Cerebellar volume reduction after photon or proton radio(chemo)therapy of glioblastoma patients


Cerebellar volume reduction after photon or proton radio(chemo)therapy of glioblastoma patients

Raschke, F.; Seidlitz, A.; Platzek, I.; Beuthien-Baumann, B.; van den Hoff, J.; Krex, D.; Kotzerke, J.; Jentsch, C.; Baumann, M.; Krause, M.; Troost, E. G. C.

Purpose or Objective:

Radio(chemo)therapy is part of the standard treatment of high-grade glioma patients and has been associated with cerebral atrophy [1,2]. Preclinical work also suggests radiation induced atrophy of the cerebellum [3,4]. Investigating cerebellar atrophy in patients treated with radiation is a further step in understanding radiation-induced deficits in both motor function and cognition. The aim of this study was to investigate cerebellar volume changes in a cohort of glioblastoma patients treated with photon or proton radio(chemo)therapy.

Material and Methods:

Data was acquired on a 3T Philips Ingenuity TF PET/MRI scanner (Philips Healthcare, Best, The Netherlands) as part of a prospective, longitudinal study investigating the effect of 11C-methionine PET/MR for tailoring the treatment of patients with glioblastoma (NCT01873469). In total, 71 patients with cerebral GBM (21 treated with proton therapy) had a baseline MR and at least one follow-up MRI, obtained in 3 monthly intervals after irradiation, available, including 3D T1-weighted (T1w) imaging (1×1×1 mm3) before and after intravenous injection of contrast agent (CE). Patients were treated with a total dose of 60 Gy(RBE=1.1) delivered in 2Gy fractions. On average 3.6 follow-ups were available covering a time period of 413 days ± 432 days (mean ± SD).

The cerebellum was cut out from each MRI by warping [5] the MNI152 brain atlas and a corresponding cerebellar mask to each brain extracted T1w MRI. Sigmoid and transverse sinuses mimicking cerebellar tissue were removed using the CE T1w MRI. The cerebellum was segmented into grey matter (GM), white matter (WM) and cerebrospinal fluid (CSF) [6]. Its volume was calculated as the sum of all GM and WM probabilities and normalized to the baseline value for each patient. The mean relative cerebellar volume change per year was estimated for each patient with a linear regression. The resulting rate of volume change per year was plotted against the mean dose delivered to the cerebellum across all patients, and the rate of volume change per year per dose was subsequently estimated using a linear regression.

Results:

Figure 1 illustrates segmentation of the cerebellum. Mean cerebellar dose for patients treated with protons and photons was 1.7Gy ± 2.1Gy and 5.7Gy ± 4.6Gy, respectively. The linear model estimated a cerebellar volume loss of approx. 1.8% per 10Gy per year (p<0.001; Fig. 2). This is similar to that estimated in the cerebrum of the same patient cohort [2].

Conclusion:

Cerebellar volume loss after radio(chemo)therapy is related to the mean cerebellar dose. The atrophy rate is similar to that previously found in the cerebrum. Additional work is needed to further validate those findings and relate them to cognitive and motor performance.

[1] Prust et al. Neurology 2015;85:683-691
[2] Petr et al. Radiother Oncol 2018;128:121-127
[3] Zhou et al. Sci Rep-UK 2017;7:46181
[4] Eekers et al. Clin Transl Radiat Oncol 2018;8:22-26
[5] Avants et al. NeuroImage 2011;54:2033–2044
[6] Avants et al. Neuroinformatics 2011;9:381-400

Keywords: radiotherapy; cerebellum; atrophy; proton therapy; photon therapy; irradiation; normal tissue

Involved research facilities

  • PET-Center
  • Contribution to proceedings
    ESTRO 2020, 03.-07.04.2020, Wien, Österreich
    Proceedings of ESTRO 2020

Permalink: https://www.hzdr.de/publications/Publ-29808