- Email: [email protected]
Disposition and tissue distribution of boron after infusion of borocaptate sodium in patients with malignant brain tumors
Int. J. Radiation Oncology Biol. Phys., Vol. 41, No. 3, pp. 631– 638, 1998 Copyright © 1998 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/98 $19.00 1 .00
PII S0360-3016(98)00069-8
●
Clinical Investigation DISPOSITION AND TISSUE DISTRIBUTION OF BORON AFTER INFUSION OF BOROCAPTATE SODIUM IN PATIENTS WITH MALIGNANT BRAIN TUMORS VLADIM´ıR HORN, D. PHARM,* JOSEF SLA´ NSKY´ , M.D.,* IVO JANKU, D.SC.,* OLDRˇ ICH SˇTROUF, PH.D.,† KAREL SˇOUREK, D.SC.‡ AND FRANTISˇ EK TOVARYSˇ , M.D.‡
*Institute of Pharmacology, Academy of Sciences of the Czech Republic, Prague, Czech Republic; †Institute of Inorganic Chemistry, ˇ ezˇ near Prague, Czech Republic; and ‡Department of Neurosurgery, Academy of Sciences of the Czech Republic, R Homolka Hospital, Prague, Czech Republic Purpose: In the frame of the Czech boron neutron capture therapy (BNCT) project, a clinical Phase I study of borocaptate sodium [Na2B12H11SH (BSH)] as the boron-10 delivery agent was performed to obtain data on disposition and tissue distribution of boron after an infusion of this compound, as well as to establish an optimal protocol for BNCT of malignant cerebral tumors. Methods and Materials: The kinetics of boron disposition after an infusion of borocaptate sodium (25 mg/kg body wt over the period of 1 h) was studied in a group of 10 patients with astrocytoma or glioblastoma of cerebral hemispheres using a modification of the Soloway–Messer colorimetric method. The boron content of tissues (tumor, healthy brain, dura mater, muscle, skin, and cranial bone) removed during the operation performed with latencies varying between 3 and 18 h was investigated by atomic emission spectrometry. Results: Compartmental analysis of boron blood concentrations has shown that in the majority of patients (four males and three females), the concentration decline can be adequately described by a two-compartment pharmacokinetic model (i.e., by a biexponential relationship). The calculated half-lives of the initial (fast) phase of the concentration decline varied between 0.85 and 3.65 h, whereas the half-life values for the terminal (slow) phase ranged between 22.2 and 111.8 h. However, in the remaining three patients (all females), the goodness of fit of the boron concentration data was significantly better when a pharmacokinetic model with three compartments was assumed. In these patients, therefore, an additional ultrafast phase with a half-life varying between 17 and 37 min was detected in the beginning of the boron blood concentration decline. On the other hand, in one of these patients, the half-life of the terminal phase was found to be 415 h (i.e., more than 17 days). Such a long persistence in the body is explained by the very high value of the total distribution volume, indicating extensive binding of BSH in peripheral tissues. Another reason may be enterohepatic recycling of BSH. Conclusion: Tumor-to-blood ratios higher than 1.5, which are necessary for an effective outcome of BNCT, can be obtained only if the time interval elapsing between the onset of surgery and termination of BSH infusion is at least 12 h. © 1998 Elsevier Science Inc. BNCT, Malignant cerebral tumors, BSH pharmacokinetics, BSH biodistribution, BSH side effects.
INTRODUCTION The fate of boron in the body after infusion of sodium borocaptate [mercaptoundecahydrododecaborate (BSH)] is of great interest for the future development of boron neutron capture therapy (BNCT) of tumors—a binary system which tries to combine chemotherapy with radiotherapy into a method characterized by greater selectivity, efficacy, and safety than the single-acting radiation and chemotherapeutic protocols commonly in use today. The principle of this method using a 10B (a stable nonradioactive isotope of boron) delivery agent such as borocaptate sodium penetrating the tumor cell lies in a nuclear fission reaction producing a-particles (malignant cell killers) which are generated in situ by irradiation with low-energy thermal neutrons.
There are several studies of boron disposition after intravascular administration of borocaptate in laboratory animals including mice, rats, rabbits, and dogs (1– 4). Moreover, borocaptate pharmacokinetics has been studied in rats using a high-performance liquid chromatography (HPLC) method specific for BSH (5). Clinical Phase I studies of boron biodistribution and pharmacokinetics following borocaptate infusion to patients with malignant brain tumors were performed in the frame of a European collaboration on BNCT (6, 7), as well as in Austria (8). However, the majority of clinical studies (6 – 8) performed on a total of 95 patients has placed more emphasis on the biodistribution of boron in tumors, normal brain, and blood, but less (six patients) on characterization of blood concentration-time data in terms
Reprint requests to: Dr. Vladimı´r Horn, Institute of Pharmacology, Academy of Sciences, Vı´den˜ska´ 1083, CZ 142 20 Prague 4,
Czech Republic. Accepted for publication 28 January 1998. 631
632
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 41, Number 3, 1998
Table 1. Basic information on patients subjected to the clinical trial Patient no.
Age
Weight (kg)
Sex
Dose (mg B)
Histology
1 2 3 4 5 6 7 8 9 10 Total mean 6 SE Males (mean 6 SE) Females (mean 6 SE)
48 60 51 42 58 32 30 44 39 59 46.20 6 3.45 50.00 6 3.54 43.67 6 5.25
68 70 109 60 65 56 78 68 56 78 70.80 6 4.91 78.75 6 10.73 65.50 6 3.48
M F M M F F F F F M 10 M 1 F 4M 6F
1003.0 1032.5 1607.8 885.0 958.8 826.0 1150.5 1003.0 826.0 1150.5 1044.31 6 72.39 1161.58 6 158.35 966.13 6 51.34
Glioblastoma Astrocytoma Glioblastoma Glioblastoma Glioblastoma Glioblastoma Glioblastoma Glioblastoma Astrocytoma Glioblastoma
of pharmacokinetic parameters other than half-lives of the boron blood concentration decline (7). In this communication, we report the results of a clinical Phase I study of borocaptate performed as a part of the Czech BNCT project (9, 10), in which equivalent attention was paid to pharmacokinetic analysis of boron disposition and to studies of boron distribution in tissues. METHODS AND MATERIALS Sodium borocaptate was synthetized at the Institute of Inorganic Chemistry, Academy of Sciences of the Czech Republic, Prague. The isotopic abundance of boron in the product was natural. An injectable pharmaceutical formulation containing lyophilized borocaptate was prepared by Le´cˇiva Pharmaceuticals (Prague) to ensure stability of the compound used in the study. Quantitation of BSH in its ionic form [(B12H11SH)22] as well as of its oxidation products [(B24H22S2)42 and (B24H22S2O)42] in the lyophilized material was carried out by HPLC as described previously (11). The content of anions in the product was 98.5% for (B12H11SH)22, 0.6% for (B24H22S2)42, and 0.5% for (B24H22S2O)42. The lyophilized formulation of borocaptate used in this study was identical to that previously used in our preclinical studies of borocaptate in rabbits (1, 12). Ten patients (four males and six females) weighing 56 – 109 kg, in the age range between 30 and 60 years, with astrocytoma or glioblastoma of cerebral hemispheres (3. or 4. malignancy grade), and with at least 2 months survival expectancy were admitted to the study after informed consent was given. The basic demographic data of the patients are listed in Table 1. None of the patients had reduced function of the kidneys or of the liver, and none suffered from another malignant disease. The plan of the study was approved by the New Drug Committee, Czech Ministry of Health, and by the local ethical commission of Homolka Hospital. Twenty-five milligrams of borocaptate (i.e., 14.75 mg of boron/kg body wt) dissolved in physiological solution were infused into each patient 3–18 h before the operation, into
the antecubital vein over the course of 1 h. The total amount of borocaptate infused varied between 1400 and 2725 mg. After termination of the BSH infusion, samples of blood (2 ml) were withdrawn from subclavian vein into heparinized polyethylene tubes at times 0, 0.25, 0.5, 1, 2, 3, 6, 12, and 24 h after the end of infusion, and later, in 24-h intervals for 7 days. Boron concentrations in whole blood were measured colorimetrically using a modification (13) of the Soloway– Messer method (14). The kidney function of the patients was monitored by measuring the clearance of inulin and p-aminohippuric acid (PAH) before, throughout, and after termination of the borocaptate infusion. The detailed results of this part of the study are reported elsewhere (15). By doing this, we have filled one of the gaps of the article by Stragliotto and Fankhauser (7) to which attention was directed in a comment on the articles by Saris (16). During the operations performed with a latency of 3 h (patients 1 and 2), 6 h (patients 3–5), 12 h (patients 6 and 7), and 18 h (patients 8 –10) after infusion, tissue samples (tumor, nontumor brain tissue, dura mater, cranial bone, muscle, and skin) for which an exposition to the neutron beam was expected were dissected from the tumor location or operation area. The content of boron in samples of tumor and nontumor brain tissue was estimated by atomic emission spectroscopy (AES) (1, 13) whereas all other tissue samples were analyzed colorimetrically using the less timeconsuming modification of the Soloway–Messer method (13). The error of BSH estimation in tissues by the former method (AES) was about 15%, whereas that of the latter amounted to 15–20% owing to interference with the biological matrix being only 5% for boric acid. Its detection limit was 0.8 ppm. The pharmacokinetic analysis of boron concentrations appearing in blood after BSH infusion was performed under the assumption of either two- or three-compartment models of boron disposition in the body. In the terminal phase of boron concentration decline, only values following a linear relationship to time after semilogarithmic plotting were included into the analysis. Also, values which were under
Disposition and distribution of boron
Fig. 1. The fit of boron blood concentration decline after termination of BSH infusion (25 mg/kg body wt over a period of 1 h) in patient 8, by a three-compartment pharmacokinetic model. CL 5 38.08 ml/min; V1 5 0.244 l/kg; Vd(ss) 5 1.051 l/kg; t1/2(1) 5 0.419 h; t1/2(2) 5 2.94 h; t1/2(3) 5 30.48 h; MRT 5 31.28 h.
