ML20081K464
| ML20081K464 | |
| Person / Time | |
|---|---|
| Site: | MIT Nuclear Research Reactor |
| Issue date: | 04/16/1991 |
| From: | Bernard J MASSACHUSETTS INSTITUTE OF TECHNOLOGY, CAMBRIDGE |
| To: | Alexander Adams Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 9106270267 | |
| Download: ML20081K464 (39) | |
Text
g&d,,,o,,'.
n e3 ,c c:O:s NUCLEAR REACTOR LABORATORY i My.r
. , . 2 ..
AN INT E RDEPARTfAENTAL CEtdER Of dff,E fAASSACHUSETIS INSilTUT E OF TECHNOLOGY oA 'Afu m 138 A:bx v Sree ' Cnb**,e. f u ir w J A etnNAnD Ja Daenar fehtu NJ f f 17 0,3 73'M Oge-ro o' He t t " Ope' Ams Isors Na ;i? 1473 f.01 C AU 1g N3 :p t, 2$34211/4202 April 16,1991 hir. Alexander Adams, Project hianager Standardization and Non Power Reactor Project Directorate Division of Reactor Projects 111, IV, V, & Special Projects Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555
Dear hit. Adams,
Enclosed is a copy of Chapter Three from the Ph.D. thesis by Richard Choi on the Design and Experimental Evaluation of an Epithennal Neutron Beam at the hilt Research Reactor. This material covens the history of the early BNCT trials. I am currently compiling information on regulatory approvals from that era and will send you that infomiation as soon as I have it.
Sincerely,
<a n ohn A. Bernard, Ph. .
Director of Reactor Operations hilt Research Reactor JAB /gw 01o l 01 0 6 2 _7 0 2 6 7m . _415 u P FDR l
1
CHAPTER THREE APPLICABILITY OF BORON NEUTRON CAPTURE THERAPY
3.1 INTRODUCTION
The basic principles of boron neutron capture therapy (BNCT) have been around since Locher nrst described them in his discussion of possible medical uses of neutrons in 1936.1 Boron, having a high affinity (cross section) for thermal neutrons, will absorb a thermal neutron and fission into an alpha particle and a lithium ion releasing 2.78 MeV of energy. This is shown schematically in Ogure 3-1. About 93% of the time,0.48 MeV of this reaction energy is released as a single gamma ray of equivalent energy. The rest of the energy released in the reaction goes into the kinetic energy of the recoiling products, an alpha particle and a lithium nucleus which move in opposite directions to consene momentum. Because these are heavy, charged particles, they rapidly lose their energy through ionization in tissue or other material. It is these secondary ionization that causes intracellular damage in tissue. Gabel et al. have calculated the rate of energy loss to the tissue from the BNCT reaction, and the results are shown in figure 3 2.2 As shown in the figure, both particles dissipate most of their energies within 10 microns. Ten 43
~.
en:,
0 -
10B v
('_ / '
" B' (q i
'N )
mj 7 'I a panicle 0.48 MeV rray (93%)
Figure 31: Graphic schematic of the boron neutron capture reaction. Slow neutron is captured by the boron 10 nucleus forming an unstable boron ll nucleus which promptly fission into an a panicle and a lithium ion with 2.78 MeV of excess energy. Gamma ray with 0.48 MeV energy is also emitted 93% of the time.
44 Chapter three
}g e i e i i I I i
1 i
1 250. l -
1
'r -
- 7 Li !
E 200 - 1 4.lle -
[.a .
l
- 1 150.
}>,.
1 ,
w O
100- 8 -
4 s '
a -
l 1
50- l -
8 I "
1 1.
0 , , ; , , , i -
6 4 2 0 2 4 6 8 .10 i
Distance in tissue (pm)
I Figure 3 2: Rate of energy loss for the two panicles generated in the BNCT reaction 2 Applicability of boron neutron capture therapy 45
microns is appmximately the dinneter of most cells. Therefore,if the (n.a) reaction occurs within the cell boundaries, most of its energy will be dissipated within the cell. This is important because the dose is then restricted only to ' hose cells containing boron and will not affect neighboring healthy cells.
Boron neutron capture therapy is in essence a combination of chemotherapy and radiotherapy. Because it requires that both the boron and the neutron be present to kill the tumor, it can be, in principle, much safer and effective then either chemotherapy or radiotherapy alone. With chemotherapy, one telles on taassive concentrations of tumor-seeking drugs to kill the tumor. Unfortunately, most af these drugs also have adverse effects on the various non tumorous organs of the body such as liver or kidneys. This puts inherent restrictions on the amount of drug that can be tolerated, thus reducing the chances for a complete cure.
Radiotherapy on the other hand, can be effective on tumors provided that a tumor is localized and has not metastasized to other organs of the body. Once the tumor has spread throughout the body or the organ, the selective advantage of the radiotherapy is lost.
Because one can no longer focus the radiation to specific tumor sites, the radiation deliven the same amount of dose to healthy tissue as to tumor. Furthermore, the conventional radiotherapy typically uses x rays or gamma rays, these have ranges on the order of tens of centimeters in tissue. Therefore, they are far less specific than the ions resulting from the BNCF reaction, which lose all their energy within 10 pm of their origin.
With BNCT, one can be far more specific, because only the cells that have incorporated boron will be rJfected by the (n,a) reactions inducca by the neutron field. Its only when these two ingredients, boron 10 in tumor er,d neutron induced (n a) reactions are present, that the lethal effects are produced.
46 Chapter three
- - .- . . - - . - - . . _~ -
Ilowever, it wasn't until the 1950's that Locher's ideas could be used in therapy. Prior to this, there was no neutron source that was intense enough to be used as a therapeutic tool.
In 1950, the large 30 MW graphite moderated air-cooled reactor (BGRR) became available at Brookhaven National Laboratory (BNL). For the first time, neutron fluxes greater than 108 n/cm2 see was available for medical research. BNCT, with its unique potential, was initiated almost immediately at the BGRR by a group from BNL and the Massachusetts Genemi Hospital (MGil) and later at the Massachusetts institute of Technology Research Reactor (MITR) by a group from MIT and MGli. These experiences were uniformly dismal, yielding only 5.7 months of mean survival time after therapy.3 Most of tht wtients treated during this time showed recurrence of tumors and in some cases necrosis of the normal brain and scalp. Despite such failures,in 19M, a Japanese post-doctoral scientist, Dr.11iroshi Hatanaka, working at Massachusetts General Hospital became interested in BNCT. After surmounting various institutional difficulties, Hatanaka started treating his patients using BNCT with thermal neutron beams and improved boron delivery agents upon his return to Japan in 1968. As of March 1989, he had treated 98 patients using BNCT; 49 of which were affected with grade til and grade IV gliomas. In some cases, his therapy has achieved a 5 year survival rate of over 50% in this chapter, the clinical trials of BNCF by the BNUMGH group, the MIT/MGH group, and of Hatanaka, using thermal ,
neutron beams, will be briefly reviewed.
3.2 CLINICAL TRIALS AT BROOKHAVEN NATIONAL LABORATORY in 1951, based on a report by Conger and Giles at Oak Ridge of the efficacy of slow neutrons in producing a large amount of radiation damage to lily bulbs as a direct result of trace amounts of boron in them, Sweet and Javid at MGH decided to use this phenomenon to treat primary brain tumors.4 Because the normal brain is protected by the blood brain Applicability of boron neutron capture therapy 47
barrier (BBB), whereas in the tumor, the BBB was thought to be absent, they theorized that they could obtain a selective concentration of boron in the tumor relative to the normal brain. With the assistance of G. Brownell at MIT, they calculated that if they could put 50 1 pg of 10B per gm of tissue into a brain tumor with only 15 pg/g 10B in healthy tissue, the i tumor would receive 3 times as much radiation as the normal brain.
f
- They immediately started human distribution studies using salt solutions containing boron or borax (Na2B40r10H
- 0) to determine the ratio of the boron concentration in tumor relative to the normal brain at a number of time intervals after intmvenous injection. Their results, based on 61 patients, showed no conclusive pattem.5 The concentration of boron in both the tumor and the brain varied widely from patient to patient. However, they found that the best tumor-to-brain ratios were obtained from about 5 min to 30 min post injection with the radcuarying from 3:1 to 48:1. Ilowever, they also noted that the concentration
- of boron in the tumor was on the same order as that in the blood (tumor / blood 10B ~ 1).
3.2.1 First clinical series at BNL (1951-1953)
Sufficiently encouraged by the distribution studies,in Februuy 15,1951, Sweet along with Farr and their colleagues at both MGH and BNL attempted the first clinical trials of BNCT on a 51-year-old business woman at the newly completed 30 MWi Brookhaven graphite air-cooled Research Reactor. During the next two years, they irradhted 9 additional patients for a total of 21 neutron capture therapies (5 patients received more than 1 treatment). In all the patients, boron in the form of borax was given by intravenous injection (i.v.) immediately prior to irradiation. The dose of borax averaged 20 grams, which provided 19 to 46 mg of 10B per kg of body weight. '
The first 5 patients treated at the BNL reactor were given only one treatment. The next 5 received multiple treatments over varying time intervals. Figures 3-3 and 3-4 show the 48 Chapter three
I Shielding blocks removed temporarily for therapy
/
/ .
