KEY POINTS
- This retrospective cohort included 91 women receiving postoperative free-breathing IMRT for unilateral breast cancer: 54 left-sided and 37 right-sided. All had normal ECG and echocardiography before RT.
- Treatment delivered 50.0–50.4 Gy in 25 fractions, with extensive regional treatment in this high-risk cohort; internal mammary nodes were irradiated in 63% of left-sided and 70% of right-sided patients.
- Dose distribution was strongly laterality-dependent. Left-sided treatment produced higher mean heart dose (7.78 vs 5.17 Gy), LV mean dose (9.49 vs 2.15 Gy), and LAD mean dose (39.38 vs 3.11 Gy; all p<0.001).
- The pattern reversed for right-sided conduction structures: sinoatrial-node mean dose was 8.22 vs 3.13 Gy, SAN maximum dose 13.16 vs 4.85 Gy, and RCA mean dose 12.38 vs 4.14 Gy for right- versus left-sided treatment (all p<0.001).
- During a median 12-month follow-up, 60/91 patients developed a new ECG abnormality: 81.5% of left-sided versus 43.2% of right-sided patients. Median ECG abnormality-free survival was 7 versus 12 months (p=0.038).
- Phenotype also differed: among affected left-sided patients, 68.2% had repolarization abnormalities such as ST-T or QTc changes, whereas 62.5% of affected right-sided patients had rhythm abnormalities (p=0.032).
- Among left-sided patients with ECG changes, LV mean dose was 10.10 Gy in the repolarization group versus 7.97 Gy in the rhythm group (p=0.003). LV mean dose predicted repolarization changes somewhat better than mean heart dose (AUC 0.776 vs 0.715); the derived 8.75-Gy LV threshold is cohort-specific and not externally validated.
CLINICAL TAKEAWAY
Mean heart dose may hide very different cardiac exposure patterns in left- and right-sided breast RT. These results support studying ventricular and conduction-system dosimetry separately, but short-term ECG abnormalities are surrogate endpoints and should not yet drive new planning constraints.