Temporal relationship between sarcoidosis and malignancies in a nationwide cohort of 1942 patients

To investigate the phenotype of sarcoidosis according to the time when a malignancy is diagnosed (preexisting to the diagnosis of sarcoidosis, concomitant, or sequential) and to identify prognostic factors associated with malignancies in a large cohort of patients with sarcoidosis. We searched for malignancies in the SARCOGEAS cohort, a multicenter nationwide database of consecutive patients diagnosed with sarcoidosis according to the ATS/ESC/WASOG criteria. Solid malignancies were classified using the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) nomenclature, and hematological malignancies using the 2016 WHO classification. We excluded patients with a biopsy-proven diagnosis of sarcoidosis based exclusively on demonstrating granulomas in tissues also involved by malignant cells. Out of 1942 patients with sarcoidosis, 233 (12%) developed 250 malignancies, including solid (n = 173), hematological (n = 57), and both types of malignancies (n = 3). Concerning the time interval between the diagnoses of both conditions, 83 (36%) patients were diagnosed with malignancy at least 1 year before sarcoidosis diagnosis, 22 (9%) had s synchronous diagnosis of both diseases, and 118 (51%) developed malignancies at least 1 year after the diagnosis of sarcoidosis (the remaining cases developed malignancies in different time intervals). The multivariate-adjusted model showed that individuals with sarcoidosis who developed a malignancy had an hazard ratio (HR) of 2.27 [95% confidence interval (CI), 1.62-3.17] for having an asymptomatic clinical phenotype at diagnosis of sarcoidosis and that spleen (presence vs. absence: HR = 2.06; 95% CI, 1.21-3.51) and bone marrow (presence vs. absence: HR = 3.04; 95% CI, 1.77-5.24) involvements were independent predictors for the development of all-type malignancies. No predictive factors were identified when the analysis was restricted to the development of solid malignancies. The analysis limited to the development of hematological malignancies confirmed the presence of involvement in the spleen (HR = 3.73; 95% CI, 1.38-10.06) and bone marrow (presence vs. absence: HR = 8.00; 95% CI, 3.15-20.35) at the time of sarcoidosis diagnosis as predictive factors. It is essential to consider the synchronous or metachronous timing of the diagnosis of malignancies in people with sarcoidosis. We found that half of the malignancies were diagnosed after a diagnosis of sarcoidosis, with spleen a