From Real-World Patient Data to Individualized Treatment Effects Using Machine Learning: Current and Future Methods to Address Underlying Challenges.
Journal
Clinical pharmacology and therapeutics
ISSN: 1532-6535
Titre abrégé: Clin Pharmacol Ther
Pays: United States
ID NLM: 0372741
Informations de publication
Date de publication:
01 2021
01 2021
Historique:
received:
05
11
2019
accepted:
14
05
2020
pubmed:
26
5
2020
medline:
21
5
2021
entrez:
26
5
2020
Statut:
ppublish
Résumé
Clinical decision making needs to be supported by evidence that treatments are beneficial to individual patients. Although randomized control trials (RCTs) are the gold standard for testing and introducing new drugs, due to the focus on specific questions with respect to establishing efficacy and safety vs. standard treatment, they do not provide a full characterization of the heterogeneity in the final intended treatment population. Conversely, real-world observational data, such as electronic health records (EHRs), contain large amounts of clinical information about heterogeneous patients and their response to treatments. In this paper, we introduce the main opportunities and challenges in using observational data for training machine learning methods to estimate individualized treatment effects and make treatment recommendations. We describe the modeling choices of the state-of-the-art machine learning methods for causal inference, developed for estimating treatment effects both in the cross-section and longitudinal settings. Additionally, we highlight future research directions that could lead to achieving the full potential of leveraging EHRs and machine learning for making individualized treatment recommendations. We also discuss how experimental data from RCTs and Pharmacometric and Quantitative Systems Pharmacology approaches can be used to not only improve machine learning methods, but also provide ways for validating them. These future research directions will require us to collaborate across the scientific disciplines to incorporate models based on RCTs and known disease processes, physiology, and pharmacology into these machine learning models based on EHRs to fully optimize the opportunity these data present.
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Research Support, U.S. Gov't, Non-P.H.S.
Review
Langues
eng
Sous-ensembles de citation
IM
Pagination
87-100Informations de copyright
© 2020 The Authors. Clinical Pharmacology & Therapeutics published by Wiley Periodicals LLC on behalf of American Society for Clinical Pharmacology and Therapeutics.
Références
Clark, L.C. et al. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin: a randomized controlled trial. JAMA 276, 1957-1963 (1996).
Léauté-Labrèze, C. et al. A randomized, controlled trial of oral propranolol in infantile hemangioma. N. Engl. J. Med. 372, 735-746 (2015).
Rothwell, P.M. External validity of randomised controlled trials: to whom do the results of this trial apply? Lancet 365, 82-93 (2005).
Bothwell, L.E., Greene, J.A., Podolsky, S.H. & Jones, D.S. Assessing the gold standard - lessons from the history of RCTs. N. Engl. J. Med. 374, 2175-2181 (2016).
Frieden, T.R. Evidence for health decision making-beyond randomized, controlled trials. N. Engl. J. Med. 377, 465-475 (2017).
Hutchins, L.F., Unger, J.M., Crowley, J.J., Coltman, C.A. Jr & Albain, K.S. Underrepresentation of patients 65 years of age or older in cancer-treatment trials. N. Engl. J. Med. 341, 2061-2067 (1999).
Ramamoorthy, A., Pacanowski, M., Bull, J. & Zhang, L. Racial/ethnic differences in drug disposition and response: review of recently approved drugs. Clin. Pharmacol. Ther. 97, 263-273 (2015).
Adler-Milstein, J., Holmgren, A.J., Kralovec, P., Worzala, C., Searcy, T. & Patel, V. Electronic health record adoption in us hospitals: the emergence of a digital “advanced use” divide”. J. Am. Med. Inform. Assoc. 24, 1142-1148 (2017).
Beaulieu-Jones, B.K. et al. Examining the use of real-world evidence in the regulatory process. Clin. Pharmacol. Ther. 107, 843-852 (2020).
Lim, B. & van der Schaar, M. Disease-atlas: navigating disease trajectories with deep learning. Machine Learning for Healthcare Conference <https://arxiv.org/abs/1803.10254> (2018).
