Purpose: Patients with haematologic malignancies have less access to palliative care and are referred later than patients with solid tumours. We developed a survey to investigate this phenomenon, with the intention of analysing palliative care perceptions among health professionals who treat haematology patients and identifying barriers and facilitators to referrals to palliative care services. Methods: This was a multicentre exploratory descriptive web-based survey. A questionnaire was administered to 320 medical and nursing staff members from five Italian haematological units and San Marino’s hospital to investigate their perception of palliative care. Quantitative and qualitative analyses were performed. Results: A total of 142/320 healthcare professionals completed the survey, achieving a 44% response rate. Most of the respondents supported the integration of haematology and palliative care and were aware of the role of palliative care. Despite this, only half had an in-hospital palliative care team, and only a few had previously attended a specific training course. The majority agreed with palliative care referral when the prognosis was less than 3 months or when the symptoms were incoercible and with blood transfusions even in the last stages of the disease. Many considered the presence of an in-hospital palliative care team or a case manager, as well as structured palliative care training, as fundamental facilitators of palliative care referrals. Conclusion: These results showed that healthcare professionals in haematology generally hold a favourable attitude and a high interest in integrating palliative care into their patients’ care. The low referral rate could depend on clinical, cultural, and organisational issues.

PURPOSE: The estimation of prognosis and life expectancy is critical in the care of patients with advanced cancer. To aid clinical decision making, we build a prognostic strategy combining a machine learning (ML) model with explainable artificial intelligence to predict 1-year survival after palliative radiotherapy (RT) for bone metastasis. MATERIALS AND METHODS: Data collected in the multicentric PRAIS trial were extracted for 574 eligible adults diagnosed with metastatic cancer. The primary end point was the overall survival (OS) at 1 year (1-year OS) after the start of RT. Candidate covariate predictors consisted of 13 clinical and tumor-related pre-RT patient characteristics, seven dosimetric and treatment-related variables, and 45 pre-RT laboratory variables. ML models were developed and internally validated using the Python package. The effectiveness of each model was evaluated in terms of discrimination. A Shapley Additive Explanations (SHAP) explainability analysis to infer the global and local feature importance and to understand the reasons for correct and misclassified predictions was performed. RESULTS: The best-performing model for the classification of 1-year OS was the extreme gradient boosting algorithm, with AUC and F1-score values equal to 0.805 and 0.802, respectively. The SHAP technique revealed that higher chance of 1-year survival is associated with low values of interleukin-8, higher values of hemoglobin and lymphocyte count, and the nonuse of steroids. CONCLUSION: An explainable ML approach can provide a reliable prediction of 1-year survival after RT in patients with advanced cancer. The implementation of SHAP analysis provides an intelligible explanation of individualized risk prediction, enabling oncologists to identify the best strategy for patient stratification and treatment selection.

Background On the basis of substantial evidence demonstrate that palliative care combined with standard care improves patient, caregiver, and society outcomes, we have developed a new healthcare model called radiotherapy and palliative care (RaP) outpatient clinic were a radiation oncologist and a palliative care physician make a joint evaluation of advanced cancer patients. Methods We performed a monocentric observational cohort study on advanced cancer patients referred for evaluation at the RaP outpatient clinic. Measures of quality of care were carried out. Results Between April 2016 and April 2018, 287 joint evaluations were performed and 260 patients were evaluated. The primary tumor was lung in 31.9% of cases. One hundred fifty (52.3%) evaluations resulted in an indication for palliative radiotherapy treatment. In 57.6% of cases was used a single dose fraction of radiotherapy (8 Gy). All the irradiated cohort completed the palliative radiotherapy treatment. An 8% of irradiated patients received the palliative radiotherapy treatment in the last 30 days of life. A total of 80% of RaP patients received palliative care assistance until the end of life. Conclusion At the first descriptive analysis, the radiotherapy and palliative care model seem to respond to the need of multidisciplinary approach in order to obtain an improvement on quality of care for advanced cancer patients.

Background/Objectives: Effective palliative care relies on accurate identification and management of symptoms, especially in patients referred for palliative radiotherapy (PRT). This study aimed to identify symptom clusters (SCs)—defined as ≥2 interrelated symptoms—in patients evaluated at a multidisciplinary Radiotherapy and Palliative Care (RaP) outpatient clinic, using the Edmonton Symptom Assessment System (ESAS). Methods: We retrospectively analyzed data from patients referred to the RaP clinic between February 2017 and April 2020. Demographic and clinical characteristics, including ESAS scores at first visit, were collected. Principal component analysis (PCA) and unsupervised K-means clustering (KMC) were used to identify SCs. Associations with ECOG performance status (PS), primary tumor site, metastases site and PRT administration were analyzed. Exploratory survival analyses were performed.Results: Among 215 patients (median age 71 years; 53% male), the mean total ESAS score was 24.03 ± 15.28. PCA identified four SCs: SCPCA1 (tiredness, drowsiness, dyspnea, malaise), SCPCA2 (depression, anxiety), SCPCA3 (nausea, loss of appetite) and SCPCA4 (pain). KMC revealed three SCs: SCKMC1 (pain, tiredness, drowsiness, malaise), SCKMC2 (nausea, loss of appetite, dyspnea), SCKMC3 (depression, anxiety). Worse ECOG PS correlated significantly with physical SCs (p < 0.05). A trend linked SCKMC1 with greater PRT use. Psychological SCs (SCPCA2, SCKMC3) were significantly associated with a lower likelihood of receiving PRT. A trend toward shorter survival was observed among patients belonging to SCKMC2. Conclusions: SC analysis could improve clinical decision-making in the PRT setting. SC profiles reflect patient complexity and may guide personalized symptom management and treatment selection in advanced cancer.