LGL: a disease rediscovered

In this issue of Blood, Jerez et al describe mutations in STAT3 in 2 subtypes of large granular lymphocyte leukemia (LGL).¹ 

Large granular lymphocyte leukemia (LGL) is a rare lymphoproliferative disorder of T cells (mostly CD8⁺ T cells) or natural killer (NK) cells with an unclear etiology.²  Recently, mutations in STAT3, a key signaling protein often mutated in tumors, were reported in 31 of 77 (40%) of T-LGL patients,³  implicating this molecule in the expansion and persistence of these cells. Jerez et al now extend these findings and show that 33 of 120 (28%) of T-LGL and 15 of 50 (30%) of NK-LGL exhibited a mutated STAT3 Src-homology 2 (SH2) domain.¹ 

This study is remarkable for the large numbers of patients investigated with this rare disorder. Mutations in the same SH2 domain, critical for activation of this protein and hence downstream pathways, were identified in both T- and NK-LGL, suggesting a common pathogenic mechanism driving expansion and persistence of lymphocytes in both disease subtypes. One of the mutations now identified in LGL, which alters the tyrosine residue at position 640 into a phenylanaline (Y640F), was previously shown to represent a gain-of-function in hepatocellular carcinoma.⁴  Jerez et al show here, in a reporter assay, that another mutation, D661V, also increases the basal activity of STAT3, independently of upstream signals.³  The SH2 domain of other STAT family members plus mutational hotspots in genes in the JAK-STAT signaling pathways were sequenced in 40 patients with wild-type STAT3, but no mutations were uncovered. Although the clinical consequences of a mutation in the SH2 domain of STAT3 remain unclear, this study provides an impressive inventory of the mutational status of this gene in T- and NK-LGL, and ample material for future investigations.

Jerez et al focused on the SH2 domain in their search for mutations.¹  However, some of the gain-of-function STAT3 mutations found in hepatocellular carcinoma are outside this domain,⁴  suggesting that all domains of the STAT3 gene should also be closely inspected before we call them truly wild-type. Detailed sequence analysis could also solve the current paradox that neither laboratory findings nor the overall survival seem to distinguish between patients with wild-type and mutant STAT3 SH2 domains.

Now that mutations in the STAT3 gene in LGL are identified, the next step is to explore the functional consequence of these alterations in CD8⁺ T cells and NK cells. The transduction of STAT3 mutants identified in LGL, which are presumably gain-of-function mutants, and, as a control, from Job syndrome patients, which represent loss-of-function or hypomorphic mutants, into CD8 T cells/NK cells from healthy donors should educate us on the effect these variants have on gene expression, proliferation, and survival on cytokine withdrawal, and leukemic behavior in the form of extended survival in a xenogenic mouse model. Furthermore, such studies should uncover differences in the way T cells and NK cells handle STAT3 mutants.

Lastly, although T-LGL is characterized by the expansion of CD57⁺ CD8⁺ T cells,²  we observed that the LGL clone is represented in both CD57⁺ and in CD57⁻ fractions.^5,6  Indeed, CD57⁺ cells lacked proliferative capacity,⁷  but were readily generated from sorted CD57⁻ LGL cells.⁵  With the recent discovery of a stem cell memory T cell (T_SCM),⁸  a burning question is whether the T-LGL clone originates in early memory cells and carries the mutant STAT3 gene. Such a finding would refocus the need to target STAT3 inhibitors on the true LGL stem cell.