the limit of detection of the colorimetric method (0.8 ppm) were excluded, as the natural levels of boron in human blood are in the range 0.06 – 0.17 mg/ml (17). The parameters of both models were estimated by nonlinear regression analysis. The goodness of fit of the measured boron concentration values to the bi- or triexponential relationships describing the models was tested statistically by means of the F test at the 5% probability level. With respect to the asymetric character of distributions of individual pharmacokinetic parameter values, geometric means with 95% limits of confidence were calculated for each parameter.
RESULTS Pharmacokinetic profile of boron after BSH infusion The multiexponential character of the postinfusion decline of boron blood concentrations is shown in Fig. 1. The maximal levels of boron in blood varied in the range 26 – 140 ppm when individually administered total doses of boron ranged between 826 and 1608 mg. The lowest values of boron blood levels smaller than 3.0 ppm were not detected earlier than the fifth day after BSH infusion. Statistical comparisons of goodness of fit of boron concentration data to two- or three-compartment pharmacokinetic models revealed that a two-compartment kinetics is sufficient for the description of boron disposition after BSH infusion in the majority of patients (four men and three women). Only in three remaining patients (all women) did a three-compartmental model fit the boron blood concentration decline significantly better. The average values of the parameter estimates of both models are presented in Table 2.
● V. HORN et al.
633
Boron tissue distribution As far as boron distribution in tissue samples dissected from the tumor and its surroundings were concerned, differences in boron content were found not only in relation to the type of tissue, but also with respect to the time interval which elapsed from the termination of BSH infusion. It is evident from Fig. 2 that 3 and 6 h after BSH infusion, a relatively high average tissue content of boron could be found with respect to the existing high boron blood concentrations not only in the tumor itself, but also in the adjacent tissues. There was a clearcut difference between maximal and average boron concentrations in the tumor in these sampling time intervals. Parallel to the decline of the average blood boron concentration (6.65 mg/g), 12 h after the end of BSH infusion, reductions of average boron content could be recognized in tissues adjacent to tumor, as well. On the other hand, the maximal and average boron concentrations in the tumor were almost identical for this sampling interval. In the last, 18-h sampling interval, a moderate increase of the boron content in nontumoral tissues was noticed notwithstanding the diminishing average concentration of boron in blood (4.33 mg/g). The difference between maximal and average boron concentrations in the tumor was small, similarly as in the previous case. The boron content in tissues of individual patients is presented in Table 3. Boron concentrations in the epitumoral healthy brain tissue varied between 6.1 and 0.8 mg/g; in all patients it was lower than in blood, tumor, and other tissues investigated, except cranial bone. Therefore, in none of the patients did the ratio of healthy brain tissue to blood exceed unity. In contrast, the tumor-to-blood ratio grew with time in relation to the diminishing boron concentration in blood, because the boron content in the tumor did not decrease in later sampling intervals. Whereas 3 and 6 h after termination of BSH infusion the average tumor-to-blood ratios did not reach unity, for the 12- and 18-h sampling intervals, the values of this ratio were higher than 2.5. The maximal and most advantageous value of the ratio of tumor to healthy brain tissue (Fig. 3) was found 12 h after BSH infusion (12.8); in all other remaining sampling intervals, the value of this ratio varied slightly above 4. Side effects of BSH infusion It should be mentioned that a hypersensitive reaction characterized by redness of the neck, respiratory disturbances, and nausea evolved in the two patients with the highest body weight. It is probable that the appearance of these adverse effects was connected with the higher rate of BSH infusion which was applied to infuse a greater amount of the compound within the predetermined time period of 60 min. The symptoms disappeared when BSH infusion was temporarily stopped or at least slowed. All other remaining patients (eight) were completely free of these hypersensitive reactions. However, when we monitored the kidney functions, we recorded a statistically significant increase in urine output in the majority of patients. The alterations of glomerular fil-
66.50 34.61–127.65
Biology
99.86 2.29–4364.02
55.83 32.07–97.19
MRT (h)
●
●
Physics
Volume 41, Number 3, 1998
tration rate as well as of renal plasma flow were of a minor degree except in one patient with a glomerular filtration rate reduction almost 50% of the initial value. The details of this part of the study are published and analyzed elsewhere (15).
5.01 1.42–17.65 0.186 0.111–0.310 19.77 11.43–34.21
1.134 0.692–1.859
0.120 0.015–0.976 12.80 1.17–140.50
1.190 0.085–16.69
0.420 0.167–1.058
49.54 30.66–80.06 1.487 0.667–3.312 0.225 0.126–0.401 23.82 12.94–43.87
1.111 0.650–1.900
44.00 25.97–74.51 2.069 1.304–3.281 1.465 0.283–7.588 0.364 0.205–0.645 35.22 3.63–341.90
M1F
F
M1F
Two compartments Geometric mean (n 5 4) 95% confidence limits Two compartments Geometric mean (n 5 3) 95% confidence limits Two compartments Geometric mean (n 5 7) 95% confidence limits Three compartments Geometric mean (n 5 3) 95% confidence limits Both models Geometric mean (n 5 10) 95% confidence limits M
F
Pharmacokinetic model Sex
CL (ml/min)
17.77 12.61–25.04
V1 (l/kg)
0.157 0.062–0.398
Vd(ss) (l/kg)
0.903 0.410–1.990
t1/2(1) (h)
1.160 0.191–7.050
t1/2(2) (h)
54.12 17.87–163.89
t1/2(3) (h)
92.80 3.26–2640.31
DISCUSSION
Patient groups
Table 2. Pharmacokinetic profile of boron after intravenous infusion of sodium borocaptate (25 mg/kg) in patients with malignant brain tumors
45.59 14.36–144.74
I. J. Radiation Oncology
65.00 21.04–200.77
634
There is some discrepancy among earlier studies investigating boron disposition kinetics after BSH infusion in glioma patients. Some authors (6, 8) were able to detect only two phases of blood boron concentration decline, but their method of pharmacokinetic data analysis comprising only an evaluation of half-lives of the biphasic boron concentration decline in blood remained unspecified; and therefore, their data cannot be regarded as reliable. On the other hand, according to Stragliotto and Fankhauser (7), who, in a limited number (six) of patients used the well-accepted nonlinear extended leastsquare regression for evaluation of parameters of the pharmacokinetic model, the time course of their boron concentrations in blood was, according to the Akaike information criterion, best described by a triexponential relationship. In the majority of our patients (7 of 10), biphasic boron kinetics was found to be statistically sufficient to describe the data; for the rest, a triexponential relationship was confirmed by the statistical test. This discrepancy is explainable by the results of the simulation study of Ludden et al. (18), indicating that the F test we used tends to select the simpler model more often than does the Akaike information criterion. Similar to previous studies, there was much variation among patients as far as the length of the individual half-lives was concerned. This was especially true for the half-lives of the terminal phase of the boron concentration decline, in which an extremely wide variation was found. It follows that in individual patients, relatively long terminal half-lives can be expected. In this respect, it is important that the longest terminal half-life reported by Stragliotto (19) was 120 h (5 days) in a patient receiving 20 mg BSH/kg. In one of our patients exhibiting three-compartment boron disposition (patient 6), the half-life of the terminal (slowest) phase of boron blood concentration decline reached 415 h (i.e., more than 17 days). With respect to our finding of boron dose-dependent disposition in rabbits (1), it is conceivable that after infusion of a higher BSH dose (25 mg/kg), not only longer terminal half-lives, but also longer mean residence times (MRT) of boron were found in our patients as compared to those of Stragliotto (19). While in the whole group of our patients the total body clearance of boron from the body was about 20 ml/min (0.285 ml/min per kg), in patient 6, a clearance value of only 6.1 ml/min (0.11 ml/min per kg) was found to be the lowest in the whole group. Such clearance values are in good agreement with the findings of Stragliotto (19), who reported boron clearances from six patients ranging between 0.107 and 0.294 ml/min per kg. However, it is unlikely that the
Disposition and distribution of boron
● V. HORN et al.
