Journal of Neuro-Oncology

, 91:199

Survival benefit of Boron neutron capture therapy for recurrent malignant gliomas

Authors

    • Department of NeurosurgeryOsaka Medical College
    • Cancer Intelligence Care System, Inc.
  • Shinji Kawabata
    • Department of NeurosurgeryOsaka Medical College
  • Kunio Yokoyama
    • Department of NeurosurgeryOsaka Medical College
  • Toshihiko Kuroiwa
    • Department of NeurosurgeryOsaka Medical College
  • Hiroyuki Michiue
    • Department of NeurosurgeryOkayama University
  • Yoshinori Sakurai
    • Particle Radiation Oncology Research Center, Research Reactor Institute, Kyoto University
  • Hiroaki Kumada
    • Department of Research Reactor and Tandem Accelerator, Nuclear Science Institute, Japan Atomic Energy Agency
  • Minoru Suzuki
    • Particle Radiation Oncology Research Center, Research Reactor Institute, Kyoto University
  • Akira Maruhashi
    • Particle Radiation Oncology Research Center, Research Reactor Institute, Kyoto University
  • Mitsunori Kirihata
    • Department of AgricultureOsaka Prefectural University
  • Koji Ono
    • Particle Radiation Oncology Research Center, Research Reactor Institute, Kyoto University
Clinical study - patient Study

DOI: 10.1007/s11060-008-9699-x

Cite this article as:
Miyatake, S., Kawabata, S., Yokoyama, K. et al. J Neurooncol (2009) 91: 199. doi:10.1007/s11060-008-9699-x

Abstract

We have applied boron neutron capture therapy (BNCT) to malignant brain tumors. Here we evaluated the survival benefit of BNCT for recurrent malignant glioma (MG). Since 2002, we have treated 22 cases of recurrent MG with BNCT. Survival time was analyzed with special reference to recursive partitioning analysis (RPA) classification, by Carson et al. (J Clin Oncol 25:2601–2606, 2007). Median survival times (MSTs) after BNCT for all patients and for glioblastoma as on-study histology at recurrence was 10.8 months (n = 22; 95% CI, 7.3–12.8 months) and 9.6 months (n = 19; 95% CI, 6.9–11.4 months), respectively. In our study, MST for the high-risk RPA classes was 9.1 months (n = 11; 95% CI, 4.4–11.0 months). By contrast, the original journal data showed that the MST of the same RPA classes was 4.4 months (n = 129; 95% CI, 3.6–5.4 months). BNCT showed a survival benefit for recurrent MG, especially in the high-risk group.

Keywords

BNCTBPA–PETGBMMGRPA

Introduction

We have applied a form of tumor-selective particle radiation, boron neutron capture therapy (BNCT), for malignant gliomas (MGs) [1, 2] and malignant meningiomas [3, 4]. BNCT comprises a binary approach [5]: a boron-10 (10B)-labeled compound is administered that delivers high concentrations of 10B to the target tumor relative to the surrounding normal tissues. This is followed by irradiation with thermal neutrons. When neutrons collide into 10B atoms, high linear-energy-transfer (LET) alpha and 7Li particles are released from the 10B (n, alpha) 7Li neutron capture reaction. The short range (5–9 micrometers) of these particles allows for relatively selective tumor killing without significant damage to the adjacent normal brain tissue.

The prognosis of recurrent MGs, especially glioblastoma multiforme (GBM) is poor [6]. We reported the effectiveness of BNCT on neuroimages for MGs [1, 2], and recently reported the survival benefit of BNCT for newly diagnosed MGs [7]. Unfortunately, the standard treatment for recurrent MG has not yet been established. Therefore, evaluation of the survival benefit of BNCT for recurrent MGs is difficult. Also with limited case numbers like this study, it is difficult to elucidate some objective assessments of the survival benefit of BNCT. To evaluate this in low and high-risk group of recurrent MGs, we adopted the recursive partitioning analysis (RPA) classification for recurrent MG advocated by Carson et al. in a 2007 article in the Journal of Clinical Oncology, in which the results of 10 recent protocols of phase-1 and -2 trials applied by the new approaches to brain tumor therapy CNS consortium (NABTT) for recurrent MG were summarized [8]. They included six systemic treatment and four local treatment trials. Originally this RPA classification was not aimed at the evaluation of the effectiveness of each trial for recurrent MG; however, this RPA classification gave us a uniform background and median survival time (MST) for each recurrent MG-type patient at the time of recurrence. So we classified our recurrent MG patients treated by BNCT and compared their survival to the MSTs presented in the above journal.