, Beam aperture
/
Concrete h shielding
- s. .. , 107 cm
@! Bismuth filter]ll XXXXX J
/
Cooling air Graphite reflector passages Reactor ( hite lus uranium)
Figure 3-3: Schematic diagram of the neutron therapy facility at the BGRR showing the beam line. The opening in the beam aperture is a rectangular hole 5 cm by 10 cm.
Applicability of boron neutron capture therapy 49 1
\'
Paraffin a u
\ ,
S Dense concrete Aluminum shielding shell 0.63 cm steel liner -
+
Figure 3-4: Detail view'of the BGRR neutron therapy beam aperture with the typical view of the patient position during irradiation t.
L l-i I
l l
l.
l L
l' l 50 Chapter three
beam line of the BGRR reactor as well as the detail of the beam port. The therapy room was actually a cavity on top of the reactor made by temporarily removing part of the reactor shielding blocks. The rectangular neutron beam port measured 5 cm by 10 cm and was located offset in the cavity but directly on top of the core. A patient during therapy would lie down on the Door with his/her head resting directly on the beam port.6 7 During the clinical trials at BNL, the patients were irradiated with their scalp and skull intact. The thennal neutron Dux at the beam aperture during typical therapy was 3 x 108 n/cm:sec.
Table 3-1 summarizes the first series of clinical trials at BNL in terms of patient's age, amount of boron used, Guence received, and number of days survival after therapy. Table 3 2 summarizes the pathological Ondings as well as best estimates of the peak background dose and the average tumor dose based on the available data. At the time of the therapy, the experimenters had no clear idea as to what the absorbed doses were to the tumor and the surrounding tissue because of the uncertainties in the level of contaminating radiations (such as gamma rays and fast neutrons) from the beam port and the relative RBEs. Only the total neutron Quence given to the patient during the therapy was measured.
In order to calculate the doses, the amounts of boron given through i.v. for each patient were correlated with the distribution studies made by Sweet to obtain the approximate boron concentration in tumor, normal brain, and in blood. His study showed that 5 g of borax given i.v. per 70 kg body weight resulted in approximately 15 ppm boron in the tumor 30 minutes post injection. The concentration in the blood was assumed to equal that in the tumor, and the concentration in the normal brain was assumed to be 1/3 that in tumor, or 5 ppm. Five grams of borax contains 0.57 g of boron. Therefore for every ppm of baron given through i.v.,1.84 ppm of bcron accumulates in the tumor and in the blood.
In normal brain, the concentration will be 1/3 that in tumor or 0.61 ppm per ppm boron injected i.v.
Applicability of boron neutron capture therapy 51 l
_ _ . _ ~ _ _ . . _ _. . ~ . _ _ _ _ _ _ _ _ _ _ _ _
4 Table 31: Summary of clinical trials at BNL from 1951 to 19536,7 Date - Date Amount Neutron. Days Patient Sex Age of of of " B Fluence Survival
- Surgerv NOT Given - Given Post NCT No.3977 F 51 Dec 1,50 Feb 15,51 1.59 g 80B 0.44 E+12 67 (PJ.) - (20 ppm) Mem peak -
No. 4055 M 38 Jan 27,51 Mar 15,5I 1.70 g 80B 1,42 E+ 12 93-(W.L.) (26 ppm) n/cm2peak No. 4045 M 60 Jan 6,51 Mar 22,51 1.47 g 80B _ l.46 E+ 12 43 (U.F.) (29 ppm) - n/cm2peak No. 4202 M $4 Feb 9,51 Jun 14,51 1.69 g 80B 1.42 E+12 141 (J.M.) n/cm2peak No,4227 M 47 Jan 13, $1 Jun 14, SI 1.69 g 10B l.42 E+12 100 (A.R.) (25 ppm) n/cm2peak No. 4653 M _58 Oct 26,51 Dec 13,51 _ 2.02 g 10B 1.47 E+12 186 (A.B.) (19 ppm) n/cm2 peak Jan 8,52 2.12g10B- 1.93 E+12 (19 ppm) n/cm2peak Mar 11,52 2.12g 30B 0.93 E+12 (19 ppm) n/cm2peak Apr 25,52 1.69 g 10B 0.98 E+12 (16 ppm) Wcm2 peak No. 4737 M 59 Jan 10,52 Feb 12,52 2.12g!?B 0.86 E+12 - 136 (T.F.) _( 27 ppm) n/cm2peak
! May 2,52 2.12g 10B 0.70 E+12
! (27 ppm) n/cm2peak No. 4709 M 54- Jan 11,52 Apr 8,52 2.12 g 80B 0.93 E+12 97 (P.P.) - (21 ppm)- n/cm2peak Apr 22,52 2.12 g 80B 0.70 E+12
' (21 ppm) n/cm2peak Jun 20,52 2.12 g 10B 0.72 E+12 (21 ppm)- ' n/cm2peak No. 5144 F- 33 May 8,52 Sep 12,52 2.12 g 40B -0 % E+12- 152
-(A.H.) (40 ppm) n/cm2peak Nov 7,52 2,12 g 10B 0.91 E+12 l' (40 ppm) n/cm2peak h Jan 9,53 2.12g 30B - 0.% E+12 i (40 ppm) _ n/cm2peak Feb 6,53 2.12g 10B - 0.90 E+12 (40 ppm) n/cm2peak
- No. 5227 F 50 Oct 4,52 Nov 7,52 2.12 g 80B - 0.91 E+12 67 (J.H.) - n/cm2peak .
Dec 5,52 2.12 g 10B - 0.82 E+ 12 n/cm2peak Jan 9,53 2.12 g l0B 0.80 E+12 n/cm2 peak 52 Chapter three
l Because the information on background doses for the BNL therapy beam was unavailable, I substituted the data from one of the thennal beams at MITR-il for the analysis. Ilowever, this will tend to reduce the estimate of dose to the healthy tissue since the BNL therapy beam probably had higher contaminants than the later generation MITR beam. Based on the measurements made on the bismuth filtered beam (M-010) at the MITR-il reactor, the peak background dose without boron (including all gamma and neutran dose) is assumed to be approximately 1.4 x 1010 RBE cGy/n. For boron capture reactions, a KERMA factor of 2.0 x 10-11 RBE cGy cm2/n/ ppm 10B was used. The macroscopic RBE factors used for these calculations were 1.0 for gamma rays,1.6 for neutrons, and 2.3 for the boron reaction. In comparison, Farr et al. used an RBE of 20 for the boron reaction, resulting in a higher estimate of the dose by a factor of ten.
Ds = Am (Kno + 0.33 CakaKa) = Scalp dose (3.la)
DT= %m (0.5 Kao + Cak Ka) B = Tumor dose (3-1b)
Da = %m (0.5 Kao + 0.33 CakaKa) = Brain dose (3 Ic)
De = Am (0.5 Kac + 0.33 Cak BKa) = Capillary dose (3-Id) where c = thermal neutron flux Kaa = background KERMA [=] 1.4 x 10-12 RBE Gy cm2/n Ka = boron KERMA [=] 2.0 x 10-13 RBE Gy em2/n ppm 10B Ca = concentration of baron given [=] gm toB/kg body weight ka = correction factor for boron concentration [=) 1.84 Using the above assumptions, scalp dose at the surface, tumor dose, brain dose, and olood capillary dose at 3 cm depth were calculated. For the calculation of the scalp dose, dose arising from the incident radiation field was added to the boron dose assuming 1/3 concentration of boron in the scalp as compared to the tumor. The thermal neutron flux at 3 cm depth was reported by Farr et al. as being 1/3 the surface thermal neutron flux. In calculating the dose to the endothelial cells of the capillaries, the effective RBE for the i
AppIleability of boron neutron capture therapy 53 l l
e .
Table 3-2: Summary of sathological results and estimates of absorbed dose to the scalp, anc to the tumor, normal brain, and blood capillaries at 3 cm depth for 8 patients treated at BNL from 1951 to 19536.7 Date Scalp Tumor Brain C a pil.
Patient Pathological Findings of dose dose dose dose NCT (RBE (RBE (RBE (RBE Gy) Gy) Gy) Gy) l No.3977 Large tumor mass present: 5.5 Feb 15, 51 1.7 1.4 0.7 0.7 (P.J.) x 2 cm cyst filled with necrode tissue.
No apparent radiadon damace.
! No.4055 Imge tumor in the left lobe. Mar 15,51 6.5 5.5 2.5 2.5 (W.L.) No apparent radiation damage to tumor or brain.
