Lung and Breast Cancer in Systemic Sclerosis: Correlations with the Disease's Features

Background/Purpose: Systemic sclerosis (SSc) shows an increased incidence of lung and cancer compared to general population. Chronic interstitial inflammation might have a causative role for lung cancer; while, it is very difficult to explain the breast cancer since mammalian glands are not involved in disease's processes. We aimed to investigate the possible correlations between SSc features and the development of lung or breast cancer in SSc.Methods: We retrospectively evaluated 312 consecutive SSc patients (30/282 M/F, mean age 60.6 ± 13.7 SD years, mean disease duration 10.1 ± 6.8 SD years). Besides clinical features, individual cumulative radiation dosage (measured in mSv equivalents) for radiological examinations since SSc diagnosis was calculated.Results: We found 14 cases of lung cancer (4.5%) and 12 cases of breast cancer (3.8%), which appeared 14.3 ± 6.7 SD years and a median period of 5 years (range 1–22) after the diseases onset, respectively. Statistical analysis evidenced significant correlations between pulmonary neoplasia and male sex (p=0.035), disease duration (p=0.019), presence of anti-Scl70 (p=0.001) and absence of anticentromere (p<0.0001) antibodies, lung fibrosis (p=0.011), forced vital capacity (FVC) reduction (p<0.0001), and cyclophosphamide therapy (p=0.005). Logistic regression identified only FVC as significant independent factor (OR 1.08, 95% confidence interval 1.03–1.13; p=0.02) for lung cancer. On the contrary, breast cancer did not correlate with any SSc clinical parameters. Finally, cumulative radiation dosage did not affect the appearance of both lung and breast cancer. Interestingly, this latter was inversely related to radiation cumulative dosage (p=0.022).Conclusion: The present findings suggest that a SSc patients' subset with more severe disease might have an increased risk to develop lung cancer. This malignancy mainly affects male patients, as well as subjects with longer disease duration and higher prevalence of lung fibrosis, presence of anti-Scl70 and absence of anticentromere autoantibodies. The correlation with previous cyclophosphamide therapy could be an indirect sign of severe lung involvement. Breast cancer did not show any significant correlations with SSc features; it involved women with significantly lower x-ray exposure, due to the medially shorter time interval between SSc diagnosis and the breast cancer onset, compared to SSc patients without this malignancy.