Abstract
Telomere length is a prognostic factor in chronic lymphocytic leukemia (CLL), yet its relevance under targeted therapies remains unclear. Here, we analyzed telomere length using qPCR in 917 samples from treatment-naïve CLL patients without TP53 aberrations enrolled in the GAIA/CLL13 trial comparing venetoclax-based combinations with rituximab (RV), obinutuzumab (GV), or obinutuzumab plus ibrutinib (GIV) versus chemoimmunotherapy (CIT). For CLL with TP53 aberrations, samples from 41 treatment-naïve patients receiving GIV in the CLL2-GIVe trial were analyzed. The median telomere lengths in GAIA/CLL13 and CLL2-GIVe were 4.0 kb (1.2-15.2) and 2.8 kb (1.4-14.0), respectively. A prognostic cutoff of 4.17 kb, based on GAIA/CLL13, was used for dichotomization into short and long telomeres. Within GAIA/CLL13, short telomeres significantly correlated with adverse clinical features including advanced Binet stage, high ECOG, high serum ß2-microglobulin and thymidine kinase levels, high lymphocyte count and large lymph nodes (all P<0.01). Among genetic features, short telomeres significantly associated with unmutated IGHV, del(11q), mutations in NOTCH1, SF3B1, XPO1, RPS15, EGR2 and NFKBIE, and karyotypic complexity (all P<0.01). Short telomeres were associated with shorter progression-free survival (PFS) within CIT, RV, GV, GIV in GAIA/CLL13 (all P<0.01) and in CLL2-GIVe (dichotomized by median; P=0.04), and impacted overall survival only in the CIT arm (P=0.024). Multivariable analysis suggested telomere length as an independent prognostic factor for PFS in the overall GAIA/CLL13 cohort. These findings propose telomere length as a prognostic marker for PFS in CLL patients receiving venetoclax-based first-line therapy and could be of relevance for risk stratification in the modern treatment era. NCT02950051