Measurement of Cerebral Blood Flow, Cerebral Blood Volume, and Cerebral Capillary Permeability in Glioma-bearing Rats Shizuka
Department
AIZAWA,
Kazuhiro
of Neurosurgery,
SAKO
Asahikawa
and
Medical
Yukichi
YONEMASU
College, Asahikawa,
Hokkaido
Abstract Effective chemotherapy and radiation therapy of brain tumors require knowledge of the cerebral cir culatory dynamics involved. In this study, regional cerebral blood flow (rCBF), regional cerebral blood volume (rCBV), and regional cerebral capillary permeability (rCP) were measured in Wistar King Aptekman rats bearing experimental KEG-1 gliomas. These parameters were assessed by auto radiography with 14C-iodoantipyrine, 14C-deoxyglucose-labeled red blood cells, and 14C-aminoiso butyric acid, respectively. rCBF within the tumor was approximately one third that in the contra lateral cortex and was consistently higher in the periphery than in the center of the tumor. In the periphery of the tumor, rCBV was approximately twice that in the contralateral cortex, but it was very low in the center of the tumor. Throughout the tumor, rCP was sharply increased relative to that measured in the contralateral cortex, and the increase was especially pronounced in the central por tion. Thus, rCBF, rCBV, and rCP each appeared to vary within the tumor, implying that the com bined use of lipid and water-soluble chemotherapeutic agents is reasonable. Measurement of these parameters
may also provide
Key words: cerebral
brain
capillary
neoplasms,
indices of radiation cerebral
permeability,
cancer
blood
permeability within gliomas is therefore necessary if we are to solve these treatment problems and im prove the prognosis. In the study presented here, we measured region al cerebral blood flow (rCBF), regional cerebral blood volume (rCBV), and regional cerebral capil lary permeability (rCP) in a rat glioma model.
February
26,
1988;
cerebral radiation
blood
volume,
therapy
Materials
Because of the limitations in the surgical removal of malignant glioma, radiation therapy, chemotherapy, and immunotherapy are commonly administered as well. Nonetheless, the outcome is unsatisfactory in most cases. This has been explained in terms of poor tumor penetration by antineoplastic agents and insen sitivity to radiation due to the anoxia of gliomas. ") Assessment of the circulatory dynamics and capillary
23, 1989
flow,
chemotherapy,
Introduction
Received
sensitivity.
Accepted
February
and
KEG-1 cells were subcultured use of a stereotactic apparatus
Methods in vitro and, with the for brain surgery, in
jected at 5 x 104 cells/ 10 ,ul into the right caudate nucleus of 20 male Wistar King Aptekman rats, weighing about 200 gm. rCBF, rCBV, and rCP were measured in six, three, and four animals, respec tively. The remaining seven rats were used for obser vation of the natural survival course. Fourteen to 16 days after the implantation of tumor cells, polyethylene catheters were inserted into the femoral artery and vein under anesthesia induced by 1.0-1.5% halothane. The lower half of the body was fixed in a plaster cast and the animal was allowed to recover from anesthesia. Blood pressure and pH were measured and blood gases analyzed immedi ately prior to auto radiographic assessments. Body temperature was maintained at 37°C with a heat lamp. rCBF was measured by quantitative autoradiog raphy,29) with 14C-iodoantipyrine (14C-IAP) as the
tracer. According to a previously described proce dure, 27.28'35) 25,uCi of 14C-IAP was administered at a constant rate over 1 minute via the femoral vein, and blood was simultaneously drawn from the fem oral artery every 4-5 seconds to measure the plasma concentration of 14C-IAP. On completion of 14C IAP administration, the rats were decapitated and the brains promptly removed, frozen at -40 C in liquid freon. Serial sections 20,um in thickness were prepared in a -22°C cryostat (Model 855; American Optical Company, Buffalo, N.Y., U.S.A.) and placed on cover glasses. The specimens were fixed on a hot plate at 60°C and placed in an x-ray cassette with 14C-methyl methacrylate standards (Amersham Co., Bucklinghamshire, England). They were then tightly attached to single-coated x-ray film (Kodak SB-5; Rochester, N.Y., U.S.A.) and placed in the dark for 5 days. After development, the x-ray film was subjected to densitometry (Sakura Micro densitometer PDM-5; Tokyo, Japan). Radioactivity was calculated from the optical density thus obtain ed. rCBF was calculated with the formula devised by Sakurada et al.") The partition coefficient used was 0.8 for both normal and tumor tissues. rCP was measured by quantitative autoradiog raphy,5.36>with the use of 14C-aminoisobutyric acid (14C-AIB). Femoral arterial blood samples were ob tained at predetermined intervals over a 10-minute period after bolus injection of 50,uCi of 14C-AIB into the femoral vein. The blood samples were im mediately centrifuged and 20 pl of plasma was used to measure the 14C-AIB concentration. After the last blood sampling, the animals were immediately decapitated and the brains quickly removed. The remainder of the procedure was the same as that for 14C-IAP autoradiography. Red blood cells labeled with 14C-deoxyglucose (14C-DG) were used to assess rCBV." ) Autologous blood (1.0-1.5 ml) was washed with a phosphate buffer solution (PBS) containing heparin, and the cells were incubated for 30 minutes at 37'C with PBS to which 14C-DG had been added. PBS was re added, the mixture was centrifuged, and the super natant was discarded. This washing procedure was repeated at least once. The red blood cells thus labeled with 14C-DG were injected into the rat fem oral vein over a period of 1 minute. Ten minutes later, the rats were sacrificed by intravenous injec tion of a saturated KCl solution. The brain was removed during immersion of the skull in liquid freon at -40 C. rCBV was calculated as follows: rCBV = (Cb/Ca) X a Where Cb represents the radioactivity per 1 gm of brain tissue, Ca the radioactivity per l ul of arterial
blood, and a the ratio of the cerebral-to-large vessel hematocrit (in this study, the value of 0.83 obtained by Everett et al.") was used). All brain specimens were embedded in paraffin after formalin fixation and sections were stained with HE. Sections used for autoradiography were also stained and examined. Results
The
mean
merely The
survival
animals
culatory
was
1.3
for
data
tumor The
are
the
seven
2.8
(mean •}
the
rats
at
of
in
after
were
pH
the
Table
cir
days
time
and
were
begun.
