Clinical survival predictors in patients with advanced cancer

Background: The clinical and epidemiological relevance of different prognos tic factors for survival in patients with advanced or terminal cancer remai ns controversial. Purposes: To establish the survival of patients with cancer after diagnosis of terminal disease and to determine the predictors of survival. Methods: An inception cohort of 227 consecutive patients aged 18 years or o lder with terminal cancer of the lung, breast, and gastrointestinal tract w ere observed from July 1, 1996, through December 31, 1998. Tumor- and treat ment-specific, clinical, laboratory, demographic, and socioeconomic variabl es were recorded at baseline. The relationships between these characteristi cs and survival time were examined using univariate Kaplan-Meier and multiv ariate Cox regression analyses. Results: Bt the time of data analysis, 208 patients (91.6%;1) had died; the overall median survival for the sample was 15.3 weeks. Shorter survival wa s independently associated (P less than or equal to.05) with a primary tumo r of the lung (vs breast and gastrointestinal tract combined), liver metast ases, moderate-to-severe comorbidity levels(vs absent-to-mild levels),weigh t loss of greater than 8.1 kg in the previous 6 months, serum albumin level s of less than 35 g/L, lymphocyte counts of less than 1 x 10(9)/L, serum la ctate dehydrogenase levels of greater than 618 U/L, and clinical estimation of survival by the treating physician of less than 2 months (vs 2-6 and >6 months). Performance status, symptoms other than nausea and vomiting, tumo r burden, and socioeconomic characteristics such as social support and educ ation and income levels did not appear to be independently associated with survival after adjusting for the effect of prognostic factors. Conclusions: Simple clinical and laboratory assessments are useful aids in the prediction of survival in patients with solid malignant neoplasms at th e onset of terminal-stages. Methodological improvements in the design and i mplementation of survival studies may reduce prognostic uncertainty and ult imately provide better care for the terminally ill patients and their famil ies.