635
Fig. 2. Mean values of tissue concentration of boron in patients with malignant brain tumors at various intervals after BSH infusion (25 mg/kg body wt over a period of 1 h). Tumor 1 5 average boron content; Tumor 2 5 maximal boron content.
extremely slow elimination of boron in patient 6 in our study was due to poor function of the kidneys representing the main eliminating organ of boron from the body after borocaptate administration (6, 20), because only patients with normal kidney function entered the study (the average pretreatment inulin clearance of patient 6
was 91.3 ml/min). It is known that BSH avidly binds to human serum albumin either by formation of a disulfide linkage between the ‘‘boron cage’’ and protein (21–23) or by electrostatic forces (24). The presence of boronated proteins following BSH administration has also been demonstrated in neoplastic cells (25). Moreover, the very
Table 3. Tissue distribution of boron at various time intervals after intravenous infusion of sodium borocaptate (25 mg/kg) in patients with malignant brain tumors Interval (h) Patient no.
3 1
6 2
3
12
18
4
5
6
7
8
9
10
10.7 15.1 10.6 4.2 2.6 4.1 16.5 17.1
16.2 4.3
7.6 7.7 9.5 2.4 0.9 2.1 22.6† 22.6†
4.5
1.4
7.1
1.4 2.0 1.5 8.9 9.6
5.7 8.3 4.3 2.0 0.8 0.8 11.9 12.8
0.093 0.549 5.933
0.140 2.088 14.875
0.276 2.974 10.762
Tissue [Boron concentration (mg/g)] Blood* Dura mater Muscle Skin Bone Brain (healthy tissue) Tumor 1 (average value) Tumor 2 (maximal value)
30.15 20.1 40.1 15.0 6.1 6.1 9.6 13.6
58.7 47.8 26.7 3.1 4.1 2.2 16.6 21.6
22.5 40.0 27.9 8.2 0.6 ,0.3‡ 10.2 16.2
8.5 2.1 2.7 4.4 15.1 19.1
8.0 3.2 6.1 1.0 4.7 5.9
11.4 13.0
Ratio Brain/blood Tumor 1/blood Tumor 1/brain
0.202 0.318 1.574
0.037 0.283 7.545
,0.013 0.453 .34
0.383 1.542 4.024
* Blood sample withdrawn at the time of surgical removal of the tissues investigated. Only one assay was performed. ‡ A value below the detection limit. †
0.978 3.356 3.432
0.714 3.357 4.700
1.606
636
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 41, Number 3, 1998
Fig. 3. Mean values of boron tissue ratios in patients with malignant cerebral tumors at various intervals after BSH infusion (25 mg/kg body wt over a period of 1 h). Tumor 1 5 average boron content.
high values for the total volume of distribution suggest that the slow elimination of boron from the body might be attributed to extensive BSH binding not only in blood, but also in tissues. Retention of BSH in deeper compartments of the body with variations among patients was also suggested by the findings of Haritz et al. (6). Therefore, in our opinion, the extremely long persistence of boron in patient 6 was connected to his very high value of total distribution volume at steady state [Vd(ss) 5 3.648 l/kg body wt], indicating extensive binding of boron in tissues of this patient. From the V1/Vd(ss) ratio of patient 6, it is evident that almost 96% of the administered amount of boron was located in the peripheral, directly nonaccessible compartments of the body. In addition to our previous findings in rabbits (1), a relatively high steady-state volume of distribution approximately threefold greater than total body water was recently reported in rats as well (5). Nevertheless, the question arises as to what extent the recently reported enterohepatic excretion pathway of BSH boron (6) can contribute to its long-lasting recycling in the body. Boron blood concentration fluctuations, which were evident in the terminal phase of the boron blood concentration decline in the majority of our patients (1, 2, 3, 4, 7, and 10) several days after BSH infusion, favored the latter hypothesis. Although the BSH product as well as the analytical methodology used in the present study (conventional AES) were different from those used by groups working in the frame of European Collaboration on BNCT, it is unlikely that our results could in some way be substantially affected by these facts with respect to the almost perfect interspecies
correlation of our rabbit data (1) on BSH total body clearance and steady-state volume of distribution (r 5 0.97; p 5 0.028, and r 5 0.99; p 5 0.005, respectively) with the mouse, rat, and human data of other authors (2, 5, 6) working with BSH produced by Centronic Ltd. (Croydon, England), and using different procedures (Quanitative Neutron Capture Radiography (QNCR) and HPLC) of measurement of boron content in blood. This indicates that the pharmacokinetic properties of our lyophilized borocaptate preparation neither differ substantially from the product used by other laboratories nor are dependent on the analytical methodology used. It also appears to be irrelevant whether only concentrations of boron as such (1, 6) or, on the other hand, of the whole molecule of BSH (5, 26) were estimated in blood. As can be seen from Fig. 2, the tissues in which the boron content was measured can in principle be classified into three categories. The first is represented by tissues in which the boron content closely follows the time course of boron blood concentrations—the muscle and dura mater belong to this category. In contrast, the maximal as well as average contents of boron in the tumor tissue (second category) do not exhibit a descending tendency over time; rather, both concentrations oscillate between their mean values: 15.6 and 13.2 mg/g, respectively. This range of concentrations appears to be optimal for the effective use of the local nuclear reaction after irradiation with the neutron beam. Moreover, the small difference between maximal and average boron concentrations in the tumor tissue indicates a quite homogenous distribution of boron in the tumor, which is in some contrast to the findings of Stragliotto et al. (27). The last tissue
Disposition and distribution of boron
group is formed by tissues characterized by low boron retention (i.e., the healthy nontumor tissue, skin, and cranial bone). In this tissue group, boron levels do not exceed 5 mg/g in all sampling time intervals. There was only one exception to this rule: skin concentration measured 3 h after the end of BSH infusion. On the basis of our findings, we conclude that the most advantageous time interval for neutron irradiation is 12 h after BSH infusion. For this time interval, as shown in Fig. 3, the tumor-to-healthy brain ratio is maximal (13 on average). Simultaneously, the risk of destruction of the epitumoral healthy tissue is lowest at this time. Another advantage is offered by the fact that at this time interval, boron blood concentrations are sufficiently low to contribute to the damage of the vascular endothelium in the brain during irradiation (28). There is also
● V. HORN et al.
