Adverse independent prognostic effect of initial lung cancer on female patients with second primary breast cancer: a propensity score-matched study based on the SEER database.

Abstract

To investigate the prognostic impact of initial lung cancer (LC) on second primary breast cancer after LC (LC-BC) and further develop a nomogram for predicting the survival of patients. All patients diagnosed with LC-BC and first primary BC (BC-1) during 2000-2017 were collected from Surveillance, Epidemiology, and End Results database. Pathological features, treatment strategies and survival outcomes were compared between LC-BC and BC-1 before and after propensity score matching (PSM). Cox regression analysis was performed to identify the prognostic factors associated with LC in patients with LC-BC. Additionally, least absolute shrinkage and selection operator regression analysis was used to select clinical characteristics for nomogram construction, which were subsequently evaluated using the concordance index (C-index), calibration curve and decision curve analysis (DCA). 827 429 patients with BC-1 and 1445 patients with LC-BC were included in the analysis. Before and after PSM, patients with BC-1 had a better prognosis than individuals with LC-BC in terms of both overall survival (OS) and breast cancer-specific survival (BCSS). Furthermore, characteristics such as more regional lymph node dissection, earlier stage and the lack of chemotherapy and radiation for LC were found to have a stronger predictive influence on LC-BC. The C-index values (OS, 0.748; BCSS, 0.818), calibration curves and DCA consistently demonstrated excellent predictive accuracy of the nomogram. In conclusion, patients with LC-BC have a poorer prognosis than those with BC-1, and LC traits can assist clinicians estimate survival of patients with LC-BC more accurately.