Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Aug 3;207(3):1085–1095. doi: 10.1111/bjh.70031

Clonal haematopoiesis associates with distinct cytometric changes of red blood cells and platelets

J B Salzbrunn 1, I A van Zeventer 1, A O de Graaf 2, S Klatte 3, L J van Pelt 4, P Kamphuis 1, M G J M van Bergen 2, A G Dinmohamed 5,6, B A van der Reijden 2, J Linssen 3, J J Schuringa 1, J H Jansen 2,✉, G Huls 1,✉
PMCID: PMC12436208  PMID: 40754673

Summary

Individuals with macrocytosis or a high RDW are at increased risk of developing haematological malignancies. The mechanisms that mediate this association remain unknown but may involve the presence of clonal haematopoiesis (CH). Here, we performed targeted next‐generation sequencing on all individuals ≥60 years with macrocytosis (MCV >100 fL; n = 269) or high RDW (RDW ≥16%; n = 242) and 1:2 age‐ and sex‐matched controls from the population‐based Lifelines cohort. Macrocytosis is not associated with CH or peripheral blood count abnormalities. In contrast, individuals with a high RDW were associated with an increased number of mutated genes, larger clone sizes and a high prevalence of blood count abnormalities. Interestingly, individuals with a high RDW and CH display a uniform population of red blood cells in the distribution plots, despite not all cells carrying the respective mutation, suggesting an indirect effect of CH on the red blood cell population. While macrocytosis is not associated with CH in general, SF3B1 mutations associate with an elevated MCV. Individuals with a combination of TET2 and SRSF2 mutations show marked disturbances in platelet morphology. In conclusion, cytometric parameters of peripheral blood cells may serve as early indicators of dysplastic changes and are associated with distinct mutational patterns in CH.

Keywords: acute leukaemia, clinical haematology, clonal evolution, clonal haematopoiesis, epidemiology, genetic analysis

Venn diagram illustrating the interrelation between clonal haematopoiesis, elevated mean corpuscular volume/red cell distribution width (MCV/RDW) and blood count abnormalities. The overlap underscores the frequent co‐occurrence of these age‐associated haematopoietic changes. Individuals presenting with all three features demonstrate an elevated risk of developing haematological malignancies. Subtle alterations in peripheral blood cell morphology, such as reflected by MCV and RDW, may serve as indicators of early dysplastic changes.

graphic file with name BJH-207-1085-g001.jpg

INTRODUCTION

Abnormal blood cell counts, morphological dysplasia and genetic alterations characterize myeloid malignancies. 1 , 2 Intriguingly, peripheral blood cell morphological alterations, like a high red blood cell volume (mean corpuscular volume, MCV) and its variance (red blood cell distribution width, RDW) are also associated with a higher incidence of solid and haematological malignancies and stratify the risk of progression of individuals with clonal haematopoiesis (CH). 3 , 4 , 5 , 6 , 7

Data on the association between MCV and CH remain inconclusive, with different findings across population‐based studies. 6 , 7 , 8 , 9 , 10 , 11 The MCV may associate only with specific mutational profiles, as was reported for SF3B1, TP53 and PPM1D mutations. 8 , 10 , 11 In contrast, higher RDW consistently associates with CH across multiple studies. 6 , 7 , 8 , 11 , 12 , 13 A high RDW was reported to be associated with larger clone sizes and various gene mutations, excluding DNMT3A and TET2, the two most commonly mutated genes in CH. 11 Additionally, RDW levels ≥14.5% have been associated with increased mortality risk for individuals with CH. 12 Analogous to the MCV and RDW, other markers of changes in peripheral cell morphology may also signal a disturbance in haematopoiesis. For example, a higher mean platelet volume (MPV) and an increase in its variance (platelet distribution width, PDW) are associated with cardiovascular disease. 14 , 15 CH is also known for its association with cardiovascular events, 16 , 17 but limited data are available on the association between CH and morphological alterations of platelets. 7 , 11 , 13

Using a case–control cohort design, we systematically analysed the mutational spectrum of CH, its association with blood abnormalities, risk of incident haematological malignancies and overall survival. The study included individuals aged ≥60 years with macrocytosis (MCV >100 fL, n = 269) and high RDW (RDW ≥16%, n = 242) from the population‐based Lifelines cohort, matched 1:2 by age and sex.

METHODS

The Lifelines cohort

This study was performed within the population‐based Lifelines cohort, a study examining 167 729 persons living in the North of the Netherlands. 18 , 19 Data on the development of haematological malignancies were retrieved after linkage to the Netherlands Cancer Registry (NCR), maintained and hosted by the Netherlands Comprehensive Cancer Organization (IKNL) (Supporting Information: Methods). 20 Cytometric data were retrieved from an automated complete blood count analyser (XN‐series, Sysmex, Kobe, Japan). 21

Cohort selection

Three subcohorts were selected based on available blood cell characteristics (Figure 1). The macrocytosis and high RDW cohort included all 269 Lifelines participants ≥60 years of age with an MCV above the clinically used cut‐off value of 100 fL and all 242 Lifelines participants ≥60 years of age with an RDW ≥16% respectively. Cases were matched 1:2 with age‐ and sex‐matched controls. To further investigate the association between CH gene mutations and markers of red blood cell and platelet morphology, we used a cohort of 3240 individuals ≥60 years with previously generated error‐corrected targeted next‐generation sequencing (NGS) data (CH cohort). 22 , 23 , 24 , 25 This cohort was enriched for individuals ≥80 years of age and those with peripheral blood count abnormalities (Figure S1).

