Background

Clonal haematopoiesis of indeterminate potential (CHIP), defined as the presence of somatic mutations and clonal expansion in haematopoietic stem cells of the blood or bone marrow, is a common ageing-associated condition that begins in middle age [1]. CHIP, a variant allele frequency (VAF) exceeding 2%, is not a malignancy but rather a predisposition to haematological malignancy and cardiovascular disease [2, 3]. Furthermore, CHIP could contribute to chronic inflammation; for example, loss of ten eleven translocation-2 (TET2) upregulated inflammatory mediators, such as interleukin-6, in mouse colitis and mouse atherosclerosis models [4, 5].

In contrast, inflammation from pre-existing comorbidities, including autoimmune diseases or infections, can cause mutagenesis via DNA damage and promote CHIP [6]. CHIP was reportedly identified in 17% of rheumatoid arthritis patients and 30.4% of patients with anti-neutrophil cytoplasmic antibody (ANCA)–associated vasculitis [7, 8]. In patients with ulcerative colitis (UC), Zhang et al. identified that 12.8% had CHIP variants and interpreted that patients with UC may show a slightly higher tendency toward a CHIP appearance after the sixth decade compared to the general population [9]. Subpopulations of CHIP with DNMT3A mutations were significantly associated with higher serum interferon-gamma levels [9]. In patients with inflammatory bowel disease co-occurring with haematologic disease, Cumbo et al. identified 85% of CHIP variants and suggested a biological link between the proinflammatory environment and haematologic malignancy onset [10].

Behçet’s disease (BD) is a chronic inflammatory immune-mediated disorder that causes recurrent oral and genital ulcers and skin, joint, eye, neurological, and gastrointestinal involvement [11, 12]. Patients with BD have a higher risk of haematological malignancy (pooled relative risk, 2.58; 95% confidence interval [CI], 1.61–3.55) [13, 14]. However, no studies have investigated the relationship between CHIP and BD. Therefore, here, we aimed to characterise the occurrence of CHIP in patients with BD and analyse the associations between CHIP and inflammatory markers in patients with BD and their outcomes.

Methods

Study design and population

For the control group, we collected blood samples from participants during a routine health check-up for screening purposes at the Seoul National University Hospital Healthcare System Gangnam Center between January 2014 and January 2017 (no. 1980–121-1056). For the BD group, we collected blood samples from patients with BD during routine outpatient visits from two tertiary hospitals (Severance Hospital in Seoul, Korea and Seoul National University Hospital in Seoul, South Korea) between March 2009 and September 2021. Systemic BD was diagnosed based on the International Criteria for Behçet’s Disease [15]. Patients with intestinal BD classified as having definite, probable, or suspected intestinal BD were enrolled in this study based on established diagnostic criteria [16]. This study was conducted in accordance with the ethical guidelines of the Declaration of Helsinki. Informed consent was obtained from all individuals, and the study protocol was approved by the Institutional Review Board of Seoul National University (no. 1606–116-772) and Severance Hospital (no. 2012–0039-030).

Targeted gene sequencing and variant annotation

Genomic DNA was extracted from the peripheral blood. Targeted next-generation sequencing was performed using a custom panel consisting of 89 CHIP genes frequently involved in CHIP, DNMT3A, TET2, ASXL1, JAK2, and TP53, with an average target depth of over 1000 × . All non-synonymous variants with a variant allele frequency of 1.5–30% were considered CHIP variants. Common germline variants listed in the gnomAD,10–0 Genomes v3, ESP6500, and ExAC databases and an internal panel of 1000 Koreans were excluded.

Data collection and study outcomes

Demographic and clinical information was collected at outpatient visits, including age, disease duration, sex, body mass index, smoking history, presence of intestinal BD, previous and current medications, and laboratory findings. Medical accelerations for BD, including immunomodulators and biologic agents, surgical requirements for BD, BD-related hospitalisation, and BD-related emergency room visits, were also investigated to determine the patients’ long-term outcomes.

Statistical analysis

Continuous data are expressed as mean and standard deviation (mean ± SD) or median and range, while categorical variables are expressed as proportion (%). An independent t-test (or Mann–Whitney U test) was used to compare the continuous variables, while the Χ2 test (or Fisher’s exact test) was used for categorical variables as appropriate. Multivariate logistic regression analyses were performed to identify the independent risk factors for CHIP after the adjustment for various confounders. All statistical analyses were assessed with the Statistical Package for Social Sciences (SPSS version 24.0; IBM Corp., Armonk, NY, USA). p values < 0.05 were considered statistically significant.

Results

Association between CHIP and BD development

Blood samples were collected from 5004 healthy controls (control group) and 117 BD patients (BD group). We identified CHIP in 13.9% of the control group and 11.1% of the BD group, indicating that there was no significant difference between the two groups (Fig. 1). In patients with BD, five variants (DNMT3A, TET2, ASXL1, STAG2, and IDH2) were detected in our cohort, and DNMT3A mutations were the most common (7.5%), followed by TET2 mutations (1.9%).

