PS1340 ABSOLUTE MONOCYTE COUNT AFFECTS PROGNOSIS IN MYELODYSPLASTIC SYNDROMES INDEPENDENTLY FROM THE IPSS‐R

Background: Monocytopenia is a hallmark of some inherited bone marrow failure syndromes, e.g. in case of germline GATA2 mutations. Its prevalence and potential prognostic impact in Myelodysplastic Syndromes (MDS) is less well studied Aims: To define the prognostic role of the absolute monocyte count (AMC) in MDS Methods: The Düsseldorf MDS‐registry was screened for patients with information about the IPSS‐R and the AMC at diagnosis. Cases with a white blood cell count (WBC) ≥13 G/l or an AMC ≥1 G/l were excluded. The influence of the AMC on overall survival (OS) was determined by Kaplan‐Meier analysis and by uni‐ and multivariate Cox regression models. Results: 976 patients with a median follow‐up of 28.5 months (mo) were identified, with a median AMC of 0.189 G/l (SD 0.239; range 0–0.996). Within higher IPSS‐R categories, lower median AMCs were observed: very‐low 0.309 G/l (range 0–0.966), low 0.232 G/l (range 0–0.996), intermediate 0.183 G/l (range 0–0.98), high 0.156 G/l (range 0–0.990) and very‐high 0.126 G/l (range 0–0.992) together with a highly significant difference between lower‐risk (IPSS‐R very‐low and low) compared to higher risk (IPSS‐R intermediate or [very] high) MDS (median AMC 0.241 vs. 0.155, p  < 0.001). Survival analyses were restricted to patients not having received allogeneic stem cell transplantation or induction chemotherapy (n = 771; IPSS‐R‐risk categories: very low n = 50 [median OS 102 mo], low n = 269 [median OS 67 mo], intermediate n = 201 [median OS 42 mo], high n = 139 [median OS 22 mo] and very high n = 112 [median OS 8 mo]). An AMC below the median was associated with a lower OS (HR for death 1.29, 95% CI 1.08–1.54, p = 0.004 and median OS 36 vs. 51 mo, p = 0.004) in univariate analysis, but without additional prognostic impact (HR 0.92, 95% CI 0.77–1.11, p = 0.397) after inclusion into a multivariate Cox regression model together with the cytogenetic risk categories according to IPSS‐R, bone marrow blast count (< or ≥5%) and presence of cytopenias (haemoglobin < or ≥10 g/dl, absolute neutrophil count < or ≥1.8 x 10 9 /l and platelets < or ≥100 x 10 9 /l). An AMC within the third quartile (0.190–0.379 G/l) was associated with a significantly longer OS compared to the first (0–0.076 G/l), second (0.077–0.189 G/l) and fourth (0.379–0.996 G/l) quartile (Q) [median OS 63 mo (Q3) versus 31 mo (Q1), 42 mo (Q2) and 41 mo (Q4), p = 0.002] and retained its independent prognostic value after inclusion into multivariate Cox regression models either together with the cytogenetic risk categories according to IPSS‐R, bone marrow blast count and presence of cytopenias as mentioned above (p = 0.015; HR 0.77, 95% CI 0.62–0.95) as well as together with the IPSS‐R categories as covariates (p = 0.001, HR 0.69, 95% CI 0.56–0.85). After stratification according to the IPSS‐R, the additional prognostic value of an AMC within Q3 was restricted to the IPSS‐R intermediate group (median OS 75 vs. 38 mo, p = 0.005). The WBC and the relative monocyte proportion were significantly higher in Q4 compared to Q1/2/3 (median 5.1 G/l vs. 3.0 G/l and 12% vs. 4%, respectively, p  < 0.001). Summary/Conclusion: Monocytopenia is an additional risk factor in MDS, but an AMC in the upper range of normal (0.38–0.99 G/l, Q4 of the cohort) is associated with an inferior survival as well. The biological basis of this unexpected phenomenon remains to be elucidated, but the significantly higher monocyte percentage suggests an accumulation of borderline cases between MDS and chronic myelomonocytic leukaemia in this subgroup. Further studies are needed to define the AMC best suitable for prognostication in MDS.