ARTICLE IN PRESS

Applied Radiation and Isotopes 61 (2004) 1069–1073

Effectiveness of BNCT for recurrent head and

neck malignancies
Itsuro Katoa,*, Koji Onob, Yoshinori Sakuraic, Masatoshi Ohmaed,
Akira Maruhashic, Yoshio Imahorie, Mitsunori Kirihataf,
Mitsuhiro Nakazawaa, Yoshiaki Yuraa
a
Department of Oral and Maxillofacial Surgery II, Osaka University, Graduate School of Dentistry, 1-8 Yamada-oka, Suita,
Osaka 565-0871, Japan
b
Radiation Oncology Research Laboratory, Research Reactor Institute, Kyoto University Noda, Kumatori-cho, Sennan-gun,
Osaka 590-0494, Japan
c
Radiation Life Science, Research Reactor Institute, Kyoto University Noda, Kumatori-cho, Sennan-gun, Osaka 590-0494, Japan
d
Department of Oral and Maxillofacial Surgery, Izumisano Municipal Hospital, Rinku General Medical Center,
2-23 Rinku Ourai-kita, Izumisano,Osaka 598-0048, Japan
e
Department of Neurosurgery, Kyoto Prefectural Medical University 465 Kajii-cho, Hirokoji-agaru, Kawara-machi, Kamigyo-ku,
Kyoto 602-8566, Japan
f
Department of Agriculture and Life Science, Osaka Prefectural University 1-1 Gakuen-cho, Sakai, Osaka 599-8531, Japan

Abstract

Recurrent head and neck malignancies (HNM) are often radio-/chemo-resistant and show extensive growth,
necessitating a wide resection including surrounding tissues. To avoid severe impairment of oro-facial structures and
functions, it is necessary to explore new treatments for HNM. Boron neutron capture therapy (BNCT) is tumor-cell
targeted radiotherapy that has significant superiority over conventional radiotherapies in principle. We report here, first
in the world, six patients with a recurrent HNM who have been treated with BNCT. The BNCT in combination with
boronophenylalanine (BPA) and borocaptate sodium (BSH) was performed using the epithermal neutrons with Kyoto
University Research Reactor (KUR). The results of BNCT were as follows: (1) 10B concentration of tumor/normal
tissue ratios (T=N ratio) of PET studies were SCC:1.8–4.4, sarcoma:3.1–4.0, parotid tumor:3.5. (2) Relative volume (%)
of each tumor to the prior were 6–46%. (3) Remarkable reduction (46–100%) of huge tumor such as 40–675 cm3
(average: 315 cm3), improvement of QOL and very mild side effects were recognized in all cases. These results indicate
that BNCT represents a new and promising treatment approach even for a huge or far advanced HNM.
r 2004 Elsevier Ltd. All rights reserved.

Keywords: Head and neck malignancies; BNCT; Epithermal neutron; BPA and BSH

1. Introduction tissues. To avoid severe impairment of oro-facial

Recurrent head and neck malignancies (HNM) are
often radio-/chemo-resistant and show extensive growth,
necessitating a wide resection including surrounding
*Corresponding author. Tel.: +81-6-6879-2941; fax: +81-6-
6876-5020.
E-mail address: katoitsu@dent.osaka-u.ac.jp (I. Kato).
structures and functions, it is necessary to explore new
treatments for HNM. Boron neutron capture therapy
(BNCT) is tumor-cells targeted radiotherapy that
significantly increases the therapeutic ratio relative to
conventional radiotherapies (Slakin, 1991; Barth et al.,
1992). When 10B absorbs thermal neutrons, the a and
7
Li particles generated in the 10B(n,a)7Li reaction are of
high-linear energy transfer radiation, and carry an

0969-8043/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apradiso.2004.05.059
 ARTICLE IN PRESS
1070 I. Kato et al. / Applied Radiation and Isotopes 61 (2004) 1069–1073

6% (22 mo) Alive (25 mo)

P.D. (2 mo) Dead (10 mo)

F: female, M: male, S.C.C.: squamous cell carcinoma, T=N ratio: B concentration ratio of tumor/normal tissue studied by F-BPA PET, a: 30% of the center dose at the margin of
average total kinetic energy of 2.34 MeV, and have a

C.R. (3 mo) Dead (7 mo)

46% (6 mo) Dead (8 mo)

27% (1 mo) Dead (2 mo)

10% (4 mo) Alive (5 mo)

short range (4–9 mm) of approximately one cell diameter,

(duration) prognosis
Relative volume %
resulting in induction of a large relative biological

17% (1.5 mo)

effectiveness and selective destruction of tumor cells

18% (11 mo)

63% (1 mo)
containing 10B. The clinical application of BNCT has

(survival)
been limited to advanced brain tumors and malignant
melanomas (Mishima et al., 1989).

mucosa
We report here, first in the world, that six patients

8.0/8.1
Skin/

5.5

6.9

3.6

5.9

7.2
with HNM who have been treated with BNCT. All

5
patients are recurrent cases after the standard treatment

Deepest-tumor
with no effective treatment left.

