Abstract 1171

T cell large granular lymphocyte leukemia (T-LGLL) is characterized by the chronic proliferation of cytotoxic T lymphocytes (CTL) and is often associated with lineage-restricted cytopenias, autoimmune disease and B cell dyscrasias. In many patients clinical features of a true leukemia are absent. Lack of transformation, absence of progressive increase in blood counts, and a paucity of recurrent chromosomal defects or mutations clearly distinguish LGL from typical leukemic development. Thus, some forms of T-LGLL resemble more of a reactive process and possibly represent an extreme pole of oligo- or polyclonal immune reactions to viruses, autoantigens, or perhaps even exaggerated immune tumor surveillance. We have shown that a highly skewed T cell receptor (TCR) variable beta (Vβ) repertoire is strongly associated with monoclonality of TCR CDR3 regions by sequencing and therefore, detection of a Vβ expansion by flow cytometric clonotyping can serve as a surrogate marker for the presence of a clonal CTL process. Flow cytometric Vβ typing offers an opportunity to study the dynamics of the CTL clonal progression during therapy and throughout the clinical course of disease.

We studied 124 patients who not only met the WHO guidelines for the diagnosis of T-LGLL but also show skewing of the Vβ spectrum by flow cytometry consistent with the mono/oligoclonal process persistent for over 6 months; 100% of cases demonstrated both TCRγ rearrangement and abnormal CTL population by flow cytometry. LGL count >900/μL was present in 73% of patients (mean 2513±3571 cells/μL). A pathologic clonal Vβ expansion was defined as > mean + 2SD of controls (n=65). Expansions identified by the Vβ panel were present in 92% (mean clone size 55±28%) and 2% had a borderline expansion (within 20% of 2SD) and 3% a δ/γ CD8+ TCR expansion. In 6% the Vβ expanded clone expressed CD4. Absolute clone count (ACC) was calculated by the Vβ contribution multiplied by the absolute CD8 (or CD4) count per μL of blood. ACC correlated well with LGL count (p<.0001, R2=.58). Clinically, patients presented with anemia (25%), neutropenia (9%), pancytopenia (19%), thrombocytopenia (3%), or multi-lineage cytopenia (29%), and 15% of patients were asymptomatic. In order to assess clonal kinetics, 62 of these patients were available for serial Vβ measurements with a follow up range of 0.5–8 years. Paired statistical analysis of clonal dynamics pre and post therapy (cyclophosphamide, alemtuzumab, methotrexate or cyclosporine) demonstrated a significant decrease in ACC between responders and refractory patients (p=.024). Unexpectedly, some patients displayed a change in the dominant clone as demonstrated by a switch in the major clonal Vβ T cell population, i.e. “clonotype switch.” Overall, 32% exhibited a clonotype switch during the study period, while others exhibited the persistence of multiple clones (22%); in 46% the initially diagnosed Vβ monoclonal expansion persisted. Those with multiple clones were more likely to change clonal dominance (p=.05). Clonotype switch was observed in both in relapsed and in refractory patients, and also was frequently accompanied by a change in clinical hematologic features.

Significant absolute lymphocytosis (>4000 lymphocytes/μL) was present in 34/124 (27%) patients while 66 had normal lymphocyte counts and 24 were lymphopenic. Of the patients followed serially who clonotype switched, only 3/20 (15%) had absolute lymphocytosis, suggesting that 2 distinct subtypes of T-LGLL may exist. Patients with high lymphocyte and LGL counts may represent a true leukemic process, are less likely to have associated autoimmune disease or second malignancy, and their dominant clone is stable, suggestive of a single transformed precursor. In contrast, in patients with clonal CTL expansions not associated with absolute lymphocytosis, clonotype switching is difficult to reconcile with a true autonomous leukemic process.

In sum, our results suggest that in a significant proportion of patients with T-LGL leukemia, the propensity for clonal CTL dominance is not inherent to an aberrant molecular event within the abnormal CTL clone but may rather be related to extrinsic chronic antigenic drive and immune dysregulation. Furthermore, our results lend credence to the body of evidence suggesting that T-LGL leukemia may not, in many instances, be a true leukemia, and may best be classified as T-LGL lymphoproliferation.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution