Shanti Neff-Baro

M.P.H, Msc Biostatistics

Amaris Consulting

Senior Biostatistician

NLST (Learn more about this study)

NLST-1096

Jul 18, 2023

Application of Predictive Modelling Using Machine Learning Techniques to Predict Survival in Oncology

Decision makers rely on economic evaluations, including cost-effectiveness assessments, to inform their choices of healthcare interventions[1] These analyses rely on accurate measurements of benefits and costs of new treatments, including extrapolation of efficacy outcomes beyond clinical trial follow-up.

Several methods have been developed to extrapolate survival curves in oncological research. However, the interplay of numerous prognostic factors and limitations of standard parametric survival models render it challenging to provide realistic predictions.

The use of artificial intelligence (AI) algorithms in predictive modeling for health-related outcomes has grown steadily over the last few years. Several algorithms have been adapted to the survival framework and perform well when applied to heterogenous clinical data in oncology, including Random Survival Forest (RSF), Support vector Machine for Survival (SVM), Cox regression (as the benchmark), Elastic-net method combined with Cox regression, Coxboost, and Deep Surv[2,3,4,5]. While these methods have been used to predict survival outcomes up to the end of study follow-up and identify prognostic factors, to our knowledge no study has evaluated the efficacy of these methods for long-term extrapolation of survival outcomes.

References
1. Latimer NR and Adler AI. Extrapolation beyond the end of trials to estimate long term survival and cost effectiveness. BMJ Med. 2022;10(1)e000094. doi: 10.1136/bmjmed-2021-000094.
2. Xiao J et al. The Application and Comparison of Machine Learning Models for the Prediction of Breast Cancer Prognosis: Retrospective Cohort Study. JMIR Med Inform. 2022;10(2):e33440. doi: 10.2196/33440.
3. Wang D et al. Development, and validation of machine learning models for predicting prognosis and guiding individualized postoperative chemotherapy: A real-world study of distal cholangiocarcinoma. Front Oncol. 2023;13:1106029. doi: 10.3389/fonc.2023.1106029.
4. Spooner A et al. A comparison of machine learning methods for survival analysis of high-dimensional clinical data for dementia prediction. Sci Rep. 2020;10(1):20410. doi: 10.1038/s41598-020-77220-w.
5. Cygu S et al. Comparing machine learning approaches to incorporate time-varying covariates in predicting cancer survival time. Sci Rep. 2023;13(1):1370. doi: 10.1038/s41598-023-28393-7.