Patients and methods

Patient enrollment

From 2002 to 2007 we treated a total of 22 cases of recurrent MG using BNCT. Our eligibility criteria for this trial were as follows: (1) age 15 years or older; (2) histologically proven supratentorial MG (GBM, AA, AO, or anaplastic oligodendroglioma, as on-study histology) that had proved to be progressive or recurrent after radiation therapy; (3) depth of the tumor from scalp less than 6 cm (if the lesion is deeper than 6 cm from the scalp, partial removal or cyst evacuation was applied to fit this criteria, see below); (4) no cerebrospinal fluid (CSF) dissemination at recurrence; (5) estimated life expectancy longer than 3 months, not pregnant or breast feeding, and having a KPS score of 60 or greater.

Clinical regimen of BNCT

After the confirmation of the tumor progression or recurrence of the original lesions on MRI, the patients received a BPA–PET to assess the distribution of boronophenylalanine (BPA) [9, 10]. The lesion/normal brain (L/N) ratio of BPA uptake can be estimated from this type of study, and dose planning was performed according to the L/N ratio, as described previously [1, 2]. If the lesions were deeper than 6 cm from the scalp, partial removal of the mass or cyst evacuation was applied. At this procedure, air instillation via an Ommaya reservoir was performed so that the neutron flux would penetrate to the deepest part of the tumor [11]. Within a month after the surgery, BNCT was performed.

In protocol 1, the patients were administered 100 mg/kg of sodium borocaptate (BSH) and 250 mg/kg of BPA for one hour intravenously 12 h prior and just prior to neutron irradiation, respectively. In protocol 2, the patients were administered 100 mg/kg of BSH intravenously for one hour, 12 h prior to neutron irradiation and 700 mg/kg of BPA continuously for 6 h before the irradiation. In both protocols, the neutron irradiation time was determined not to exceed 13 Gy-Eq to the normal brain by simulation. Here, Gy-Eq (Gy: Gray) corresponds to the biologically equivalent X-ray dose that would have equivalent effects on tumors and on the normal brain. For some deep tumors, air instillation was performed as stated above just prior to neutron irradiation.

Patient follow-up

Patients were followed up by bimonthly Gd-enhanced MRI. When the lesions became enlarged or new lesions appeared on the follow-up MRI, we applied BPA–PET to evaluate the tumor activity [12]. If the positron emission tomography (PET) results suggested tumor progression (TP), additional treatments were applied. If PET suggested the high possibility of radiation necrosis (RN), medical treatments for this pathology or surgical resections were applied [12, 13].

Patient characteristics

The patients’ age, gross tumor volume (GTV) (Gd-enhanced lesions on MRI at relapse, use of temozolomide (TMZ) and absorbed dose by BNCT (minimum tumor dose and maximum brain dose) are summarized in Table 1. In 12 cases surgery was applied before BNCT, as a form of cyst evacuation or partial tumor removal to make a cavity to establish an Ommaya reservoir as described above. Ten cases were administered TMZ, three before the relapse and seven after BNCT. Individual information of TMZ usage is listed in Table 2. In Table 2, two histological diagnoses were prepared. One is initial histology and the other is on-study histology. Here, on-study histology means the histology that was confirmed at the last surgery for each patient, prior to BNCT.
Table 1

Patient characteristics

Description

  

Age (median (range))

51

(15–67)

aGTV at the relapse (median (range))

42.0

(4.1–64.5) ml

bReoperated cases at relapse

12

 

TMZ

10

 

    Before BNCT

3

 

    After BNCT

7

 

aGTV was measured on contrast-enhanced MRI at the relapse

bCyst puncture or partial removal to make cavity for air instillation

Table 2

 

Case No

Age

Sex

Histology

RPA by Carson et al.

