Biochemical Assessment of the Effects of Acivicin and Dipyridamole Given As a Continuous 72-Hour Intravenous Infusion1

Home , CTP synthetase

(CANCER RESEARCH 48, 5591-5596. October 1, 1988] Biochemical Assessment of the Effects of Acivicin and Dipyridamole Given as a Continuous 72-Hour Intravenous Infusion1

Paul H. Fischer,1 James K. V. Willson,2 ConcepciÃ³n RisueÃ±o,Rendra l titsch, Joan Bruggink, Alan Ranhosky, and Donald L. Trump Department of Human Oncology, University of Wisconsin Clinical Cancer Center, Madison, Wisconsin 53792 [P. H. F., J. K. V. W., C. K., K. T., J. B., D. L. T.J; William S. Middleton Memorial Veterans Medical Center, Madison, Wisconsin 53705 Â¡J. K. V. W., D. L. T.]; and Boehringer-lngelheim, Inc., Ridgffleld, Connecticut [A. R.J

ABSTRACT in mediating resistance to a variety of antimetabolites, we have conducted a Phase I trial of acivicin and DP3 simultaneously Since this Phase I trial was based on a strategy of biochemical administered as 72-h continuous i.v. infusions. This trial was modulation, namely, the inhibition of nucleoside uptake by dipyridamole, a biochemical assessment of the actions of acivicin and dipyridamole was based on the ability of DP, an inhibitor of nucleoside transport undertaken in order to aid our interpretation of the clinical findings. The (1-4), to potentiate the cytotoxic effects of acivicin (5-8), an primary biochemical objectives of this trial were: (a) to determine whether inhibitor of de novo pyrimidine and purine biosynthesis (9-11). plasma levels of dipyridamole sufficient to inhibit nucleoside uptake The initial studies demonstrated that cancer cells could avidly could be achieved with a 72-h continuous i.v. infusion; (b) to monitor the salvage nucleic acid precursors and replete their nucleotide effects of acivicin on two key enzymatic targets, CTP synthetase and pools, thus bypassing the metabolic blocks of acivicin (5). DP GMP synthetase; and (c) to evaluate changes in cellular ribonucleoside effectively reduced nucleoside salvage and restored the cytotox- triphosphate pools during therapy. Since peripheral blood mononuclear icity of acivicin (5-8). Since this Phase I trial was based on a cells have relevant biochemical targets and can be serially obtained during the course of therapy, the biochemical effects of acivicin and dipyridamole strategy of biochemical modulation, we felt that interpretation were determined in these cells. At the maximally tolerated dose of of the clinical findings obtained in this study would be enhanced dipyridamole (23.1 mg/kg/72 h), the steady-state concentrations of total if a biochemical assessment of the effects of acivicin and DP and free dipyridamole averaged 11.9 n\i and 27.8 IIM,respectively. These could be obtained. The key questions were: (a) could plasma levels were sufficient to inhibit cytidine (l /IM) uptake by greater than levels of DP sufficient to inhibit nucleoside uptake be sustained 50% in the lymphocytes of five of six patients so treated. Using lympho cytes obtained from 14 normal volunteers the concentration of free for 72 h, and (b) was acivicin effectively inhibiting the synthesis dipyridamole needed to inhibit the uptake of l MM cytidine by 50% of de novo purine or pyrimidine biosynthesis? averaged 13.8 Â±1.1 IIM.The plasma levels of a,-acid glycoprotein, which Ideally such biochemical measurements should be made in tightly binds dipyridamole, ranged from 60 to 300 mg/dl in the patients the tumor cells. However, in patients with solid tumors, it is in this study. As a consequence there were wide variations in the per very difficult to obtain serial tumor biopsies, and another source centage of dipyridamole present as the unbound, pharmacologically active of tissue was needed. Peripheral blood mononuclear cells were form and in the rates of dipyridamole clearance. The decreased rate of dipyridamole clearance seen in patients with high levels of in-acid used to assess the biochemical effects of acivicin and DP for glycoprotein resulted in higher plasma concentrations of total dipyrida several reasons: (a) they can be routinely obtained prior to and mole and compensated for the reduced fraction of free drug. Therefore, during therapy; (h) nucleoside transport in lymphocytes is the plasma concentration of free dipyridamole varied much less than the facilitated by a DP-sensitive carrier; (c) these cells express the total drug concentration in these patients. enzymes of de novo nucleotide biosynthesis, including the aciv- CTP synthetase and GMP synthetase activities were measured in icin-sensitive targets CTP synthetase and GMP synthetase; and patients' peripheral mononuclear cells prior to and at various times during (d) the RTP pools in these cells are large enough to be routinely therapy. CTP synthetase activity was inhibited in a time-dependent fashion by greater than 75% in seven of 13 Ã©valuablecourses; GMP assayed. synthetase was similarly inhibited in only three of ten cases. Ribonucle The results from this study indicate that concentrations of oside triphosphate pools were also measured in the patient's lymphocytes. DP sufficient to inhibit cytidine transport can be attained with CTP pool reductions of 30 to 50% were seen in nine of 19 courses, but a 72-h continuous infusion. However, interpretation of total in only four cases was the inhibition greater than 50%. Similarly, in six DP levels and extrapolation of in vitro findings to the clinical of 19 courses GTP pool reduction of 30 to 50% was evident, and in four situation are complicated by the presence of AGP. Since DP of 19 cases the inhibition was greater than 50%. Considering data from avidly binds to AGP (12-14), it was necessary to measure the all courses, drug therapy did not significantly reduce any of the ribonu cleoside triphosphate pools. Taken together, these results suggest that plasma levels of this protein and then determine whether blood levels of dipyridamole sufficient to inhibit nucleoside salvage can changes in AGP levels alter the pharmacology of DP. We found be achieved in vivo; however, the lack of a consistent, pronounced effect that AGP levels varied widely among the patients and signifi of acivicin on de novo nucleotide biosynthesis precludes analysis of the cantly influenced DP,, the percentage of unbound DP and the role of salvage in modulating the toxicity of acivicin in vivo. rate of DP clearance. Techniques for the reproducible, serial measurement of CTP synthetase, GMP synthetase, and ribo INTRODUCTION nucleoside triphosphate pools in the lymphocytes of patients on study were developed; however, marked inhibition of de novo As part of an effort to evaluate the role of nucleoside salvage nucleotide synthesis by acivicin was not consistently seen in the Received 10/8/87; revised 4/7/88; accepted 6/24/88. lymphocytes of these patients during the course of this trial, The costs of publication of this article were defrayed in part by the payment and it was difficult to evaluate the influence of nucleoside 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. salvage on the action of acivicin in this study. ' Supported in part by National Cancer Institute Grant CM-47663-28 and a grant from Boehringer-lngelheim. Inc. 3The abbreviations used are: DP. dipyridamole; DP,, concentration of free and 1To whom requests for reprints should be addressed, at Hematology/Oncol- bound DP; DPr, concentration of free DP; AGP, Â«i-acidglycoprotein; DMEM, ogy. University Hospitals of Cleveland, Case Western Resene University School Dulbecco's modified Eagle's medium; HPLC, high-performance liquid chroma- of Medicine, 2074 Abington Road. Cleveland, OH 44106. tography; PBS, phosphate-buffered saline; RTP. ribonucleoside 5'-triphosphate. 5591

