Which is more accurate for predicting OS and TTFT, CLL-IPI or MDACC-PI?

Massimo Gentile from the Azienda Ospedaliera of Cosenza, Consenza, Italy, and colleagues from across Italy and the US recently published data in a letter to the editor on the topic of comparing two prognostic tools for CLL in a community based cohort in Blood, October, 2016. The authors set out to compare and assess the validity of the recently published CLL-IPI and the MD Anderson Cancer Center Prognostic Index (MDACC-PI) scores for predicting OS and TTFT in newly diagnosed CLL patients. The data was collected from 858 patients across 5 Italian centers and, in a separate data set, 506 patients at the Mayo Clinic in the US.

Highlights:

• 5-year OS of CLL-IPI groups in this study, similar to original publication
• OS C-statistic: CLL-IPI = 0.71 (P<0.0001) vs. MDACC-PI = 0.68 (P<0.001)
• TTFT C-statistic: CLL-IPI = 0.72 (P<0.001) vs. MDACC-PI = 0.63 (P<0.001)
• Mayo Clinic OS C-statistic: CLL-IPI 0.75 (P<0.001) vs. MDACC-PI = 0.66 (P<0.001)
• Mayo Clinic TTFT C-statistic: CLL-IPI 0.74 (P<0.001) vs. MDACC-PI = 0.66 (P<0.001)

The CLL-IPI is a more accurate prognostic predictor than MDACC-PI in terms of both OS and TTFT, in two separate populations, confirming the use of CLL-IPI. However, as MDACC-PI can be used without FISH (TP53) or IGHV, unlike CLL-IPI, it may still be useful in some settings. When taken together, this work validates the CLL-IPI for use in predicting both OS and TTFT in patients with newly diagnosed CLL.