FIGURE 1.

FIGURE 1

Open in a new tab

Cohort overview. Overview of cohort selection taking place at 1the baseline visit and 2second screening visit of the population‐based Lifelines cohort. CH, clonal haematopoiesis; MCV, mean corpuscular volume; RDW, red blood cell distribution width.

Targeted error‐corrected NGS

Error‐corrected targeted NGS was performed with a custom panel of single‐molecule tagged molecular inversion probes (smMIP) covering regions in 27 myeloid and lymphoid malignancy‐associated driver genes (Illumina NovaSeq 6000 platform; Table S1). Somatic variants were called based on a variant allele frequency (VAF) ≥1% and ≥10 consensus variant reads. The panel design, library preparation and data analyses were previously described. 26 The mean coverage across all individuals and genes was 2986 (Figure S2). A list of all variants is provided in the Supporting Information Data S2: ‘Variant Data’.

Statistical analysis

Non‐parametric data were represented as medians with interquartile ranges and compared using the Mann–Whitney test. Categorical data were presented as absolute numbers with percentages and compared using the chi‐squared test. Cumulative incidences of haematological malignancies and cause‐specific mortality were visualized using the Aalen–Johansen estimator, with death and death from other causes, respectively, considered as a competing risk. Multivariable models for overall survival and malignancy development were constructed using Cox proportional hazards regression. For cause‐specific mortality, multivariable analyses were performed using competing risk regression according to the method of Fine and Gray, with death from other causes considered a competing risk. Multivariable linear and logistic regression analyses were performed to investigate the relation between individual driver gene mutations and cell morphological parameters. Throughout the manuscript, B denotes the non‐standardized regression coefficient, and β denotes the standardized coefficient. In all multivariable analyses, age and sex were included as covariables. For gene‐specific analyses, we performed multiple testing corrections using the false discovery rate (FDR) and reported q‐values to control for type I error. Statistical analyses were performed using R version 4.2.0.

RESULTS

Macrocytosis and a high RDW are associated with an increased risk of developing haematological malignancies

To evaluate the risk of incident haematological malignancies for individuals with abnormal red blood cell or platelet morphology, we analysed all individuals ≥60 years in the population‐based Lifelines cohort with available blood cell cytometric data (total n = 17 588; Figure 1). Macrocytosis and a high RDW are associated with an increased risk of developing haematological malignancies (MCV >100 fL: hazard ratio [HR], 5.11; 95% confidence interval [CI], 2.75–9.49, p < 0.001 and RDW ≥16%: HR, 6.49; 95% CI, 3.57–11.81, p < 0.001) and inferior overall survival (Figure 2A,B, Figure S3). The association between macrocytosis and incident haematological malignancies remained after correction for vitamin B12 deficiency and smoking (Table S2). Cause of death analysis revealed that the excess death rate of individuals with a high RDW was predominantly accounted for by CH‐associated causes, including cardiovascular death, death from haematological malignancies and death from solid cancers (Figure 2C, Table S3). For individuals with macrocytosis, death could not be attributed to a specific cause. Abnormal elevated MPV or PDW, markers of a disturbed platelet morphology, was not associated with overall survival nor the risk of developing haematological malignancies. However, cause of death analysis revealed both parameters to be associated with an increased risk of cardiovascular death (Figure 2C).

FIGURE 2.

FIGURE 2

Open in a new tab

Macrocytosis and a high RDW are associated with an increased risk of developing haematological malignancies. (A, B) The risk of incident haematological malignancies and overall survival associated with abnormal blood cell morphological parameters. Hazard ratios are derived from Cox proportional hazard regression with age and sex as covariables. Based on data from all individuals ≥60 years of age at the second screening visit with successful linkage with the Netherlands Cancer Registry (n = 17 588) and Statistics Netherlands (n = 17 760). (C) Cumulative incidence curves for cause‐specific mortality for individuals with abnormal cell morphological parameters. The primary cause of death was classified based on ICD‐10 codes; see Supporting Information: Methods. Statistical significance is reported from competing risk regression, with death from other causes considered a competing risk and age and sex as covariables. Multiple testing corrections were applied, and statistical significance is reported as q‐values **q < 0.05, ***q < 0.01, ****q < 0.001. RDW, red blood cell distribution width.