TMZ

BNCT protocol

Absorbed dose (Gy-Eq)

Survival (months)after BNCT

Cause of death

Initial

On-study

Before

After

Min tumor

Max brain

1

42

M

AA

GB

1

+

1

15.5

12.4

43.1

TP

2

57

F

AA

GB

1

+

2

37.3

8.3

22.0

D

3

15

F

AA

AA

2

+

2

56.3

10.7

33.4

A

4

53

M

Oligo

GB

2

2

73.9

13.2

6.9

D

5

51

M

AOA

AOA

2

2

27.4

8.1

32.4

D

6

33

F

G2

AA

2

1

12.7

7.1

15.0

OC

7

61

M

AA

GB

3

1

34.4

3.7

10.8

RN

8

29

F

AA

GB

3

1

25.7

5.9

9.6

B

9

62

M

AA

GB

3

1

23.4

9.9

2.5

OC

10

31

M

G2

GB

3

1

29.3

14.2

4.4

D

11

51

M

AA

GB

3

+

2

44.9

13.6

9.1

TP

12

48

M

GB

GB

4

1

27.2

11.1

7.8

D

13

46

F

GB

GB

4

+

2

49.2

12.1

12.8

D

14

41

M

GB

GB

4

+

2

54.3

12.7

10.3

D

15

35

M

GB

GB

5

2

37.7

13.4

6.0

D

16

45

M

GB

GB

5

+

2

59.0

13.8

11.4

RN

17

59

M

GB

GB

7

1

32.8

11.2

8.6

TP

18

50

M

GB

GB

7

1

32.6

13.6

15.3

RN

19

63

M

GB

GB

7

2

34.7

9.4

11.0

D

20

67

F

GB

GB

7

+

2

58.0

11.7

12.3

D

21

60

F

GB

GB

7

+

+

2

34.9

7.5

5.8

TP

22

54

M

GB

GB

7

+

2

19.7

10.7

7.4

TP

M, male; F, female; AA, anaplastic astrocytoma; Oligo, oligodendroglioma; AOA, anaplastic oligoastrocytoma; G2, grade 2 astrocytoma; GB, glioblastoma; RPA, Recursive partitioning analysis; TMZ, temozolomode; Gy-Eq, Gray equivalent; Min, minimum; Max, maximum; BNCT, boron neutron capture therapy; TP, local tumor progression; A, alive; D, CSF dissemination; RN, radiation necrosis; OC, other cause; B, both dissemination and local tumor progression

Statistical methods

Continuous data are summarized with medians, ranges and P-values. Univariate analysis was performed using chi-square log-rank testing. Survival distributions (MSTs and 95% CIs) were estimated using the product limit method. The analysis was intention-to-treat and included all eligible patients. Data were entered into Microsoft Excel (Microsoft Corporation) and analyzed using JMP software version 7 (SAS Institute, Cary, NC, USA).

RPA classification

To objectively evaluate the survival benefit of BNCT for recurrent MG, we classified our BNCT cases according to the RPA classification advocated in some journals [8]. These classifications can be summarized as follows: class 1, not GBM (initial histology), KPS ≥ 80, frontal (tumor location); RPA class 2, not GBM, KPS ≥ 80, not frontal; RPA class 3, not GBM, KPS ≤ 70; RPA class 4, GBM, Age ≤ 50, KPS ≥ 90; RPA class 5, GBM, Age ≤ 50, 60 ≤ KPS ≤ 80; RPA class 6, GBM, Age ≥ 50, no steroid use; RPA class 7, GBM, Age ≥ 50, steroid use. Individual class of RPA of our cases treated by BNCT is listed in Table 2.

