Venetoclax May Overcome the Prognostic Disadvantage of Short Telomeres Regarding Progression Free Survival in Previously Untreated CLL Patients with Coexisting Conditions: A Study By the German CLL Study Group (GCLLSG)

The prognostic role of telomere length (TL) has already been investigated for CLL patients treated with chemo-immunotherapy. However, the role of TL in the era of targeted therapies remains unclear. The CLL14 trial demonstrated superior long-term efficacy of fixed-duration therapy with venetoclax-obinutuzumab (Ven-Obi) compared to chlorambucil-obinutuzumab (Clb-Obi). Previous studies have mainly correlated TL and clinical outcomes in younger and fit CLL patients. Given that TL is associated with both proliferation and age, we aimed to provide data from the CLL14 trial consisting of a previously untreated patient population with coexisting conditions, featuring a median age of 72 years and a median CIRS score of 8.

We analyzed baseline TL data by flow-FISH from randomly chosen and IGHV balanced 210 (n=93 Clb-Obi and n=107 Ven-Obi) of 432 (48.6%) patients in the CLL14 trial, examining its association with clinical outcomes after a median follow-up of 79.8 months. The flow-FISH technique offers high precision, reproducibility, and sensitivity, enabling measurement of TL in CLL cells without contamination by non-malignant cells.

TL, measured in triplicate for each patient, ranged from 0.931 to 18.463 kilo bases (kB), with a median of 4.725 kB. There was no significant difference in TL between the Clb-Obi arm with a median of 4.653 kB and the Ven-Obi arm with a median of 4.824 kB. An optimal TL cutoff value was chosen to minimize the Akaike information criterion (AIC, an estimator of prediction error) value for Cox proportional hazards regression of progression-free survival (PFS: short vs. long), ensuring at least 10 PFS events per group. Short telomeres were defined as <6.293 kB. In total, 151 of 210 (71.9%) patients had short telomeres (n=65 Clb-Obi and n=86 Ven-Obi) and 59 (28.1%) had long telomeres (n=28 in Clb-Obi and n=31 in Ven-Obi). Short TL was significantly associated with high-risk clinical and genetic parameters such as CLL-IPI, TLS risk, creatinine clearance, β2m, serum thymidine kinase, nodal size, ALC, mutational status of IGHV, mutated ATM and del11q.

In the full cohort, patients with short TL had a significantly shorter PFS (median PFS 48.4 vs. 62.2 months; hazard ratio [HR] 1.709 [95% CI 1.130-2.584], P=0.011). The estimated OS rate in the full cohort at six years was 71.1% for patients with short TL vs. 82.5% for patients with long TL (HR 1.918 [95% CI 0.966-3.806], P=0.068). A significant association between TL and PFS was also observed in the Clb-Obi arm (median PFS 28.3 vs. 47.6 months; HR 2.165 [95% CI 1.249-3.754], P=0.006), but not in the Ven-Obi arm (median PFS 64.7 months vs. not reached, HR 1.590 [95% CI 0.843-2.997], P=0.152). This significance persisted in the Clb-Obi arm when dichotomizing TL by median value (<4.725 kB) or considering TL as a continuous variable. Additionally, patients with short TL had significantly longer PFS in the Ven-Obi group compared to the Clb-Obi group (HR 0.344 [95% CI 0.230-0.513], P<0.001), contrary to patients with long TL (HR 0.531 [95% CI 0.253-1.114], P=0.094).

In our population, TL was associated with genetic and clinical markers indicative of proliferation. This study confirms that short TL is linked to inferior PFS in chemo-immunotherapy-treated (Clb-Obi) CLL patients, whereas in the Ven-Obi group, outcomes were not associated with TL, suggesting that the predictive impact of TL was neutralized by Ven-Obi treatment. To refine our understanding, we plan to compare our results with further trials on targeted therapies, considering methodologies for TL determination and patient age.