![[PDF]](/pb-assets/Icons/adobe-pdf-icon-1498649896243.png){kind=icon}
Abstract
For the analysis of competing risks data, three different types of hazard functions have been considered in the literature, namely the cause-specific hazard, the sub-distribution hazard, and the marginal hazard function. Accordingly, medical researchers can fit three different types of the Cox model to estimate the effect of covariates on each of the hazard function. While the relationship between the cause-specific hazard and the sub-distribution hazard has been extensively studied, the relationship to the marginal hazard function has not yet been analyzed due to the difficulties related to non-identifiability. In this paper, we adopt an assumed copula model to deal with the model identifiability issue, making it possible to establish a relationship between the sub-distribution hazard and the marginal hazard function. We then compare the two methods of fitting the Cox model to competing risks data. We also extend our comparative analysis to clustered competing risks data that are frequently used in medical studies. To facilitate the numerical comparison, we implement the computing algorithm for marginal Cox regression with clustered competing risks data in the R joint.Cox package and check its performance via simulations. For illustration, we analyze two survival datasets from lung cancer and bladder cancer patients.
References
| 1. | Cox, DR. Regression models and life-tables (with discussion). J Royal Stat Soc Ser B 1972; 34: 187–220. Google Scholar |
| 2. | Kalbfleisch, JD, Prentice, RL. The statistical analysis of failure time data. 2nd ed. New York, NY: John Wiley and Sons. Google Scholar |
| 3. | Fine, JP, Gray, RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 1999; 94: 548–560. Google Scholar | Crossref | ISI |
| 4. | Jeong, JH, Fine, J. Direct parametric inference for the cumulative incidence function. J Royal Stat Soc Ser C (Appl Stat) 2006; 55: 187–200. Google Scholar | Crossref | ISI |
| 5. | Zheng, M, Klein, JP. Estimates of marginal survival for dependent competing risks based on an assumed copula. Biometrika 1995; 82: 127–138. Google Scholar | Crossref | ISI |
| 6. | Escarela, G, Carriere, JF. Fitting competing risks with an assumed copula. Stat Meth Med Res 2003; 12: 333–349. Google Scholar | SAGE Journals | ISI |
| 7. | Braekers, R, Veraverbeke, N. A copula-graphic estimator for the conditional survival function under dependent censoring. Can J Stat 2005; 33: 429–447. Google Scholar | Crossref | ISI |
| 8. | Chen, YH. Semiparametric marginal regression analysis for dependent competing risks under an assumed copula. J Royal Stat Soc Ser B 2010; 72: 235–251. Google Scholar | Crossref |
| 9. | Lo, SM, Wilke, RA. A copula model for dependent competing risks. J Royal Stat Soc Series C (Appl Stat) 2010; 59: 359–376. Google Scholar | Crossref | ISI |
| 10. | Crowder, MJ. Classical competing risks. New York, NY: Chapman and Hall/CRC, 2001. Google Scholar | Crossref |
| 11. | Pintilie, M. Competing risks: a practical perspective. New York, NY: John Wiley & Sons, 2006. Google Scholar | Crossref |
| 12. | Bakoyannis, G, Touloumi, G. Practical methods for competing risks data: a review. Stat Meth Med Res 2012; 21: 257–272. Google Scholar | SAGE Journals | ISI |
| 13. | Ha, ID, Jeong, JH, Lee, Y. Statistical modelling of survival data with random effects: h-likelihood approach. Singapore: Springer, 2017. Google Scholar | Crossref |
| 14. | Emura, T, Chen, YH. Analysis of survival data with dependent censoring, copula-based approaches, JSS research series in statistics. Singapore: Springer, 2018. Google Scholar | Crossref |
| 15. | Rivest, LP, Wells, MT. A martingale approach to the copula-graphic estimator for the survival function under dependent censoring. J Multivariate Analys 2001; 79: 138–155. Google Scholar | Crossref | ISI |
| 16. | Emura, T. joint.Cox: joint frailty-copula models for tumour progression and death in meta-analysis, CRAN, version 3.6. 2019. Google Scholar |
| 17. | Cox, DR, Oakes, D. Analysis of survival data. New York, NY: CRC Press, 1984. Google Scholar |
| 18. | de Uña-Álvarez, J, Veraverbeke, N. Generalized copula-graphic estimator. Test 2013; 22: 343–360. Google Scholar | Crossref |
| 19. | Emura, T, Chen, YH. Gene selection for survival data under dependent censoring, a copula-based approach. Stat Meth Med Res 2016; 25: 2840–2857. Google Scholar | SAGE Journals | ISI |
| 20. | Klein, JP, Moeschberger, ML. Survival analysis: techniques for censored and truncated data. New York, NY: Springer, 2003. Google Scholar | Crossref |
