This study was designed to investigate the clinical relevance of polyamine determination in human tissue, serum, and urine as diagnostic markers in the diagnosis or postoperative control of patients with gastrointestinal cancer. Total, free, and acetylated polyamines were measured simultaneously in tissue, serum, and urine of patients with esophageal (n = 16), gastric (n = 40), pancreatic (n = 20) or colorectal (n = 50) carcinoma, in 40 patients with non-malignant gastrointestinal diseases, and 30 healthy volunteers. Though sensitivity of polyamines for gastrointestinal cancer proved to be high, non-malignant gastrointestinal diseases partly showed similar elevated polyamine concentrations in urine and serum which resulted in a low specificity of polyamines for gastrointestinal cancer. Therefore polyamines are not suitable as diagnostic markers in tumor screening. Furthermore, since polyamine concentrations in serum and urine normalized after curative operation while they were further increased and elevated in patients with proven relapse of the tumor or metastases, polyamines might be of clinical value in postoperative monitoring of therapeutical success or predicting relapse of the tumor.