Therapeutic Efficacy of Fluoropyrimidines Depends on the Duration of Thymidylate Synthase Inhibition in the Murine Colon 26-B Carcinoma Tumor Model1

5-Fluorouracil (FUra) and 5-fluoro-2’ -deoxyuridine (FdUrd) are common chemotherapeutic drugs for the treatment of advanced colorectal cancer. Two recognized mechanisms of action of these agents are inhibition of thymidylate synthase (TS) and incorporation of fluorinated UTP into cellular RNA. In previous studies on drug scheduling of both fluoropyrimidines, we observed the highest therapeutic efficacy by using a weekly i.v. push schedule. Furthermore, weekly 400-mg/kg FdUrd is superior to equitoxic weekly 80-mg/kg FUra in murine Colon 26-B carcinoma. We evaluated the most important pharmacokinetic and pharmacodynamic effects of both fluoropyrimidines to delineate the biochemical mechanisms underlying their differences in therapeutic activity in this tumor model. FUra concentrations and elimination in tumors after FdUrd or FUra administration were comparable, and the level of FUra incorporation into cellular RNA following treatment with FUra or FdUrd was similar Free tumoral 5-fluoro-dUMP levels were initially 3-fold higher after FdUrd but diminished rapidly thereafter. The number of free [3H]5-fluoro-dUMPbinding sites decreased to about 25 and 15% of control values within 2 h after treatment with equitoxic doses of FUra and FdUrd and remained low for 72 h. The duration of TS inhibition was significantly longer following treatment with FdUrd compared with FUra, 168 and 72 h, respectively. The superiority of the antitumor activity of an i.v. push of FdUrd over FUra in the treatment of Colon 26-B tumors correlates with maintenance of TS inhibition and repeated drug administration when TS remains low, whereas FUra incorporation into RNA does not appear to distinguish the antitumor response of FdUrd from that of FUra in this tumor model. INTRODUCTION The fluoropyrimidines FdUrd3 and FUra are the most effective drugs in the treatment of patients with advanced cobrectal cancer ( 1 , 2). These agents exert their action after being metabolized to the nucleotide level. One of these nucleotides, FdUMP, is a potent inhibitor of TS, the key enzyme in pyrimidine de tioi’o synthesis (2-6). The inhibition can be enhanced by the natural cosubstrate of TS, 5, lO-methylenetetrahydrofolate. LV is the precursor of this reduced folate. which is essential for the formation of a stable ternary complex with TS and FdUMP (3, 7-10). FdUrd is metabolized by either phosphorylation into FdUMP catalyzed by thymidine kinase or by cleavage catalyzed by TP into FUra and can thus act as a prodrug for FUra (I 1-14). FUra is further anabolized via 5-fluoro-UMP into 5-fluoro-UDP and FUTP and subsequently into FdUMP or 5-fluoro-dUTP. FUTP and the latter can be incorporated into RNA and DNA. respectively (2, 15-17). The antiproliferative effects of both drugs are believed to be caused by the inhibition of TS (2, 5, 9, 18, 19), although the RNA incorporation of FUTP has also been associated with the antitumor activity of FUra therapy ( I 5, 20). Iii vitro FdUrd is generally more cytotoxic than FUra (18. 2 1 ); the highest cytotoxicity is seen after prolonged exposure (22). FdUrd was already being given by continuous systemic infusion in clinics in the l960s; however, it was associated with considerable toxicity, and a 15to 20-fold dose reduction was required (23). More importantly, these continuous infusions of FdUrd were not more effective than other treatments (24). At present, the use of FdUrd is restricted to selective high-dose regimens by hepatic arterial infusion in patients with colorectal liver metastases (1, 25), or it is given in a chronomodulated modulated schedule (26). When given as a continuous infusion, FUra does not require dose reduction. It can be administered in different schedules (27). Recently, our group showed clinically relevant superiority of high-dose iv. push FdUrd administration in a FUra-resistant murine colon tumor. Colon carcinoma 26 (here designated Cobon 26-B) was considered resistant, because treatment with FUra did not increase the median life span of mice substantially, and no partial or complete responses had been achieved. Furthermore, in this tumor model, weekly iv. push administration of FdUrd at its MTD resulted in greater therapeutic activity and selectivity than with continuous infusions during 4 days, three Received 1 1/2 1/95: revised 4/22/96: accepted 5/6/96. t This study was supported in part by Dutch Cancer Society Grant IKA VU 92-88 and American Cancer Society Grant OHP 71474 (Y. M. R.). 2 To whom requests for reprints should be addressed. 3 The abbreviations used are: FdUrd, 5-fluoro-2’-deoxyuridine: FUra. 5-fluorouracil: FdUMP. 5-fluoro-dUMP: TS, thymidylate synthase: LV. leucovorin; FUTP, 5-fluoro-UTP, MTD. maximum tolerated dose: ww. wet weight: TP. thymidine phosphorylase; AUC. area under the concentration curve. Research. on October 16, 2017. © 1996 American Association for Cancer clincancerres.aacrjournals.org Downloaded from 1328 Efficacy of Fluoropyrimidines in Colon 26-B Tumors and TS weekly 1-h infusions, or iv. push daily for 4 days (28). These observations led to the start of a Phase I and pharmacokinetic trial with a high-dose iv. push of FdUrd at the Roswell Park Cancer Institute (14). The present study was designed to delineate biochemical mechanisms associated with the observed differences in antitumor activity of FdUrd and FUra in Colon 26-B carcinoma. Using equitoxic doses of FdUrd or FUra, tumor tissue concentrations of FUra and free FdUMP, levels of FUTP incorporation into cellular RNA, and the duration of TS inhibition were measured. The results of this study indicate that the duration of TS inhibition is consistent with the observed differences in antitumor activity between the two