637
agreement between our observations and those of Stragliotto and Fankhauser (7) in that a high rate of BSH infusion may be critical as far as the appearance of some reversible adverse reactions is concerned. Our findings also demonstrate that retention of boron in the body of individual patients can be abnormally high, which could contribute to an undesirable accumulation of BSH in the kidney, the main organ of BSH elimination in the body (5–7, 20). In this way, a potential nephrotoxic risk (12) could arise when repeated BSH infusions advocated in some proposals of BNCT clinical trials (7, 11, 29) should be used. Another risk could result from competition between BSH and concomitant drug therapy for binding sites on the albumin molecule, especially if binding of BSH is by electrostatic forces (24).
REFERENCES 1. Buchar, E.; Bedna´ˇrova´, S.; Gru¨ner, B.; Walder, P.; Sˇtrouf, O.; Janku˚, I. Dose-dependent disposition kinetics and tissue accumulation of boron after intravenous injection of sodium mercaptoundecahydrododecaborate in rabbits. Cancer Chemother. Pharmacol. 29:450 – 454; 1992. 2. Gregoire, V.; Begg, A. C.; Huiskamp, R.; Verrijk, R.; Bartelink, H. Selectivity of boron carriers for boron neutron capture therapy: Pharmacological studies with borocaptate sodium, L-borophenylalanine and boric acid in murine tumors. Radiother. Oncol. 27:46 –54; 1993. 3. Joel, D.; Slatkin, D.; Fairchild, R.; Micca, P.; Nawrocky, M. Pharmacokinetics and tissue distribution of the sulfhydryl boranes (monomers and dimer) in glioma-bearing rats. Strahlenther. Onkol. 165:167–170; 1989. 4. Kraft, S. L.; Gavin, P. R.; Leathers, C. W.; DeHann, C. E.; Bauer, W. F.; Miller, D. L.; Dorn, R. W.; Griebenow, M. L. Biodistribution of boron in dogs with spontaneous intracranial tumors following borocaptate administration. Cancer Res. 54: 1259 –1263; 1994. 5. Mehta, S. C.; Boudinot, F. D.; Lu, D. R. Pharmacokinetics of sodium mercaptoundecahydrododecaborate after intravenous injection in rats. Drug. Metab. Disp. 23:1368 –1371; 1995. 6. Haritz, D.; Gabel, D.; Huiskamp, R. Clinical Phase-I study of Na2B12H11SH (BSH) in patients with malignant glioma as precondition for boron neutron capture therapy (BNCT). Int. J. Radiat. Oncol. Biol. Phys. 28:1175–1181; 1994. 7. Stragliotto, G.; Fankhauser, H. Biodistribution of boron sulfhydryl for boron neutron capture therapy in patients with intracranial tumors. Neurosurgery 36:285–293; 1995. 8. Haselsberger, K.; Radner, H.; Pendl, G. Boron neutron capture therapy: Boron biodistribution and pharmacokinetics of Na2B12H11SH in patients with glioblastoma. Cancer Res. 54: 6318 – 6320; 1994. 9. Burian, J.; Janku˚, I.; Kvı´tek, J.; Maresˇ, V.; Prouza, Z.; Spurny´, F.; Sˇourek, K.; Sˇtı´br, B.; Sˇtrouf, O. BNCT project in Czechoslovakia. In: Fairchild, R. G.; Bond, V. P.; Woodhead, W. D., eds. Clinical aspects of neutron capture therapy. New York: Plenum Press; 1989:39. 10. Sˇourek, K.; Tovarysˇ, F. Perspectives of the clinical BNCT of malignant gliomas in Czechoslovakia. Strahlenther. Onkol. 165:247–249; 1989. 11. Dewit, L.; Mijnheer, B.; Moss, R.; Gabel, D. A proposal for clinical pilot studies for BNCT. In: Abstracts of the Fourth International Symposium on Neutron Capture Therapy for Cancer, 1990:167.
12. Janku˚, I.; Buchar, E.; Jirˇicˇka, Z. Nephrotoxicity of borocaptate after short-term administration in rabbits. Toxicology 79:99 – 108; 1993. 13. Sˇtrouf, O.; Mertenova´, E.; Schneiderova´, L.; Za´mecˇnı´kova´, H.; Janku˚, I. Boron determination for neutron capture therapy by colorimetry and emission spectrometry. Strahlenther. Onkol. 165:174 –176; 1989. 14. Soloway, A. H.; Messer, J. R. Detection of hydrolytically stable hydrides in biological materials. Anal. Chem. 36:433– 434; 1964. 15. Horn, V.; Sla´nsky´, J.; Buchar, E.; Janku˚, I.; Sˇourek, K.; Tovarysˇ, F. The diuretic effect of borocaptate sodium in rats and in patients with brain tumors. Methods Find. Exp. Clin. Pharmacol. 19:559 –566; 1997. 16. Saris, S. Comment to the paper of G. Stragliotto and H. Fankhauser: Biodistribution of boron sulfhydryl for boron neutron capture therapy in patients with intracranial tumors. Neurosurgery 16:293; 1995. 17. Abou-Shakra, F. R.; Havercroft, J. M.; Ward, N. I. Lithium and boron in biological tissues and fluids. Trace Elements Med. 6:142–146; 1989. 18. Ludden, T. M.; Beal, S. L.; Sheiner, L. B. Comparison of the Akaike information criterion, the Schwartz criterion and the F test as guides to model selection. J. Pharmacokin. Biopharm. 22:431– 445; 1994. 19. Stragliotto, G. Biodistribution and pharmacokinetics of boron sulfhydryl for boron neutron capture therapy in patients with intracranial tumours. Thesis, Faculte´ de Me´decine de l Universite´ de Lausanne, 1993. 20. Sweet, W. H.; Messer, J. R.; Hatanaka, H. Supplementary pharmacological study between 1972 and 1977 on purified mercaptoundecahydrododecaborate. In: Hatanaka, H., ed. Boron neutron capture therapy for tumors. Niigata: Nishimura; 1986:59 –76. 21. Nakagawa, T.; Nagai, T. Interaction between serum albumin and mercaptoundecahydrododecaborate ion (an agent for boron neutron capture therapy of brain tumor). Chem. Pharm. Bull. Tokyo 24:2934 –2948; 1976. 22. Samsel, E. G.; Miller, D. L. High resolution 10B and 11 B nuclear magnetic resonance (NMR) spectroscopy of Na2B12H11SH impurities and metabolites. Strahlenther. Onkol. 165:140 –141; 1989. 23. Soloway, A. H.; Hatanaka, H.; Davis, M. A. Penetration of brain and brain tumor. VII. Tumor-binding sulfhydryl boron compounds. J. Med. Chem. 10:714 –717; 1967. 24. Zhu-Tang, P.-P. P.; Schweizer, M. P.; Bradshaw, K. M.;
638
I. J. Radiation Oncology
●
Biology
●
Physics
Bauer, W. F. 11B nuclear magnetic resonance studies of the interaction of borocaptate sodium with serum albumin. Biochem. Pharmacol. 49:625– 632; 1995. 25. Soloway, A. H.; Alam, F.; Barth, R. F.; Bapat, B. V. Boron chemistry and target cell affinity. Strahlenther. Onkol. 165: 118 –120; 1989. 26. Mehta, S. C.; Lu, D. R. Interspecies pharmacokinetic scaling of BSH in mice, rats, rabbits, and humans. Biopharm. Drug. Disp. 16:735–744; 1995. 27. Stragliotto, G.; Munafo, A.; Biollaz, J.; Fankhauser, H. Pharmacokinetics of boron sulfhydryl (BSH) in patients with in-
Volume 41, Number 3, 1998
tracranial tumours. In: Allen, B. J., et al., eds. Progress in neutron capture therapy for cancer. New York: Plenum Press; 1992:549 –550. 28. Asbury, A. K.; Ojeman, R. G.; Nielsen, S. L.; Sweet, W. H. Neuropathologic study of fourteen cases of malignant brain tumor treated by boron-10 slow neutron capture therapy. J. Neuropathol. Exp. Neurol. 31:278 –303; 1972. 29. Gabel, D. Minutes of the meeting of the Project Management Group Lausanne. In: Newsletter of the European Collaboration on Boron Neutron Capture Therapy of Tumors. 12:2–3; 1992.
