Integrating Machine Learning With Microsimulation to Classify Hypothetical, Novel Patients for Predicting Pregabalin Treatment Response Based on Observational and Randomized Data in Patients With Painful Diabetic Peripheral Neuropathy

Joe Alexander Jr, Roger A Edwards, Luigi Manca, Roberto Grugni, Gianluca Bonfanti, Birol Emir, Ed Whalen, Steve Watt, Marina Brodsky, Bruce Parsons, Joe Alexander Jr, Roger A Edwards, Luigi Manca, Roberto Grugni, Gianluca Bonfanti, Birol Emir, Ed Whalen, Steve Watt, Marina Brodsky, Bruce Parsons

Abstract

Purpose: Variability in patient treatment responses can be a barrier to effective care. Utilization of available patient databases may improve the prediction of treatment responses. We evaluated machine learning methods to predict novel, individual patient responses to pregabalin for painful diabetic peripheral neuropathy, utilizing an agent-based modeling and simulation platform that integrates real-world observational study (OS) data and randomized clinical trial (RCT) data.

Patients and methods: The best supervised machine learning methods were selected (through literature review) and combined in a novel way for aligning patients with relevant subgroups that best enable prediction of pregabalin responses. Data were derived from a German OS of pregabalin (N=2642) and nine international RCTs (N=1320). Coarsened exact matching of OS and RCT patients was used and a hierarchical cluster analysis was implemented. We tested which machine learning methods would best align candidate patients with specific clusters that predict their pain scores over time. Cluster alignments would trigger assignments of cluster-specific time-series regressions with lagged variables as inputs in order to simulate "virtual" patients and generate 1000 trajectory variations for given novel patients.

Results: Instance-based machine learning methods (k-nearest neighbor, supervised fuzzy c-means) were selected for quantitative analyses. Each method alone correctly classified 56.7% and 39.1% of patients, respectively. An "ensemble method" (combining both methods) correctly classified 98.4% and 95.9% of patients in the training and testing datasets, respectively.

Conclusion: An ensemble combination of two instance-based machine learning techniques best accommodated different data types (dichotomous, categorical, continuous) and performed better than either technique alone in assigning novel patients to subgroups for predicting treatment outcomes using microsimulation. Assignment of novel patients to a cluster of similar patients has the potential to improve prediction of patient outcomes for chronic conditions in which initial treatment response can be incorporated using microsimulation.

Clinical trial registries: www.clinicaltrials.gov: NCT00156078, NCT00159679, NCT00143156, NCT00553475.

Keywords: agent-based modeling and simulation; coarsened exact matching; hierarchical cluster analysis; machine learning; time series regressions.

Conflict of interest statement

Birol Emir, Bruce Parsons, Stephen Watt, and Ed Whalen are employees of Pfizer. Joe Alexander Jr and Marina Brodsky were employed by Pfizer at the time the study was conducted. Roger A Edwards is an employee of Health Services Consulting Corporation who was a paid consultant by Pfizer in connection with this study and development of this manuscript. Luigi Manca, Roberto Grugni, and Gianluca Bonfanti are employees of Fair Dynamics Consulting, who were paid subcontractors to Health Services Consulting Corporation in connection with this study and the development of this manuscript. The authors report no other conflicts of interest in this work.

© 2019 Alexander Jr et al.

Figures

Figure 1
Figure 1
Simulation steps. Reproduced from Alexander J, Edwards RA, Brodsky M, et al Using time-series analysis approaches for improved prediction of pain outcomes in subgroups of patients with painful diabetic peripheral neuropathy. PLoS One. 2018;13(12):e0207120. Creative commons license and disclaimer available from http://creativecommons.org/licenses/by/4.0/legalcode. Abbreviations: OS, observational study; PDF, probability density function; RCT, randomized controlled trial.
Figure 2
Figure 2
Accuracy results for the kNN method only, SFCM method only, and the ensemble method in (A) training dataset by cluster, (B) testing dataset by cluster, and (C) overall testing and training datasets. Abbreviations: kNN, k-nearest neighbors; SFCM, supervised fuzzy c-means.
Figure 3
Figure 3
Ensemble method flowchart.