Tomašev, N. et al. A clinically applicable approach to continuous prediction of future acute kidney injury. Nature 572, 116-119 (2019).
McCauley, M. & Darbar, D. A new paradigm for predicting risk of torsades de pointes during drug development: commentary on: improved prediction of drug- induced torsades de pointes through simulations of dynamics and machine learning algorithms. Clin. Pharmacol. Ther. 100, 324-326 (2016).
Alaa, A.M. & van der Schaar, M. Prognostication and risk factors for cystic fibrosis via automated machine learning. Sci. Rep. 8, 1-19 (2018).
Athreya, A.P. et al. Pharmacogenomics-driven prediction of antidepressant treatment outcomes: a machine-learning approach with multi-trial replication. Clin. Pharmacol. Ther. 106, 855-865 (2019).
Cave, A., Brun, N.C., Sweeney, F., Rasi, G., Senderovitz, T. & HMA-EMA Joint Big Data Taskforce. Big data-how to realize the promise. Clin. Pharmacol. Ther. 107, 753-761 (2020).
Badillo, S. et al. An introduction to machine learning. Clin. Pharmacol. Ther. 107, 871-885 (2020).
Ribba, B., Dudal, S., Lavé, T. & Peck, R.W. Model-informed artificial intelligence: reinforcement learning for precision dosing. Clin. Pharmacol. Ther. 107, 853-857 (2020).
Moore, B.L., Sinzinger, E.D., Quasny, T.M. & Pyeatt, L.D. Intelligent control of closed-loop sedation in simulated ICU patients. Flairs conference, pp. 109-114<https://www.aaai.org/Papers/FLAIRS/2004/Flairs04-023.pdf> (2004).
Gaweda, A.E., Muezzinoglu, M.K., Aronoff, G.R., Jacobs, A.A., Zurada, J.M. & Brier, M.E.Individualization of pharmacological anemia management using reinforcement learning. Neural Netw. 18, 826-834 (2005).
Johansson, F., Shalit, U. & Sontag, D. Learning representations for counterfactual inference. International Conference on Machine Learning, pp. 3020-3029<https://dl.acm.org/doi/10.5555/3045390.3045708> (2016).
Alaa, A.M. & van der Schaar, M. Bayesian inference of individualized treatment effects using multi-task gaussian processes. Advances in Neural Information Processing Systems, 3424-3432 <http://papers.nips.cc/paper/6934-bayesian-inference-of-individualized-treatment-effects-using-multi-task-gaussian-processes.pdf> (2017). Accessed May 7, 2020.
Alaa, A. & van der Schaar, M. Limits of estimating heterogeneous treatment effects: Guidelines for practical algorithm design. Proceedings of the 35th International Conference on Machine Learning. Proc. Machine Learn. Res. 80, 129-138 (2018).
Yoon, J., Jordon, J. & van der Schaar, M. GANITE: Estimation of individualized treatment effects using generative adversarial nets. International Conference on Learning Representations (ICLR) <https://openreview.net/forum?id=ByKWUeWA-> (2018).
Chen, P., Dong, W., Lu, X., Kaymak, U., He, K. & Huang, Z. Deep representation learning for individualized treatment effect estimation using electronic health records. J. Biomed. Inform. 100, 103303 (2019).
Bica, I., Jordon, J. & van der Schaar, M. Estimating the effects of continuous-valued interventions using generative adversarial networks <https://arxiv.org/abs/2002.12326> (2020).
Neyman, J. Sur les applications de la théorie des probabilités aux experiences agricoles: Essai des principes. Roczniki Nauk Rolniczych 10, 1-51 (1923).
Rubin, D.B. Bayesian inference for causal effects: the role of randomization. Ann. Stat. 6, 34-58 (1978).
Robins, J.M. & Hernán, M.A. Estimation of the causal effects of time-varying exposures. In: Longitudinal Data Analysis (eds. Fitzmaurice, G., Davidian, M., Verbeke, G., Molenberghs, G.) 547-593 (Chapman and Hall/CRC, London, UK, 2008).
