Time for a new prognostic score in CLL?

In this issue of Blood, Langerbeins et al¹ evaluate the prognostic value of the current chronic lymphocytic leukemia International Prognostic Index (CLL-IPI) using a pooled data set of patients with CLL received first-line treatment with targeted drugs (n = 991) or chemoimmunotherapy (n = 1256). This work is timely, as all prognostic scores need reevaluation as standard therapies change. The study was limited by a short median observation time (40.5 months), and the actual number of chemotherapy-free patients in the cohort receiving targeted drugs was 891 (100 patients in the targeted drug cohort had received bendamustine debulking). Nevertheless, the authors were able to conclude that CLL-IPI retained prognostic value for progression-free survival (PFS), but with diminished discriminatory impact for the prediction of overall survival (OS).

The “targeted drugs” treatment in the study by Langerbeins et al included combinations of anti-CD20 monoclonal antibodies with inhibitors of the B-cell receptor signaling-associated pathways (BCRi, mainly ibrutinib) with or without the BCL2 inhibitor (BCL2i) venetoclax, but only 26% of these patients had received venetoclax and ibrutinib in combination.

The CLL-IPI score was originally generated by selecting 5 prognostic factors independently able to predict OS from 27 genetic, biochemical, and clinical parameters in patients with CLL who were treated in clinical trials with chemoimmunotherapy regimens as first-line treatment between 1997 and 2010, at a time when targeted treatments were not available.² The study describing the CLL-IPI was published in 2016, and the 5 factors were TP53 status (no abnormalities vs TP53 mutation or deletion or both), IGHV mutational status (mutated vs unmutated), serum β2-microglobulin concentration (≤3.5 mg/L vs >3.5 mg/L), clinical stage (Binet A or Rai 0 vs Binet B-C or Rai I-IV), and age (≤65 years vs >65 years). These factors were assigned a value (deleted/mutated TP53 = 4, unmutated IGHV = 2, high β2-microglobulin = 2, advanced clinical stage = 1, advanced age = 1) by approximating the hazard ratio (HR) value for OS risk, and a score was generated to discriminate 4 prognostic groups (low risk = 0-1, intermediate = 2-3, high = 4-6, very high = 7-10).

Langerbeins et al documented a significant improvement of PFS and OS in patients receiving targeted drugs compared with chemoimmunotherapy. The improved PFS gained using targeted treatments was observed in all CLL-IPI categories. However, the study by Langerbeins et al highlighted that only β2-microglobulin concentration (HR = 1.7), IGHV status (HR = 2.6), and TP53 status (HR = 1.6) maintained prognostic value and that the relative weights of these 3 factors to PFS risk assessment were different from those calculated for OS in the 2016 study.² 

Langerbeins at al also demonstrated that the improvement gained using targeted drugs instead of chemoimmunotherapy on OS was particularly evident in the CLL-IPI high- and very-high-risk groups, but not in the low- and intermediate-risk groups. Although there was a pairwise significant difference between each risk group in the 2016 study,² the current study documented a difference only between the intermediate- and high-risk groups. Also, only Binet stage A vs C, β2-microglobulin, and TP53 status, but not Binet stage A vs B or B vs C and IGHV status, maintained their prognostic value for OS in the patients receiving targeted drugs with the current observation time. This suggested that although an extension of the observation time was needed before making firm conclusions on the role of the CLL-IPI score in predicting OS, new parameters would need to be identified and weighted to predict PFS and OS in CLL in the context of targeted treatments. Although there is always attention toward the role of genetic evolution,³ environmental factors explaining CLL resistance to novel treatments should also be sought.^4,5 

The better understanding of the main biological “drivers” of CLL, namely BCR and BCL2, and how to target them therapeutically has led to a significant improvement of the treatment strategies in CLL.^6,7 The current most appealing strategies include those combining BCRi with BCL2i,^8-10 which were given to only 26% of the patients in the trials considered by Langerbeins et al, or BCL2i with anti-CD20 antibodies. However, concepts are rapidly evolving, and the period of observation of patients having received BCRi or BCL2i or their combinations remains short in a condition like CLL, where survivals can be measured in decades. Also, the new clinical trials with BCRi plus BCL2i combinations or BCL2i plus anti-CD20 antibody combinations will tell us if these therapeutic approaches may find a different indication for the CLL with unmutated IGHV (U-CLL) or mutated IGHV (M-CLL). U-CLL and M-CLL carry distinctive cellular origin, biology, epigenetics and genetics, and clinical behavior.⁶ Therefore, prognostic algorithms may need to be generated separately in these 2 types of CLL. Only extended observation time and a better understanding of how to combine new targeted drugs will make it possible to identify the best prognostic factors of OS. In the interim, when combined with the outcome data in Langerbeins et al, the CLL-IPI remains a tool that assists prognostication for patients with CLL in the era of chemotherapy-free targeted regimens.

Conflict-of-interest disclosure: The author declares no competing financial interests.