The Association of Leukocyte Immunoglobulin-like Receptor B4 (LILRB-4) with Central Nervous System Involvement in Patients with Acute Myeloid Leukemia.

The prevalence of central nervous system (CNS) involvement in patients with acute myeloid leukemia (AML) is variable with a reported range of 0.6% to 46%. Unfortunately, the clinical consequences of untreated CNS involvement in AML which include complete paraplegia and bilateral total blindness, can be devastating. In addition, both the diagnostic evaluation and treatment of CNS involvement are associated with potential significant morbidity. Thus, there is a need to predict CNS involvement in a noninvasive manner in patients with AML. The leukocyte immunoglobulin-like receptors are a family of receptors that regulate the activity of cancer stem cells, cancer development and relapse. These receptors are expressed on many cell types and the leukocyte immunoglobulin-like receptor B4 (LILRB-4) is expressed on monocytic myeloid cells. To investigate whether measurement of LILRB-4 is predictive of CNS involvement in patients with AML. Fifty-six patients with AML and followed at the University of Texas Southwestern medical system had measurement of the LILRB-4 by flow cytometry. Demographic, laboratory, risk stratification, cytogenetics and clinical variables were acquired from electronic medical chart review. The study cohort of 56 patients with AML had a median age of 60 years (range: 0.83 - 92); 52% male; and racial/ethnic composition of African American 20%, Asians 4%, White Hispanic 13% and White Non-Hispanic 64%. Eleven (20%) patients were diagnosed with CNS involvement (CNS+). LILRB-4 was positive in 91% of patients with CNS+ compared to 38% without CNS (CNS-) (p<0.002). In univariate logistic analysis: age, White Non-Hispanic and positive LILRB-4 (LILRB-4+) were statistically predictive for CNS+ [OR, 95% CI, p value]: 0.95, 0.92-0.99, p<0.01; 4.31, 1.08-17.25, p<0.04; 16.46, 1.93-140.2, p<0.02, respectively. Traditional risk factors such as white blood cell count, FLT-internal tandem duplication mutation, National Comprehensive Cancer Network (NCCN) risk stratification scheme and serum lactate dehydrogenase levels were not predictive for CNS+. A positive LILRB-4 has a sensitivity of 91% and specificity of 62% for CNS+. A positive LILRB-4 by flow cytometry on leukemia cells is strongly associated with CNS involvement in patients with AML. These novel findings suggest that LILRB-4+ may be useful in risk stratification for CNS+. Moreover, combined with preclinical studies, LILRB-4 may provide insight into the pathogenesis of AML seeding the CNS. Further prospective studies are needed to confirm these intriguing results. [ABSTRACT FROM AUTHOR]