Long-term changes in pulmonary function after incidental lung irradiation for breast cancer: a prospective study with 7-year follow-up.

Purpose: To evaluate late pulmonary function changes after incidental pulmonary irradiation for breast cancer.

Methods And Materials: Forty-three consecutive female patients diagnosed with breast carcinoma and treated with postoperative radiation therapy (RT) at the same dose (50 Gy) and fractionation (2 Gy/fraction, 5 days/week) were enrolled. Pulmonary function tests (PFT) and ventilation/perfusion scans were performed before RT and 6, 12, 24, and 84 months afterward.

Results: Forty-one patients, mean age 55 years, were eligible for the analysis. No differences were found in the baseline PFT values for age, smoking status and previous chemotherapy; women undergoing mastectomy showed baseline spirometric PFT values lower than did women treated with conservative surgery. The mean pulmonary dose was 10.9 Gy, being higher in women who also received lymph node RT (15.8 vs 8.6, P<.01). Only 1 patient experienced symptomatic pneumonitis. All PFT values showed a reduction at 6 months. From then on, the forced vital capacity and forced expiratory volume in 1 second began their recovery until reaching, and even exceeding, their baseline values at 7 years. Diffusing capacity of the lungs for carbon monoxide and ventilation/perfusion scans continued to reduce for 24 months and then partially recovered their baseline values (-3.5%, -3.8%, and -5.5%, respectively). Only the percentage difference at 7 years in the ventilation scan correlated with the dosimetric parameters studied. Other variables, such as age, smoking status, previous chemotherapy, and concomitant tamoxifen showed no significant relation with changes in PFT (ΔPFT) values at 7 years.

Conclusions: The study of reproducible subclinical parameters, such as PFT values, shows how their figures decrease in the first 2 years but practically recover their baseline values in the long term. The extent of the reduction in PFT values was small, and there was no clear association with several dosimetric and clinical parameters.