MATERIALS AND METHODS Co. using a specific microbiological assay which has been described in detail (17). During the course of the Phase I trial of acivicin (60 mg/nr, contin uous infusion over 72 h) and escalating doses of dipyridamole,3 the Nucleoside Uptake. Since DP is bound to AGP, nucleoside uptake assays were carried out using lymphocytes resuspended in DMEM to pharmacodynamics and biochemistry of acivicin and dipyridamole were determine the sensitivity of the cells to inhibition by DP. Lymphocytes studied as follows: heparinized peripheral venous blood samples were obtained in 10-ml plain glass Vacutainer tubes at the following times resuspended in plasma were used to determine the in vivoconcentration of DP required to inhibit nucleoside uptake and to assess the influence during the course of treatment: predrug and at 24, 48, and 72 h during of AGP concentration on DP's actions. the infusion. The plasma was separated promptly and kept frozen at -70Â°Cuntil analyzed by the assay methods described below. Mononu- Cells (approximately 1 x IO6)were incubated for 60 min at 37Â°Cin a 5% CÃ›2atmosphere with either 1 ^M [3H]uridine (10 (iCi/nmol) or clear cells were obtained from heparinized blood samples by centrifu- [3H]cytidine (10 ^Ci/nmol) in the presence or absence of various con gation on a Ficoll-Hypaque gradient for 30 min at 1000 rpm. The cells at the interface of the gradient were removed, resuspended in 10 ml of centrations of DP. The cells were collected by centrifugation for 10s at 12,000 rpm and then extracted for 15 min with 1.0 ml of ice-cold DMEM, and counted using a hemacytometer. A portion of the cells was collected by centrifugation for analysis of RTF pools. The remain 0.5 M HCIO4. The acid was neutralized with 2 volumes of freorr.alanine ing cells were diluted with an additional 10 ml of DMEM and then (18), and the aqueous layer was collected. The nucleosides and nucleo- collected by centrifugation: this wash procedure was repeated 2 more tides were separated by ion exchange chromatography using Dowex times. The cells were counted, split into two aliquots, centrifuged, and AG 1-X8, 100- to 200-mesh resin as the formate. The nucleosides were then resuspended in either DMEM or the patient's plasma. These cells eluted with 10 ml of water, and the nucleotides were removed using 10 were used for the enzymatic assays and nucleoside uptake experiments. ml of 8 M formic acid:l M ammonium formate. Amersham (Arlington Heights. IL) supplied the [G-3H]hypoxanthine, CTP Synthetase Assay. An in situ assay measuring the conversion of and the [5-3H]cytidine and [5-3H]uridine were purchased from Moravek UTP to CTP was used to determine the activity of CTP synthetase in Biochemicals (Brea, CA). Basic drugs known to be bound to Â«i-acid the intact cells. Approximately 1 x IO6 peripheral mononuclear cells glycoprotein were not administered during the dipyridamole infusion. were incubated with 1 pM [5-3H]uridine (specific activity, 20 ^Ci/nmol) Patients who required continuation of such drugs were excluded. for 60 min. The cells were collected by centrifugation, washed with Dipyridamole Assay. The HPLC assay for DP in plasma, adapted PBS, centrifuged, and then extracted for 20 min in 0.5 ml of 0.5 N from published methods (15, 16), utilized a Spectra-Physics SP8770 HC1O4. The extract was clarified by centrifugation, neutralized, lyoph- pump, Gilson Spectroglo Fluorometric detector (285-nm excitation, ilized, and resuspended in water. A portion of the sample was mixed 470-nm emission), LDC/Milton Roy Model C1-10B integrator, and a with authentic CTP, UTP, GTP, and ATP, and the [3H]UTP and Rheodyne Model 7125 injector with a 50-Â¿illoop. A Waters Cig- [3H]CTP were separated by HPLC using a Spherisorb SAX 10- x 250- Bondapak reverse-phase column was used with a mobile phase of 65% mm ion exchange column monitored at 254 nm. The retention times methanol:35% water containing 5 HIMheptane sulfonic acid and 0.1% averaged 11.3 and 14.8 min for CTP and UTP, respectively, using a acetic acid. At a flow rate of 1.8 mi/min the retention time for DP was 0.3 M Na2HPO4 buffer at pH 3.5 and a flow rate of 2 ml/min. Fractions 4.1 min, and that of the internal standard, RA 433, a trimorpholine were collected every 0.5 min, and the radioactivity was determined by analogue of DP (Boehringer Ingelheim, Ridgefield, CT), was 3.1 min. liquid scintillation spectrometry after the addition of 3 ml of ACS DP was extracted from 0.5 ml of plasma with 10 ml of 5% isobutanol (Amersham). In order to account for possible differences in the degree in dichloromethane after the addition of 100 ng of internal standard of [3H]uridine uptake, CTP synthetase activity was expressed as both and 1.0 ml of 2 N NaOH. The extract was dried and reconstituted in the pmol of [3H]CTP formed per 10" cells and the ratio of intracellular 100 Â»iIof the mobile phase