Advertisement

Using PSA to detect prostate cancer onset: An application of Bayesian retrospective and prospective changepoint identification

  • Elizabeth H. Slate
  • Larry C. Clark
Part of the Lecture Notes in Statistics book series (LNS, volume 140)

Abstract

Serum prostate-specific antigen (PSA) concentrations are now widely used to aid in the detection of prostate cancer. When prostate cancer is present, PSA levels typically increase. But a number of benign conditions will also cause elevated PSA levels and, conversely, prostate cancer has been diagnosed in the absence of raised PSA.

We analyze one of the most extensive data sets currently available for longitudinal PSA readings, obtained by an historical prospective study of frozen serum samples from the Nutritional Prevention of Cancer Trial (Clark et al.1996). These data consist of serial readings for over 1200 men taken at approximate six-month intervals over an 11 year period.

We fit a fully Bayesian hierarchical changepoint model to the longitudinal PSA readings. Our ojectives include better understanding the natural history of PSA levels in patients who have and have not been diagnosed with prostate cancer and identifying subject-specific changepoints that are indicative of cancer onset. With the goal of accurate early detection, we perform a prospective sequential analysis to compare several diagnostic rules, including a rule based on the posterior distribution of individual changepoints.

Keywords

Prostate Cancer Receiver Operating Characteristic Curve Posterior Distribution Benign Prostatic Hyperplasia Linear Mixed Effect Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Breslow, N.E. and Lin, X. (1995). Bias correction in generalised linear mixed models with a single component of dispersionBiometrika 82, 81–91.MathSciNetzbMATHCrossRefGoogle Scholar
  2. Carlin, B.P., Gelfand, A.E. and Smith, A.F.M. (1992). Hierarchical Bayesian analysis of changepoint problems.Appl. Statist 41, No. 2, 389–405.zbMATHCrossRefGoogle Scholar
  3. Carter, H.B., Pearson, J.D., Metter, E.J., Brant, L.J., Chan, D.W., Andres, R., Fozard, J.L. and Walsh, P.C. (1992a). Longitudinal evaluation of prostate-specific antigen levels in men with and without prostate disease.J. Amer. Med. Assoc. 267, No. 16, 2215–2220.CrossRefGoogle Scholar
  4. Carter, H.B., Morrell, C.H., Pearson, J.D., Brant, L.J., Plato, C.C., Metter, E.J., Chan, E.W., Fozard, J.L. and Walsh, P.C. (1992b). Estimation of prostatic growth using serial prostate-specific antigen measurements in men with and without prostate diseaseCancer Research 52, 2232–3328.Google Scholar
  5. Carter, H.B., Epstein, J.I., Chan, D.W., Fozard, J.L. and Pearson, J.D. (1997). Recommended prostat-specific antigen testing intervals for the detection of curable prostate cancer J. Amer. Med. Assoc. 277, 1456–1460.CrossRefGoogle Scholar
  6. Catalona, W.J., Smith, D.S., Ratliff, T.L. and Basler, J.W. (1993). Detection of organ-confined prostate cancer is increased through prostate-specific antigen-based screening. J. Amer. Med. Assoc. 270, No. 8, 948–954.CrossRefGoogle Scholar
  7. Catalona, W.J., Smith, D.S., Ratliff, T.L., Dodds, K.M., Copien, D.E., Yuan, J.J., Petros, J.A. and Andriole, G.L. (1991). Measurement of prostate-specific antigen in serum as a screening test for prostate cancer. New Engl. J. Med. 324, No. 17,1156–1161.CrossRefGoogle Scholar
  8. Clark, L.C., Combs, G. F., Turnbull, B.W., Slate, E.H.. Chalker, D.K., Chow, J., Davis, L.S., Glover, R.A., Graham, G.F., Gross, E.G., Krongrad, A., Lesher, J.L., Park, H.K., Sanders, B.B., Smith, C.L., Taylor, J.R. and the Nutritional Prevention of Cancer Study Group (1996). Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin: A randomized clinical trial J. Amer. Med. Assoc. 276, No. 24,1957–1963.CrossRefGoogle Scholar
  9. Cleveland, W.S., and Devlin, S.J. (1988). Locally-weighted Regression: An Approach to Regression Analysis by Local FittingJ. Amer. Statist. Assoc. 83, 596–610.zbMATHCrossRefGoogle Scholar
  10. Cronin, K.A. (1995). Detection of Changepoints in Longitudinal Data Ph.D. Thesis, School of Operations Research, Cornell University.Google Scholar
  11. Cronin, K.A., Slate, E.H., Turnbull, B.W. and Wells, M.T. (1994). Using the Gibbs sampler to detect changepoints: application to PSA as a longitudinal marker for prostate cancer Computing Science and Statistics 26, eds. J. Sall and A. Lehman, Interface Foundation of North America, pp. 314–318.Google Scholar