Wallington, M. et al. 30-day mortality after systemic anticancer treatment for breast and lung cancer in England: a population-based, observational study. Lancet Oncol. 17, 1203-1216 (2016).
Rosenbaum, P.R. & Rubin, D.B. The central role of the propensity score in observational studies for causal effects. Biometrika 70, 41-55 (1983).
Pearl, J. Understanding propensity scores. In Causality: Models, Reasoning, and Inference 2nd edn., 348-352 (Cambridge University Press, Cambridge, UK, 2009).
Fogarty, C.B., Mikkelsen, M.E., Gaieski, D.F. & Small, D.S. Discrete optimization for interpretable study populations and randomization inference in an observational study of severe sepsis mortality. J. Am. Stat. Assoc. 111, 447-458 (2016).
D’Amour, A., Ding, P., Feller, A., Lei, L. & Sekhon, J. Overlap in observational studies with high-dimensional covariates <https://arxiv.org/abs/1711.02582> (2017).
Johansson, F.D. et al. Characterization of overlap in observational studies <https://arxiv.org/abs/1907.04138> (2020).
Schölkopf, B., Platt, J.C., Shawe-Taylor, J., Smola, A.J. & Williamson, R.C. Estimating the support of a high-dimensional distribution. Neural Comput. 13, 1443-1471 (2001).
Brat, G.A. et al. Postsurgical prescriptions for opioid naive patients and association with overdose and misuse: retrospective cohort study. BMJ 360, j5790 (2018).
Mente, A., de Koning, L., Shannon, H.S. & Anand, S.S. A systematic review of the evidence supporting a causal link between dietary factors and coronary heart disease. Arch. Intern. Med. 169, 659-669 (2009).
Młyńczak, M. & Krysztofiak, H.Discovery of causal paths in cardiorespiratory parameters: a time-independent approach in elite athletes. Front. Physiol. 9, 1455 (2018).
Robins, J.M., Rotnitzky, A. & Scharfstein, D.O. Sensitivity analysis for selection bias and unmeasured confounding in missing data and causal inference models. In Statistical Models in Epidemiology, the Environment, and Clinical Trials 1-94 (Springer, New York, NY, 2000).
Scharfstein, D., McDermott, A., Díaz, I., Carone, M., Lunardon, N. & Turkoz, I. Global sensitivity analysis for repeated measures studies with informative drop-out: a semi-parametric approach. Biometrics 74, 207-219 (2018).
Franks, A., Damour, A. & Feller, A. Flexible sensitivity analysis for observational studies without observable implications. J. Am. Stat. Assoc. 1-38 (2019). https://doi.org/10.1080/01621459.2019.1604369.
Lash, T.L., Fox, M.P., MacLehose, R.F., Maldonado, G., McCandless, L.C. & Greenland, S. Good practices for quantitative bias analysis. Int. J. Epidemiol. 43, 1969-1985 (2014).
Kuroki, M. & Pearl, J. Measurement bias and effect restoration in causal inference. Biometrika 101, 423-437 (2014).
Kallus, N., Mao, X. & Udell, M. Causal inference with noisy and missing covariates via matrix factorization. Advances in Neural Information Processing Systems, 6921-6932 <http://papers.nips.cc/paper/7924-causal-inference-with-noisy-and-missing-covariates-via-matrix-factorization.pdf> (2018). Accessed May 7, 2020.
Louizos, C., Shalit, U., Mooij, J.M., Sontag, D., Zemel, R. & Welling, M. Causal effect inference with deep latent-variable models. Adv. Neural Inf. Process. Syst. 6446-6456 <http://papers.nips.cc/paper/7223-causal-effect-inference-with-deep-latent-variable-models.pdf> (2017). Accessed May 7, 2020.
Lee, C., Mastronarde, N. & van der Schaar, M. Estimation of individual treatment effect in latent confounder models via adversarial learning. NIPS Machine Learning for Health Workshop <https://arxiv.org/abs/1811.08943> (2018).
Wang, Y. & Blei, D.M. The blessings of multiple causes <https://arxiv.org/abs/1805.06826> (2019).