11.7(5.8 cm)
4.4(9.5 cm)

6.9(7.5 cm)
19.7(4.5 cm)
14.4(5.0 cm)

21.0(5.5 cm)
15.6a(6 cm)
7.4(8 cm)

7.8(8 cm)
This BNCT was characterized by (1) fluoride-18-

(depth)
labeled p-boronophenylalanine (18F-BPA) positron
emission tomography (PET) studies before BNCT, (2)
use of epithermal neutrons, and (3) administration of

Tumor-peak

23.0(2.5 cm)

37.6(2.5 cm)
29.1(2.5 cm)
36.5(1.5 cm)
27.3(2.0 cm)

41.1(2.0 cm)
22.9(3 cm)

27.9(3 cm)

28.4(3 cm)
boronophenylalanine (BPA) in combination with bor-

(depth)

collimator=4.7 Gy-Eq, mo: month(s), P.D.: Progressive Disease, C.R.: Complete Response, G: use of gelatin (5 mm-thickness).
ocaptate sodium (BSH) as 10B-carriers.

Total dose (Gy-Eq)

18
Tumor-surface
(Gy-Eq)

15.3(G)
11.5(G)

20.7(G)
2. Materials and methods

8.8

7.5
16.3

11.9

17.1
12
2.1. Patients

(E+12 cm
2)
T-max of thermal neutron

1.7(3.0 cm)

2.0(2.5 cm)
1.7(2.4 cm)
2.6(2.5 cm)
2.7(2.5 cm)
1.3(2.5 cm)
1.8(2.0 cm)
2.0(3.0 cm)

2.5(2.0 cm)
Fluence
Six patients with a HNM having 3 squamous cell
carcinomas (SCCs), 2 sarcomas and one parotid grand
cancer. All patients are recurrent cases after the
Irradiation
time (min)

standard therapy with no effective treatment left. All

the cases had the approval of the ethical committees, 60

70
90
40

77
120

100
120

120
medical committee of KUR and that of Osaka
University, Graduate School of dentistry.
irradiation

Characteristics of each case was summarized in

(ppm)
After
B concentration in Blood (ppm)

32.2

32.9

49.1
39.1
57.9
42.1
32.3

36.2
(—)

Table 1, such as patients’ gender and age, clinical

diagnosis, histopathological diagnosis, initial tumor
irradiation

volume (average: 315 cm3), 10B concentration ratio of

Before

(ppm)

tumor/normal tissue (T=N) of 18F-BPA, etc.

50.8

59.3
81.7
77.2
73.2
57.6
(—)

10
44

59
of 18F-BPA
T=N ratio

18
2.2. F-BPA PET study
3.5–4.0
2.4/1.8
4.0/4.4
3.3

3.5

3.1
10

The accumulation of BPA to tumor and normal tissue

was imaged and quantified by 18F-BPA-PET study
volume (cm3)
Initial tumor

before BNCT.
The synthesis method and preparation of L-18F-BPA
675

423
110
451

450
123
294
40

(Imahori et al., 1998a) and the choice of region of

interests (ROIs) (Imahori et al., 1998b) are detailed in
Mucoepidermoid

Osteosarcoma

Inflammatory

our previous report.

fibrosarcoma
Histopathol.

carcinoma
diagnosis
Treatment summary of 6 cases

S.C.C.
S.C.C.
S.C.C.

2.3. Boron compounds

Borocaptate sodium (BSH) and para-boronophenyla-

Case Gender (age) BNCT

2nd (supine)

2nd (seated)
3rd (seated)
F (67) 1st (supine)

F (58) 1st (seated)

lanine (BPA) were used as 10B-carriers (Ono et al.,

M (61) (supine)
M (56) (supine)

M (73) (seated)
F (60) (seated)
no. (posture)

1999). BSH (5 g/body) in 50% physiological saline and

BPA (250 mg/kg) in fructose solution were administered
intravenously for 1 h. The epithermal neutron irradia-
Table 1

tion with KUR was started 12- and 1-h after the
no.

administration of BSH and BPA, respectively.

