Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Distribution of Adriamycin in mice bearing mammary adenocarcinoma 16/C

Distribution of Adriamycin in mice bearing mammary adenocarcinoma 16/C Soufhern Research Institute, Birmingham, A L , USA (Received 26 June 1990; revision accepted 12 November 1990) Abstract. The response of solid mammary adenocarcinoma 16/C to treatment with Adriamycin is highly variable and ranges from growth under treatment to complete regression. Tumour and host factors were evaluated to determine the influence of each on the response. We determined that the concentration of Adriamycin in plasma and tumour was a function of tumour size and treatment history in mice bearing mammary adenocarcinoma 16/C. The plasma concentrations following a single dose of Adriamycin (10 mg/kg) increased in proportion to tumour mass without a concurrent increase in tumour concentration. When mice bearing large tumours (>1.0 g) were treated with a multidose protocol, the plasma concentrations were higher and the tumour concentrations lower following the initial dose than following subsequent doses; in tumour-free mice, prior treatment with Adriamycin did not affect the plasma level achieved after a second dose. The magnitude of the decrease in plasma and increase in tumour concentrations was a function of the initial tumour size and the treatment schedule. The increase in tumour levels represented the sum of residual Adriamycin and drug bound as a result of the dose immediately prior to analysis. At the time of the initial treatment, the Adriamycin was distributed within each tumour in proportion to vascular perfusion. The percent of the tumour mass that was wellperfused appeared to increase with repeated treatments. The results indicate that the plasma concentration of Adriamycin did not pecessarily reflect the tumour exposure in the mammary adenocarcinoma 16/C model. In hosts bearing mammary adenocarcinoma 16/C-or, possibly, other tumours that produce similar effects on the host-a low initial dose of Adriamycin might modify the distribution, possibly reduce the toxicity and allow escalation of subsequent doses with increased exposure of the tumour. Although Adriamycin (ADR)has shown antitumour activity in a broad spectrum of experimental and human solid tumours, leukaemias and lymphomas, the response of individual tumours is unpredictable. This variability of response is also evident in individual tumours transplanted from a single tumour into inbred mice and staged according to size at the time of treatment (Schabel er al. 1982).The variable responses of solid mammary adenocarcinoma 16/C (mam ad 16/C)to melphalan-L-PAM (Simpson-Herren, Noker & Wagoner 1988) and ADR (SimpsonCorrespondence: Dr Linda Simpson-Herren, Southern Research Institute, PO Box 55305, Birmingham. AL 35255-5305, USA. L. Simpson-Herren and P. E . Noker Herren & Noker 1989) have been attributed to variations in intratumour drug concentration within individual tumours. We demonstrated that the amount of L-PAM or ADR that reached a given tumour region was directly related to the vascular perfusion of the region, with little penetration of the drugs to cells located in the poorly perfused regions. Regions of mam ad 16/C that were poorly vascularized (and exposed to low drug levels) ranged from <lo% to of the mass of individual tumours. The presence of clonogenic cells in these regions has been demonstrated by bioassay (Simpson-Herren & Noker 1989, Simpson-Herren et al. 1988). Limited diffusion of ADR in solid tumours is consistent with reports that ADR penetrates only a few cell diameters into cellular masses of intra-abdominal ovarian tumours (Ozols et al. 1979) or into spheroids (Durand 1989). A concentration gradient of ADR fluorescence from the outside to the inside of mammary tumour cell (EMT6) spheroids was reported by Sutherland et al. (1979). In addition to t h e possible limitations of effective treatment of solid tumour masses imposed by the penetration characteristics of ADR, the pharmacokinetics of distribution and elimination of ADR may limit the effectiveness of therapy, but these factors are poorly understood. It has been suggested that the wide variations in plasma levels reported in both patients and experimental animals may be related to impairment of liver function (Benjamin, Wiernik & Bachur 1974, Preiss et al. 1987), but they are not believed to be related to renal insufficiency (Benjamin 1974). Prior chemotherapy is reported to decrease the plasma levels of ADR achieved in patients (Gessner et al. 1980, Piazza et al. 1989), and possibly to accelerate disappearance of ADR from plasma (Tipping el al. 1982). Clinical studies suggest that plasma kinetics may be non-linear (Erttman et al. 1988, Boston & Phillips 1983, Robert et al. 1983) and possibly time-dependent (Robert et al. 1983, Gil et al. 1983). Further, abnormal haematologic parameters may alter the drug distribution (Broggini et al. 1980). The goal of the studies reported here was to investigate the intratumour distribution of ADR, to study the effects of the tumour on pharmacokinetics of ADR and to seek means of increasing exposure of clonogenic cells to cytotoxic levels of the drug. As the studies progressed, it became apparent that tumour concentration and distribution of ADR were not direct functions of the plasma concentration. Investigation of the plasma pharmacokinetics as well as tumour distribution in the same system was undertaken to more clearly define the relationship. MATERIALS AND METHODS Biological systems The transplantable mam ad 16/C was derived from a spontaneous mammary tumour that arose in a C3H mouse in 1974 (Corbet et al. 1978). The tumour has been routinely transplanted by subcutaneous implantation of tumour fragments in C3H female mice. For experiments, tumour fragments were implanted subcutaneously in female B6C3F1 (C57BV6 x C3H) mice. Tumour growth was followed by caliper measurements of two perpendicular diameters and tumour weight was calculated as Y2 (length X width2). Mice were randomized to control and treated groups either immediately after tumour implantation or after staging of tumours to meet the specifications of the study. Mice were maintained under pathogen-free conditions in accordance with institutional guidelines established for animal welfare. Additional details of the tumour history and biological techniques have been reported previously by Corbett et al. (1978) and Simpson-Herren, Noker & Wagoner (1987). A D R in mice with mam ad 161C Drug treatment ADR (NSC 123127) was obtained from Sigma (St Louis, MO, USA) and was administered i.p. in sterile water at 10 mg/kg body weight, unless otherwise specified, and on the schedule indicated for individual studies. Treatment solutions were prepared immediately prior to use and were given i.p. unless otherwise specified. Sample collection and preparation At selected times after administration of ADR, mice were anaesthetized with ether and a single terminal blood sample was collected from the axillary region into a Microtainer plasma separator. Plasma was obtained after centrifugation of cooled samples and subsequently extracted with 4 volumes of methanol. Portions of each extract were assayed for ADR. Tumours were removed, blotted on paper and frozen. Grossly necrotic tissue was discarded. Each tumour was homogenized in 2 volumes of water. Portions (0.5 ml) of each homogenate were mixed with 0.5 ml of water and then precipitated with 0.25 ml of 33% silver nitrate. After standing on ice for 15 min, the mixtures were diluted with 2.5 ml of acetonitri1e:water (52:48) containing 30 m heptanesulphonic acid and 15 m phosphoric acid. After centrifuM M gation, portions of each extract were assayed for ADR. Sample analysis HPLC analyses were accomplished with a Waters Associates (Milford, MA, USA) high pressure liquid chromatograph equipped with a Model 6000A high pressure delivery pump, a Model 710 automatic sample injector (WISP), a Model 730 data module and a Spectroflow 980 fluorescence detector (Kratos Analytical, Ramsey, NJ, USA). For the assay of ADR, samples were injected onto a pBondapak C18 column (Waters Associates) and eluted with acentonitri1e:water (35:65) containing 20 m heptanesulphonic acid and 10 m phosphoric acid at a M M flow rate of 