CLL-IPI: valid in the era of oral inhibitors?

In this issue of Blood, Parikh et al¹  show in their analysis for the first time that the chronic lymphocytic leukemia international prognostic index (CLL-IPI) can predict time to first diagnosis and overall survival (OS) in a cohort of 969 patients with monoclonal B-cell lymphocytosis (MBL) and Rai 0 stage CLL. This might prove to be a very useful tool for the management of these classical watch-and-wait patients who need to be monitored for progression or disease-related symptoms and potential treatment indication on a regular basis.

The CLL-IPI²  is a well-established score in CLL and was originally developed based on data from 8 phase 3 studies including 3472 treatment-naive CLL patients. It discriminates 4 prognostic subgroups with different 5-year overall survival rates resulting in different treatment implications and combines the parameters TP53 status, IGHV mutational status, serum β2-microglobulin concentration, clinical stage, and age, with different weighting in a prognostic score. The CLL-IPI was initially introduced in 2016 and comprises data of patients who received chemoimmunotherapy.² 

A plethora of other prognostic scores with focus on different clinical outcome parameters like OS, progression-free survival, or treatment-free survial have been introduced to the field of CLL research, but a meta-analysis by the Cochrane group recently revealed that the CLL-IPI still shows the best discrimination, despite overestimation.³ 

However, the treatment landscape in CLL has dramatically changed in the last years with the introduction of novel oral inhibitors. Against the background of quickly changing treatment possibilities and recommendations, the CLL-IPI might be outdated now. Therefore, several further scores have been introduced to the new treatment landscape.

Soumerai et al⁴  developed the first validated risk score to predict OS in patients with relapsed/refractory CLL who were treated with targeted therapies. It identified 4 factors that differ widely from the CLL-IPI including serum β2-microglobulin, lactate dehydrogenase, hemoglobin, and time from initiation of last therapy and helps to identify patients who are at a high risk for early death.

The BALL score identifies a subset of patients with CLL, accounting for about 50% of the whole population, who benefit, in particular, from single agent ibrutinib. It comprises β2-microglobulin, hemoglobin, lactate dehydrogenase, and time elapsed from last therapy <24 months as parameters.⁵  However, this score did not show satisfying results in a real-world patient cohort, so the improved survival risk score for ibrutinib, which excludes time to last therapy, was then developed to determine OS in relapsed/refractory patients with CLL treated with ibrutinib.⁶ 

Furthermore, a 4-factor model consisting of TP53 status, prior treatment, β-2 microglobulin, and lactate dehydrogenase levels was introduced. It identifies patients with an increased risk of ibrutinib failure at treatment initiation and remained significant when applied to either treatment-naive or relapsed/refractory patient cohorts treated with ibrutinib.⁷ 

In a completely different approach paying tribute to the changing dynamics within the course of the disease, the continuous individualized risk index was evaluated in CLL including the CLL-IPI and minimal residual disease levels. This dynamic risk model seems to be superior to established risk assessment scores in determining clinical outcomes.⁸  However, this model is probably too complex to be used broadly in clinical practice but could be a very helpful tool in clinical trials otherwise. Prospectively, an easy-to-use tool will have to be developed for a more individualized management of patients that can be easily implemented into clinical routine and therefore finds broad acceptance. Hence, the CLL-IPI should be reevaluated as soon as more mature data on first-line treatments with oral inhibitors are available.