Clinical and Mechanistic Relationship of T-Cell Large Granular LymphocyticLeukemia (TLGL) andmonoclonal Gammopathy of Undetermined Significance(MGUS)

Introduction: The incidental finding of asymptomatic clonal expansion of plasma cells (MGUS), and T cells (TCUS) is not uncommon and its prevalence increases with age. While the propensity of such conditions to progress to multiple myeloma (MM) and TLGL, respectively is not fully elucidated, we observed in our clinical practice an intertwine between TLGL and MGUS, suggesting a possible mechanistic relationship. Per definition, MGUS should be asymptomatic and without cytopenia, the presence of which would upstage this condition. Thus, most often cytopenias in patients (pts) with MGUS and TLGL coexistence have been primarily attributed to the autoimmune sequelae of the T cell clonopathy. However, TLGL can also be asymptomatic in 20-50% of cases. In such a scenario the identification of the culprit etiology of cytopenia is crucial for the selection of proper therapies. Moreover, understanding the relationship between MGUS and TLGL may provide important lessons regarding the nature of the latter. One could stipulate that TLGL cells are the closest natural equivalent to the chimeric antigen receptor T cells (CART) and naturally arise from the immune trigger provided by clonal plasma cells. The initially polyclonal CTL responses may subsequently evolve to a clonal process, in particular upon acquisition of STAT3 mutations, ultimately controlling the MGUS clone. Alternatively, both humoral and CTL responses are triggered by the same offending agent and its persistence leads to the selection of dominant MGUS and TLGL clones as markers of an overshooting immune response.

Methods/Results: We accrued a cohort of 280 pts with TLGL and explored both deep clinical and genomic phenotypes on 171 informative pts for whom complete data was available. Overall, pts had a median age of 67 years (IQR 58-73) with almost 1.09 M:F ratio. Among them, 23% had a concomitant plasma cell dyscrasia, with the majority being MGUS (19%), whereas only 2% and 1% of pts presented with Waldenstrom's macroglobulinemia/lymphoplasmacytic lymphoma (WM-LPL) and MM, respectively. Of note, no MGUS was found in TLGL pts <50 years, suggesting other pathogenetic drivers in younger cases. Overall, TLGL/MGUS had IgG (63%), IgM (25%) and IgA (9%) monoclonal protein phenotypes. In addition, polyclonal hypergammaglobulinemia was present in another 12% of pts. Whereas the overall expected MM evolution is 1-2% per year in typical MGUS, at a median follow up time of 12.5 years (8-17) only 1 MGUS/TLGL case progressed to MM as opposed to the estimated 3-4 pts. Furthermore, while the median overall survival (OS) of pts with MGUS has been reported as 8 years in historical cohorts, in our TLGL/MGUS cases reached 19 years. Looking at clinical characteristics, 13.5% of pts were asymptomatic, 61% had anemia, 53% neutropenia, and 34.5% thrombocytopenia. Anemia (76% vs 58%, p=.04), and bilineage cytopenia (38% vs 22%, p=.04) were more common in pts with co-associated MGUS. Gender, age distribution and disease-specific features such as TLGL counts (median 1785 vs 1680/µL), presence of STAT3 mutations (26% vs 33%) and co-occurring rheumatoid arthritis (15% vs 14%) did not differ between the two groups. Overall response rate (ORR) to T cell-directed therapy achieved 87% in pts with TLGL vs 64% in pts with concomitant MGUS (p=.0017). Indeed, because of the higher refractoriness rate in the latter group (median line of therapies 4 vs 3 in non-MGUS cases), 6 of these immunosuppression-refractory pts received bortezomib-dexamethasone resulting in a 67% ORR with a median time-to-next-treatment of 6.5 months (3-11). These observations argue against TLGL being the culprit of cytopenia in all TLGL/MGUS, suggesting that the monoclonal protein may have a specificity against myeloid or erythroid precursors.

Conclusion: Our study shed light on the intertwining relation of the TLGL/MGUS clinical dyad. Rather than a coincidental finding in older pts, such clinical presentation may represent the first paradigmatic example of natural CART cells, analogous to those currently used in MM clinical practice. Finally, the etiology of cytopenias in TLGL need to be cautiously investigated, not automatically assuming a relation to TLGL. To further dissect such complicated dynamics, we are investigating a large cohort of MGUS pts for the presence of TCUS/TLGL and assessing the specificity of polyclonal and monoclonal proteins in MGUS pts with cytopenia using 2 million 12-mers peptide arrays.

Maciejewski:Alexion: Consultancy; Apellis Pharmaceuticals: Consultancy.