1 mumin. Excitation and emission wavelengths of 233 nm and 470 nm, respectively, were used. Separation of A D R from the metabolite, adriamycinol, was demonstrated using a standard sample of adriamycinol generously supplied by Adria Laboratories). Under the conditions of our studies there was no evidence of measurable concentrations of adriamycinol in either tumours or plasma. Pharmacokinetics of ADR Pharmacokinetic half-lives of A D R in plasma were estimated from HPLC data'with a modified form of NONLIN (Metzler, Elfring & McEwen 1974) and CSTRIP (Sedman & Wagner 1976). The data were fitted to one-, two- and three-compartment models. A model was accepted as best fit if an additional term, or compartment, failed to reduced significantly ( P < 0.05) the weighted sums of squared errors as estimated by the F-test with appropriate degrees of freedom. Statistical weights were determined from the measured concentrations and were the same for each model. Clinical chemistry Blood urea nitrogen (BUN), albumin (ALB), aspartate aminotransferase (AST), and alanine aminotransferase (ALT) were quantitated using a Roche Cobras Fara analyser. The results were compared with concurrent standards and historical data for mice from the same strain. RESULTS Response of mam ad 16/C to ADR A single dose of ADR, administered i.p. at 10 mglkg to mice bearing 0.5-1.0 g mam ad 16/C L. Simpson-Herren and P . E . Noker Time post-implant (days) Figure 1. Growth curves for mammary adenocarcinoma 16/C treated a with vehicle only (control) or b with a single i.p. dose of ADR (Adriamycin; 10 mg/kg) on day 9. produced tumour responses ranging from minimal to complete regression; however, the majority of tumours exhibited a period of stasis or incomplete regression that reached a nadir 4-6 days after treatment (Figure 1). In other experiments where tumour-bearing mice were treated in a similar manner, effects of therapy on tumour mass were not evident after a single dose but became evident when a second dose was administered 4 days later. N o cures were observed following a single dose of A D R and the rate of regrowth approximated that of the untreated control tumours. Plasma concentrations as a function of dose The results of early studies indicated that plasma concentrations of ADR frequently were highly variable in tumour-bearing animals. To determine whether this was a function of the tumour, we first measured the plasma concentration of ADR in tumour-free mice at 10 min ADR in mice with mam ad 16/C E .3.0 [ r 10 15 ADR administered (mglkg) Figure 2. Plasma (0)and tumour ( X ) levels of Adriamycin (ADR) 10 min after administration of ADR at doses of 5 , 10 and 20 mg/kg in mice bearing 1.0 to 2.0 g tumours (10 or more per time point) and (0) control, tumour-free mice after in administration of the same ADR doses (4 per time point). The data are presented as mean values k SD. post-injection of doses ranging from 5 mg/kg to 20 mg/kg (Figure 2). Four mice per point were used to obtain to data on tumour-free animals. The values shown for the tumour-bearing mice are means ~ S of four studies, totaling 10 or more mice bearing tumours from 1.0 to 2.0 g for D each point. Plasma levels achieved at 10 min following administration of A D R increased 3- to >lO-fold with each 2-fold increase in dose administered in individual studies. The larger increases occurred as the dose was increased from 10 to 20 mg/kg. Plasma levels of ADR in tumour-free and tumour-bearing mice The variability of plasma A D R was evaluated in 13 studies involving five tumour-free mice per experiment. The concentration of A D R in plasma at 10 min after i.p. injection of 10 mg/kg A D R was0.62 t 0.07pglml. The inter-experiment variation was 11%. These results indicated that the plasma levels of A D R following a 10 mg/kg dose to tumour-free mice were consistent within experiments and between experiments. Thirty-nine mice bearing mam ad 16/C were treated with A D R as above at 11 to 18 days after tumour implantation. The harvested tumour weights (when sacrificed 10 min after treatment) ranged from less than 0.5 t o more than 6.0 g. In these mice, the plasma concen- L. Simpson-Herren and P . E. Noker trations ranged from less than 0.5 to more than 5.0 pg/ml. Linear regression analysis of the plasma levels as a function of tumour size (correlation coefficient r = 0.79) indicated that a direct relationship existed between the tumour mass and the plasma levels of A D R under the conditions studied (Figure 3). These studies included tumours as large as 6.5 g o r approximately 25% of the body weight. The larger tumours were included to more clearly define the relationships between plasma drug levels and tumour size, since clinical symptoms and lethality occur at higher relative tumour burdens (tumour weightlhost weight) in murine models than in man. The relationship between the elevated plasma levels in tumour-bearing mice and toxicity has not been investigated. I- ._ I 4 - 2.0 - r I a m 1.5- m a 1.0 - I - _ _ _ L _ L 1 - ~ - 1 ~ - I Tumourfree Harvested tumour weight ( g ) Figure 3. The concentration of Adriamycin (ADR) in plasma 10 min after injection of ADR at 10 mg/kg as a function of the harvested weight of mammary adenocarcinoma 16/Cturnours. The correlation coefficient ( r ) = 0.79. The mean f SD for ADR concentration in plasma of tumour-free mice from 13 experiments is shown on the left. Plasma and tumour uptake and elimination of ADR In view of the variability of A D R pharmacokinetics in the reported literature, we felt that it was necessary to investigate the plasma and tumour uptake and elimination of A D R in the animal model used for these studies. Tumour-free B6C3F1 female mice were administered A D R (10 mg/kg) and terminal blood samples were obtained at periods ranging from 5 to 60 min. The concentration of A D R in plasma was determined and the data are shown in Figure 4a. Computer analysis indicated that the data were well described ( 3 = 0.996) a by two-phase A D R i mice with mam ad 16IC n 10’ lo-’ 0 ._ a n [L loo- lo-’ -+% W o a ‘+7 Time after ADR administration (min) Figure 4. The concentration of Adriamycin (ADR) in a plasma of tumour-free mice, b plasma of mice bearing mammary adenocarcinoma 16/C tumours and c turnours of tumour-bearing mice at intervals of 5 min to 96 h following a single dose of ADR. The plasma elimination curves were fitted by computer analysis. L. Simpson-Herren and P . E. Noker elimination curve, with an initial phase of 3.2 min and a second or terminal phase of 40.7 min. Under the same experimental conditions, the concentration of A D R in plasma and tumours of mice bearing mam ad 16/C tumours (1-2 g) was measured (Figure 4b & c). In plasma, there was no evidence of an uptake phase for A D R , and the data were best described (? = 0.828) by a single phase with an elimination half-life of 25 min. As was anticipated from earlier studies, the variability in plasma levels of A D R in these tumour-bearing mice was considerably greater than the variability observed in tumour-free mice. A D R concentrations in tumours from these same mice reached a maximum within 15 min, and remained essentially constant through 60 min. (Additional tumours assayed 96 h after dosing showed that A D R levels were maintained at the same level for the 4-day period). There was a poor correlation between plasma and tumour levels of A D R in these animals. Effects of prior treatment with ADR on plasma and tumour levels of ADR following subsequent treatments Further investigation indicated that plasma and tumour concentrations of A D R were functions, not only of tumour mass, but also of the treatment history of the host. These relationships were deduced from studies at different stages of tumour growth (Figure 5 & Table 1) and from studies of mice bearing 18-day tumours with different treatment histories (Figure 6 &Table 2). 