Decision Criteria for the Selection Between 3DCRT and VMAT in High-Grade Gliomas Based on the Normal Tissue Complication Probability of Normal Brain

Abstract

Although VMAT has been increasingly used for the radiation of high-grade gliomas as a substitute for 3DCRT, it has not been clear which patients will truly benefit from VMAT over 3DCRT. Here we investigated the normal tissue complication probability (NTCP) of normal brains in order to create the criteria which can be easily used for the selection of the optimal treatment modality. Twenty consecutive patients with high-grade gliomas treated by 3DCRT (60 Gy in 30 fractions) at our hospital between April 2014 and February 2016 were enrolled. First, 1-arc VMAT plans were created based on our treatment protocol for each 3DCRT plan, retrospectively. For both plans, the NTCP for normal brain (i.e., the whole brain minus the GTV) was calculated using a Lyman-Kutcher-Burman model and the following parameters: α/β = 2.9, n = 0.25, m = 0.15, and TD50 = 60 Gy. Second, the patients were divided into two groups according to the NTCPs of the 3DCRT plans for normal brain by the threshold of 5% (<5%-3DCRT-NTCP group and ≥5%-3DCRT-NTCP group). We then compared the NTCPs for the 3DCRT and VMAT plans, and the GTV size between these groups by Wilcoxon rank sum test. For the latter, we used receiver operating characteristic (ROC) curves to identify the cut-off value. Finally, the correlation between the GTV size and the largest cross-sectional diameter of the GTV, and the largest cross-sectional area (largest cross-sectional diameter x largest diameter perpendicular to it) were evaluated, respectively. The difference in the NTCPs for normal brain in the 3DCRT and VMAT plans in the <5%-3DCRT-NTCP group was not significant (median 1.2% [0.2%–4.7%] vs. 0.8% [0.1%–3.0%], p=0.09). In the ≥5%-3DCRT-NTCP group, the NTCPs for the VMAT plans were significantly lower than the NTCPs for the 3DCRT plans (median 9.6% [5.4%–13.2%] vs. 4.0% [2.8%–5.8%], p=0.0011). The ≥5%-3DCRT-NTCP group's GTV size was significantly larger than that of the <5%-3DCRT-NTCP group (median 64.2 cc [14.0 cc–246.4 cc] vs. 149.5 cc [45.4 cc–391.5 cc], p=0.0016), and the cut-off value of the GTV size was 130.5 cc (AUC: 0.93, 95% CI: 0.78–1.0). The GTV size was strongly correlated with the largest cross-sectional diameter of the GTV (R2=0.82) and the largest cross-sectional area (R2=0.94), which produced the cut-off values of 7.5 cm and 41 cm2. Our findings show that VMAT is more appropriate than 3DCRT in cases in which the GTV size is 130.5 cc or larger, which corresponds to the threshold of 7.5 cm in the largest cross-sectional diameter of GTV and to 41 cm2 in the largest cross-sectional area on the diagnostic images.