Further evidence for the use of polyamines as biochemical markers for malignant tumors.

One hundred ninety patients with a variety of tumor presented within a specified time period and fit a specified protocol. Multiple serial urinary putrescine, spermidine, and spermine levels were obtained in these patients, and their disease activity over time, defined as either active or nonactive, was determined by clinical examination, the results of laboratory tests, and radiological criteria. Twenty-four-hr urine collections were used for analysis of polyamine levels. A linear mixed-effects model and the method of maximum likelihood estimation were used for statistical analysis. Statistically significant differences were found in polyamine levels between patients with active or nonactive disease for tumors of the breast, stomach, prostate, female genital tract, and a variety of metastatic carcinomas of unknown origin. There were 105 patients with these tumors; 319 polyamine determinations were obtained from this subset of patients. Our results suggest that serial determination of polyamine levels in urine may have clinical utility for monitoring the disease states for these tumors.