Flexible modeling of the hazard rate and treatment effects in long-term survival studies

Abstract

The effects of predictors on time to failure may be difficult to assess in cancer studies with longer follow-up, as the commonly used assumption of proportionality of hazards holding over an extended period is often questionable. Motivated by a long-term prostate cancer clinical trial, we contrast and compare four powerful methods for estimation of the hazard rate. These four methods allow for varying degrees of smoothness as well as covariates with effects that vary over time. We pay particular attention to an extended multiresolution hazard estimator, which is a flexible, semi-parametric, Bayesian method for joint estimation of predictor effects and the hazard rate. We compare the results of the extended multiresolution hazard model to three other commonly used, comparable models: Aalen’s additive model, Kooperberg’s hazard regression model, and an extended Cox model. Through simulations and the analysis of a large-scale randomized prostate cancer clinical trial, we use the different methods to examine patterns of biochemical failure and to estimate the time-varying effects of androgen deprivation therapy treatment and other covariates.

Keywords

Biochemical failure hazard rate multiresolution hazard non-proportional hazards prostate cancer survival analysis

Get full access to this article

View all access options for this article.

References

Pilepich

Winter

John

et al.

Phase III Radiation Therapy Oncology Group (RTOG) trial 86-10 of androgen deprivation adjuvant to definitive radiotherapy in locally advanced carcinoma of the prostate. Int J Radiat Oncol Biol Phys 2001; 50: 1243–1252.

Pilepich

Winter

Lawton

et al.

Androgen suppression adjuvant to definitive radio-therapy in prostate carcinoma: long-term results of phase III RTOG 85-31. Int J Radiat Oncol Biol Phys 2005; 61: 1285–1290.

Horwitz EM, et al. Ten-year follow-up of radiation therapy oncology group protocol 92-02: a phase III trial of the duration of elective androgen deprivation in locally advanced prostate cancer. Journal of Clinical Oncology2008; 26(15): 2497–2504.

Albertsen

Hanley

Fine

. 20-year outcomes following conservative management of clinically localized prostate cancer. J Am Med Assoc 2005; 293: 2095–2101.

Nguyen

Levy

Lee

et al.

Long-term outcomes for men with high-risk prostate cancer treated definitively with external beam radiotherapy with or without androgen deprivation. Cancer 2013; 119: 3265–3271.

Taira

Merrick

Butler

et al.

Time to failure after definitive therapy for prostate cancer: implications for importance of aggressive local treatment. J Contemp Brachytherapy 2013; 5: 215–221.

Amling

Blute

Bergstralh

et al.

Long-term hazard of progression after radical prostatectomy for clinically localized prostate cancer: continued risk of biochemical failure after 5 years. J Urol 2000; 164: 101–105.

Dillioglugil

Leibman

Kattan

et al.

Hazard rates for progression after radical prostatectomy for clinically localized prostate cancer. Adult Urol 1997; 50: 93–99.

Hanlon

Hanks

. Failure patterns and hazard rates for failure suggest the cure of prostate cancer by external beam radiation. Adult Urol 2000; 55: 725–729.

10.

Walz

Chun

Klein

et al.

Risk-adjusted hazard rates of biochemical recurrence for prostate cancer patients after radical prostatectomy. Eur Urol 2008; 55: 412–421.

11.

Bouman

Dukic

Meng

. Bayesian multiresolution hazard model with application to an AIDS reporting delay study. Statistica Sinica 2005; 15: 325–357.

12.

Bouman

Dignam

Dukic

et al.

A multiresolution hazard model for multi-center survival studies: application to tamoxifen treatment in early stage breast cancer. J Am Stat Assoc 2007; 102: 1145–1157.

13.

Dignam

Dukic

Anderson

et al.

Hazard of recurrence and adjuvant treatment effects over time in lymph node-negative breast cancer. Breast Cancer Res Treat 2009; 116: 595–602.

14.

Dukic

Dignam

. Bayesian hierarchical multiresolution hazard model for the study of time-dependent failure patterns in early stage breast cancer. Bayesian Anal 2007; 2: 591–610.

15.

Dignam

Dukic

Anderson

et al.

Hazard of recurrence and adjuvant treatment effects over time in lymph node-negative breast cancer. Breast Cancer Res Treat 2009; 116: 595–602.

16.

Chen

Hagar

Dignam

et al.

Pruned Multiresolution Hazard (PMRH) models for time-to-event data. Bayesian Analysis 2014. In Review.

17.

Hagar

Albers

Pivovarov

et al.