The prevalence and spectrum of clonal haematopoiesis is not altered in individuals with macrocytosis

For the macrocytosis cohort, NGS data were successfully generated for 257/269 cases and 529/538 matched controls (Figure 1, Table S4). Macrocytosis showed no significant difference in CH prevalence (43.2% vs. 37.8%, p = 0.17; Figure 3A–D), mutated gene count (p = 0.94) or clone size (p = 0.83). No differences were observed in the prevalence of the most commonly mutated genes or their clone sizes (Figure 3E). In line with recent studies which investigated the MCV as a continuous variable, 10 , 11 there was a trend towards enrichment of SF3B1 mutations (p = 0.08) among individuals with macrocytosis (Table S5). Reduced overall survival was observed in individuals with macrocytosis and CH (HR, 2.86; 95% CI, 1.67–4.89; p < 0.001) and without CH (HR, 1.91; 95% CI, 1.07–3.40; p = 0.027; P for interaction = 0.65; Figure 3F). Over a median follow‐up period of 7.3 years, 3.7% of individuals with macrocytosis and CH developed a haematological malignancy (HR, 12.02; 95% CI, 1.33–108.53; p = 0.027; Figure 3G, Table S6). Among 564 individuals with follow‐up MCV measurements available, stable macrocytosis was observed in 62% of cases. This was not linked to increased CH prevalence or an altered mutational landscape (Figure S4).

FIGURE 3.

FIGURE 3

Open in a new tab

The prevalence and spectrum of clonal haematopoiesis is not altered in individuals with macrocytosis. (A–C) Shown is the prevalence of CH (bar graph), distribution in number of mutations (violin plot) and proportion of individuals with a recurrent gene mutation for macrocytosis cases and age‐ and sex‐matched controls. The darker shade indicates individuals with an isolated gene mutation, while the lighter shade indicates individuals with a mutation in the respective gene and ≥1 additional mutation in another gene. (D, E) Highest observed VAF per individual (box plot) and VAF per variant for recurrent gene mutations detected in macrocytosis cases and controls. As the VAF per variant is reported in this figure, one individual may be represented multiple times. (F) Kaplan–Meier curve representing overall survival for macrocytosis cases and controls stratified by the presence of CH. (G) Cumulative incidence of haematological malignancies for macrocytosis cases and controls stratified by the presence of CH. (H) Proportion of individuals with a concurrent blood count abnormality for macrocytosis cases and controls. (I) Mutational landscape for macrocytosis cases stratified by the presence of a peripheral blood count abnormality. (J) Kaplan–Meier curve representing the overall survival for macrocytosis cases and controls stratified by the presence of CH and concurrent blood count abnormalities. For all time‐to‐event analyses, p‐values are reported from Cox proportional hazards regression with age and sex as covariables. BC, blood count abnormality (see Supporting Information: Methods); CH, clonal haematopoiesis; ref., statistical reference group; VAF, variant allele frequency.

We next questioned whether the triad of macrocytosis, CH and a blood count abnormality represents a more profound disturbance of haematopoiesis. Macrocytosis was associated with only a subtle increase in the prevalence of blood count abnormalities (e.g. anaemia 7.4% vs. 4.2%, p = 0.082; thrombocytopenia 5.8% vs. 2.3%, p = 0.018; Figure 3H, Table S7). The triad of macrocytosis, CH and a concurrent blood count abnormality showed no distinct mutational patterns, except for a higher prevalence of SF3B1 mutations (p = 0.007; Figure 3I, Figure S5). This triad (n = 24) associated with inferior survival (HR, 3.12; 95% CI, 1.39–6.98, p = 0.006; Figure 3J) and from this subgroup of individuals, 13% were diagnosed with haematological malignancies over a median follow‐up of 7.0 years (Table S6).

A high RDW is associated with a high‐risk mutational spectrum of clonal haematopoiesis

NGS data were successfully generated for 239/242 high RDW cases and 454/484 matched controls (Figure 1, Table S8). No difference in the overall prevalence of CH was found for individuals with CH ≥1% VAF (49.8% vs. 45.6%, p = 0.33; Figure 4A). However, restricting analysis to clone sizes ≥10% VAF revealed a higher CH prevalence in high RDW cases (15.9% vs. 7.9%, p = 0.002). Overall, high RDW cases exhibited a distinct mutational spectrum, with an increased number of mutated genes (p < 0.001) and a higher prevalence of JAK2 (4.6% vs. 0.2%, p < 0.001) and TET2 (20.9% vs. 15.0%, p = 0.048) mutations (Figure 4B,C). Notably, JAK2 and TET2 gene mutations were only enriched in the context of co‐mutations (Table S9). Clone sizes were significantly higher for individuals with a high RDW (median VAF, 3.6% vs. 2.9%, p = 0.007; Figure 5D), but this effect was not observed for specific gene mutations (Figure 4E). Individuals with high RDW and CH had lower overall survival (HR, 3.48; 95% CI, 1.85–6.56; p < 0.001; 77% 5‐year survival), whereas isolated high RDW was not associated with lower overall survival (HR, 1.58; 95% CI, 0.74–3.37, p = 0.24; interaction p = 0.057; Figure 4F). Over a median follow‐up period of 4.25 years, 6.5% of individuals with a high RDW and CH developed a haematological malignancy (HR, 15.21; 95% CI, 1.86–124.68, p < 0.001; Figure 4G, Table S6). No differences in the incidence of myeloid and lymphoid malignancies were observed, although the low number of incidence cases limited statistical analysis (Figure S6). Among 315 individuals with an additional RDW measurement available, stable high RDW (25.5% of baseline cases) was associated with an increased number of mutated genes (p = 0.003, Figure S7).

