Dosimetric Comparison of Upfront Boosting With Stereotactic Radiosurgery Versus Intraoperative Radiotherapy for Glioblastoma.

PURPOSE

To simulate and analyze the dosimetric differences of intraoperative radiotherapy (IORT) or pre-operative single-fraction stereotactic radiosurgery (SRS) in addition to post-operative external beam radiotherapy (EBRT) in Glioblastoma (GB).

METHODS

Imaging series of previously treated patients with adjuvant radiochemotherapy were analyzed. For SRS target definition, pre-operative MRIs were co-registered to planning CT scans and a pre-operative T1-weighted gross target volume (GTV) plus a 2-mm planning target volume (PTV) were created. For IORT, a modified (m)GTV was expanded from the pre-operative volume, in order to mimic a round cavity as during IORT. Dose prescription was 20 Gy, homogeneously planned for SRS and calculated at the surface for IORT, to cover 99% and 90% of the volumes, respectively. For tumors > 2cm in maximum diameter, a 15 Gy dose was prescribed. Plan assessment was performed after calculating the 2-Gy equivalent doses (EQD2) for both boost modalities and including them into the EBRT plan. Main points of interest encompass differences in target coverage, brain volume receiving 12 Gy or more (V), and doses to various organs-at-risk (OARs).

RESULTS

Seventeen pre-delivered treatment plans were included in the study. The mean GTV was 21.72 cm (SD ± 19.36) and mGTV 29.64 cm (SD ± 25.64). The mean EBRT and SRS PTV were 254.09 (SD ± 80.0) and 36.20 cm (SD ± 31.48), respectively. Eight SRS plans were calculated to 15 Gy according to larger tumor sizes, while all IORT plans to 20 Gy. The mean EBRT D was 97.13% (SD ± 3.48) the SRS D 99.91% (SD ± 0.35) and IORT D 83.59% (SD ± 3.55). Accounting for only-boost approaches, the brain V was 49.68 cm (SD ± 26.70) and 16.94 cm (SD ± 13.33) (p<0.001) for SRS and IORT, respectively. After adding EBRT results respectively to SRS and IORT doses, significant lower doses were found in the latter for mean D of chiasma (p=0.01), left optic nerve (p=0.023), right (p=0.008) and left retina (p<0.001). No significant differences were obtained for brainstem and cochleae.

CONCLUSION

Dose escalation for Glioblastoma using IORT results in lower OAR exposure as conventional SRS.