was
and
auto
days.
cerebral
14-16
that
to
that SD)
experiments
prior
summarized
study alive
pressure
just
the
all
diameter
blood
obtained
studies
were
when
mean mm.
of
20.9 •}
employed parameters
implantation, The
period
observed
5.2 •@
blood
gas
radiographic
1.
The results of rCBF, rCBV, and rCP measure ments are given in Table 2. Representative auto radiograms are shown in Fig. 1 upper, HE-stained sections of the specimen used for autoradiography are presented in Fig. 1 middle, and the densitomet ric profiles of the autoradiograms in Fig. 1 lower. The mean cortical rCBF and rCBV were somewhat
Table
1
Physiological
Values
are
blood
pressure.
Table
2
Values
means
rCBF, normal
are
parameters
mea•@
SD.
rCBV, brain
SD.
MABP:
and rCP tissues
mean
in brain
arterial
tumor
and
Fig. I
upper: Representative autoradiograms obtained with 14C-IAP (left), 14C-DG-labeled red blood cells (center), and 14C-AIB (right) in coronal brain sections from tumor-bearing rats. middle: HE-stained histological sections used for autoradiography. lower: Densitometric profiles of autoradiograms obtained through the sites indicated by horizontal lines in the upper figure.
lower on the ipsilateral side than on the contra lateral side, whereas rCP was almost the same on both sides. Relative to both cortices, rCBF was markedly reduced within the tumor, somewhat more so at the center (0.42 ml/min/gm) than at the periphery (0.59 ml/min/gm). The difference in in tratumoral rCBV was more pronounced, being 11.5 ,ul/gm at the center and 53.9,u1/gm at the periphery. The latter value was about twice that in the con tralateral cortex. rCP was markedly augmented throughout the tumor, particularly in the central por tion of the tumor, where it was 10 times greater than that in the contralateral cortex. Both rCBF and rCBV in the white matter adjacent to the tumor were substantially reduced. On the other hand, rCP in the adjacent white matter was markedly increased (15.2 ml/min/kg). Discussion The brain established days) are
tumor model used in this study is an one, and our survival data (mean, 20.9 in agreement with those reported by
Kaneko et al.") One requirement of such a model is a constant rate of tumor growth. In this study, at 14 16 days after implantation, the tumor size averaged 5.2 mm, which suggests a relatively constant growth rate. The two types of experimental brain tumors are those induced by carcinogenic agents (primary tu mors) and those induced by implantation of tumor cells. For measurement of rCP, the former model, in which mechanical damage is not inflicted, is thought to be desirable. Those tumors, however, have the disadvantage of being inconsistent in terms of loca tion, size, and histological type. Therefore, we used a transplantation model. The mean diameter of the tumors produced indicates that the influence of mechanical injury at the time of implantation was negligible. There have been numerous reports on rCBF in experimental rat brain tumors .1,1,4,6,12,14,16,19,24,33) The rCBF values vary, probably because of differences in the types of tumors and their ages at the time of measurement. Whereas rCBF has been reported as relatively constant in primary experimental tu
mors,1,12) that in transplanted tumors is reportedly low in the central portion and higher in the periph ery of tumors that have reached a size of more than 2 mm2 in cross-section.',', 11,16,24) A similar ten dency was noted in the present study. In our study, the results of autoradiography in dicated that the central portion of the tumor was viable (non-necrotic), but rCBF was low and there were fewer new blood vessels than in the periphery. In other words, tumor growth was faster in this transplantation model than in primary brain tumor models, so that vascular development lagged behind tumor growth. This suggests a close relationship be tween the heterogeneity of rCBF within the tumor and the rate of tumor growth (i.e., the degree of malignancy). rCBF in the ipsilateral cortex was de creased by 19% relative to that of the contralateral cortex, but was reduced by 37% in the white matter adjacent to the tumor. This may have been because the tumor had a more pronounced effect on the ad jacent white matter through direct compression or edema. rCBV values paralleled those of rCBF. rCBF in the contralateral (parietal) cortex was 84% of the value (1.72 ml/min/gm) we obtained in non-tumor-bearing rats previously.27) This reduction in rCBF on the contralateral side may be explained in terms of the decrease in perfusion pressure con sequent to elevated intracranial pressure, and the impaired function on the ipsilateral side may even tually "spread" to the contralateral side via nerve fibers, constituting a fall of rCBF (diaschisis). How ever, this sequence of events is hypothetical at pre