The aim of this study is to evaluate the effect of twelve weeks of a physical and psycho-social program consisting of: supervised, structured exercise training in a group of advanced lung cancer patients (cardio and strength training, relaxation training) twice weekly. Primary outcome is (VO2peak). The hypotheses are that patients who undergo this intervention will increase maximal oxygen uptake (VO2peak), strength (1RM), functional capacity (6MWD) and quality of life (HRQOL) and reduce the level of anxiety and depression compared to those who do not.

Control - usual care The patients randomized to the control group received no training but are offered the chance to participate in the supervised training after they have completed their antineoplastic treatment, at least after twelve weeks. Patients in early 2nd line treatment ("switch maintenance") will be offered training after 12 weeks, although they have not completed chemotherapy.

Intervention The supervised training was carried out in groups of 12-16 patients and each session had a duration of 1.5 hours, was administered twice weekly and was supervised by a research physiotherapist. The training comprised warm up exercises, strength and fitness training as well as stretching. Warm up exercises consisted of 10 minutes of light, stationery cycling, adjusted to 60-90% of the patient's maximum HR. Strength training was carried out using 6 machines (Technogym: Leg press, chest press, lateral machine, leg extension, abdominal crunch, and lower back). The practical aim of strength training was to complete 3 series of 5-8 sets, with 70-90% of 1RM. The exercises were specifically selected to involve the largest possible number of muscle groups in the least number of exercises. To ensure progression in strength training, each patient was instructed in carrying out the 1RM test using each of the above-mentioned strength training machines once every second week, after which their program would be adjusted. Cardiovascular training was carried out as interval training on stationery bikes. Intensity was equivalent to 85-95% of each patient's maximum HR and lasted approximately 10-15 minutes. After the training session, 5-10 minutes were dedicated to stretching the large muscle groups in order to increase agility. Following each training session, progressive relaxation of 15-20 minutes was performed.

Pre training screening Each Patient was screened by a clinical nurse specialist prior to participating in each physical training session and before the physiological tests (35). If one of the following criteria were met, the patient was prohibited from exercising/being tested on that day: diastolic blood pressure <45 or >95, heart rate (HR) at rest >115/min, temperature > 38 0C, respiratory rate at rest >30/min, infection requiring treatment, fresh bleeding, total leucocyte count <1.0 109/L or platelets <50 109/L. Physical tests and HRQOL evaluation were performed at baseline and after six weeks of training.

Statistical analysis Sample size The applied calculations are performed on 55 patients on the basis of VO2peak; through six weeks of training 55 patients achieved an increase of 0.85 ml / kg / min in VO2peak (SD = 2,48). It is assumed that the control group in the current study will have a reduction of 0.5 ml / kg / min for VO2peak (SD = 2.48). Assumptions for calculation of patient numbers are as follows: Type 1 error 0.05, type 2 error 0.20, and success level 1.35, SD 2.48. This leads to a total number of patients of 108 (54 in each arm). This study expects a drop-out rate of 50% and therefore another 108 patients must be included, which gives a sample size of 216 patients.

Data entry is carried out in the Open Clinica and analysis will be performed using the computer program SAS. As regards the choice of the statistical tests applied to data in which there will be no continuous differences, that will be analyzed by paired t-test or a corresponding non-parametric test, as categorical data are analyzed using the Pearson χ2 tests. Other data (e.g., from questionnaires) are analyzed using logistic regression and chi square or Mantel Haenzel test. Significance level is set at 0.05.

Inclusion Criteria:

• Non small cell lung cancer stage IIIb-IV
• Small cell lung cancer Extensiv disease
• Patients >18 years
• WHO performance status 0-2
• undergoing chemotherapy

Exclusion Criteria:

• brain or bone metastases;
• prolonged bone marrow suppression
• anti-coagulant treatment
• symptomatic heart disease
• congestive heart failure
• arrhythmia
• myocardial infarction diagnosed within the last three months
• inability to provide informed consent.