0 , c F 0 a , C n n ula lb w 2a 2b 3a -3b Group Figure 5. Concentrations of Adriamycin (ADR) in plasma of tumour-free mice (0; la) and in plasma (0)and tumours ( 8 ) mice bearing lo-, 14- and 18-day implants of of mammary adenocarcinoma ( l b 3 b ) The bars represent the mean k SD of data from 5 individual mice. The groups, treatments and measured haematocrit values are shown in Table 1. A D R in mice with mam ad 161C Table 1. Groups, times of Adriamycin (ADR) treatment(s) and sacrifice (days after implant) and haematocrit levels in mice bearing lo-, 14- and 18-day implants of mammary adenocarcinoma 16/C. Values represent the mean + S O of the data from 5 individual mice. The haematocrit was determined at the time of sacrifice; the value for tumoui-free mice is the historical mean for these studies. Plasma and tumour ADR values for these groups of mice are shown in Figure 5 Group la lb 2a 2b 3a 3b Tumour weight (g) Tumour-free 0.69 f 0.24 3.05 1.21 0.97 f 0.38 1.73 4.83 0.47 k 0.09 Treatment day 10 10 14 10, 14 18 10, 14, 18 Sacrifice day 10 10 14 14 18 18 Haematocrit (YO) 33 f 5 26 f 7 16 4 20 f 7 19 f 3 18 f 7 In an initial experiment, mice from a single tumour implant group were randomly assigned to groups of five on day 10 post-implant, and treated as shown in Table 1. A D R concentrations in plasma of mice bearing 0.69 g tumours were not significantly different from plasma levels in tumour-free mice, and the tumour A D R levels approximated the plasma levels. In groups of mice administered the initial dose of A D R at 14 days after tumour implant (Group 2a), the ADR levels were higher in plasma and lower in tumours than the corresponding levels in a companion group of mice treated on day 10 as well as day 14 (Group 2b). Following a single dose of A D R on day 18 to previously untreated mice (Group 3a), high plasma and low tumour ADR concentrations were observed, but in mice treated on days 10,14 and 18 (Group 3b), the plasma levels of A D R were less than half those observed in Group 3a. In contrast, the tumour levels in the previously treated group (days 10 and 14 in addition to day 18) were more than three times those in the group treated on day 18 only. From the results of these experiments, we conclude that plasma levels of A D R following a single dose of 10 mglkg increased with tumour size without a concurrent increase in tumour levels. Further, when A D R (10 mglkg) administered every 4 days for two o r three doses inhibited tumour growth or induced tumour regression, the treatment prevented the increase in plasma A D R levels and increased the tumour A D R concentration. The contribution of residual A D R from previous treatments was not evaluated in this study. The contribution of A D R from prior treatments to plasma and tumour concentrations was evaluated when mice from a single source were either implanted with tumour fragments o r left tumour-free and, after 10 days, assigned to groups of five (Table 2). Two groups of tumourbearing mice were treated with A D R on day 10 post-implant, four groups (including the Table 2. Groups, times of Adriamycin (ADR) treatment(s) and sacrifice and haematocrit levels in mice bearing 18-day implants of mammary adenocarcinoma 16/C. Values represent the mean ~ S of the data from 5 individual D mice. The haematocrit was determined at the time of sacrifice; the value for tumour-free mice is the historical mean for these studies. Plasma and tumour ADR values for these groups of mice are shown in Figure 6 Group la lb Tumour weight ( 8 ) Tumour-free 4.4 k 0.8 2.3 k 0.4 2.0 ? 0.5 0.93 0.26 0.72 f 0.27 Treatment day 18 18 14 (only) 14, 18 10, 14 10, 14, 18 Sacrifice day 18 18 18 18 18 18 Haematocrit (YO) 32 k 5 17 f 4 14 k 1 17 k 4 17 f 5 19 f 6 2a 2b 3a 3b L . Simpson-Herren and P . E. Noker cn 0 E 'E 0, c a 0 l [ r 1.c nl L la lb 2a 2b Group 3a 3b Figure 6 . Concentrations of Adriamycin (ADR) in plasma of tumour-free mice ( 0 ; la) and in plasma (0) and tumours (a) mice bearing 18-day implants of mammary of adenocarcinoma 16/C tumours with different treatment histories. The bars indicate the mean values ? SD deviation of data from 5 individual mice. The groups, treatments and measured haematocrit values are shown in Table 2. previously treated groups) were treated on day 14 and four groups were treated on day 18, 10 min prior to sacrifice. The groups of mice that were not treated on the day of sacrifice were included to provide an estimate of the residual ADR.Thus, the concentrations of A D R found in the plasma and tumours of mice in Group 2a (treated on day 14 only) may be considered to be a measure of the residual A D R contribution to plasma and tumour concentration in mice of Group 2b (treated on days 14 and 18). Similarly, the A D R in the plasma and tumours of mice in Group 3a (treated days 10 and 14) may be considered as the contribution of residual A D R to plasma and tumour levels for Group 3b (treated days 10, 14 and 18). A D R in mice with mam ad 16/C In mice bearing previously untreated 18-day-old tumours, 4.4 k 0.8 g in weight (Group lb), plasma levels were higher and more variable than those observed in the tumour-free mice or any of the previously treated groups. The results indicate that these high plasma levels of ADR were not indicative of comparable exposure of the tumours to the drug. The tumour levels of ADR in these mice (Group l b ) were lower than those of any previously treated group, including the two groups that were evaluated 4 days after the last treatment. Plasma levels following the second treatment with ADR administered on day 18, 4 days after the initial dose on day 14 (Group 2b), were lower than those observed in the group treated on day 18only (Group lb). In the group treated on day 14 and then sacrificed on day 18 with no additional treatment, the residual plasma ADR was insignificant (Group 2a). Tumour concentrations approximated the sum of the residual (Group 2a) and new ADR (Group lb). Following the third dose of ADR (Group 3b), the plasma levels were similar to the plasma levels observed following two doses (Group 2b). The tumour ADR (Group 3b) exceeded the level that might be expected from the sum of the residual ADR (Group 3a) and the new ADR from a single treatment (Group lb). The possibility that the tumour biology and vascular supply were changed in tumours that regressed following the initial treatment(s) was indicated in studies where lissamine green (LG; Simpson-Herren et al. 1988) was administered 4 days after an initial dose of ADR. These tumours were well-perfused with minimal regions that appeared poorly stained. The concentration of LG in the total tumour approximated the dye concentration in the well-perfused regions of untreated tumours. These observations were consistent with a reduction in the percent of poorly vascularized tumour following effective treatment. To determine whether prior treatment also reduced the plasma concentrations achieved in tumour-free mice, we administered ADR (10 mg/kg) as a single dose or two doses 4 days apart to groups of five mice each and collected plasma 10 min after the first or second injection. The plasma ADR concentrations were 0.80 k 0.09 pg/ml in the pretreated group and 0.79 k 0.05 pg/ml in the group given a single dose. It should be noted that the presence of a mam ad 16/C tumour for 10 days or longer produced anaemia (reduced haematocrit) that was not modified by treatment with ADR that induced tumour stasis or regression (Tables 1 & 2). That is, the haematocrit did not return to control values in those mice that responded to treatment. On day 18 post-implant of tumours, the haematocrit ranged from 13.5 k 1.3% to 18.8 k 6.0% in five groups of mice, regardless of the treatment history. Further, there was no identifiable relationship between the haematocrit and plasma levels of ADR in individual mice (Figures 5 & 6; Tables 1 & 2), indicating that reduced binding of ADR to red blood cells was not responsible for the high plasma levels observed in mice bearing large tumours. Relationship between plasma ADR levels and clinical chemistry parameters Blood chemistry parameters were measured to assess the possible relationship between the elevated plasma levels of ADR in tumour-bearing mice and renal or liver function. Since the determination of plasma ADR levels and a clinical chemistry profile required a larger volume of blood than could be obtained from an individual mouse, blood was pooled from 19 mice bearing mam ad 16/C >2.0 g and from 10 tumour-free mice. Pooled blood samples, obtained 10 min after administration of 10 mg/kg ADR, were divided into three aliquots, and independent analyses were performed on t h e aliquots. The results are shown in Table 