Measurement of plasma 5-fluorouracil by high-performance liquid chromatography with comparison of results to tissue drug levels observed using in vivo 19F magnetic resonance spectroscopy in patients on a protracted venous infusion with or without interferon-alpha.

PURPOSE To measure plasma 5-fluorouracil (5-FU) levels using high-performance liquid chromatography (HPLC) and compare the findings to the tissue metabolism of 5-FU evaluated using 19F magnetic resonance spectroscopy (MRS), during a protracted venous infusion (PVI) with or without interferon-alpha. METHODS Patients receiving PVI 5-FU (300 mg/m2/day) with or without interferon-alpha (5 x 10(6) units 3 times per week), had 2 weekly plasma 5-FU levels evaluated using reverse-phase ion-pairing HPLC. These samples were drawn just prior to the patient undergoing MRS using a 1.5T Siemens Magnetom whole body magnetic resonance system with a 16 cm surface coil placed over normal liver or metastatic tumour. Semi-quantitated MRS values were compared with the plasma 5-FU levels using linear regression analysis. Data were available from patients given interferon-alpha with PVI 5-FU from day 1 or at the point of 5-FU refractory disease. RESULTS A total of 30 patients were studied. Plasma 5-FU concentrations while on a protracted venous infusion varied from <25 ng/ml (0.192 mu M) to 25,000 ng/ml (192 mu M). A high plasma 5-FU concentration was associated with an increase in patient toxicity. Patients given interferon-alpha with 5-FU had higher median plasma 5-FU levels higher than patients on 5-FU alone (6138 vs. 218 ng/ml; p = 0.03). There was no correlation between the plasma 5-FU concentration and tumour response. A comparison of the plasma 5-FU data to the MRS studies in normal liver revealed a positive correlation between plasma 5-FU and liver catabolite signal (r = 0.68; p = 0.016) but a negative correlation with the log plasma 5-FU concentration and 5-FU liver signal (r = -0.63; p = 0.022). The patients experiencing toxicity, in addition to having a higher plasma 5-FU concentration did not exhibit a liver 5-FU signal, while the reverse was true for those having no toxicity. CONCLUSIONS Plasma 5-FU levels may show greater interpatient variation when given as a protracted venous infusion. Levels of 5-FU correlated with treatment toxicity but not with anti-tumour activity. The addition of interferon-alpha to 5-FU increases plasma 5-FU levels. MRS findings suggest patients with low plasma 5-FU levels have higher 5-FU levels in normal liver tissue than in those with higher plasma levels.