Premium
WE‐C‐AUD B‐06: Dose Response Analysis of Radiation‐Induced Retinopathy (RIRP) Single‐Institution Data Using the Lyman NTCP Model
Author(s) -
Song W,
Bhandare N,
Moiseenko V
Publication year - 2008
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.2962693
Subject(s) - nuclear medicine , confidence interval , medicine , radiation therapy , dosimetry , incidence (geometry) , ophthalmology , mathematics , surgery , geometry
Purpose: To analyze the dose‐response characteristics of radiation‐induced retinopathy (RIRP) to high‐dose radiation therapy using the Lyman NTCP model. Method and Materials: A retrospective analysis of 108 patients treated QD (23 RIRPs observed) and 78 patients treated BID (7 RIRPs observed) in the head‐and‐neck sites were performed. Using the α/β ratio determined with the QD group (2.65 Gy; [−1.06,∞] 95% CI), the total dose for each patient was converted to the normalized “2 Gy/fraction” total dose (NTD) before analysis. Subsequently, the Lyman NTCP parameters were estimated using the maximum‐likelihood method with asymmetric confidence intervals determined using the profile‐likelihood analysis (68% CI). The chi‐square goodness‐of‐fit test was performed to evaluate the modeling. Results: Due to the minute thickness and volume of the organ a DVH analysis was not performed. Instead, the dose to the retina was specified as the highest isodose covering at least 1/3 of the retinal surface that geometrically covered the observed retinal damage. Therefore, the volume effect parameter, n , was not investigated. The estimated values of NTD 50 , and m were 72.4 Gy [67.5,79.5] and 0.356 [0.29,0.44], respectively, for QD dataset and 73.0 Gy [67.0,85.5] and 0.209 [0.15,0.30], respectively, for BID dataset. While the NTD 50 was generally in good agreement between the two groups, the BID group showed steeper slope resulting in lower incidence at clinical dose range (∼ 30 to 60 Gy). The chi‐square test showed that the model was adequate (p = 1.00) for both the QD and BID groups. Conclusion: The Lyman NTCP model was used to describe the RIRP data. Results show that QD and BID groups give comparable NTD 50 values when NTD conversion is made. However, the BID group showed steeper slope. If the validity of the linear‐quadratic formalism is assumed, then hyperfractionation may be beneficial for RIRP given the small α/β ratio estimated.