Radiation-induced lung parenchymal changes will be assess with non-contrast enhanced computerized tomography (CT) scans and incidence will be estimated in each radiation group by the cumulative incidence function and compared using Gray's test. Among patients who do not have any changes detected, death and disease progression will be considered as competing risks.

Incidence of thyroid insufficiency will be estimated in each radiation group by the cumulative incidence function and compared using Gray's test. In the absence of thyroid dysfunction, death and disease progression will be considered as competing risks.

Incidence of ipsilateral arm lymphedema will be estimated in each radiation group by the cumulative incidence function and compared using Gray's test.

Serial arm measurements may be analyzed using mixed effects model to compare the severity of ipsilateral arm lymphedema over time between the radiation groups.

Breast cosmesis will be assessed with characteristics graded on a four-point scale. Assessments will be performed by a trained nurse/nurse practitioner using the European Organization for Research and Treatment of Cancer (EORTC) Cosmetic Rating System for Breast Cancer. Incidence of adverse cosmesis (2=fair or 3=poor score) will be estimated in each radiation group by the cumulative incidence function and compared using Gray's test. Among patients without adverse breast cosmesis, death and disease progression will be considered as competing risks.