| 21. | Crowder, MJ. Multivariate survival analysis and competing risks. New York, NY: CRC Press, 2012. Google Scholar | Crossref |
| 22. | Tsiatis, A. A nonidentifiability aspect of the problem of competing risks. Proc Natn Acad Sci USA 1975; 72: 20–22. Google Scholar | Crossref | Medline | ISI |
| 23. | Nelsen, RB. An introduction to copulas. 2nd ed. New York, NY: Springer Series in Statistics, Springer-Verlag, 2006. Google Scholar |
| 24. | Rotolo, F, Legrand, C, Van Keilegom, I. A simulation procedure based on copulas to generate clustered multi-state survival data. Comput Meth Programs Biomed 2013; 109: 305–312. Google Scholar | Crossref | Medline |
| 25. | Gray, RJ. cmprsk: subdistribution analysis of competing risks, CRAN, version 2.2-7. 2017. Google Scholar |
| 26. | Shih, J-H, Emura, T. Likelihood-based inference for bivariate latent failure time models with competing risks under the generalized FGM copula. Computat Stat 2018; 33: 1293–1323. Google Scholar | Crossref |
| 27. | Heckman, JJ, Honoré, BE. The identifiability of the competing risks model. Biometrika 1989; 76: 325–330. Google Scholar | Crossref | ISI |
| 28. | Abbring, JH, Van den Berg, GJ. The identifiability of the mixed proportional hazards competing risks model. J Royal Stat Soc: Series B (Statistical Methodology) 2003; 65: 701–710. Google Scholar | Crossref | ISI |
| 29. | Scheike, TH, Zhang, MJ. Flexible competing risks regression modeling and goodness-of-fit. Lifetime Data Analys 2008; 14: 464. Google Scholar | Crossref | Medline |
| 30. | Sfumato, P, Filleron, T, et al. Goftte: A R package for assessing goodness-of-fit in proportional (sub) distributions hazards regression models. Comput Meth Program Biomed 2019; 177: 269–275 Google Scholar | Crossref | Medline |
| 31. | Chen, HY, Yu, SL, Chen, CH, et al. A five-gene signature and clinical outcome in non–small-cell lung cancer. New Engl J Med 2007; 356: 11–20. Google Scholar | Crossref | Medline | ISI |
| 32. | Rondeau, V, Pignon, JP, Michiels, S. A joint model for dependence between clustered times to tumour progression and deaths: a meta-analysis of chemotherapy in head and neck cancer. Stat Meth Med Res 2015; 24: 711–729. Google Scholar | SAGE Journals | ISI |
| 33. | Lee, M, Ha, ID, Lee, Y. Frailty modeling for clustered competing risks data with missing cause of failure. Stat Meth Med Res 2017; 26: 356–373. Google Scholar | SAGE Journals | ISI |
| 34. | Emura, T, Nakatochi, M, Murotani, K, et al. A joint frailty-copula model between tumour progression and death for meta-analysis. Stat Meth Med Res 2017; 26: 2649–2666. Google Scholar | SAGE Journals | ISI |
| 35. | Peng, M, Xiang, L, Wang, S. Semiparametric regression analysis of clustered survival data with semi-competing risks. Computat Stat Data Analys 2018; 124: 53–70. Google Scholar | Crossref |
| 36. | Brazauskas, R, Le-Rademacher, J. Methods for generating paired competing risks data. Comput Meth Programs Biomed 2016; 135: 199–207. Google Scholar | Crossref | Medline |
| 37. | Ha, ID, Christian, NJ, Jeong, JH, et al. Analysis of clustered competing risks data using subdistribution hazard models with multivariate frailties. Stat Meth Med Res 2016; 25: 2488–2505. Google Scholar | SAGE Journals | ISI |
| 38. | Emura, T, Matsui, S, Rondeau, V. Survival analysis with correlated endpoints, joint frailty-copula models, JSS Research series in statistics. Singapore: Springer, 2019. Google Scholar | Crossref |
| 39. | Ha, ID, Noh, M, Kim, J, et al. frailtyHL: frailty models using h-likelihood, CRAN, version 2.1. 2018, http://CRAN.Rproject.org/package=frailtyHL. Google Scholar |
| 40. | Ramsay, J. Monotone regression spline in action. Stat Sci 1988; 3: 425–461. Google Scholar | Crossref |
| 41. | Sylvester, RJ, van der Meijden, AP, Oosterlinck, W, et al. Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials. Eur Urology 2006; 49: 466–477. Google Scholar | Crossref | Medline | ISI |
| 42. | Scheike T. timereg: flexible regression models for survival data, CRAN, version 1.9.4. 2019. Google Scholar |
| 43. | Emura, T, Matsui, S, Chen, HY. compound.Cox: univariate feature selection and compound covariate for predicting survival. Comput Meth Programs Biomed 2019; 168: 21–37. Google Scholar | Crossref | Medline |
| 44. | Park, J, Bakoyannis, G, Yiannoutsos, C. Semiparametric competing risks regression under interval censoring using the R package intccr. Comput Meth Programs Biomed 2019; 173: 167–176. Google Scholar | Crossref | Medline |
| 45. | Sabatier, R, Finetti, P, Adelaide, J, et al. Down-regulation of ECRG4, a candidate tumor suppressor gene, in human breast cancer. PLoS One 2011; 6: e27656. Google Scholar | Crossref | Medline | ISI |
