Abstract
Clinicians often have to make treatment decisions based on the likelihood that individual patients will receive benefit. This information has to be derived from appropriate studies and provided in a way that easily transforms from the statistical model into clinical practice. This chapter discusses four methodological strategies to provide information for treatment decisions: subgroup analysis, regression models, models for potential outcome, and prediction models. Based on how statistical techniques are able to provide information for stratified, precision, or individualized medicine we present the challenges that exist in analysis, reporting and applying those results. We also identify relevant ethical issues. There is still a substantial gap between formal statistical results on patient treatment interaction and ways in which they can be presented to clinicians as easy to use tools in clinical decision-making. At the moment, most energy is spent in the discovery of biomarkers which help to grasp patient treatment interaction. Less energy is spent in corresponding validation studies and the process of translating statistical results into clinical practice. Furthermore, clinical studies that provide evidence for successful translation of predictive biomarkers into clinical practice are needed.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Anderson, T. W. (1957). Maximum likelihood estimates for a multivariate Normal distribution when some observations are missing. Journal of the American Statistical Association, 52(278), 200–203. https://doi.org/10.1080/01621459.1957.10501379.
Bhandari, M., Guyatt, G., Walter, S. D., Tornetta, P., Schemitsch, E. H., Swiontkowski, M., & Sanders, D. (2008). Randomized trial of reamed and Unreamed intramedullary nailing of Tibial shaft fractures. The Journal of Bone and Joint Surgery-American Volume, 90(12), 2567–2578. https://doi.org/10.2106/jbjs.g.01694.
Bodemer, N., Meder, B., & Gigerenzer, G. (2014). Communicating relative risk changes with baseline risk. Medical Decision Making, 34(5), 615–626. https://doi.org/10.1177/0272989x14526305.
Breiman, L. (2001). Statistical modeling: The two cultures (with comments and a rejoinder by the author). Statistical Science, 16(3), 199–231. https://doi.org/10.1214/ss/1009213726.
Callegaro, A., Spiessens, B., Dizier, B., Montoya, F. U., & van Houwelingen, H. C. (2016). Testing interaction between treatment and high-dimensional covariates in randomized clinical trials. Biometrical Journal, 59(4), 672–684. https://doi.org/10.1002/bimj.201500194.
Cameron, A. C., & Trivedi, P. K. (1998). Regression analysis of count data. New York: Cambridge University Press. https://doi.org/10.1017/cbo9780511814365.
Casella, G. (2008). Statistical design. New York: Springer. https://doi.org/10.1007/978-0- 387-75965-4.
Collett, D. (2002). Modelling binary data (2nd ed.). London: Chapman Hall/CRC Press.
Collett, D. (2014). Modelling survival data in medical research. Boca Raton: Chapman Hall/CRC Press.
Collins, G. S., Reitsma, J. B., Altman, D. G., & Moons, K. G. M. (2015). Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD). Annals of Internal Medicine, 162(10), 735–736. https://doi.org/10.7326/l15-5093-2.
Cuzick, J. (2005). Forest plots and the interpretation of subgroups. The Lancet, 365(9467), 1308. https://doi.org/10.1016/s0140-6736(05)61026-4.
Engelhardt, A., Shen, Y. M., & Mansmann, U. (2016). Constructing an ROC curve to assess a treatment-predictive continuous biomarker. Studies in Health Technology and Informatics, 228, 745–749.
Gadbury, G. L., & Iyer, H. K. (2000). Unit-treatment interaction and its practical consequences. Biometrics, 56(3), 882–885. https://doi.org/10.1111/j.0006-341x.2000.00882.x.
Gadbury, G. L., Iyer, H. K., & Allison, D. B. (2001). Evaluating subject-treatment interaction when comparing two treatments. Journal of Biopharmaceutical Statistics, 11(4), 313–333. https://doi.org/10.1081/BIP-120008851.
Gadbury, G. L., Iyer, H. K., & Albert, J. M. (2004). Individual treatment effects in randomized trials with binary outcomes. Journal of Statistical Planning and Inference, 121(2), 163–174. https://doi.org/10.1016/s0378-3758(03)00115-0.
Gail, M., & Simon, R. (1985). Testing for qualitative interactions between treatment effects and patient subsets. Biometrics, 41(2), 361–372. https://doi.org/10.2307/2530862.
