1. Introduction to Large Granular Lymphocytic Leukemia

Large Granular Lymphocytic Leukemia (LGLL) represents about 2–5% of chronic lymphoproliferative disorders and predominantly affects the elderly (median age at diagnosis 66 years) ^[1]. Although the reported incidence is around 0.2 cases/1,000,000, the true prevalence may be much higher due to a large proportion of indolent, undiagnosed cases and, often, an inability to distinguish it from reactive cytotoxic T-cells (CTL) lymphoproliferation. The majority of LGLL cases are of the T subtype (i.e., derived from CTL), whereas only <10% are of natural killer (NK) cells origin ^[2].

The diagnosis of LGLL is based on evidence of chronic clonal T- or NK-cell proliferation via cytology, immunophenotypic analysis, and TCR repertoires assessment. In cytology, circulating LGLL cells should be ≥0.5 × 10 9/L in the peripheral blood smear ^[3]. Cell phenotype by flow cytometry helps in identifying T-cells and NK-cells with an aberrant phenotype. In particular, T-LGLL shows CD3 + , CD8 + , and CD57 + expression whereas NK-LGLL expresses CD8 + , CD16 + , and CD56 + ^[2]. The evidence of T-LGLL clonal expansion may be assessed by using TCR γ -polymerase chain reaction analysis and Vβ gene rearrangement testing, which help to distinguish between reactive and leukemic LGLL cells. In rare situations, a bone marrow biopsy with immunohistochemistry studies may be helpful, especially when LGLL counts are too low or in the case of the NK-subtype because, in this case, it is not possible to study TCR-repertoires’ clonality ^[3][4].

Unlike other lymphomas, LGLL expansion is usually self-limited, persistent, and not associated with lymphadenopathy or “B symptoms”, although splenomegaly is present in about 24–50% of cases ^[5]. Cytopenias, a major presentation of LGLL, may result from direct cell-mediated targeting of myeloid precursors (neutropenia) or erythroid precursors (reticulocytopenic anemia), or indirectly through cytokine-mediated destruction of hematopoietic stem cells, representing an indication for treatment ^[6]. In addition to cytopenia(s), LGLL is frequently associated with autoimmune disorders or other accompanying conditions ( Figure 1 ) ^[7]. The pathogenetic link to these frequently co-occurring conditions remains speculative.

Figure 1. Diseases commonly associated with LGLL and schematic representation of underlying pathogenesis. MGUS: Monoclonal gammopathy of undetermined significance; PRCA: pure red cell aplasia; AIHA: autoimmune hemolytic anemia; MDS: Myelodysplastic syndrome; HSC: hematopoietic stem cell. AICD: activation-induced cell death; STAT: Signal Transducer and Activator of Transcription 3; TCR: T-cell receptor; CTL: cytotoxic T-cells; ITP: Immune thrombocytopenic purpura; MM: multiple myeloma; RCC: Renal cell carcinoma; RF: rheumatoid factor; ANA: antinuclear antibodies. Figures were generated with BioRender.com (accessed on 20 August 2021).

Clinically, LGLL patients may remain asymptomatic for many years due to the aforementioned indolent course of the disease ^[1][8]. However, severe neutropenia, moderate symptomatic neutropenia, transfusion-dependent anemia, or autoimmune hemolytic anemia may necessitate therapy.

2. Overview of the Main Pathways Involved in the Pathogenesis of LGLL with Possible Therapeutic Implications

The pathogenesis of LGLL suggests potential new therapeutic approaches, many of which can be adopted from the therapeutic armamentarium designed for autoimmune diseases. Indeed, multiple pathways also involved in immune responses are found to be dysregulated in LGLL. Two major categories can be identified: (i) pathways promoting survival and (ii) pathways contributing to cell ability to escape apoptosis ^[2][9].

