THU031 White Blood Cells (WBC) As Biological Markers Of Cortisol Status In Patients With Cushing's Disease (CD)

Abstract

Abstract Disclosure: B.M. Mascarenhas-Nakano: None. P.F. Santos-Neto: None. L.C. Portari: None. S.R. Correa-Silva: None. J. Abucham: None. CD is a rare disease caused by ACTH-secreting pituitary adenomas and responsible for ∼80% of cases of endogenous hypercortisolism. Hypercortisolism affects all tissues, leading to several nonspecific and few specific anatomical and functional changes which are not always easy to recognize and quantify. Clinical suspicion is the first step in the diagnosis but is rather impressionistic. The influence of endogenous cortisol on WBC, recognized a long time ago, has been widely forgotten/ignored in clinical practice. In this study, we aimed to assess the potential of WBC as biological markers of cortisol status in patients with CD. Patients and Methods: We retrospectively compared Leukocyte (Le), Neutrophil (Ne), Lymphocyte (Ly), Monocyte (Mo), Eosinophil (Eo), and Basophil (Ba) counts at diagnosis in patients with proven CD (CDdx, n=64), at remission after treatment (pituitary surgery, adrenalectomy, or radiotherapy) (CDrem=36), and in a control group of patients with hypercortisolism suspected and ruled out using ES guidelines (CT, n=63). In addition, we retro-prospectively evaluated WBC in a subgroup of 33 patients before and after treatment [25 in remission (RE) and 8 in non-remission (NRE)]. Comparisons >2 groups: ANOVA followed by multiple comparison tests (MCT). Before vs after comparisons: paired tests. Variations (Δ= after-before) between different groups: unpaired tests. Associations: Fisher´s exact test. ROC curves were generated cut off values chosen by Youden index. Significance was set at 0.05. Results: At Diagnosis. Le, Ne, Ly, and Mo were increased and Ly decreased: CDdx vs CT (0.0075<P<0.0001), CDdx vs CDre (0.0015<P<0.0001), CDre vs CT (0.34<P<0.99). Eo and Ba showed no differences in CDdx vs CT (0.20<P<0.67). Diagnostic ROC curves were significant (0.047<P<0.0001) for all cells and various indexes (ratios and differences). AUC (0.8705), Youden index (0.696)(cut off 3,047), sensitivity (87.3%), specificity (82.26%), PPV (0.83) and NPV (0.86) were all highest for the difference Ne-Ly with overlapping confidence intervals with other indexes. At Remission. In paired analysis, most cells showed significant changes after remission. However, comparisons of Δs between RE and NRE groups were only significant for ΔLy and ΔNe% [nearly significant (P=0.06) for ΔNe%-Ly% and ΔLy%]. Remission was associated with ΔLy>-3 (P=0.0049) and ΔNe%<-12.5% (P=0.0463). Combining both parameters (ΔLy>-3 and/or ΔNe%<-12.5%), a stronger association (P= 0.0001) was found. Evaluation of cortisol status after treatment using the combined criterion was able to identify 21/25 (84%) patients in remission and 8/8(100%) innon-remission. Conclusions: Our results show that WBC, mainly Ne and Ly, are reliable biological markers of cortisol status and potentially useful in different steps of the diagnosis of endogenous hypercortisolism as well as in the follow up of patients with CD after treatment. Presentation: Thursday, June 15, 2023