- No. 4045 Large tumor with necrotic Mar 22,51 7.2 6.2 2.7 2.7 (G.F.) tissue present.
No apmrent radiation damace.
No. 4227 Large tumor filled with cysts Jun 14,51 6.4 5.4 2.5 2.5 l (A.R.) found in the right hemisphere.
No apparent radiation damace. l No. 4653 Tumor with numerous necrotic Dec 13,51 5.5 4.5 2.2 2.2 1 (A.B.) sites found. Thickening of the Jan 8,52 7.2 5.9 2.9 2.9 vessels as wellas the presence Mar 11,52 3.5 2.8 1.4 1.4 l of giant tumor cells indicate Apr 25,52 3.3 2.6 1.3 1.3 l
radiation damage, Scalp also showed signs of Total 19.5 15.8 7.8 7.8 radiation damace.
No.4737 Tumor with necrotic foci were Feb 12,52 4.0 3.4 1.5 1.5 (T.F.) found. May 2,52 3.3 2.8 1.2 1.2
- No signs of radiation damage.
Total 7.3 6.2 2.7 2.7 No. 4709 Large tumor with numerous Apr 8,52 3.7 3.1 1.5 1.5 (P.P.) necrouc cites found. Giant cell Apr 22,52 2.8 2.3 1.1 1.1 formanon and hyper.chromauc Jun 20,52 2.9 2.4 1.1 1.1 nuclei suggest radiation effect.
Scalp shows radiation damage. Total 9.4 7.8 3.7 3.7 No. 5144 Scalp shows significant Sep 12,52 6.0 5.4 2.2 2.2 (A.H.) ruimnno damage; blood vessels Nov 7,52 5.7 5.1 2.1 2.1 are dilated and sweat glands Jan 9,53 6.0 5.4 2.2 2.2 showed damage. Brain showed Feb 6,53 5.6 5.1 2.1 2.1 scars, edema, and tumor with giant cells suggesting radiation Total 23.3 21.0 8.6 8.6 damage.
54 Chapter three
9 boron reaction was reduced to 1/3 the normal value. This is based on the Monte Carlo calculation by Deutsch and Murray which indicates that the bomn reaction which originates in the capillaries is only 1/3 as effective as those originating in the cells.8 What is immediately apparent from table 3-2 is that,in most cases, the dose delivered to the tumor is only a fraction of the 60 G3 usually given in conventional radiation therapy.
Appropriately, in most of these patients (except patients No. 4653,4709, and 5144), no sign of radiation damage could be found in either the tumor or the healthy tissue in the subsequent pathological examinations.
According to Farr et al. they noted temporary improvement in two of the patients who received single treatments. In patient No. 3977, there was noticeable improvement for approximately 6 weeks. Within 7 days of therapy, she was able to speak and relate ir.cidents that occurred before treatment. Patient No. 4055 showed improvement for about 4 weeks. During that time, he was able to stand, speak, and understr.d simple commands.
liis swelling in the cranial cavity was also under control during this time. Ilowever, all conditions became progressively worse after this period. In the other three patients, NCT did not seem to affect the progression of their disease.
In the series of patients that received only one treatment, there does not seem to be any correlation between the doses that they received and their outcome. Strangely, it was the patient that received the least dose (No. 3977), that showed the most improvement after therapy. Again the doses calculated in table 3-2 are only estimates. Because the boron in the form of borax seems to show large variations in its distribution in tumor as well as normal brain,it is possible that she received a much higher tumor dose and a much lower background dose than estimated in the table 3-2.
Applicability of boron neutron capture therapy 55
Among the 5 patients who received multiple treatments,3 (No. 4653,4737, and 5144) showed temporary improvements. Once again, it is difficult to draw any conclusions from the data. Patients No. 4653 and 5144 received the highest doses among those treated at BNL However, patient no. 4737 rueived only 7 Gy to his scalp and probably only 6 Gy to the tumor. Radiation damage was indicated in 3 of those receiving multiple treatments.
They were patients No. 4653,4709, and 5144, receiving 20 Gy,9 Gy, and 23 Gy to the scalp, respectively. Those receiving less dose than these 3 did not show any signs of radiation damage.
It is apparent from these data, that the first series of BNL clinical trials were deficient in three areas. The first is that not enough dose was given to treat even shallow tumors. This was the result of an over optimistic value of the RBE used for the boron reaction. With the RBE value of 20, the therapists were probably overestimating their dose by up to 500% or more. When they thought that they were giving 40 Gy, they were actually giving 6 Gy, using the current macroscopic RBE value of 2.3. Even for patients to whom they had given 20 Gy. this was done over a period of 4 to 5 months. In no instance did they give more than 8 Gy per treatment. In view of this,it would appear unlikely that they gave enough tumor dose to achieve tumor eradication in any of the patients.
Another apparent area in which these original trials were deficient was in neutron penetrability. Thermal neutron beams have inherently very low penetrability. Useful penetration using thermal neutrons for BNCI' can be shown to be only 3 to 4 cm The neutron beam at BNL was most likel) very thermal. In addition the therapists were handicapped by the fact that the beam port was located on the floor. They were unable to perform a craniotomy and reflection of the scalp, the skull, and the dura, which would have given them an additional I to 2 cm of beam penetration. Therefore, they had no chance of reach;ng the deep-seated tumor cells that may have caused the regrowth of the tumors i 56 Chapter three
elsewhere in the brain. In many of the pathological examinations of these patients, very large tumors, often encompassing the entire half of the brain, were found.
The third area in which they needed improvement was in finding a better boronated agent.
The borax which they were using relied only on the BBB to achieve a selective concentration in tumor. The tumor to normal brain ratio varied widely over individuals and therefore no reliable prediction could be made of the do' age to the various tissues in the brain. In addition, some damage was found in the blood vessels of the brain; no doubt due to the high concentration of boron that remained in the blood. Finally according to Farr et al, the borax in the amount used for the therapy appears to have toxic effects.6 The intake of borax was followed by retching and often by the evacuation of bladder and bowels.
Respiratory arrest was noted in one patient and poor color was observed in most of the patients following the injection.
During the two year's span from February of 1951 to February of 1953, the group at BNL.
along with the group from MGH, used BNCT to treat 10 patients at the 30 MW, Brookhaven graphite air-cooled reactor. Although all of the patients died shortly after therapy, many important lessons had been teamed. 'Ihe possibility of using NCT to treat malignant brain tumors was shown. About half the patients showed remarkable, albeit brief, recovery after the therapy. It was shown that no serious complications occurred directly from the therapy. Perhaps most importantly, this pioneering work laid the ground work for further BNCT trials of malignant CNS tumors which cannot be successfully treated with conventional modalities.
3.3 CLINICAL TRIALS AT THE MASSACHUSETTS INSTITUTE OF TECHNOLOGY RESEARCH REACTOR Ten years following the initiation of NCT at BNL, Sweet and his group from MGH along I
1 Applicability of boron neutron capture therapy 57
e with the group at MIT led by Bmwnell conducted the fifth major clinical trial of BNC at the Massachusetts Institute of Technology Res: arch Reactor. The MITR 1, unlike the BGRR that was used for the clinical trials at BNL, was heavy water moderated and cooled with graphite reDectors outside the core tank. Running at 2 MW power, MITR-1 could provide a thermal neutron flux of > 1010 n/cm2 sec at its medical beam port. This was almost 30 times the available flux of the BGRR, running at 10 times the power level. In addition, the MITR I was built with BNCr in mind and had a complete operating room built into the medical therapy facility. It is essentially identical to the current MITR Il except for the reactor core and the surrounding reflector. Figures 3-5 and 3-6 show the isometric cutaway of the MITR I and the neutron therapy beam line, ,
In the ten years following the first clinical trial, a number ofimprovements have been made to BNCT. The first was the availability of a reactor facility with a neutron flux high enough for NCT. It was found from the BNL trial, that the tumor-to-brain ratio of bomn falls rapidly with time. Therefore, a facility that can deliver a therapeutic dose in ten minutes or less was sought.9 i
The second major improvement was in the boronated agent. In the first and second series of patients treated at BNL, borax was given intravenously before therapy. Because of the high concentration of boron in the skin, major necrosis occurred in the epithelial layer exposed to neutrons. Farr et al. discovered that this could be controlled if the irradiation time was kept below 10 min so that borax would not have enough time to concentrate in the epithelial layer of the skin. However, the toxicity of the borax did not allow such short time with the available BGRR facility (with the available neutmn flux, ~300 ppm of 10B l would need to be administered to allow total therapy time of 10 minutes). 'Iherefore, for the third series, sodium pentaborate (Na2 B io o t6) was used in solution with glucose l
delivered by internal carotid infusion. This allowed an increase in the concentration of 10B l by up to 135% with no toxic effect. However even this agent did not provide the high l
58 Chapter three
/"
/
\ 'YAi. g ,g * .,,,
Dense # 3 O i M iP ': ,
concrete # Experimental 7 y pE. 7,-;"
ijl
.h ::y i in g'ffg Fuel e 4 . ,
i l-gl, Mi - :- - sn y.