3. Plasma ADR levels were elevated 4.4-fold in the tumour-bearing mice, but the clinical chemistry profiles of these mice were within normal ranges, indicating normal renal (BUN and ALB) or liver (ALB, AST and ALT) function. L. Simpson-Herren and P . E. Noker Table 3. Relationship of clinical chemistry profile to elevated plasma levels of Adriamycin (ADR): results are expressed as mean values t SD Plasma ADR (Irdml) Turnour-free Turnour-bearing Clinical dhemistry BUN (mg/dl) ALB (g/dl) AST (U/I) ALT (U/I) 0.39 t 0.12 1.72 t 0.32 25 f 4 23 6 2.9 & 0.1 2.5 f 0.2 271 t 45 298 f 70 55 f 8 56 t 8 ALB = serum albumin; BUN = blood urea nitrogen; AST = aspartate amhotransferase; ALT = alanine aminotransferase. DISCUSSION A major problem in chemotherapy of cancer in man is the variability of response of tumours of the same histological type and size to treatment with an effective agent. Similarly, murine mam ad 16/C from the same implant group, staged according to size and age, respond t o therapy with effective agents in a variable manner (Simpson-Herren et al. 1988, SimpsonHerren & Noker 1989), ranging from complete regression to growth under treatment. We used this model to investigate factors that may be related to the response of tumours t o chemotherapy. The dose range in which we found that plasma levels increased more rapidly than would be predicted if the plasma level were a linear function of the dose administered (Figure 2 ) included doses in the therapeutically effective range for A D R (10 mg/kgor 30 mg/m’). This observation would be consistent with the dose-dependent pharmacokinetics of ADR suggested by the reports of several investigators (Powis, Ames & Kovach 1983), which indicated that shorter plasma half-lives occurred in patients given lower doses of ADR. These observations contrast with the linear pharmacokinetic models frequently used for this drug. Plasma A D R concentrations were elevated and highly variable following administration of ADR to mice bearing large mam ad 16/C tumours (>1.0 g) when compared with plasma concentrations in tumour-free mice. We previously reported that plasma levels of L-PAM were elevated in the same system, but the magnitude of the elevation was not as great (SimpsonHerren et al. 1987). This might suggest that a single mechanism was responsible for both observations. The most likely explanation would be that the volume of distribution was reduced in mice bearing large mam ad 16/C tumours. However, to date, we have been unable to demonstrate this using Evan’s blue techniques (Schultz et af. 1953). It is possible that the elimination and/or metabolism of A D R (and L-PAM) was affected by the presence of the tumour. In the present study, this is supported by the observation that the presence of large tumours modified the plasma elimination curve of ADR, i.e. a two-phase elimination was observed in tumour-free mice, but only a single phase elimination was seen in tumour bearing mice following i.p. administration of the drug. The major routes of elimination for the two drugs differ: chemical hydrolysis is a major determinant of L-PAM disposition (Farmer & Newel1 1983) and half-life in vivo (Alberts e al. 1979a,b); while metabolism by the liver is the f major route of elimination of A D R (Powis et al. 1983). The reduction in body temperature that occurs in many animals with tumour growth could slow the hydrolysis of L-PAM, but the effects of low temperature on A D R metabolism are unclear. Our clinical chemistry data indicate that neither renal nor hepatic insufficiency was present in this model at the time that the elevation of plasma A D R levels became evident. Prior treatment with A D R modified the plasma drug levels achieved and the relationship between plasma and tumour concentrations. In mice bearing large tumours, the initial dose of ADR resulted in high plasma levels (compared with the levels achieved in tumour-free mice A D R in mice with m u m ad 16/C following the same dose) and tumour concentrations with were 25% (or less) of the plasma levels. Following a second o r third dose at 4-day intervals, the plasma levels achieved were reduced; the reduction may be related to the anti-tumour effects of the drug-either directly as a function of the tumour regression o r by reduction of the toxic effects of the tumour on the host. In the same mice, tumour levels of ADR increased. The tumour concentrations of ADR following the second o r third treatment reflected both residual ADR from earlier treatments and the ADK bound as a result of the treatment 10 min prior to sacrifice. The amount of ADR bound in the tumour appeared t o increase with each dose. Preliminary data suggest that this may have resulted from a change in the biology and vascularity of the tumour, with a reduction in the percent of the tumour that was poorly vascularized. There appeared t o be an optimal interval between doses to take advantage of the tumour changes, i.e. 4 days in this system (Simpson-Herren & Noker, unpublished data). The reduction in plasma levels following successive doses of ADR supports the observation of Gessner et al. (1981) that patients who had received prior treatment with ADR achieved lower plasma levels during subsequent courses of treatment with the same drug. In the mam ad 16/C model, the reduction in plasma levels coincided with tumour response (regression) and may reflect changes in the tumour-host relationship and/or in plasma elimination of ADR. The observations reported by Broggini er al. (1980) suggest that plasma levels of ADR in anaemic animals may be elevated as a result of a reduction in the amount of ADR bound to red blood cells, with a concurrent increase in the amount of free ADR to raise the apparent plasma levels. The data relating haematocrit t o plasma levels (Figures 5 & 6) suggest that the plasma levels of ADR are not a function of haematocrit in the mam ad 16/C. Persistence of ADR in tumour tissue has been reported (Wassermann & Rasmussen 1987), but persistence of ADR at constant concentrations in regressing tumours is, as yet, unexplained. We would expect that the cells binding the highest levels at ADR would be killed and removed from the tumour mass with a resultant decrease in the concentration of ADR. We have found no evidence of this phenomenon to date. Preliminary data indicate that, following treatment of M5076 sarcomas implanted in B6C3F1 mice (the same host used for the mam ad 16/C), the plasma levels achieved were not a function of tumour mass (Simpson-Herren & Noker, unpublished data). Plasma levels of ADR in mice bearing M5076 tumours as large as 3.5 g did not differ significantly from those of tumour-free mice. The observations reported here suggest that the variability of ADR pharmacokinetic parameters may be associated with the debilitating effects of the mam ad 16/C on the host. Tumour-induced cachexia is frequently observed in man, but is infrequently observed in anima! models, suggesting that the mam ad 16/C system may provide insight into the highly variable pharmacokinetics of ADR in man. These studies indicate that the plasma level achieved following the initial dose of ADR was a function of the tumour size in this system. The reduction in the plasma ADR level achieved following a second or third treatment at equivalent dose levels coincided with tumour regression and appeared to be an indicator of tumour response in this experimental model. The data suggest that a low initial dose of ADR escalated at the second and third treatment might increase tumour exposure without increasing the total dose and the host toxicity. This approach to therapy with ADR is under investigation. ACKNOWLEDGEMENTS This investigation was supported by Public Health Service Grant R 0 1 CA37132 awarded by L. Simpson-Herren and P . E. Noker the National Cancer Institute, National Institutes of Health, Department of Health and Human Services. The authors would like to thank Melissa Edminston, Laurie Cutts and Margaret A. Thomson for excellent technical assistance. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cell Proliferation Wiley

Distribution of Adriamycin in mice bearing mammary adenocarcinoma 16/C

Simpson-Herren, L.; Noker, P. E.
Cell Proliferation , Volume 24 (3) – May 1, 1991
Download PDF
15 pages

Loading next page...
