Prevalence and Significance of Clonal Monocytosis of Undetermined Significance (CMUS) Amongst 431,531 United Kingdom Biobank Participants

Introduction

Clonal Monocytosis of Undetermined Significance (CMUS) and Clonal Cytopenia and Monocytosis of Undetermined Significance (CCMUS) were introduced by the 2022 International Consensus Classification (ICC) of Myeloid Neoplasms to delineate precursor entities characterized by clonal hematopoiesis (CH)-associated mutations and concurrent monocytosis, but not meeting diagnostic criteria for chronic myelomonocytic leukemia (CMML). However, these were introduced based primarily on expert opinion, with no studies yet performed to evaluate their practical relevance as discrete entities in the general population. Here, we assess the prevalence and significance of CMUS and CCMUS amongst 431,531 UK Biobank (UKB) participants.

Methods

CH driver mutations were identified in blood whole exome sequencing data from 431,531 UKB participants using Mutect2. In line with ICC criteria, CMUS was defined as the presence of a CH driver mutation, monocyte count ≥0.5 x 10⁹/L, monocytes ≥10% of total white blood cell count (WBC), and no prevalent hematological malignancy diagnosis. When cytopenia was also present, individuals were labelled as CCMUS. Those with a CH driver mutation and cytopenia, but no monocytosis, were classed as Clonal Cytopenia of Undetermined Significance (CCUS), and those with a driver and no cytopenia or monocytosis as CH. Cumulative incidence of myeloid neoplasia (MN) with death as a competing risk was examined, and forward stepwise regression used to fit a parsimonious model of factors associated with time to MN.

Results

We found that 1746/431,531 (0.4%) UKB participants met criteria for CMUS or CCMUS, of whom 70.6% were male. Notably, individuals with non-clonal monocytosis (i.e. meeting CMUS blood indices but lacking a CH driver mutation; n = 30,852) were also disproportionately male (70.3%), with males having higher median monocyte proportions of total WBC (7.4% vs 6.4% in females). CMUS and CCMUS were associated with a high risk of incident MN, with a 10-year cumulative incidence of 2.4%, 9.1% and 18.6% for CMUS with monocytes 0.5-1, CMUS with monocytes ≥1 and CCMUS respectively, contrasting with 1.8% and 5.3% respectively for CH and CCUS (without monocytosis).

Although DNMT3A was the most common driver mutation in CMUS/CCMUS, the mutation landscape varied according to blood indices and concurrent cytopenia state, with stepwise increase in frequency of TET2 (26.3%, 31.1% and 37.6%) and SRSF2 (1.6%, 6.6% and 16.3%) mutations amongst CMUS with monocytes 0.5-1, CMUS with monocytes ≥1 and CCMUS, respectively. Indeed, in CMUS with monocytes ≥1 and CCMUS, TET2 was the most common driver mutation.

Compared to CMUS driven by other mutations, CMUS with DNMT3A mutations was associated with a reduced risk of progression to MN (HR 0.28, 95% CI 0.12-0.63 for males; HR 0.16, 95% CI 0.05-0.53 for females, p=0.002/0.003 respectively), whilst CMUS with SRSF2 mutations was associated with increased risk of MN (HR 7.42, 95% CI 4.15 - 13.28 for males and HR 5.20, 95% CI 2.18 - 12.4 for females, p=1.38x10^-11/0.002 respectively). Similarly, CMUS with multiple TET2 mutations was associated with a relatively increased risk of progression to MN in males (HR 2.62, 95% CI 1.20 - 5.68, p=0.02), but single TET2 mutations were not (HR 1.15, 95% CI 0.66 - 2.01, p=0.62). Interestingly, amongst males with CMUS, the presence of mosaic loss of Y chromosome (mLOY) (in addition to their CMUS-defining mutation) was associated with decreased MN risk (HR 0.24, 95% CI 0.07-0.78, p = 0.02).

Amongst those who did not develop MN, both CMUS and non-clonal monocytosis were associated with reduced overall survival only when the absolute monocyte count exceeded 1 x 10⁹/L (age- and sex-adjusted HR 1.89 and 2.92 for non-clonal monocytosis ≥1 and CMUS with monocytes ≥1, p < 2.0 x 10^-16 p = 1.30 x 10^-15 respectively).

Conclusions

Supporting their introduction as distinct diagnostic entities, CMUS and CCMUS were associated with a substantially increased risk of incident MN, exceeding that associated with CCUS. However, our findings suggest that their definitions can be refined to enhance their prognostic significance by considering different monocyte count thresholds for males versus females, and by amending the classification of CMUS/CCMUS to exclude DNMT3A mutations (with the possible exception of R882 hotspot mutations), since these lacked the association with increased risk of MN seen for other driver mutations.