Refined Continuous Risk Index for Accurately Predicting Outcomes of Patients With Chronic Lymphocytic Leukemia After Limited-Duration Therapy

Vollständiger Abstract

Worum geht es in dieser Arbeit?

PURPOSE Measurable residual disease (MRD) status after limited-duration treatment for chronic lymphocytic leukemia (CLL) strongly predicts survival times and is considered a surrogate end point. However, most prognostic indices, including the International Prognostic Index (CLL-IPI), rely solely on pretreatment features. To incorporate dynamic MRD data in prognostic models, the continuous individualized risk index (CIRI) was proposed in 2019. Since developed in the context of chemoimmunotherapy, we here propose a refined version of CIRI, in the context of fixed-duration targeted therapy, which includes MRD status at interim, end-of-treatment (EoT) and, in addition, 12-month post-EoT (CIRI2-CLL). METHODS After developing model parameters from the CLL8/10/11 and MURANO trials, we applied CIRI2-CLL to independent validation sets from the CLL14 (venetoclax-obinutuzumab v chlorambucil-obinutuzumab) and CLL13 (venetoclax-obinutuzumab with or without ibrutinib v chemoimmunotherapy) trials and assessed model calibration, stratification, and performance. RESULTS We compared CIRI2-CLL with individual indices and found good calibration with an approximately 5% difference between observed and predicted event probabilities for progression-free survival (PFS). CIRI2-CLL demonstrated superior performance compared with individual indices as measured by C-statistics between 0.82 and 0.98 at all time points, including MRD and CLL-IPI, improving PFS prediction by 14%-37% from 2 to 5 years. When stratifying patients by CIRI2-CLL into low-, intermediate-, and high-risk groups, the corresponding 3-year PFS rates were 100.0%, 70.2%, and 10.8%. Performance was maintained for overall survival. CONCLUSION These results validate CIRI2-CLL in the context of limited-duration CLL therapy. Our approach suggests the models' adaptability to emerging additional longitudinal MRD and/or outcome data. By introducing CIRI2-CLL, we offer a dynamic tool for investigators to reliably identify patients with increased risk of disease relapse after limited-duration therapy with venetoclax and obinutuzumab, freely accessible at CIRI2 (Stanford University), with important implications for informed decision making and for future risk-adapted CLL trial design.