A feasibility study of functional preservation in craniospinal irradiation with photon for pediatric medulloblastoma

TL;DR

This study introduces a new radiation technique for treating pediatric medulloblastoma that reduces damage to critical brain and body structures, potentially lowering long-term side effects.

Contribution

A novel functional preservation craniospinal irradiation technique was developed and tested to spare multiple critical organs while maintaining target coverage.

Findings

FP-CSI significantly reduced radiation doses to the hippocampus, hypothalamus, and pituitary gland compared to standard CSI.

FP-CSI lowered predicted risks of neurocognitive impairment and endocrine dysfunction by over 60% and 40%, respectively.

Despite slightly reduced target conformity, FP-CSI maintained clinically acceptable PTV coverage and accurate dose delivery.

Abstract

Craniospinal irradiation (CSI) is essential for treating pediatric medulloblastoma (MB) but causes significant long‐term toxicities. Existing dose‐reduction or partial‐sparing strategies improve neurocognitive outcomes but may compromise survival or fail to address other late effects.

A new functional preservation CSI (FP‐CSI) technique was developed to spare the hippocampus, hypothalamic‐pituitary axis (HPA), cochlea, and scalp while ensuring homogeneous vertebral coverage. Eight pediatric patients with average‐risk MB were retrospectively planned with volumetric modulated arc therapy (VMAT) using both FP‐CSI and standard CSI (S‐CSI). Dosimetric parameters for the planning target volume (PTV) and organs at risk (OARs), radiobiological effects, plan robustness, plan complexity, and plan quality assurance (QA) were compared.

FP‐CSI significantly reduced mean doses to the hippocampus (12.4 vs. 23.9 Gy), hypothalamus (14.7 vs. 23.9 Gy), and pituitary gland (15.4 vs. 24.1 Gy, all p < 0.01). Vertebral dose gradients were halved (4.7 vs. 8.7 Gy). Moderate dose reductions were also achieved for the cochlea and scalp. Compared with S‐CSI, FP‐CSI exhibited slightly inferior PTV homogeneity (HI: 0.16 vs. 0.07) and conformity (CI: 0.88 vs. 0.93), but coverage remained clinically acceptable. Normal tissue complication probability (NTCP) modeling showed pronounced decreases in predicted neurocognitive and endocrine toxicity risks, with probability of neurocognitive impairment reduced from 84.5% to 24.9% and probability of endocrine dysfunction from 44.7% to 27.3%. FP‐CSI increased modulation complexity and produced slightly lower gamma passing rates for cranial beams, while spinal beam deliverability remained similar to S‐CSI. Robustness analysis indicated greater sensitivity of FP‐CSI to setup and rotational errors. Nevertheless, 3D dose reconstruction confirmed accurate delivery, with volumetric dose deviations generally below 1 Gy.

FP‐CSI effectively spares critical functional structures while maintaining clinically acceptable target coverage, and offers a promising strategy to reduce long‐term radiotherapy‐induced toxicities in pediatric MB.

Figures