Graphite ,l lp
,, ,g,s . ,,
Fuel N,[ !
- I $ ,7 t
- elements - i i .;$]
Experimental -I
. f?,.;
P l cunn kam
- ,.c I Hermal
&i, l" q. 9 ' ' . ** huupy ,,{ ,,; column
- ' j:. door f" .
c . , o/ 4 ' ' j/ % .
f y'
f
'Y' \acilit r
Observation Heavy wate)p window \p /
Medical therapy room Wa:et shutter Figure 3-5: Iso-metric cut-away view of the MIT Research Reactor showing major components and experimental facilities. The reactor is MITR I and has a different beam line then the current upgraded MITR II.
Applicability of boron neutron capture therkpy 59
l I
I l
i 1l' . . -
- h* * '
E l
=
l Reactor i
( ,q ps sm Core p es s _
4: w s g-l it'
,y 3 6
p
.. ..:- - _B_ismu_t_h_ m
. . ,.**d,;' .
< t _
R*w
'I r.m l 'l~ Q. .1.~; .=-; tf.3 J s ,
l
.i , :? .% .- r E
.,',f.:..'
r !.w.
. , . I...,. . -
l f*. , . .
.4.:'\l,.$
'--1 6 ., ,. > c. .m. s s s s s s. s s ,
- 0: : . ;. . w. ,
-1 11.
i- -
\
_ i ..
. A j .
?.
p.y
..g
- . n.
.(#.?, .p. . I: . . ..ti;;W:. s:. J'@ Light
,. r- r:4 I M P
. . w.c.,.
... .." = ,. .:
, . . 5*9.
g .
. . ., . water . ' '.S.W.
j
. :- ...;;..,! 4<; t . . . . : . . shutter
.w: .. .. ..s.. .n ,&, ; . \ \.
- .
- ,..:. .u+ .e
. w. ;m .s,.a:==..
, ?.
y
, y. . , , .
[ ,
. ..cr. . .. ., .. . ,:p - , .,.
. . :. ' . . . . . . ::? ' .
Y:b.' f,.l.'hh. If.y -,.* ..
- b.fl ___., y .
- , ,.e....__-- -
- __- _ _ _ e. o 1
Bismuth I collimator l
l 1
Operating couch I
\\ fr k I I
)
I I I I
- . Figure 3-6: The cross sectional view of the medical therapy beam at
! MITR-I used for human clinical trials from 1960 to 1962.
I i
60 Chapter threo
m ... .. . .
i tumor / brain ratio and low toxicity that was needed. Soloway et al. at htGli had been conducting animal and human distribution studies on a boronated compound that could be used for therapy. The compound that showed the most promise for NCT was paracarboxybenzeneboronic acid. It provided tumor / brain ratio of 5 to 7,30 minutes after injection, and maintained this high ratio for about one hour, after which the ratio fell to 4 for the next two hours.
The third major improvement in procedure was the ability to renect the scalp, the skull, and the dura, and expose the tumor directly to the neutron beam. nis was possible because the aperture for the hilTR beam was located in the ceiling of a complete aseptic operating room. De surgeon was able to perfomi the necessary craniotomy safely before the patient would be raised into position for the radiation therapy. Then through the viewing window, he and an anesthesiologist could monitor the patient and stop the irradiation procedure if needed.
3.3.1 First clinical series at MITR (1960-1961)
Between November of 1960 and August of 1961,16 patients were treated at htITR with BNCT. Because it was now possible to perform craniotomies before irradiation, the procedures for NCT differed greatly from those used at BNL De Rougemont and Soloway demonstrated that the BBB recovers from cerebral lobectomy within 3 weeks.11 Based on this data, NCT was administered 2 to 3 weeks following resection of the tumor. After the original surgery, the wound was closed and allowed to heal. Immediately prior to neutron therapy, the wound was reopened and renection of the scalp, the skull, and the dura was performed to expose the tumor directly to the neutron source. This was done in the operating room built into the hilTR. Because themial neutrons are attenuated rapidly by fluid, all cerebrospinal fluid was removed from the Applicability of boron neutron capture therapy 61 {
l tumor site and replaced with an air balloon. During the course of the therapy, suction was applied to the tumor region to prevent further accumulation of fluid.
Once the craniotomy was complete, boron was given in the form of a solution of p-carboxybenzeneboronic acid through i.v. The concentration administered ranged from 15 to 30 mg 10B per kg body weight. The concentration factor for boron in tumor was 1.87.
He concentration factor in normal brain was about 1/4 this ve'ue, or 0.43. His meant that 30 ppm of boron injected resulted in about 56 ppm in tumor and about 13 ppm in brain.
The infusion of boron took anywhere from 10 min to over an hour depending cn desired concentration and patient tolerance.
Following the administration of the boron, the patient was elevated to the beam aperture by a hydraulic lift built into the floor. Once the patient was secured, everyone left the room and the built-in shutters were opened, allowing an intense beam of thermal neutrons to irradiate the open brain. The patients were irradiated for 45 min to 90 min for a total neutron fluence of 5 x 1012 to 2 x 1013 n/cm2 Table 3-3 sununarizes the first MITR clinical trial in terms of patient's age, amount of boron, fluence given, and survival, much like table 3-1 for the BNL series. One can see that there is no apparent increase in the survival time for the patients treated at MIT vs.
those treated at BNL The average survival time after therapy is little less than 6 months.
Table 3-4 summarizes the pathological results and my estimates of the dose delivered to the tumor, normal brain and the endothelial cells of the blood capillaries. The scalp dose is not included, since with craniotomy the scalp dose is inconsequential. This time, all doses were calculated at 3 cm below the cut surface of the brain. The concentrations of boron in the tumor and the normal brain were obtained from the concentration factors described previously. The concentration of boron in the blood, based on pathological samples of I
62 Chapter three '
o l
Table 3-3: Summary of clinical trials at MIT from 1960 to 1961 Date Date Amount Neutron Days Patient Sex Age of of of " B Fluence Surs is al Surgerv NCT Given Given Post NOT No. I F 39 Jun 20,60 Nov 15,60 0.91 g '"Il 1.50 E+ 13 $U w ks (S.E.) (20 ppm) n/cm2 No.2 F 43 Oct 6,60 Not (G.R.) Irrat No.3 M 14 Aug 12,60 Nov 30,60 1.14 g 80B l.00 E+ 13 6 wks (G.S.) (20 ppm) n/cm2 No. 4 F 20 Dec 17,60 Dec 30,60 1.06 g '0B 2.00 E+ 13 2I wks (P.D.) (21 ppm) Wem2 No.5 F 51 Dec 7,60 Jan 6,61 0.69 g OB 1.35 E+13 8 wLs (ll.C.) (15 ppm) n/cm2 No.6 M 35 Dec 15,60 Jan 18,61 1.59 g IVB 1.10 E+13 26 wis 01.G.) (25 ppm) n/cm2 No. 7 M 43 Dec 12,60 Jan 25,61 1.68 g AUB 1.95 E+13 26 wLs (J.D.) (30 ppm) n/cm2 No.8 F 59 Dec 23,60 Jan 31,61 1.92 g toB 2.60 E+13 21 wks (F.L.) (30 ppm) n/cm2 No.9 M 14 Jan 28,61 Feb 15,61 1.35 g 20B 1.30 E+13 39 wLs (S.W.) (30 ppm) n/cm2 No.10 F $5 Feb 23,61 Irr. 0.17 g 50B (C.F.) Cancelled (3 ppm)
No.I1 F 47 Feb 27,61 Mar 21,61 3.30 E+13 26 wks (N.L) (31 ppm) n/cm2 No.12 F 55 Feb 23,61 Apr 12,61 1.74 g 10B 1.65 E+13 26 wks (C.F.) (31 ppm) n/cm2 No.13 M 64 Mar 27,61 Apr 18,61 3.30 E+13 2I w Ls (NJ.) (26 ppm) n/cm2 No.14 F 29 Apr 5,61 May 3,61 1.23 g lob 2,10 E+13 see No.17 (P.P.) (30 ppm) n/cm2 No.15 M $7 Apr 61 May 16,61 2.33 g 10B 1.80 E+13 35 wLs (S.W.) (30 ppm) n/cm2 No.16 M 34 Jun 9,61 Jun 30,61 1.90 E+13 26 wks (G.G.) (30 ppm) n/cm2
~
No.17 F 29 Jun 22,61 Jul 19,61 1.23 g $0B 2.00 E+13 30 w ks (P.P.) (30 ppm) n/cm2 No.18 M 45 Jul 28,61 Aug 4,61 2.09 g 10B 2.80 E+13 13 wks (M.3.) (30 ppm) n/cm2 No.19 M 63 Aug1,61 Aug 18,61 1.70 g 10B 2.35 E+13 35 wks (M.H.) (30 ppm) n/cm2 Applicability of boron neutron capture therapy 63
tumor collected during operations, was found to be, on average, twice that in tumor and therefore, the dose to the capillaries was determined to be 2/3 that in the tumor.10 What is immediately apparent is that the doses given to the patients were enormous. Often the doses delivered ~to the normal brain were over 30 Gy and this was 3 cm below the surface. At the surface, the doses to the healthy tissue could have been 50 Gy or more.
