72 Chapter 5 Discussion We observed myocardial damage in patients with EC after nCRT with a prescribed dose of only 41.4 Gy. To our knowledge, this study shows for the first time a linear dose-response relationship between mean radiation dose per myocardial segment and ECV. Since ECV is a good surrogate for histological collagen burden, our results demonstrate that co-irradiation of the heart causes direct damage to the myocardium in terms of diffuse fibrosis.[25] It is well established that elevated ECV is associated with increased risk of heart failure and cardiac death.[26–28] In line with this, myocardial fibrosis might contribute to the evaluated risk of cardiac complications and/or mortality after thoracic radiation therapy. The literature shows that the mean heart dose is an independent predictor of cardiac complications and OS in patients with EC treated with nCRT.[2,29] We demonstrated that radiation to the heart causes an increase in ECV in a dose-dependent way. This myocardial fibrosis might be one of the mechanisms that contributes to worse OS when the heart is exposed to higher radiation doses. However, we found no differences in CMR parameters between the groups and therefore could not demonstrate the clinical impact of elevated ECV in our study, which might be explained by the smaller sample size. Pre-clinical studies have already shown that direct injury of the myocardium due to irradiation is associated with decreased microvascular density, vessel leakage and perivascular fibrosis.[12,30] In line with this, several studies reported regional hypoperfusion within the radiation field using myocardial scintigraphy for breast cancer patients.[31–33] Similarly, gated Myocardial Perfusion Imaging revealed significantly more myocardial perfusion defects within the first year after concurrent chemoradiation for esophageal cancer compared to control patients. [34] An autopsy study found diffuse pericellular or perivascular fibrosis, especially when the radiation dose to the heart exceeded 30 Gy.[35] In the present study, the radiation dose in these segments was on average 24 Gy if ECV was elevated (> 27.7%) versus 11 Gy in segments with normal ECV (p = 0.03). Even in segments that received a relatively low radiation dose (mean dose < 5 Gy), ECV was slightly higher than in control segments, 26.3% versus 24.0% respectively (p =0.03) (figure 3). Concurrent chemotherapy might have contributed to the effect of the radiotherapy on ECV, especially in the low dose segments. However, we demonstrated an independent linear dose effect relation between mean dose per segment and ECV.
RkJQdWJsaXNoZXIy MTk4NDMw