Skip to main content
PMC home page
Cancer Science logoLink to Cancer Science
. 2005 Aug 19;94(5):473–477. doi: 10.1111/j.1349-7006.2003.tb01467.x

Prognostic models in patients with non‐small‐cell lung cancer using artificial neural networks in comparison with logistic regression

Taizo Hanai 1,6,✉, Yasushi Yatabe 2,6, Yusuke Nakayama 1, Takashi Takahashi 4, Hiroyuki Honda 1, Tetsuya Mitsudomi 3, Takeshi Kobayashi 1
PMCID: PMC11160259  PMID: 12824896

Abstract

It is difficult to precisely predict the outcome of each individual patient with non‐small‐cell lung cancer (NSCLC) by using conventional statistical methods and ordinary clinico‐pathological variables. We applied artificial neural networks (ANN) for this purpose. We constructed a prognostic model for 125 NSCLC patients with 17 potential input variables, including 12 clinico‐pathological variables (age, sex, smoking index, tumor size, p factor, pT, pN, stage, histology) and 5 immunohistochemical variables (p27 percentage, p27 intensity, p53, cyclin D1, retinoblastoma (RB)), by using the parameter‐increasing method (PIM). Using the resultant ANN model, prediction was possible in 104 of 125 patients (83%, judgment ratio (JR)) and accuracy for prediction of survival at 5 years was 87%. On the other hand, JR and survival prediction accuracy in the logistic regression (LR) model were 37% and 78%, respectively. In addition, ANN outperformed LR for prediction of survival at 1 or 3 years. In these cases, PIM selected p27 intensity and cyclin D1 for the 3‐year survival model and p53 for the 1‐year survival model in addition to clinico‐pathological variables. Finally, even in an independent validation data set of 48 patients, who underwent surgery 10 years later, the present ANN model could predict outcome of patients at 5 years with the JR and accuracy of 81% and 77%, respectively. This study demonstrates that ANN is a potentially more useful tool than conventional statistical methods for predicting survival of patients with NSCLC and that inclusion of relevant molecular markers as input variables enhances its predictive ability. (Cancer Sci 2003; 94: 473–477)