Analysis of the cause of death after BNCT

Unfortunately, 21 out of the 22 patients died during the observation period, as listed in Table 2. The cause of death was analyzed with the following categories: local TP, CSF dissemination, RN, and other cause of death. These classifications were based on Gd-enhanced MRI, BPA–PET, histology of the surgical specimen and autopsy. In one case, both CSF dissemination and local TP occurred simultaneously and it was impossible to determine a single pathology as the major cause of death.

Results

Survival after BNCT and after diagnosis

Individual histology (initial and on-study at relapse), RPA class, TMZ use, BNCT protocol (1 or 2), absorbed dose by BNCT, survival period after BNCT, and cause of death are summarized in Table 2. Survival after BNCT (n = 22) and that from initial GBM diagnosis (n = 19, on-study histology as GBM) are shown in Fig. 1. MST after BNCT for all patients (n = 22) was 10.8 months (95% CI, 7.3–12.8 months). MST after BNCT for GBM cases as on-study histology at recurrence (n = 19) was 9.6 months (95% CI, 6.9–11.4 months). MST after initial GBM diagnosis (n = 19) was 19.1 months (95% CI, 11.6–23.0 months).
https://static-content.springer.com/image/art%3A10.1007%2Fs11060-008-9699-x/MediaObjects/11060_2008_9699_Fig1_HTML.gif
Fig. 1

Kaplan–Meier survival curves for recurrent MG cases treated by BNCT. The continuous line shows the survival of all patients after BNCT ( = 22). The broken line shows the survival of GBM (on-study histology) after diagnosis of GBM ( = 19)

Survival with special reference to RPA classes

The MSTs (months) of our BNCT cases classified according to RPA classes are shown in Table 3 and compared in each case with the values from Carson et al.: Class 1 (n = 2): 32.6 vs. 25.7 (Carson et al.), Class 2 (n = 4): 23.7 vs. 17.2, Class 3 (n = 5): 9.1 vs. 3.8, Class 4 (n = 3): 10.2 vs. 10.4, Class 5 (n = 2): 8.5 vs. 6.4, Class 7 (n = 6): 9.8 vs. 4.9. The tendencies in patient survival of our cases after BNCT were very similar to those of the original report in terms of RPA classification. Since our cases were so limited in number, we joined the worst prognosis classes (Class 3 and 7) together into one class. The MST of our cases in this combined class was 9.1 months (n = 11; 95% CI, 4.4–11.0 months), while that in Carson et al. was 4.4 months (n = 129; 95% CI, 3.6–5.4 months).
Table 3

Comparison of NABTT trials and our BNCT series

 

All patients

RPA 3 + 7

MST

95% CI

Number in series

MST

95% CI

Number in series

NABTT

7.0

6.2–8.0

n = 310

4.4

3.6–5.4

n = 129

BNCT

10.8

7.3–12.8

n = 22

9.1

4.4–11.0

n = 11

aNew Approaches to brain tumor therapy CNS Consortium; 10 phase-1 and -2 trials reported by Carson et al. (J Clin Oncol 25:2601–2606, 2007)

MST, Median survival time; CI, confidence interval

Cause of death after BNCT

We lost 21 cases out of 22. The causes of deaths were CSF dissemination (10 cases), local TP (5), both (1), RN (3), and other (2), as shown in Table 2. With regard to RN, we discuss more extensively in “Discussion”.

Adverse effects of BNCT

No serious adverse effects were observed both in protocols 1 and 2 in this study of BNCT for recurrent MGs, even though all patients were applied with radiotherapy previously. Hematuria was reported in the literature using large amounts of BPA in BNCT [14]. Fortunately, we did not experience this adverse effect at all, however, three cases in protocol 2 showed transient decrease volume and turbidity of urine and fever during the first 24 h after BNCT. We concluded these side effects were caused by recrystallization of BPA in urine. Thereafter, we over hydrated the remaining patients after BNCT, and no such side effects were observed again.

