Chemotherapy in advanced non-small cell lung cancer.

Chemotherapy in advanced non-small cell lung cancer (NSCLC) has been evaluated with response rate, survival and quality of life as criteria. The data were collected from 142 phase II and III trials. The three main conclusions are: (1) multivariate landmark analyses show that a response to chemotherapy has an independent significant value for prognosis, and all studies of chemotherapy vs. best supportive care show some survival benefit in favour of chemotherapy; (2) NSCLC patients are a heterogeneous group with a large variation of known and unknown prognostic factors, such as the treatment centre, for both response and survival; and (3) retrospective analysis of the data show that response rate and survival are significantly correlated at response rate over 30% and in limited disease patients. The following recommendations are made: (1) new drugs should be compared in randomised phase II trials with a standard active drug; (2) randomised phase III trials of single agents or best supportive care vs. combination chemotherapy should be repeated in a well-defined subgroup of patients with a high performance score and limited tumour load; and (3) the palliative effect of chemotherapy in randomised trials in patients with symptoms should be investigated with relief of symptoms and quality of life as endpoints.