Cancer is a leading cause of death worldwide and the prognostic evaluation of cancer patients is of great importance in medical care. The use of artificial neural networks in prediction problems is well established in human medical literature. The aim of the current study was to assess the prognostic value of a series of clinical and molecular variables with the addition of γ-H2AX—a new DNA damage response marker—for the prediction of prognosis in patients with early operable non-small cell lung cancer by comparing the γ-H2AX-based artificial network prediction model with the corresponding LR one. Two prognostic models of 96 patients with 27 input variables were constructed by using the parameter-increasing method in order to compare the predictive accuracy of neural network and logistic regression models. The quality of the models was evaluated by an independent validation data set of 11 patients. Neural networks outperformed logistic regression in predicting the patient’s outcome according to the experimental results. To assess the importance of the two factors p53 and γ-H2AX, models without these two variables were also constructed. JR and accuracy of these models were lower than those of the models using all input variables, suggesting that these biological markers are very important for optimal performance of the models. This study indicates that neural networks may represent a potentially more useful decision support tool than conventional statistical methods for predicting the outcome of patients with non-small cell lung cancer and that some molecular markers, such as γ-H2AX, enhance their predictive ability. 1. Introduction Prediction is one of the most interesting areas where intelligent systems are utilized [1]. In particular, prediction is an attempt to accurately forecast the evolution or outcome of a specific situation, using as input information a concrete set of variables that describe this situation [2]. In medicine, the valid and effective interpretation of medical data and the correct and early diagnosis along with a documented prognostic evaluation of the clinical and pathological data are very important parameters for a better management of the disease [3]. Prediction is a very difficult task because the expert human can hardly process the huge amount of data and usually suffers from absence of good and accurate analysis of these laboratory data [4, 5]. Lung cancer is the most common cause of cancer mortality worldwide for both men and women, causing approximately 1.2 million deaths per year. In the United States, there were 221.000
References
[1]
B. Pandey and R. B. Mishra, “Knowledge and intelligent computing system in medicine,” Computers in Biology and Medicine, vol. 39, no. 3, pp. 215–230, 2009.
[2]
J. A. Gómez-Ruiz, J. M. Jerez-Aragonés, J. Mu？oz-Pérez, and E. Alba-Conejo, “A neural network based model for prognosis of early breast cancer,” Applied Intelligence, vol. 20, pp. 231–238, 2004.
[3]
O. Er, N. Yumusak, and F. Temurtas, “Chest diseases diagnosis using artificial neural networks,” Expert Systems with Applications, vol. 37, no. 12, pp. 7648–7655, 2010.
[4]
I. Saritas, “Prediction of breast cancer using artificial neural networks,” Journal of Medical Systems, vol. 36, no. 5, pp. 2901–2907, 2011.
[5]
F. Feng, Y. Wu, Y. Wu, G. Nie, and R. Ni, “The effect of artificial neural network model combined with six tumor markers in auxiliary diagnosis of Lung Cancer,” Journal of Medical Systems, vol. 36, pp. 2973–2980, 2011.
[6]
R. Siegel, E. Ward, O. Brawley, and A. Jemal, “Cancer statistics, 2011: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths,” CA-Cancer Journal for Clinicians, vol. 61, no. 4, pp. 212–236, 2011.
[7]
P. Goldstraw, D. Ball, J. R. Jett et al., “Non-small-cell lung cancer,” The Lancet, vol. 378, no. 9804, pp. 1727–1740, 2011.
[8]
C. Stephan, H. Cammann, H. Rittenhouse et al., “New biomarkers and application of multivariate models for detection of prostate cancer,” Aktuelle Urologie, vol. 40, no. 4, pp. 221–230, 2009.
[9]
L. Cagini, M. Monacelli, G. Giustozzi et al., “Biological prognostic factors for early stage completely resected non-small cell lung cancer,” Journal of Surgical Oncology, vol. 74, pp. 53–60, 2000.
[10]
M. G？nen, “Bias, biostatistics, and prognostic factors,” Journal of Thoracic Oncology, vol. 6, no. 7, pp. S1705–S1709, 2011.
[11]
Q. Hu, B. Li, D. Garfield et al., “Prognostic factors for survival in a Chinese population presenting with advanced non-small cell lung cancer with an emphasis on smoking status: a regional, single-institution, retrospective analysis of 4552 patients,” Thoracic Cancer, vol. 3, pp. 162–168, 2012.
[12]
J. S. Dickey, C. E. Redon, A. J. Nakamura, B. J. Baird, O. A. Sedelnikova, and W. M. Bonner, “H2AX: functional roles and potential applications,” Chromosoma, vol. 118, no. 6, pp. 683–692, 2009.
[13]
M. Podhorecka, A. Skladanowski, and P. Bozko, “H2AX phosphorylation: its role in DNA damage response and cancer therapy,” Journal of Nucleic Acids, vol. 2010, Article ID 920161, 9 pages, 2010.
[14]
D. Matthaios, D. Bouros, and S. Kakolyris, “H2AX and lung cancer:is it the Ariadne’s thread?” DNA Repair, vol. 12, no. 2, pp. 90–91, 2013.