PII S0360-3016(98)00069-8
●
Clinical Investigation DISPOSITION AND TISSUE DISTRIBUTION OF BORON AFTER INFUSION OF BOROCAPTATE SODIUM IN PATIENTS WITH MALIGNANT BRAIN TUMORS VLADIM´ıR HORN, D. PHARM,* JOSEF SLA´ NSKY´ , M.D.,* IVO JANKU, D.SC.,* OLDRˇ ICH SˇTROUF, PH.D.,† KAREL SˇOUREK, D.SC.‡ AND FRANTISˇ EK TOVARYSˇ , M.D.‡
*Institute of Pharmacology, Academy of Sciences of the Czech Republic, Prague, Czech Republic; †Institute of Inorganic Chemistry, ˇ ezˇ near Prague, Czech Republic; and ‡Department of Neurosurgery, Academy of Sciences of the Czech Republic, R Homolka Hospital, Prague, Czech Republic Purpose: In the frame of the Czech boron neutron capture therapy (BNCT) project, a clinical Phase I study of borocaptate sodium [Na2B12H11SH (BSH)] as the boron-10 delivery agent was performed to obtain data on disposition and tissue distribution of boron after an infusion of this compound, as well as to establish an optimal protocol for BNCT of malignant cerebral tumors. Methods and Materials: The kinetics of boron disposition after an infusion of borocaptate sodium (25 mg/kg body wt over the period of 1 h) was studied in a group of 10 patients with astrocytoma or glioblastoma of cerebral hemispheres using a modification of the Soloway–Messer colorimetric method. The boron content of tissues (tumor, healthy brain, dura mater, muscle, skin, and cranial bone) removed during the operation performed with latencies varying between 3 and 18 h was investigated by atomic emission spectrometry. Results: Compartmental analysis of boron blood concentrations has shown that in the majority of patients (four males and three females), the concentration decline can be adequately described by a two-compartment pharmacokinetic model (i.e., by a biexponential relationship). The calculated half-lives of the initial (fast) phase of the concentration decline varied between 0.85 and 3.65 h, whereas the half-life values for the terminal (slow) phase ranged between 22.2 and 111.8 h. However, in the remaining three patients (all females), the goodness of fit of the boron concentration data was significantly better when a pharmacokinetic model with three compartments was assumed. In these patients, therefore, an additional ultrafast phase with a half-life varying between 17 and 37 min was detected in the beginning of the boron blood concentration decline. On the other hand, in one of these patients, the half-life of the terminal phase was found to be 415 h (i.e., more than 17 days). Such a long persistence in the body is explained by the very high value of the total distribution volume, indicating extensive binding of BSH in peripheral tissues. Another reason may be enterohepatic recycling of BSH. Conclusion: Tumor-to-blood ratios higher than 1.5, which are necessary for an effective outcome of BNCT, can be obtained only if the time interval elapsing between the onset of surgery and termination of BSH infusion is at least 12 h. © 1998 Elsevier Science Inc. BNCT, Malignant cerebral tumors, BSH pharmacokinetics, BSH biodistribution, BSH side effects.
INTRODUCTION The fate of boron in the body after infusion of sodium borocaptate [mercaptoundecahydrododecaborate (BSH)] is of great interest for the future development of boron neutron capture therapy (BNCT) of tumors—a binary system which tries to combine chemotherapy with radiotherapy into a method characterized by greater selectivity, efficacy, and safety than the single-acting radiation and chemotherapeutic protocols commonly in use today. The principle of this method using a 10B (a stable nonradioactive isotope of boron) delivery agent such as borocaptate sodium penetrating the tumor cell lies in a nuclear fission reaction producing a-particles (malignant cell killers) which are generated in situ by irradiation with low-energy thermal neutrons.
There are several studies of boron disposition after intravascular administration of borocaptate in laboratory animals including mice, rats, rabbits, and dogs (1– 4). Moreover, borocaptate pharmacokinetics has been studied in rats using a high-performance liquid chromatography (HPLC) method specific for BSH (5). Clinical Phase I studies of boron biodistribution and pharmacokinetics following borocaptate infusion to patients with malignant brain tumors were performed in the frame of a European collaboration on BNCT (6, 7), as well as in Austria (8). However, the majority of clinical studies (6 – 8) performed on a total of 95 patients has placed more emphasis on the biodistribution of boron in tumors, normal brain, and blood, but less (six patients) on characterization of blood concentration-time data in terms
Reprint requests to: Dr. Vladimı´r Horn, Institute of Pharmacology, Academy of Sciences, Vı´den˜ska´ 1083, CZ 142 20 Prague 4,
Czech Republic. Accepted for publication 28 January 1998. 631
632
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 41, Number 3, 1998
Table 1. Basic information on patients subjected to the clinical trial Patient no.
Age
Weight (kg)
Sex
Dose (mg B)
Histology
1 2 3 4 5 6 7 8 9 10 Total mean 6 SE Males (mean 6 SE) Females (mean 6 SE)
48 60 51 42 58 32 30 44 39 59 46.20 6 3.45 50.00 6 3.54 43.67 6 5.25
68 70 109 60 65 56 78 68 56 78 70.80 6 4.91 78.75 6 10.73 65.50 6 3.48
M F M M F F F F F M 10 M 1 F 4M 6F
1003.0 1032.5 1607.8 885.0 958.8 826.0 1150.5 1003.0 826.0 1150.5 1044.31 6 72.39 1161.58 6 158.35 966.13 6 51.34
Glioblastoma Astrocytoma Glioblastoma Glioblastoma Glioblastoma Glioblastoma Glioblastoma Glioblastoma Astrocytoma Glioblastoma
of pharmacokinetic parameters other than half-lives of the boron blood concentration decline (7). In this communication, we report the results of a clinical Phase I study of borocaptate performed as a part of the Czech BNCT project (9, 10), in which equivalent attention was paid to pharmacokinetic analysis of boron disposition and to studies of boron distribution in tissues. METHODS AND MATERIALS Sodium borocaptate was synthetized at the Institute of Inorganic Chemistry, Academy of Sciences of the Czech Republic, Prague. The isotopic abundance of boron in the product was natural. An injectable pharmaceutical formulation containing lyophilized borocaptate was prepared by Le´cˇiva Pharmaceuticals (Prague) to ensure stability of the compound used in the study. Quantitation of BSH in its ionic form [(B12H11SH)22] as well as of its oxidation products [(B24H22S2)42 and (B24H22S2O)42] in the lyophilized material was carried out by HPLC as described previously (11). The content of anions in the product was 98.5% for (B12H11SH)22, 0.6% for (B24H22S2)42, and 0.5% for (B24H22S2O)42. The lyophilized formulation of borocaptate used in this study was identical to that previously used in our preclinical studies of borocaptate in rabbits (1, 12). Ten patients (four males and six females) weighing 56 – 109 kg, in the age range between 30 and 60 years, with astrocytoma or glioblastoma of cerebral hemispheres (3. or 4. malignancy grade), and with at least 2 months survival expectancy were admitted to the study after informed consent was given. The basic demographic data of the patients are listed in Table 1. None of the patients had reduced function of the kidneys or of the liver, and none suffered from another malignant disease. The plan of the study was approved by the New Drug Committee, Czech Ministry of Health, and by the local ethical commission of Homolka Hospital. Twenty-five milligrams of borocaptate (i.e., 14.75 mg of boron/kg body wt) dissolved in physiological solution were infused into each patient 3–18 h before the operation, into
the antecubital vein over the course of 1 h. The total amount of borocaptate infused varied between 1400 and 2725 mg. After termination of the BSH infusion, samples of blood (2 ml) were withdrawn from subclavian vein into heparinized polyethylene tubes at times 0, 0.25, 0.5, 1, 2, 3, 6, 12, and 24 h after the end of infusion, and later, in 24-h intervals for 7 days. Boron concentrations in whole blood were measured colorimetrically using a modification (13) of the Soloway– Messer method (14). The kidney function of the patients was monitored by measuring the clearance of inulin and p-aminohippuric acid (PAH) before, throughout, and after termination of the borocaptate infusion. The detailed results of this part of the study are reported elsewhere (15). By doing this, we have filled one of the gaps of the article by Stragliotto and Fankhauser (7) to which attention was directed in a comment on the articles by Saris (16). During the operations performed with a latency of 3 h (patients 1 and 2), 6 h (patients 3–5), 12 h (patients 6 and 7), and 18 h (patients 8 –10) after infusion, tissue samples (tumor, nontumor brain tissue, dura mater, cranial bone, muscle, and skin) for which an exposition to the neutron beam was expected were dissected from the tumor location or operation area. The content of boron in samples of tumor and nontumor brain tissue was estimated by atomic emission spectroscopy (AES) (1, 13) whereas all other tissue samples were analyzed colorimetrically using the less timeconsuming modification of the Soloway–Messer method (13). The error of BSH estimation in tissues by the former method (AES) was about 15%, whereas that of the latter amounted to 15–20% owing to interference with the biological matrix being only 5% for boric acid. Its detection limit was 0.8 ppm. The pharmacokinetic analysis of boron concentrations appearing in blood after BSH infusion was performed under the assumption of either two- or three-compartment models of boron disposition in the body. In the terminal phase of boron concentration decline, only values following a linear relationship to time after semilogarithmic plotting were included into the analysis. Also, values which were under
Disposition and distribution of boron
Fig. 1. The fit of boron blood concentration decline after termination of BSH infusion (25 mg/kg body wt over a period of 1 h) in patient 8, by a three-compartment pharmacokinetic model. CL 5 38.08 ml/min; V1 5 0.244 l/kg; Vd(ss) 5 1.051 l/kg; t1/2(1) 5 0.419 h; t1/2(2) 5 2.94 h; t1/2(3) 5 30.48 h; MRT 5 31.28 h.