Univariate analysis for the survival after BNCT

In Table 4, we analyzed factors for survival after BNCT, such as sex, age, TMZ or steroid usage, KPS, minimum absorbed dose in tumors, initial histology, GTV at the relapse, BNCT protocol (1 or 2) and RPA classes. Among them, only RPA class (RPA class 3 and 7 or others) showed a statistical significant influence on survival after BNCT.
Table 4

Univariate analysis of factors for survival after BNCT

Factor

Group

Survival (months)

P-Value

Median

95% CI

 

Sex

Male (n = 15)

9.1

6.0

11.0

P = 0.2456

 

Female (n = 7)

12.8

5.8

22.0

 

Age

≦50 (n = 11)

11.4

6.0

15.3

P = 0.2482

 

>50 (n = 11)

9.1

5.8

12.3

 
 

≦57 (n = 16)

11.4

7.4

15.3

P = 0.0982

 

>57 (n = 6)

10.8

2.5

  

KPS

≦80 (n = 13)

9.6

6.0

11.4

P = 0.1271

 

>80(n = 9)

12.8

5.8

  

Initial Histology

GBM (n = 11)

10.3

6.0

12.3

P = 0.1329

 

Not GBM (n = 11)

10.8

4.4

32.4

 

TMZ

Used (n = 10)

12.3

5.8

22.0

P = 0.1468

 

Not used (n = 12)

9.6

4.4

15.0

 

Steroid

Used (n = 13)

9.6

6.9

11.4

P = 0.1445

 

Not used (n = 9)

12.8

2.5

  

GTV (ml)

≦37.2(n = 11)

9.1

4.4

12.8

P = 0.5273

 

>37.2 (n = 11)

10.8

7.4

15.3

 

Minimum tumor

≦34.0(n = 12)

9.6

2.5

15.3

P = 0.9110

Dose (Gy-Eq)

>34.0(n = 10)

11.0

6.0

12.8

 
 

≦37.0(n = 13)

9.6

5.8

15.0

P = 0.6548

 

>37.0(n = 9)

11.4

6.0

22.0

 

BNCT protocol

1 (n = 9)

9.6

2.5

15.3

P = 0.8184

 

2 (n = 13)

11.0

6.9

12.8

 

RPA class

RPA 3&7 (n = 11)

9.1

4.4

11.0

P = 0.0216

 

RPA not 3&7 (n = 11)

12.8

6.9

32.4

 

Representative case

A 48-year-old man with a right temporal mass was operated emergently for consciousness disturbance in a hospital. The operation was partial tumor removal and histological diagnosis was GBM. He received fractionated X-ray radiation therapy (XRT) with a total dose of 80 Gy and chemotherapy consisting of nimustine and vincristine. Even during the radiotherapy, the tumor continued to enlarge, and the patient was referred to our institute for BNCT (Fig. 2 a, a′). He was classified as RPA class 4. The BNCT was performed with the minimum tumor absorbed dose of 27.2 Gy-Eq, and maximum brain absorbed dose of 11.1 Gy-Eq. One week after BNCT the mass shrunk rapidly (Fig. 2 b, b′). Three months after BNCT, the original mass became enlarged in Gd-MRI. He was operated on again. The histology was mainly necrosis with small pocket of residual tumor cells. He was well for another 4 months. We lost this case 7.8 months after BNCT and 13.5 months after initial surgery, due to CSF dissemination (Fig. 2 c, c′). This is a representative case of recurrent MG treated by BNCT, with regard to the rapid tumor shrinkage after BNCT and the occurrence of radiation necrosis and CSF dissemination as the cause of death.
https://static-content.springer.com/image/art%3A10.1007%2Fs11060-008-9699-x/MediaObjects/11060_2008_9699_Fig2_HTML.jpg
Fig. 2

A representative case of recurrent GBM treated by BNCT. (a, a′) MRI, prior to BNCT. Gd-enhanced lesions were at the right temporo-occipital lobe; (b, b′) MRI, 48 h after BNCT. Marked shrinkage of the lesions was recognized; (c, c′) MRI, 7 months after BNCT. CSF dissemination was prominent