sent. rCP within the tumor was quite high, particularly in the central portion, which corroborates previous ly reported observations .1,3,7,13,15,22,23,33) According to Molnar et al.,23) the rate of transcapillary transfer of materials within the tumor becomes greater as the tumor enlarges, is low in necrotic or cystic regions, and is dependent on the properties of the blood vessels supplying the tumor. These authors added that the rate is generally higher in tumors fed by the blood vessels of the choroid plexus and meninges. In our study, the tumors were situated within the brain parenchyma and, with only minor variation, were consistent in size. Since the tumor tissue was heterogeneous, higher rCP was expected in the periphery of the tumor, where rCBF was abundant and blood vessels were dense. On the contrary, however, it tended to be higher in the central por tion, which suggests a difference in the permeability of blood vessels among various sites within the tumor. The results of this study may be relevant to the
selection of treatment in the clinical setting. With such nitrosourea compounds as ACNU, distribution appears to depend on rCBF, since this agent is lipid-soluble.',",") Therefore, in our tumor model, poor ACNU penetration would be expected in the central portion. Although the blood-brain barrier is an obstacle to water-soluble agents, the rCP is in creased throughout the tumor, so that even at its center, an effective concentration may be achieved with a combination of a nitrosourea compound and a water-soluble agent. This problem requires fur ther study, as the many reports concerning anti cancer drug distribution pertain to tumors as a whole",") and do not describe local uptake by specific regions within tumors. The relationship between radiation sensitivity and the oxygen concentration in tumor tissues has been known for many years.") A few years ago, Pal lavicini and Hill") reported a decrease in the radia tion sensitivity of a KHT sarcoma transplanted into mouse muscle in response to a reduction in in tratumoral rCBF induced by anesthetics and sever al vasoactive agents. Like our experimental glioma, most tumors are poorly perfused with blood and hypoxic.20'21) The effects of radiation therapy have been evaluated clinically under hyperbaric oxygen conditions') and in laboratory mice following the in crease of oxygen transport by means of administra tion of a perfluorochemical emulsion.") Oxygen transport is believed to occur through simple diffu sion along the concentration gradient between capil lary and brain tissue, and Tannock30) reported the arrival of oxygen up to 150,um from the capillary. While the rCBF and capillary density may not be identical, they probably change in parallel. There fore, estimates of the tissue oxygen concentration will probably be possible through measurement of rCBF and rCBV, and will serve as a useful index of radiation sensitivity. Acknowledgments The authors deeply appreciate the cooperation of Drs. Nozomi Suzuki and Shigeki Yura; the Tumor Study Group of the Department of Neurosurgery, Hokkaido University School of Medicine, for the kind donation of the tumor cells; and the technical assistance of Mr. Hironari Isobe in the preparation of the manuscript. The abstract was presented at the 45th Annual Meeting of the Japan Neurosurgical Society, in Tokyo, 1986.
References 1) Blasberg RG, Kobayashi T, Horowitz M, Rice JM, Groothuis D, Molnar P, Fenstermacher JD: Regional blood flow in ethylnitrosourea-induced brain tumors. Ann Neurol 14: 198-201, 1983 2) Blasberg RG, Kobayashi T, Horowitz M, Rice JM, Groothuis D, Molnar P, Fenstermacher JD: Regional blood-to-tissue transport in ethylnitrosourea-induced brain tumors. Ann Neurol 14: 202-215, 1983 3) Blasberg RG, Kobayashi T, Patlak CS, Shinohara M, Miyaoka M, Rice JM, Shapiro WR: Regional blood flow, capillary permeability, and glucose utilization in two brain tumor models: Preliminary observation and pharmacokinetic implantations. Cancer Treat Rep 65: 3-12, 1981 4) Blasberg RG, Molnar P, Horowitz M, Kornblith P, Pleasants G, Fenstermacher JD: Regional blood flow in RT-9 brain tumors. JNeurosurg 58: 868-873, 1983 5) Blasberg RG, Patlak CS, Jehle JW, Fenstermacher JD: An autoradiographic technique to measure the permeability of normal and abnormal brain capil laries. Neurology (Minneap) 28: 363, 1978 (abstract) 6) Blasberg RG, Shapiro WR, Molnar P, Patlak CS, Fenstermacher JD: Local blood flow in Walker 256 metastatic brain tumors. J Neurooncol 2: 195-204, 1984 7) Blasberg RG, Shapiro WR, Molnar P, Patlak CS, Fenstermacher JD: Local blood-to-tissue transport in Walker 256 metastatic brain tumors. J Neurooncol 2: 205-218, 1984 8) Chang CH: Hyperbaric oxygen and radiation therapy in the management of glioblastoma, in Evans AE (ed): Modern