Final results of a phase 1 trial with soquelitinib (SQL), a selective interleukin-2-inducible T cell kinase (ITK) inhibitor for treatment of relapsed/refractory (R/R) T cell lymphomas (TCL)

Abstract Introduction ITK is expressed in T cells and NK cells and is involved in multiple signaling pathways including T cell receptor (TCR) activation and differentiation of T helper (Th) cells. In mice, gene knock-out of ITK leads to Th1 skewing of Th cells. SQL is an oral, covalent inhibitor of ITK that suppresses Th2 response with sparing of Th1 function and cytokine production (Hsu et al. npj Drug Discov, 2024). In vivo studies in murine tumor models have demonstrated anti-tumor activity in several tumor types including tumors that do not express ITK. These studies demonstrated increased infiltration of normal CD8+ T cells with enhanced cytolytic capacity based on elevated production of IFNγ and TNFα. Tonic TCR signaling has been implicated in TCL, and ITK inhibition with SQL was found to reduce GATA3 expression (master transcriptional regulator of Th2), which is ITK dependent (Geng et al. Blood Cancer J, 2022). GATA3+ subtype of peripheral TCL is associated with a poor prognosis; median overall survival (OS) <1 year (Amador et al. Blood, 2019). We hypothesized that SQL would be active in TCL due to direct effects on tumor cells as well as induction of anti-tumor immunity. We now report the results of a Phase 1 dose escalation trial in R/R TCL (NCT03952078). Methods Twenty-eight patients (pts) were enrolled in the dose escalation portion of the trial evaluating 100, 200, 400 and 600mg oral BID until tumor progression. ITK target occupancy was measured in blood T cells. An expanded cohort was enrolled at the 200mg BID dose. Safety and tumor responses were evaluated every 2-3 months. Lymphocyte subsets and gene expression analysis in blood and tumor were evaluated by flow cytometry and RNA sequencing. Tumor expression of GATA3 was determined by immunohistochemistry (IHC). Results A total of 75 R/R TCL pts were enrolled; 28 in dose escalation, 47 in expansion. Median age, 62; male 57%; median prior therapies 3 (range 1-19). Histologies included peripheral T cell lymphoma not otherwise specified (PTCL-NOS, n=32); angioimmunoblastic T cell lymphoma (AITL, n=12); cutaneous T cell lymphoma (CTCL, n=19) and others (n=12). Complete target occupancy was achieved at doses of ≥ 200mg BID and additional pts were enrolled in an expansion cohort. At the 200mg BID dose (n=37 total from expansion and dose escalation), objective response rate of 39% (6/23) was seen in pts who had received treatment with 1-3 prior cytotoxic regimens including 6 durable CRs (28, 25, 22+, 21, 17+ and 12 months) and 3 PRs (29+, 7 and 7 months) vs 0% in 14 pts with > 3 priors. 11 of 11 pts with available tissue tested prior to SQL therapy were GATA3+ by IHC (H score >100). These included 6 PTCL-NOS, 2 of which achieved CR. In pts treated with 1-3 prior cytotoxic regimens, the median progression free survival and duration of response was 6.2 months and 18.5 months, respectively. Similar OS was seen for PTCL and CTCL with median OS for all pts who received 200mg BID of 28.1 months with 24-month OS of 66%. SQL was well tolerated, with Grade 3+ AEs reported in 29% of pts. The most common (≥ 5%) Grade 3+ AEs included pruritus and anemia considered not drug related. Analysis of blood and tumor tissue demonstrated an increase in Th1 and T effector memory cells (TEMRA) in evaluable responding pts, and increases in granzymes and perforins in normal T cells infiltrating the tumors. Treatment was associated with significant reductions of T cell exhaustion markers, LAG3 and TIM3, in blood CD4 and CD8 cells. ConclusionSQL monotherapy is active and well tolerated in pts with R/R TCL, particularly those pts with limited exposure to prior immunosuppressive therapies. Durable CRs and PRs and relatively long OS were seen at 200mg BID, a dose that achieves full target occupancy. Selective blockade of ITK results in Th1 skewing, increases in TEMRA cells, increase T cell cytolytic capacity and reduction of T cell exhaustion. These findings indicate that the mechanism of action involves the induction of a host anti-tumor immune response that requires a baseline level of immunocompetence. These data suggest that effective therapeutic strategies for these lymphomas should extend beyond direct targeting of malignant T cells. SQL is now being evaluated vs standard therapy (belinostat or pralatrexate) in a randomized Phase 3 registration trial in PTCL pts with 1-3 prior therapies (NCT06561048).