Proton beam therapy reduces the incidence of acute haematological and gastrointestinal toxicities associated with craniospinal irradiation in pediatric brain tumors

Figures & data

Table I. Patient characteristics.

Characteristic	PhBCSI (n = 13)	PrBCSI (n = 30)	p-value
Age (years)	11 (3–18)	10 (2–18)	0.254
Sex			0.619
Male	8 (62)	16 (53)
Female	5 (39)	14 (47)
Histology
Medulloblastoma	4 (31)	9 (30)
Mixed germ cell tumours	3 (23)	5 (17)
Germinoma	1 (8)	6 (20)
Non-germinomatous germ cell tumours	1 (8)	3 (10)
Other histology	2 (15)	6 (20)
Treatment aim			1.000
Prophylactic	9 (69)	22 (73)
Leptomeningeal seeding or recurrent tumour	4 (31)	8 (27)
Chemotherapy before CSI	10 (77)	26 (87)	0.655
CSI dose (CGE or Gy)	32.1 (23.4–39.6)	29.4 (19.8–39.6)	0.236
≤ 23.4	4 (31)	13 (43)
> 23.4–≤ 33	2 (15)	5 (17)
> 33–36	2 (15)	9 (30)
> 36–≤ 39	4 (31)	0 (0)
> 39	1 (8)	3 (10)
Dose per fraction (CGE or Gy)			0.023
1.5	3 (23)	0 (0)
1.8	10 (77)	30 (100)
Dose applied to the primary site (CGE or Gy)	53.2 (39.6–60.6)	51.8 (30.6–61.2)	0.858

Values are presented as the mean (range) or number (per cent).

CGE, cobalt grey equivalent; CSI, craniospinal irradiation; PhBCSI, photon beam craniospinal irradiation; PrBCSI, proton beam craniospinal irradiation.

Figure 1. Representative examples of dose distribution for the craniospinal irradiation method. (A) Sagittal view (left: proton beam therapy; right: photon beam therapy). (B) Axial view (top: proton beam therapy, bottom: photon beam therapy). (C) Dose-volume histograms for the brain and spinal planning target volumes, and of specific tissues of interest (solid line: proton beam therapy, dashed line: photon beam therapy). PTV, planning target volume.

Table II. Acute toxicities.

Toxicity	PhBCSI (n = 13)	PrBCSI (n = 30)	p-value
Leukopenia			0.069
Grade 3	6 (46)	14 (57)
Grade 4	4 (31)	2 (7)
G-CSF administration	4 (31)	12 (40)	0.655
Anaemia			0.493
Grade 3	2 (15)	0 (0)
Grade 4	0 (0)	0 (0)
RBC transfusion	5 (39)	15 (50)	0.486
Thrombocytopenia			0.012
Grade 3	4 (31)	6 (20)
Grade 4	3 (23)	1 (3)
Platelet transfusion	6 (46)	5 (17)	0.042
Nausea	6 (46)	10 (33)	0.424
Dysphagia	2 (15)	14 (47)	0.086
Anorexia	4 (31)	11 (37)*	1.000
Skin disorders	4 (31)	11 (37)	1.000
Vomiting	4 (31)	9 (30)*	1.000
Neurological disorders	3 (23)	4 (13)	0.655
Diarrhoea	3 (23)*	0 (0)	0.023
Ophthalmic disorders	1 (8)	2 (7)	1.000
Cough	1 (8)	2 (7)	1.000

Values are presented as number (percent).

G-CSF, granulocyte colony-stimulating factor; PhBCSI, photon beam craniospinal irradiation; PrBCSI, proton beam craniospinal irradiation; RBC, red blood cell.

*Non-haematological toxicities were classified as Grade 1 or 2, except in three grade 3 patients labelled with an asterisk.

Figure 2. (A,B) Changes in thrombopoietin levels (A) and platelet counts (B) during craniospinal irradiation. (C,D) Changes in thrombopoietin (C) and platelet counts (D) in five patients who did not undergo chemotherapy after craniospinal irradiation. Values in A and B are means with 95% confidence intervals. Values in C and D are the actual values in each patient.

Table III. Changes in haematological parameters from before starting CSI to 1 month after treatment.

	PhBCSI (n = 13)	PrBCSI (n = 30)	p-value
White blood cell count (K/μl)	−2.61 ± 2.27	−0.57 ± 2.22	0.009
Haemoglobin (g/dl)
Uncorrected	−0.70 ± 1.89	+ 0.23 ± 1.04	0.115
Corrected for transfusion*	−1.16 ± 2.06	−0.57 ± 1.48	0.294
Platelet count (× 10^5 cells/μl)
Uncorrected	−1.37 ± 0.96	−0.49 ± 0.64	0.008
Corrected for transfusion*	−2.74 ± 2.28	−0.68 ± 0.72	0.007

Values are presented as the mean ± standard deviation change in each parameter from before starting CSI to 1 month after completing CSI.

PhBCSI, photon beam craniospinal irradiation; PrBCSI, proton beam craniospinal irradiation.

*A standardised unit represents one pack of red blood cells or platelets in adults, adjusted for body weight. We assumed that transfusion of 1 U increased the haemoglobin level by 1 g/ml, and the platelet count by 12 000 cells/μl.