Likewise, the doses to the capillary wall were equally high. It is then no wonder that significant radiation damage was noted throughout the brain and the blood vessels.
Although it is true that the tumors near the surface also received very high doses, because the neutrons attenuate very rapidly with depth, tumor cells at 6 or 7 cm depth only received a minor fraction of the dose near the surface. Therefore, for most of the patients, they suffered signincant radiation damage, and at the same time, suffered recurrence of the tumors.
In two of the cases (G.S. and H.G.), the patients developed infections from the procedure, pointing out the risks of open craniotomy during therapy. In 5 of the 16 patients (S.E.,
J.D., S.W., G.G., and M.H.), there were some improvements following the therapy. The rest showed no improvement following therapy. There doesn't seem to be a correlation between dose and patient outcome. If we divide the 16 patients into two groups with one receiving 20 Gy or more to the brain at 3 cm and the other receiving less than 20 cGy and compare the survival rates, we find an average of 25 weeks' survival for those receiving a higher dose, and 26 weeks for those receiving a lower dose.
The pathological data from Asbury et al. on 14 patients irradiated at MITR indicates that all of the patients died from cerebral related causes. These included severe radiation necrosis of the brain, regrowth of tumors, massive intracranial hemorrhage, and acute bacterial meningitis. But even here, the amount of necrosis, regrowth of tumors, and vessel changes does not conelate well with the calculated doses. For example, no radiation 64 Chapter three
4 Table 3-1: Summary of pathological results and estimates of absorbed dose to the tumor, normal brain, and blood capillaries at 3 cm below the cut surface for 16 patients treated at Mir from 1960 to 196110 Patient Pathological Findings Date of Tumor Brain Capil.
NCT dose dose dose (RBE (RRE (RHE Gy) Gy) G3)
No.1 Tumor recurrence in area opposite to Nov 15,60 47.9 19.1 35.4 (S.E.) trradiation. Extensive radiadon necrosis of brain and vessels.
No. 3 Massive intracranial hemorrhage. No Nov 30,60 31.9 12.7 23.6 (G.S.) residual tumor or neemsis found.
No.4 Extensive tumor nodules on dura and surface Dec 30,60 66.4 26.0 48.9 (P. D-) of hemisphere.
No.5 Extensive recurrence of tumor throughout Jan 6,61 34.7 15.3 26.3 (H.C.) the brain. Mild vascular damage.
No.6. No residual tumor found. Extensive Jan 18,61 42.0 15.6 30.6 (H G.) radiation necrosis and herniadon of brain.
No.7. No data available. Jan 25,61 86.6 30.4 62.3 (J.D.)
No.8 Tumor recunence in the fornix and adjacent Jan 31,61 115.4 40.6 83.0 (F.L.) medial thalamus. Extensive radiadon necrosis of brain.
l~ No.9 Several nests of tumor near resected site. Feb 15,61 57.7 20.3 41.5
(: (S.W.) Damage of cervical cord and medulla related
!' to radiadon.
No.1I Severalislands of tumor, Extensive Mar 21,61 150.6, $?,4 108.1 (N.lJ radiation necmsis of brain.
No.12 No data available. Apr 12,61 75.3 26.2 54.1 (C.F.)
No.13 Tumor recurrence found in area opposite to Apr 18,61 130.1 47,7 94.4 (N.J.) trradiated zone. Extensive radiation necrosis m irr. side.
No.14 No data avaitable. Msy 3,61 93.2 32.8 67.1 (P.P.) Jul 19,61 88.8 31.2 63.9 Total 182.0 64.0 131.0 No.15- Tumor recurrence near resecdon area. May 16,61 79.9- 28.1 57.5 l- (S.W.) Extensive radiadon necrosis of brain.
No.16 Extensive tumor infiltration and radiation - Jun 30,61 84.4 29.6 60.7 (G.G.) necrosis.
No.IS Tumor recurrence near resection area. Acute Aug 4,61 124.3 43.7 89.4 (M.J ) hacterial meningids.
No.19 No data available. Aug 18,61 104.3 36.7 75.0 .
(M.H.)
Applicability of boron neutron capture therapy 65
.-.- - - _.~...- -- - -
d damage was noted in patient No.18 (M.J.), even though, he received one of the highest doses during the trial. On the other hand, patient No.1 (S.E.), who received approximately 1/3 the dose that patient No.18 received, was found to have severe radiation necrosis of the brain and vessels.
3.3.2' Second clinical series at MITR (1961-1962)
The next series of patients irradiated at MITR differed from the first series in two ways.
First, they were injected with a. new boron agent. This compound, sodium perhydrodecaborate (Na2B ioH io) was tested extensively by Soloway and Sweet and found to be superior to p-carboxyphenylboronic acid in several ways:ll 1) this compound had a-far greater percentage of boron per weight: 2) it was also less toxic then the previous compound (doses of 50 mg per kg of body weight could be given without deleterious effect); 3)it maintained its tumor-to-brain ratio far better than the p-carboxy-compound.
Whereas in the previous compound, the T/B ratio fell to 4 after about an hour, the new compound maintained a ratio of 7:1 for over 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> in mouse gliomas.
The second improvement was that now the reactor could be operated at twice the power level' This reduced the irradiation time by half and thus afforded the physician the
_ possibility to funher optimize the time between the injection and the irradiation.
~ In the second series, two patients were treated with these new improvements. Table 3 5 summarizes the treatment conditions and the length of survival after NCT, For these patients, only the concentration of boron given was available.- Although it is difficult to interpret trends from just two data points, the results are worse than in series one. Both patients showed no improvements following therapy and died soon thereafter.
66 Chapter three
i l .'.
f rm f this compound. A team led by H. Wat-Table 3-5: Sun, compound enriched with boron 10. Early distribu I c mp und showed tumor to-normal brain rat Patient sex Age 8"' importantly, the new compound also showed incret No. 20 F 48 Ju (W.C.) blood approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after injection, with No. 21 F 31 3 wk (H.S .) to if delivered by common carotid artery and as hi cantid artery. Independent confirmation of this i his human distribution trials, found no increased tx Table 3-6 shows the summary blood.18 However,in his trials, Sweet had adm absorbed dose to the tumor, the i debulking. Hatanaka administers the drug only table 3-2. There were two differ llatanaka theorizes "that a tumor excision decrease boron concentration factors havt respectively. This was based on the tortured blood vessels in the area surrou-circulation, and eventually enables a good trans Sweet et al. and reDects the incre cells " To date, his theory has not been proved.
The second difference is the tumc Hatanaka had also conducted experiments with c:
Both patients died of severe ceret verify that high boron concentration in blood was the first patient, there was regrow to the normal brain found in the patients treated a near the surface for normal brain, MITR-1 after the administration of perhydrod demonstrated that an acute cerebral capillary i Table 3-6: Summary of p patients could be duplicated. However, th. . .is inj the tumor, non surface for 2 p. were given large doses of an adrenocortical stero electmn micrographs of the capillary in these ca Patient Pathological NCT.13 There is no proven explanation as to w!
not provide any reasons.
No. 20 Evidence of recurrent t (W.C.) tissue. Radiation damag the surface of the brain.
After it was discovered that the patients at MIT n No. 21 Dicd of severe cerebral ede (H .S .)
Hatanaka, with the advice of N. Watanabe, a Ap Applicability of
fluence to give a maximum of 20 Gy to the 3.4 TEIKYO UNIVERSITY HUMAN consensus on the maximum dose that a health:
According to Hatanaka's calculations, this Upon his return to Japan, Hatanaka organized a te fluence of 1013to the surface of the brain foi boron neutron capture therapy based on various i
(~30 to 80 ppm administered per body weighi MIT clinical trials. These can be divided into 3 m E such as Na2 B i2 H SH, which was found to have s g On August 20,1968, Hatanaka treated his fi ii previous compounds; 2) large doses of adrenocorti patient's family. She was a 29 year old f:ma radiation damage to the blood vessels in the brain; a and a 50 Gy of telecobalt therapy with little Gy or less.
E irradiation, she weighed 30 kg, and had a fist.
E-head. He was author"ed to perform the t) During the pathological studies of the patients treat Japanese Atomic Energy Commission's speci patients had suffered acute radiation damage to the and physicians. Although she survived a thought to be caused by a high concentration of bor remarkable improvement, including gaining : time of irradiation. Indeed, the boron concentration i as that in the tumor. Therefore, a new boron compo 3.4.1 Hatanaka's revised therapi a higher tumor-to-blood concentration ratio.
