Purpose
nPreoperative chemoradiation therapy (CRT) is part of the standardntreatment of locally advanced rectal cancers. Tumor regression atnthe time of operation is desirable, but not much is known about thenrelationship between radiation dose and tumor regression. In thenpresent study we estimated radiation dose-response curves fornvarious grades of tumor regression after preoperative CRT.
Methods and Materials
nA total of 222 patients, treated with consistent chemotherapy andnradiation therapy techniques, were considered for the analysis.nRadiation therapy consisted of a combination of external-beamnradiation therapy and brachytherapy. Response at the time ofnoperation was evaluated from the histopathologic specimen andngraded on a 5-point scale (TRG1-5). The probability of achievingncomplete, major, and partial response was analyzed by ordinalnlogistic regression, and the effect of including clinicalnparameters in the model was examined. The radiation dose-responsenrelationship for a specific grade of histopathologic tumornregression was parameterized in terms of the dose required for 50%nresponse, D50,i, and the normalized dose-response gradient,nγ50,i.
Results
nA highly significant dose-response relationship was found (P=.002).nFor complete response (TRG1), the dose-response parameters werenD50,TRG1 = 92.0 Gy (95% confidence interval [CI] 79.3-144.9 Gy),nγ50,TRG1 = 0.982 (CI 0.533-1.429), and for major response (TRG1-2)nD50,TRG1&2 = 72.1 Gy (CI 65.3-94.0 Gy), γ50,TRG1&2 = 0.770n(CI 0.338-1.201). Tumor size andcategory both had a significantneffect on the dose-response relationships.
Conclusions
nThis study demonstrated a significant dose-response relationshipnfor tumor regression after preoperative CRT for locally advancednrectal cancer for tumor dose levels in the range of 50.4-70 Gy,nwhich is higher than the dose range usually considered.
Ane L. Appelt, MSc; John Pløen, MD; Ivan R. Vogelius, PhD; SørennM. Bentzen, PhD, DSc; Anders Jakobsen, DMSc