Risk factors and prevalence of monoclonal gammopathy of undetermined significance (MGUS) by duffy genotype

Abstract Introduction: There is a 3x higher prevalence of MGUS in people identifying as Black. However, race is a social construct making this an inadequate explanatory model. Instead, race may be a proxy for environmental exposures and/or genetic variations such as Duffy null genotype (CC). CC genotype is found in 66% of people identifying as Black in the United States and <1% of White people. Duffy antigen maintains homeostatic levels of chemokines and cytokines; lack of expression (encoded by the CC genotype) is associated with tumorigenesis and metastasis in solid tumors as well as inflammation. Thus, we aimed to evaluate if the CC genotype is a risk factor for MGUS. Methods: Participants were identified from 2010 - 2021 from Mass General Brigham Biobank, a large repository of biospecimens and data linked to electronic health record data and survey data. Participants with a plasma cell malignancy diagnosed prior to screening were excluded, and participants with Duffy genotype and assessment for MGUS included. MALDI-TOF mass spectrometry and EXENT-iQ software were used to screen for and quantify monoclonal gammopathies. Duffy genotype (rs2814778; CC or non-CC) of each participant was determined from whole-exome sequencing data. Selected longitudinal clinical data was recorded. Chi-squared tests were used to compare categorial variables, Wilcoxon rank-sum test used to compare age and BMI, and multiple logistic regression was used to assess the presence of MGUS predicted by age, sex, BMI, Duffy genotype, family history of blood cancer, heavy alcohol use, and 30+ pack years of smoking (STATA 16.1). Results: 13,861 patients with MGUS testing and Duffy genotyping were identified; 9.2% (n=1,275) were CC of whom 86.6% (n=1106) identified as Black or African American. 58.8% (n=1106) of Black participants had the CC genotype compared to 1.4% (n=169) of non-Black participants (p<0.001). There was no difference in male sex by genotype (p=0.653). Median age at sample collection was 52 (IQR: 39-62) for CC and 55 (IQR: 41-65) for non-CC participants (p<0.001). Median BMI was 30.2 (IQR: 25.8-35.4) for CC and 26.5 (IQR: 23.3-30.6) for non-CC participants (p<0.001) with 51.5% of CC vs 27.6% of non-CC participants with BMI >30 (p<0.001). CC participants had similar rates of 30+ pack years of smoking (4.4% vs 6.7%; p=0.058) and higher rates of alcohol use >30g/day (7.0% vs 4.4%; p=0.009) than non-CC participants. Non-CC participants had a higher proportion of family history of blood cancer (17.8% vs 11.8%; p<0.001). In unadjusted analysis, CC participants had greater odds of MGUS than non-CC participants (14.3% vs 11.4%, OR 1.30 [95% CI: 1.10-1.53], p=0.002), but a similar pattern was seen when assessing MGUS and Black race (OR 1.33 [95% CI 1.15-1.53], p>0.001). Collinearity between Duffy genotype and race is 0.68 (moderate). In a multiple regression analysis without race, CC genotype (p=0.033), older age (p<0.001), male sex (p=0.002), and 30+ pack years of smoking (p<0.001) were significantly associated with MGUS while elevated BMI (p=0.467), heavy alcohol use (p=0.989), and family history of blood cancer (p=0.151) were not. In a multiple regression of Black participants only, older age (p<0.001) and smoking (p=0.003) remained significantly associated with MGUS, but male sex (p=0.521) and CC genotype (p=0.304) were not. CC participants with MGUS had a higher proportion of IgA isotype (22.4% vs 11.5%, p<0.001) but no differences in IgM isotype (19.2% vs 18.1%, p=0.720). Within the Black population there was no association between CC genotype and IgA isotype (22.1% vs 17.4%, p=0.398).Conclusion: This is the first assessment of Duffy genotype and MGUS prevalence. There is significant overlap between the social identity of Black race, CC genotype, and African genetic ancestry which complicates analysis. Although the unadjusted analysis showed 30% higher odds of MGUS in CC patients, there was no significant difference in MGUS prevalence by Duffy genotype within a Black population in adjusted analyses. Future analyses will assess a larger cohort, integrate African genetic ancestry, and further address multicollinearity as well as analyze Duffy genotype and rate of progression from MGUS to multiple myeloma (MM)—especially given the higher rate of IgA isotype in CC patients. Race is an imperfect proxy for genetic or environmental risk factors for MGUS, and further study is needed to elucidate root underlying risk factors and evolution to MM.