Even with the clinical staging of Rai and Binet for determining prognosis, 50% of patients with early-stage CLL are at risk of early disease progression and premature death.1 The major dilemma regarding the management of CLL is the decision of when to begin treatment for the patients. While the currently available prognostic tools only predict the overall survival (OS) of early-stage CLL patients at high risk of rapid disease progression, the recent validated international prognostic score (IPS) for asymptomatic early-stage CLL could offer valuable insights of treatment initiation to the patients who are already tired of the “watch-and-wait” approach.
Unlike most of the existing prognostic tools aiming only at predicting the OS of early-stage CLL patients in need of therapy, the newly validated IPS-E (early, asymptomatic disease) tool can predict the probability of time-to-first treatment (TTFT), improve the overall treatment planning and the design of clinical trials.2 “The IPS-E is a simple and robust prognostic tool based on routine clinical and laboratory variables. The simplicity of IPS-E should facilitate its translation to the clinic,” highlighted the lead investigator of the study Dr. Davide Rossi, MD, Institute of Oncology Research in Bellinzona, Switzerland.
Like most cancers, CLL is multifactorial, and the disease at an early stage is heterogeneous. While some patients with early-stage CLL display no symptoms and may not require treatment for years, an estimated 30-50% of them are at risk of progressive disease within a short time.3 Besides, early-stage CLL patients can also become symptomatic without a change in the stage of their cancer. The primary dilemma of when to initiate CLL treatment is mainly because of the lack of patient stratification. At present, approximately 70% to 80% of early-stage CLL patients follow the international guideline recommendations of the “watch-and-wait” strategy.3 Thus, the IPS-E tool serves as a point-of-care testing resource by updating patients on the expected treatment timeframe.
The validation of IPS-E score involves a forecast of the likelihood of TTFT based on a retrospective observational study. The clinical design of the study consists of 4,933 patients with early-stage CLL from 11 international cohorts.2 Using univariate and multivariate analyses on the training group of 333 patients, identified 3 covariates to be consistently and independently correlated with TTFT. The 3 covariates included unmutated IGHV genes, absolute lymphocyte count of >15×109/l, and presence of palpable lymph nodes. The information led to the development of a prognostic score, with each factor counting as one point. The IPS-E tool was therefore invented based on the sum of these 3 covariates. Subsequently, it facilitates the stratification of early-stage CLL patients into subgroups comprising of low-risk (score 0), intermediate-risk (score 1), and high-risk patients (score 2-3). These scores thus demonstrated a definite TTFT.2
Validating the accuracy of the scoring system consists of using data from 10 cohorts, inclusive of 9 cohorts staged by the Binet system and 1 cohort staged by the Rai system. The findings established 30% of patients as low risk, 35% as intermediate risk, and 35% as high risk.2,4 The IPS score correlated closely with actual results from a five-year metanalysis of early-stage CLL patients in both the training and validation cohorts. The actual outcomes demonstrated a cumulative risk of 8.4%, 28.4% and 61.2% among low-risk, intermediate-risk and high-risk CLL patients requiring treatment within five years.2 Furthermore, the C-index was 0.74 and 0.70 in the training series and the aggregate of validation series.4
Dr. Rossi summarized the IPS-E model by commenting, “The IPS-E can be regarded as a building block to which newly discovered independent outcome predictors for patients with early-stage CLL could be added. A prospective study would help further assess and eventually strengthen this prognostic tool.” The investigators highlighted that the risk of bias related to the timing of evaluations and premature censoring requires further attention. Likewise, the IPS-E could be refined and improved by bringing in more patient data.2
The IPS-E model may potentially aid in the management of CLL. Furthermore, with the growing emergence of efficacious small molecule anti-CLL drugs, the IPS-E tool could prove critical in supporting the design of clinical trials for early-stage CLL patients with the highest risk of disease progression.