1

2
3
4
5
6
 ARTICLE IN PRESS
I. Kato et al. / Applied Radiation and Isotopes 61 (2004) 1069–1073
10
2.4. Measurement of neutron fluence, B-concetration
and dose estimation
The blood samples were taken from central vain route
at the time of just finished BSH-infusion (0-h), 3-h later,
6-h later, 9-h later, just before BPA-infusion, just before
irradiation, and just after irradiation. 10B concentration
in the blood was measured by the prompt g-ray analysis.
Each 10B concentration and its exact time were plotted.
As 10B concentration-curve of BSH has a linear portion
after 6-h of BSH-infusion, the average BSH concentra-
tion during the clinical irradiation could be estimated by
the extrapolation of the measured data before the BPA
infusion. The average BSH+BPA concentration could
be estimated by the interpolation of the measured data
after the BPA infusion in the same manner. By the
subtraction of the BSH+BPA concentration and the
BSH concentration, the BPA concentration could be
estimated.
The distributions of neutron fluences, and the physical
doses of neutrons and gamma-rays were calculated by
the dose-planning system ‘‘SERA’’ (Wessol et al., 1999).
The SERA-calculated distributions were normalized
using the thermal and epi-thermal neutron fluences
measured by gold-wire activation analysis at the
patient’s surface. Using the normalized data for
the thermal neutron fluence, and the fast and g-ray
physical doses, the CBE and RBE-weighted doses, ETotal
(Gy-Eq), were calculated by the following equations:
ETotal ¼ EB10 þ EThermal þ EFast þ Eg ;
EB10 ¼ ðCBSH  CBEBSH þ CBPA  CBEBPA Þ
 7:43  10
14  FThermal ;
EThermal ¼ N  RBEThermal  6:78  10
14  FThermal ;
EFast ¼ RBEFast  DFast ;
Eg ¼ RBEg  Dg ; ð1Þ
D: Physical Absorbed doseðGyÞ;
FThermal : Thermal neutron fluenceðcm
2 Þ;
14
N concentrationð%Þ; where 14
N ¼ 2%;
10
C: B concentration ðppmÞ; ð2Þ
RBEThermal ¼ 3:0; RBEFast ¼ 3:0; and RBEg ¼ 1:0;
CBEBSH ¼ 2:5 and CBEBPA ¼ 3:8 for tumor;
CBEBSH ¼ 0:8 and CBEBPA ¼ 2:5 for normal skin;
CBEBSH ¼ 0:37 and CBEBPA ¼ 1:35 for normal tissue:
ð3Þ
3. Results
The results of all 6 cases are summarized in Table 1.
(1) 10B concentration of T=N ratio of PET studies were
SCC: 1.8–4.4, sarcoma: 3.1–4.0, parotid tumor: 3.5.
1071
Especially in the patients with SCC, T=N ratios were
varied individually. (2) Tumor sizes were 40–675 cm3
(average: 315 cm3). Relative volume (%) of each tumor
to the prior was as follows. 3 SCCs: CR (complete
response) for 3 month (mo), 27% after 1 mo and PD
(progressive disease) after 2 mo, 2 sarcomas: 46% after 6
mo and 10% after 4 mo, parotid tumor: 6% after 25 mo,
respectively. (3) Improvement of QOL such as a relief of
symptoms of severe pain, bleeding, and flow of mucous
exudates at the local lesion were recognized in all 6
cases. Especially in case 2, improvement of the
performance status from 4 to 2 was observed. (4) A
few side effects such as transient mucositis and alopecia
less than Grade-2 by NCI-CTC were recognized in all
cases.
Case 1: A 67-year-old woman was diagnosed with
mucoepidermoid carcinoma of parotid gland in 1998
and underwent a parotidectomy, followed by radio-
therapy. In March 1999 the tumor recurred and
additional chemotherapy was ineffective. In October,
2001 the ulcerated tumor had grown to 13.5  12.5 
8 cm3 and caused pain, bleeding and mucous exudates
(Fig. 1 upper).
Although the tumor had shrunk by 63%, one-month
after the first BNCT, the second BNCT was performed
with gelatin-sheet because insufficient radiation dose at
the tumor surface had resulted in re-growth. BNCT
caused great effects on the patient such as tumor
reduction, relief of pain and exudates-secretion from the
ulceration, in spite of slight side effects such as transient
mucositis, and alopecia (less than Grade-2 by NCI-CTC),
with the third BNCT, one-year after the first BNCT has
been gradually brought a 6% of relative tumor volume to
the prior for 25 M, remarkable reduction (94%) in the size
