Association of Clonal Hematopoiesis and Dysgranulopoiesis with Germline ETV6 Carriers without Hematological Malignancies

Introduction

Germline predisposition to hematological malignancies (HM) includes a category characterized by thrombocytopenia and an increased risk of HM due to germline mutations in RUNX1, ETV6, and ANKRD26. ETV6 is the most recently discovered gene and, thus, the least understood. Individuals with germline ETV6 mutations (gETV6) often exhibit mild-moderate thrombocytopenia, bleeding tendencies and a lifetime risk of 30% of developing HM. Potential pre-malignant findings in gETV6 carriers, such as clonal hematopoiesis (CH) and bone marrow (BM) dysplasia, are understudied. In contrast, CH has been described in ~40% and 14% of unaffected carriers of gRUNX1 and gANKRD26, respectively, and dysmegakaryopoiesis has been reported in both.

This study aims to outline the clinical and genetic features of gETV6 carriers with the goal of better understanding the pathophysiology of HM in which gETV6 mutations are implicated.

Methods

Individuals with gETV6 mutation without HM diagnosis are referred to as “carriers-without HM”, while those with gETV6 and a HM diagnosis are referred to as “carriers-with HM”. Noncarrier family members (i.e. ETV6 wild type) are referred to as “controls”. All germline variants were confirmed in hair follicles in carriers-with HM and in peripheral blood (PB) in carriers-without HM. The ACMG criteria were followed for variant classification. Carriers-without HM underwent baseline BM aspirates (BMA) and annual follow-ups with a complete blood count and BMA when indicated. The bleeding phenotype was assessed using the ISTH-SSC Bleeding Assessment Tool (ISTH-BAT), and CH was evaluated with a NGS gene panel including recurrently myeloid genes, in either BM or PB.

Results

We enrolled 16 carriers and 17 controls from 5 unrelated families from 2 centers. The 5 families had 4 unique gETV6 variants, with two sharing the same amino acid change: c.1196G>T (R399L), c.1195C>T (R399L), c.163C>T (R55C), and c.463+1_463+2dupGT (p.?). The HM incidence in our cohort was 17%, with 2 carriers-with HM: one with acute myeloid leukemia at age 52 (gETV6: R55C) and one with chronic myelomonocytic leukemia type 2 at age 63 (gETV6: c.463+1_463+2dupGT).

The median age at germline assessment for carriers-without HM was 48 years (IQR: 28-54) and 46 years (IQR: 26-54) for controls. The mean follow-up time of carriers-without HM was 23 months (IQR 15-31). Carriers-without HM had decreased platelet counts compared to controls: 129.0 x10⁹/L (IQR: 116-186) vs. (260.4 x10⁹/L (IQR: 237-282) (p< 0.01), with 64% (9/14) of them with counts <150 x10⁹/L. The average mean platelet volume was similar between carriers-without HM and controls (10.3 vs. 9.6 fL, p=0.15), even between those with thrombocytopenia. Other blood count values were also comparable. Sixty-four percent (9/14) of carriers-without HM reported a bleeding tendency, but only 36% (5/14) had an abnormal ISTH-BAT score, all from families with the R399L missense variant, while all controls had normal ISTH-BAT scores (p< 0.01).

BMA was performed in 71% (10/14) of carriers-without HM. Abnormal cellularity was observed in 50%: 20% were hypercellular and 30% hypocellular. Moreover, 70% exhibited increased number of megakaryocytes with normal shape. Notably, 80% showed significant dysgranulopoiesis (i.e. dysplasia > 30%).

Finally, NGS was performed in 12 carriers-without HM, revealing that 42% (5/12) had ≥1 acquired mutation in genes: KRAS (G12R, VAF 1.3%), EZH2 (Y579S, VAF 40%), TP53 (K394Q VAF 4.3%), DNMT3A (I898N, 3%), JAK2 (V617F, VAF 16%), NFE2 (Y257Cfs, VAF 10%), PTPN11 (L262H VAF 5%), CBL (F434I, VAF 2.3% & C384R VAF2.4%). Notably, all of them belonged to the two families with the R399L gETV6.

Conclusion

To our knowledge, this is the first study that associates CH with gETV6 carriers-without HM and the largest series to explore dysplastic BM features. Our study contributes to the limited understanding of gETV6 within the hematologic malignancy predisposition category. We showed that CH in gETV6 is common but limited to R399L carriers in our cohort, with its incidence closer to that described for gRUNX1 than for gANKRD26 carriers. Unlike the other two predisposition entities with pre-existing thrombocytopenia, dysgranulopoiesis was frequent while dysmegakaryopoiesis was absent. Further longitudinal studies with larger cohorts are essential to validate these findings and to determine how CH and BM dysplasia may affect HM development in gETV6 carriers.