Neopterin and Prognosis in Patients with Adenocarcinoma of the Colon

Concentrations of neopterin, a sensitive indicator for the activation of cellular immunity, were measured in urine samples of 44 patients with adenocarcinoma of the colon at diagnosis. To judge the relative predictive value of neopterin concentrations, other routine clinical and laboratory variables were concomitantly determined. The patients were then followed up to I0 yr, and the abilities of all variables to predict death from cancer were assessed. Neopterin concentrations were not correlated with either tumor stage or Dukes' stage. In univariate analyses using the product-limit approach, four variables were significant indicators of poor prognosis: presence of distant metastases (P = 0.0001); high Dukes' stage (P = 0.0009); high urinary neopterin concentration (P = 0.0034); and advanced stage (P = 0.030). Presence v e r s u s absence of lymph node metastases was not associated with prognosis. Multivariate survival analyses by the proportional hazards technique demonstrated that neopterin provided statistically independent predictive information in addition to either presence v e r s u s absence of distant metastases or Dukes' stage. When neopterin and tumor stage were investigated for joint prediction, stage failed to be included in the model. Thus, neopterin concentrations provide valuable and statistically independent prognostic information in patients with adenocarcinoma of the colon. I N T R O D U C T I O N Neopterin belongs to the class of pteridines which are pyrazino[2,3-d]pyrimidine compounds biosynthesized from GTE In hepatic and neurological tissue the key enzyme of pteridine biosynthesis, GTP cyclohydrolase I (EC 3.5.4.16), is constitutively present, whereas in other tissues, including macrophages and fibroblasts but also in several tumor cell lines, the enzyme exists in a cytokine-inducible form (1). In 'human monocytes/macrophages, 7-interferon is the most potent inducer of this enzyme (2), and its effects can be potentiated by tumor necrosis factor ct and lipopolysaccharide (3). In contrast to all other cellular systems investigated hitherto, human monocytes/ macrophages are unique because they produce by far the highest amounts of neopterin after cytokine induction, whereas all other cell types produce equal or higher amounts of biopterin derivatives. This peculiarity of human monocytes/macrophages is due to the fact that, in these cells, the activities of the constitutively present subsequent enzymes of the pteridine pathway, 6-pyruvoyl-tetrahydropterin synthase and sepiapterin reductase (EC 1.1.1.153), are much lower than the activity of the 7-interferon-induced GTP cyclohydrolase I (1). Thus, the 7,8-dihydroneopterin triphosphate being formed from GTP by GTP cyclohydrolase I accumulates, and aromatic neopterin is formed by the action of omnipresent phosphatases and by nonenzymatic oxidation. Based on these biochemical aspects of pteridine synthesis in human cells, neopterin has been recognized to be a useful in vivo indicator of the activation state of the cellular immune system (4-6). For a broad spectrum of diseases, all of which are characterized by involvement of activation of T-lymphocytes and macrophages,, interesting applications of neopterin measurements have been reported by a multitude of Received 5/11/92; accepted 10/30/92. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. To whom requests for reprints should be addressed, at Institut fiir Medizinische Chemie und Biochemie, Fritz Pregl Strasse 3, A-6020 lnnsbruck, Austria. research groups (for a comprehensive review, see Ref. 6). In malignant diseases, increased neopterin concentrations have been observed in various tumor types. Frequently, a positive correlation was found between high neopterin levels and advanced tumor stage, and in all cancer types investigated so far with respect to long-term prognosis, high neopterin levels were significantly pointing to a poor outcome of the disease (reviewed in Ref. 7). In a small pilot study, only about 50% of patients with colorectal carcinomas showed pretherapeutically raised neopterin levels, and no dependence on tumor stage was apparent (8). Thus, it was particularly interesting to study the potential predictive value of pretherapeutically measured neopterin concentration in this patient group because several earlier investigations have pointed out that neopterin as a predictor of death from cancer is statistically independent of other clinically important predictors such as tumor stage. M A T E R I A L S AND M E T H O D S Patients. Forty-four patients with adenocarcinoma of the colon were included in the study. All investigations reported here were performed before initiation of therapy. There were 25 men and 19 women; the ages ranged from 45 to 84 yr with a median age of 71 yr and an interquartile range from 59 to 76 yr. There were 3 patients with tumor Stage Tt, 7 with Stage T2, 28 with Stage T3, and 6 with Stage T4. The distribution of patients according to Dukes' classification was: 6 patients with Dukes' Stage A; 14 with Dukes' Stage B; 15 with Dukes' Stage C; and 9 with Dukes' Stage D. There were 15 patients with highly differentiated tumors, 26 with moderately differentiated tumors, and 3 patients with poorly differentiated tumors. The patients were uniformly treated by surgery. In addition, one patient received radiation therapy, and one other patient received chemotherapy. The possible effects of chemotherapy and radiation therapy were not analyzable because of the small number of patients receiving such treatment. During follow-up, four patients experienced a recurrence of their disease; these four plus one additional patient were operated on again. Two patients were diagnosed later to have a secondary malignancy. Because of the small number of these complications, no statistical analyses were performed to investigate their effects on survival. Laboratory Examinations. Neopterin concentrations in first morning urine specimens were determined by an optimized and fully automated highpressure liquid chromatography technique without oxidative pretreatment. The procedure is described in detail elsewhere (6). In brief, we used an LC 5500 chromatograph (Varian, Palo Alto, CA) controlled by a Vista 402 data system (Varian), Fluorichrom florescence detector (Varian), UV 200 UV detector (Varian), and a 10-1al injection valve (Rheodyne, Berkeley, CA). A ready-to-use C~8 reversed-phase column (4 x 125 mm) was used, with 7-pm-diameter packing (Merck, Darmstadt, Germany) together with a guard column (4 x 4 mm) with the same material (Merck). The mobile phase was potassium phosphate buffer (15 mM, pH 6.4), and the flow rate was 0.8 ml per min. One hundred pi of urine were diluted in 1 ml of mobile phase, and 10 pl were injected. By this reverse-phase technique, urinary creatinine is simultaneously determined in the same chromatographic run. To compensate for physiological variations of urine concentrations, neopterin levels are related to these creatinine values and are expressed as lamol of neopterin/mol of creatinine. The native fluorescence of neopterin at 353-nm excitation wavelength and 438-nm emission wavelength is used for detection; creatinine is quantified based on the UV absorption at 235 nm. The performance characteristics of this technique were reported previously (6). Analytical sensitivity is 120 fmol of neopterin per injection and 36 pmol of creatinine per injection at a peak:noise ratio of 5:1. Thus, the detection limit is 72 nmol of neopterin per 1 liter of urine which is one order of magnitude