Discussion

Here we reported the survival benefit of BNCT for recurrent MG cases, mainly GBM. The MST after BNCT for GBM cases as on-study histology at recurrence (n = 19) was 9.6 months (95% CI, 6.9–11.4 months). In the literature, we found a summary of a large series of eight phase-2 trials of chemotherapies for recurrent GBM cases [15]. In this report, the authors mentioned the MST of GBM after relapse as 25 weeks (5.8 months; 95% CI, 21–28 weeks, 4.9–6.5 months; n = 225). In comparison with this result, our data for the survival benefit of BNCT in recurrent GBM was not bad.

As to BNCT for recurrent GBM, two small series have been reported in the literature. A Swedish group and a Finnish group reported that MSTs for recurrent GBM after BNCT were 8.7 (n = 12) [16] and 7.5 months (n = 7) [17], respectively. Our data in the current report is almost equal to/somewhat better than the findings in these reports.

Kaplan–Meyer analysis in Fig. 1 showed that MST after BNCT for all patients (n = 22) was 10.8 months (95% CI, 7.3–12.8 months). We are not sure whether this result is reliable, as this is the result of a small series from a single institute. To evaluate the survival benefit of BNCT in low and high-risk group of recurrent MGs, we applied RPA to our cases as advocated in the literature [8]. Inclusion criteria for our trial and the 10 NABTT phase-1 and -2 trials reported in Carson et al. were not very different. Our case numbers for each RPA class were so limited, however, that the MST of our cases in each RPA class were relatively better in comparison with original NABTT results, as listed above. In the original article, RPA class 3 (Not GBM, KPS ≤ 70) and class 7 (GBM, Age ≥ 50, steroid use) showed extremely poor prognosis (supplementary Table 1). The MST of our combined class 3 and class 7 cases was 9.1 months (n = 11; 95% CI, 4.4–11.0 months), while that in the original article was 4.4 months (n = 129; 95% CI, 3.6–5.4 months). We cannot know whether our current MST data is significantly better than that of each NABTT trial because their raw data were not available. But at least, BNCT showed a good survival benefit even for the highest-risk group, RPA class 3 and 7.

TMZ is the sole promising drug for GBM so far. A Swedish BNCT group reported potential TMZ effects with combination of BNCT at the relapse of GBM [16]. However, in our univariate analysis, TMZ did not contribute prominently to the prolongation of survival in our series (Table 4). In our 22 cases, we used TMZ in 10 cases, before BNCT in 3 cases (Cases 11, 14 and 16) and after BNCT in 7 (Cases 1, 2, 3, 13, 20, 21 and 22). For the former three cases, TMZ could not control the tumor growth and methylation-specific PCR showed an unmethylated O6-methylguanine DNA methyltransferase (MGMT) promoter [18] (data not shown). We stopped the administration of TMZ after BNCT as we judged TMZ was not efficacious for these three cases. Among the latter seven cases, only two (Cases 1 and 2, both classified as RPA class 1) showed methylated promoter status for MGMT, with good prognoses. For the other five cases, we were not sure of the MGMT expression status of the tumor. In the high-risk group in our series (RPA class 3 and 7), three cases were administered TMZ after BNCT (Cases 20, 21 and 22). Among them, Case 21 and 22 showed a relatively short survival after BNCT. We do not deny the meaning of TMZ use at relapse; however, in our series for this high-risk group, the survival benefit of TMZ was limited. In the literature, TMZ has actually shown modest survival benefit at relapse of recurrent GBM [19]. Brada et al. reported only 5.4 months prolongation as MST with TMZ at relapse in the report.

There are several reports with relatively good results for recurrent MG, with an MST of around 10 months after the stereotactic radiosurgery (SRS) [20] or stereotactic radiotherapy (SRT) [21] at relapse. However, there was big difference in GTV at the relapse between these SRS or SRT cases and ours. The median GTV of the former two was 10.1 and 12.7 ml, while the median GTV of our cases was 42.0 ml. There might also be a difference as to performance status or age between the SRS or SRT reports and our cases. The result of re-irradiation for recurrent GBM was poor [22]. The MST of this report was 26 weeks after the treatment. In addition, BNCT can be applied in only one day. Taken together, BNCT could be one of the promising radiation treatment options for recurrent MG at relapse.