 
/lp/wiley/distribution-of-adriamycin-in-mice-bearing-mammary-adenocarcinoma-16-c-lJrI0X0LGO

References (22)

Publisher
Wiley
Copyright
1991 Blackwell Science Limited
ISSN
0960-7722
eISSN
1365-2184
DOI
10.1111/j.1365-2184.1991.tb01154.x
Publisher site
See Article on Publisher Site

Abstract

Soufhern Research Institute, Birmingham, A L , USA (Received 26 June 1990; revision accepted 12 November 1990) Abstract. The response of solid mammary adenocarcinoma 16/C to treatment with Adriamycin is highly variable and ranges from growth under treatment to complete regression. Tumour and host factors were evaluated to determine the influence of each on the response. We determined that the concentration of Adriamycin in plasma and tumour was a function of tumour size and treatment history in mice bearing mammary adenocarcinoma 16/C. The plasma concentrations following a single dose of Adriamycin (10 mg/kg) increased in proportion to tumour mass without a concurrent increase in tumour concentration. When mice bearing large tumours (>1.0 g) were treated with a multidose protocol, the plasma concentrations were higher and the tumour concentrations lower following the initial dose than following subsequent doses; in tumour-free mice, prior treatment with Adriamycin did not affect the plasma level achieved after a second dose. The magnitude of the decrease in plasma and increase in tumour concentrations was a function of the initial tumour size and the treatment schedule. The increase in tumour levels represented the sum of residual Adriamycin and drug bound as a result of the dose immediately prior to analysis. At the time of the initial treatment, the Adriamycin was distributed within each tumour in proportion to vascular perfusion. The percent of the tumour mass that was wellperfused appeared to increase with repeated treatments. The results indicate that the plasma concentration of Adriamycin did not pecessarily reflect the tumour exposure in the mammary adenocarcinoma 16/C model. In hosts bearing mammary adenocarcinoma 16/C-or, possibly, other tumours that produce similar effects on the host-a low initial dose of Adriamycin might modify the distribution, possibly reduce the toxicity and allow escalation of subsequent doses with increased exposure of the tumour. Although Adriamycin (ADR)has shown antitumour activity in a broad spectrum of experimental and human solid tumours, leukaemias and lymphomas, the response of individual tumours is unpredictable. This variability of response is also evident in individual tumours transplanted from a single tumour into inbred mice and staged according to size at the time of treatment (Schabel er al. 1982).The variable responses of solid mammary adenocarcinoma 16/C (mam ad 16/C)to melphalan-L-PAM (Simpson-Herren, Noker & Wagoner 1988) and ADR (SimpsonCorrespondence: Dr Linda Simpson-Herren, Southern Research Institute, PO Box 55305, Birmingham. AL 35255-5305, USA. L. Simpson-Herren and P. E . Noker Herren & Noker 1989) have been attributed to variations in intratumour drug concentration within individual tumours. We demonstrated that the amount of L-PAM or ADR that reached a given tumour region was directly related to the vascular perfusion of the region, with little penetration of the drugs to cells located in the poorly perfused regions. Regions of mam ad 16/C that were poorly vascularized (and exposed to low drug levels) ranged from <lo% to of the mass of individual tumours. The presence of clonogenic cells in these regions has been demonstrated by bioassay (Simpson-Herren & Noker 1989, Simpson-Herren et al. 1988). Limited diffusion of ADR in solid tumours is consistent with reports that ADR penetrates only a few cell diameters into cellular masses of intra-abdominal ovarian tumours (Ozols et al. 1979) or into spheroids (Durand 1989). A concentration gradient of ADR fluorescence from the outside to the inside of mammary tumour cell (EMT6) spheroids was reported by Sutherland et al. (1979). In addition to t h e possible limitations of effective treatment of solid tumour masses imposed by the penetration characteristics of ADR, the pharmacokinetics of distribution and elimination of ADR may limit the effectiveness of therapy, but these factors are poorly understood. It has been suggested that the wide variations in plasma levels reported in both patients and experimental animals may be related to impairment of liver function (Benjamin, Wiernik & Bachur 1974, Preiss et al. 1987), but they are not believed to be related to renal insufficiency (Benjamin 1974). Prior chemotherapy is reported to decrease the plasma levels of ADR achieved in patients (Gessner et al. 1980, Piazza et al. 1989), and possibly to accelerate disappearance of ADR from plasma (Tipping el al. 1982). Clinical studies suggest that plasma kinetics may be non-linear (Erttman et al. 1988, Boston & Phillips 1983, Robert et al. 1983) and possibly time-dependent (Robert et al. 1983, Gil et al. 1983). Further, abnormal haematologic parameters may alter the drug distribution (Broggini et al. 1980). The goal of the studies reported here was to investigate the intratumour distribution of ADR, to study the effects of the tumour on pharmacokinetics of ADR and to seek means of increasing exposure of clonogenic cells to cytotoxic levels of the drug. As the studies progressed, it became apparent that tumour concentration and distribution of ADR were not direct functions of the plasma concentration. Investigation of the plasma pharmacokinetics as well as tumour distribution in the same system was undertaken to more clearly define the relationship. MATERIALS AND METHODS Biological systems The transplantable mam ad 16/C was derived from a spontaneous mammary tumour that arose in a C3H mouse in 1974 (Corbet et al. 1978). The tumour has been routinely transplanted by subcutaneous implantation of tumour fragments in C3H female mice. For experiments, tumour fragments were implanted subcutaneously in female B6C3F1 (C57BV6 x C3H) mice. Tumour growth was followed by caliper measurements of two perpendicular diameters and tumour weight was calculated as Y2 (length X width2). Mice were randomized to control and treated groups either immediately after tumour implantation or after staging of tumours to meet the specifications of the study. Mice were maintained under pathogen-free conditions in accordance with institutional guidelines established for animal welfare. Additional details of the tumour history and biological techniques have been reported previously by Corbett et al. (1978) and Simpson-Herren, Noker & Wagoner (1987). A D R in mice with mam ad 161C Drug treatment ADR (NSC 123127) was obtained from Sigma (St Louis, MO, USA) and was administered i.p. in sterile water at 10 mg/kg body weight, unless otherwise specified, and on the schedule indicated for individual studies. Treatment solutions were prepared immediately prior to use and were given i.p. unless otherwise specified. Sample collection and preparation At selected times after administration of ADR, mice were anaesthetized with ether and a single terminal blood sample was collected from the axillary region into a Microtainer plasma separator. Plasma was obtained after centrifugation of cooled samples and subsequently extracted with 4 volumes of methanol. Portions of each extract were assayed for ADR. Tumours were removed, blotted on paper and frozen. Grossly necrotic tissue was discarded. Each tumour was homogenized in 2 volumes of water. Portions (0.5 ml) of each homogenate were mixed with 0.5 ml of water and then precipitated with 0.25 ml of 33% silver nitrate. After standing on ice for 15 min, the mixtures were diluted with 2.5 ml of acetonitri1e:water (52:48) containing 30 m heptanesulphonic acid and 15 m phosphoric acid. After centrifuM M gation, portions of each extract were assayed for ADR. Sample analysis HPLC analyses were accomplished with a Waters Associates (Milford, MA, USA) high pressure liquid chromatograph equipped with a Model 6000A high pressure delivery pump, a Model 710 automatic sample injector (WISP), a Model 730 data module and a Spectroflow 980 fluorescence detector (Kratos Analytical, Ramsey, NJ, USA). For the assay of ADR, samples were injected onto a pBondapak C18 column (Waters Associates) and eluted