  12. Emir, B., Wieand, S., Su, J.Q. and Cha, S. (1995). The analysis of repeated markers used to predict progression of cancer. Proceedings of the International Biometric Society (ENAR). Google Scholar
  13. Gann, P.H., Hennekens, C.H. and Stampfer, J.J. (1995). A prospective evaluation of plasma prostate-specific antigen for detection of prostatic cancer J. Amer. Med. Assoc. 273, 289–294.CrossRefGoogle Scholar
  14. Gelfand, A.E. and Smith, A.F.M. (1990). Sampling-based approaches to calculating marginal densities J. Amer. Statist. Assoc. 85, 398–409.MathSciNetzbMATHCrossRefGoogle Scholar
  15. Gelfand, A.E., Sahu, S.K. and Carlin, B.P. (1995). Efficient parametrisations for normal linear mixed models Biometrika 82, 479–88.MathSciNetzbMATHCrossRefGoogle Scholar
  16. Geman, S. and Geman, D. (1984). Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images IEEE Transactions on Pattern Analysis and Machine Intelligence 6, 721–741.zbMATHCrossRefGoogle Scholar
  17. Lange, N., Carlin, B.P. and Gelfand, A.E. (1992). Hierarchical Bayes models for the progression of HIV infection using longitudinal CD4 T-cell numbers J. Amer. Statist. Assoc. 87, No. 419, 615–632.zbMATHCrossRefGoogle Scholar
  18. Lindstrom, M.J. and Bates, D.M. (1988). Newton-Raphson and EM algorithms for linear mixed-effects models for repeated-measures data, J. Amer. Statist. Assoc. 83, 1014–1022.MathSciNetzbMATHGoogle Scholar
  19. Lindstrom, M.J. and Bates, D.M. (1990). Nonlinear Mixed Effects Models for Repeated Measures Data Biometrics 46, 673–687.MathSciNetCrossRefGoogle Scholar
  20. Morrell, C.H., Pearson, J.D., Carter, H.B. and Brant, L.J. (1995). Estimating unknown transition times using a piecewise nonlinear mixed-effects model in men with prostate cancer J. Amer. Statist. Assoc. 90 45–53.CrossRefGoogle Scholar
  21. Murtagh, P.A. (1995). ROC curves with multiple marker measurements. Biometrics 51, No. 4, 1514–1522.CrossRefGoogle Scholar
  22. Murtagh, P.A. and Schaid, D.J. (1991). Application of R.O.C. curve methodology when markers are repeated measures. Proc. of the International Biometric Society (ENAR) Houston, TX, March 24–27.Google Scholar
  23. Oesterling, J.E., Jacobsen, S.J., Chute, C.G., Guess, H.A., Girman, C.J., Panser, L.A. and Lieber, M.M. (1993). Serum prostate-specific antigen in a community-based population of healthy men. J. Amer. Med. Assoc. 270, No. 5, 660–664.Google Scholar
  24. Pearson, J.D., Luderer, A.A., Metter, E.J., Partin, A.W., Chan, D.W., Fozard, J.L. and Carter, H.B. (1996). Longitudinal analysis of serial measurements of free and total PSA among men with and without prostatic cancer Urology 48, (6A), 4–9.CrossRefGoogle Scholar
  25. Pearson, J.D., Morrell, C.H., Landis, P.K., Carter, H.B. and Brant, L.J. (1994). Mixed-effects regression models for studying the natural history of prostate disease Statist. Med. 13, 587–601.CrossRefGoogle Scholar
  26. Pearson, J.D., Kaminski, P., Metter, E.J., Fozard, J.L., Brant, L.J., Morrell, C.H. and Carter, H.B. (1991). Modeling longitudinal rates of change in prostate specific antigen during aging, Proc. Social Statistics Section of the Amer. Statist. Assoc. Google Scholar
  27. Press, S.J. (1982). Applied Multivariate Analysis: Using Bayesian and Frequentist Methods of Inference Robert E. Krieger Publishing, New York.zbMATHGoogle Scholar
  28. Seber, G.A.F. and Wild, C.J. (1989). Nonlinear Regression Wiley, New York.zbMATHCrossRefGoogle Scholar
  29. Slate, E.H. and Cronin, K.A. (1997). Changepoint modeling of longitudinal PSA as a biomarker for prostate cancer in C. Gatsonis, J.S. Hodges, R.E. Kass, R. McCulloch, P. Rossi and N.D. Singpurwalla (eds) Case Studies in Bayesian Statistics III, Springer-Verlag, New York, pp. 444–456.Google Scholar
  30. Stephens, D.A. (1994). Bayesian retrospective multiple-changepoint identification. Appl. Statist 43, No. 1, 159–178.zbMATHCrossRefGoogle Scholar
  31. Vessella, R.L., Noteboom, J. and Lange, P.H. (1992). Evaluation of the Abbott IMx automated immunoassay of prostate-specific antigen Clinical Chemistry 38, 044–2054.Google Scholar
  32. Whittemore, A.S., Lele, C., Friedman, G.C., Stamey, T. Vogelman, J.J., and Orentreich, N. (1995). Prostate-specific antigen as predictor of prostate cancer in black man and white men J. Nat. Cancer Inst. 87, No. 5,354–360.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 1999

Authors and Affiliations

  • Elizabeth H. Slate
  • Larry C. Clark

There are no affiliations available

Personalised recommendations