Gentles, S. J., Stacey, D., Bennett, C., Alshurafa, M., & Walter, S. D. (2013). Factors explaining the heterogeneity of effects of patient decision aids on knowledge of outcome probabilities: A systematic review sub-analysis. Systematic Reviews, 2(1). https://doi.org/10.1186/2046-4053-2-95.
Huang, Y., Gilbert, P. B., & Janes, H. (2012). Assessing treatment-selection markers using a potential outcomes framework. Biometrics, 68(3), 687–696. https://doi.org/10.1111/j.1541-0420.2011.01722.x.
Kasenda, B., Schandelmaier, S., Sun, X., von Elm, E., You, J., Blumle, A., Tomonaga, Y., et al. (2014). Subgroup analyses in randomised controlled trials: Cohort study on trial protocols and journal publications. BMJ, 349(Jul16 1), g4539–g4539. https://doi.org/10.1136/bmj.g4539.
Laubender, R. P., & Mansmann, U. (2014). Estimating individual treatment effects from responses and a predictive biomarker in a parallel group RCT. Department of Statistics: Technical Reports 176. https://epub.ub.uni-muenchen.de/22207/1/TR176.pdf.
Lord, F. M. (1955a). Equating test ScoresA maximum likelihood solution. Psychometrika, 20(3), 193–200. https://doi.org/10.1007/bf02289016.
Lord, F. M. (1955b). Estimation of parameters from incomplete data. Journal of the American Statistical Association, 50(271), 870. https://doi.org/10.2307/2281171.
McGinn, T. G., Guyatt, G. H., Wyer, P. C., Naylor, C. D., Stiell, I. G., Richardson, W. S., & for the Evidence-Based Medicine Working Group. (2000). Users guides to the medical literature. JAMA, 284(1), 79. https://doi.org/10.1001/jama.284.1.79.
Neyman, J. (1923). Sur Les Applications de La Theorie Des Probabilites Aux Experiences Agricoles: Essai Des Principes. Statistical Science, 5, 463–472. (Dabrowska DM and Speed TP, Translators).
Olotu, A., Fegan, G., Wambua, J., Nyangweso, G., Awuondo, K. O., Leach, A., Lievens, M., et al. (2013). Four-year efficacy of RTS,S/AS01E and its interaction with malaria exposure. New England Journal of Medicine, 368(12), 1111–1120. https://doi.org/10.1056/nejmoa1207564.
Rao, C. R., Toutenburg, H., & Shalabh, H. (2008). Linear models and generalizations: Least squares and alternatives (3rd ed.). Berlin: Springer Series in Statistics.
Rothwell, P. M., Eliasziw, M., Gutnikov, S. A., Fox, A. J., Taylor, D. W., Mayberg, M. R., Warlow, C. P., & Barnett, H. J. M. (2003). Analysis of pooled data from the randomised controlled trials of endarterectomy for symptomatic carotid stenosis. The Lancet, 361(9352), 107–116. https://doi.org/10.1016/s0140-6736(03)12228-3.
Rothwell, P. M., Mehta, Z., Howard, S. C., Gutnikov, S. A., & Warlow, C. P. (2005). From subgroups to individuals: General principles and the example of carotid endarterectomy. The Lancet, 365(9455), 256–265. https://doi.org/10.1016/s0140-6736(05)17746-0.
Royston, P., & Sauerbrei, W. (2008). Multivariable model-building: A pragmatic approach to regression anaylsis based on fractional polynomials for modelling continuous variables (Vol. 777). New York: Wiley.
Royston, P., Sauerbrei, W., & Ritchie, A. (2004). Is treatment with interferon-α effective in all patients with metastatic renal carcinoma? A new approach to the investigation of interactions. British Journal of Cancer, 90(4), 794–799. https://doi.org/10.1038/sj.bjc.6601622.
Rubin, D. B. (1974). Estimating causal effects of treatments in randomized and nonrandomized studies. Journal of Educational Psychology, 66(5), 688–701. https://doi.org/10.1037/h0037350.
Senn, F., & Harrell, S. (1997). On wisdom after the event. Journal of Clinical Epidemiology, 50(7), 749–751. https://doi.org/10.1016/s0895-4356(97)00023-1.
Shen, Y.-M., Le, L. D., Wilson, R., & Mansmann, U. (2017). Graphical presentation of patient-treatment interaction elucidated by continuous biomarkers. Methods of Information in Medicine, 56(01), 13–27. https://doi.org/10.3414/me16-01-0019.