The Nuclear Factor kappa β (NF-κβ) signaling pathway is another transcription factor that has been implicated in the constitutive activation of LGLL cells. Once NF-κβ protein is translocated to the nucleus, it activates several proto-oncogenes such as cyclin D1 , c-myc , and the anti-apoptotic genes BCL-2 and MCL-1 , promoting the production of IL-2 ^[10][11] and allowing LGLL cells to escape the physiological mechanism of activation-induced cell death (AICD) ^[10][12] ( Figure 1 ).

Platelet-Derived Growth Factor (PDGF) receptor heterodimeric complexes (aa, bb, and ab) can be found on CTLs. An increased expression of both PDGF receptor type b and PDGF-bb results in an autocrine loop and activation of several pathways involved in LGLL (e.g., JAK-STAT, Ras-RAF-MEK1-ERK, and PI3K-Akt) ^[10][13]. Consistent with these findings, antibodies targeting PDGF-bb were shown to be cytotoxic to LGLL cells ^[14]. Whether activated by PDGF pathways or other mechanisms (see also below), RAS-RAF-MEK1-ERK signaling is hyperactive in LGLL (particularly in the NK subtype), and its inhibition showed encouraging results in this setting via restoration of Fas sensitivity ^[10][15]. It is worth mentioning that a significant shift towards pro-survival sphingolipids such as sphingosine-1-phosphate (S1P) has been identified in LGLL along with downregulation of the corresponding proapoptotic ceramide and sphingosine ^[16]. In particular, the upregulation of rate-limiting enzymes in LGLL such as sphingosine kinase 1 (SphK1), converting sphingosine to S1P, and the upregulation of N-acylsphingosine amidohydrolase 1 (ASAH1), converting ceramide to sphingosine, may represent one of the mechanisms responsible for enhanced LGLL cell survival ^[10].

IL-15 is a member of the IL-2 family that controls the activation and proliferation of T- and NK-cells. This cytokine is produced by antigen-presenting cells (APCs) and exerts its effects through interaction with the soluble or membrane receptor IL-15Rα. This receptor presents IL-15 to IL-2/IL-15Rβ subunits, which are expressed in both T- and NK-LGL cells. In particular, IL-15 forms a complex with IL-2Rβ, finally resulting in activation of JAK-STAT and Ras/MAPK pathways, in addition to an unbalanced generation of antiapoptotic (Bcl-2 and Bcl-xL) and pro-apoptotic signals (Bim and Puma) ^[17]. Furthermore, it has been shown that LGLL cells overexpress CD122 (a receptor subunit shared by IL-2 and IL-15). Increased soluble IL-15Rα concentration has been found in patients’ sera, pointing towards an essential role of the IL-15 signaling pathway in LGLL pathogenesis and, possibly, treatment ^{[13][17][18][19]}.

3. Current Therapy Approaches and Their Results

Due to the lack of randomized prospective trials, current standard treatment options in LGLL mostly rely on the metanalysis of phase II trials and case series. The mainstay of first-line therapies involves immunosuppression administered in a chronic and protracted fashion, rather than in pulse/cycle mode as is typical in many B cell lymphomas. This route of administration (chronic vs. pulses) is preferred because the proliferative fraction of memory cells for LGLL clone is low. The greatest amount of experience has been reported with methotrexate (MTX, 10 mg/m 2/weekly) , cyclosporine A (CsA, 3–5 mg/kg/day), or low dose cyclophosphamide (Cy, 50 to 100 mg/day), with or without short prednisone taper ( Table 1 ). Treating patients for at least 4–6 months is recommended before assessing response ^[1][2].

Table 1. Response to different therapies reported in the largest study cohorts of LGLL patients. The data presented are derived from a meta-analysis of the existing literature.

* 83 of 99 patients received CsA plus steroids.

