[Prognostic factors in operated non-small cell cancer of the lung. Study from a randomized therapeutic trial]

This article presents the results of a prognostic study of primary resected lung cancer (non-small cell). The data result from a randomised clinical trial of immunotherapy with a non-specific adjuvant; the follow-up was between four to seven years. Thirty-five clinical, biological and anatomo-pathological parameters were gathered at the time of inclusion in the trial. The response criteria used were survival without recurrence and total survival. A multivariate analysis using the Cox's model was carried out for each criterion. At the reference date of the 1st April 1985, 125 relapses and 132 deaths were counted amongst 219 patients; there was only one patient lost to follow-up and only 39 missing data were observed. The negative therapeutic results of the immunotherapy used were confirmed by this new intermediate analysis. The rate of survival without recurrence at 5 years was 43% and the overall survival at five years was 42%. The use of Cox's model to show the prognostic information at the 5% level for survival without recurrence could be summarised by five factors: main staging (the prognostic factor), leucocytosis, the cutaneous reaction to proteus, Karnofsky index and presence of physical signs. For stages I and II the outcome was identical and no factor was predictive at the 5% level. For stage III the cutaneous reaction to proteus and leucocytosis were prognostic. For overall survival, the prognostic information at the 5% level could be summarised by five factors: staging (main prognostic factor), leucocytosis, Karnofsky index, presence of physical signs and lymphocytosis. For stages I and II whose outcome was identical only Karnofsky index and lymphocytosis were predictive at the 5% level.(ABSTRACT TRUNCATED AT 250 WORDS)