DP129 | REFINING PROGNOSIS IN ESSENTIAL THROMBOCYTHEMIA: VALIDATION OF THE AAA+A SCORE AND DISCOVERY OF A NOVEL MARKER, THE LYMPHOCYTE-TO-MONOCYTE RATIO

Background: Recently, the AAA+A score—which includes monocytosis into the AAA score—was developed for patients (pts) with essential thrombocythemia (ET), and has proven effective in predicting overall survival (OS). However, external validation is still lacking. New data highlights the prognostic value of absolute lymphocyte (ALC) and monocyte counts (AMC), with the lymphocyte-to-monocyte ratio (LMR) emerging as a potential OS marker in hematologic disease. Aims: The primary aim of this trial is to assess the prognostic value of the AAA+A score for OS in a cohort of ET pts. The secondary aim is to assess the impact of LMR and other clinical-laboratory data on OS. Methods: Between 1995 -2025, 472 pts diagnosed according to current WHO criteria at Fondazione IRCCS S. Gerardo in Monza were enrolled in the INFLA-ME (INFLAmmation in myeloproliferative disease) trial. The AAA+A score was calculated as described by Tefferi et al. (Blood, 2024), and LMR as the ratio of ALC to AMC value at diagnosis. OS was defined from diagnosis to death/last follow-up (fu). Kaplan-Meier and Cox regression analyses were used to evaluate OS. Results: Pts median age was 64.5 (18-92) years (yrs), with a F/M ratio of 1.61. Most pts were JAK2-mutated (69%),79% had intermediate-high IPSET score. Median values for ALP and AMC were, respectively, 2 (0.1-8) and 0.59 (0.02-2.2) x10^9/l. After 9.4 yrs of median fu, 71 pts (15%) died. According to the AAA+A score, 105 pts (22.1%) were classified as low risk, 171 (36%) as intermediate-1 (INT-1), 166 (35.2%) as intermediate-2 (INT-2), and 30 (6.4%) as high risk. After 30 yrs of fu, median OS was not reached in low and INT-1 group, while it was 13.5 yrs in the INT-2 and 11 yrs in the high risk group (p<0.001). To simplify the AAA+A score and reduce age-related bias, we assessed the prognostic value of LMR, which includes 2 of the 4 AAA+A variables excluding age. ROC analysis identified 2.835 as the optimal cut-off. Pts with LMR <2.835 had significantly lower OS (p < 0.001). In univariate analysis, high-risk IPSET score (p<0.001) and age >60 yrs (p<0.001) were also associated with lower OS. In multivariate analysis, only age >60 yrs (p<0.001) and LMR <2.835 (p<0.001) retained independent prognostic significance. Conclusion: This study confirms the prognostic value of the AAA+A score in a large ET cohort and highlights LMR—an age-independent simplification of AAA+A—as a predictor of OS. Prospective studies are needed to validate our data.