fluoropyrimidines. MATERIALS AND METHODS Chemicals. FUra and FdUrd were obtained from Hoffmann-La Roche, Inc. (Basel, Switzerland) [5-3H]dUMP was obtained from Amersham International (Buckinghamshire, United Kingdom), and [6-3H1-FdUMP was from Moravek (Brea, CA). All other chemicals were of the highest quality commercially available. Mice. Female BALB/c mice (Harlan/Cpb, Zeist, the Netherlands), 6-7 weeks old, were kept six per cage with water and food ad libitum. Colon 26-B tumor (28, 29) fragments were inoculated in both flanks of the mice under slight ether anesthesia. Ten days after transplantation, when the tumors had reached a size of about 200 mm’, drugs were administered. The antitumor effects of FUra (80 mg/kg) and FdUrd (400 mg/kg) given as bolus injections were evaluated as described (28). Tissue Preparation. Tumors for biochemical analyses were removed 2, 8, 24, 48, 72, 168, and 240 h after drug administrations. These are the time points when tumor volumes were measured. More points were not considered more informative and would have resulted in needless sacrifice of too many mice. Tumors were removed rapidly during ether anesthesia, immediately frozen in liquid nitrogen, and subsequently kept at -80#{176}C until assayed. Frozen tumors were pulverized with a microdismembrator, and the still-frozen powder was suspended in ice-cold Tris buffer [200 ms Tris, 15 mn CMP, 100 nmi NaF, and 20 mM 3-mercaptoethanol (pH 7.4)] at a concentration of 0.33 g/ml. The suspension was centrifuged at 4000 X g, and the supernatant was subsequently centrifuged at 7000 X g. This enzyme-containing supernatant (enzyme suspension) was split in several parts for the different assays, thus allowing the use of one tumor sample for all analyses. Fifty p.1 were used for the estimation of the protein content of the tumor, as measured with Coomassie blue staining (Bio-Rad; according to the method of Bradford; Ref. 30); 400 p.1 were used for the determination of FUra and FdUrd and free FdUMP levels. The remaining volume was used for the two assays, by which TS inhibition could be evaluated (free FdUMP-binding sites and TS catalytic activity). The 4000 x g precipitate was immediately frozen (80#{176}C)and later used for RNA incorporation measurements. FUra, FdUrd, and FdUMP Measurements. Because prolonged high concentrations of FUra in human and murine tumors have been observed (3 1 ), we evaluated the tumoral FUra concentrations together with FdUrd concentrations. After extraction of the 7000 X g supernatant with trichboroacetic acid (final concentration, 5%), precipitation, and neutralization with trioctylamine/l,l,2-trifluorotrichloroethane, the measurement of FUra levels in tumors was performed by gas chromatography coupled with mass spectrometry with [‘5N, ‘5N]FUra as an internal standard (31). FdUrd concentrations were analyzed by further extraction with ethyl acetate and methanol as described for plasma evaluations (14). Tumoral levels of FdUMP were determined in these trichloroacetic acid extracts using the isotope dilution assay with Lactobacillus casei TS. [6-3H]FdUMP was used as the radioactive substrate (8). The detection limit was 10 fmol/mg ww. FdUMP levels were too low for direct measurement by high-performance liquid chromatographic procedures; in addition, in tumor samples, too many interfering peaks preclude selective determination of nonradioactive FdUMP. Evaluation of TS Inhibition. Two assays were used for the evaluation of TS levels in tumors from drug-treated mice. The ligand-binding assay of [6-3HJFdUMP to TS gives an indication of the number of free binding sites for FdUMP of TS after treatment. In the TS catalytic activity assay, the conversion of [5-3H]dUMP into dTMP and 3H20 is measured. In tumors from early time points (2 h-3 days), TS was measured immediately after preparation of the extracts essentially as described (8, 9). Tumors from untreated mice served as references. Tumor samples from the later two time points (168 and 240 h) were split into two parts. In one part, a dissociation buffer [0.75 M NH4CO3, 100 mrvi NaF, 20 mrvi 3-mercaptoethanol, and 15 mrvi CMP (pH 7.8) with 0.05 total volume of 1 .6 mM dUMP] was used to remove all the FdUMP from the ternary complex, leaving the total amount of FdUMP-binding sites (TS-tot) or the total activity of the enzyme (TS-total). These samples were incubated for 3 h at 30#{176}C. The other part was frozen (-80#{176}C) during the dissociation procedure, and TS levels were measured thereafter. In these samples, inhibition was measured as the number of free FdUMP-binding sites or residual TS catalytic activity. Values were then calculated as relative to those of untreated tumors. Under these conditions, no loss of enzyme levels were observed (8). Further details on both assays have been described elsewhere (8). Determination of RNA Incorporation. After addition of the Tris buffer to the powder and centrifugation, the precipitated part was stored and used for further analysis of RNA incorporation by FUTP. For this purpose, a recently developed technique, described in detail, was used (32). This assay is based on degradation of purified RNA substituted by FUra, to FUra, by incubation of the RNA with RNase, alkaline phosphatase, and uridine phosphorylase. Hereafter, FUra was extracted and derivatized for measurement by gas chromatography coupled with mass spectrometry as described (31, 32). Pharmacokinetic and Statistical Evaluation. The AUC was determined according to the trapezoidal method. Other pharmacokinetic parameters such as t’/2 were determined as described previously (28, 31). Statistical evaluation was accomplished with Student’s t test for unpaired data. Statistical significance was assumed when P < 0.05. Research. on October 16, 2017. © 1996 American Association for Cancer clincancerres.aacrjournals.org Downloaded from