Summary
The objective of this study is to compare the predictive accuracy of a neural network (NN) model versus the standard Cox proportional hazard model. Data about the 3811 patients included in this study were collected within the ‘El Álamo’ Project, the largest dataset on breast cancer (BC) in Spain. The best prognostic model generated by the NN contains as covariates age, tumour size, lymph node status, tumour grade and type of treatment. These same variables were considered as having prognostic significance within the Cox model analysis. Nevertheless, the predictions made by the NN were statistically significant more accurate than those from the Cox model (p<0.0001). Seven different time intervals were also analyzed to find that the NN predictions were much more accurate than those from the Cox model in particular in the early intervals between 1–10 and 11–20 months, and in the later one considered from 61 months to maximum follow-up time (MFT). Interestingly, these intervals contain regions of high relapse risk that have been observed in different studies and that are also present in the analyzed dataset.
Similar content being viewed by others
References
Hilsenbeck SG, Ravdin PM, de Moor CA, (1998) Time-dependence of hazard ratios for prognostic factors in primary breast cancer Breast Cancer Res Treat 52:227–237
Cox DR, (1972) Regression models and life tables J R Stat Soc 34:187–202
Ohno-Machado L, (1997) A comparison of Cox proportional hazards and artificial neural network models for medical prognosis Comput Biol Med 27:55–65
Xiang A, Lapuerta P, Ryutov A, Buckley J, Azen S, (2000) Comparison of the performance of neural network methods and Cox regression for censored survival data Comput Stat and Data Anal 34:243–257
Burke H, Goodman P, Rosen D, Henson D, Weinstein J, Harrel F, Marks J, Winchester D, Bostwick D: Artificial neural networks improve the accuracy of cancer survival prediction. Cancer 79: 857–862, 1997
Brown SF, Branford AJ, Moran W, (1997) On the use of artificial neural networks for the analysis of survival data IEEE Trans Neural Networks 8:1071–1077
Faraggi D, Simon R, Yaskil E, Kramar A, (1997) Bayesian neural network models for censored data Biometrica J 5:519–532
Ripley RM, Harris AL, Tarassenko L, (1998) Neural network models for breast cancer prognosis Neural Comput Appl 7:367–375
Lundin M, Lundin J, Burke HB, Toikkanen S, Pylkkänen Joensuu H, (1999) Artificial neural networks applied to survival prediction in breast cancer Oncology 57:281–286
Jerez JM, Gómez JA, Ramos G, Muñoz J, Alba E, (2003) A combined neural network and decision trees model for prognosis of breast cancer relapse Artif Intell Med 27:45–63
Ravdin PM, Clark GM, Hilsenbeck G, Owens MA, Vendely P, Pandian MR, McGuire WL, (1992) A demonstration that breast cancer recurrence can be predicted by neural network analysis Breast Cancer Res Treat 21:47–53
De Laurentis M, Ravdin PM, (1994) A technique for using neural network analysis to perform survival analysis of censored data Cancer Lett 77:127–138
De Laurentiis M, De Placido S, Bianco AR, Clark GM, Ravdin PM, (1999) A prognostic model that makes quantitative estimates of probability of relapse for breast cancer patients Clin Cancer Res 5:4133–4139
Boracchi P, Biganzoli E, Marubini E, (2001) Modelling cause-specific hazards with radial basis function artificial neural networks: application to 2233 breast cancer patients Stat Med 20:3677–3694
Liestol K, Andersen PK, (2002) Updating of covariates and choice of time origin in survival analysis: problems with vaguely defined disease states Stat Med 21:3701–3714
Biganzoli E, Boracchi P, Coradini D, Daidone MG, Marubini E, (2003) Prognosis in node-negative primary breast cancer: a neural network analysis of risk profiles using routinely assessed factors Ann Oncol 14:1484–1493
Biganzoli E, Boracchi P, Mariani L, Marubini E, (1998) Feed forward neural networks for the analysis of censored survival data: a partial logistic regression approach Stat Med 17:1169–1186
Dreiseitl S, Ohno-Machado L, (2002) Logistic regression and artificial neural network classification models: a methodology review J Biomed Inf 35:352–359
