Lung Cancer: Optimal Treatment Strategies

Objective: Search of best treatment plan for non-small lung cancer (LC) patients (LCP) was realized. Methods: In trial (1985-2008) the data of consecutive 535 LCP after complete resections (R0) (age = 57.3 ± 8.2 years; male-482, female-53; tumor diameter: D = 4.7 ± 2.2 cm; pneumonectomies-222, lobectomies-313, combined procedures with resection of pericardium, atrium, aorta, VCS, carina, diaphragm, ribs-155; only surgery-S-316, adjuvant chemoimmunoradiotherapy-AT-117: CAV/gemzar + cisplatin + thymalin/taktivin + radiotherapy 45-50Gy, postoperative radiotherapy 45-50Gy-RT-102; squamous-341, adenocarcinoma-153, large cell-41; stage IA-105, IB-130, IIA-21, IIB-122, IIIA-116, IIIB-41; T1-150, T2-230, T3-114, T4-41; N0-310, N1-118, N2-107; G1-126, G2-152, G3-257) were reviewed. Variables selected for 5-year survival (5YS) study were input levels of blood, biochemic and hemostatic factors, sex, age, TNMG, D. Survival curves were estimated by Kaplan-Meier method. Differences in curves between groups were evaluated using a log-rank test. Neural networks computing, Cox regression, clustering, structural equation modeling, Monte Carlo and bootstrap simulation were used to determine any significant regularity. Results: For total of 535 LCP overall life span (LS) was 1723.3 ± 1294.9 days and cumulative 5YS reached 63.6%, 10 years – 52.8%. 304 LCP (LS = 2597.3 ± 1037 days) lived more than 5 years without LC progressing. 186 LCP (LS = 559.8 ± 383.1 days) died because of LC during first 5 years after surgery. 5YS of LCP with N1-2 was superior significantly after AT (65.6%) compared with RT (39.5%) (P = 0.0003 by log-rank test) and S (28.3%) (P = 0.000). Cox modeling displayed that 5YS significantly depended on: phase transition (PT)“early-invasive LC”, PT N0-N12, AT, age, weight, histology, G, T, D, blood cell subpopulations, cell ratio factors, ESS, prothrombin index, heparin tolerance, recalcification time, bilirubin, (P = 0.000-0.046). Neural networks computing, genetic algorithm selection and bootstrap simulation revealed relationships between 5YS and PT N0-N12 (rank = 1), procedure type, G, T, histology, AT, PT “earlyinvasive LC”, RT, S, sex, ESS, prothrombin index, fibrinogen, Hb, protein, weight, lymphocytes. Correct prediction of 5YS was 99.6% by neural networks computing (error = 0.045; urea under ROC curve = 0.995). Conclusion: Optimal treatment strategies for LCP are: 1) screening and early detection of LC; 2) availability of experienced surgeons because of complexity of radical procedures; 3) aggressive en block surgery and adequate lymphadenectomy for completeness; 4) precise prediction; 5) adjuvant chemoimmunoradiotherapy for LCP with unfavorable progno-