Concept in Brain Tumor Therapy. Beccles, England, Castle House, 1979, pp 155-161 9) Diksic M, Sako K, Feindel W, Farrokhzad S, Yamamoto YL: Pharmacokinetics of positron-label ed 1,3-bis(2-chloroethyl)nitrosourea (BCNU) in human brain tumors using positron emission tomography. Cancer Res 44: 3120-3124, 1984 10) Everett NB, Simmons B, Lasher EP: Distribution of blood (Fe-59) and plasma (1-131) values of rats deter mined by liquid nitrogen freezing. Circ Res 44: 419 426, 1956 11) Gray LH, Conger AD, Ebert M: The concentration of oxygen dissolved in tissue at the time of irradiation as a factor in radiotherapy. Brit J Radiol 26: 638, 1953 12) Groothuis DR, Blasberg RG, Molnar P, Bigner D, Fenstermacher JD: Regional blood flow in avian sar coma virus (ASV)-induced brain tumor. Neurology (Cleveland) 33: 686-696, 1983 13) Groothuis DR, Fischer JM, Pasternak JF, Blasberg RG, Vick NA, Bigner DD: Regional measurement of blood to tissue transport in experimental RD-2 rat gliomas. Cancer Res 43: 3368-3373, 1983 14) Groothuis DR, Pasternak JF, Fischer JM, Blasberg RG, Bigner DD, Vick NA: Regional measurement of
Res Mogami blood Tewson local ischemia. S185, Kobayashi W: Fenstermacher tumor. 809-811, manipulations Feindel short-lived autoradiography Oncol LCBF 25: tumor Regional 310-325, 43:198718) flow hematocrit Biol and model 304-310, Nippon 3362-3367, W: H: TJ: 198527) Kaneko 198316) 1983 tumors. Jin and Neurology LCGU. Phys 18F Changes S, autoradiography. Sako H: Fenstermacher Pharmacokinetics Hossman 25-32, Cereb glucose (Tokyo) An Abe 197822) experimental 198217) and K, KA, 20: (in Biskirk on glucose JD: Kato Niebuhr 997-1005, Igaku and Molnar H, Experimental Izumiyama JJ, 9: autoradiographic A, for JJapanese)25) Low P, in immunochemotherapy radiation Blood Diksic M, Neurooncol (Cleveland) Stroke 198315) 1321-1325, Aida Blasberg utilization Regional long-lived of Kogure M: Pallavicini experimental simultaneous the The MG, RG, Hasegawa Hoshasen 1980 consumption the I, Hill glioblastoma. K, metabolism RP: H, M, Tamura Flow T, Effect early RG-2 Ushio Groothuis 15: blood-brain Lockwood (in Yamamoto Cancer response. Y, Tsuru blood-to-tissue Regional 896-900, of Hayakawa and 14C study Japanese)19) 33: 198326) of Metab 3: cerebral rat 11C-BCNU phase Gakkai blood M: tumor Kato 271-283, Sako blood 702-711, measurement gliomas. in flow M, T, A, Res of K, Local Sako D, double AH, of in Int Diksic barrier 7the rats K, 198428) Yamada of Kodama LY: M, experimental blood Diksic by flow Bigner Zasshi [Suppl 27: Farrokhzad Jimmunochemotherapy M, Sako glioblastoma: with cerebral Yap 198520) Yamamoto Cancer aof 198323) S, K, Radiat Use Pathobiological 2212-2220, transport in tracer Kobatake in Yamamoto YL, brain combination Kirsch 196721) correlates. flow K, Molnar in Feindel in 43: YL, tumors. experimental Yamamoto D, K, E, 1]: T, of WM, Clin of metastatic P, RT-9 Experimental Schulz gliomas. Kirsch brain YL, brain avian PS-K Diksic brain D, brain Neurosurg Blasberg Jtumor. brain Leitner M: tumors WM, sarcoma tumors. Correlation JW: Cereb and JRG, tumors. JTucker studied The virus of JHorowit Neuroon Neurosu Blood effect ACNU WS, local (ASV) by of 3: 59: Neuro M, cereb Flow Neuro Tabu prolo 229dou brai bloo Sm 58: Me Me K, isc flo 8trgB Fu2
occlusion in the rat. Stroke 16: 828-834, 1985 Sakurada 0, Kennedy C, Jehle J, Brown JD, Carbin GL, Sokoloff L: Measurement of local cerebral blood flow with iodo(14C)antipyrine. Am JPhysiol 3: H59 H66, 1978 30) Tannock IF: The relation between cell proliferation and the vascular system in a transplanted mouse mam mary tumor. Brit J Cancer 22: 258-273, 1963 31) Teicher BA, Rose CM: Oxygen-carrying perfluoro chemical emulsion as an adjuvant to radiation t herapy in mice. Cancer Res 44: 4285-4288, 1984 32) Walker MD, Weiss HD: Chemotherapy in the treat ment of malignant brain tumors, in Friedlander WJ (ed): Advances in Neurology, vol 13. New York, Raven, 1975, pp 149-191 33) Yamada K, Hayakawa T, Ushio Y, Arita N, Kato A, Mogami H: Regional blood flow and capillary permeability in the ethylnitrosourea-induced rat glioma. J Neurosurg 55: 922-928, 1981 34) Yamamoto YL, Diksic M, Sako K, Arita N, Feindel 29)
Address reprint requests of Neurosurgery, 3-11
Nishikagura,
to: S. Aizawa, M.D., Department Asahikawa Medical College, 4-5 Asahikawa,
Hokkaido
078,
New Kojima flow. brain Japanese) W, Thompson studies (ed): Functional York, damage: edema. No inM: human Raven, The To No Relevance CJ: Kojima Shinkei effect Radionuclide malignant M: To The Pharmacokinetics 1983, effect Shinkei of hyperglycemia pp 38: toglioma, the 327-33535) 39: Imaging 1117-1125, Yura local S, 571-578, in K, Sako Magistretti cerebral and Aizawa ofon S,Suzuki the metabolic N, 1986 ischemic 1987 Yonemasu Brain. blood PL (in (in Y, Yura Japanes S,Sako K,A Japan.