Sun, X., Briel, M., Walter, S. D., & Guyatt, G. H. (2010). Is a subgroup effect believable? Updating criteria to evaluate the credibility of subgroup analyses. BMJ, 340, c117–c117. https://doi.org/10.1136/bmj.c117.
Ternès, N., Rotolo, F., Heinze, G., & Michiels, S. (2016). Identification of biomarker-by-treatment interactions in randomized clinical trials with survival outcomes and high-dimensional spaces. Biometrical Journal, 59(4), 685–701. https://doi.org/10.1002/bimj.201500234.
Thall, P. F., & Vail, S. C. (1990). Some covariance models for longitudinal count data with Overdispersion. Biometrics, 46(3), 657–671. https://doi.org/10.2307/2532086.
Tian, L., Alizadeh, A. A., Gentles, A. J., & Tibshirani, R. (2014). A simple method for estimating interactions between a treatment and a large number of covariates. Journal of the American Statistical Association, 109(508), 1517–1532. https://doi.org/10.1080/01621459.2014.951443.
Vandenbroucke, J. P., Broadbent, A., & Pearce, N. (2016). Causality and causal inference in epidemiology: The need for a pluralistic approach. International Journal of Epidemiology, 45(6), 1776–1786. https://doi.org/10.1093/ije/dyv341.
Vickers, A. J., Kattan, M. W., & Sargent, D. J. (2007). Method for evaluating prediction models that apply the results of randomized trials to individual patients. Trials, 8(1). https://doi.org/10.1186/1745-6215-8-14.
Wegwarth, O. (2012). Do physicians understand cancer screening statistics? A national survey of primary care physicians in the United States. Annals of Internal Medicine, 156(5), 340–349. https://doi.org/10.7326/0003-4819-156-5-201203060-00005.
Zikmund-Fisher, B. J. (2012). The right tool is what they need, not what we have. Medical Care Research and Review, 70(1_suppl), 37S–S. https://doi.org/10.1177/1077558712458541.
Zikmund-Fisher, B. J., Fagerlin, A., & Ubel, P. A. (2010). Risky feelings: Why a 6% risk of Cancer does not always feel like 6%. Patient Education and Counseling, 81(December), S87–S93. https://doi.org/10.1016/j.pec.2010.07.041.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Statistical Excursus: Cox Regression and Logistic Regression
Statistical Excursus: Cox Regression and Logistic Regression
The Cox proportional Hazards model quantifies the effect of a treatment (G) and a biomarker (X) on an event time T by specifying the survival probability. P[T > t|X, G] represents the probability to survive the time t given biomarker measurement X and treatment G. The proportional hazards model is given by
Here α0(s) describes an unspecific baseline hazard and exp{ β G · G + β X · X+ β GX · G · X} quantifies the hazards ratio between subjects treated by IN (G = 1) versus subjects treated by ST (G = 0) and presenting a biomarker measurement of X. In a more general sense the model can also handle nonlinear influence of the biomarker on outcome by
Here f GX(X) = 0 for G = 0 and f GX(X) = h X(X) for G = 1. Using the survival function of the baseline group P[T > t |X = 0, G = 0] = P 0[T > t], the above expression is equivalent to
The term β G · G + f X(X) + f GX (X) or β G · G + β X · X+ β GX · G · X are called linear predictors and written as LP(X,G). Therefore the above formulae translate to
and
The logistic regression quantifies the effect of a biomarker measured by X and treatment defined by G on a binary outcome Y using the model:
The linear predictor is different from the linear predictor of the Cox regression because it also contains a baseline term β0:
or
The value exp{LP(X, 1)} quantifies the odds ratio (OR) of a person showing a biomarker measurement of value X and being treated by the innovative treatment versus a person showing a biomarker measurement of value X = 0 and being treated by the standard. The odds ratio for a person with biomarker measurement X = x 0 between both treatments is
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Mansmann, U., Boulesteix, AL. (2020). Modelling Individual Response to Treatment and Its Uncertainty:A Review of Statistical Methods and Challenges for Future Research. In: LaCaze, A., Osimani, B. (eds) Uncertainty in Pharmacology. Boston Studies in the Philosophy and History of Science, vol 338. Springer, Cham. https://doi.org/10.1007/978-3-030-29179-2_14
Download citation
DOI: https://doi.org/10.1007/978-3-030-29179-2_14
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-29178-5
Online ISBN: 978-3-030-29179-2
eBook Packages: Religion and PhilosophyPhilosophy and Religion (R0)