Survival analysis with Electronic Health Record data: experiments with chronic kidney disease. Stat Anal Data Mining 2014; 7: 385–403.

18.

R Development Core Team. R: A language and environment for statistical computing, Vienna, Austria: R Foundation for Statistical Computing, 2008.

19.

Aalen

. A linear regression model for the analysis of life times. Stat Med 1989; 8: 907–925.

20.

Aalen

. Further results on the non-parametric linear model in survival analysis. Stat Med 1993; 12: 1569–1588.

21.

Kooperberg

Stone

Truong

. Hazard regression. J Am Stat Assoc 1995; 90: 78–94.

22.

Martinussen

Scheike

. Dynamic regression models for survival data, New York, NY: Springer, 2006.

23.

Hagar Y, Chen Y and Dukic V. MRH package in R, http://cran.r-project.org/web/packages/MRH/index.html (2014, accessed 4 January 2017).

24.

Roach

III Hanks

Thames

Jr et al.

Defining biochemical failure following radiotherapy with or without hormonal therapy in men with clinically localized prostate cancer: recommendations of the RTOG-ASTRO Phoenix consensus conference. Int J Radiat Oncol Biol Phys 2006; 65: 965–974.

25.

Cooner

Mosley

Rutherford

et al.

Prostate cancer detection in clinical urological practice by ultrasonography, digital rectal examination and prostate specific antigen. J Urol 1990; 143: 1146–1152.

26.

Catalona

Smith

Ratliff

et al.

Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. N Engl J Med 1991; 324: 1156–1161.

27.

Brawer

Chetner

Beatie

et al.

Screening for prostatic carcinoma with prostate specific antigen. J Urol 1992; 147: 841–845.

28.

Radiation Therapy Oncology Group. RTOG 9202 protocol information, www.rtog.org/ClinicalTrials/ProtocolTable/StudyDetails.aspx?study=9202, 2014 (accessed 11 March 2014).

29.

Epstein

Albsbrook

Jr Amin

et al.

The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason grading of prostatic carcinoma. Am J Surg Pathol 2005; 29: 1228–1242.

30.

American Joint Committee on Cancer. What is cancer staging? http://cancerstaging.org/references-tools/Pages/What-is-Cancer-Staging.aspx (accessed 1 October 2014).

31.

Held-Warmkessel

. Contemporary issues in prostate cancer, Sudbury, MA: Jones and Bartlett Publishers, 2006.

32.

Grambsch

Therneau

. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 1994; 81: 515–526.

33.

Andersen

Borgan

Gill

et al.

Statistical methods based on counting processes, Berlin: Springer-Verlag, 1993.

34.

Sinha

Dey

. Semiparameteric Bayesian analysis of survival data. J Am Stat Assoc 1997; 92: 1195–1212.

35.

Müller

Rodriguez

. Nonparametric Bayesian inference, Beachwood, Ohio and Alexandria, Virginia, USA: Institute of Mathematical Statistics and American Statistical Association, 2013.

36.

Holford

. Life tables with concomitant information. Biometrics 1976; 32: 587–597.

37.

Holford

. The analysis of rates and of survivorship using log-linear models. Biometrics 1980; 36: 299–305.

38.

Laird

Olivier

. Covariance analysis of censored survival data using log-linear analysis techniques. J Am Stat Assoc 1981; 76: 231–240.

39.

Taulbee

. A general model for the hazard rate with covariables. Biometrics 1979; 35: 439–450.

40.

Dabrowska

Doksum

Song

. Graphical comparison of cumulative hazards for two populations. Biometrika 1989; 76: 763–773.

41.

Parzen

Wei

Ying

. Simultaneous confidence intervals for the difference of two survival functions. Scand J Stat 1997; 24: 309–314.

42.

McKeague

Zhao

. Simultaneous confidence bands for ratios of survival functions via empirical likelihood. Stat Probab Lett 2002; 60: 405–415.

43.

Wei

Schaubel

. Estimating cumulative treatment effects in the presence of nonproportional hazards. Biometrics 2008; 64: 724–732.

44.

Dong

Matthews

. Empirical likelihood for cumulative hazard ratio estimation with covariate adjustment. Biometrics 2012; 68: 408–418.

45.

Buckley

James

. Linear regression with censored data. Biometrika 1979; 66: 429–436.

46.

Kalbfleisch

Prentice

. The statistical analysis of failure time data, Chichester: Wiley, 2002.

47.

Berry

Natarajan

et al.