Fig. 1
figure 1

Forest plot of adjusted odds ratios for the incidence of Behçet’s disease adjusted for age and sex. CHIP, clonal haematopoiesis of indeterminate potential; OR, odds ratio

Full size image

CHIP and disease activity of BD

Patients with BD and CHIP were older and had a higher platelet count, erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) level and lower albumin level at diagnosis than BD patients without CHIP (Table 1). Patients with a higher baseline ESR showed a higher CHIP-positive tendency than those with a lower baseline ESR (4.0%, 8.0%, 13.3%, and 28.6%, respectively; p trend = 0.067; Fig. 2A). Patients with a higher baseline CRP level showed a significantly higher CHIP-positive rate than those with a lower baseline CRP level (0.0%, 10.0%, 19.2%, and 27.9%, respectively; p trend = 0.015; Fig. 2B). Patients with a higher baseline albumin level showed a significantly lower CHIP-positive rate than those with a lower baseline albumin level (23.8%, 9.4%, 16.0%, and 2.6%, respectively; p trend = 0.035; Fig. 2C). Older patients showed a significantly higher CHIP positivity rate than younger patients (0.0%, 6.5%, 15.4%, and 23.3%, respectively; p trend = 0.003; Fig. 2D).

Table 1 Baseline characteristics of Behcet’s disease (N = 117)
Full size table
Fig. 2
figure 2

CHIP positivity rate according to A ESR, B CRP level, C albumin level at diagnosis, and D age at blood sampling. CHIP, clonal haematopoiesis of indeterminate potential; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate

Full size image

A univariate logistic analysis was performed to determine independent factors associated with CHIP in patients with BD. Older age at blood sampling (odds ratio [OR], 1.090; 95% CI, 1.030–1.153; p = 0.003), high platelet count at diagnosis (OR, 1.000; 95% CI, 1.000–1.000; p = 0.012), and high ESR at diagnosis (OR, 1.027; 95% CI, 1.009–1.046; p = 0.004) were positively associated with CHIP (Table 2). A high albumin level (OR, 0.228; 95% CI, 0.058–0.887; p = 0.033) was negatively associated with CHIP (Table 2). Age at blood sampling was the only independent factor for CHIP after multivariate logistic analysis including various confounders (OR, 1.099; 95% CI, 1.020–1.184; p = 0.013) (Table 2).

Table 2 Logistic regression analysis for CHIP (N = 117)
Full size table

CHIP and long-term clinical outcomes of BD

The development of CHIP in patients with BD was not associated with medical acceleration, surgical requirement, BD-related hospitalisation, or BD-related emergency room visits (Table 1). Moreover, CHIP was not an independent risk factor for poor clinical composite outcomes, including medical acceleration, surgery, hospitalisation, or emergency room visits, in patients with BD, in the Cox regression analysis (data not shown).

There was one patient with aplastic anaemia and two patients with myelodysplastic syndrome in our cohort. However, there was no CHIP mutation in the three patients.

Discussion

To our knowledge, this is the first report to describe the development of CHIP in patients with BD and analysing the association between CHIP and inflammation in BD. We determined that the rate of CHIP in BD patients was not increased compared to the general population. The DNMT3A and TET2 mutations were dominant in the control and BD groups. The CHIP rate significantly increased with age at the time of blood sampling. Furthermore, it was associated with a higher CRP level and ESR but negatively with albumin level at diagnosis. Finally, it was not associated with the clinical outcomes of BD.

Ageing, the strongest risk factor for CHIP development in patients with BD, is widely accepted and considered the biggest contributor to the accumulation of somatic mutations. Its intrinsic mechanisms include haematopoietic stem cell–induced age-dependent DNA damage, telomere attrition, and epigenetic dysregulation involving DNMT3A and TET2 [17]. Interestingly, in addition to intrinsic mechanisms, various environmental stressors such as cytotoxic stress, autoimmunity, and infection contribute to CHIP emergence [18]. Here, we investigated the association between CHIP and elevated levels of inflammatory factors in patients with BD. Although the significance disappeared after the adjustment for age, ESR, CRP, and albumin levels were confirmed as highly correlated with CHIP in patients with BD by univariate logistic regression analysis. Inflammatory markers, including white blood cell and neutrophil counts, ESR, and CRP level, were not elevated in the general population with CHIP in a previous study [19]. However, CHIP carriers have higher CRP levels among patients with prior coronary artery disease [20]. In line with the results of previous studies, our results support the theory that CHIP is regulated by chronic inflammation, reflecting higher serum ESR and CRP and albumin levels in CHIP( +) BD patients. Regarding elevated platelet numbers in CHIP( +) patients, similar results were reported in a previous study (294,000/µL vs. 241,000/µL, p = 0.021) [21]. Regarding long-term clinical outcomes, CHIP could not predict medical acceleration, surgical requirement, BD-related hospitalisation, or BD-related emergency room visits. We appreciate the reviewer’s comment. There were some limitations in our study. First, there was a lack of validation using an independent cohort. Second, this study had low statistical power due to the small number of cases. Third, there is the possibility of other unknown variants associated with BD. We added these issues in the “Discussion” section.

Conclusions

In summary, our study found that older age and degree of inflammation in BD might be associated with CHIP emergence.