Differences In Delineation Guidelines For Head-And-Neck Cancer Result In Inconsistent Dose And Corresponding Ntcp Values

Purpose/Objective(s)Several guidelines for organ at risk (OAR) delineation have been published. However, the definition, selection, and delineation of OARs varies widely among the different publications. This may lead to unjustified comparisons between institutes that apply different guidelines, jeopardizing the translation of published results into routine clinical practice. The main purpose of this study was to test the hypothesis that the delineation of swallowing OARs (SWOARs) based on different guidelines results in inconsistent dose-volume parameters and subsequent normal tissue complication probability (NTCP) values for dysphagia-related endpoints.Materials/MethodsTen different SWOARs were delineated according to five different published delineation guidelines in 29 head and neck cancer patients. Reference delineation guidelines were defined as those corresponding to the guidelines and NTCP-models of Christianen et al. The geometric differences between the guidelines were quantified with the Concordance Index (CI). Furthermore, dosimetric consequences of using different guidelines were studied, as well as differences in the subsequent NTCP values (see Table for the studied models).ResultsThe Table gives an overview of the results. The largest CI was found for the larynx (mean CI of 0.86), while the smallest CI was 0.00 for the esophageal inlet muscle. The maximum difference in mean dose to the SWOARs between the different guidelines (maxΔD) was 19.3 Gy. NTCP values based on another guideline than the reference guideline resulted in a mean absolute ΔNTCP of 2.6 ± 3.4%. For 6 patients, ΔNTCP exceeded 10%. ΔNTCP depended on patient's anatomy and posture as well as on the primary tumor site.ConclusionsPoster Viewing Abstract 2710; TableConcordance index and dose and NTCP differences in 29 patientsNTCP model(associated swallowing organs at risk)CI [±SD]Mean dose difference [±SD] (Gy)between 5 guidelinesMean dose difference [±SD] (Gy)between 5 guidelinesMax NTCP difference (%)between 2 guidelines1. Swallowing dysfunction RTOG grade 2-4(superior PCM and supraglottic larynx)0.43 [0.17]3.2 [3.4]4.2 [2.5]11.62. Swallowing solid food(superior PCM and supraglottic larynx)0.43 [0.17]3.2 [3.4]4.3 [3.0]14.53. Swallowing soft food(middle PCM)0.58 [0.15]1.8 [2.1]1.8 [2.3]10.94. Swallowing liquids(supraglottic larynx)0.33 [0.12]3.8 [3.3]1.5 [1.7]7.15. Choking when swallowing(supraglottic larynx and EIM)0.22 [0.11]3.8 [3.8]3.5 [5.7]25.0Abbreviation: SD = standard deviation, RTOG = Radiation Therapy Oncology Group, CI = Concordance Index, NTCP = normal tissue complication probability, PCM = pharyngeal constrictor muscle, EIM = esophageal inlet muscle Open table in a new tab Purpose/Objective(s)Several guidelines for organ at risk (OAR) delineation have been published. However, the definition, selection, and delineation of OARs varies widely among the different publications. This may lead to unjustified comparisons between institutes that apply different guidelines, jeopardizing the translation of published results into routine clinical practice. The main purpose of this study was to test the hypothesis that the delineation of swallowing OARs (SWOARs) based on different guidelines results in inconsistent dose-volume parameters and subsequent normal tissue complication probability (NTCP) values for dysphagia-related endpoints. Several guidelines for organ at risk (OAR) delineation have been published. However, the definition, selection, and delineation of OARs varies widely among the different publications. This may lead to unjustified comparisons between institutes that apply different guidelines, jeopardizing the translation of published results into routine clinical practice. The main purpose of this study was to test the hypothesis that the delineation of swallowing OARs (SWOARs) based on different guidelines results in inconsistent dose-volume parameters and subsequent normal tissue complication probability (NTCP) values for dysphagia-related endpoints. Materials/MethodsTen different SWOARs were delineated according to five different published delineation guidelines in 29 head and neck cancer patients. Reference delineation guidelines were defined as those corresponding to the guidelines and NTCP-models of Christianen et al. The geometric differences between the guidelines were quantified with the Concordance Index (CI). Furthermore, dosimetric consequences of using different guidelines were studied, as well as differences in the subsequent NTCP values (see Table for the studied models). Ten different SWOARs were delineated according to five different published delineation guidelines in 29 head and neck cancer patients. Reference delineation guidelines were defined as those corresponding to the guidelines and NTCP-models of Christianen et al. The geometric differences between the guidelines were quantified with the Concordance Index (CI). Furthermore, dosimetric consequences of using different guidelines were studied, as well as differences in the subsequent NTCP values (see Table for the studied models). ResultsThe Table gives an overview of the results. The largest CI was found for the larynx (mean CI of 0.86), while the smallest CI was 0.00 for the esophageal inlet muscle. The maximum difference in mean dose to the SWOARs between the different guidelines (maxΔD) was 19.3 Gy. NTCP values based on another guideline than the reference guideline resulted in a mean absolute ΔNTCP of 2.6 ± 3.4%. For 6 patients, ΔNTCP exceeded 10%. ΔNTCP depended on patient's anatomy and posture as well as on the primary tumor site. The Table gives an overview of the results. The largest CI was found for the larynx (mean CI of 0.86), while the smallest CI was 0.00 for the esophageal inlet muscle. The maximum difference in mean dose to the SWOARs between the different guidelines (maxΔD) was 19.3 Gy. NTCP values based on another guideline than the reference guideline resulted in a mean absolute ΔNTCP of 2.6 ± 3.4%. For 6 patients, ΔNTCP exceeded 10%. ΔNTCP depended on patient's anatomy and posture as well as on the primary tumor site. ConclusionsPoster Viewing Abstract 2710; TableConcordance index and dose and NTCP differences in 29 patientsNTCP model(associated swallowing organs at risk)CI [±SD]Mean dose difference [±SD] (Gy)between 5 guidelinesMean dose difference [±SD] (Gy)between 5 guidelinesMax NTCP difference (%)between 2 guidelines1. Swallowing dysfunction RTOG grade 2-4(superior PCM and supraglottic larynx)0.43 [0.17]3.2 [3.4]4.2 [2.5]11.62. Swallowing solid food(superior PCM and supraglottic larynx)0.43 [0.17]3.2 [3.4]4.3 [3.0]14.53. Swallowing soft food(middle PCM)0.58 [0.15]1.8 [2.1]1.8 [2.3]10.94. Swallowing liquids(supraglottic larynx)0.33 [0.12]3.8 [3.3]1.5 [1.7]7.15. Choking when swallowing(supraglottic larynx and EIM)0.22 [0.11]3.8 [3.8]3.5 [5.7]25.0Abbreviation: SD = standard deviation, RTOG = Radiation Therapy Oncology Group, CI = Concordance Index, NTCP = normal tissue complication probability, PCM = pharyngeal constrictor muscle, EIM = esophageal inlet muscle Open table in a new tab Abbreviation: SD = standard deviation, RTOG = Radiation Therapy Oncology Group, CI = Concordance Index, NTCP = normal tissue complication probability, PCM = pharyngeal constrictor muscle, EIM = esophageal inlet muscle