Department
AIZAWA,
Kazuhiro
of Neurosurgery,
SAKO
Asahikawa
and
Medical
Yukichi
YONEMASU
College, Asahikawa,
Hokkaido
Abstract Effective chemotherapy and radiation therapy of brain tumors require knowledge of the cerebral cir culatory dynamics involved. In this study, regional cerebral blood flow (rCBF), regional cerebral blood volume (rCBV), and regional cerebral capillary permeability (rCP) were measured in Wistar King Aptekman rats bearing experimental KEG-1 gliomas. These parameters were assessed by auto radiography with 14C-iodoantipyrine, 14C-deoxyglucose-labeled red blood cells, and 14C-aminoiso butyric acid, respectively. rCBF within the tumor was approximately one third that in the contra lateral cortex and was consistently higher in the periphery than in the center of the tumor. In the periphery of the tumor, rCBV was approximately twice that in the contralateral cortex, but it was very low in the center of the tumor. Throughout the tumor, rCP was sharply increased relative to that measured in the contralateral cortex, and the increase was especially pronounced in the central por tion. Thus, rCBF, rCBV, and rCP each appeared to vary within the tumor, implying that the com bined use of lipid and water-soluble chemotherapeutic agents is reasonable. Measurement of these parameters
may also provide
Key words: cerebral
brain
capillary
neoplasms,
indices of radiation cerebral
permeability,
cancer
blood
permeability within gliomas is therefore necessary if we are to solve these treatment problems and im prove the prognosis. In the study presented here, we measured region al cerebral blood flow (rCBF), regional cerebral blood volume (rCBV), and regional cerebral capil lary permeability (rCP) in a rat glioma model.
February
26,
1988;
cerebral radiation
blood
volume,
therapy
Materials
Because of the limitations in the surgical removal of malignant glioma, radiation therapy, chemotherapy, and immunotherapy are commonly administered as well. Nonetheless, the outcome is unsatisfactory in most cases. This has been explained in terms of poor tumor penetration by antineoplastic agents and insen sitivity to radiation due to the anoxia of gliomas. ") Assessment of the circulatory dynamics and capillary
23, 1989
flow,
chemotherapy,
Introduction
Received
sensitivity.
Accepted
February
and
KEG-1 cells were subcultured use of a stereotactic apparatus
Methods in vitro and, with the for brain surgery, in
jected at 5 x 104 cells/ 10 ,ul into the right caudate nucleus of 20 male Wistar King Aptekman rats, weighing about 200 gm. rCBF, rCBV, and rCP were measured in six, three, and four animals, respec tively. The remaining seven rats were used for obser vation of the natural survival course. Fourteen to 16 days after the implantation of tumor cells, polyethylene catheters were inserted into the femoral artery and vein under anesthesia induced by 1.0-1.5% halothane. The lower half of the body was fixed in a plaster cast and the animal was allowed to recover from anesthesia. Blood pressure and pH were measured and blood gases analyzed immedi ately prior to auto radiographic assessments. Body temperature was maintained at 37°C with a heat lamp. rCBF was measured by quantitative autoradiog raphy,29) with 14C-iodoantipyrine (14C-IAP) as the
tracer. According to a previously described proce dure, 27.28'35) 25,uCi of 14C-IAP was administered at a constant rate over 1 minute via the femoral vein, and blood was simultaneously drawn from the fem oral artery every 4-5 seconds to measure the plasma concentration of 14C-IAP. On completion of 14C IAP administration, the rats were decapitated and the brains promptly removed, frozen at -40 C in liquid freon. Serial sections 20,um in thickness were prepared in a -22°C cryostat (Model 855; American Optical Company, Buffalo, N.Y., U.S.A.) and placed on cover glasses. The specimens were fixed on a hot plate at 60°C and placed in an x-ray cassette with 14C-methyl methacrylate standards (Amersham Co., Bucklinghamshire, England). They were then tightly attached to single-coated x-ray film (Kodak SB-5; Rochester, N.Y., U.S.A.) and placed in the dark for 5 days. After development, the x-ray film was subjected to densitometry (Sakura Micro densitometer PDM-5; Tokyo, Japan). Radioactivity was calculated from the optical density thus obtain ed. rCBF was calculated with the formula devised by Sakurada et al.") The partition coefficient used was 0.8 for both normal and tumor tissues. rCP was measured by quantitative autoradiog raphy,5.36>with the use of 14C-aminoisobutyric acid (14C-AIB). Femoral arterial blood samples were ob tained at predetermined intervals over a 10-minute period after bolus injection of 50,uCi of 14C-AIB into the femoral vein. The blood samples were im mediately centrifuged and 20 pl of plasma was used to measure the 14C-AIB concentration. After the last blood sampling, the animals were immediately decapitated and the brains quickly removed. The remainder of the procedure was the same as that for 14C-IAP autoradiography. Red blood cells labeled with 14C-deoxyglucose (14C-DG) were used to assess rCBV." ) Autologous blood (1.0-1.5 ml) was washed with a phosphate buffer solution (PBS) containing heparin, and the cells were incubated for 30 minutes at 37'C with PBS to which 14C-DG had been added. PBS was re added, the mixture was centrifuged, and the super natant was discarded. This washing procedure was repeated at least once. The red blood cells thus labeled with 14C-DG were injected into the rat fem oral vein over a period of 1 minute. Ten minutes later, the rats were sacrificed by intravenous injec tion of a saturated KCl solution. The brain was removed during immersion of the skull in liquid freon at -40 C. rCBV was calculated as follows: rCBV = (Cb/Ca) X a Where Cb represents the radioactivity per 1 gm of brain tissue, Ca the radioactivity per l ul of arterial
blood, and a the ratio of the cerebral-to-large vessel hematocrit (in this study, the value of 0.83 obtained by Everett et al.") was used). All brain specimens were embedded in paraffin after formalin fixation and sections were stained with HE. Sections used for autoradiography were also stained and examined. Results
The
mean
merely The
survival
animals
culatory
was
1.3
for
data
tumor The
are
the
seven
2.8
(mean •}
the
rats
at
of
in
after
were
pH
the
Table
cir
days
time
and
were
begun.