with acentonitri1e:water (35:65) containing 20 m heptanesulphonic acid and 10 m phosphoric acid at a M M flow rate of 1 mumin. Excitation and emission wavelengths of 233 nm and 470 nm, respectively, were used. Separation of A D R from the metabolite, adriamycinol, was demonstrated using a standard sample of adriamycinol generously supplied by Adria Laboratories). Under the conditions of our studies there was no evidence of measurable concentrations of adriamycinol in either tumours or plasma. Pharmacokinetics of ADR Pharmacokinetic half-lives of A D R in plasma were estimated from HPLC data'with a modified form of NONLIN (Metzler, Elfring & McEwen 1974) and CSTRIP (Sedman & Wagner 1976). The data were fitted to one-, two- and three-compartment models. A model was accepted as best fit if an additional term, or compartment, failed to reduced significantly ( P < 0.05) the weighted sums of squared errors as estimated by the F-test with appropriate degrees of freedom. Statistical weights were determined from the measured concentrations and were the same for each model. Clinical chemistry Blood urea nitrogen (BUN), albumin (ALB), aspartate aminotransferase (AST), and alanine aminotransferase (ALT) were quantitated using a Roche Cobras Fara analyser. The results were compared with concurrent standards and historical data for mice from the same strain. RESULTS Response of mam ad 16/C to ADR A single dose of ADR, administered i.p. at 10 mglkg to mice bearing 0.5-1.0 g mam ad 16/C L. Simpson-Herren and P . E . Noker Time post-implant (days) Figure 1. Growth curves for mammary adenocarcinoma 16/C treated a with vehicle only (control) or b with a single i.p. dose of ADR (Adriamycin; 10 mg/kg) on day 9. produced tumour responses ranging from minimal to complete regression; however, the majority of tumours exhibited a period of stasis or incomplete regression that reached a nadir 4-6 days after treatment (Figure 1). In other experiments where tumour-bearing mice were treated in a similar manner, effects of therapy on tumour mass were not evident after a single dose but became evident when a second dose was administered 4 days later. N o cures were observed following a single dose of A D R and the rate of regrowth approximated that of the untreated control tumours. Plasma concentrations as a function of dose The results of early studies indicated that plasma concentrations of ADR frequently were highly variable in tumour-bearing animals. To determine whether this was a function of the tumour, we first measured the plasma concentration of ADR in tumour-free mice at 10 min ADR in mice with mam ad 16/C E .3.0 [ r 10 15 ADR administered (mglkg) Figure 2. Plasma (0)and tumour ( X ) levels of Adriamycin (ADR) 10 min after administration of ADR at doses of 5 , 10 and 20 mg/kg in mice bearing 1.0 to 2.0 g tumours (10 or more per time point) and (0) control, tumour-free mice after in administration of the same ADR doses (4 per time point). The data are presented as mean values k SD. post-injection of doses ranging from 5 mg/kg to 20 mg/kg (Figure 2). Four mice per point were used to obtain to data on tumour-free animals. The values shown for the tumour-bearing mice are means ~ S of four studies, totaling 10 or more mice bearing tumours from 1.0 to 2.0 g for D each point. Plasma levels achieved at 10 min following administration of A D R increased 3- to >lO-fold with each 2-fold increase in dose administered in individual studies. The larger increases occurred as the dose was increased from 10 to 20 mg/kg. Plasma levels of ADR in tumour-free and tumour-bearing mice The variability of plasma A D R was evaluated in 13 studies involving five tumour-free mice per experiment. The concentration of A D R in plasma at 10 min after i.p. injection of 10 mg/kg A D R was0.62 t 0.07pglml. The inter-experiment variation was 11%. These results indicated that the plasma levels of A D R following a 10 mg/kg dose to tumour-free mice were consistent within experiments and between experiments. Thirty-nine mice bearing mam ad 16/C were treated with A D R as above at 11 to 18 days after tumour implantation. The harvested tumour weights (when sacrificed 10 min after treatment) ranged from less than 0.5 t o more than 6.0 g. In these mice, the plasma concen- L. Simpson-Herren and P . E. Noker trations ranged from less than 0.5 to more than 5.0 pg/ml. Linear regression analysis of the plasma levels as a function of tumour size (correlation coefficient r = 0.79) indicated that a direct relationship existed between the tumour mass and the plasma levels of A D R under the conditions studied (Figure 3). These studies included tumours as large as 6.5 g o r approximately 25% of the body weight. The larger tumours were included to more clearly define the relationships between plasma drug levels and tumour size, since clinical symptoms and lethality occur at higher relative tumour burdens (tumour weightlhost weight) in murine models than in man. The relationship between the elevated plasma levels in tumour-bearing mice and toxicity has not been investigated. I- ._ I 4 - 2.0 - r I a m 1.5- m a 1.0 - I - _ _ _ L _ L 1 - ~ - 1 ~ - I Tumourfree Harvested tumour weight ( g ) Figure 3. The concentration of Adriamycin (ADR) in plasma 10 min after injection of ADR at 10 mg/kg as a function of the harvested weight of mammary adenocarcinoma 16/Cturnours. The correlation coefficient ( r ) = 0.79. The mean f SD for ADR concentration in plasma of tumour-free mice from 13 experiments is shown on the left. Plasma and tumour uptake and elimination of ADR In view of the variability of A D R pharmacokinetics in the reported literature, we felt that it was necessary to investigate the plasma and tumour uptake and elimination of A D R in the animal model used for these studies. Tumour-free B6C3F1 female mice were administered A D R (10 mg/kg) and terminal blood samples were obtained at periods ranging from 5 to 60 min. The concentration of A D R in plasma was determined and the data are shown in Figure 4a. Computer analysis indicated that the data were well described ( 3 = 0.996) a by two-phase A D R i mice with mam ad 16IC n 10’ lo-’ 0 ._ a n [L loo- lo-’ -+% W o a ‘+7 Time after ADR administration (min) Figure 4. The concentration of Adriamycin (ADR) in a plasma of tumour-free mice, b plasma of mice bearing mammary adenocarcinoma 16/C tumours and c turnours of tumour-bearing mice at intervals of 5 min to 96 h following a single dose of ADR. The plasma elimination curves were fitted by computer analysis. L. Simpson-Herren and P . E. Noker elimination curve, with an initial phase of 3.2 min and a second or terminal phase of 40.7 min. Under the same experimental conditions, the concentration of A D R in plasma and tumours of mice bearing mam ad 16/C tumours (1-2 g) was measured (Figure 4b & c). In plasma, there was no evidence of an uptake phase for A D R , and the data were best described (? = 0.828) by a single phase with an elimination half-life of 25 min. As was anticipated from earlier studies, the variability in plasma levels of A D R in these tumour-bearing mice was considerably greater than the variability observed in tumour-free mice. A D R concentrations in tumours from these same mice reached a maximum within 15 min, and remained essentially constant through 60 min. (Additional tumours assayed 96 h after dosing showed that A D R levels were maintained at the same level for the 4-day period). There was a poor correlation between plasma and tumour levels of A D R in these animals. Effects of prior treatment with ADR on plasma and tumour levels of ADR following subsequent treatments Further investigation indicated that plasma and tumour concentrations of A D R were functions, not only of tumour mass, but also of the treatment history of the host. These relationships were deduced from studies at different stages of tumour growth (Figure 5 & Table 1) and from studies of mice bearing 18-day tumours with different treatment histories (Figure 6 &Table 2). 