Chronic neutropenia (CrN) is defined as neutropenia lasting longer than 3 months and has various underlying etiologies, including congenital neutropenia (CN). We aimed to determine the underlying etiologies of patients with CrN and define the characteristics at diagnosis suggestive for CN as a final diagnosis. The study included 197 pediatric and adolescent patients who were diagnosed with CrN between 2010 and 2022 in a single-center. Patients with transient neutropenia, splenomegaly, alloimmune, secondary autoimmune, myelodysplastic syndrome and hematological malignancies, other bone marrow failure syndromes such as Fanconi anemia, and aplastic anemia due to CrN were excluded from the study. The patients included in the study were sub-classified into 5 groups according to their final diagnosis. Group 1, patients with CN (n=64, %32.5) included all genetically verified CrN patients. Group 2, patients with primary autoimmune neutropenia (pIN) group (n=10, %5.1) with anti-granulocyte antibody (AGA) positivity. Group 3, patients with unclassified chronic idiopathic neutropenia (UCIN) (n=73, %37.1). Group 3 included patients who had a clinical severity resembling immune neutropenia, but whose AGA testing was either negative or not available; but the CrN resolved during follow-up. Group 4, included the patients with unclassified chronic benign neutropenia (UCBN) (n=34, %17.3) and included the patients with ongoing neutropenia, who do not have a history of serious infections, no underlying genetic cause for neutropenia, no AGA positivity. Group 5, unspecified congenital neutropenia (UCN) group (n=16, %8,1), was defined as the group of patients whose genetic tests did not detect a mutation in genes related to CN, but history reveales severe infections and a possible CN as an underlying cause. Genetic testing of patients varies from single gene evaluation with Sanger sequencing analysis to Whole Exome Sequencing (WES). The median age at diagnosis of neutropenia of the patients included in the study was 12 (0-168) months 53.8% of the patients were male (n=106). There was consanguinity between parents in 34.5% and at least 1 family member was diagnosed with neutropenia in the in 18.8%. Among the patients in Group 1, HAX-1 mutation was detected in 19 (29.7%) patients and it was found to be the most common genetic defect causing CN. Other common genetic disorders include ELANE mutations (18.8%), Shwachman Diamond Syndrome (10.9%), Hermansky Pudlak Syndrome (7.8%) and glycogen storage type 1 b (6.3%). G6PC3 deficiency, VPS45 deficiency, Barth syndrome and Kabuki syndrome in 2 each (3.1%), ADA2 deficiency, Cohen Syndrome, Poikiloderma with neutropenia, Prolidase deficiency, CLPB deficiency, Majeed syndrome, GINS4 deficiency, reticular dysgenesis, germline RUNX1 mutation were detected in 1 patient. There are 3 patients with acute myeloid leukemia and all of them died before hematopoietic stem cell transplantation (HSCT). Two of these patients were diagnosed with SDS, and one had ELANE mutation. One patient with the HAX-1 mutation developed non-Hodgkin lymphoma and was treated with HSCT and is still alive. There were 7 patients (3.6%) who underwent HSCT. 9 patients (4.6%) died. Except for 3 patients who died from AML and 1 who died after HSCT; 1 patient with Barth syndrome deceased related to associated dilated cardiomyopathy, Another patient with G6PC3 mutation died after complications of cardiac surgery due to associated cardiac anomaly and 3 patients (2 VPS45, 1 CLPB) died due to infections. Patients in Group1 and Group 2+Group3 were compared, in order to answer the question of “Can we create a model that predicts CN?”. Many parameters were compared and a multivariate model was created with statistically significant risk factors. CN was seen 6,3 times more in those with parental consanguinity, 8,9 times in those with a family history of neutropenia, 4,8 times in those with more than 2 hospitalizations due to infection, and 4,1 times more in those with recurrent oral involvement (p<0.05). Patients diagnosed with CN were estimated based on the results of the multifactor logistic regression model. The rate of correct identification of the patients by the model is 88.9%. In conclusion, our study is one of the largest single-center studies and suggests a novel classification of CrN patients.
Read full abstract