was
and
auto
days.
cerebral
14-16
that
to
that SD)
experiments
prior
summarized
study alive
pressure
just
the
all
diameter
blood
obtained
studies
were
when
mean mm.
of
20.9 •}
employed parameters
implantation, The
period
observed
5.2 •@
blood
gas
radiographic
1.
The results of rCBF, rCBV, and rCP measure ments are given in Table 2. Representative auto radiograms are shown in Fig. 1 upper, HE-stained sections of the specimen used for autoradiography are presented in Fig. 1 middle, and the densitomet ric profiles of the autoradiograms in Fig. 1 lower. The mean cortical rCBF and rCBV were somewhat
Table
1
Physiological
Values
are
blood
pressure.
Table
2
Values
means
rCBF, normal
are
parameters
mea•@
SD.
rCBV, brain
SD.
MABP:
and rCP tissues
mean
in brain
arterial
tumor
and
Fig. I
upper: Representative autoradiograms obtained with 14C-IAP (left), 14C-DG-labeled red blood cells (center), and 14C-AIB (right) in coronal brain sections from tumor-bearing rats. middle: HE-stained histological sections used for autoradiography. lower: Densitometric profiles of autoradiograms obtained through the sites indicated by horizontal lines in the upper figure.
lower on the ipsilateral side than on the contra lateral side, whereas rCP was almost the same on both sides. Relative to both cortices, rCBF was markedly reduced within the tumor, somewhat more so at the center (0.42 ml/min/gm) than at the periphery (0.59 ml/min/gm). The difference in in tratumoral rCBV was more pronounced, being 11.5 ,ul/gm at the center and 53.9,u1/gm at the periphery. The latter value was about twice that in the con tralateral cortex. rCP was markedly augmented throughout the tumor, particularly in the central por tion of the tumor, where it was 10 times greater than that in the contralateral cortex. Both rCBF and rCBV in the white matter adjacent to the tumor were substantially reduced. On the other hand, rCP in the adjacent white matter was markedly increased (15.2 ml/min/kg). Discussion The brain established days) are
tumor model used in this study is an one, and our survival data (mean, 20.9 in agreement with those reported by
Kaneko et al.") One requirement of such a model is a constant rate of tumor growth. In this study, at 14 16 days after implantation, the tumor size averaged 5.2 mm, which suggests a relatively constant growth rate. The two types of experimental brain tumors are those induced by carcinogenic agents (primary tu mors) and those induced by implantation of tumor cells. For measurement of rCP, the former model, in which mechanical damage is not inflicted, is thought to be desirable. Those tumors, however, have the disadvantage of being inconsistent in terms of loca tion, size, and histological type. Therefore, we used a transplantation model. The mean diameter of the tumors produced indicates that the influence of mechanical injury at the time of implantation was negligible. There have been numerous reports on rCBF in experimental rat brain tumors .1,1,4,6,12,14,16,19,24,33) The rCBF values vary, probably because of differences in the types of tumors and their ages at the time of measurement. Whereas rCBF has been reported as relatively constant in primary experimental tu
mors,1,12) that in transplanted tumors is reportedly low in the central portion and higher in the periph ery of tumors that have reached a size of more than 2 mm2 in cross-section.',', 11,16,24) A similar ten dency was noted in the present study. In our study, the results of autoradiography in dicated that the central portion of the tumor was viable (non-necrotic), but rCBF was low and there were fewer new blood vessels than in the periphery. In other words, tumor growth was faster in this transplantation model than in primary brain tumor models, so that vascular development lagged behind tumor growth. This suggests a close relationship be tween the heterogeneity of rCBF within the tumor and the rate of tumor growth (i.e., the degree of malignancy). rCBF in the ipsilateral cortex was de creased by 19% relative to that of the contralateral cortex, but was reduced by 37% in the white matter adjacent to the tumor. This may have been because the tumor had a more pronounced effect on the ad jacent white matter through direct compression or edema. rCBV values paralleled those of rCBF. rCBF in the contralateral (parietal) cortex was 84% of the value (1.72 ml/min/gm) we obtained in non-tumor-bearing rats previously.27) This reduction in rCBF on the contralateral side may be explained in terms of the decrease in perfusion pressure con sequent to elevated intracranial pressure, and the impaired function on the ipsilateral side may even tually "spread" to the contralateral side via nerve fibers, constituting a fall of rCBF (diaschisis). How ever, this sequence of events is hypothetical at pre