0 , c F 0 a , C n n ula lb w 2a 2b 3a -3b Group Figure 5. Concentrations of Adriamycin (ADR) in plasma of tumour-free mice (0; la) and in plasma (0)and tumours ( 8 ) mice bearing lo-, 14- and 18-day implants of of mammary adenocarcinoma ( l b 3 b ) The bars represent the mean k SD of data from 5 individual mice. The groups, treatments and measured haematocrit values are shown in Table 1. A D R in mice with mam ad 161C Table 1. Groups, times of Adriamycin (ADR) treatment(s) and sacrifice (days after implant) and haematocrit levels in mice bearing lo-, 14- and 18-day implants of mammary adenocarcinoma 16/C. Values represent the mean + S O of the data from 5 individual mice. The haematocrit was determined at the time of sacrifice; the value for tumoui-free mice is the historical mean for these studies. Plasma and tumour ADR values for these groups of mice are shown in Figure 5 Group la lb 2a 2b 3a 3b Tumour weight (g) Tumour-free 0.69 f 0.24 3.05 1.21 0.97 f 0.38 1.73 4.83 0.47 k 0.09 Treatment day 10 10 14 10, 14 18 10, 14, 18 Sacrifice day 10 10 14 14 18 18 Haematocrit (YO) 33 f 5 26 f 7 16 4 20 f 7 19 f 3 18 f 7 In an initial experiment, mice from a single tumour implant group were randomly assigned to groups of five on day 10 post-implant, and treated as shown in Table 1. A D R concentrations in plasma of mice bearing 0.69 g tumours were not significantly different from plasma levels in tumour-free mice, and the tumour A D R levels approximated the plasma levels. In groups of mice administered the initial dose of A D R at 14 days after tumour implant (Group 2a), the ADR levels were higher in plasma and lower in tumours than the corresponding levels in a companion group of mice treated on day 10 as well as day 14 (Group 2b). Following a single dose of A D R on day 18 to previously untreated mice (Group 3a), high plasma and low tumour ADR concentrations were observed, but in mice treated on days 10,14 and 18 (Group 3b), the plasma levels of A D R were less than half those observed in Group 3a. In contrast, the tumour levels in the previously treated group (days 10 and 14 in addition to day 18) were more than three times those in the group treated on day 18 only. From the results of these experiments, we conclude that plasma levels of A D R following a single dose of 10 mglkg increased with tumour size without a concurrent increase in tumour levels. Further, when A D R (10 mglkg) administered every 4 days for two o r three doses inhibited tumour growth or induced tumour regression, the treatment prevented the increase in plasma A D R levels and increased the tumour A D R concentration. The contribution of residual A D R from previous treatments was not evaluated in this study. The contribution of A D R from prior treatments to plasma and tumour concentrations was evaluated when mice from a single source were either implanted with tumour fragments o r left tumour-free and, after 10 days, assigned to groups of five (Table 2). Two groups of tumourbearing mice were treated with A D R on day 10 post-implant, four groups (including the Table 2. Groups, times of Adriamycin (ADR) treatment(s) and sacrifice and haematocrit levels in mice bearing 18-day implants of mammary adenocarcinoma 16/C. Values represent the mean ~ S of the data from 5 individual D mice. The haematocrit was determined at the time of sacrifice; the value for tumour-free mice is the historical mean for these studies. Plasma and tumour ADR values for these groups of mice are shown in Figure 6 Group la lb Tumour weight ( 8 ) Tumour-free 4.4 k 0.8 2.3 k 0.4 2.0 ? 0.5 0.93 0.26 0.72 f 0.27 Treatment day 18 18 14 (only) 14, 18 10, 14 10, 14, 18 Sacrifice day 18 18 18 18 18 18 Haematocrit (YO) 32 k 5 17 f 4 14 k 1 17 k 4 17 f 5 19 f 6 2a 2b 3a 3b L . Simpson-Herren and P . E. Noker cn 0 E 'E 0, c a 0 l [ r 1.c nl L la lb 2a 2b Group 3a 3b Figure 6 . Concentrations of Adriamycin (ADR) in plasma of tumour-free mice ( 0 ; la) and in plasma (0) and tumours (a) mice bearing 18-day implants of mammary of adenocarcinoma 16/C tumours with different treatment histories. The bars indicate the mean values ? SD deviation of data from 5 individual mice. The groups, treatments and measured haematocrit values are shown in Table 2. previously treated groups) were treated on day 14 and four groups were treated on day 18, 10 min prior to sacrifice. The groups of mice that were not treated on the day of sacrifice were included to provide an estimate of the residual ADR.Thus, the concentrations of A D R found in the plasma and tumours of mice in Group 2a (treated on day 14 only) may be considered to be a measure of the residual A D R contribution to plasma and tumour concentration in mice of Group 2b (treated on days 14 and 18). Similarly, the A D R in the plasma and tumours of mice in Group 3a (treated days 10 and 14) may be considered as the contribution of residual A D R to plasma and tumour levels for Group 3b (treated days 10, 14 and 18). A D R in mice with mam ad 16/C In mice bearing previously untreated 18-day-old tumours, 4.4 k 0.8 g in weight (Group lb), plasma levels were higher and more variable than those observed in the tumour-free mice or any of the previously treated groups. The results indicate that these high plasma levels of ADR were not indicative of comparable exposure of the tumours to the drug. The tumour levels of ADR in these mice (Group l b ) were lower than those of any previously treated group, including the two groups that were evaluated 4 days after the last treatment. Plasma levels following the second treatment with ADR administered on day 18, 4 days after the initial dose on day 14 (Group 2b), were lower than those observed in the group treated on day 18only (Group lb). In the group treated on day 14 and then sacrificed on day 18 with no additional treatment, the residual plasma ADR was insignificant (Group 2a). Tumour concentrations approximated the sum of the residual (Group 2a) and new ADR (Group lb). Following the third dose of ADR (Group 3b), the plasma levels were similar to the plasma levels observed following two doses (Group 2b). The tumour ADR (Group 3b) exceeded the level that might be expected from the sum of the residual ADR (Group 3a) and the new ADR from a single treatment (Group lb). The possibility that the tumour biology and vascular supply were changed in tumours that regressed following the initial treatment(s) was indicated in studies where lissamine green (LG; Simpson-Herren et al. 1988) was administered 4 days after an initial dose of ADR. These tumours were well-perfused with minimal regions that appeared poorly stained. The concentration of LG in the total tumour approximated the dye concentration in the well-perfused regions of untreated tumours. These observations were consistent with a reduction in the percent of poorly vascularized tumour following effective treatment. To determine whether prior treatment also reduced the plasma concentrations achieved in tumour-free mice, we administered ADR (10 mg/kg) as a single dose or two doses 4 days apart to groups of five mice each and collected plasma 10 min after the first or second injection. The plasma ADR concentrations were 0.80 k 0.09 pg/ml in the pretreated group and 0.79 k 0.05 pg/ml in the group given a single dose. It should be noted that the presence of a mam ad 16/C tumour for 10 days or longer produced anaemia (reduced haematocrit) that was not modified by treatment with ADR that induced tumour stasis or regression (Tables 1 & 2). That is, the haematocrit did not return to control values in those mice that responded to treatment. On day 18 post-implant of tumours, the haematocrit ranged from 13.5 k 1.3% to 18.8 k 6.0% in five groups of mice, regardless of the treatment history. Further, there was no identifiable relationship between the haematocrit and plasma levels of ADR in individual mice (Figures 5 & 6; Tables 1 & 2), indicating that reduced binding of ADR to red blood cells was not responsible for the high plasma levels observed in mice bearing large tumours. Relationship between plasma ADR levels and clinical chemistry parameters Blood chemistry parameters were measured to assess the possible relationship between the elevated plasma levels of ADR in tumour-bearing mice and renal or liver function. Since the determination of plasma ADR levels and a clinical chemistry profile required a larger volume of blood than could be obtained from an individual mouse, blood was pooled from 19 mice bearing mam ad 16/C >2.0 g and from 10 tumour-free mice. Pooled blood samples, obtained 10 min after