Keywords:
Radiotherapy techniques, Biological effects, Radiobiology, Radiation therapy / Oncology, Cone beam CT, Radiation physics, Head and neck, Computer applications
Authors:
W. Harriss-Phillips, E. Bezak; ADELAIDE/AU
DOI:
10.1594/ranzcr2014/R-0162
Conclusion
Discussion
Virtual well oxygenated tumours were controlled with mean EQD2’s of 69±11Gy,
an expected and uniform dose value for all SBRT schemes analysed.
BED conversions also had uniform outcomes(83±11 Gy) across the schedules.
Assessing total cell death compared to having 1 or 5 cells remain,
bordered on significance and required approx.
5 Gy less dose. These cells may be controlled by natural immune function and hence tumour control still achieved,
however this hypothesis is beyond the scope of this work.
Hypoxic SBRT simulations required,
on average,
an extra +28±4 Gy,
compared to oxic simulations.
With reoxygenation applied,
this reduced to approximately +15 Gy for the higher doses per fraction. However,
reoxygenation only had independent significance for hypoxic tumours when the dose per fraction was ≤9 Gy per fraction (RT over ≥2 weeks).
Delivering 12 and 15 Gy per fraction was potent enough to kill all living oxic and hypoxic cells during the last delivered fraction.
Reoxygenation was assumed to be dependent on current tumour volume in this study.
Converting hypoxic total control dose predictions into EQD2 values did not have uniform results (Figure 5) with dose per fraction,
illustrating the usefulness of complex MC algorithms for RT response predictions when implementing dynamic effects,
such as oxygenation.
LQC utilisation was only (just) significant for the higher doses per fraction,
as expected,
based on Figure 3 & 6 data.
However this has not been validated clinically,
in fact the opposite effect may apply,
if the theory of vessel ablation at higher doses per fraction (and hence increased necrotic cell kill) is proven valid.
Simulating endothelial cell death however was beyond the scope of HYP-RT in this work.
Final Conclusions
YES – Tumour HYPOXIA and REOXYGENATION MATTER in SBRT
SBRT simulations had uniform EQD2 tumour control outcomes for oxic tumours.
More complex hypoxia and reoxygenation simulations showed that: i) much more dose is needed for hypoxic tumours (28±8 Gy),
ii) reoxygenation reduces this extra dose by approximately 50% for schedules delivering 9 Gy/# or less,
and ii) conversion of dose predictions to EQD2 had unpredictable results,
increasing with dose per fraction.
Using a DL parameter bordered on significance for 9 Gy/# or more.
Future simulation will bring to light fraction per week effects on cell kill efficacy.