FIGURE 4.

FIGURE 4

Open in a new tab

A high RDW is associated with a high‐risk mutational spectrum of clonal haematopoiesis. (A–C) The prevalence of CH (bar graph), distribution in number of mutations (violin plot) and proportion of individuals with a recurrent gene mutation for high RDW cases and age‐ and sex‐matched controls. The darker shade indicates individuals with an isolated gene mutation, while the lighter shade indicates individuals with a mutation in the respective gene and ≥1 additional mutation in another gene. (D, E) Highest observed VAF per individual (box plot) and VAF per variant for recurrent gene mutations detected in high RDW cases and controls. As the VAF per variant is reported in this figure, one individual may be represented multiple times. (F) Kaplan–Meier curve representing overall survival for high RDW cases and controls stratified by the presence of CH. (G) Cumulative incidence of haematological malignancies for high RDW cases and controls stratified by the presence of CH. (H) Proportion of individuals with a concurrent blood count abnormality for high RDW cases and controls. (I) Mutational landscape for high RDW cases stratified for the presence of a peripheral blood count abnormality. (J) Kaplan–Meier curve representing the overall survival for high RDW cases and controls stratified by the presence of CH and concurrent blood count abnormalities. For all time‐to‐event analyses, p‐values are reported from Cox proportional hazards regression with age and sex as covariables. BC, blood count abnormality (see Supporting Information: Methods); CH, clonal haematopoiesis; RDW, red blood cell distribution width; ref., statistical reference group; VAF, variant allele frequency.

FIGURE 5.

FIGURE 5

Open in a new tab

Myeloid driver gene mutations are associated with distinct morphological changes of red blood cells and platelets. (A) The number of individuals carrying a recurrent gene mutation in the CH cohort (total n = 3240). (B) Association between blood cell morphological parameters and the presence of driver gene mutations at ≥1% VAF (upper panel) and ≥10% VAF (lower panel). Colours represent standardized regression coefficients (number of SD changes for the respective parameter). To allow comparison of the effect sizes, the standardized Beta (β) is reported. (C) Correlation between the clone size of TET2 or SRSF2 mutations and the MPV and PDW respectively. (D) Boxplots show the absolute value of MPV (upper panel) and PDW (lower panel) for individuals with a combined presence of TET2 and SRSF2 mutations (n = 17), isolated TET2 (n = 449) or isolated SRSF2 mutations (n = 16) compared to individuals without the presence of these mutations. (E) Platelet distribution curve for four selected individuals with co‐occurring TET2 and SRSF2 mutations. The upper and lower boundaries of the platelet distribution curve are indicated by the vertical green lines. (F) Red blood cell distribution curve for four selected individuals with the presence of multiple mutated genes and high RDW. The upper and lower boundaries of the red blood cell distribution curve are indicated by the vertical green lines. p‐values are reported from linear regression analysis with age and sex as covariables. In panel (B), multiple testing corrections were applied and statistical significance reported as q‐values **q < 0.05, ***q < 0.01, ****q < 0.001. CH, clonal haematopoiesis; MPV, mean platelet volume; PDW, platelet distribution width; RDW, red blood cell distribution width; SD, standard deviation, VAF, variant allele frequency.

High RDW was associated with blood count abnormalities, including anaemia (p < 0.001), thrombocytosis (p < 0.001) and leucocytosis (p = 0.037) (Figure 4H, Table S10). These individuals, and not those with an isolated high RDW, had a higher VAF and higher prevalence of TET2 and JAK2 mutations (Figure 4I, Figure S8). The triad of high RDW, CH and a blood count abnormality was associated with markedly inferior overall survival (HR, 5.24; 95% CI, 2.71–10.12; p < 0.001; Figure 4J). A blood count abnormality in multiple lineages was observed in 27.4% of individuals with this triad, and 7.9% of individuals with the triad were diagnosed with a haematological malignancy during a median follow‐up of 3.8 years (Table S6). Interestingly, individuals with CH and a blood count abnormality (representative of clonal cytopenia of unknown significance [CCUS]), but in the absence of a high RDW, had no inferior overall survival (HR, 0.70; 95% CI, 0.16–3.10; p = 0.64). There was minimal overlap between individuals with a high RDW and macrocytosis (4.13%, p = 0.21) and only 2/15 individuals with a high RDW who developed a haematological malignancy had macrocytosis. In addition to somatic mutations, mosaic chromosomal abnormalities (mCAs) may also confer a fitness advantage, leading to the expansion of a haematopoietic clone and the development of CH. While we could not assess all mCAs, we evaluated the prevalence of loss of chromosome Y (LoY) based on previously generated data. 27 Among 74 individuals from the high RDW cohort and 41 individuals from the macrocytosis cohort, no significant differences in the prevalence of LoY were found (Table S11).

