The metabolism of 5-fluorouracil (5-FU) was monitored non-invasively in two xenografts, a hypopharynx carcinoma and a colon carcinoma (CSM) by 19F-magnetic resonance spectroscopy following an i.v. bolus injection of 130 mg kg-1 5-FU. Both the level of fluoronucleotides (FNuc) and the tumor growth delay were significantly higher in the CSM colon carcinoma than in the hypopharynx carcinoma (both parameters, P < 0.001). Administration of 100 mg kg-1 methotrexate (MTX) at 15 h before treatment with 5-FU caused a significantly increased conversion of 5-FU to FNuc in both tumors (P < 0.05) as compared with the application of 5-FU alone. However, only in the CSM tumor was a significantly increased growth delay (P < 0.01) observed. Pre-treatment of both xenografts with 400 mg kg-1 thymidine enhanced the conversion of 5-FU to FNuc in both tumors. In the CSM tumour this treatment modality caused a significantly (P < 0.05) higher growth delay as compared with the results obtained with 5-FU alone, whereas in the hypopharynx carcinoma the additional application of thymidine caused no significant change in tumor growth. It is known that both thymidine and MTX can reduce the DNA-directed cytotoxicity of 5-FU, whereas the RNA-directed cytotoxicity is increased. It is concluded that the DNA-mediated toxicity may be more important in the hypopharynx carcinoma than in the CSM colon carcinoma. As a consequence, pre-treatment with MTX or thymidine enhances FNuc formation, although only in the CSM carcinoma is there an increased tumor growth delay. Thus, in the hypopharynx carcinoma the measurement of FNuc did not serve as a predictor for the treatment efficacy of the combined treatment modality. Pre-treatment with MTX did not influence the catabolism of 5-FU, whereas thymidine actually prolonged the half-life of 5-FU without alpha-fluoro-beta-alanine becoming detectable.

Purpose: An early indicator of tumor sensitivity to irradiation could provide useful information on the eectiveness of therapy and may facilitate more individual designs of treatment protocols. The aim of the present study was to evaluate the potential of in vivo 31 P nuclear magnetic resonance spectroscopy in predicting the response of a xenografted human hypopharynx carcinoma to radiotherapy. Methods: The tumor had been serially heterotransplanted to athymic mice. 31 P NMR spectra were collected before and at four intervals (24, 48, 72, and 120 h) after irradiation with 15 Gy or 30 Gy. Alterations of phosphorus metabolism were compared with the growth delays, the histological appearance, and the mitotic activity of the treated tumors. Results: Radiation with 30 Gy induced increases of the phosphodiester level (P < 0.001) as well as of the tumor pH (P < 0.05) and decreases of the phosphomonoester level (P < 0.001) within 48 h. The changes clearly preceded measurable tumor responses and were accompanied by severe histological destruction and marked depression of mitotic indices. However, none of these spectral alterations was signi®cantly correlated with individual delays of tumor growth. The only parameters allowing a prediction of radiation-induced tumor responses were the pre-treatment levels of phosphomonoesters and -diesters. The 31 P NMR spectroscopic changes observed after therapy with 15 Gy were either unsystematic or insigni®cant. Conclusions: Pretreatment levels of tumor phospholipids were indicative of radiosensitivity in the xenografted human hypopharynx carcinoma investigated here. However, since phosphorus metabolism varies considerably among different tumor lines, it seems unlikely that there exists a uniform 31 P NMR spectroscopic parameter predicting tumor response to radiation therapy.Key words 31 P NMR spectroscopy á Hypopharynx carcinoma á Radiotherapy Abbreviations PCr phosphocreatine á PME phosphomonoesters á PDE phosphodiesters á P i inorganic phosphate á TPC total phosphorus content

Between May 1995 and March 1996 we were able to diagnose and remove glomus tumors of the left nasal cavity from three female patients. Ages of the patients were 9, 36 and 74 years. All patients suffered from a frequent epistaxis and all were extremely sensitive to the slightest nasal touch. One patient reported breathing difficulties due to nasal congestion. Examination revealed a tumor that filled the entire nasal cavity. The method of choice in treatment of these tumors is surgical removal. Hemangiopericytoma, non-chromaffin paraganglioma, hidradenoma, cavernous hemangioma and nevus cell nevi have to be excluded by histology and immunohistochemical techniques. From a clinical perspective the bleeding septal polyp (granuloma telangiectaticum sive pyogenicum sive pediculatum) has to be considered because it often comes from Kieselbachi's plexus, has a mushroom-like appearance and bleeds slightly.

A xenografted hypopharynx carcinoma growing subcutaneously in nude mice was studied by in vivo 31P NMR spectroscopy during uninfluenced growth and following treatment with cisplatin (CDDP). Parallel to the NMR experiments, the cytokinetic and histological changes in the tumor were investigated. The most significant change in the growing tumor was a decline in the level of phosphocreatine (PCr), whereas the tumor pH did not change. Following treatment with CDDP (4, 8 and 12 mg/kg), a dose-dependent decrease in the level of phosphomonoesters (PME) took place, whilst no dose dependence could be observed for the increase of PCr. the pH shifted to alkaline only after administration of the highest CDDP dose. Tumor cytokinetics revealed a cell arrest at the G1/S boundary 24 h after chemotherapy. At this time, the histological sections showed a dilatation of capillaries, whereas first necroses appeared on day 3. The proliferative activity of the tumor showed a sharp decline 24 h after CDDP application, followed by a revival of cell proliferation that was proportional to the dose applied between days 5 and 7. This increase in proliferative activity was paralleled by a marked increase in the PME/phosphodiesters ratio. Thus, in the tumor investigated the PME were the best indicators of tumor response to therapy. A precise correlation between the cytokinetic data and the re-energization of the tumor was not possible because histological changes, which may contribute to improved tumor energy status took place at the same time.