the limit of detection of the colorimetric method (0.8 ppm) were excluded, as the natural levels of boron in human blood are in the range 0.06 – 0.17 mg/ml (17). The parameters of both models were estimated by nonlinear regression analysis. The goodness of fit of the measured boron concentration values to the bi- or triexponential relationships describing the models was tested statistically by means of the F test at the 5% probability level. With respect to the asymetric character of distributions of individual pharmacokinetic parameter values, geometric means with 95% limits of confidence were calculated for each parameter.
RESULTS Pharmacokinetic profile of boron after BSH infusion The multiexponential character of the postinfusion decline of boron blood concentrations is shown in Fig. 1. The maximal levels of boron in blood varied in the range 26 – 140 ppm when individually administered total doses of boron ranged between 826 and 1608 mg. The lowest values of boron blood levels smaller than 3.0 ppm were not detected earlier than the fifth day after BSH infusion. Statistical comparisons of goodness of fit of boron concentration data to two- or three-compartment pharmacokinetic models revealed that a two-compartment kinetics is sufficient for the description of boron disposition after BSH infusion in the majority of patients (four men and three women). Only in three remaining patients (all women) did a three-compartmental model fit the boron blood concentration decline significantly better. The average values of the parameter estimates of both models are presented in Table 2.
● V. HORN et al.
633
Boron tissue distribution As far as boron distribution in tissue samples dissected from the tumor and its surroundings were concerned, differences in boron content were found not only in relation to the type of tissue, but also with respect to the time interval which elapsed from the termination of BSH infusion. It is evident from Fig. 2 that 3 and 6 h after BSH infusion, a relatively high average tissue content of boron could be found with respect to the existing high boron blood concentrations not only in the tumor itself, but also in the adjacent tissues. There was a clearcut difference between maximal and average boron concentrations in the tumor in these sampling time intervals. Parallel to the decline of the average blood boron concentration (6.65 mg/g), 12 h after the end of BSH infusion, reductions of average boron content could be recognized in tissues adjacent to tumor, as well. On the other hand, the maximal and average boron concentrations in the tumor were almost identical for this sampling interval. In the last, 18-h sampling interval, a moderate increase of the boron content in nontumoral tissues was noticed notwithstanding the diminishing average concentration of boron in blood (4.33 mg/g). The difference between maximal and average boron concentrations in the tumor was small, similarly as in the previous case. The boron content in tissues of individual patients is presented in Table 3. Boron concentrations in the epitumoral healthy brain tissue varied between 6.1 and 0.8 mg/g; in all patients it was lower than in blood, tumor, and other tissues investigated, except cranial bone. Therefore, in none of the patients did the ratio of healthy brain tissue to blood exceed unity. In contrast, the tumor-to-blood ratio grew with time in relation to the diminishing boron concentration in blood, because the boron content in the tumor did not decrease in later sampling intervals. Whereas 3 and 6 h after termination of BSH infusion the average tumor-to-blood ratios did not reach unity, for the 12- and 18-h sampling intervals, the values of this ratio were higher than 2.5. The maximal and most advantageous value of the ratio of tumor to healthy brain tissue (Fig. 3) was found 12 h after BSH infusion (12.8); in all other remaining sampling intervals, the value of this ratio varied slightly above 4. Side effects of BSH infusion It should be mentioned that a hypersensitive reaction characterized by redness of the neck, respiratory disturbances, and nausea evolved in the two patients with the highest body weight. It is probable that the appearance of these adverse effects was connected with the higher rate of BSH infusion which was applied to infuse a greater amount of the compound within the predetermined time period of 60 min. The symptoms disappeared when BSH infusion was temporarily stopped or at least slowed. All other remaining patients (eight) were completely free of these hypersensitive reactions. However, when we monitored the kidney functions, we recorded a statistically significant increase in urine output in the majority of patients. The alterations of glomerular fil-
66.50 34.61–127.65
Biology
99.86 2.29–4364.02
55.83 32.07–97.19
MRT (h)
●
●
Physics
Volume 41, Number 3, 1998
tration rate as well as of renal plasma flow were of a minor degree except in one patient with a glomerular filtration rate reduction almost 50% of the initial value. The details of this part of the study are published and analyzed elsewhere (15).
5.01 1.42–17.65 0.186 0.111–0.310 19.77 11.43–34.21
1.134 0.692–1.859
0.120 0.015–0.976 12.80 1.17–140.50
1.190 0.085–16.69
0.420 0.167–1.058
49.54 30.66–80.06 1.487 0.667–3.312 0.225 0.126–0.401 23.82 12.94–43.87
1.111 0.650–1.900
44.00 25.97–74.51 2.069 1.304–3.281 1.465 0.283–7.588 0.364 0.205–0.645 35.22 3.63–341.90
M1F
F
M1F
Two compartments Geometric mean (n 5 4) 95% confidence limits Two compartments Geometric mean (n 5 3) 95% confidence limits Two compartments Geometric mean (n 5 7) 95% confidence limits Three compartments Geometric mean (n 5 3) 95% confidence limits Both models Geometric mean (n 5 10) 95% confidence limits M
F
Pharmacokinetic model Sex
CL (ml/min)
17.77 12.61–25.04
V1 (l/kg)
0.157 0.062–0.398
Vd(ss) (l/kg)
0.903 0.410–1.990
t1/2(1) (h)
1.160 0.191–7.050
t1/2(2) (h)
54.12 17.87–163.89
t1/2(3) (h)
92.80 3.26–2640.31
DISCUSSION
Patient groups
Table 2. Pharmacokinetic profile of boron after intravenous infusion of sodium borocaptate (25 mg/kg) in patients with malignant brain tumors
45.59 14.36–144.74
I. J. Radiation Oncology
65.00 21.04–200.77
634
There is some discrepancy among earlier studies investigating boron disposition kinetics after BSH infusion in glioma patients. Some authors (6, 8) were able to detect only two phases of blood boron concentration decline, but their method of pharmacokinetic data analysis comprising only an evaluation of half-lives of the biphasic boron concentration decline in blood remained unspecified; and therefore, their data cannot be regarded as reliable. On the other hand, according to Stragliotto and Fankhauser (7), who, in a limited number (six) of patients used the well-accepted nonlinear extended leastsquare regression for evaluation of parameters of the pharmacokinetic model, the time course of their boron concentrations in blood was, according to the Akaike information criterion, best described by a triexponential relationship. In the majority of our patients (7 of 10), biphasic boron kinetics was found to be statistically sufficient to describe the data; for the rest, a triexponential relationship was confirmed by the statistical test. This discrepancy is explainable by the results of the simulation study of Ludden et al. (18), indicating that the F test we used tends to select the simpler model more often than does the Akaike information criterion. Similar to previous studies, there was much variation among patients as far as the length of the individual half-lives was concerned. This was especially true for the half-lives of the terminal phase of the boron concentration decline, in which an extremely wide variation was found. It follows that in individual patients, relatively long terminal half-lives can be expected. In this respect, it is important that the longest terminal half-life reported by Stragliotto (19) was 120 h (5 days) in a patient receiving 20 mg BSH/kg. In one of our patients exhibiting three-compartment boron disposition (patient 6), the half-life of the terminal (slowest) phase of boron blood concentration decline reached 415 h (i.e., more than 17 days). With respect to our finding of boron dose-dependent disposition in rabbits (1), it is conceivable that after infusion of a higher BSH dose (25 mg/kg), not only longer terminal half-lives, but also longer mean residence times (MRT) of boron were found in our patients as compared to those of Stragliotto (19). While in the whole group of our patients the total body clearance of boron from the body was about 20 ml/min (0.285 ml/min per kg), in patient 6, a clearance value of only 6.1 ml/min (0.11 ml/min per kg) was found to be the lowest in the whole group. Such clearance values are in good agreement with the findings of Stragliotto (19), who reported boron clearances from six patients ranging between 0.107 and 0.294 ml/min per kg. However, it is unlikely that the
Disposition and distribution of boron
● V. HORN et al.