Myeloid driver gene mutations are associated with distinct morphological changes of red blood cells and platelets

We next analysed a cohort of 3240 individuals ≥60 years old with NGS data (CH cohort; Figures 1 and 5A, Table S12). In this cohort, we analysed routinely reported cell morphological parameters of red blood cells and platelets as continuous variables (Figure 5B, Figure S9, Table S13). MCV associated with mutations in SF3B1 (B, 3.29 fL; SE, 0.70; β, 0.88; q < 0.001) and JAK2 (B, 2.17 fL; SE, 0.71; β, 0.46; q = 0.018). The association between JAK2 mutations and MCV disappeared after correction for the reticulocyte count (Figure S10). The RDW associated with a broad spectrum of CH mutations, including ASXL1, JAK2, SRSF2 and U2AF1. Mutations in DNMT3A (detected in 62.7% of individuals with CH) were not associated with any morphological parameter apart from a lower RDW (B, −0.13 fL; SE, 0.04; β, −0.12; q = 0.012). All regression estimates increased when we restricted the analyses to CH clones with a VAF ≥10%, indicating a dose–response relationship. When restricting the analysis to CH with a VAF ≥10%, additional associations were found between TET2 mutations and MPV (B, 0.34 fL; SE, 0.12; β, 0.34; q = 0.035) and PDW (B, 0.74 fL; SE, 0.27; β, 0.33; q = 0.036), as well as between SRSF2 mutations and MPV (B, 1.13 fL; SE, 0.35; β, 1.13; q = 0.011) and PDW (B, 2.91 fL; SE, 0.78; β, 1.31; q = 0.002). In a linear regression analysis, the VAF of both TET2 and SRSF2 gene mutations associated with the MPV and PDW respectively (Figure 5C).

Only the combination of TET2 and SRSF2 mutations (mean MPV, 11.9 fL vs. 10.9 fL, p = 0.001; median PDW, 15.6 fL vs. 13.1 fL, p < 0.001), but not the isolated presence of either TET2 or SRSF2 mutations, is associated with a disturbance in platelet morphology (Figure 5D, Figure S11). Only individuals with this combination revealed an evident skewing towards larger platelets in the platelet size distribution curve (n = 17 in total, curves displayed for n = 4; Figure 5E, Figure S12). The data from the case–control cohort (Figure 4) indicated that the association between RDW and CH may not be driven by enrichment or expansion of a specific driver gene mutation, but rather an overall disturbed mutational landscape characterized by multiple gene mutations and a higher VAF. In line with this, we observed the disappearance of all gene‐specific associations with RDW when restricting the analysis to individuals with a single gene mutation, while individuals with multiple mutated genes had a significantly higher RDW (14.1% vs. 13.3%, p < 0.001, Figure S13). In contrast to the skewed platelet size distribution curve, individuals with a high RDW and multiple mutations showed a uniform increase in the variance of red blood cell sizes, suggesting a cell extrinsic effect (Figure 5F; Figure S14).

DISCUSSION

Identifying individuals at high risk of progression to haematological malignancy remains a challenge in clinical practice. As was first indicated in a comprehensive analysis of the UK Biobank population by Weeks et al., 6 peripheral cell morphological alterations may reflect an early disruption in haematopoiesis. In their clonal haematopoiesis risk score, the authors assigned the same statistical weight to morphological alterations of red blood cells (both macrocytosis and a high RDW) and a set of high‐risk gene mutations while the prognostic value of cytopenia, number of mutations and clone size was estimated to be lower.

Data on the association between MCV and CH remain inconclusive, with different findings across population‐based studies. 6 , 7 , 8 , 9 , 10 , 11 In our cohort, macrocytosis did not associate with CH in terms of prevalence, clone size and number and type of mutated genes, also after correction for potential confounders. Cause of death analyses revealed no specific enrichment of disease categories and macrocytosis associated with a higher risk of haematological malignancy development and inferior overall survival with and without concomitant presence of CH. We speculate that macrocytosis may serve as a surrogate measure for an overall adverse health status or common risk factors, that may be mostly independent of CH or blood count abnormalities. While gene‐specific analysis of high‐risk mutated genes in CH was limited due to the low prevalence of these mutations in the general population, we found a specific association between macrocytosis and SF3B1 mutations that persisted independently of blood count alterations. These findings are in line with previously reported data obtained from population‐based studies, MDS patients and mice with mutated sf3b1. 10 , 11 , 28 , 29 , 30

A high RDW is a measure of an increased variance in red blood cell sizes and may signal a disturbance in haematopoiesis or early dysplastic changes better than a mean increase in the volume of red blood cells (MCV). Indeed, in contrast to macrocytosis, a high RDW is associated with CH ≥10% VAF, blood count abnormalities and a high‐risk mutational spectrum of CH, characterized by larger clone sizes and enrichment of several gene mutations, including mutated JAK2 and TET2. These associations were not driven by the clonal expansion of isolated gene mutations but rather by an overall increase in co‐mutational spectra and clonal complexity. Of note, the association between high RDW and high‐risk mutational spectra of CH was observed only in the presence of a peripheral blood count abnormality. In line with this, only the triad of a high RDW, CH and blood count abnormalities was associated with decreased overall survival. This indicates that an isolated elevated RDW has limited clinical significance and may be relevant only in cases of pre‐existing haematopoietic dysregulation, such as in individuals with CCUS. In individuals with CCUS, we demonstrate that high RDW further stratifies the risk of death, underscoring its potential clinical value. Intriguingly, we show that the RBC distribution curve for evaluable individuals with high RDW and CH showed a uniform increase in the variance of red blood cell sizes. In the case of a direct effect of certain mutations on affected RBCs, one would expect a suggestion of different populations in the RBC distribution curve. Since this is not the case, our data suggest that CH may induce a general cell extrinsic effect (e.g. inflammation in the bone marrow microenvironment) modulating the size of all cells.

