GI Toxicity After Radiotherapy for Gynecological Malignancies
GI Toxicity After Radiotherapy for Gynecological Malignancies
SABR involved multiple single radiation beams converging on single or multiple closely associated upper abdomen cancer targets. A SABR example case is depicted in Figure 2. All women had 8 Gy × three fraction SABR prescription; SABR treatment delivery parameters are listed in Table 2. The mean abdominopelvic cancer PTV irradiated was 118 cm (range: 4–613 cm). Median clinical follow-up was 10 months (range: 1–55 months). No surviving woman was lost to follow-up; 17 (50%) women died.
(Enlarge Image)
Figure 2.
Case example. (A) A 'pin cushion' map of 156 delivered treatment beams by the robotic Cyberknife device while targeting a malignant mesenteric implant next to the duodenum. (B–D) Axial, sagittal, and coronal projections of radiosurgical treatment. The clinical target is contoured in red (shaded structure on the right). The duodenum is contoured in blue (shaded structure on the left or below). The 24 Gy prescription isodose is highlighted in green (inner isodose line) and the duodenal volume receiving 80% of the 24 Gy dose is outlined in white (outer isodose line).
For color images please see online www.futuremedicine.com/doi/pdf/13.215
Table 3 catalogs all post-therapy SABR-related GI toxicities. In total, 14 (41%) of the 34 women had at least one grade 2 or higher post-therapy GI adverse event. A total of 12 (86%) of the 14 women had their symptoms resolved at the time of the 1-month post-therapy visit. One woman who had a large (301 cm) liver lesion treated had grade 4 hyperbilirubinemia. In the late observation period, a single (3%) grade 3 episode of enterocolitis (day 169) and a single (3%) grade 2 duodenal ulcer (day 163) occurred.
Radiation dose volume histograms for cancer targets and the duodenum are illustrated in Figure 3. Overall, a dose of 24 Gy was given to a mean duodenal volume of 1 cm (range: 0–7 cm) among the 34 women. Of the V0%–V100% parameters, V70% (p = 0.033) and V80% (p = 0.027) reached, while V60% (p = 0.058) and V90% (p = 0.061) approached, statistical significance for increased odds of having grade 2 or higher GI toxicity. Goodness-of-fit tests identified V80% as having the lowest -2 log likelihood statistics among V60%, V70% and V80% parameters.
(Enlarge Image)
Figure 3.
Dose–volume histogram of the patient in Figure 2 who underwent stereotactic ablative radiotherapy for a malignant mesenteric implant target (red line; right) next to the duodenum (blue line; left). The duodenal volume receiving 80% of the 24 Gy dose (dose = 19.2 Gy) is marked.
CTV: Clinical target volume; V: Volume.
For color images please see online www.futuremedicine.com/doi/pdf/13.215
In a binary logistic regression analysis including the duodenal V80%, only age at SABR treatment (odds ratio: 0.87; 95% CI: 0.77–0.98; p = 0.02) was significantly associated with grade 2 or higher post-therapy GI adverse events. A trend for V80% to predict GI toxicity was observed (odds ratio: 1.43; 95% CI: 0.97–2.12; p = 0.07). The receiver operating characteristic area under the curve, or the probability that V80% for a randomly chosen positive case exceeds that of a randomly chosen negative case, was 0.90 (95% CI: 0.79–1.00; p < 0.001). This indicates that when the V80% is used with age, performance status, and the number and volume of the SABR targets, the logistic regression yields a sensitivity of 71% and specificity of 80%. Nine (64%) of the 14 women with grade 2 or higher GI toxicity had more than 1 cm of the duodenum receiving 80% of the prescription dose, suggesting that V80% may be a clinically useful parameter predicting GI toxicity.
Results
SABR Dose Distribution for the Intestine
SABR involved multiple single radiation beams converging on single or multiple closely associated upper abdomen cancer targets. A SABR example case is depicted in Figure 2. All women had 8 Gy × three fraction SABR prescription; SABR treatment delivery parameters are listed in Table 2. The mean abdominopelvic cancer PTV irradiated was 118 cm (range: 4–613 cm). Median clinical follow-up was 10 months (range: 1–55 months). No surviving woman was lost to follow-up; 17 (50%) women died.
(Enlarge Image)
Figure 2.
Case example. (A) A 'pin cushion' map of 156 delivered treatment beams by the robotic Cyberknife device while targeting a malignant mesenteric implant next to the duodenum. (B–D) Axial, sagittal, and coronal projections of radiosurgical treatment. The clinical target is contoured in red (shaded structure on the right). The duodenum is contoured in blue (shaded structure on the left or below). The 24 Gy prescription isodose is highlighted in green (inner isodose line) and the duodenal volume receiving 80% of the 24 Gy dose is outlined in white (outer isodose line).
For color images please see online www.futuremedicine.com/doi/pdf/13.215
Table 3 catalogs all post-therapy SABR-related GI toxicities. In total, 14 (41%) of the 34 women had at least one grade 2 or higher post-therapy GI adverse event. A total of 12 (86%) of the 14 women had their symptoms resolved at the time of the 1-month post-therapy visit. One woman who had a large (301 cm) liver lesion treated had grade 4 hyperbilirubinemia. In the late observation period, a single (3%) grade 3 episode of enterocolitis (day 169) and a single (3%) grade 2 duodenal ulcer (day 163) occurred.
Radiation dose volume histograms for cancer targets and the duodenum are illustrated in Figure 3. Overall, a dose of 24 Gy was given to a mean duodenal volume of 1 cm (range: 0–7 cm) among the 34 women. Of the V0%–V100% parameters, V70% (p = 0.033) and V80% (p = 0.027) reached, while V60% (p = 0.058) and V90% (p = 0.061) approached, statistical significance for increased odds of having grade 2 or higher GI toxicity. Goodness-of-fit tests identified V80% as having the lowest -2 log likelihood statistics among V60%, V70% and V80% parameters.
(Enlarge Image)
Figure 3.
Dose–volume histogram of the patient in Figure 2 who underwent stereotactic ablative radiotherapy for a malignant mesenteric implant target (red line; right) next to the duodenum (blue line; left). The duodenal volume receiving 80% of the 24 Gy dose (dose = 19.2 Gy) is marked.
CTV: Clinical target volume; V: Volume.
For color images please see online www.futuremedicine.com/doi/pdf/13.215
Clinical Toxicity After SABR
In a binary logistic regression analysis including the duodenal V80%, only age at SABR treatment (odds ratio: 0.87; 95% CI: 0.77–0.98; p = 0.02) was significantly associated with grade 2 or higher post-therapy GI adverse events. A trend for V80% to predict GI toxicity was observed (odds ratio: 1.43; 95% CI: 0.97–2.12; p = 0.07). The receiver operating characteristic area under the curve, or the probability that V80% for a randomly chosen positive case exceeds that of a randomly chosen negative case, was 0.90 (95% CI: 0.79–1.00; p < 0.001). This indicates that when the V80% is used with age, performance status, and the number and volume of the SABR targets, the logistic regression yields a sensitivity of 71% and specificity of 80%. Nine (64%) of the 14 women with grade 2 or higher GI toxicity had more than 1 cm of the duodenum receiving 80% of the prescription dose, suggesting that V80% may be a clinically useful parameter predicting GI toxicity.
