Risk of radiation-induced second malignant neoplasms from photon and proton radiotherapy in paediatric abdominal neuroblastoma

Abstract

Background and purpose: State-of-the-art radiotherapy modalities have the potential of reducing late effects of treatment in childhood cancer survivors. Our aim was to investigate the carcinogenic risk associated with 3D conformal (photon) radiation (3D-CRT), intensity modulated arc therapy (IMAT) and pencil beam scanning proton therapy (PBS-PT) in the treatment of paediatric abdominal neuroblastoma.

Materials and methods: The risk of radiation-induced second malignant neoplasm (SMN) was estimated using the concept of organ equivalent dose (OED) for eleven organs (lungs, rectum, colon, stomach, small intestine, liver, bladder, skin, central nervous system (CNS), bone, and soft tissues). The risk ratio (RR) between radiotherapy modalities and lifetime absolute risks (LAR) were reported for twenty abdominal neuroblastoma patients (median, 4y; range, 1-9y) historically treated with 3D-CRT that were also retrospectively replanned for IMAT and PBS-PT.

Results: The risk of SMN due to primary radiation was reduced in PBS-PT against 3D-CRT and IMAT for most patients and organs. The RR across all organs ranged from 0.38 ± 0.22 (bladder) to 0.98 ± 0.04 (CNS) between PBS-PT and IMAT, and 0.12 ± 0.06 (rectum and bladder) to 1.06 ± 0.43 (bone) between PBS-PT and 3D-CRT. The LAR for most organs was within 0.01-1% (except the colon) with a cumulative risk of 21 ± 13%, 35 ± 14% and 35 ± 16% for PBS-PT, IMAT and 3D-CRT, respectively.

Conclusions: PBS-PT was associated with the lowest risk of radiation-induced SMN compared to IMAT and 3D-CRT in abdominal neuroblastoma treatment. Other clinical endpoints and plan robustness should also be considered for optimal plan selection.

Keywords: Abdominal neuroblastoma; Childhood cancer; Intensity modulated arc therapy (IMAT); Paediatric; Pencil beam scanning; Proton therapy; Second malignant neoplasm.

Fig. 1

Example of (a) CT and overlaid anatomy and doses distributions for 3D-CRT, IMAT and PBS-PT plans, (b) volume rendering of the eleven organs included in the risk assessment analysis (lungs, rectum, colon, stomach, small intestine, liver, bladder, skin, central nervous system (CNS), bone, and soft tissues), and (c) dose-volume histograms (DVHs) for selected organs. Note that the vertebra is homogeneously irradiated to reduce growth asymmetry. Colour version online.

Fig. 2

Dose-response relationship for the eleven organs considered. Note that the rectum and skin curves overlap since both used a linear dose–response relationship. Colour version online.

Fig. 3

Boxplots of organ equivalent doses (OEDs) for 3D-CRT, IMAT and PBS-PT plans. Outliers (diamonds) fall outside the ± 2.7std range.

Fig. 4

Boxplot of (a) the risk ratios (RRs) and (b) ratio of mean organ doses between modalities for all subjects included. Outliers (diamonds) fall outside the ± 2.7std range. Note that the data is similar in both plots for rectum and skin, as their organ equivalent doses (OED) were also calculated using a linear dose–response relationship.

Fig. 5

Boxplot of the Lifetime Attributable Risk (LAR) for 3D-CRT, IMAT and PBS-PT plans. Asterisks indicate statistically significant results (p < 0.05, Wilcoxon signed-rank test). Outliers (diamonds) fall outside the ± 2.7std range.