sent. rCP within the tumor was quite high, particularly in the central portion, which corroborates previous ly reported observations .1,3,7,13,15,22,23,33) According to Molnar et al.,23) the rate of transcapillary transfer of materials within the tumor becomes greater as the tumor enlarges, is low in necrotic or cystic regions, and is dependent on the properties of the blood vessels supplying the tumor. These authors added that the rate is generally higher in tumors fed by the blood vessels of the choroid plexus and meninges. In our study, the tumors were situated within the brain parenchyma and, with only minor variation, were consistent in size. Since the tumor tissue was heterogeneous, higher rCP was expected in the periphery of the tumor, where rCBF was abundant and blood vessels were dense. On the contrary, however, it tended to be higher in the central por tion, which suggests a difference in the permeability of blood vessels among various sites within the tumor. The results of this study may be relevant to the
selection of treatment in the clinical setting. With such nitrosourea compounds as ACNU, distribution appears to depend on rCBF, since this agent is lipid-soluble.',",") Therefore, in our tumor model, poor ACNU penetration would be expected in the central portion. Although the blood-brain barrier is an obstacle to water-soluble agents, the rCP is in creased throughout the tumor, so that even at its center, an effective concentration may be achieved with a combination of a nitrosourea compound and a water-soluble agent. This problem requires fur ther study, as the many reports concerning anti cancer drug distribution pertain to tumors as a whole",") and do not describe local uptake by specific regions within tumors. The relationship between radiation sensitivity and the oxygen concentration in tumor tissues has been known for many years.") A few years ago, Pal lavicini and Hill") reported a decrease in the radia tion sensitivity of a KHT sarcoma transplanted into mouse muscle in response to a reduction in in tratumoral rCBF induced by anesthetics and sever al vasoactive agents. Like our experimental glioma, most tumors are poorly perfused with blood and hypoxic.20'21) The effects of radiation therapy have been evaluated clinically under hyperbaric oxygen conditions') and in laboratory mice following the in crease of oxygen transport by means of administra tion of a perfluorochemical emulsion.") Oxygen transport is believed to occur through simple diffu sion along the concentration gradient between capil lary and brain tissue, and Tannock30) reported the arrival of oxygen up to 150,um from the capillary. While the rCBF and capillary density may not be identical, they probably change in parallel. There fore, estimates of the tissue oxygen concentration will probably be possible through measurement of rCBF and rCBV, and will serve as a useful index of radiation sensitivity. Acknowledgments The authors deeply appreciate the cooperation of Drs. Nozomi Suzuki and Shigeki Yura; the Tumor Study Group of the Department of Neurosurgery, Hokkaido University School of Medicine, for the kind donation of the tumor cells; and the technical assistance of Mr. Hironari Isobe in the preparation of the manuscript. The abstract was presented at the 45th Annual Meeting of the Japan Neurosurgical Society, in Tokyo, 1986.
References 1) Blasberg RG, Kobayashi T, Horowitz M, Rice JM, Groothuis D, Molnar P, Fenstermacher JD: Regional blood flow in ethylnitrosourea-induced brain tumors. Ann Neurol 14: 198-201, 1983 2) Blasberg RG, Kobayashi T, Horowitz M, Rice JM, Groothuis D, Molnar P, Fenstermacher JD: Regional blood-to-tissue transport in ethylnitrosourea-induced brain tumors. Ann Neurol 14: 202-215, 1983 3) Blasberg RG, Kobayashi T, Patlak CS, Shinohara M, Miyaoka M, Rice JM, Shapiro WR: Regional blood flow, capillary permeability, and glucose utilization in two brain tumor models: Preliminary observation and pharmacokinetic implantations. Cancer Treat Rep 65: 3-12, 1981 4) Blasberg RG, Molnar P, Horowitz M, Kornblith P, Pleasants G, Fenstermacher JD: Regional blood flow in RT-9 brain tumors. JNeurosurg 58: 868-873, 1983 5) Blasberg RG, Patlak CS, Jehle JW, Fenstermacher JD: An autoradiographic technique to measure the permeability of normal and abnormal brain capil laries. Neurology (Minneap) 28: 363, 1978 (abstract) 6) Blasberg RG, Shapiro WR, Molnar P, Patlak CS, Fenstermacher JD: Local blood flow in Walker 256 metastatic brain tumors. J Neurooncol 2: 195-204, 1984 7) Blasberg RG, Shapiro WR, Molnar P, Patlak CS, Fenstermacher JD: Local blood-to-tissue transport in Walker 256 metastatic brain tumors. J Neurooncol 2: 205-218, 1984 8) Chang CH: Hyperbaric oxygen and radiation therapy in the management of glioblastoma, in Evans AE (ed): Modern