Points vs. Volumes: How Are They Related in Assessing Doses to the Organs at Risk during Intracavitary High-Dose-Rate Brachytherapy for Cervical Cancer?

PurposeTo explore the relationship between the maximum rectal, sigmoid and bladder point doses; the ICRU rectal and ICRU bladder point doses; and the D0.1cc, D1cc and D2cc for the rectum, sigmoid and bladder in patients undergoing intracavitary brachytherapy for cervical cancer over the course of 5 HDR fractions.Materials and MethodsFrom 2004-2010, 40 patients with Stage I-III cervical cancer underwent an MR at fraction 1 and a CT scan for fractions 1-5 for image-based brachytherapy treatment planning. For each fraction, 30cc of contrast was injected into the rectosigmoid and bladder. Points closest to the applicator and pear-shaped dose distribution were identified as maximum rectal, maximum sigmoid, and maximum bladder point doses. An ICRU rectal point and ICRU bladder point were also identified according to the criteria of the International Commission on Radiation Units and Measurements (ICRU) report 38. The D0.1cc, D1cc, and D2cc for the contoured rectum, sigmoid and bladder were calculated for each fraction. Modifications in the shape of the pear were made based on these normal tissue dose constraints as well as coverage of the GEC ESTRO-defined high-risk CTV on the fraction 1 MR. The relationships of these critical organ point doses and volumes for each fraction were then analyzed using the Pearson correlation and confirmed using a mixed-effects repeated measures model. Patients were followed for assessment of late toxicities which were graded according the Common Toxicity Criteria (CTC). Median followup was 4.08 years.ResultsFor the rectal point doses, there was a moderate but statistically significant relationship between the ICRU rectal point and the max rectal point (p= 0.0002). When comparing points to volumes, the rectal ICRU point was significantly related to each of the volume doses, but most strongly correlated to the D 2cc (p<0.0001). The max rectal point most strongly correlated with the D1cc (p<0.0001). For the sigmoid, the max sigmoid point dose most strongly correlated to the D0.1cc but was also statistically related to the D1cc and D2cc (p< 0.0001). For the bladder, the ICRU bladder point dose was moderately related to the max bladder point dose (p=0.0002). The ICRU bladder point was most statistically related to D2cc (p< 0.0001), but also correlated with the other volumes. The max bladder point was most strongly correlated to the D0.1cc (p<0.0001).There were 4 patients (10%) with grade 1 or 2 rectal toxicities. There was one patient (2.5%) with grade 1 bladder toxicity. Four patients (10%) were found to have any grade sigmoid toxicity and 2 of these patients (5%) had grade 4 toxicity. There was no statistically significant correlation between any of the point doses or volume doses with these toxicities.ConclusionsImage-guided HDR brachytherapy enables assessment of both point doses and volume doses. The relationship between the points and the volumes appears to be complex in this comparative analysis and not absolutely predictive of outcome. Our results confirm the correlation between the ICRU rectal point and the rectal D2cc and also suggest a correlation between the ICRU bladder point and the bladder D2cc. The maximum sigmoid and maximum bladder points were significantly correlated to the D0.1cc, but the rectal maximum point was more closely correlated to the D1cc. PurposeTo explore the relationship between the maximum rectal, sigmoid and bladder point doses; the ICRU rectal and ICRU bladder point doses; and the D0.1cc, D1cc and D2cc for the rectum, sigmoid and bladder in patients undergoing intracavitary brachytherapy for cervical cancer over the course of 5 HDR fractions. To explore the relationship between the maximum rectal, sigmoid and bladder point doses; the ICRU rectal and ICRU bladder point doses; and the D0.1cc, D1cc and D2cc for the rectum, sigmoid and bladder in patients undergoing intracavitary brachytherapy for cervical cancer over the course of 5 HDR fractions. Materials and MethodsFrom 2004-2010, 40 patients with Stage I-III cervical cancer underwent an MR at fraction 1 and a CT scan for fractions 1-5 for image-based brachytherapy treatment planning. For each fraction, 30cc of contrast was injected into the rectosigmoid and bladder. Points closest to the applicator and pear-shaped dose distribution were identified as maximum rectal, maximum sigmoid, and maximum bladder point doses. An ICRU rectal point and ICRU bladder point were also identified according to the criteria of the International Commission on Radiation Units and Measurements (ICRU) report 38. The D0.1cc, D1cc, and D2cc for the contoured rectum, sigmoid and bladder were calculated for each fraction. Modifications in the shape of the pear were made based on these normal tissue dose constraints as well as coverage of the GEC ESTRO-defined high-risk CTV on the fraction 1 MR. The relationships of these critical organ point doses and volumes