Ranganath, R. & Perotte, A. Multiple causal inference with latent confounding <https://arxiv.org/pdf/1805.08273.pdf> (2018).
Bica, I., Alaa, A.M. & van der Schaar, M. Time series deconfounder: estimating treatment effects over time in the presence of hidden confounders <https://arxiv.org/pdf/1902.00450.pdf> (2019).
Zhang, L., Wang, Y., Ostropolets, A., Mulgrave, J.J., Blei, D.M. & Hripcsak, G. The medical deconfounder: assessing treatment effects with electronic health records (EHRs) <https://arxiv.org/abs/1904.02098v1> (2019).
Shalit, U., Johansson, F.D. & Sontag, D. Estimating individual treatment effect: Generalization bounds and algorithms. In: Proceedings of the 34th International Conference on Machine Learning-Volume 70, pp. 3076-3085 <http://proceedings.mlr.press/v70/shalit17a.html> (2017).
Chipman, H.A. et al. Bart: Bayesian additive regression trees. Ann. Appl. Stat. 4, 266-298 (2010).
Hill, J.L. Bayesian nonparametric modeling for causal inference. J. Comput. Graph. Stat. 20, 217-240 (2011).
Athey, S. & Imbens, G. Recursive partitioning for heterogeneous causal effects. Proc. Natl. Acad. Sci. USA 113, 7353-7360 (2016).
Wager, S. & Athey, S. Estimation and inference of heterogeneous treatment effects using random forests. J. Am. Stat. Assoc. 113, 1228-1242 (2018).
Hahn, P.R., Murray, J.S. & Carvalho, C. Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effects <https://arxiv.org/abs/1706.09523> (2017).
Yao, L., Li, S., Li, Y., Huai, M., Gao, J. & Zhang, A. Representation learning for treatment effect estimation from observational data <https://paperswithcode.com/paper/representation-learning-for-treatment-effect> (2018).
Alaa, A.M., Weisz, M. & van der Schaar, M. Deep counterfactual networks with propensity-dropout. ICML 2017 Workshop on Principled Approaches to Deep Learning <https://arxiv.org/pdf/1706.05966.pdf> (2017).
Goodfellow, I. et al. Generative adversarial nets <https://dl.acm.org/doi/10.5555/2969033.2969125> (2014).
Alaa, A.M. & van der Schaar, M. Bayesian nonparametric causal inference: Information rates and learning algorithms. IEEE J. Select. Top. Signal Proc. 12, 1031-1046 (2018).
Wendling, T., Jung, K., Callahan, A., Schuler, A., Shah, N. & Gallego, B. Comparing methods for estimation of heterogeneous treatment effects using observational data from health care databases. Stat. Med. 37, 3309-3324 (2018).
Khera, S. et al. Association between hospital volume and 30-day readmissions following transcatheter aortic valve replacement. JAMA Cardiol. 2, 732-741 (2017).
Kolte, D. et al. Thirty-day readmissions after endovascular or surgical therapy for critical limb ischemia: analysis of the 2013 to 2014 nationwide readmissions databases. Circulation 136, 167-176 (2017).
Imbens, G.W. The role of the propensity score in estimating dose-response functions. Biometrika 87, 706-710 (2000).
Imai, K. & Van Dyk, D.A. Causal inference with general treatment regimes: generalizing the propensity score. J. Am. Stat. Assoc. 99, 854-866 (2004).
Hirano, K. & Imbens, G.W. The propensity score with continuous treatments. Applied Bayesian Modeling and Causal Inference from Incomplete-Data Perspectives 226164, 73-84 <https://scholar.harvard.edu/imbens/files/hir_07feb04.pdf> (2004).
Schwab, P., Linhardt, L., Bauer, S., Buhmann, J.M. & Karlen, W. Learning counterfactual representations for estimating individual dose-response curves <https://arxiv.org/pdf/1902.00981.pdf> (2019).
Robins, J. A new approach to causal inference in mortality studies with a sustained exposure period-application to control of the healthy worker survivor effect. Mathemat. Model. 7, 1393-1512 (1986).