Concept in Brain Tumor Therapy. Beccles, England, Castle House, 1979, pp 155-161 9) Diksic M, Sako K, Feindel W, Farrokhzad S, Yamamoto YL: Pharmacokinetics of positron-label ed 1,3-bis(2-chloroethyl)nitrosourea (BCNU) in human brain tumors using positron emission tomography. Cancer Res 44: 3120-3124, 1984 10) Everett NB, Simmons B, Lasher EP: Distribution of blood (Fe-59) and plasma (1-131) values of rats deter mined by liquid nitrogen freezing. Circ Res 44: 419 426, 1956 11) Gray LH, Conger AD, Ebert M: The concentration of oxygen dissolved in tissue at the time of irradiation as a factor in radiotherapy. Brit J Radiol 26: 638, 1953 12) Groothuis DR, Blasberg RG, Molnar P, Bigner D, Fenstermacher JD: Regional blood flow in avian sar coma virus (ASV)-induced brain tumor. Neurology (Cleveland) 33: 686-696, 1983 13) Groothuis DR, Fischer JM, Pasternak JF, Blasberg RG, Vick NA, Bigner DD: Regional measurement of blood to tissue transport in experimental RD-2 rat gliomas. Cancer Res 43: 3368-3373, 1983 14) Groothuis DR, Pasternak JF, Fischer JM, Blasberg RG, Bigner DD, Vick NA: Regional measurement of
Res Mogami blood Tewson local ischemia. S185, Kobayashi W: Fenstermacher tumor. 809-811, manipulations Feindel short-lived autoradiography Oncol LCBF 25: tumor Regional 310-325, 43:198718) flow hematocrit Biol and model 304-310, Nippon 3362-3367, W: H: TJ: 198527) Kaneko 198316) 1983 tumors. Jin and Neurology LCGU. Phys 18F Changes S, autoradiography. Sako H: Fenstermacher Pharmacokinetics Hossman 25-32, Cereb glucose (Tokyo) An Abe 197822) experimental 198217) and K, KA, 20: (in Biskirk on glucose JD: Kato Niebuhr 997-1005, Igaku and Molnar H, Experimental Izumiyama JJ, 9: autoradiographic A, for JJapanese)25) Low P, in immunochemotherapy radiation Blood Diksic M, Neurooncol (Cleveland) Stroke 198315) 1321-1325, Aida Blasberg utilization Regional long-lived of Kogure M: Pallavicini experimental simultaneous the The MG, RG, Hasegawa Hoshasen 1980 consumption the I, Hill glioblastoma. K, metabolism RP: H, M, Tamura Flow T, Effect early RG-2 Ushio Groothuis 15: blood-brain Lockwood (in Yamamoto Cancer response. Y, Tsuru blood-to-tissue Regional 896-900, of Hayakawa and 14C study Japanese)19) 33: 198326) of Metab 3: cerebral rat 11C-BCNU phase Gakkai blood M: tumor Kato 271-283, Sako blood 702-711, measurement gliomas. in flow M, T, A, Res of K, Local Sako D, double AH, of in Int Diksic barrier 7the rats K, 198428) Yamada of Kodama LY: M, experimental blood Diksic by flow Bigner Zasshi [Suppl 27: Farrokhzad Jimmunochemotherapy M, Sako glioblastoma: with cerebral Yap 198520) Yamamoto Cancer aof 198323) S, K, Radiat Use Pathobiological 2212-2220, transport in tracer Kobatake in Yamamoto YL, brain combination Kirsch 196721) correlates. flow K, Molnar in Feindel in 43: YL, tumors. experimental Yamamoto D, K, E, 1]: T, of WM, Clin of metastatic P, RT-9 Experimental Schulz gliomas. Kirsch brain YL, brain avian PS-K Diksic brain D, brain Neurosurg Blasberg Jtumor. brain Leitner M: tumors WM, sarcoma tumors. Correlation JW: Cereb and JRG, tumors. JTucker studied The virus of JHorowit Neuroon Neurosu Blood effect ACNU WS, local (ASV) by of 3: 59: Neuro M, cereb Flow Neuro Tabu prolo 229dou brai bloo Sm 58: Me Me K, isc flo 8trgB Fu2
occlusion in the rat. Stroke 16: 828-834, 1985 Sakurada 0, Kennedy C, Jehle J, Brown JD, Carbin GL, Sokoloff L: Measurement of local cerebral blood flow with iodo(14C)antipyrine. Am JPhysiol 3: H59 H66, 1978 30) Tannock IF: The relation between cell proliferation and the vascular system in a transplanted mouse mam mary tumor. Brit J Cancer 22: 258-273, 1963 31) Teicher BA, Rose CM: Oxygen-carrying perfluoro chemical emulsion as an adjuvant to radiation t herapy in mice. Cancer Res 44: 4285-4288, 1984 32) Walker MD, Weiss HD: Chemotherapy in the treat ment of malignant brain tumors, in Friedlander WJ (ed): Advances in Neurology, vol 13. New York, Raven, 1975, pp 149-191 33) Yamada K, Hayakawa T, Ushio Y, Arita N, Kato A, Mogami H: Regional blood flow and capillary permeability in the ethylnitrosourea-induced rat glioma. J Neurosurg 55: 922-928, 1981 34) Yamamoto YL, Diksic M, Sako K, Arita N, Feindel 29)
Address reprint requests of Neurosurgery, 3-11
Nishikagura,
to: S. Aizawa, M.D., Department Asahikawa Medical College, 4-5 Asahikawa,
Hokkaido
078,
New Kojima flow. brain Japanese) W, Thompson studies (ed): Functional York, damage: edema. No inM: human Raven, The To No Relevance CJ: Kojima Shinkei effect Radionuclide malignant M: To The Pharmacokinetics 1983, effect Shinkei of hyperglycemia pp 38: toglioma, the 327-33535) 39: Imaging 1117-1125, Yura local S, 571-578, in K, Sako Magistretti cerebral and Aizawa ofon S,Suzuki the metabolic N, 1986 ischemic 1987 Yonemasu Brain. blood PL (in (in Y, Yura Japanes S,Sako K,A Japan.
