Dr Sekeres on the Methods and Design of a Study Evaluating Smoking Implications in MDS

“This is a study we’ve been conducting for 7 years. It’s the only study of its kind. We have gone to [approximately] 150 sites in the United States, and we’ve [suggested to hematologists]: If you’re seeing a patient who has cytopenias and in whom you suspect a diagnosis of MDS, enroll them to this study. We received, through the generosity of patients, samples of their bone marrow, clinical data, and quality-of-life data.”

Mikkael A. Sekeres, MD, a professor of medicine and chief of the Division of Hematology in the Leukemia Section at the University of Miami Health System and Sylvester Comprehensive Cancer Center, discussed the design of a study investigating the association between smoking intensity, genetic mutations, and disease progression in patients with myelodysplastic syndrome (MDS). 

Sekeres explained that he and coinvestigators carried out a unique prospective cohort study to assess the relationship between smoking intensity/duration and the quantity/types of genetic mutations detected in patients diagnosed with MDS starting in June 2016. This study was led by Sangeetha Venugopal, MD, MS, of Sylvester Comprehensive Cancer Center. The investigators used prospective data from patients across approximately 150 sites in the United States who were enrolled in the National Heart, Lung, and Blood Institute MDS Natural History Study, Sekeres described. Hematologists were asked to consider enrollment to this study for patients with cytopenias and suspected MDS, he explained. Bone marrow samples, clinical data, and quality-of-life data were collected from 1898 enrolled patients, who were also followed over time, according to Sekeres.

Univariate analyses revealed that the total number of mutations was comparable between smokers and nonsmokers, including within specific disease categories, such as idiopathic cytopenia of undetermined significance, clonal cytopenia of undetermined significance, MDS, MDS/myeloproliferative neoplasm overlap syndrome, and acute myeloid leukemia. Nonetheless, the overall prevalence of mutations was higher in smokers than in nonsmokers, particularly for mutations in gene pathways for chromatin modification (smokers, 15% vs nonsmokers, 11%; P < .01) and RNA splicing (26% vs 19%; P < .001), as well as mutation in individual genes like ASXL1 (12% vs 8%; P < .01), SF3B1 (9% vs 6%; P < .05), U2AF1 (6% vs 3%; P < .05), and ZRSR2 (2% vs 1%; P < .05). Furthermore, multivariable regression models showed that after adjusting for sex, age, and disease group, the average number of mutations was significantly higher among smokers (2.0) vs nonsmokers (1.4; P = .04).