administration of 10 mg/kg ADR, were divided into three aliquots, and independent analyses were performed on t h e aliquots. The results are shown in Table 3. Plasma ADR levels were elevated 4.4-fold in the tumour-bearing mice, but the clinical chemistry profiles of these mice were within normal ranges, indicating normal renal (BUN and ALB) or liver (ALB, AST and ALT) function. L. Simpson-Herren and P . E. Noker Table 3. Relationship of clinical chemistry profile to elevated plasma levels of Adriamycin (ADR): results are expressed as mean values t SD Plasma ADR (Irdml) Turnour-free Turnour-bearing Clinical dhemistry BUN (mg/dl) ALB (g/dl) AST (U/I) ALT (U/I) 0.39 t 0.12 1.72 t 0.32 25 f 4 23 6 2.9 & 0.1 2.5 f 0.2 271 t 45 298 f 70 55 f 8 56 t 8 ALB = serum albumin; BUN = blood urea nitrogen; AST = aspartate amhotransferase; ALT = alanine aminotransferase. DISCUSSION A major problem in chemotherapy of cancer in man is the variability of response of tumours of the same histological type and size to treatment with an effective agent. Similarly, murine mam ad 16/C from the same implant group, staged according to size and age, respond t o therapy with effective agents in a variable manner (Simpson-Herren et al. 1988, SimpsonHerren & Noker 1989), ranging from complete regression to growth under treatment. We used this model to investigate factors that may be related to the response of tumours t o chemotherapy. The dose range in which we found that plasma levels increased more rapidly than would be predicted if the plasma level were a linear function of the dose administered (Figure 2 ) included doses in the therapeutically effective range for A D R (10 mg/kgor 30 mg/m’). This observation would be consistent with the dose-dependent pharmacokinetics of ADR suggested by the reports of several investigators (Powis, Ames & Kovach 1983), which indicated that shorter plasma half-lives occurred in patients given lower doses of ADR. These observations contrast with the linear pharmacokinetic models frequently used for this drug. Plasma A D R concentrations were elevated and highly variable following administration of ADR to mice bearing large mam ad 16/C tumours (>1.0 g) when compared with plasma concentrations in tumour-free mice. We previously reported that plasma levels of L-PAM were elevated in the same system, but the magnitude of the elevation was not as great (SimpsonHerren et al. 1987). This might suggest that a single mechanism was responsible for both observations. The most likely explanation would be that the volume of distribution was reduced in mice bearing large mam ad 16/C tumours. However, to date, we have been unable to demonstrate this using Evan’s blue techniques (Schultz et af. 1953). It is possible that the elimination and/or metabolism of A D R (and L-PAM) was affected by the presence of the tumour. In the present study, this is supported by the observation that the presence of large tumours modified the plasma elimination curve of ADR, i.e. a two-phase elimination was observed in tumour-free mice, but only a single phase elimination was seen in tumour bearing mice following i.p. administration of the drug. The major routes of elimination for the two drugs differ: chemical hydrolysis is a major determinant of L-PAM disposition (Farmer & Newel1 1983) and half-life in vivo (Alberts e al. 1979a,b); while metabolism by the liver is the f major route of elimination of A D R (Powis et al. 1983). The reduction in body temperature that occurs in many animals with tumour growth could slow the hydrolysis of L-PAM, but the effects of low temperature on A D R metabolism are unclear. Our clinical chemistry data indicate that neither renal nor hepatic insufficiency was present in this model at the time that the elevation of plasma A D R levels became evident. Prior treatment with A D R modified the plasma drug levels achieved and the relationship between plasma and tumour concentrations. In mice bearing large tumours, the initial dose of ADR resulted in high plasma levels (compared with the levels achieved in tumour-free mice A D R in mice with m u m ad 16/C following the same dose) and tumour concentrations with were 25% (or less) of the plasma levels. Following a second o r third dose at 4-day intervals, the plasma levels achieved were reduced; the reduction may be related to the anti-tumour effects of the drug-either directly as a function of the tumour regression o r by reduction of the toxic effects of the tumour on the host. In the same mice, tumour levels of ADR increased. The tumour concentrations of ADR following the second o r third treatment reflected both residual ADR from earlier treatments and the ADK bound as a result of the treatment 10 min prior to sacrifice. The amount of ADR bound in the tumour appeared t o increase with each dose. Preliminary data suggest that this may have resulted from a change in the biology and vascularity of the tumour, with a reduction in the percent of the tumour that was poorly vascularized. There appeared t o be an optimal interval between doses to take advantage of the tumour changes, i.e. 4 days in this system (Simpson-Herren & Noker, unpublished data). The reduction in plasma levels following successive doses of ADR supports the observation of Gessner et al. (1981) that patients who had received prior treatment with ADR achieved lower plasma levels during subsequent courses of treatment with the same drug. In the mam ad 16/C model, the reduction in plasma levels coincided with tumour response (regression) and may reflect changes in the tumour-host relationship and/or in plasma elimination of ADR. The observations reported by Broggini er al. (1980) suggest that plasma levels of ADR in anaemic animals may be elevated as a result of a reduction in the amount of ADR bound to red blood cells, with a concurrent increase in the amount of free ADR to raise the apparent plasma levels. The data relating haematocrit t o plasma levels (Figures 5 & 6) suggest that the plasma levels of ADR are not a function of haematocrit in the mam ad 16/C. Persistence of ADR in tumour tissue has been reported (Wassermann & Rasmussen 1987), but persistence of ADR at constant concentrations in regressing tumours is, as yet, unexplained. We would expect that the cells binding the highest levels at ADR would be killed and removed from the tumour mass with a resultant decrease in the concentration of ADR. We have found no evidence of this phenomenon to date. Preliminary data indicate that, following treatment of M5076 sarcomas implanted in B6C3F1 mice (the same host used for the mam ad 16/C), the plasma levels achieved were not a function of tumour mass (Simpson-Herren & Noker, unpublished data). Plasma levels of ADR in mice bearing M5076 tumours as large as 3.5 g did not differ significantly from those of tumour-free mice. The observations reported here suggest that the variability of ADR pharmacokinetic parameters may be associated with the debilitating effects of the mam ad 16/C on the host. Tumour-induced cachexia is frequently observed in man, but is infrequently observed in anima! models, suggesting that the mam ad 16/C system may provide insight into the highly variable pharmacokinetics of ADR in man. These studies indicate that the plasma level achieved following the initial dose of ADR was a function of the tumour size in this system. The reduction in the plasma ADR level achieved following a second or third treatment at equivalent dose levels coincided with tumour regression and appeared to be an indicator of tumour response in this experimental model. The data suggest that a low initial dose of ADR escalated at the second and third treatment might increase tumour exposure without increasing the total dose and the host toxicity. This approach to therapy with ADR is under investigation. ACKNOWLEDGEMENTS This investigation was supported by Public Health Service Grant R 0 1 CA37132 awarded by L. Simpson-Herren and P . E. Noker the National Cancer Institute, National Institutes of Health, Department of Health and Human Services. The authors would like to thank Melissa Edminston, Laurie Cutts and Margaret A. Thomson for excellent technical assistance.

Journal

Cell Proliferation – Wiley

Published: May 1, 1991

There are no references for this article.