for Chromatographie analysis. [3H]UTP to [3H]CTP. The standard curve (peak area ratio of DP:RA 433 versus DP CM P Synthetase Assay. An in situ assay measuring the conversion concentration) was linear from 0.5 to 20 ÃŸM.Recovery of RA433 of XMP to GMP, GDP, and GTP was used to estimate GMP synthetase averaged 81%. Absolute recovery of DP averaged 68 Â±6%(n = 31 over activity in intact cells. Approximately 1 x IO6cells were incubated for the entire concentration range) and did not vary with amount of drug. 60 min with 1 pM [G-3H]hypoxanthine. The cells were extracted as At 5 nM DP, within-day variation was 2.3% (n = 3), and day-to-day described for the CTP synthetase assay, and the tritiated XMP, GMP, variation was 2.9% (n = 8 over 8 wk). The addition of known amounts GDP, and GTP were separated using HPLC with a Spherisorb SAX of DP to patient samples yielded values averaging 98% of the expected 10- x 250-mm ion exchange column. A portion of the sample was amounts for both high (>10 ^M) and low (<5 pM) levels of DP. mixed with a solution containing authentic nucleotide markers. Sepa Since DP is extensively bound to AGP (12-14), an assay was devel ration was achieved using a gradient derived from a mixture of 0.003 oped to measure free DP. Plasma ultrafiltrates were prepared using the M and 0.3 M Na2HPO4 buffers (pH 5) run at a flow rate of 2 ml/min Amicon Centrifree micropartition system according to the manufactur er's instructions: 0.5 ml of plasma was centrifuged at 2000 x g for 30 and monitored at 254 nm. Initially the mobile phase was the low concentration buffer. Over the first 8 min the percentage of the 0.3 M min to produce about 200 /ulof ultrafiltrate. Fifty >i\of the ultrafÃltrate buffer was increased linearly from 0 to 15% and then maintained at were assayed by HPLC without further preparation. DP content of that level for the next 7 min. Between 15 and 30 min the high patient ultrafÃltratewas quantified by comparison with standards pre pared from drug-free ultrafiltrate spiked with appropriate amounts of concentration buffer was increased to 100% and then continued at that DP. The standard curve, obtained by adding DP to ultrafiltrates of level for the remaining 20 min of the run. Under these conditions the normal plasma, was linear from 10 to 200 nM. These low concentrations following retention times (in min) were typically obtained: AMP, 7; of DP could be detected using the HPLC system previously described UMP, 7.7; IMP, 8.3; GMP, 10.1; XMP, 16.7; UDP, 22.5; CDP, 23.8; with the addition of a much more sensitive fluorometric detector ADP, 24.7; GDP, 28.1; UTP, 32.9; CTP, 35.9; ATP, 37.8; and GTP, (Shimadzu Model RF-530) with excitation optimized at 286 nM and 46.5. Thus, the conversion of [3H]XMP to [3H]guanine nucleotides emission at 497 nM. DP did not bind to the filter units, as shown by could be determined without interference from radiolabeled adenine complete recovery of DP when spiked ultrafÃltrate standards were nucleotides. GMP synthetase activity was expressed as both the pmol/ IO6cells formed and the ratio of [3H]XMP to [3H]guanine nucleotides subjected to a second filtration. The coefficient of variation in the in order to correct for variations in the amount of [3H]hypoxanthine ultrafiltration step, as shown by assay of separate ultrafiltrates prepared from the same plasma sample, was 2.2% for a 5 ^M DP plasma sample uptake. (n = 4) and 4.5% for a 10 /JM plasma sample (n = 3). Replicate assays Ribonucleoside Triphosphate Pool Determinations. Approximately of the same 100 nM ultrafÃltratesolution showed a within-day variability 4 x IO6cells were extracted with 2 ml of cold 0.5 N HC1O4 for 15 min. of 5.8% (n = 3) and a day-to-day variability of 7.0% (n = 6 over 8 The supernatant was clarified by centrifugation, neutralized, lyophi- days). Specificity of the method was demonstrated by standard addition lized, and stored at â€”20Â°C.The samples were reconstituted with 100 of known amounts of DP to previously assayed patient ultrafiltrates M!of water, and the ribonucleoside triphosphates were quantified by (yielding values which averaged 105% of the expected values). HPLC as previously described (5). Acivicin Assay. Plasma acivicin levels were measured by the Upjohn ofi-Acid Glycoprotein. Plasma levels of AGP were assayed by radial 5592 Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1988 American Association for Cancer Research. BIOCHEMICAL EFFECTS OF ACIVICIN AND DIPYRIDAMOLE GIVEN i.v. immunodiffusion using NOR-Partigen (Behring Diagnostics, La Jolla, CA) plates. Plasma samples (5 ^1) were placed in each well and read 48 Aâ€¢ o-o Patient PatientB\\\N\=Â»>\â€¢-â€¢ h later. Linear standard curves were obtained in the range of 54 to 205 0C mg/dl, and the control samples, 80.1 mg/dl, yielded levels within 4% of the expected values. 0ua!S'S.^15<5~tÃ¬\j1009080706050403020IOn__-â€”Medium^^^T//

