Fluoropyrimidine chemotherapy in a rat model: comparison of fluorouracil metabolite profiles determined by high‐field ¹⁹F‐NMR spectroscopy of tissues ex vivo with therapy response and toxicity for locoregional vs systemic infusion protocols

Abstract

A Sprague-Dawley rat model with DS sarcoma transplanted in the thigh was used to compare transcatheter locoregional i.a. and systemic i.v. administration of 5-fluorouracil (FU) at 12 dose-rate schedules: 25, 50 and 100 mg/kg; bolus, 1, 5 and 24 h infusions. In experiment A tumor (62/67 animals) as well as liver and kidney (56/67 animals) were excised 1 h after a single bolus or 1 h infusion or at the end of 5 and 24 h infusions. (19)F-NMR spectroscopy at 11.7 T was used to quantitate FU and its metabolites in ca. 1 g of tissue at 4 degrees C. In experiment B analogous FU treatments were repeated for 5 days (rats 80+11 controls). Tumor volumes vs time, various blood parameters and survival times were recorded, and a log growth rate parameter log GR, a response index RI, and a toxicity index TI were calculated. The i.a. vs i.v. ratios for tumor concentrations of FU and total anabolites (F-Nucl) were >1 for nearly all treatments and increased with infusion time at the higher doses. F-Nucl in tumor correlated linearly with total fluorine concentration (Tot. F range 30-1100 nmol/g) over all treatments (r=0.92, slope=0.45, p<0.0001). For non-bolus i.v. treatments [FU+F-Nucl] decreased linearly with decreasing FU dose rate (r(2)=0.74, zero intercept), while i.a. treatments showed non-linear behavior. For non-bolus treatments the mean log GR per treatment group showed a negative correlation (r=-0.87) with log[F-Nucl]. The most effective non-toxic treatments were 25 mg/kg over 5 or 24 h; the i.a. route was superior to i.v. on the basis of [FU+F-Nucl], RI, the reduction in log GR, and Kaplan-Meier survival statistics. For liver and kidney Tot. F (>83% FU catabolites) reached ca. 3-4 and 6-7 micromol/g, respectively, at the highest dose rates for either route; F-Nucl were detected only for Tot. F>500 nmol/g and increased exponentially as Tot. F increased (toxic treatments). The concentrations of the main catabolite (alpha-fluoro-beta-alanine, FBAL) in tumor did not correlate with Tot. F but rather with FBAL levels in kidney (r=0.90, all treatments), indicating that uptake of liver-derived FBAL from the circulation is the major source of FBAL in tumor.

Fluorine-19 or phosphorus-31 NMR spectroscopy: A suitable analytical technique for quantitative in vitro metabolic studies of fluorinated or phosphorylated drugs

“…Hull and coworkers are the sole to develop a 19 F NMR methodology for the quantification of fluoropyrimidine metabolites in intact human or animal excised tissues at 4 • C [42,45,76]. Their previous study on FdUrd metabolism in Novikoff hepatoma in the rat [42] is completed by an important study of FU metabolism in a rat model with DS sarcoma transplanted in the thigh for a wide range of therapy protocols [45].…”

“…For 19 F NMR analyses, FBEN in D 2 O is employed by Martino's group [34][35][36][37] and Kamm et al [38,39], C 6 F 6 and, since 1990, 1,2-difluorotetrachloroethane (DFTCE) in benzene-d6 by Hull's group [40][41][42][43][44][45], NaF in D 2 O by Wyrwicz and coworkers [46]. Hexachlorocyclotriphosphazene (HCCP) in toluene-d8 [23,24] then MPA in D 2 O [20] are utilised by Martino's group for 31 P NMR analyses.…”

Monitoring fluoropyrimidine metabolism in solid tumors with in vivo 19F magnetic resonance spectroscopy

“…administration of FBAL. Also, in ex vivo 19 F MRS studies of excised tumors the FBAL concentration in tumor did not correlate with the total fluorine concentration in these tumors but rather with plasma levels or concentrations in kidney [23,68]. Thus, FBAL detected in tumor may not necessarily have been produced in tumor tissue but may be taken up from the circulation after clearance from the liver [69].…”

“…In clinical studies, in vivo concentrations of fluorine-containing metabolites must be larger than ca. 200 nmol/g tissue to be detectable in reasonable measurement times (<5-15 min) at B 0 = 1-2 T. For animal studies at 7 T, 50 nmol/g can be quantitated in 30 min and in ex vivo studies 5-10 nmol can be detected in 0.5-1 g tissue samples in 1-2 h at 11.7 T on a high-resolution MR system (see [23] and references therein). Therefore, any method which provides an increase in SNR is relevant in 19 F MRS, especially in clinical applications.…”

“…High levels of FBAL have been associated with neurotoxicity and cardiotoxicity, possibly due to its conversion into ␣-fluoro-␤-hydroxypropionate (FHPA) and fluoroacetate (FAC) [20][21][22]. Finally, fluoride (F − ) can be released from FBAL via enzymatic reactions, which are particularly efficient in rat liver [23].…”