References

  • 1. Vital statistics of Japan. Vol. 3. Tokyo : Statistics and Information Department, Ministry of Health and Welfare; 1998. p. 384–411. [Google Scholar]
  • 2. Wingo PA, Ries LA, Giovino GA, Miller DS, Rosenberg HM, Shopland DR, Thun MJ, Edwards BK. Annual report to the nation on the status of cancer, 1973–1996, with a special section on lung cancer and tobacco smoking. J Natl Cancer Inst 1999; 91: 675–90. [DOI] [PubMed] [Google Scholar]
  • 3. Moore DFJ, Lee JS. Staging and prognostic factors: non‐small cell lung cancer. In: Pass IBM, Johnson DH, Turrisi AT, editors. Lung cancer: principles and practice. Philadelphia : Lippincott‐Raven; 1996. p. 481–94. [Google Scholar]
  • 4. Lau CL, D'Amico TA, Harpole DH. Clinical and molecular prognostic factors and models for non‐small cell lung cancer. In: Pass HI, Mitchell JB, Johnson DH, Turrisi AT, Minna JD, editors. Lung cancer: principles and practice. Philadelphia : Lippincott Williams & Wilkins; 2000. p. 602–611. [Google Scholar]
  • 5. Mountain CF. International staining system for lung cancer. In: Pass HI, Mitchell JB, Johnson DH, Turrisi AT, Minna, JD , editors. Lung cancer: principles and practice. Philadelphia : Lippincott Williams & Wilkins; 2000. p. 591–601. [Google Scholar]
  • 6. Binsberg RJ, Vokes, EE , Rosenzweig, K. Non‐small cell lung cacner. In: DDevita T, Hellman S, Rosenberg, SA , editors. Cancer: principles & practice of oncology. Philadelphia : Lippincott Williams & Wilkins; 2000. p. 925–83. [Google Scholar]
  • 7. Sobin LH, Wittekind, C. TNM classification of malignant tumours. 5th ed. New York : Wiley‐Liss Inc; 1997. [Google Scholar]
  • 8. Sobin LH. TNM: principles, history, and relation to other prognostic factors. Cancer 2001; 91: 1589–95. [DOI] [PubMed] [Google Scholar]
  • 9. Montie JE, Wei JT. Artificial neural networks for prostate carcinoma risk assessment. An overview. Cancer 2001; 91: 1947–52. [DOI] [PubMed] [Google Scholar]
  • 10. Nishio M, Koshikawa T, Yatabe Y, Kuroishi T, Suyama M, Nagatake M, Sugiura T, Ariyoshi Y, Mitsudomi T, Takahashi, T. Prognostic significance of cyclin D1 and retinoblastoma expression in combination with p53 abnormalities in primary, resected non‐small cell lung cancers. Clin Cancer Res 1997; 3: 1051–8. [PubMed] [Google Scholar]
  • 11. Nishio M, Koshikawa T, Kuroishi T, Suyama M, Uchida K, Takagi Y, Washimi O, Sugiura T, Ariyoshi Y, Takahashi T, Ueda R. Prognostic significance of abnormal p53 accumulation in primary, resected non‐small‐cell lung cancers. J Clin Oncol 1996; 14: 497–502. [DOI] [PubMed] [Google Scholar]
  • 12. Yatabe Y, Masuda A, Koshikawa T, Nakamura S, Kuroishi T, Osada H, Takahashi T, Mitsudomi T. p27KIP1 in human lung cancers: differential changes in small cell and non‐small cell carcinomas. Cancer Res 1998; 58: 1042–7. [PubMed] [Google Scholar]
  • 13. Dayhoff JE, DeLeo JM. Artificial neural networks: opening the black box. Cancer 2001; 91: 1615–35. [DOI] [PubMed] [Google Scholar]
  • 14. Baxt WG, Skora J. Prospective validation of artificial neural network trained to identify acute myocardial infarction. Lancet 1996; 347: 12–5. [DOI] [PubMed] [Google Scholar]
  • 15. Ball G, Mian S, Holding F, Allibone RO, Lowe J, Ali S, Li G, McCardle S, Ellis IO, Creaser C, Rees RC. An integrated approach utilizing artificial neural networks and SELDI mass spectrometry for the classification of human tumors and rapid identification of potential biomarkers. Bioinformatics 2002; 18: 395–404. [DOI] [PubMed] [Google Scholar]
  • 16. Milik M, Sauer D, Brunmark AP, Yuan L, Vitiello A, Jackson MR, Peterson PA, Skolnick J, Glass CA. Application of an artificial neural network to predict specific class I MHC binding peptide sequences. Nat Biotechnol 1998; 16: 753–6. [DOI] [PubMed] [Google Scholar]
  • 17. Lubbert A, Simuits R. Using measuremant data in bioprocess modelling and control. TIBTECH 1994; 12: 304–11. [Google Scholar]
  • 18. Rumelhart DE, Hinton GE, Williams RL. Learning representations by back‐propagating errors. Nature 1986; 323: 533–6. [Google Scholar]
  • 19. Wasserman PD. Advanced methods in neural computing. New York : Van Nostrand Reinhold, a Division of Wadsworth Inc; 1993. [Google Scholar]
  • 20. Hanai T, Katayama A, Honda H, Kobayashi T. Automatic fuzzy modeling for Ginjo sake brewing process using fuzzy neural networks. J Chem Eng Jpn 1997; 30: 94–100. [Google Scholar]
  • 21. Mitsudomi T, Hamajima N, Ogawa M, Takahashi T. Prognostic significance of p53 alterations in patients with non‐small cell lung cancer: a meta‐analysis. Clin Cancer Res 2000; 6: 4055–63. [PubMed] [Google Scholar]
  • 22. Jefferson MF, Pendleton N, Lucas SB, Horan MA. Comparison of a genetic algorithm neural network with logistic regression for predicting outcome after surgery for patients with nonsmall cell lung carcinoma. Cancer 1997; 79: 1338–42. [DOI] [PubMed] [Google Scholar]
  • 23. Bellotti M, Eisner B, Paez De Lima A, Esteva H, Marchevsky AM. Neural networks as a prognostic tool for patients with non‐small cell carcinoma of the lung. Mod Pathol 1997; 10: 1221–7. [PubMed] [Google Scholar]
  • 24. Sargent DJ. Comparison of artificial neural networks with other statistical approaches: results from medical data sets. Cancer 2001; 91: 1636–42. [DOI] [PubMed] [Google Scholar]
  • 25. Schena M, Shalon D, Davis RW, Brown PO. Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 1995; 207: 467–70. [DOI] [PubMed] [Google Scholar]
  • 26. Gress TM, Hoheisel JD, Lennon GG, Zehetner G, Lehrach H. Hybridization fingerprinting of high‐density cDNA‐library arrays with cDNA pools derived from whole tissues. Mamm Genome 1992; 3: 609–19. [DOI] [PubMed] [Google Scholar]
  • 27. Kononen J, Bubendorf L, Kallioniemi A, Barlund M, Schraml P, Leighton S, Torhorst J, Mihatsch MJ, Sauter G, Kallioniemi OP. Tissue microarrays for high‐throughput molecular profiling of tumor specimens. Nat Med 1998; 4: 844–7. [DOI] [PubMed] [Google Scholar]
  • 28. Khan J, Wei JS, Ringner M, Saal LH, Ladanyi M, Westermann F, Berthoud F, Schwab M, Antonescu CR, Peterson C, Meltzer PS. Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks. Nat Med 2001; 7: 673–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Cancer Science are provided here courtesy of Wiley

RESOURCES