N\ v. \ NA\

RESULTS \\^r--^ x^' Dipyridamole Levels and Inhibition of Nucleoside Uptake. At the maximally tolerated dose of 23.1 mg/kg/72 h, plasma levels Vxx>i of total DP averaged 11.9 ^M (19). Although these levels are much higher than those necessary to inhibit nucleoside uptake 1Plasma i i in vitro, their adequacy in an in vivo setting is unclear because 50 500 5,000 50,000 much of the DP administered to humans is bound to AGP. For Dipyridamole (nM) that reason we investigated the relationship between DP con Fig. I. Effect of dipyridamole on the uptake of cytidine in peripheral blood mononuclear cells. Approximately 1 X IO6 cells obtained prior to the start of centrations, AGP levels, and inhibition of nucleoside uptake by therapy were resuspended in either DMEM or the patient's own plasma. The DP. The amount of DP required to inhibit the uptake of 1 MM effects of the indicated concentrations of DP on the uptake of 1 u\\ [3H)cytidine cytidine or 1 Â¿iMuridineby 50% was estimated using peripheral (10 jiCi/nmoI) in a 1-h incubation were then determined. The concentrations of blood mononuclear cells resuspended in either the patient's AGP were 94 and 300 mg/dl for Patients A and B, respectively. plasma or in DMEM. This was done to determine the inherent sensitivity of the cells to DP and to ascertain the effects of AGP on the action of DP. AGP levels, DP,, DPf, and inhibition of nucleoside uptake by DP in a given patient's cells could then 15.0 Human Lymphocytes be compared. In order to establish base-line values and the degree of vari â€” 12.5 ability associated with these measurements, the effects of DP on the uptake of several nucleic acid precursors in peripheral g 100 blood mononuclear cells obtained from normal volunteers were O determined (Table 1). The cells were resuspended in DMEM to eliminate binding to AGP. DP inhibited the uptake of l ÃŸM 7.5 cytidine, uridine, and deoxycytidine by 50% at concentrations i of 13.8, 24.7, and 47.5 ^M, respectively, and little variation was I 5.0 - seen from individual to individual. In contrast, the uptake of Q two nucleobases, adenine and hypoxanthine, was not blocked 2.5 - by dipyridamole. The effect of plasma on DP inhibition of cytidine uptake in lymphocytes from two patients is shown in Fig. 1. In these U0 50 100 150 200 250 300 studies the uptake assays were carried out using cells resus a|-Acid-Glycoprotein (mg/dl) pended in DMEM or in the patients' plasma. The presence of Fig. 2. Relationship between Â»,-acidglycoprotein and inhibition of cytidine plasma shifted the dose-response curve for inhibition of cytidine uptake by dipyridamole. The amount of DP required to inhibit the uptake of 1 uptake by DP to the right approximately 100-fold. These results JIMcytidine by 50% in lymphocytes obtained prior to therapy and resuspended in the patient's own plasma is plotted as a function of the AGP concentration in are consistent with the marked binding of DP to AGP. The that patient's plasma. sensitivity of the patients' cells to DP, while similar when the uptake assay was run in DMEM, was different when the cells were resuspended in plasma. These findings suggested that were compared to the concentrations of DP required to inhibit differences in a plasma constituent, probably AGP, accounted the uptake of 1 nM cytidine by 50% in the mononuclear cells for the decreased inhibition seen in Patient B. In fact, AGP obtained from the same patients (Fig. 2). The assays were done levels in Patient A were 94 mg/dl, whereas much higher values, using cells resuspended in that patient's plasma. The data 300 mg/dl, were evident in Patient B. Plasma levels of AGP shown in Figs. 2 and 3 demonstrate a wide variation in AGP were determined in patients receiving doses of DP greater than levels, from about 60 mg/dl to 300 mg/dl. As plasma levels of or equal to 17.4 mg/kg/72 h. Plasma concentrations of AGP AGP increased, there was a coordinate increase in the concen tration of DP required to inhibit cytidine uptake. In 5 of 6 Table 1 Inhibition of nucleoside uptake by dipyridamole peripheral blood patients treated with the maximum tolerated dose of DP, 23.1 mononuclear cells mg/kg/72 h, levels of DP sufficient to inhibit cytidine uptake Peripheral blood mononuclear cells obtained from normal volunteers were by greater than 50% were achieved. This was estimated from resuspended in DMEM to eliminate binding to AGP. DP inhibited the uptake of the DP, and the dose-response curves for DP inhibition of 1 nM cytidine, uridine, and deoxycytidine by 50% at concentrations of 13.8, 24.7, and 47.5 /i\i. respectively. Uptake of adenine and hypoxanthine was not blocked cytidine uptake for each of these patients. by DP. The variation in the amount of DP required to inhibit cytidine PrecursorCytidine (nM)13.8 uptake in the different patients was probably related to changes in the percentage of free DP caused by alterations in plasma Uridine 24.7 2.5 7 AGP levels. This was evaluated by determining the DPr in the Deoxycytidine 47.5 6.3n14 655 Hypoxanthine >5000 plasma of these patients. As shown in Fig. 3A, the percentage AdenineIDÂ»' >5000SE1.1 of free DP decreased from about 1% at an AGP level of 60 mg/ ' IDÂ«,,50% inhibitor} dose. dl to approximately 0.2% at 300 mg/dl of AGP. Interestingly, 5593 Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1988 American Association for Cancer Research. BIOCHEMICAL EFFECTS OF ACIV1CIN AND DIPYR1DAMOLE GIVEN i.v.