We lost many cases of recurrent MGs after BNCT by CSF dissemination, as we reported (in preparation) and as shown in Table 2 and Fig. 2. In other words, local control by BNCT for even recurrent MG was fairly good. There was a tendency for CSF dissemination to occur in relatively long-term survivors from diagnosis (data not shown). On the other hand, a major problem in BNCT for recurrent MG was the occurrence of RN. We experienced RN by BNCT especially for recurrent MG, because the patients had been treated by radiotherapy prior to BNCT. Although BNCT is cell-selective particle radiation, some particle dose is inevitably absorbed by the normal brain tissue as shown in Table 2. The diagnosis of this pathology is difficult; however, amino acid PET may give us good clue for it, as stated above [12]. Most of RN could be controlled with medical or surgical treatments as above; however, we lost three cases by RN in our series. Preventive medical treatments such as by anticoagulants or by vitamin E must be considered after BNCT, especially for recurrent cases. This is not mentioned in other BNCT reports for recurrent MG [16, 17]; however, it should be seriously considered. In Swedish reports of BNCT for recurrent GBM, the authors mentioned a median time to tumor progression of 6 months after BNCT, but there was no statement as to how TP was judged in their report. It is very difficult to differentiate RN and TP on MRI, especially with high-dose radiation treatment. So we did not apply the analysis of time to tumor progression in our series. In univariate analysis (Table 4), there was no correlation of minimum tumor dose by BNCT and survival after BNCT. Especially for recurrent cases, if we increase the minimum tumor dose by BNCT, the incidence of RN probably increases, as discussed here. Therefore, it is very difficult to elucidate the most suitable dose of BNCT at relapse. Regardless, RN is a serious problem to be overcome in the field of BNCT.

XRT plus concomitant TMZ (Stupp’s regimen) has been the global standard so far for newly diagnosed GBM [23]. Pellettieri et al. reported that BNCT at relapse after Stupp’s regimen might be the best treatment of GBM [16]. Also in our series BNCT at relapse showed a good MST after the initial GBM diagnosis of 19.1 months (n = 19; 95% CI, 11.6–23.0 months). But it cannot be concluded so easily that BNCT at relapse after Stupp’s regimen is the best for the treatment of GBM because 19 cases in our series were referred to our institute at relapse with a significant interval after initial treatments. This interval might prolong the survival after initial GBM diagnosis at a glance.

In summary, the RPA classification advocated by Carson et al. predicted the patient survival trends of our BNCT series; however, BNCT showed the most prominent survival benefit in the high-risk group (RPA classes 3 and 7).

Acknowledgments

This work was partly supported by Grants-in-Aid for Scientific Research (B) (16390422 and 19390385) from the Japanese Ministry of Education, Science and Culture, by a Grant-in-Aid for Scientific Research from the Ministry of Health, Labor and Welfare of Japan to S–I.M. (P·I., Hideki Matsui) and by the Regional Science Promotion Program of the Japan Science and Technology Corporation, as well as by the “Second-term Comprehensive 10-Year Strategy for Cancer Control” of the Ministry of Health, Labor, and Welfare of Japan to S–I.M. This work was also supported in part by the Takeda Science Foundation for Osaka Medical College, by a Grant-in-Aid for Cancer Research from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (12217065) to K. O., and by Grants-in-Aid for Scientific Research by young researchers (B) (18791030) from the Japanese Ministry of Education, Science, and Culture to S. K. The top two authors contributed equally in this study as primary co-investigators.

Supplementary material

11060_2008_9699_MOESM1_ESM.doc (34 kb)
MOESM1 (DOC 33 kb)

Copyright information

© Springer Science+Business Media, LLC. 2008