The patient is operated on approximately on Follow ng the clinical trials, Sweet and Soloway at M excised completely whenever possible in ord development. One compound which showed some p neutron beam. The wound is then closed hydrododecaborate, Na2B i2HiiSH. However this l Soloway, Hatanaka claims that the BBB ca unstable and would oxidize in the presence of water steroids.
addition, it was difficult to determine the purity or level On the day before neutron irradiation, the I
~ " " * "
boron-10 per kg of body weight enriched N Hatanaka at this time was working as a post-doctoral sc common carotid anery through a motorize of this compound on various animals. He was su mg/kg body weight prednisolone succinate assistance from Shionogi Research Laboratory in Osa!-
injection, the patient is once again opera 68 Chapter three 70 Chapter three
l ..
Table 3-5: Summary of clinical trials at MIT (1961-1962)
Date Date Amount Neutron Days Patient Sex Age of of of HB Fluence S ursis al Su ritery NCT Given Given Post NCT No. 20 F 48 July 61 Aug 22,61 30 ppm 2.50 E+13 11 wks (W.C.)
No. 21 F 31 3 wks pnor N/A 30 ppm -2.50 E+ 13 2 wis (H.S.) to in.
Table 3-6 shows the summary of the pathological findings and my estimates for the absorbed dose to the tumor, the normal brain, and to the capillaries in a manner similar to table 3-2. There were two differences fmm the first series. The first difference is that the boron concentration factors have been changed to 2.92 and 0.47 for the tumor and brain respectively. This was based on clinical distribution studies made on human subjects by Sweet et al. and reflects the increased tumor to brain ratio obtained by the new compound.
The second difference is the tumor to blood ratio have been changed to 0.79.12 Both patients died of severe cerebral edema and were worse after treatment than before. In the first patient, there was regrowth of tumors deep in the brain as well as radiation damage near the surface for normal brain.
Table 3-6: Summary of pathological results and estimates of absorbed dose to the tumor, normal brain, and blood capillaries at 3 cm below the cut surface for 2 patients treated at MIT from 1961 to 196212 Patient Pathological Findings Date of Tumor Brain C a p il.
NCT dose dose dose (RBE (RBE (RBE Gy) Gy) Gy)
No. 20 Evidence of recurrent tumor in the deep Aug 22. 61 82.9 27.4 38.8 (W.C.) tissue. Radiation damage to the scalp and the surface of the brain.
No.21 Died of severe cerebral edema. N/A N/A N/A N/A (H.S.)
Applicability of boron neutron capture therapy 67
t 3.4 TEIKYO UNIVERSITY HUMAN TRIALS (HATANAKA)
Upon his return to Japan, Hatanaka organized a team to pursue additional clinical trials of boron neutron capture therapy based on various improvements discovered since the last MIT clinical trials. These can be divided into 3 major areas: 1) a new boron compound such as Na2B12 H itSH, which was found to have superior tumor-to-blood ratio than any previous compounds; 2) large doses of adrenocorticosieroids, which appear to suppress radiation damage to the blood vessels in the brain; and 3) limiting the radiation dose to 20 Gy or less.
During the pathological studies of the patients treated at MIT,it was discovered that the patients had suffered acute radiation damage to the capillary walls of the brain. This was thought to be caused by a high concentration of boron present in the blood vessels at the time of irradiation. Indeed, the bomn concentration in the blood was almost twice as high as that in the tumor. Therefore, a new boron compound was sought which would produce a higher tumor-to-blood concentration ratio.
Following the clinical trials, Sweet and Soloway at MGH continued to work on compound development. One compound which showed some promise was sodium mercaptoundeca-hydrododecaborate, Na2B12H iiSH. However this compound was found to be highly unstable and would oxidize in the presence of water or air to form toxic derivatives. In addition, it was difficult to determine the purity or level of oxidation present in the chemical solution of this compound.
Hatanaka at this time was working as a post-doctoral scientist at MGH studying the toxicity of this compound on various animals. He was sufficiently impressed and solicited assistance from Shionogi Research Laboratory in Osaka in preparing an enriched, purified 68 Chapter three
- _ 4 '
e 4*
y a
form of this compound. A team led by H. Watanabe was successful in preparing the-compound enriched with boron-10. Early distribution studies in human patients with this -
compound showed tumor-to normal brain ratios ranging from 25 to 7.16 More importantly, the new compound also showed increased concentration in tumor compared to -
blood approximately 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after injection, with tumor to-blood ratio ranging from 1 to 3 if delivered by common carotid artery and as high as 8 when delivered by the internal .
carotid'anery. Independent confinnation of this finding proved to be difficult. Sweet,in ;
his hum.m distribution trials, found no incirased boron concentration in tumor compared to blood.ts However, in his trials,' Swcet had administered the compound prior to tumor debulking. - Hatanaka administers the drug only after complete debulking of the tumor.
Hatanaka theorizes "that a tumor excision decreases the intracranial pressure and straightens 1 the tortured blood vessels in the arca surrounding the tumor, improves the blood circulation, and eventually enables a good transport of boron 10 to the residual tumor
" cells " To date his theory has not been proved Hatanaka had also conducted experiments with cats during his study at MGH, in order to verify that high boron concentration in blood was responsible for the late radiation' damage to the normal brain found in the patients treated at MIT, he irradiated a series of cats at the __
MITR-I after the administration of perhydrodecaborate. In those studies, Hatanaka demonstrated that an acute cerebral capillary injury similar to those observed in MIT patients could be duplicated. However, this injury was reduced significantly if the cats
were given large doses of an adrenocortical steroid (20 mg prednisonlone succinate). The
~e lectron micrographs of the capillary in these cats showed no radiation damage following NCT.13 There is no proven explanation as to why this occurs and Hatanaka himself does not provide any reasons.
After it was discovered that the patients at MIT may have received lethal doses of radiation, Hatanaka, with the advice of N. Watanabe, a radiologist, decided to limit the neutron Appilcability-of boron neutron capture therapy 69
\
l .
Cuence to give a maximum of 20 Gy to the surface of the brain. This was the general consensus on the maximum dose that a healthy brain can tolerate without radiation damage.
According t' Hatanaka's calculations, this dose corTesponded to a maximum neutron fluence of 1013t o the surface of the brain for clinically useful concentrations of bomn 10
(~30 to 80 ppm administered per body weight).
On August 20,1968. Hatanaka treated his first patient using BNCT at the request of the patient's family. She was a 29 year old female patient who had undergone 2 craniotomies and a 50 Gy of telecobalt therapy with little change in condition. At the time of neutron irradiation, she weighed 30 kg, and had a fist size tumor extruding from her craniectomized head. He was authorized to perform the treatment only after extensive review by the Japanese Atomic Energy Commission's special committee composed of several radiologists and physicians. Although she survived only 4 months following NCT, she showed remarkable improvement, including gaining 13 kg in body weight prior to her death.
3.4.1 Hatanaka's revised therapeutic protocol for BNCT The patient is operated on approximately one week prior to BNCT.14 The visible tumor is excised completely whenever possible in order to maximize the effectiveness of the thermal neutron beam. The wound is then closed and allowed to heal. In contrast to data by Soloway, Hatanaka claims that the BBB can be repaired in one week with large doses of steroids.
On the day before neutron irradiation, the patient is given appmximately 30 to 80 mz of boron-10 per kg of body weight enriched Na2 B 12 H iSH i compound by internal carotid or common carotid artery through a motorized infusion pump. The patient is also given 5 mg/kg body weight prednisolone succinate by I.V. injection. Ten to twelve hours after injection, the patient is once again operated on at the reactor site. A craniotomy is 70 Chapter three
performed to expose the tumor to the neutron beam aperture. Instead i a balloon as used by Sweet. Hatanaka uses a sterilized " ping pong ball" to keep the normal brain from collapsing into the tumor cavity and to extend the usefulness of the horizontal thermal l neutron beam at depth. Gold foils and TLDs are placed inside the head to measure the incident thermal neutron Quence and gamma rays. Then the patient's head is covered with a helmet shaped shield made with either boronated plastic or lithium 6 Ouoride loaded l plastic. This is to protect the head and the healthy scalp from the radiation field. The shield has a hole cut in it corresponding to the craniotomy to expose only the open tumor area to the neutmn beam. The entire operative area is then covered with a sterile plastic film to give aseptic protection. During the entire operative procedure and irradiation, the patient is kept under remote anesthesia developed by Drs. T. Niinobe and hi. Kamiyama.12 The patient is then irradiated at one of four nuclear facilities: the 100 kW Hitachi Training Reactor (HTR) in Kawasaki City, the 10 htW Japan Research Reactor (JRR-3) in Tokai Village, Ibaragi Prefecture, the 5 htW Kyoto University Reactor (KUR) in Kumatori, Osaka, and the 100 kW hiusashi Institute of Technology Research Reactor (htulTR) in Kawasaki. Because of the low neutron flux at the therapy beam port for most of these reactors (maximum thermal neutron Oux at the 10 htW JRR-3 was 5 x 108 n/cm2 sec), the irradiation took several hours.