of the tumor (Fig. 1 lower), and disappearance of tumor
were ulceration achieved with normal skin cover and
continuing improvement in facial palsy.
Case 2: A 61-year-old man with recurrent maxillary
sarcoma (Fibroblastic osteosarcoma) after neo-adjuvant
and adjuvant chemotherapy and operation was treated
with 68 Gy of radiation therapy in vain. With the
purpose of improvement of QOL the BNCT was
performed for the patient at the terminal stage. In this
case T=N ratio (3.5–4.0) was comparatively high and
10
B blood concentration just before the irradiation was
the highest (81.7 ppm) out of the 6 cases. A peak of the
total dose equivalent at the tumor was 28.4 Gy-Eq (3 cm
depth) and deepest tumor (9.5 cm) was 4.4 Gy-Eq. Per-
formance status (PS) of the patient remarkably im-
proved: before (PS: 4) and after BNCT(PS: 2). Though
he was in the bed all day long because of severe
headache, bleeding, nausea and vomiting before BNCT,
he came to our hospital on his own foot for 2 months
without those symptoms, after BNCT. We suppose that
BNCT might have made him live longer for 8 months
after all.
 ARTICLE IN PRESS
1072 I. Kato et al. / Applied Radiation and Isotopes 61 (2004) 1069–1073
Fig. 1. Clinical appearance of case 1: Before BNCT (the upper
panel) and 22 months after the first BNCT (the lower panel).
Effectiveness of BNCT such as remarkable reduction of the
huge tumor, disappearance of ulceration with normal skin
cover and continuing improvement of facial palsy testify to the
tumor-selective character of BNCT.
Case 3: A 56-year-old man with recurrent oro-
pharynx carcinoma (SCC) after the operation, che-
motherapy, conventional radiotherapy and Cyberknife.
Low value of T=N ratio (1.8–2.4) and the deep location
from the skin surface made BNCT unsuccessful and
resulted in progressive disease (PD), in spite of the
highest value of thermal neutron fluence and high values
of total dose equivalent.
Case 4: A 60-year-old woman with recurrent gingival
carcinoma at mandible (SCC) after standard therapy.
The tumor volume was 450 cm3 and the huge tumor was
pushing more than half of the left oral cavity,
pharyngeal space. T=N ratio (4.0–4.4) was the highest
out of the 6 cases. Relative volume prior to and 1 mo
after BNCT was 27%, although the tumor was
chemotherapy-resistant and grew fast. BNCT, more-
over, completely stopped her severe pain at mandible the
next day and made her free from morphine.
Case 5: A 73-year-old man with recurrent gingival
carcinoma at mandible (SCC) after standard therapy,
operation, chemotherapy and conventional external
radiation therapy (60 Gy). The cancer was infiltrating
deep into the marrow of the mandible as a consequence
of cortex bone destruction. The clinical findings of
disappeared ulceration at the gingival and CT scan 1
month after BNCT revealed that the lesion disappeared
for 3 months. BNCT seemed effective for the cancer
infiltrating into the bone. But overdose exposure of
radiation made him radio-osteonecrosis, which needed
mandiblectomy to reveal several histo-pathological nests
of recurrent tumor cells.
Case 6: Recurrent maxillary sarcoma (Inflammatory
myofibrosarcoma) of a 58-year-old woman after opera-
tion invaded into the left of maxilla and infra-orbital
fossa. The patient did not choose the surgical operation
including removal of eye ball. The 50 Gy of conventional
external radiation therapy brought down approximately
60% of the relative volume of the tumor. Then two
BNCTs (1.5 M interval) brought down the lesion by
about 10% of the relative volume compared with that
status 4 months before BNCT.
4. Discussion
The effects of BNCT depend on a sufficient cellular
uptake of boron followed by an exposure to epithermal
neutron beam from a nuclear reactor. The critical factor
for the application of BNCT is the selective accumula-
tion of 10B-compound into the tumor-tissue, rather than
in the surrounding normal-tissue. 18F-BPA-PET study
showed a preferential accumulation of 18F-BPA in the