Both macrocytosis and high RDW are associated with an increased incidence of haematological malignancy. In individuals with macrocytosis, this risk was observed only in those with concurrent CH, supporting the hypothesis that macrocytosis may serve as a surrogate marker for overall adverse health risk. In contrast, high RDW was linked to haematological malignancy development independently of CH. This finding may be explained by a higher prevalence of mCAs in individuals with high RDW. However, while our findings are not representative of all mCAs, we did not observe an increased presence of LoY in individuals with high RDW. Additionally, it should be noted that our panel for assessing CH mutations only included genes associated with myeloid malignancies. We observed no differences in the incidence of myeloid and lymphoid malignancies, although the low number of incidence myeloid malignancy cases (5 out of 15 in the RDW cohort) limited statistical analysis.

Furthermore, TET2 and SRSF2 mutations are associated with an increased platelet volume and distribution width. TET2 mutations are of particular interest due to their link to cardiovascular disease. Larger platelets exhibit higher thrombogenic properties and are associated with cardiovascular events. 31 , 32 For individuals with a TET2 and SRSF2 mutation, a distinct right‐hand shoulder was observed in the platelet distribution curve. This population of larger platelets may thus be the result of the mutated gene combination itself or another related cell‐intrinsic process. These findings are exploratory, as incident cases of cardiovascular events involving this gene combination were limited. Nevertheless, the potential impact of this gene combination on platelet morphology and cardiovascular risk profiles warrants further investigation.

In conclusion, different signs of dysplasia may be present in patients with CH or CCUS. Our findings emphasize the importance of recognizing individuals exhibiting the triad of CH, blood count abnormalities and peripheral cell morphological alterations as indicators of early dysplastic changes. Specifically, cytometric parameters reflecting increased variability in cell morphology, such as a high RDW, associate with high‐risk mutational patterns of CH and are valuable for further stratifying the risk of progression to haematological malignancies.

AUTHOR CONTRIBUTIONS

J.B.S., I.A.v.Z. and A.O.d.G. contributed to study design, data collection, analysis and interpretation of the data; P.K., M.G.J.M.v.B., A.G.D., B.A.v.d.R. and J.J.S. were involved in the interpretation of the data; S.K., L.J.v.P. and J.L. were involved in the data collection. G.H. and J.H.J. were principal investigators and involved in the study design, data collection and interpretation of the results.

FUNDING INFORMATION

The Lifelines initiative has been made possible by subsidy from the Dutch Ministry of Health, Welfare and Sport, the Dutch Ministry of Economic Affairs, the University Medical Center Groningen (UMCG), Groningen University and the Provinces in the North of the Netherlands (Drenthe, Friesland, Groningen). This work is part of the MDS‐RIGHT project, which has received funding from the European Union's Horizon 2020 Research and Innovation Program under grant agreement No 634789—‘Providing the right care to the right patient with MyeloDysplastic Syndrome at the right time’. This work was further supported by a grant from the Dutch Cancer Foundation (KWF10813).

CONFLICT OF INTEREST STATEMENT

Stefanie Klatte and Jo Linssen are employees of Sysmex Europe SE. The other authors declare to have no competing interests.

ETHICS STATEMENT

The Lifelines protocol was approved by the University Medical Groningen medical ethical committee under number 2007/152 and the study was performed in accordance with the Declaration of Helsinki.

PATIENT CONSENT STATEMENT

All participants provided written informed consent before participating in the study.

Supporting information

Data S1.

BJH-207-1085-s001.docx (2.1MB, docx)

Data S2.

BJH-207-1085-s002.xlsx (173.7KB, xlsx)

ACKNOWLEDGEMENTS

The authors wish to acknowledge the services of the Lifelines Cohort Study, the contributing research centres delivering data to Lifelines and all the study participants. We thank the registration team of the Netherlands Comprehensive Cancer Organisation (IKNL) for the collection of data for the Netherlands Cancer Registry as well as IKNL staff for scientific advice. We thank the team of Statistics Netherlands for collection and providing access to statistical data for the population of the Netherlands. Finally, we thank the Genome Technology Center, Radboud University Medical Center, for performing NovaSeq sequencing.

Salzbrunn JB, van Zeventer IA, de Graaf AO, Klatte S, van Pelt LJ, Kamphuis P, et al. Clonal haematopoiesis associates with distinct cytometric changes of red blood cells and platelets. Br J Haematol. 2025;207(3):1085–1095. 10.1111/bjh.70031

J. H. Jansen and G. Huls Co‐senior authors and contributed equally to this work.

[Correction added on 15 September 2025, after first online publication: The subcategory has been changed.]

