Abstract P1-16-03: Quantitative versus semi-quantitative assessments of radiation-induced pulmonary fibrosis post adjuvant breast radiotherapy

Abstract

Abstract Objective To evaluate the quantitative versus semi-quantitative assessments of radiation induced pulmonary fibrosis (RIPF) post adjuvant breast radiotherapy (RT). Methods High resolution computed tomography (HRCT) assessed lung physical density changes (CTD) and physician identified HRCT visual grading scores (CTS) were analysed at the minimum of 12 months post RT at one institution. The treated side in-portal lung regions for CTD and CTS assessments were: central-axis (CA) + regions 5cm superior & inferior to CA and the corresponding mid anterolateral region respectively. Respiratory motion was accounted for by subtracting the untreated side lung density from the treated side. Mean lung densities correspond to each voxels were automatically calculated by Pinnacle software (Phillips, Eindhoven, The-Netherlands). Grading of CTS was according to the RTOG/EORTC (grade 0, 1, 2, and 3 defined as none, slight, patchy, and dense HRCT appearance respectively) and analysed by a radiologist (JL) and re-checked a radiation oncologist (PG). Results Total numbers of 403 patients were analysed. A substantial association was verified between CTD and CTS assessment. An increase of ∼0.01 g/ml (95% CI 0.003-0.02) in CTD with each CTS score increase of 1 was observed (Table-1a). The RIPF can be categorised quantitatively into three groups of CTS 0 vs. 1-2 vs. 3 based on the mean CTD (Table-1b). Table-1a. Correlation between CTD and CTS method. Table-1b. Grouping of CTS based on CTD methodTable-1a. Correlation between CTD and CTS methodCTSTreated side mean CTD - LeftTreated side mean CTD - RightMean Total00.055 (119)-0.011 (136)0.020 (255)10.065 (60)0.005 (56)0.036 (116)20.083 (15)0.012 (13)0.050 (28)30.108 (4)00.108 (4)Total0.061 (198)-0.005 (205)0.028 (403)Table-1b. Grouping of CTS based on CTD methodCTSTreated side mean CTD - LeftTreated side mean CTD - RightMean Total95% Confidence interval00.055 (119)-0.011 (136)0.020 (255)0.012-0.0271-20.069 (75)0.006 (69)0.039 (144)0.029-0.04830.108 (4)00.108 (4)0.079-0.137 A cut off CTD of 0.089 g/ml exemplified the best compromise between sensitivity (100%) and specificity (88.2%) for dense HRCT appearance. However, there was no good compromise of CTD cut off for slight and patchy HRCT score possibly due to intra observer variation and the scale of the CTD measure (small increase in CTD may not be detected visually by the observer). Multivariable analysis revealed increasing age, current smoker, V20 ≥ 10% (the volume of lung that was covered by the 20Gy isodose line), central lung distance ≥ 2cm (the distance between posterior RT tangents and the chest wall), combined endocrine & chemotherapy, and treated side mean CTD to be significantly associated with development of grade ≥1 RIPF. Conclusions There was a good correlation between quantitative (CTD) and semi-quantitative (CTS) assessment of RIPF post adjuvant breast RT. The CTD method could be advantageous for both routine clinical practice and future clinical trials that require more detailed quantification of dense RIPF. Citation Format: Pramana A, Browne L, Cox H, Saba A, Pham K, Trakis S, Crawford K, Hall M, Batchelor N, Lim J, Graham P. Quantitative versus semi-quantitative assessments of radiation-induced pulmonary fibrosis post adjuvant breast radiotherapy. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P1-16-03.