tumors (T=N ¼ 3:124:4) except that of case 3
(T=N ¼ 1:8=2:4) which showed poor effect of BNCT.
Tumor whose T=N ratio is less than 2.5 or tumors at
deep location from the skin surface seemed to be a
contraindication for BNCT such as case 3.
In case 1, we tried fractionated BNCT because the
tumor seemed too huge for a sufficient effect. Our
strategy was that if the first BNCT would shrink the
tumor to some extent, the second or the third BNCT
 ARTICLE IN PRESS
I. Kato et al. / Applied Radiation and Isotopes 61 (2004) 1069–1073
Fig. 2. Dose Profile of the second BNCT in case 1. Epithermal
neutron beam generates a peak of the thermal neutron flux at
2.5–3-cm beneath the tissue-surface. To increase the radiation
dose for the tumor surface, a 5 mm-thick gelatin-sheet was
placed on the tumor, which shifted the total dose profile to left,
resulted in a 5–7 Gy-Eq increase at the center of the tumor.
might bring sufficient effect even at the deepest of the
tumor because of shrinkage. Epithermal neutron beam
generates a peak of the thermal neutron flux at 2–3-cm
beneath the tissue-surface (Fig. 2). Although the tumor
had shrunk by 63%, one-month after the first BNCT,
the second BNCT was performed with gelatin sheet to
compensate for low dose at the tumor surface. A 5-mm
thick gelatin sheet was placed on the ulceration, which
resulted in 15.3 Gy-Eq (6.5 Gy-Eq increased) of radia-
tion doses at the tumor surface (Fig. 2). Two BNCTs
(one-month interval) brought 18% of relative volume
compared with the initial volume. This tumor-reduction
effect lasted at least for 11 months.
BNCT brought each patient great improvement of
QOL such as remarkable reduction of the huge tumor
(5 cases out of 6), improvement of PS (case 2), relief of
pain (cases 1, 2 and 4), decreased bleeding (cases 1 and
2) and exudates (case 1). Especially results of BNCT
such as disappearance of ulceration (cases 1 and 5),
normal skin cover and continuing improvement of facial
palsy (case 1) testify to the tumor-selective character of
BNCT. There were a few side effects such as transient
mucositis and alopecia less than Grade-2 by NCI-CTC
in all cases.
1073
We report here, first in the world that 6 patients with
HNM who have been treated with BNCT.
5. Conclusion
We could ascertain that the effectiveness of BNCT
used, ‘‘epithermal neutron beam’’, for head and neck
tumors in 5 out of 6 cases. These results indicate that
BNCT represents a new and promising treatment
approach even for a huge or far advanced HNM.
Acknowledgements
This work was supported in part by a grant
(16390589) from the Ministry of Education, Culture,
Sports, Science and Technology, Japan.
References
Barth, R.F., Soloway, A.H., Fairchild, R.G., Brugger, R.M.,
1992. Boron neutron capture therapy for cancer. Realities
and prospects. Cancer 70, 2995–3007.
Imahori, Y., Ueda, S., Ohmori, Y., Kusuki, T., Ono, K., Fujii,
R., Ido, T., 1998a. Fluorine-18-labeled fluoroboronpheny-
lalanine PET in patients with glioma. J. Nucl. Med. 39,
325–333.
Imahori, Y., Ueda, S., Ohmori, Y., Sakae, K., Kusuki, T.,
Kobayashi, T., Takagaki, M., Ono, K., Ido, T., Fujii, R.,
1998b. Positron Emission Tomography-based boron
neutron capture therapy using boronophenylalanine
for high-grade gliomas: Part II. Clin. Cancer Res. 4,
1833–1841.
Mishima, Y., Kobayashi, T., Kanda, K., Yoshino, K., 1989.
Treatment of malignant melanoma by single thermal
neutron capture therapy with melanoma-seeking 10B-com-
pound. Lancet 12, 388–389.
Ono, K., Masunaga, S., Suzuki, M., Kinashi, Y., Takagaki, M.,
Akaboshi, M., 1999. The combined effect of boronopheny-
lalanine and borocaptate in boron neutron capture therapy
for SCCVII tumors in mice. Int. J. Radiat. Oncol. Biol.
Phys. 43, 431–436.
Slakin, D.N., 1991. A history of boron neutron capture therapy
of brain tumors-postulation of a brain radiation dose
tolerance limit. Brain 114, 1609–1629.
Wessol, D., Cohen, M., Harkin, G., Rossmeier, M., Wemple,
C., Wheeler, F., 1999. INEEL/EXT-99-00766.