References

    1. Collins FS, Varmus H. A new initiative on precision medicine. N Engl J Med. 2015;372(9):793–795. doi:10.1056/NEJMp1500523
    1. Sim I. Two ways of knowing: big data and evidence-based medicine. Ann Intern Med. 2016;164(8):562–563. doi:10.7326/M15-2970
    1. Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, and cost. Health Aff (Millwood). 2008;27(3):759–769. doi:10.1377/hlthaff.27.3.759
    1. Amarasingham R, Patzer RE, Huesch M, Nguyen NQ, Xie B. Implementing electronic health care predictive analytics: considerations and challenges. Health Aff (Millwood). 2014;33(7):1148–1154. doi:10.1377/hlthaff.2014.0352
    1. Hannan EL. Randomized clinical trials and observational studies: guidelines for assessing respective strengths and limitations. JACC Cardiovasc Interv. 2008;1(3):211–217. doi:10.1016/j.jcin.2008.01.008
    1. Pfizer Inc. Lyrica (pregabalin) Prescribing Information. New York, NY; 2018.
    1. European Medicines Agency. Lyrica (pregabalin) Summary of Product Characteristics. London, UK; 2017.
    1. Alexander J Jr., Edwards RA, Brodsky M, et al. Using time series analysis approaches for improved prediction of pain outcomes in subgroups of patients with painful diabetic peripheral neuropathy. PLoS One. 2018;13(12):e0207120. doi:10.1371/journal.pone.0207120
    1. Alexander J Jr., Edwards RA, Manca L, et al. Dose titration of pregabalin in patients with painful diabetic peripheral neuropathy: simulation based on observational study patients enriched with data from randomized studies. Adv Ther. 2018;35(3):382–394. doi:10.1007/s12325-018-0664-6
    1. Alexander J, Edwards RA, Savoldelli A, et al. Integrating data from randomized controlled trials and observational studies to predict the response to pregabalin in patients with painful diabetic peripheral neuropathy. BMC Med Res Methodol. 2017;17(1):113. doi:10.1186/s12874-017-0389-2
    1. Alexander J Jr., Edwards RA, Brodsky M, et al. Assessing the value of time series real-world and clinical trial data vs. baseline-only data in predicting responses to pregabalin therapy for patients with painful diabetic peripheral neuropathy. Clin Drug Investig. 2019;39(8):775–786. doi:10.1007/s40261-019-00812-6
    1. Kazemi M, Moghimbeigi A, Kiani J, Mahjub H, Faradmal J. Diabetic peripheral neuropathy class prediction by multicategory support vector machine model: a cross-sectional study. Epidemiol Health. 2016;38:e2016011. doi:10.4178/epih/e2016011
    1. Rubio TG, Castillo AS. Evaluation of ANN and SVM for the classification and prediction of patients with diabetic neuropathy. XVIII Congreso Argentino de Ciencias de la Computación; Bahia Blanca, Argentina; 2012. Available from:
    1. Gujral S. Early diabetes detection using machine learning: a review. IJIRST. 2017;3(10):57–62.
    1. Kavakiotis I, Tsave O, Salifoglou A, Maglaveras N, Vlahavas I, Chouvarda I. Machine learning and data mining methods in diabetes research. Comput Struct Biotechnol J. 2017;15:104–116. doi:10.1016/j.csbj.2016.12.005
    1. Emir B, Johnson K, Kuhn M, Parsons B. Predictive modeling of response to pregabalin for the treatment of neuropathic pain using 6-week observational data: a spectrum of modern analytics applications. Clin Ther. 2017;39(1):98–106. doi:10.1016/j.clinthera.2016.11.015
    1. Ho TK. Random decision forest. 3rd International Conference on Document Analysis and Recognition; Montreal, QC, Canada; 1995.
    1. Alpaydin E. Introduction to Machine Learning. 2nd ed. Cambridge, MA: MIT Press; 2010.
    1. Bishop C. Neural Networks for Pattern Recognition. New York, NY: Oxford University Press; 1995.
    1. Broomhead DS, Lowe D. Radial Basis Functions, Multi-Variable Functional Interpolation and Adaptive Networks. UK: Royal Signals and Radar Establishment Malvern; 1988.