Following the irradiation, the open wound was closed; however, a plastic draining tube was left inserted in the tumor cavity area to assure discharge of tumor debris and CSF.
This ingenious device eliminated the need for many cerebral shunt and spinal taps that had to be made on patients irradiated in the U.S. o relieve pressure from build up of CSF. In one patient, Hatanaka collected tissue sediments totalling 89 g from the tumor cavity. For most of the patients, the discharge from the cavity lasted a few weeks. The Guids were drained continuously until they turned clear, at which time the tube was removed. In addition, for 1 to 3 weeks following the therapy, Hatanaka irrigated the tumor cavity Applicability of boron neutron capture therapy 71
through the plastic tube with 5 mg solution of methotrexate solution. Provided that there were no complications, the patients were usually allowed to go home within 20 days following the initial surgery.
In 1985, Hatanaka also introduced the use of heavy water (D20) to treat tumors located at greater than 6 cm depth. By replacing the water that is found in normal brain with heavy water, he claims to be able to extend the useful penetration depth of the thermal neutron beam. According to his study, the acute toxicity from such procedure in terms of LD50 is the same as that of water. However, there are no other independent studies to suppon his toxicity claim.
3.4.2 Selected case history for Hatanaka's clinical trials of BNCT Because the published data on patients treated by Hatanaka is incomplete and superficial at best, and also due to the large number of patients treated by him (over 100), in the following section only a small number of patient case histories will be summarized. The overall data on the rest of the patients will be summarized in a later section.
Although Hatanaka has published several papers on the clinical case histories for a number of patients, they are all lacking information needed to determine the approximate radiation dose delivered to the tumor. This informat6n is vital to correlating the survival time with both the tumor dose and the healthy tissue dose. Even in his published report on eighteen select patients autopsied after their deaths following NCT, he fails to include the amount of boron given to the patients for the therapy.14 In another publication, he gives detailed information on boron concentration found in tumor as well as blood for each patient, but leaves out the neutron fluence information.15 Even though he provides detailed information on the patient's symptoms and the outcome, this information is of little use without the information on dose to determine the correct correlation.
72 Chapter three
i.-
Altogether, Hatanaka had treated 98 patients using BNCT as of March 1989. Of these patients, specific neutron fluence information was published on 20 patients. Specific boron concentration information was published for 6 patients. Among these, only 3 patients could be found with both the boron concentration data and the neutron fluence data. Table 3 7 summarizes the irradiation information in a manner similar to that shown in tables 3 1, '
3-3, and 3 5 for the US clinical trials.
Among the patients listed, E.H. belongs in a category of patients that have been operated en by neurosurgeons other than Hatanaka. According to Hatanaka, this is an important criteria in judging BNCT cffectiveness, Patients M.S. and M.T. are what Hatanaka claims as true BNCT patients. These patients were diagnosed, operated on, cad irradiated by Hatanaka.
Table 3-7: Summary of select clinical trials by Hatanaka (1968-1974)l11314 Date Date Amount Neutron Days Patient Sex Age of of of "B Fluence Survival Surgery NCT Given Given Post NCT (35 ppm)
No. 7 F 22 Sept 71 Sept 71 34 ppm in 9.70 E+12 30 weeks (E.H.) tumor,10 mm in blood (40 ppm)
No, 7 M 28 Dec 24,71 Jan 29,72 37 ppm in 1.00 E+13 greater than (M.S.) tumor,23 34 months mm in blood (40 ppm)
No.10 M 50 Jun 20,72 Jun 29,72 15 ppm in 0.96 E+13 greater than (M.T.) tumor,28 13 years l mm in blood Table 3-8 summarizes my calculated estimate of the approximate dose delivered to the tumor and the capillary wall 3 cm below the open craniotomy ba:,ed on the measured boron concentration in the tissue and the thermal neutron fluence. All of the patients listed in the table were irradiated at the HTR. Although the thermal neucon flux and the incident l Applicability of soron neutron capture therapy 73
! l
core gamma ray dose from this reactor are well characterized, there is no infomution on the incident fast neutron dose. Hoy ever, based on the published cadmium ratio of 150-300, it was assumed that the fast nertron dose would be only a minor fraction of the overall dose given to the patients. The incident core gamma ray dose to the patient during the therapy was 0.5 Gy/hr. With aserage neutron flux of 3.5 x 108 n/cm2sec, the patient received j approximately 0.4 Gy/1012 n/cm2in incident gamma ray dose. For the dose calculations, the patient was assumed to receive 0.62 Gy/1012 n/cm2 in induced gamma ray dose. This was based on measurements made at MIT in similar neutron beams. The boron KERMA ,
l factor of 2.0 x 10-11 RBE cGy cm2/n/ ppm 10B and thermal neutron KERMA factor of 2.2 I x 10-11 RBE cGy cm 2/n were also assumed.
Table 3-8: Summary of post-operative results and estimates made by this l author of absorbed dose to the tumor, and blood capillaries at 3 cm !
below the open craniotomy for select patients treated at HTR from 1968 to 197412.13.14 Patient Post operative findings Date of Tumor Capil.
NCT dose dose (RBE (RBE G y) Gy)
She died from bacterial infection of the scalp. Her No. ? skin flap from the craniotomy she received before Sept 71 27.7- 7.8 (E.H.) Hatanaka's treatment was already showing signs of atrophy before BNCT.
Patient suffered from rapid regrowth of tumor before No. ? NCT. After NCT,89g of sediments were collected Jan 29,72 30.5 10.9 (M.S.) from his tumor cavity over 2 weeks. He _ was dis-harged within 2 months and returned to his normal occupation as a salesman.
He showed improvements in his speech and motor No.10 coordination and was discharged 3 weeks after NCT. Jun 29,72 15.2- 11.6 (M.T.) He tuer showed symptoms of degeneration of his left retina and eventual formation of a cataract that could be auributed to late radiation damage.
Although there are only three examples froia Hatanaka's trials that could be used to compare with the work of other therapists, some unique points do stand out fmm his trials.
The doses given to the patient are clearly in between those of the BNL trials from '51 '53 and the MIT trials. At the BNL trials, the highest tumor dose given at any single irradiation 74 Chapter three
t*
was -6 Gy. Although they noted some improvement in their patients even with this low dose, they were unable to achieve a cure. The patients from the MIT trials received 3 to 4 times the dose that flatanaka's patients received and, subsequently, all showed signs of serious radiation damage to the nonnal brain. It is interesting, however, that the MIT patients showed remains of tumor sites at autopsy in almost every case, while the patients treated by llatanaka with much lower dose apparently show a complete cure. Even E.ll.
died of bacterial infection that is attributed to her therapy prior to NCT. The only other significant difference during therapy between the MIT trials and the llTR trials was that the irradiation time for MIT was less than half an hour in some cases where as at ilTR the patients were irradiated for several hours each.
Perhaps it was the drain tube that made the difference in llatanaka's trials in patient M.S.,
79 g of tumor debris was removed by this tube. Without the tube, the debris would have remained in the tumor cavity where it could have reproliferated or have been transponed to other areas of the brain. If this is indeed a vital part of Hatanaka's success, it should be further investigated. For example, a drain tube might be introduced after the debulking procedure.
3.4.3 Overall view of Hatanaka's clinical trials to date From August 1968 to March 1989, Hatanaka has treated 98 patients. lie has used 4 different nuclear reactors in that 20 year span. Table 3 9 outlines the physical characteristics of each reactor. They range from the 100 kW HTR and the 100 kW TRIGA type reactor (MulTR) to the large 10 MW JRR 3. He has encountered great difficulty in achieving this feat. He had to persuade not only govemment agencies, but also the families of the patients, and other physicians around the world. Although data on the specific nature of his therapy for each patient is sorely lacking, it cannot be denied that he has shown significant increases in the patient survival rate from this deadly disease.
Applicability of boron neutron capture therapy 75 i.maisii-si
llatanaka separates his patients into four distinct groups when discussing his results.
These four groups are: 1) those that have received prior treatment from other physicians including photon and chemotherapy; 2) all BNCT patients treated between 1968 and 1985 with exception of those in group 1; 3) those patients from group 2 that had tumor within 6 cm from the surface; and 4) those patients treated since 1985. There are 46 cases from group 1,38 cases from group 2,12 cases in group 3, and 11 cases from group 4, for a total of 95 cases as of February 1989 (12 cases from group 3 are also included in the group 2 tally,46 + 38 + 11 = 95).