635
Fig. 2. Mean values of tissue concentration of boron in patients with malignant brain tumors at various intervals after BSH infusion (25 mg/kg body wt over a period of 1 h). Tumor 1 5 average boron content; Tumor 2 5 maximal boron content.
extremely slow elimination of boron in patient 6 in our study was due to poor function of the kidneys representing the main eliminating organ of boron from the body after borocaptate administration (6, 20), because only patients with normal kidney function entered the study (the average pretreatment inulin clearance of patient 6
was 91.3 ml/min). It is known that BSH avidly binds to human serum albumin either by formation of a disulfide linkage between the ‘‘boron cage’’ and protein (21–23) or by electrostatic forces (24). The presence of boronated proteins following BSH administration has also been demonstrated in neoplastic cells (25). Moreover, the very
Table 3. Tissue distribution of boron at various time intervals after intravenous infusion of sodium borocaptate (25 mg/kg) in patients with malignant brain tumors Interval (h) Patient no.
3 1
6 2
3
12
18
4
5
6
7
8
9
10
10.7 15.1 10.6 4.2 2.6 4.1 16.5 17.1
16.2 4.3
7.6 7.7 9.5 2.4 0.9 2.1 22.6† 22.6†
4.5
1.4
7.1
1.4 2.0 1.5 8.9 9.6
5.7 8.3 4.3 2.0 0.8 0.8 11.9 12.8
0.093 0.549 5.933
0.140 2.088 14.875
0.276 2.974 10.762
Tissue [Boron concentration (mg/g)] Blood* Dura mater Muscle Skin Bone Brain (healthy tissue) Tumor 1 (average value) Tumor 2 (maximal value)
30.15 20.1 40.1 15.0 6.1 6.1 9.6 13.6
58.7 47.8 26.7 3.1 4.1 2.2 16.6 21.6
22.5 40.0 27.9 8.2 0.6 ,0.3‡ 10.2 16.2
8.5 2.1 2.7 4.4 15.1 19.1
8.0 3.2 6.1 1.0 4.7 5.9
11.4 13.0
Ratio Brain/blood Tumor 1/blood Tumor 1/brain
0.202 0.318 1.574
0.037 0.283 7.545
,0.013 0.453 .34
0.383 1.542 4.024
* Blood sample withdrawn at the time of surgical removal of the tissues investigated. Only one assay was performed. ‡ A value below the detection limit. †
0.978 3.356 3.432
0.714 3.357 4.700
1.606
636
I. J. Radiation Oncology
●
Biology
●
Physics
Volume 41, Number 3, 1998
Fig. 3. Mean values of boron tissue ratios in patients with malignant cerebral tumors at various intervals after BSH infusion (25 mg/kg body wt over a period of 1 h). Tumor 1 5 average boron content.
high values for the total volume of distribution suggest that the slow elimination of boron from the body might be attributed to extensive BSH binding not only in blood, but also in tissues. Retention of BSH in deeper compartments of the body with variations among patients was also suggested by the findings of Haritz et al. (6). Therefore, in our opinion, the extremely long persistence of boron in patient 6 was connected to his very high value of total distribution volume at steady state [Vd(ss) 5 3.648 l/kg body wt], indicating extensive binding of boron in tissues of this patient. From the V1/Vd(ss) ratio of patient 6, it is evident that almost 96% of the administered amount of boron was located in the peripheral, directly nonaccessible compartments of the body. In addition to our previous findings in rabbits (1), a relatively high steady-state volume of distribution approximately threefold greater than total body water was recently reported in rats as well (5). Nevertheless, the question arises as to what extent the recently reported enterohepatic excretion pathway of BSH boron (6) can contribute to its long-lasting recycling in the body. Boron blood concentration fluctuations, which were evident in the terminal phase of the boron blood concentration decline in the majority of our patients (1, 2, 3, 4, 7, and 10) several days after BSH infusion, favored the latter hypothesis. Although the BSH product as well as the analytical methodology used in the present study (conventional AES) were different from those used by groups working in the frame of European Collaboration on BNCT, it is unlikely that our results could in some way be substantially affected by these facts with respect to the almost perfect interspecies
correlation of our rabbit data (1) on BSH total body clearance and steady-state volume of distribution (r 5 0.97; p 5 0.028, and r 5 0.99; p 5 0.005, respectively) with the mouse, rat, and human data of other authors (2, 5, 6) working with BSH produced by Centronic Ltd. (Croydon, England), and using different procedures (Quanitative Neutron Capture Radiography (QNCR) and HPLC) of measurement of boron content in blood. This indicates that the pharmacokinetic properties of our lyophilized borocaptate preparation neither differ substantially from the product used by other laboratories nor are dependent on the analytical methodology used. It also appears to be irrelevant whether only concentrations of boron as such (1, 6) or, on the other hand, of the whole molecule of BSH (5, 26) were estimated in blood. As can be seen from Fig. 2, the tissues in which the boron content was measured can in principle be classified into three categories. The first is represented by tissues in which the boron content closely follows the time course of boron blood concentrations—the muscle and dura mater belong to this category. In contrast, the maximal as well as average contents of boron in the tumor tissue (second category) do not exhibit a descending tendency over time; rather, both concentrations oscillate between their mean values: 15.6 and 13.2 mg/g, respectively. This range of concentrations appears to be optimal for the effective use of the local nuclear reaction after irradiation with the neutron beam. Moreover, the small difference between maximal and average boron concentrations in the tumor tissue indicates a quite homogenous distribution of boron in the tumor, which is in some contrast to the findings of Stragliotto et al. (27). The last tissue
Disposition and distribution of boron
group is formed by tissues characterized by low boron retention (i.e., the healthy nontumor tissue, skin, and cranial bone). In this tissue group, boron levels do not exceed 5 mg/g in all sampling time intervals. There was only one exception to this rule: skin concentration measured 3 h after the end of BSH infusion. On the basis of our findings, we conclude that the most advantageous time interval for neutron irradiation is 12 h after BSH infusion. For this time interval, as shown in Fig. 3, the tumor-to-healthy brain ratio is maximal (13 on average). Simultaneously, the risk of destruction of the epitumoral healthy tissue is lowest at this time. Another advantage is offered by the fact that at this time interval, boron blood concentrations are sufficiently low to contribute to the damage of the vascular endothelium in the brain during irradiation (28). There is also
● V. HORN et al.