E- 40- 'SIOO)e 0^10o

d-O 0.81o o IO Â«p|2O oRi w.i t J * nB-o000ODo i i O 24 48 72 Hours of Therapy 0,-Acid Glycoprotein (mg/dl) Fig. 4. CTP synthetase activity following acivicin therapy. Enzyme activity was measured by quantifying the formation of [3H]CTP from [3H]UTP in periph Fig. 3. Relationship between Â«i-acidglycoprotein and the percentage of free eral blood mononuclear cells exposed to I Â»iM[3H]uridinefor 1 h. The cells were DP, as well as the total-body clearance of DP, DP,, and DPr, was determined for patients receiving 5.8 mg/kg/24 h or greater of DP. The percentage of unbound obtained prior to and at the indicated times after therapy with acivicin (10 mg/ DP was calculated and plotted as a function of the plasma AGP concentration kg/24 h) and dipyridamole (13 mg/kg/24 h). The data are from a single patient (A ). The total-body clearance of DP was calculated (CLTB = infusion rate/steady- given 4 courses of therapy and are presented as the mean Â±SE (n = 4) for the state concentration of DP) and plotted as a function of the plasma AGP concen 0-, 24-, and 48-h time points. The 72-h value was from a single course of treatment. tration (B). a 'S 0.2 0.4 -g despite wide variations in the percentage of free DP in these individuals, DPf was relatively consistent and averaged 27.8 Â± 015 0.3 y Â« 3.2 HM(mean Â±SD; n = 4) at Level 11 (23.1 mg/kg/72 h) and *' 30.4 Â±6.0 (n = 5) at Level 12 (30 mg/kg/72 h). These data ^Q. Ol 02lÃ¬ suggested that, in patients with high levels of AGP, the clear a ance of DP would be decreased due to a lower percentage of x -05 O.I ? I free DP. A plot of DP clearance versus AGP concentration (Fig. 3Ã„)strongly supports this suggestion. Although levels of 0.0 AGP varied widely from patient to patient, AGP concentrations 24 48 O 24 48 changed little from course to course in the same patient. Fur Hours of Therapy Fig. 5. GMP synthetase activity following acivicin treatment. Enzyme activity thermore, the plasma levels of DP were essentially the same at was measured by quantifying the formation of [3HJguanine nucleotides from 24, 48, and 72 h after initiation of DP therapy. [3H)xanthine monophosphate in peripheral blood mononuclear cells exposed to The importance of determining the DPf rather than DP, was 1 //M [3H]hypoxanthine. The cells were obtained prior to and at 24 h and 48 h after therapy with acivicin (20 mg/kg/24 h). These data are from a single patient. emphasized in a comparison of patients treated with 23.1, 30, and 39 mg of DP/kg/72 h Even though the amount of DP elevation in this ratio and a greater than 75% inhibition of administered increased, the average steady-state DP, actually [3H]CTP formation occurred 7 times. decreased from 11.9 to 5.5 to 4.3 nM. In contrast, the plasma GMP Synthetase Activity. GMP synthetase, which catalyzes DPr increased from 27.8 to 30.4 to 37.8 nM as the dose of DP the glutamine-dependent formation of GMP from XMP, is also was escalated. This unusual circumstance was accounted for by an enzymatic target of acivicin (10, 11, 20). An in situ assay the fact that the patients treated at 23.1 mg/kg/72 h happened was used to measure GMP synthetase activity in peripheral to have higher average AGP levels (217 mg/dl) and a lower blood mononuclear cells. The data presented in Fig. 5 show the percentage of free DP, 0.3%, than the group administered 30 time-dependent accumulation of [3H]XMP and depletion of mg/kg/72 h which had AGP levels averaging 150 mg/dl and tritiated guanine nucleotides in the cells taken from a patient 0.6% free DP. The single patient who received 39 mg/kg/72 h on study. To account for possible differences in the degree of [3H]hypoxanthine uptake, enzyme activity was also measured DP (5 courses) had an average AGP level of 73 mg/dl and 0.9% as the ratio of [3H]XMP to labeled guanine nucleotides. In only free DP. Thus DPf but not DP, increased with the dosage of DP administered. 3 of 10 courses was the ratio elevated greater than 3-fold. CTP Synthetase Activity. CTP synthetase, which catalyzes Ribonucleoside Triphosphate Pools. RTP pools in peripheral the synthesis of CTP from UTP, is a potentially important blood mononuclear cells were measured prior to and at various target for acivicin (10, 11, 20). The activity of this enzyme in times after the initiation of therapy. The data in Fig. 6 were peripheral blood mononuclear cells, obtained prior to and at obtained from a single patient treated 4 times. With the excep tion of the pretreatment values for GTP, variability was reason various times after the initiation of therapy, was measured using ably low. In this patient there was a suggestion of a modest an i/i situ assay. The reproducibility of the method is illustrated in Fig. 4, which depicts the time-dependent decrease in CTP decrease in the GTP levels. In 16 Ã©valuablecourses a selective decrease in either CTP or GTP was seen only 6 times. In only synthetase activity seen in a patient following treatment with one patient was the CTP pool markedly (greater than 90%) acivicin and DP. These data were averaged from four separate decreased. If all the data on RTP pools are averaged, no courses of therapy given over a 5-mo period. In this individual, therapy-related changes were evident (Fig. 7). [3H]CTP formation was reduced about 90% at 48 h. In order to account for the possibility that reduced uptake of [3H]uridine DISCUSSION may lead to apparent inhibition of CTP synthetase activity, enzyme activity was also measured as the ratio of [3H]UTP to Nucleoside salvage may ameliorate the cytotoxic effects of [3H]CTP. In the 13 Ã©valuablecourses, both a 3-fold or greater antimetabolites and diminish the clinical efficacy of such agents 5594

