Background In the United States, national guidelines suggest that aggressive cancer care should be avoided in the final months of life. However, guideline compliance currently requires clinicians to make judgments based on their experience as to when a patient is nearing the end of their life. Machine learning (ML) algorithms may facilitate improved end-of-life care provision for patients with cancer by identifying patients at risk of short-term mortality. Objective This study aims to summarize the evidence for applying ML in ≤1-year cancer mortality prediction to assist with the transition to end-of-life care for patients with cancer. Methods We searched MEDLINE, Embase, Scopus, Web of Science, and IEEE to identify relevant articles. We included studies describing ML algorithms predicting ≤1-year mortality in patients of oncology. We used the prediction model risk of bias assessment tool to assess the quality of the included studies. Results We included 15 articles involving 110,058 patients in the final synthesis. Of the 15 studies, 12 (80%) had a high or unclear risk of bias. The model performance was good: the area under the receiver operating characteristic curve ranged from 0.72 to 0.92. We identified common issues leading to biased models, including using a single performance metric, incomplete reporting of or inappropriate modeling practice, and small sample size. Conclusions We found encouraging signs of ML performance in predicting short-term cancer mortality. Nevertheless, no included ML algorithms are suitable for clinical practice at the current stage because of the high risk of bias and uncertainty regarding real-world performance. Further research is needed to develop ML models using the modern standards of algorithm development and reporting.

BACKGROUND In the U.S., national guidelines suggest that aggressive cancer care should be avoided in the final months of life. However, guideline compliance currently requires clinicians to make judgments based on their experience as to when a patient is nearing the end of their life. Machine learning (ML) algorithms may facilitate improved end-of-life care provision for cancer patients by identifying patients at risk of short-term mortality. OBJECTIVE This study sought to summarize the evidence in applying ML in 1-year or shorter cancer mortality prediction for assisting with the transition to end-of-life care for cancer patients. METHODS We searched MEDLINE, Embase, Scopus, Web of Science, and IEEE to identify relevant articles. We included the studies describing ML algorithms predicting 1-year or shorter mortality for oncology patients. We used the prediction model risk of bias assessment tool to assess quality of included studies. RESULTS We included 15 articles involving 110,058 patients in the final synthesis. Twelve studies have a high or unclear risk of bias. Model performance was good: area under the receiver operating characteristic curve ranged from 0.72 - 0.92. We identified common issues leading to biased models, including using a single performance metric, incomplete report of or inappropriate modeling practice, and small sample size. CONCLUSIONS We found signs of encouraging ML performance in predicting short-term cancer mortality. Nevertheless, no included ML algorithms are suitable for clinical practice at the current stage due to the high risk of bias and uncertainty regarding real-world performance. Further research is needed to develop ML models using modern standards of algorithm development and reporting.

12128 Background: Major stressful life events have been shown to be associated with an increased risk of lung cancer, breast cancer and the development of various chronic illnesses. The stress response generated by our body results in a variety of physiological and metabolic changes which can affect the immune system, endocrine system and metabolism which has been shown to be associated with tumor progression. There is an indication that stress may need to be considered as a risk factor for malignancies. Methods: This is a matched case control study. The objective of this study was to determine if major stressful life events are associated with the incidence of head, neck, and pancreatic cancer (HNPC). Cases (CA) were HNPC patients diagnosed within the previous 12 months. Controls (CO) were patients without a prior history of malignancy and were matched with the cases by age and smoking status. Basic demographic data and medical information were collected from the patient’s medical records. Data on major stressful life events were collected using the modified Holmes-Rahe stress scale, and the following variables: death of a spouse, death of a child/immediate family member, serious personal illness, divorce/separation, loss of a job, caring for ill family member, financial difficulties, relocation, stress at work, detention/incarceration and retirement. A total sample of 300 was needed (100 cases, 200 controls) to achieve at least 80% power to detect odds ratios (OR) of 2.00 or higher at 5% level of significance. Results: From January 2018 to August 2021, 278 patients were enrolled (CA = 77, CO = 201) matched for mean age (years) (CA = 63, CO = 64), median smoking exposure (years) (CA = 36, CO = 38). About 65% of patients in CA group and 49% of CO group were male and 54% and 46% of the CA and CO groups respectively were of white race. In a multivariable logistic regression analysis after controlling for potential confounding variables (including sex, age, race, education, marital status, smoking history), there was no difference in lifetime incidence of major stressful event between the cases and controls. However, patients with HNPC were significantly more likely to report a major stressful life event within past 5 years when compared to CO [OR = 2.59 (1.24, 5.44), p = 0.012]. Conclusions: Patients with head, neck and pancreatic cancers are significantly associated with having a major stressful life event within 5 years of their diagnosis. This study highlights the potential need to recognize stressful life events as risk factors for developing malignancies and consider incorporating early rehabilitative efforts for major life stressful events.