17 January 2020 in Nature Communications
Infections have become the major cause of morbidity and mortality among patients with chronic lymphocytic leukemia (CLL) due to immune dysfunction and cytotoxic CLL treatment. Yet, predictive models for infection are missing. In this work, we develop the CLL Treatment-Infection Model (CLL-TIM) that identifies patients at risk of infection or CLL treatment within 2 years of diagnosis as validated on both internal and external cohorts. CLL-TIM is an ensemble algorithm composed of 28 machine learning algorithms based on data from 4,149 patients with CLL. The model is capable of dealing with heterogeneous data, including the high rates of missing data to be expected in the real-world setting, with a precision of 72% and a recall of 75%. To address concerns regarding the use of complex machine learning algorithms in the clinic, for each patient with CLL, CLL-TIM provides explainable predictions through uncertainty estimates and personalized risk factors.
24 December 2022 in Leukemia
Recent evidence suggests that the prognostic impact of gene mutations in patients with chronic lymphocytic leukemia (CLL) may differ depending on the immunoglobulin heavy variable (IGHV) gene somatic hypermutation (SHM) status. In this study, we assessed the impact of nine recurrently mutated genes (BIRC3, EGR2, MYD88, NFKBIE, NOTCH1, POT1, SF3B1, TP53, and XPO1) in pre-treatment samples from 4580 patients with CLL, using time-to-first-treatment (TTFT) as the primary end-point in relation to IGHV gene SHM status. Mutations were detected in 1588 (34.7%) patients at frequencies ranging from 2.3–9.8% with mutations in NOTCH1 being the most frequent. In both univariate and multivariate analyses, mutations in all genes except MYD88 were associated with a significantly shorter TTFT. In multivariate analysis of Binet stage A patients, performed separately for IGHV-mutated (M-CLL) and unmutated CLL (U-CLL), a different spectrum of gene alterations independently predicted short TTFT within the two subgroups. While SF3B1 and XPO1 mutations were independent prognostic variables in both U-CLL and M-CLL, TP53, BIRC3 and EGR2 aberrations were significant predictors only in U-CLL, and NOTCH1 and NFKBIE only in M-CLL. Our findings underscore the need for a compartmentalized approach to identify high-risk patients, particularly among M-CLL patients, with potential implications for stratified management.
8 March 2018 in Haematologica
CECasper Tabassum Eriksen
+1
7 May 2021 in Clinical Cancer Research
Purpose: TP53 aberration (TP53 mutation and/or 17p deletion) is the most important predictive marker in chronic lymphocytic leukemia (CLL). Although each TP53 aberration is considered an equal prognosticator, the prognostic value of carrying isolated (single-hit) or multiple (multi-hit) TP53 aberrations remains unclear, particularly in the context of targeted agents. Patients and Methods: We performed deep sequencing of TP53 using baseline samples collected from 51 TP53 aberrant patients treated with ibrutinib in a phase II study (NCT01500733). Results: We identified TP53 mutations in 43 patients (84%) and del(17p) in 47 (92%); 9 and 42 patients carried single-hit and multi-hit TP53, respectively. The multi-hit TP53 subgroup was enriched with younger patients who had prior treatments and unmutated immunoglobulin heavy-chain variable region gene status. We observed significantly shorter overall survival, progression-free survival (PFS), and time-to-progression (TTP) in patients with multi-hit TP53 compared with those with single-hit TP53. Clinical outcomes were similar in patient subgroups stratified by 2 or >2 TP53 aberrations. In multivariable analyses, multi-hit TP53 CLL was independently associated with inferior PFS and TTP. In sensitivity analyses, excluding mutations below 1% VAF demonstrated similar outcome. Results were validated in an independent population-based cohort of 112 patients with CLL treated with ibrutinib. Conclusions: In this study, single-hit TP53 defines a distinct subgroup of patients with an excellent long-term response to single-agent ibrutinib, whereas multi-hit TP53 is independently associated with shorter PFS. These results warrant further investigations on prognostication and management of multi-hit TP53 CLL. See related commentary by Bomben et al., p. 4462