Polarized CTL Responses Detected in Patients with Autoimmune Neutropenia.

Drugs and intrinsic bone marrow diseases can explain most of the cases of neutropenia, and true autoimmune neutropenia (AIN) is a diagnosis of exclusion. Anti-neutrophil antibodies are not reliable, and their absence does not preclude the diagnosis of AIN. Lineage-restricted cytopenias, including neutropenia, were associated with T cell Large Granular Lymphocyte leukemia (T-LGL), but the diagnosis of this condition involves positive TCR rearrangement and flow cytometric identification of a pathologic cytotoxic T cell (CTL) population. These routinely applied methods have a limited sensitivity and rely on the presence of a high frequency of clonal cells in the sample. AIN, similar to T-LGL, may be related to a CTL-mediated process. We hypothesize that AIN, in a portion of patients, is a CTL-mediated disease in which myeloid progenitor cells are the targets. Consequently, in those patients, polarized expansions of CTL clones may be detected if efficient and sensitive diagnostic methods will be applied. Previously, we developed a diagnostic algorithm for the identification and quantification of clonal expansions in T-LGL based on the molecular analysis of TCR- utilization pattern. We studied a cohort of patients with various degrees of neutropenia (N=23) that was unexplained by clinical grounds and standard laboratory testing. Anti-neutrophil antibodies were found in 6 of these patients; in 3 patients, serum-mediated inhibition (>20%) of colony formation by normal hematopoietic progenitor was found, but there was no correlation between antibodies and serum inhibition. For detection of CTL expansions in AIN, VB typing and VB specific RT-PCR were applied followed by PCR cloning and sequencing of a large number of clones, and determination of expanded CDR3 clonotypes. When no expansion was detected by flow cytometry, multiplex PCR was used to amplify the whole VB spectrum. If identical CDR3 regions were detected by sequencing of at least 22 clones, CDR3 fragments of appropriate VB families were subcloned and sequenced, and immunodominant (identical clones occurring repetitively) were identified. Using this approach, we found only 2 expanded clones in 24 healthy donors. Those expanded clones accounted for 20% of a given VB family, or 0.7% of the CD8+ repertoire (as calculated by multiplication of clonal expansion within VB family by VB family contribution to the whole CTL population). In AIN we found expansion in 9 of 21 patients (3 of them were not detected by VB flow cytometry). Clonal frequency was 40%± 13% of a given VB family or 13%± 14% of the total CD8+ population. The presence of expanded CTL did not correlate with anti-neutrophil antibodies of serum-mediated colony inhibition. By comparison, CTL clones found in patients with T-LGL leukemia (N=75) comprised 68% of a given VB family, or 43% of the entire VB repertoire. We conclude that, using sensitive approaches, CTL expansions can be detected in a significant proportion of patients with AIN. These cases may represent minor variants of an autoimmune process that operates in T-LGL leukemia. The antigens that trigger these expansions likely may be shared. Clinically, detection of the CTL-mediated process in neutropenia may point toward rational immunosuppressive therapy aimed at T cells.