The use of risk estimation models for the induction of secondary cancers following radiotherapy

Figures & data

Figure 1. The difference between risk prediction of the linear model (dashed line) and the LQ-based competition model for single doses (grey solid line) and fractionated irradiations (black solid line). The linear model appears to be valid only for organ doses below 1 Gy. (α₁=0.05 Gy⁻¹, α₂=0.25 Gy⁻¹, α/β = 5 Gy)

Figure 2. Competitive risk model and the irradiation of the lung at a breast treatment with 50 Gy in 25 fractions. Upper panel-the dose response curve of the competitive model for the risk of secondary cancers induction (Equation 3) in the lung. Lower panel-the dose volume histograms for the irradiation of the lung. (α₁=0.017 Gy⁻¹, α₂=0.25 Gy⁻¹, α/β = 4.5 Gy)

Figure 3. Competitive risk model and the irradiation of the bladder at a prostate treatment with 78 Gy in 39 fractions. Upper panel-the dose response curve of the competitive model for the risk of secondary cancers induction (Equation 3) in the bladder. Lower panel-the dose volume histograms for the irradiation of the bladder. (α₁=0.006 Gy⁻¹, α₂=0.25 Gy⁻¹, α/β = 5.4 Gy)

Figure 4. Competitive risk model and the irradiation of the rectum at a prostate treatment with 78 Gy in 39 fractions. Upper panel-the dose response curve of the competitive model for the risk of secondary cancers induction (Equation 3) in the rectum. Lower panel-the dose volume histograms for the irradiation of the rectum. (α₁=0.017 Gy⁻¹, α₂=0.25 Gy⁻¹, α/β = 4 Gy)

Table I.  Risk predictions from different risk models and different methods to deal with dose heterogeneity throughout the irradiated organ and comparisons with clinically observed results.

Organ	Lung	Bladder	Rectum
Average organ dose	1.8 Gy	32.2 Gy	45.7 Gy
Linear risk model	3.00%	19.3%	77.7%
Average dose and competition model	1.95%	0.003%	<0.001%
Full dose distribution and competition model	0.67%	0.29%	0.002%
Clinically observed risk	0.9%	0.15-0.32%	0.05–0.20%
Average organ dose	10 Gy	>30 Gy	–
Reference	Inskip et al. (1994)	Boice et al. (1985)	Boice et al. (1985)