Contributor Information

J. H. Jansen, Email: joop.jansen@radboudumc.nl.

G. Huls, Email: g.huls@umcg.nl.

DATA AVAILABILITY STATEMENT

The manuscript is based on data from the Lifelines cohort study. The data catalogue of Lifelines is publicly accessible at www.lifelines.nl. More information about how to request Lifelines data and the conditions of use can be found on their website (https://www.lifelines.nl/researcher/how‐to‐apply). Results from cause of death analyses are based on calculations by the authors using non‐public microdata from Statistics Netherlands. Under certain conditions, these microdata are accessible for statistical and scientific research. For further information: microdata@cbs.nl. Next‐generation sequencing data generated in this study are available in the Supporting Information Data S2: ‘Variant Data’.

REFERENCES

  • 1. Khoury JD, Solary E, Abla O, Akkari Y, Alaggio R, Apperley JF, et al. The 5th edition of the World Health Organization classification of Haematolymphoid Tumours: myeloid and histiocytic/dendritic neoplasms. Leukemia. 2022;36(7):1703–1719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Arber DA, Orazi A, Hasserjian RP, Borowitz MJ, Calvo KR, Kvasnicka HM, et al. International consensus classification of myeloid neoplasms and acute leukemias: integrating morphologic, clinical, and genomic data. Blood. 2022;140(11):1200–1228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Abelson S, Collord G, Ng SWK, Weissbrod O, Mendelson Cohen N, Niemeyer E, et al. Prediction of acute myeloid leukaemia risk in healthy individuals. Nature. 2018;559(7714):400–404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Bolton KL, Ptashkin RN, Gao T, Braunstein L, Devlin SM, Kelly D, et al. Cancer therapy shapes the fitness landscape of clonal hematopoiesis. Nat Genet. 2020;52(11):1219–1226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Montagnana M, Danese E. Red cell distribution width and cancer. Ann Transl Med. 2016;4(20):399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Weeks LD, Niroula A, Neuberg D, Wong W, Lindsley RC, Luskin MR, et al. Prediction of risk for myeloid malignancy in clonal hematopoiesis. NEJM Evid. 2023;2(5):EVIDoa2200310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Gu M, Kovilakam SC, Dunn WG, Marando L, Barcena C, Mohorianu I, et al. Multiparameter prediction of myeloid neoplasia risk. Nat Genet. 2023;55:1523–1530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Coombs CC, Zehir A, Devlin SM, Kishtagari A, Syed A, Jonsson P, et al. Therapy‐related clonal hematopoiesis in patients with non‐hematologic cancers is common and associated with adverse clinical outcomes. Cell Stem Cell. 2017;21(3):374–382.e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Buscarlet M, Provost S, Zada YF, Barhdadi A, Bourgoin V, Lépine G, et al. DNMT3A and TET2 dominate clonal hematopoiesis and demonstrate benign phenotypes and different genetic predispositions. Blood. 2017;130(6):753–762. [DOI] [PubMed] [Google Scholar]
  • 10. Saiki R, Momozawa Y, Nannya Y, Nakagawa MM, Ochi Y, Yoshizato T, et al. Combined landscape of single‐nucleotide variants and copy number alterations in clonal hematopoiesis. Nat Med. 2021;27(7):1239–1249. [DOI] [PubMed] [Google Scholar]
  • 11. Kar SP, Quiros PM, Gu M, Jiang T, Mitchell J, Langdon R, et al. Genome‐wide analyses of 200,453 individuals yield new insights into the causes and consequences of clonal hematopoiesis. Nat Genet. 2022;54(8):1155–1166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Jaiswal S, Fontanillas P, Flannick J, Manning A, Grauman PV, Mar BG, et al. Age‐related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371(26):2488–2498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Dawoud AAZ, Tapper WJ, Cross NCP. Clonal myelopoiesis in the UK biobank cohort: ASXL1 mutations are strongly associated with smoking. Leukemia. 2020;34(10):2660–2672. [DOI] [PubMed] [Google Scholar]
  • 14. Sato Y, Yoshihisa A, Watanabe K, Hotsuki Y, Kimishima Y, Yokokawa T, et al. Association between platelet distribution width and prognosis in patients with heart failure. PLoS One. 2020;15(12):e0244608. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Sansanayudh N, Anothaisintawee T, Muntham D, McEvoy M, Attia J, Thakkinstian A. Mean platelet volume and coronary artery disease: a systematic review and meta‐analysis. Int J Cardiol. 2014;175(3):433–440. [DOI] [PubMed] [Google Scholar]
