Paper Details

ABSTRACT

Introduction

Telomere biology disorders (TBDs) comprise a group of syndromes caused by pathogenic germline variants (PGVs) in telomere-maintenance genes. Typical clinical manifestations include bone marrow failure, pulmonary fibrosis, liver disease, and cancer predisposition. The commonest hematologic cancers are myeloid neoplasms (MNs), particularly, myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). Risk of MN in TBDs has been associated with preceding clonal hematopoiesis, primarily affecting the DNA damage response pathway. Long-term outcomes and response to MN therapies remains poorly described.

Methods

Patients had a TBD diagnosis, lymphocyte telomere length (TL) <10^th percentile for age (except biallelic CTC1 variants), and a pathogenic/likely pathogenic variant in a TBD-associated gene or, if a variant of uncertain significance, had familial disease segregation, functional data demonstrating telomere impairment, or at least two of the following: 1) evidence of hematologic involvement; 2) interstitial lung disease; 3) chronic liver disease; 4) mucocutaneous features of TBD; 5) significant family history. Patients with one allele of only known autosomal recessive disease (AR) or females with X-linked recessive (XLR) genes were excluded. Data were retrospectively extracted from medical records from 15 institutions. Cumulative incidence (CI) of MN was determined from date of birth, and censored at hematopoietic cell transplant (HCT). Overall survival from birth and from MN development were determined by Kaplan-Meier analysis. All sites obtained IRB approval.

Results

A total of 795 TBD patients were included: autosomal dominant (AD) genotypes included: TERT (n=242), TERC (n=121), RTEL1 (n=118), PARN (n=23), NAF1 (n=4), POT1 (n=3), ZCCHC8 (n=2). AR and XLR genotypes included: DKC1 (n=32), RTEL1 (n=16), TERT (n=15), CTC1 (n=8), WRAP53 (n=5), PARN (n=2), ACD (n=1). There were 29 patients with TINF2 and 174 with gene unknown. MN occurred in 111 patients (14%). Median time to development of first MN was 50 years.

MN was rare in children. Incidence of MN increased with age: CI=1% by age 20, 5% by age 40 and 17% by age 60. There was no significant difference in CI of MN by inheritance mode. Patients with TERC had increased CI of MN compared with TERT (P=0.03); 9% vs 4% by age 40 and 19% vs 11% by age 60, respectively.

Data for categorization were available for 98 patients at first MN diagnosis. Fifty-eight (59%) had low blast count (<5%) disease, MDS (n=52), MPN/CMML (n=5), or CCUS with high VAF TP53 (n=1) while 40 patients had high blast count (>5%) disease, MDS with increased blasts (IB) (n=22) or AML (n=18). Karyotype in those presenting initially with MDS-IB or AML was complex (n=16; 48%), followed by chromosome 7 abnormalities (n=7; 21%), normal (n=5; 15%), other isolated (n=5; 15%), and unknown (n=7). Somatic mutation profile at MN diagnosis in MDS-IB and AML patients was limited to 27. Most (n=20; 87%) had >1 somatic mutation (VAF >2%). Commonest somatic mutation was TP53 (n=13 patients; 48%), multi-hit in 6 patients. Splicing factor mutations occurred in 6, most commonly U2AF1^S34F, never in isolation. Ras pathway mutations occurred in 6 patients, most commonly PTPN11 (n=5), none of whom had MPN/CMML. Only two patients had FLT3 mutations, one of whom had concurrent NPM1.

Median OS after MN diagnosis was 2.4 years. There was no difference in OS from MN diagnosis by inheritance or genotype. Median OS for MDS-IB or AML was 1.1 years, and 3.5 years in those with low blast count MDS or MPN/CMML. Initial therapies for MDS-IB and AML included intensive chemotherapy (n=14), hypomethylating agent (HMA) + venetoclax (n=17), HMA alone (n=6), HMA + targeted therapy (n=1), and upfront HCT (n=1). Median OS was 1.2 years in those who received induction, 1.3 years for HMA + venetoclax, and 1.75 years for HMA alone. Ultimately 18 (41%) MDS-IB or AML patients proceeded to HCT; 8 relapsed, 7 had disease remission, 2 died shortly after HCT, due to PTLD and multiorgan failure. One patient who achieved remission post-HCT later died of lung disease.

Conclusion

Risk of MN in TBD patients increases with age and is rare in children. The clonal profile of MDS-IB and AML is distinctive, dominated by a high risk somatic profile: complex karyotype and TP53 mutations. Median OS is poor in TBD-patients once MN occurs, especially MDS-IB and AML. Further studies are needed to determine optimal intervention timing and therapy for MNs in TBD.