Fitzgibbons PL, Page DL, Weaver D, Thor AD, Allred DC, Clark GM, Ruby SG, O’Malley F, Simpson JF, Connolly JL, Hayes DF, Edge SB, Lichter A, Schnitt SJ, (2000) Prognostic factors in breast cancer. College of American Pathologists Consensus Statement 1999 Arch Pathol Lab Med 124:966
Kroman N, Jensen MB, Wohlfahrt J, Mouridsen HT, Andersen PK, Melbye M, (2000) Factors influencing the effect of age on prognosis in breast cancer: population based study BMJ 320:474–478
Colleoni M, Rotmensz N, Robertson C, Orlando L, Viale G, Renne G, Luini A, Veronesi P, Intra M, Orecchia R, Catalano G, Galimberti V, Nole F, Martinelli G, Goldhirsch A, (2002) Very young women (<35 years) with operable breast cancer: features of disease at presentation Ann Oncol 13:273–279
Therneau TM, Grambsch PM, (2000) Modelling Survival Data: Extending Survival Data. Springer-Verlag, New York, NY
Fox J, (2002) An R and S-Plus companion to applied regression. Sage Publications, Thousand Oaks, CA
Breslow NE, (1974) Covariance analysis of censored survival data Biometrics 30:89–99
Shao J, Tu D, (1995) The Jacknife and Bootstrap. Springer Verlag, New York, NY
Harrell FE, Califf RM, Pryor DB, Lee KL, Rosati RA, (1982) Evaluating the yield of medical tests J Am Med Assoc 247:2543–2546
Harrell FE, Lee KL, Califf RM, Pryor DB, Rosati RA, (1984) Regression modeling strategies for improved prognostic prediction Stat Med 3:143–152
J. Hanley B Mc Neil 1982 The meaning and use of the area under the receiver operating characteristic (ROC) curve. Radiology 143: 29–36
Hanley JA, McNeil BJ, (1983) A method for comparing the areas under receiver operating characteristic curves derived from the same cases Radiology 148:839–843
Clark TG, Bradburn MJ, Love SB, Altman DG, (2003) Survival Analysis Part IV: Further concepts and methods in survival analysis Br J Cancer 89:781–786
S Dreiseitl L Ohno-Machado H Harald Kittler S Vinterbo H Billhardt M Binder 2001 A comparison of machine learning methods for the diagnosis of pigmented skin lesions J Biomed Inform 34: 28–36
Greiner M, Pfeiffer D, Smith RD, (2000) Principles and practical application of the receiver-operating characteristic analysis for diagnostic tests Prev Vet Med 45:23–41
Demicheli R, Abbattista A, Miceli R, Valagussa P, Bonadonna G, (1996) Time distribution of the recurrence risk for breast cancer patients undergoing mastectomy: further support about the concept of tumor dormancy Breast Cancer Res Treat 41:177–185
Saphner T, Tormey DC, Gray R, (1996) Annual hazard rates of recurrence for breast cancer after primary therapy J Clin Oncol 14:2738–2746
Karrison TG, Ferguson DJ, Meier P, (1999) Dormancy of mammary carcinoma after mastectomy J Natl Cancer Inst 91:80–85
Retsky MW, Wardwell RH, Swartzendruber DE, Headley DL, (1987) Prospective computerized simulation of breast cancer: comparison of computer predictions with nine sets of biological and clinical data Cancer Res 47:4982–4987
Demicheli R, Valagussa P, Bonadonna G, (2001) Does surgery modify growth kinetics of breast cancer micrometastases? Br J Cancer 85:490–492
Bloom HJG, Richardson WW, Harries EJ, (1962) Natural history of untreated breast cancer (1905–1933) BMJ 2:213–221
Demicheli R, Miceli R, Valagussa P, Bonadonna G, (2000) Re: dormancy of mammary carcinoma after mastectomy J Natl Cancer Inst 92:347–348
Acknowledgements
Support from APOMA (Málaga, Spain) under contract No. 8.06/47/2176 is acknowledged. The authors also thank Esther Mahillo (GEICAM) for her collaboration and coordination of this project.
Author information
Authors and Affiliations
Corresponding author
Additional information
Address for offprints and correspondence: Leonardo Franco, Depto. de Lenguajes y Ciencias de la Computación, Universidad de Málaga, Campus de Teatinos S/N, 29071, Málaga, Spain; Tel.: +34-952-133304; Fax: +34-952-133397; E-mail: Leonardo.Franco@psy.ox.ac.uk
Rights and permissions
About this article
Cite this article
Jerez, J., Franco, L., Alba, E. et al. Improvement of breast cancer relapse prediction in high risk intervals using artificial neural networks. Breast Cancer Res Treat 94, 265–272 (2005). https://doi.org/10.1007/s10549-005-9013-y
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10549-005-9013-y