Robins, J.M. Correcting for non-compliance in randomized trials using structural nested mean models. Commun. Statist. Theory Meth. 23, 2379-2412 (1994).
Robins, J.M., Hernan, M.A. & Brumback, B. Marginal structural models and causal inference in epidemiology. Epidemiology 11, 550-560 (2000).
Xu, Y., Xu, Y. & Saria, S. A Bayesian nonparametric approach for estimating individualized treatment-response curves. Machine Learning for Healthcare Conference, pp. 282-300 <https://arxiv.org/pdf/1608.05182v1.pdf> (2016).
Roy, J., Lum, K.J. & Daniels, M.J. A Bayesian nonparametric approach to marginal structural models for point treatments and a continuous or survival outcome. Biostatistics 18, 32-47 (2016).
Schulam, P. & Saria, S. Reliable decision support using counterfactual models <https://arxiv.org/pdf/1703.10651.pdf> (2017).
Soleimani, H., Subbaswamy, A. & Saria, S. Treatment-response models for counterfactual reasoning with continuous-time, continuous-valued interventions <https://arxiv.org/abs/1704.02038> (2017).
Lim, B., Alaa, A. & van der Schaar, M. Forecasting treatment responses over time using recurrent marginal structural networks <http://papers.nips.cc/paper/7977-forecasting-treatment-responses-over-time-using-recurrent-marginal-structural-networks.pdf> (2018).
Bica, I., Alaa, A.M., Jordon, J. & van der Schaar, M. Estimating counterfactual treatment outcomes over time through adversarially balanced representations. International Conference on Learning Representations <https://openreview.net/pdf?id=BJg866NFvB> (2020)
Alaa, A. & van der Schaar, M. Validating causal inference models via influence functions. International Conference on Machine Learning, pp. 191-201<http://proceedings.mlr.press/v97/alaa19a/alaa19a.pdf> (2019).
Hampel, F.R., Ronchetti, E.M., Rousseeuw, P.J. & Stahel, W.A. Robust statistics: the approach based on influence functions, Vol. 196, (John Wiley & Sons, Hoboken, NJ, 2011).
Robins, J., Tchetgen, E., van der Vaart, A. & Lingling, L. Higher order influence functions and minimax estimation of nonlinear functionals. In Probability and Statistics: Essays in Honor of David A. Freedman (eds. Nolan, D. & Speed, T.) 335-421 (Institute of Mathematical Statistics, Beachwood, OH, 2008).
Silva, R. Observational-interventional priors for dose-response learning. In: Advances in Neural Information Processing Systems 1561-1569 <http://papers.nips.cc/paper/6107-observational-interventional-priors-for-dose-response-learning.pdf> (2016). Accessed May 7, 2020.
Kallus, N., Puli, A.M. & Shalit, U. Removing hidden confounding by experimental grounding. In: Advances in Neural Information Processing Systems 10888-10897 <http://papers.nips.cc/paper/8286-removing-hidden-confounding-by-experimental-grounding.pdf> (2018). Accessed May 7, 2020.
Koch, G., Pfister, M., Daunhawer, I., Wilbaux, M., Wellmann, S. & Vogt, J.E. Pharmacometrics and machine learning partner to advance clinical data analysis. Clin. Pharmacol. Ther. 107, 926-933 (2020).
Geng, C., Paganetti, H. & Grassberger, C. Prediction of treatment response for combined chemo-and radiation therapy for non-small cell lung cancer patients using a bio-mathematical model. Sci. Rep. 7, 13542 (2017).
Goulooze, S.C. et al. Beyond the randomized clinical trial: innovative data science to close the pediatric evidence gap. Clin. Pharmacol. Ther. 107, 786-795 (2019).
Xu, H., Li, J., Jiang, X. & Chen, Q. Electronic health records for drug repurposing: Current status, challenges, and future directions. Clin. Pharmacol. Ther. 107, 712-714 (2020).
Eichler, H.-G. et al. Are novel, nonrandomized analytic methods fit for decision making? the need for prospective, controlled, and transparent validation. Clin. Pharmacol. Ther. 107, 773-779 (2020).