DP despite wide variations in AGP levels and DP,. Lymphocytes-_-1UTPÃ•Tjâ€¢rCTPm\GTPÃ•-LIJ'â€¢'Ã¬ATP_ ig The effects of plasma binding on drug disposition have re n.20l~~ cently been reviewed (29, 30), and the findings in this study are consistent with theoretical considerations. In the case of drug m.15a; Â«|1-0 binding to albumin, protein concentrations are generally much OÃŒ greater than those of the drug, and variations in protein levels -HA1-2.0^0)o (nn35.10Ribonucle Â£a.0.5 are not of concern. In the case of AGP, with a concentration typically in the range of 10 Â¿Â¿M,variationsin protein levels can D01Human 0 < be of importance (29, 30), particularly since DP, was also in this range. An inverse relationship between the percentage of n n free drug and AGP concentration has been reported for several Hours of Therapy agents (29, 31, 32). In the present study, using a continuous infusion of DP, changes in the percentage of free drug have Fig. 6. Ribonucleoside triphosphate pools following acivicin therapy. RTF pools were measured in peripheral blood mononuclear cells prior to and at 24 also been shown to correlate with changes in the rate of clear and 48 h after theapy with acivicin (20 mg/kg/24 In and dipyridamole (13 mg/ ance. However, the alterations in DP, and the clearance rate kg/24 h). The data are from a single patient given 4 course of treatment and are presented as the mean Â±SE (n = 4). were compensatory, and relatively constant values for DPf were seen. Acivicin inhibits a variety of glutamine-dependent reactions and, therefore, can interfere with pyrimidine and purine biosyn t/}QJI rhGTPâ€”~i1TÂ¿â€¢/P thesis at several steps. An earlier study by McGovren et al. (33) 251.^ demonstrated inhibition of carbamoyl phosphate synthetase II in leukocytes and malignant ascites from patients receiving 20â€¢r- 2.0 1la E C"Ã¶ c.10l of CTP synthetase and, to a lesser extent, GMP synthetase T TTT|1Ã•tATPk 1PlÃŽW.T10 activity was evident in lymphocytes obtained from patients administered 20 mg of acivicin/m2/day as a 3-day continuous 05cnnnUTP\ 0.5 infusion. Concurrent drops in RTP pool sizes were infrequently

00 seen. Insufficient enzyme inhibition, nucleoside salvage, and low triphosphate utilization could have contributed to the lack Hours of Therapy of RTP pool depletion. In particular, the fact that the peripheral Fig. 1. Ribonucleoside triphosphate pools following acivicin therapy. RTF mononuclear cells used in these studies are largely nonprolifer- pools were measured in peripheral pool mononuclear cells prior to and at 24 and 48 h after therapy with acivicin (20 mg/kg/24 h) and dipyridamole (from 5.8 to ating may significantly reduce the demands on ribonucleoside 13 mg/kg/24 h). These data are the average changes in all Ã©valuablecoursesand triphosphates and, consequently, the effects of inhibitors of are presented as the mean Â±SE (n = 18 to 21) nucleotide biosynthesis. In parallel studies using a human colon cancer cell line (HCT 116) with a high growth fraction, we (5-8, 21-24). This trial was conducted as a part of an effort to found that inhibition of CTP synthetase by acivicin was accom test this possibility. Although this study did not allow us to panied by a significant drop in CTP.4 Thus, lymphocytes may conclude whether nucleoside salvage modulates the effects of be well suited for biochemical analysis of certain targets, such acivicin in a clinical setting, several important observations as nucleoside transporters and biosynthetic enzymes, but of less regarding the clinical pharmacology of DP and the biochemical value in evaluating the functional consequences of drug action, effects of acivicin were made. such as RTP pool size depletion. Based on the biochemical We determined that the maximally tolerated dose of i.v. evidence for inhibition of nucleoside transport obtained in this administered DP, 23.1 mg/kg/72 h, yielded plasma concentra trial, we initiated a second clinical study utilizing methotrexate tions sufficient to inhibit cytidine uptake in human peripheral and DP in combination (34). The results of that investigation, mononuclear cells. This information was critical since the ex demonstrating clear modulation of methotrexate toxicity by tensive binding of DP to AGP (12-14) invalidates direct ex DP, support the contention that pharmacologically active con trapolation of data obtained in tissue culture systems to the centrations of DP were obtained in the present study. clinic. Interpretation of DP plasma levels was potentially fur The lack of effect of DP on the clinical toxicity of acivicin ther complicated by the wide range of AGP plasma levels in may have been related to insufficient inhibition of de novo cancer patients (25-27). Our Undings illustrated that the per purine and pyrimidine biosynthesis. Neurotoxicity is dose lim centage of free DP and the clearance of DP were inversely iting in the continuous infusion studies (35, 36), whereas mye- related to the plasma levels of AGP and, as a consequence, DPf losuppression was limiting when acivicin was given daily for 5 was relatively constant from patient to patient. In patients with days (37, 38). Perhaps the combination of acivicin and DP lower levels of AGP and, therefore, a higher percentage of free should be evaluated using a different schedule of administration. DP, it is likely that glucuronidation of DP (28) has occurred at Another possible explanation for the lack of a drug interaction a faster rate and increased DP clearance. The enhanced rate of is that, clinically, nucleoside salvage is not an important media clearance then led to lower DP,. Conversely, in patients with tor of acivicin's action. The relatively low plasma concentra high levels of AGP, a greater percentage of DP was protein tions of cytidine and guanosine may be insufficient to restore bound, and clearance was diminished. Therefore, DPf, which is nucleotide pools in vivo. Finally, recent data suggest that several the product of DP, and the percentage of unbound DP, was relatively constant in patients administered the same dose of 4 Unpublished results. 5595