In addition to its effects on DNA replication, MTX suppresses the activation of JAK/STAT signaling ^[23]. Indeed, STAT3 Y640F mutant cases appear to be more likely to respond to MTX treatment ^[21], which is also the preferred choice when treating LGLL patients with associated rheumatoid arthritis (RA), neutropenia, or other autoimmune conditions ^[8]. CsA, as a calcineurin inhibitor, blocks NF-AT, IL-2, and IFN-γ expression ^[12]. Interestingly, the HLA-DR4 allele has been found to be overrepresented in LGLL patients with co-occurring RA. Moreover, it has been shown that HLA-DR4 carriers may have a higher likelihood of CSA responsiveness, suggesting the presence of underlying antigen-driven mechanisms in different contexts characterized by specific immunogenetic predisposition ^[24]. Cy appears to be a good option, especially for cases with concurrent PRCA or profound anemia ^[8]. Chronic daily oral administration seems to be the preferred approach rather than periodic intravenous (IV) boluses. However, it is worth noting that a lack of response to either MTX, CsA, or Cy is not uncommon, and a switch between the three should be considered in such cases ^[2]. In addition, Cy should not be used for more than 12 months to avoid long-term complications, as demonstrated by studies involving large cohorts of patients with rheumatologic conditions ^[2]. Refractory cases are not uncommon and usually present with persistent transfusion-dependent anemia or severe neutropenia, requiring >1 line of treatment in up to 30% of cases ^[20][22]. Response to therapy ( Table 1 ) varies between cohorts, and patients often require trials, switching among different treatments.

4. Potentially Available New Therapeutic Options and Clinical Evidence

The pathogenic overlap between autoimmune T-cell-mediated reactions and the reactive-to-semiautonomous nature of LGLL implies that many shared targets may exist in these conditions. Thus, refractory LGLL cases may be analyzed for the presence of clinical clues for rationally applied salvage therapies.

One hypothesis behind the persistent activation and proliferation of LGLL cells is the chronic exposure to “un-cleared” antigens ^[2]. For this purpose, medications inhibiting the cycle of T-cell activation, particularly in patients with associated autoimmune processes, provide a valid avenue for targeted treatment. Abatacept is a hybrid protein consisting of the extracellular domain of CTLA-4 linked to the Fc region of human IgG1 (CTLA4-Ig). Upon binding to CD80/CD86 on APC, abatacept blocks CD28 co-stimulation of naïve T-cells, attenuating their activation. Effective in RA ^[25] , we previously reported the successful treatment of patients with refractory LGLL in association with RA and neutropenia with a response seen in 4/8 patients ^[22][26].

IL-6 is a key cytokine activator of the JAK/STAT pathway ^[27]. In a homeostatic state, IL-6 binds to its membrane-bound receptor IL-6R or the soluble form sIL-6R, forming a complex with the corresponding transducer protein (gp130) ^[28]. This complex activates the JAK-STAT3 pathway and triggers the expression of the suppressor of the cytokine-signaling 3 ( SOCS3 ) gene responsible for this physiologic negative-feedback found to be disrupted in LGLL ^[27][29]. Therefore, targeting IL-6 may represent a valid approach in refractory LGLL. Tocilizumab (anti-IL6 receptor) and Siltuximab (anti-IL6) are two approved drugs for the treatment of various autoimmune conditions (e.g., RA and Castleman disease) ^[28][30]. Considering the association of LGLL and RA ^[31], IL-6 antagonists may represent another therapeutic alternative, especially for LGLL patients with RA-like features. The associated risk of transient neutropenia with tocilizumab ^[31] warrants caution and suggests preferential use in refractory LGLL patients with transfusion-dependent anemia.

Bortezomib is a proteasome inhibitor that downregulates the NF-kβ pathway activity, ultimately blocking the degradation of different pro-apoptotic factors ^[32][33]. Given its efficacy in patients with multiple myeloma, this drug may be exploited in LGLL cases refractory to T-cell-directed immunosuppression, especially when associated with monoclonal gammopathies ( Table 1 ). The successful treatment with Bortezomib of patients with PRCA/LGLL with associated MGUS was reported ^[34][35].