[15]
J. Beane, P. Sebastiani, T. H. Whitfield et al., “A prediction model for lung cancer diagnosis that integrates genomic and clinical features,” Cancer Prevention Research, vol. 1, no. 1, pp. 56–64, 2008.
[16]
G. Santos-García, G. Varela, N. Novoa, and M. F. Jiménez, “Prediction of postoperative morbidity after lung resection using an artificial neural network ensemble,” Artificial Intelligence in Medicine, vol. 30, no. 1, pp. 61–69, 2004.
[17]
E. Bartfay, W. J. Mackillop, and J. L. Pater, “Comparing the predictive value of neural network models to logistic regression models on the risk of death for small-cell lung cancer patients,” European Journal of Cancer Care, vol. 15, no. 2, pp. 115–124, 2006.
[18]
C. M. Bishop, Neural Networks for Pattern Recognition, Oxford University Press, New York, NY, USA, 1995.
[19]
K. Balachandran and R. Anitha, “Supervised learning processing techniques for pre-diagnosis of lung cancer disease,” International Journal of Computer Applications, vol. 1, article 17, 2010.
[20]
G. Schwarzer, W. Vach, and M. Schumacher, “On the misuses of artificial neural networks for prognostic and diagnostic classification in oncology,” Statistics in Medicine, vol. 19, pp. 541–561, 2000.
[21]
P. J. G. Lisboa, “A review of evidence of health benefit from artificial neural networks in medical intervention,” Neural Networks, vol. 15, no. 1, pp. 11–39, 2002.
[22]
S. A. Grumett and P. B. Snow, “Artificial neural networks: a new model for assessing prognostic factors,” Annals of Oncology, vol. 11, no. 4, pp. 383–384, 2000.
[23]
E. Biganzoli, P. Boracchi, D. Coradini, M. G. Daidone, and E. Marubini, “Prognosis in node-negative primary breast cancer: a neural network analysis of risk profiles using routinely assessed factors,” Annals of Oncology, vol. 14, no. 10, pp. 1484–1493, 2003.
[24]
E. Chatzimichail, E. Paraskakis, M. Sitzimi, and A. Rigas, “Predicting the long-term outcome of preschool children with asthma symptoms,” in Proceedings of the E-Health and Bioengineering Conference (EHB '11), pp. 1–4, November 2011.
[25]
A. Ciampi and F. Zhang, “A new approach to training back-propagation artificial neural networks: empirical evaluation on ten data sets from clinical studies,” Statistics in Medicine, vol. 21, no. 9, pp. 1309–1330, 2002.
[26]
V. Van Belle, K. Pelckmans, J. A. K. Suykens, and S. Van Huffel, “Additive survival least-squares support vector machines,” Statistics in Medicine, vol. 29, no. 2, pp. 296–308, 2010.
[27]
E. A. Chatzimichail, A. G. Rigas, and E. N. Paraskakis, “An artificial intelligence technique for the prediction of persistent asthma in children,” in Proceedings of the 10th International Conference on Information Technology and Applications in Biomedicine (ITAB '10), pp. 1–4, November 2010.
[28]
E. Pagano, C. Filippini, D. Di Cuonzo et al., “Factors affecting pattern of care and survival in a population-based cohort of non-small-cell lung cancer incident cases,” Cancer Epidemiology, vol. 34, no. 4, pp. 483–489, 2010.
[29]
L. Liu, E. Zhao, C. Li, et al., “TRIM28, a new molecular marker predicting metastasis and survival in early-stage non-small cell lung cancer,” Cancer Epidemiology, vol. 37, pp. 71–78, 2013.
[30]
E. Biganzoli, P. Boracchi, L. Mariani, and E. Marubini, “Feed forward neural networks for the analysis of censored survival data: a partial logistic regression approach,” Statistics in Medicine, vol. 17, pp. 1169–1186, 1998.
[31]
C. R. Mehta and N. R. Patel, “Exact logistic regression: theory and examples,” Statistics in Medicine, vol. 14, no. 19, pp. 2143–2160, 1995.
[32]
J. V. Tu, “Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes,” Journal of Clinical Epidemiology, vol. 49, no. 11, pp. 1225–1231, 1996.
[33]
B. Eftekhar, K. Mohammad, H. E. Ardebili, M. Ghodsi, and E. Ketabchi, “Comparison of artificial neural network and logistic regression models for prediction of mortality in head trauma based on initial clinical data,” BMC Medical Informatics and Decision Making, vol. 5, article 3, 2005.
[34]
D. J. Sargent, “Comparison of artificial neural networks with other statistical approaches: results from medical data sets,” Cancer, vol. 91, no. 8, supplement, pp. 1636–1642, 2001.
[35]
D. Matthaios, P. G. Foukas, M. Kefala et al., “γ-H2AX expression detected by immunohistochemistry correlates with prognosis in early operable non-small cell lung cancer,” OncoTargets and Therapy, vol. 5, pp. 309–314, 2012.
