Clonal Hematopoiesis Risk Score for Myeloid Neoplasms in Patients With Myeloid Precursor Conditions

Get Permission

As reported in NEJM Evidence, Weeks et al identified a clonal hematopoiesis risk score for prediction of risk for myeloid neoplasms among individuals with clonal hematopoiesis of indeterminate potential (CHIP) and clonal cytopenia of undetermined significance (CCUS).

As stated by the investigators, “CHIP and CCUS are defined by somatic mutations in genes associated with myeloid neoplasms at a variant allele fraction (VAF) of 0.02 or greater in the absence and presence of cytopenia, respectively. CHIP/CCUS is highly prevalent in adults, and defining predictors of myeloid neoplasm risk would aid clinical management and research.”

Study Details

The study involved analysis of sequenced exomes of 438,890 healthy UK Biobank (UKB) participants in derivation and validation cohorts. Mutations, laboratory values, and myeloid neoplasm outcomes were used in conditional probability-based recursive partitioning and Cox regression to determine predictors of incident myeloid neoplasms. Combined statistical weights were used to define the clonal hematopoiesis risk score.

Key Findings

In a UKB derivation cohort of 193,743 individuals, 11,337 met criteria for CHIP/CCUS. Among these 11,337 individuals, 269 (2.37%) had incident myeloid neoplasm events during follow-up.

Recursive partitioning distinguished individuals with CHIP/CCUS in the derivation cohort with 10-year probabilities of developing a myeloid neoplasm ranging from 0.0077 to 0.85. Multivariate analysis validated partitioning variables as predictors of myeloid neoplasms: factors significantly associated with risk of myeloid neoplasms and incorporated into the clonal hematopoiesis risk score consisted of single DNMT3A mutations (hazard ratio [HR] = 0.578, P = .00307), high-risk mutations (HR = 4.72,  P < 2×10–16), presence of two or more mutations (HR = 2.32, P = 2.88×10–9), VAF of  ≥ 0.2 (HR = 2.63, P = 3.66×10–12), age65 years (HR = 1.53, P = .000596), CCUS vs CHIP (HR = 1.74, P = .000324), and the red blood cell indices of red cell distribution width of ≥ 15% (HR = 3.63, P < 2×10–16) and mean corpuscular volume ≥ 100 fL (HR = 4.03, P < 2×10–16).

The clonal hematopoiesis risk score was used to define low-risk (n = 10,018, 88.4%), intermediate-risk (n = 1,196, 10.5%), and high-risk (n = 123, 1.1%) groups in the derivation cohort. The 10-year cumulative incidence of myeloid neoplasms was 0.669 ± 0.0827%, 7.83 ± 0.807%, and 52.2 ± 4.96% in the low-, intermediate-, and high-risk groups, respectively.

In a validation cohort of 245,147 UKB participants, the c-index for clonal hematopoiesis risk score was 0.799 ± 0.015.

In two clinical cohorts of 646 (29.6% clonal hematopoiesis risk score = high-risk) and 99 (34.3% clonal hematopoiesis risk score = high-risk) hematology patients with CHIP/CCUS, most myeloid neoplasm events occurred in clonal hematopoiesis risk score high-risk patients (23 of 31 events and 20 of 34 events). Hazard ratios for myeloid neoplasms compared with patients with clonal hematopoiesis risk score low-risk scores in the two cohorts were 4.53 and 2.87 among intermediate-risk patients, and 40.3 and 8.80 among high-risk patients.     

The clonal hematopoiesis risk score calculator can be accessed at

The investigators concluded, “The clonal hematopoiesis risk score provides a simple prognostic framework for CHIP/CCUS, distinguishing a high-risk minority from the majority of CHIP/CCUS, which has a minimal risk of progression to myeloid neoplasm.”

Benjamin L. Ebert, MD, PhD, of Dana-Farber Cancer Institute, is the corresponding author for the NEJM Evidence article.  

Disclosure: The study was funded by the National Institutes of Health, Harold Amos Medical Faculty Development Program, and others. For full disclosures of the study authors, visit

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.