    1. Cortes C, Vapnik V. Support vector machine. Mach Learn. 1995;20(3):273–297. doi:10.1007/BF00994018
    1. Bezdek JC. Pattern Recognition With Fuzzy Objective Function Algorithms. New York, NY: Plenum Press; 1981.
    1. Gagliardi F. Instance-based classifiers applied to medical databases: diagnosis and knowledge extraction. Artif Intell Med. 2011;52(3):123–139. doi:10.1016/j.artmed.2011.04.002
    1. Lesser H, Sharma U, LaMoreaux L, Poole RM. Pregabalin relieves symptoms of painful diabetic neuropathy: a randomized controlled trial. Neurology. 2004;63(11):2104–2110. doi:10.1212/01.WNL.0000145767.36287.A1
    1. Richter RW, Portenoy R, Sharma U, Lamoreaux L, Bockbrader H, Knapp LE. Relief of painful diabetic peripheral neuropathy with pregabalin: a randomized, placebo-controlled trial. J Pain. 2005;6(4):253–260. doi:10.1016/j.jpain.2004.12.007
    1. Rosenstock J, Tuchman M, LaMoreaux L, Sharma U. Pregabalin for the treatment of painful diabetic peripheral neuropathy: a double-blind, placebo-controlled trial. Pain. 2004;110(3):628–638. doi:10.1016/j.pain.2004.05.001
    1. Freynhagen R, Strojek K, Griesing T, Whalen E, Balkenohl M. Efficacy of pregabalin in neuropathic pain evaluated in a 12-week, randomised, double-blind, multicentre, placebo-controlled trial of flexible- and fixed-dose regimens. Pain. 2005;115(3):254–263. doi:10.1016/j.pain.2005.02.032
    1. Arezzo JC, Rosenstock J, Lamoreaux L, Pauer L. Efficacy and safety of pregabalin 600 mg/d for treating painful diabetic peripheral neuropathy: a double-blind placebo-controlled trial. BMC Neurol. 2008;8:33. doi:10.1186/1471-2377-8-33
    1. Hoffman DL, Sadosky A, Dukes EM, Alvir J. How do changes in pain severity levels correspond to changes in health status and function in patients with painful diabetic peripheral neuropathy? Pain. 2010;149(2):194–201. doi:10.1016/j.pain.2009.09.017
    1. Satoh J, Yagihashi S, Baba M, et al. Efficacy and safety of pregabalin for treating neuropathic pain associated with diabetic peripheral neuropathy: a 14 week, randomized, double-blind, placebo-controlled trial. Diabet Med. 2011;28(1):109–116. doi:10.1111/j.1464-5491.2010.03152.x
    1. Tölle T, Freynhagen R, Versavel M, Trostmann U, Young JP Jr. Pregabalin for relief of neuropathic pain associated with diabetic neuropathy: a randomized, double-blind study. Eur J Pain. 2008;12(2):203–213. doi:10.1016/j.ejpain.2007.05.003
    1. Toelle RT, Varvara R, Nimour M, Emir B, Brasser M. Pregabalin in neuropathic pain related to DPN, cancer and back pain: analysis of a 6-week observational study. Open Pain J. 2012;5:1–11. doi:10.2174/1876386301205010001
    1. Baratloo A, Hosseini M, Negida A, El Ashal G. Part 1: simple definition and calculation of accuracy, sensitivity and specificity. Emerg (Tehran). 2015;3(2):48–49.
    1. Alexander J Jr., Edwards RA, Brodsky M, et al. Using time series analysis approaches for improved prediction of pain outcomes in subgroups of patients with painful diabetic peripheral neuropathy. PLoS One. 2018;13(12):e0207120. doi:10.1371/journal.pone.0207120
    1. Acion L, Kelmansky D, van der Laan M, Sahker E, Jones D, Arndt S. Use of a machine learning framework to predict substance use disorder treatment success. PLoS One. 2017;12(4):e0175383. doi:10.1371/journal.pone.0175383
    1. Allyn J, Allou N, Augustin P, et al. A comparison of a machine learning model with EuroSCORE II in predicting mortality after elective cardiac surgery: a decision curve analysis. PLoS One. 2017;12(1):e0169772. doi:10.1371/journal.pone.0169772
    1. Zheng T, Xie W, Xu L, et al. A machine learning-based framework to identify type 2 diabetes through electronic health records. Int J Med Inform. 2017;97:120–127. doi:10.1016/j.ijmedinf.2016.09.014

Source: PubMed

Sponsors and Collaborators

Medical Conditions

Drug Interventions

3
Subscribe