  • 16. Jaiswal S, Natarajan P, Silver AJ, Gibson CJ, Bick AG, Shvartz E, et al. Clonal hematopoiesis and risk of atherosclerotic cardiovascular disease. N Engl J Med. 2017;377(2):111–121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Bick AG, Pirruccello JP, Griffin GK, Gupta N, Gabriel S, Saleheen D, et al. Genetic interleukin 6 signaling deficiency attenuates cardiovascular risk in clonal hematopoiesis. Circulation. 2020;141(2):124–131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Scholtens S, Smidt N, Swertz MA, Bakker SJ, Dotinga A, Vonk JM, et al. Cohort profile: LifeLines, a three‐generation cohort study and biobank. Int J Epidemiol. 2015;44(4):1172–1180. [DOI] [PubMed] [Google Scholar]
  • 19. Klijs B, Scholtens S, Mandemakers JJ, Snieder H, Stolk RP, Smidt N. Representativeness of the LifeLines cohort study. PLoS One. 2015;10(9):e0137203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Schouten LJ, Höppener P, van den Brandt PA, Knottnerus JA, Jager JJ. Completeness of cancer registration in Limburg, The Netherlands. Int J Epidemiol. 1993;22(3):369–376. [DOI] [PubMed] [Google Scholar]
  • 21. van Pelt JL, Klatte S, Hwandih T, Barcaru A, Riphagen IJ, Linssen J, et al. Reference intervals for Sysmex XN hematological parameters as assessed in the Dutch Lifelines cohort. Clin Chem Lab Med. 2022;60(6):907–920. [DOI] [PubMed] [Google Scholar]
  • 22. van Zeventer IA, Salzbrunn JB, de Graaf AO, van der Reijden BA, Boezen HM, Vonk JM, et al. Prevalence, predictors, and outcomes of clonal hematopoiesis in individuals aged ≥80 years. Blood Adv. 2021;5(8):2115–2122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. van Zeventer IA, de Graaf AO, Koorenhof‐Scheele TN, van der Reijden BA, van der Klauw MM, Dinmohamed AG, et al. Monocytosis and its association with clonal hematopoiesis in community‐dwelling individuals. Blood Adv. 2022;6(14):4174–4184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Kamphuis P, van Bergen M, van Zeventer IA, de Graaf AO, Dinmohamed AG, Salzbrunn JB, et al. Abnormal platelet counts and clonal hematopoiesis in the general population. Hema. 2023;7(1):e821. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Kamphuis P, van Zeventer IA, de Graaf AO, Salzbrunn JB, van Bergen M, Dinmohamed AG, et al. Sex differences in the Spectrum of clonal hematopoiesis. Hema. 2023;7(2):e832. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. van Zeventer IA, de Graaf AO, Salzbrunn JB, Nolte IM, Kamphuis P, Dinmohamed A, et al. Evolutionary landscape of clonal hematopoiesis in 3,359 individuals from the general population. Cancer Cell. 2023;41(6):1017–1031.e4. [DOI] [PubMed] [Google Scholar]
  • 27. Kamphuis P, van Zeventer IA, de Graaf AO, Genovese G, Salzbrunn JB, Dinmohamed AG, et al. Clonal hematopoiesis defined by somatic mutations infrequently co‐occurs with mosaic loss of the Y chromosome in a population‐based cohort. Hema. 2023;7(10):e956. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Shingai N, Harada Y, Iizuka H, Ogata Y, Doki N, Ohashi K, et al. Impact of splicing factor mutations on clinical features in patients with myelodysplastic syndromes. Int J Hematol. 2018;108(6):598–606. [DOI] [PubMed] [Google Scholar]
  • 29. Cui R, Gale RP, Xu Z, Qin T, Fang L, Zhang H, et al. Clinical importance of SF3B1 mutations in Chinese with myelodysplastic syndromes with ring sideroblasts. Leuk Res. 2012;36(11):1428–1433. [DOI] [PubMed] [Google Scholar]
  • 30. Obeng EA, Chappell RJ, Seiler M, Chen MC, Campagna DR, Schmidt PJ, et al. Physiologic expression of Sf3b1(K700E) causes impaired erythropoiesis, aberrant splicing, and sensitivity to therapeutic spliceosome modulation. Cancer Cell. 2016;30(3):404–417. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Chu SG, Becker RC, Berger PB, Bhatt DL, Eikelboom JW, Konkle B, et al. Mean platelet volume as a predictor of cardiovascular risk: a systematic review and meta‐analysis. J Thromb Haemost. 2010;8(1):148–156. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Gasparyan AY, Ayvazyan L, Mikhailidis DP, Kitas GD. Mean platelet volume: a link between thrombosis and inflammation? Curr Pharm Des. 2011;17(1):47–58. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1.

BJH-207-1085-s001.docx (2.1MB, docx)

Data S2.

BJH-207-1085-s002.xlsx (173.7KB, xlsx)

Data Availability Statement

The manuscript is based on data from the Lifelines cohort study. The data catalogue of Lifelines is publicly accessible at www.lifelines.nl. More information about how to request Lifelines data and the conditions of use can be found on their website (https://www.lifelines.nl/researcher/how‐to‐apply). Results from cause of death analyses are based on calculations by the authors using non‐public microdata from Statistics Netherlands. Under certain conditions, these microdata are accessible for statistical and scientific research. For further information: microdata@cbs.nl. Next‐generation sequencing data generated in this study are available in the Supporting Information Data S2: ‘Variant Data’.

Articles from British Journal of Haematology are provided here courtesy of Wiley

RESOURCES