Clonal Hematopoiesis in Nonhuman Primates: Harnessing Shared Evolution to Model Clonal Selection in Human Disease

Clonal hematopoiesis (CH) describes the insidious expansion of hematopoietic stem and progenitor cells (HSPCs) that have acquired somatic mutations in healthy persons without overt cytopenias, myelodysplasia, or leukemia.^1,2  Numerous studies have shown that this phenomenon becomes more common as humans age and that the clone size expands over time.^3-6  The clinical importance of CH is due to its association with an increased risk of hematologic malignancies (particularly myeloid neoplasms, although data is emerging for increased risk of lymphoid neoplasms as well), cardiovascular and other inflammatory diseases and all-cause mortality.^2,7-10  Given the systemic implications of CH, several large academic centers have developed multidisciplinary clinics with a goal of preventative management, early intervention and coordination of collaborative clinical trials. However, the long natural history of CH creates a challenge for the design of prospective studies in humans, which remain sorely needed to elucidate underlying mechanisms and target for intervention.

The subject of CH has been chosen for several previous “Year’s Best” articles (2014, Dr. Ravi Majeti; 2015, Dr. Pamela Becker; 2017, Drs. Joana Conant and Tracy George; 2018, Dr. Andrew Roberts; 2020 Drs. Mark Evans and Annette Kim; and 2021 – Dr. Adam Mead). We selected the topic of CH as one of the Year’s Best in 2022 for the work done in the laboratory of Dr. Cynthia Dunbar at the National Heart, Lung, and Blood Institute at the National Institutes of Health. Dr. Dunbar’s lab is one of only a few in the world that has the capability to study hematopoiesis in nonhuman primates, and she and her colleagues have used this primate model to investigate CH in aged rhesus macaques (RM).¹¹  They have furthered their investigations with the use of CRISPR-Cas9 technology to engineer loss-of-function mutations in DNMT3A, TET2, and ASXL1 (epigenetic regulators most commonly implicated in human CH^2,5,7 ) in HSPCs and transplanted these altered autologous cells into young adult RM to examine the effect of the mutations on hematopoiesis, clonal evolution, and inflammation. This work, first presented as an abstract at the 2020 ASH Annual Meeting,¹²  represents a marked innovation in the area of CH. While murine models have greatly contributed to our understanding of the properties of HSPCs and hematopoiesis, studying CH in mice has intrinsic limitations. However, nonhuman primates have similar marrow and HSPC function as compared to humans.¹³  Thus, this model more closely mimics the natural occurrence of CH in humans, allows for expedited investigation of the mechanisms underlying CH and malignant transformation and provides a preclinical platform for testing of interventions that could have far-reaching implications on hematologic, cardiovascular, and inflammatory disease in our aging populations.

In the article, Dr. Tae-Hoon Shin and colleagues followed 60 aged macaques (median age of 25 years) and performed error-corrected deep sequencing of 56 genes that are recurrently mutated in human CH.¹¹  In 12 animals, somatic mutations (13 total; a single mutation in 11 animals and 2 mutations in 1 animal) were identified with a median variant allele frequency (VAF) greater than 1 percent. Four animals had a mutation with VAF of 2 percent or higher. The mutation profile was similar, though not identical to that found in human CH: DNMT3A, RUNX1, TP53, NOTCH1, CREBBP, and TET2. The DNMT3A gene contained the most frequent driver mutations in the cohort (4 animals), had the highest VAF, and demonstrated clonal expansion in longitudinal samples obtained during a study period of one to five years. All animals showed normal complete blood counts and had no evidence of hematologic malignancies.

For the autologous transplant model, HSPCs from three young adult macaques (ages 3-9 years) were edited to generate CRISPR-Cas9–mediated loss-of-function mutations in DNMT3A, TET2, and ASXL1. The animals were treated with total body irradiation and underwent autologous transplantation using the edited HSPCs as the stem-cell product. All animals engrafted and maintained normal blood counts for 55 months following transplant. The animals demonstrated marked clonal expansion of TET2-mutated HSPCs, limited clonal expansion of DNMT3A-mutated HSPCs, and no significant expansion of ASXL1-mutated HSPCs. Morphologic examination of the marrow from two animals with the highest TET2 VAFs revealed hypercellularity with myeloid shift, but neither marrow had dysplasia or an increase in blast forms. Secondary somatic mutations have not been detected in the animals to date.

The studies exploring the inflammatory characteristics associated with CH in this model were of equal importance. A series of experiments performed on TET2-mutant colony-forming units, isolated macrophages, bone marrow plasma and blood from the CH macaques, revealed increases in inflammatory mediators, particularly IL-1B and IL-6 as compared to control animals. It is proposed that the inflammation associated with CH provides favorable conditions for clonal expansion.¹⁴  Thus, the authors used the model to test whether pharmacological blockade of IL-6 signaling with the monoclonal antibody tocilizumab could disrupt clonal expansion. The three young adult CH macaques were treated with tocilizumab 10 mg/kg weekly for four months. The treatment was well-tolerated and appeared to slow the growth rate of the clone and decrease the clone size (as measured by TET2 VAF) in one animal.

In summary, CH has significant clinical implications for patient outcomes beyond the field of hematology. The publication by Dr. Shin and colleagues reaffirms the relationship between driver mutations in epigenetic regulators, CH, and inflammation in other primates. This innovative model should continue to enrich our understanding of the pathobiology of CH and its impact on human health, as well as reveal new avenues for translational discovery. Since the first observation that CH occurs in individuals without cancer,¹⁵  determining the mechanism(s) that controls the evolution from indolent CH to a lethal hematologic malignancy has been a primary goal for the field. As hematologists, a key objective is to identify the individuals at risk of such a malignant transformation so that, ultimately, an intervention may occur. The RM model will undoubtedly serve as an important tool to accomplish these objectives.