for each fraction were then analyzed using the Pearson correlation and confirmed using a mixed-effects repeated measures model. Patients were followed for assessment of late toxicities which were graded according the Common Toxicity Criteria (CTC). Median followup was 4.08 years. From 2004-2010, 40 patients with Stage I-III cervical cancer underwent an MR at fraction 1 and a CT scan for fractions 1-5 for image-based brachytherapy treatment planning. For each fraction, 30cc of contrast was injected into the rectosigmoid and bladder. Points closest to the applicator and pear-shaped dose distribution were identified as maximum rectal, maximum sigmoid, and maximum bladder point doses. An ICRU rectal point and ICRU bladder point were also identified according to the criteria of the International Commission on Radiation Units and Measurements (ICRU) report 38. The D0.1cc, D1cc, and D2cc for the contoured rectum, sigmoid and bladder were calculated for each fraction. Modifications in the shape of the pear were made based on these normal tissue dose constraints as well as coverage of the GEC ESTRO-defined high-risk CTV on the fraction 1 MR. The relationships of these critical organ point doses and volumes for each fraction were then analyzed using the Pearson correlation and confirmed using a mixed-effects repeated measures model. Patients were followed for assessment of late toxicities which were graded according the Common Toxicity Criteria (CTC). Median followup was 4.08 years. ResultsFor the rectal point doses, there was a moderate but statistically significant relationship between the ICRU rectal point and the max rectal point (p= 0.0002). When comparing points to volumes, the rectal ICRU point was significantly related to each of the volume doses, but most strongly correlated to the D 2cc (p<0.0001). The max rectal point most strongly correlated with the D1cc (p<0.0001). For the sigmoid, the max sigmoid point dose most strongly correlated to the D0.1cc but was also statistically related to the D1cc and D2cc (p< 0.0001). For the bladder, the ICRU bladder point dose was moderately related to the max bladder point dose (p=0.0002). The ICRU bladder point was most statistically related to D2cc (p< 0.0001), but also correlated with the other volumes. The max bladder point was most strongly correlated to the D0.1cc (p<0.0001).There were 4 patients (10%) with grade 1 or 2 rectal toxicities. There was one patient (2.5%) with grade 1 bladder toxicity. Four patients (10%) were found to have any grade sigmoid toxicity and 2 of these patients (5%) had grade 4 toxicity. There was no statistically significant correlation between any of the point doses or volume doses with these toxicities. For the rectal point doses, there was a moderate but statistically significant relationship between the ICRU rectal point and the max rectal point (p= 0.0002). When comparing points to volumes, the rectal ICRU point was significantly related to each of the volume doses, but most strongly correlated to the D 2cc (p<0.0001). The max rectal point most strongly correlated with the D1cc (p<0.0001). For the sigmoid, the max sigmoid point dose most strongly correlated to the D0.1cc but was also statistically related to the D1cc and D2cc (p< 0.0001). For the bladder, the ICRU bladder point dose was moderately related to the max bladder point dose (p=0.0002). The ICRU bladder point was most statistically related to D2cc (p< 0.0001), but also correlated with the other volumes. The max bladder point was most strongly correlated to the D0.1cc (p<0.0001). There were 4 patients (10%) with grade 1 or 2 rectal toxicities. There was one patient (2.5%) with grade 1 bladder toxicity. Four patients (10%) were found to have any grade sigmoid toxicity and 2 of these patients (5%) had grade 4 toxicity. There was no statistically significant correlation between any of the point doses or volume doses with these toxicities. ConclusionsImage-guided HDR brachytherapy enables assessment of both point doses and volume doses. The relationship between the points and the volumes appears to be complex in this comparative analysis and not absolutely predictive of outcome. Our results confirm the correlation between the ICRU rectal point and the rectal D2cc and also suggest a correlation between the ICRU bladder point and the bladder D2cc. The maximum sigmoid and maximum bladder points were significantly correlated to the D0.1cc, but the rectal maximum point was more closely correlated to the D1cc. Image-guided HDR brachytherapy enables assessment of both point doses and volume doses. The relationship between the points and the volumes appears to be complex in this comparative analysis and not absolutely predictive of outcome. Our results confirm the correlation between the ICRU rectal point and the rectal D2cc and also suggest a correlation between the ICRU bladder point and the bladder D2cc. The maximum sigmoid and maximum bladder points were significantly correlated to the D0.1cc, but the rectal maximum point was more closely correlated to the D1cc.