies investigating this outcome.

Conclusion

Quantitative evidence for the efficacy of psychodynamic psychotherapy in this population is inconclusive. More research is needed with regard to effective intervention length and adequate outcome measures, especially in patients with incurable illness.

Introduction

This study demonstrates the potential benefits of large language models (LLMs) such as ChatGPT-4o and DeepSeek-R1 in supporting multidisciplinary uro-oncologic tumor boards (MUTBs) discussing complex urological-oncological cases. In particular, DeepSeek-R1 has demonstrated better performance than ChatGPT-4o in specific types of urological malignancies and specific treatment protocols.

Objective

This study examined the agreement between the AI-based systems ChatGPT-4o and DeepSeek-R1 and the decision making of a MUTB, which is a team of experts who assess uro-oncological cases.

Methods

A total of 97 uro-oncological cases covering a period of one year were retrospectively evaluated. The treatment plans for the cases were determined using traditional tumor board methods and separately presented to ChatGPT-4o and DeepSeek-R1. Comprehensive medical records were created for each case, and the planned treatments were classified as surgical intervention, radiotherapy/chemotherapy (RT-CT), advanced imaging, or follow-up. Agreement between the board and AI decisions was analyzed using Cohen’s kappa and with precision, recall, and F1 scores.

Results

In prostate cancer cases, DeepSeek-R1 showed substantial agreement (κ = 0.787, p < 0.001) with the board decisions, while ChatGPT-4o showed moderate agreement (κ = 0.593, p < 0.001). For renal malignancies, DeepSeek-R1 showed moderate agreement (κ = 0.603,p < 0.001), while ChatGPT-4o showed slight agreement (κ = 0.172, p < 0.001). In bladder cancer cases, DeepSeek-R1 showed moderate agreement (κ = 0.604, p < 0.05), but ChatGPT-4o did not show statistically significant agreement (κ = 0.217, p > 0.05). When all cases were considered together, DeepSeek-R1 showed high agreement, while ChatGPT-4o showed moderate agreement (DeepSeek-R1: κ = 0.773, p < 0.001; ChatGPT-4o: κ = 0.608, p < 0.001).

Conclusion

Our findings suggest that DeepSeek-R1’s decision support performance is better than that of ChatGPT-4o in various oncological scenarios. The results highlight the potential for publicly available AI models to be used in complex clinical fields, such as oncology.