Table 3-9: Physical characteristics of the nuclear reactors used by llatanaka for BNCT from 1968 to 198916 Name Type Thermal power Max. % inc. rdose Gold Cd. # Patients (LW) (n/cm2 sec) (Gy/1012 n/cm2) ratio treated llTR Swimming pool 100 3.5 x 108 0.40 150-300 13 JRR 3 -- 10,000 4.8 x 108 0.93 -- 1 KUR -- 5,000 1.1 x 109 0.13 5,000 1 MulTR TRIGA-Il 100 1.5 x 109 0.09 27-100 83 The 5-year survival rate with group I was less than 3%. His is not much different from the survival rate using conventional multimodality therapy for grade !!! and IV gliomas (5%).17 The 5 year survival rate with group 2 is much better at 19%. liowever, this is not significantly better than the rate achieved by conventional therapy group for all glioma patients (10-15%) when one considers that, most likely, not all of flatanaka's patients suffered from grade 111 and IV gliomas. It is the select patients from his group 2 that showed a trmarkable 5-year survival rate of 58%. There have been critics ofliatanaka for picking only these good results to constitute his group 3. Ilowever, these survival rates are greater than any that %ve been reported for any type of malignant brain tumor. In addition, his group 4 patients seem to be following a similar trend in survival to his group 3 patients.
76 Chapter three
1 .
Because the number of patients in this group is, so far, low (11 cases), it is too early to evaluate the outcome on the effectiveness of BNCT.
Table 310 summarizes the effectiveness of DNCT for each group in tem s of survival rates l
and performance rating (Karnofsky) and compares this to the best conventional multimodality therapy for grade 111 and IV gliomas, in the conventional group, grade ill gliomas accounted for 41 patients and grade IV gliomas accounted for 60 patients.
Karnorsky rating is a measure of the deterioration of function and a score of 1009 indicates no loss of function. What is apparent from this table is that although 11NCT seems to provide better survival rates if used as a primary therapy, if it is used in conjunction with conventional therapy (group 1), it is less effective than conventional therapy alone. This is explained by liatanaka to be caused by 3 effects: 1) delayed cerebral radio necrosis,2) surgical complications, and 3) chemotherapy complications.
Table 310: Comparison of survival rates for patients treated by llatanaka from 1%8 to June 1985 and conventional multirrv.xiality therapy for grade til and IV gliomas ,
Bomn neutron capture therapyis Conventional Group 1 Group 2 Group 3 therapyt6
- patients 46 38 12 101 Age (years) 43.Si2.1 50.211.9 49.8i4.1 -
Mean survival (days) 482.0164.8 612.7tl48.6 1320.4i402.I 675 Median survival (days) 349 289 768 --
5 year survival rate 4.6% 19.3 % 58.3 % 4.0%
10-year survival rate 0% 9.6% 29.2 % -
1.ive as of June 30,85 1/46 8/38 5/12 -
Kamorsky rating 57.2i2.9 73.4i3.9 86.714.1 71.2*
- Average rating at 2 years post op for those patients surviving at least 2 years (34/101).
The normal brains of those patients treated with conventional therapy are often subjected to l 60 Gy or more of photon therapy on a fractionated schedule. This limits the brain's tolerance to other radiation therapy including NCT and, subsequently, the patient often j Appliesbility of boron neutron capture therapy 77 l
4 suffers from severe radiation damage following NCT. In addition,llatanaka states that the skill of the surgeon in diagnosing and operating on the glioma patient is very important for NCT. lie has lost a number of patients during early trials through surgical complications from previous therapy such as serum hepatitis, atrophic scalp, topical misdiagnosis, and misleading information. In addition, the optimal craniotomy window is much larger for BNCT than it is for conventional therapy. Complications arising from previous chemotherapy include generalized weakness, cachexia, thromboaneritis of the carotid anery after continuous infusion, and immune suppression.14
3.5 CONCLUSION
Although llatanaka has shown remarkable hope for treating this otherwise fatal disease, grades 111 and IV gliomas, his results have pointed out a need for .* neutron beam with greater effectiveness at depth. The difference he has shown for survival rates of those patients with superficial tumors (<6 cm depth) from the overall group (5 year survival rates of 58.3% compared to 19.3%) indicate that BNCT is only as effective as the ability of those neutrons to penetrate deep into the tissue. One approach to achieve this is through replacement of light water in the brain with heavy water. Although llatanaka claims that there is no acute effect from this procedure, there is some evidence that the maximum allowable concentration of D20 may be only 20% (this apparent contradiction may relate to acute verses chronic exposure to D2 0). If this is true, then some other method must be found to extend the effective range of the neutron beam in tissue. The use of epithermal neutron beams is a proven method to extend the useful range of neutrons in tissue.
Whereas a thermal neutron beam loses over 90% of its peak flux at 6 cm depth, an epithermal neutron beam can retain over 50% of its peak flux at the same depth. The advantages of using such a beam is the subject of the next chapter.
78 Chapter three ;
- sg 5
t REFERENCES
- 1. G. Locher, " Biological effects and therapeutic possibilities of neutrons," American Journal of Roentgenology, 1936,36:113.
- 2. D. Gabel, S. Foster, and R. Fairchild, "The Monte Carlo Simulation of the Biological Effect of the !?B(n.a)7Li Reaction in Cells and Tissues and its implication for Boron Neutron Capture Therapy," Radiation Research , 1987,111:14 25.
- 3. Il. llatanaka, " Introduction," Boron Neutron Cacture Thernov for Tumors,
~
Nishimura Co., Niigata, Japan,1986.
- 4. W. Sweet,"The uses of nuclear disintegration in the diagnosis and treatment of brain tumor," New England Journal of Medicine,195 l, 245: 875-878.
- 5. W. Sweet and M. Javid, "The possible use of neutron capturing isotopes such as boron 10 in the treatment of neoplasms.1. Intracranial Tumors," Journal of Neurosurgery, 1952,9: 200-209.
- 6. L. Farr, W. Sweet, J. Robertson, C. Foster,11. Locksley, D. Sutherland, M.
Mendelson, and E. Stickley, " Neutron capture therapy with baron in the tscatment of glioblastoma multiforme," American Journal of Roentgenology, 1954, 71: 279 291.
- 7. J. Godwin, L. Farr, W. Sweet, and J. Robertson, " Pathological study of eight aatients with glioblastoma multiforme treated by neutron capture therupy using
- oron 10," Cancer, 1955, 8: 601 615,
- 8. O. Deutsch and B. Murray, " Monte Carlo dosimetry calculation for boron neutron capture therapy in the treatment of brain tumors," Nuclear Technology,1975, 26: 320-339,
- 9. L. Farr, J. Robertson, E. Stickley, H. Bagnall, O. Easterday, and W. Kahle, "Recent advances in neutron capture therapy," Procress in Nuclear Enerev Series Vil -
Medical Sciences Vol. 2, London, Pergamon Press, 1959,128-138.
- 10. A. Asbury, R. Ojemann, S. Nielsen, and W. Sweet, "Neuropathogic study of fourteen cases of malignant brain tumor treated by boron-10 slow neutron capture radiation," Journal of Neuropathology and Experimental Neurology, 1972, 31:278.
- 11. W. Sweet, A. Soloway, and G. Brownell, " Boron-slow neutron capture therapy of gliomas," Acta Radiology, 1963,1: 114 121.
- 12. W. Sweet " Final report on grant # AT(301) 1093 The use of thermal and epithermal neutrons in the treatment of neoplasms," from private notes.
I Applicability of boron neutron capture therapy 79
- b 13.11. llatanaka, and K. Sano, "A revised boron neutron capture therapy for malignant brain tumours, I. Experience on terminally ill patients after Co 60 radiotherapy,' Journal of Neurology, 1973,204:309-332.
- 14. !!. llatanaka, " Eighteen autopsy cases of malignant brain tumors treated by boron-neutron capture therapy between 1968 and 1985," Boron Neutron Caoture
'neraov for Tumors, Nishimura Co., Niigata, Japan,1986.
15.11. llatanaka, "A revised boron neutron capture therapy for malignant brain tumors,
- 11. Interim clinical result with the patients excluding previous treatments,"
Jourt al ofNeurology, 1975,209:8l 94.
16.11.11atanaka, " Clinical experience of boron neutron capture therapy for gliomas - a comparison with conventional chemo immuno-radiotherapy," Boron neutron caoture theraov for tumors. Nishimura Co., Niigita, Japan,1986.
- 17. K. Jellinger, "Present limits of conventional treatment for malignant brain tumors,"
Doron neutron caoture thernov for tumors, Nishimura Co., Niigita, Japan, 1986.
I8. W. Sweet," Supplementary phamiacological study between 1972 and 1977 on purified mercaptoundecahydrododecaborate," Daron neutron caoture thernov for tumors, Nishimura Co., Niigita, Japan,1986.
b0 Chapter three
.. . - - - . - - - _ _ _ - - - - - .