637
agreement between our observations and those of Stragliotto and Fankhauser (7) in that a high rate of BSH infusion may be critical as far as the appearance of some reversible adverse reactions is concerned. Our findings also demonstrate that retention of boron in the body of individual patients can be abnormally high, which could contribute to an undesirable accumulation of BSH in the kidney, the main organ of BSH elimination in the body (5–7, 20). In this way, a potential nephrotoxic risk (12) could arise when repeated BSH infusions advocated in some proposals of BNCT clinical trials (7, 11, 29) should be used. Another risk could result from competition between BSH and concomitant drug therapy for binding sites on the albumin molecule, especially if binding of BSH is by electrostatic forces (24).
REFERENCES 1. Buchar, E.; Bedna´ˇrova´, S.; Gru¨ner, B.; Walder, P.; Sˇtrouf, O.; Janku˚, I. Dose-dependent disposition kinetics and tissue accumulation of boron after intravenous injection of sodium mercaptoundecahydrododecaborate in rabbits. Cancer Chemother. Pharmacol. 29:450 – 454; 1992. 2. Gregoire, V.; Begg, A. C.; Huiskamp, R.; Verrijk, R.; Bartelink, H. Selectivity of boron carriers for boron neutron capture therapy: Pharmacological studies with borocaptate sodium, L-borophenylalanine and boric acid in murine tumors. Radiother. Oncol. 27:46 –54; 1993. 3. Joel, D.; Slatkin, D.; Fairchild, R.; Micca, P.; Nawrocky, M. Pharmacokinetics and tissue distribution of the sulfhydryl boranes (monomers and dimer) in glioma-bearing rats. Strahlenther. Onkol. 165:167–170; 1989. 4. Kraft, S. L.; Gavin, P. R.; Leathers, C. W.; DeHann, C. E.; Bauer, W. F.; Miller, D. L.; Dorn, R. W.; Griebenow, M. L. Biodistribution of boron in dogs with spontaneous intracranial tumors following borocaptate administration. Cancer Res. 54: 1259 –1263; 1994. 5. Mehta, S. C.; Boudinot, F. D.; Lu, D. R. Pharmacokinetics of sodium mercaptoundecahydrododecaborate after intravenous injection in rats. Drug. Metab. Disp. 23:1368 –1371; 1995. 6. Haritz, D.; Gabel, D.; Huiskamp, R. Clinical Phase-I study of Na2B12H11SH (BSH) in patients with malignant glioma as precondition for boron neutron capture therapy (BNCT). Int. J. Radiat. Oncol. Biol. Phys. 28:1175–1181; 1994. 7. Stragliotto, G.; Fankhauser, H. Biodistribution of boron sulfhydryl for boron neutron capture therapy in patients with intracranial tumors. Neurosurgery 36:285–293; 1995. 8. Haselsberger, K.; Radner, H.; Pendl, G. Boron neutron capture therapy: Boron biodistribution and pharmacokinetics of Na2B12H11SH in patients with glioblastoma. Cancer Res. 54: 6318 – 6320; 1994. 9. Burian, J.; Janku˚, I.; Kvı´tek, J.; Maresˇ, V.; Prouza, Z.; Spurny´, F.; Sˇourek, K.; Sˇtı´br, B.; Sˇtrouf, O. BNCT project in Czechoslovakia. In: Fairchild, R. G.; Bond, V. P.; Woodhead, W. D., eds. Clinical aspects of neutron capture therapy. New York: Plenum Press; 1989:39. 10. Sˇourek, K.; Tovarysˇ, F. Perspectives of the clinical BNCT of malignant gliomas in Czechoslovakia. Strahlenther. Onkol. 165:247–249; 1989. 11. Dewit, L.; Mijnheer, B.; Moss, R.; Gabel, D. A proposal for clinical pilot studies for BNCT. In: Abstracts of the Fourth International Symposium on Neutron Capture Therapy for Cancer, 1990:167.
12. Janku˚, I.; Buchar, E.; Jirˇicˇka, Z. Nephrotoxicity of borocaptate after short-term administration in rabbits. Toxicology 79:99 – 108; 1993. 13. Sˇtrouf, O.; Mertenova´, E.; Schneiderova´, L.; Za´mecˇnı´kova´, H.; Janku˚, I. Boron determination for neutron capture therapy by colorimetry and emission spectrometry. Strahlenther. Onkol. 165:174 –176; 1989. 14. Soloway, A. H.; Messer, J. R. Detection of hydrolytically stable hydrides in biological materials. Anal. Chem. 36:433– 434; 1964. 15. Horn, V.; Sla´nsky´, J.; Buchar, E.; Janku˚, I.; Sˇourek, K.; Tovarysˇ, F. The diuretic effect of borocaptate sodium in rats and in patients with brain tumors. Methods Find. Exp. Clin. Pharmacol. 19:559 –566; 1997. 16. Saris, S. Comment to the paper of G. Stragliotto and H. Fankhauser: Biodistribution of boron sulfhydryl for boron neutron capture therapy in patients with intracranial tumors. Neurosurgery 16:293; 1995. 17. Abou-Shakra, F. R.; Havercroft, J. M.; Ward, N. I. Lithium and boron in biological tissues and fluids. Trace Elements Med. 6:142–146; 1989. 18. Ludden, T. M.; Beal, S. L.; Sheiner, L. B. Comparison of the Akaike information criterion, the Schwartz criterion and the F test as guides to model selection. J. Pharmacokin. Biopharm. 22:431– 445; 1994. 19. Stragliotto, G. Biodistribution and pharmacokinetics of boron sulfhydryl for boron neutron capture therapy in patients with intracranial tumours. Thesis, Faculte´ de Me´decine de l Universite´ de Lausanne, 1993. 20. Sweet, W. H.; Messer, J. R.; Hatanaka, H. Supplementary pharmacological study between 1972 and 1977 on purified mercaptoundecahydrododecaborate. In: Hatanaka, H., ed. Boron neutron capture therapy for tumors. Niigata: Nishimura; 1986:59 –76. 21. Nakagawa, T.; Nagai, T. Interaction between serum albumin and mercaptoundecahydrododecaborate ion (an agent for boron neutron capture therapy of brain tumor). Chem. Pharm. Bull. Tokyo 24:2934 –2948; 1976. 22. Samsel, E. G.; Miller, D. L. High resolution 10B and 11 B nuclear magnetic resonance (NMR) spectroscopy of Na2B12H11SH impurities and metabolites. Strahlenther. Onkol. 165:140 –141; 1989. 23. Soloway, A. H.; Hatanaka, H.; Davis, M. A. Penetration of brain and brain tumor. VII. Tumor-binding sulfhydryl boron compounds. J. Med. Chem. 10:714 –717; 1967. 24. Zhu-Tang, P.-P. P.; Schweizer, M. P.; Bradshaw, K. M.;
638
I. J. Radiation Oncology
●
Biology
●
Physics
Bauer, W. F. 11B nuclear magnetic resonance studies of the interaction of borocaptate sodium with serum albumin. Biochem. Pharmacol. 49:625– 632; 1995. 25. Soloway, A. H.; Alam, F.; Barth, R. F.; Bapat, B. V. Boron chemistry and target cell affinity. Strahlenther. Onkol. 165: 118 –120; 1989. 26. Mehta, S. C.; Lu, D. R. Interspecies pharmacokinetic scaling of BSH in mice, rats, rabbits, and humans. Biopharm. Drug. Disp. 16:735–744; 1995. 27. Stragliotto, G.; Munafo, A.; Biollaz, J.; Fankhauser, H. Pharmacokinetics of boron sulfhydryl (BSH) in patients with in-
Volume 41, Number 3, 1998
tracranial tumours. In: Allen, B. J., et al., eds. Progress in neutron capture therapy for cancer. New York: Plenum Press; 1992:549 –550. 28. Asbury, A. K.; Ojeman, R. G.; Nielsen, S. L.; Sweet, W. H. Neuropathologic study of fourteen cases of malignant brain tumor treated by boron-10 slow neutron capture therapy. J. Neuropathol. Exp. Neurol. 31:278 –303; 1972. 29. Gabel, D. Minutes of the meeting of the Project Management Group Lausanne. In: Newsletter of the European Collaboration on Boron Neutron Capture Therapy of Tumors. 12:2–3; 1992.