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1988 American Association for Cancer Research. BIOCHEMICAL EFFECTS OF ACIVICIN AND DIPYRIDAMOLE GIVEN i.V. processes, including sodium-dependent, inhibitor-insensitive 17. McGovren, J. P., Neil, G. L., Sem, P. L. C., and Stewart, J. C. Sex- and age- related mouse toxicity and disposition of the amino acid antitumor agent, mechanisms, are involved in nucleoside transport and that acivicin. J. Pharmacol. Exp. Ther., 216:433-440, 1981. different tissues may have different mechanisms of nucleoside 18. Khyin. J. X. An analytical system for rapid separation of tissue nucleotides transport (39, 40). For these reasons, it is still unclear if at low pressures on conventional aniÃ³nexchangers. Clin. Chem., 21: 1245- 1252, 1975. nucleoside salvage is an important mechanism of resistance for 19. Willson, J. K. V., Fischer, P. H., Tutsch, K., Alberti, D., Simon, K., a variety of anticancer drugs and if inhibitors of nucleoside Hamilton, R. D., Bruggink, J., Kodier, J. M., Earhart, R. H., Ranhosky, A., transport can offer a therapeutic approach to address this and Trump, D. L. Dipyridamole and acivicin: a Phase I clinical trial of a combination based upon inhibition of nucleoside salvage. Cancer Res., 48: problem. 5585-5590, 1988. 20. Weber, G., Lui, M. S., Seboldt, J., and Faderan, M. A. Molecular targets of anti-glutamine therapy with acivicin in cancer cells. In: D. HÃ¤ssingerand H. ACKNOWLEDGMENTS Sies (eds.), Glutamine Metabolism in Mammalian Tissues, pp. 278-291. Berlin: Springer Verlag, 1984. 21. Chan, T. C. K., Young, B., King, M. E., Taetle, R., and Howell, S. B. The authors thank Ranjit Singh and Tim VanMouwerik for their Modulation of the activity of A'-(phosphonacetyl)-L-aspartate by dipyrida technical expertise and Bonnie Rayho for her help in the preparation mole. Cancer Treat. Rep., 69: 425-430, 1985. of this manuscript. 22. Chan, T. C. K., and Howell, S. B. Mechanism of synergy between N- phosphonacetyl-/-aspartate and dipyridamole in a human ovarian carcinoma cell line. Cancer Res., 45: 3598-3604, 1985. 23. Nelson, J. A., and Drake, S. Potentiation of methotrexate toxicity by dipyr REFERENCES idamole. Cancer Res., Â¥4:2493-2496, 1984. 24. Van Mouwerik, T. J., Pangallo, C. A., Willson, J. K. V., and Fischer, P. H. 1. Scholtissek, C. Studies on the uptake of nucleic acid precursors into cells in Augmentation of methotrexate cytotoxicity in human colon cancer cells tissue culture. Biochim. Biophys. Acta, I5S: 435-447, 1968. achieved through inhibition of thymidine salvage by dipyridamole. Biochem. 2. Paterson, A. R. B., Kolassa, N., and Cass, C. E. Transport of nucleoside Pharmacol., 36: 809-814, 1987. drugs in animal cells. Pharmacol. Ther., 12: 515-536, 1981. 25. Snyder, S., and Ashwell, G. Quantitation of specific serum glycoproteins in 3. Paterson, A. R. P.. Lau, E. Y., Dahlig, E., and Cass, C. E. A common basis malignancy. Clin. Chem. Acta, 34:449-455, 1971. for inhibition of nucleoside transport by dipyridamole and nitrobenzylthioi- 26. Weiss, J. F., Morantz, R. A., Bradley, W. P., and Chretien, P. B. Serum nosine? Mol. Pharmacol., 18: 40-44, 1980. acute-phase proteins and immunoglobulins in patients with gliomas. Cancer 4. Plagemann, P. G. W., and Wohlhueter, R. M. Permeation of nucleosides, Res., 39: 542-544, 1979. nucleic acid bases, and nucleotides in animal cells. Curr. Top. Membr. 27. Chio, L-F., and Oon, C. J. Changes in serum Â«,-antitrypsin, Â«i-acidglyco Transp., 14: 255-330, 1980. protein, and ft-glycoprotein l in patients with malignant hepatocellular 5. Fischer, P. H., Pamukcu, R., Bittner, G., and Willson, J. K. V. Enhancement carcinoma. Am. Cancer Soc., 43: 594-604, 1979. of the sensitivity of human colon cancer cells to growth inhibition by acivicin 28. Zak, S. B., Tallan, H. H., Quinn, G. P., Fratta, I., and Greengard, P. The achieved through inhibition of nucleic acid precursor salvage by dipyridamole. determination and physiological distribution of dipyridamole and its glucu- Cancer Res., 44: 3355-3359, 1984. ronides in biological material. J. Pharmacol. Exp. Ther., 141:392-398,1963. 6. Zhen, Y-S., Lui, M. S., and Weber, G. Effects of acivicin and dipyridamole 29. Piafsky, K. M. Disease-induced changes in the plasma binding of basic drugs. on hepatoma 3924A cells. Cancer Res., 43: 1616-1619, 1983. Clin. Pharmacokinet., 5: 246-262, 1980. 7. Weber, G. Biochemical strategy of cancer cells and the design of chemother 30. Wilkinson, G. R. Plasma and tissue binding considerations in drug disposi apy: G. H. A. Clowes Memorial Lecture. Cancer Res., 43: 3466-3492, 1983. tion. Drug. Metab. Rev., 14:427-465, 1983. 