Bayesian survival analysis with nonproportional hazards. J Am Stat Assoc 2004; 99: 515–526.

48.

Nieto-Barajas

. Bayesian semiparametric analysis of short- and long-term hazard ratios with covariates. Comput Stat Data Anal 2014; 71: 477–490.

49.

Hennerfeind

Brezger

Fahrmeir

. Geoadditive survival models. J Am Stat Assoc 2006; 101: 1065–1075.

50.

Cai

Meyer

. Bayesian semi parametric modeling of survival data based on mixtures of B-spline distributions. Comput Stat Data Anal 2011; 55: 1260–1272.

51.

De Iorio

Johnson

Müller

et al.

Bayesian nonparametric nonproportional hazards survival modeling. Biometrics 2009; 65: 762–771.

52.

Ferguson

. Prior distributions on spaces of probability measures. Ann Stat 1974; 2: 615–629.

53.

Lavine

. Some aspects of Polya tree distributions for statistical modelling. Ann Stat 1992; 20: 1222–1235.

54.

Muliere

Walker

. A Bayesian non-parametric approach to survival analysis using Polya trees. Scand J Stat 1997; 24: 331–340.

55.

Hanson

Johnson

. Modeling regression error with a mixture of Polya trees. J Am Stat Assoc 2002; 97: 1020–1033.

56.

Hanson

. Inference for mixtures of finite Polya tree models. J Am Stat Assoc 2006; 101: 1548–1565.

57.

Zhao

Hanson

Carlin

. Mixtures of Polya trees for flexible spatial frailty survival modelling. Biometrika 2009; 96: 263–276.

58.

Zhao

Hanson

. Spatially dependent Polya tree modeling for survival data. Biometrics 2011; 67: 391–403.

59.

Wong

. Optional Polya tree and Bayesian inference. Ann Stat 2010; 38: 1433–1459.

60.

Nieto-Barajas

Müller

. Rubbery Polya tree. Scand J Stat 2012; 39: 166–184.

61.

Geman

. Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images. IEEE Trans Pattern Anal Mach Intell 1984; 6: 721–741.

62.

Komárek

Lesaffre

. Bayesian accelerated failure time model for correlated interval-censored data with a normal mixture as error distribution. Statistica Sinica 2007; 17: 549–569.

63.

Komárek A. bayessurv package in R, http://cran.r-project.org/web/packages/bayesSurv/bayesSurv.pdf (2015, accessed 4 January 2017).

64.

Jara A, Hanson T, Quintana F, et al. Dppackage package in R. cran.r-project.org/web/packages/DPpackage/DPpackage.pdf (2012, accessed 4 January 2017).

65.

Yang

Prentice

. Semiparametric analysis of short-term and long-term hazard ratios with two sample survival data. Biometrika 2005; 92: 1–17.

66.

Fleming

Harrington

. Counting processes and survival analysis, Hoboken, NJ: John Wiley & Sons, 2011.

67.

Andersen

Borgan

Hjort

et al.

Counting process models for life history data: a review with discussion and reply. Scand J Stat 1985; 12: 97–158.

68.

Therneau TM. A package for survival analysis in s, http://CRAN.R-project.org/package=survival (accessed 4 January 2017).

69.

Schwarz

. Estimating the dimension of a model. Ann Stat 1978; 6: 461–464.

70.

Kooperberg C. polspline: polynomial spline routines. https://cran.r-project.org/web/packages/polspline/index.html, 2016 (accessed 4 January 2017).

71.

Scheike T. timereg package in R. http://cran.r-project.org/web/packages/timereg/timereg.pdf (2014, accessed 4 January 2017).

72.

Geweke

. Evaluating the accuracy of sampling-based approaches to calculating posterior moments in Bayesian Statistics 4, Oxford, UK: Clarendon Press, 1992.

73.

Gelman

Rubin

. Inference from iterative simulation using multiple sequences. Stat Sci 1992; 7: 457–511.

74.

Brooks

Gelman

. General methods for monitoring convergence of iterative simulations. J Comput Graph Stat 1998; 7: 434–455.

75.

Prentice

Kalbfleisch

Peterson

Jr et al.

The analysis of failure times in the presence of competing risks. Biometrics 1978; 34: 541–554.

76.

Akaike

. A new look at the statistical model identification. IEEE Trans Pattern Anal Mach Intell 1974; 19: 716–723.

77.

Kullback

Leibler

. On information and sufficiency. Ann Math Stat 1951; 22: 79–86.

78.

Kaplan

Meier

. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.