8. Weber, G., Lui, M. S., Natsumeda, Y., and Faderan, M. A. Salvage capacity 31. Piafsky, K. M., Borja, D., Odar-Cedarlof, I., Johansson, C., and SjÃ¶vist,F. of hepatoma 3924A and action of dipyridamole. Adv. Enzyme Regul., 21: Increased plasma protein binding of propranolol and chlorpromazine me 53-69, 1983. diated by disease-induced elevations of plasma M, adii glycoprotien. N. Engl. 9. Jayaram, H. N., Cooney, D. A., Ryan, J. A., Neil, G., Dion, R. L., and Bono, J. Med., 299: 1435-1439, 1978. V. H. L-[Â«S,5S]-a-Amino-3-chIoro-4,5-dihydro-5-isoxazoleaceticacid (NSC- 32. Kroboth, P. D., Smith, R. B., Sorkin, M. I., Silver, M. R., Rault, R., Garry, 163501): a new amino acid antibiotic with the properties of an antagonist of M., and Juhl, R. P. Triazolam protein binding and correlation with aracid L-glutamine. Cancer Chemother. Rep., 59:481-491, 1975. glyprotein concentration. Clin. Pharmacol. Ther., 36: 379-383, 1984. 10. Neil, G. L., Berger, A. E., McPartland, R. P., Grindey, G. B., and Bloch, A. 33. McGovren, J. P., Pratt, E. A., Belt, R. J., Taylor, S. A., Benjamin, R. S., Biochemical and pharmacological effects of the fermentation-derived anti- Ardalan, B., and Ohnuma, T. Pharmacokinetic and biochemical studies of tumor agent. (aS,5S)-o-amino-3-chloro-4,5-dihydro-5-isoxazoleacetic acid acivicin in Phase I clinical trials. Cancer Res., 45: 4460-4463, 1985. (AT-125). Cancer Res., 39: 852-856, 1979. 34. Willson, J. K. V., Fischer, P. H., Tutsch, K., Remick, S. C., Grem, J. L., 11. DentÃ³n, J. E., Lui, M. S., Aoki, T., Sebolt, J., and Weber, G. Rapid in vivo Bruggink, J., Alveiti, D., Simon, K., and Trump, D. L. Dipyridamole activation by acivicin of CTP synthetase, carbomoylphosphate synthetase II, modulates methotrexate cytotoxicity: results of a clinical trial based upon and amidophosphoribosyltransferase in hepatoma. Life Sci., 30:1073-1080, inhibition of nucleoside salvage. Proc. Am. Soc. Clin. Oncol., 6: 137, 1986. 1982. 35. Weiss, G. R., McGovren, J. P., Schade, O., and Kufe, D. W. Phase I and 12. Kopitar, V. Z., and Weisenberger, H. Specific binding of dipyridamole to pharmacological study of acivicin by 24-hour continuous infusion. Cancer human serum protein: its isolation, identification, and characterization as Res., Â«.-3892-3895,1982. Â«i-acidicglycoprotein. Arzneim.-Forsch., 6: 859-862, 1971. 36. Earhart, R. H., Koeller, J. M., Davis, T. E., Borden, E. C., McGovren, J. P., 13. Niewiarowski, S., Lukasiewicz, H., Nath, N., and Tai-Sha, A. Inhibition of Davis, H. L., and Tormey, D. L. Phase I trial and pharmacokinetics of human platelet aggregation by dipyridamole and two related compounds and acivicin administered by 72-hour infusion. Cancer Treat Rep., 67: 683-692, its modification by acid glycoproteins and human plasma. J. Lab. Clin. Med., 1983. Â«6:64-76,1975. 37. Sridhar, K. S., Ohnuma, T., Chahinian, A. P., and Holland, J. F. Phase I 14. Subbarao. K.. Rucinski, B., Rausch, M. A., Schmid, K., and Niewiarowski, study of acivicin in patients with advanced cancer. Cancer Treat Rep., 67: S. Binding of dipyridamole to human platelets and to Â«i-acidglycoprotein 701-703, 1983. and its significance for the inhibition of adenosine uptake. J. Clin. Invest., 38. Murphy, W. K., Burgess, M. A., Valdivierso, M., and Bodey, G. P. Phase I 60:936-943, 1977. evaluation of acivicin (AT-125). Proc. Am. Soc. Clin. Oncol., /: 23, 1982. 15. Wolfram, K. M.. and Bjornsson. T. D. High-performance liquid Chromato 39. Jakobs, E. S., and Paterson, A. R. P. Sodium-dependent, concentrative graphie analysis of dipyridamole in plasma and whole blood. J. Chromatog., nucleoside transport in cultured intestinal epithelial cells. Biochem. Biophys. 183: 57-64. 1980. Res. Commun., 140: 1028-1035, 1986. 16. Schmid, J., Beschke, K.. Roth, W., Bozler, G., and Koss, F. W. Rapid, 40. Darnowski, J. W., Holdridge, C., and Handschumacher, R. E. Concentrative sensitive determination of dipyridamole in human plasma by high-perform uridine transport by murine splenocytes: kinetics, substrate specificity, and ance liquid chromatography. J. Chromatog. 163: 239-243, 1979. sodium dependency. Cancer Res., 47: 2614-2619, 1987.

5596

Downloaded from cancerres.aacrjournals.org on October 1, 2021. © 1988 American Association for Cancer Research. Biochemical Assessment of the Effects of Acivicin and Dipyridamole Given as a Continuous 72-Hour Intravenous Infusion

Paul H. Fischer, James K. V. Willson, Concepcion Risueno, et al.

Cancer Res 1988;48:5591-5596.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/48/19/5591

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/48/19/5591. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.