Combined use of tofacitinib (pan-JAK inhibitor) and ruxolitinib (a JAK1/2 inhibitor) for refractory T-cell prolymphocytic leukemia (T-PLL) with a JAK3 mutation

Case presentation

A 62-year-old male presented with lymphocytosis and splenomegaly. A diagnosis of T-PLL was made based on bone marrow examination. The patient was initially observed but after a year developed progressive disease with worsening lymphadenopathy and splenomegaly. Treatment with alemtuzumab was initiated. Figure 1 summarizes the timeline of treatments starting from initial diagnosis. After 3 months of treatment, the patient achieved a complete response and was consolidated with a matched unrelated donor stem cell transplant using BEAM conditioning. Graft versus host disease (GVHD) prophylaxis was given with alemtuzumab and tacrolimus. He relapsed after 16 months. Comorbidities at the time of relapse included chronic skin GVHD responsive to extracorporeal photophoresis, poor graft function with anemia, CMV reactivation, and chronic renal insufficiency.

Figure 1.

Treatment timeline. PD: progressive disease; CR: complete response.

At the time of relapse, the bone marrow was mildly hypercellular with an interstitial and focally nodular proliferation of monotonous, small lymphocytes with scant cytoplasm, convoluted nuclei, and condensed chromatin. Peripheral blood smear showed 35% atypical lymphocytes with similar morphology (Figure 2(A,B)). Flow cytometry demonstrated these lymphocytes expressed CD3, CD2, CD5, CD7, and were double positive for CD4 and CD8. They also expressed TCRa/b and had relatively dim CD45 expression. Conventional cytogenetic analysis showed a normal male karyotype, but FISH analysis detected deletions of MYB, 11q, and 13q. FISH studies performed on subsequent samples identified TCL1 rearrangement and isochromosome 8q, resulting in additional copies of MYC. Molecular testing for the T cell receptor gamma chain identified a monoclonal rearrangement, consistent with the presence of relapsed T-PLL.

Figure 2.

(A) Peripheral blood smear showing increased small and intermediate size lymphocytes with a monotonous appearance. Some lymphocytes have cytoplasmic blebs. (B) Mildly hypercellular bone marrow with an interstitial proliferation of lymphoid cells and reduced trilineage hematopoiesis. (C,D) Skin biopsy showing diffuse dermal infiltrate of small, monotonous cells, sparing the epidermis.

The patient received salvage therapy on a clinical trial with romidepsin and lenalidomide [1], He had a partial response but eventually developed grade 4 thrombocytopenia that required treatment interruption which led to subsequent disease progression. He was retreated with a single dose of alemtuzumab; however, due to complications of myelosuppression and CMV reactivation, his treatment was interrupted again followed by disease progression. Table 1 summarizes the responses, duration of treatment and side effects after the patient relapsed following transplant.

Table 1.

Description of therapies used after patient relapsed after transplant including absolute atypical lymphocyte count (ALC) upon initiation of treatment, median ALC during treatment, best response and main side effects (SE).

Treatment	ALC at the start of treatment (109/L)	Median ALC during treatment (109/L)	Response during treatment by ALC	Time on treatment (Months	Time to next therapy (months)	Reson to stop	Side effects
Romidepsin/lenalidomide	20	N/A	PR	~3	5	SE	Grade 4 thrombocytopenia
Alemtuzumab	24	0.1	PR	1	8	SE	Rash, fever, sepsis, CMV reactivation, grade 4 pancytopenia
Tocifitnib	85	112	SD	1	1	N/A	Grade 1 edema
Tocifitnib/Jakafi	150	98	SD	10	10	PD	Grade 1 edema, grade 3 anemia, grade 3 thrombocytopenia
Alemtuzumab	235	128	PR	0.25	2	SE	Grade 3 AKI, grade 4 neutropenia and thrombocytopenia
PEP-C	24	18	SD	2.5	2.5	PD	Grade 3 neutropenia, grade 4 thrombocytopenia, grade 3 anemia

PD: progressive disease; AKI: acute kidney injury.

At the time of progression, his absolute lymphocyte count had increased to 85,000. He developed new erythematous skin lesions and had daily fevers and night sweats. Biopsy of a left flank erythematous skin lesion showed a dense, focally periadnexal and angiocentric lymphocytic infiltrate in the dermis without involvement of the epidermis. The lymphocytes were very monotonous with nuclear contour irregularity and prominent nucleoli. Frequent mitotic figures were presented (Figure 2(C,D)). These lymphoid cells expressed CD2, CD3, CD4, CD5, CD7, and CD8 as well as TCL-1 and beta-F1, an immunophenotype similar to that seen in the bone marrow. Targeted next-generation sequencing performed on a peripheral blood sample identified two missense JAK3 mutations (p.M511I, variant allele frequency (VAF): 50%; p.L875H, VAF: 10%). In addition, a missense mutation of uncertain significance in ATM was also present (p.I1740N, VAF: 96%). The JAK3 p.M511I mutation had been previously reported in T-PLL and was felt to be pathogenic [2].

In vitro experiments on primary samples were established to assess the response to tumor cells to various novel therapeutic options (Figure 3). Patientderived leukemia cells were cultured in the presence of control peptides, tofacitinib (a pan-JAK inhibitor that is FDA-approved for the treatment of rheumatoid arthritis), rituximab, and doxorubicin, at varying concentrations. After 24 h, the cells were lysed and viability was measured with the ATPlite assay (PerkinElmer Life Sciences, Waltham, MA) and the Polarstar Optima microplate reader (BMG Technologies, Durham NC). Patient cells were somewhat sensitive to inhibition with tofacitinib while no response was seen with doxorubicin or rituximab, as expected. Additional experiments tested the combination of tofacitinib and ruxolitinib, a JAK1/2 inhibitor. These tests demonstrated the synergistic effect of the dual combination against leukemic cells.

Figure 3.

Level of ATP in murine cells treated for 24 h with control peptides, tofacitinib (tof), ruxolitinib (rux), doxorubicin or a combination of tofacitinib and ruxolitinib. ATP levels were assayed using ApoSENSOR ADP/ATP kit. Following 24 h incubation under the various conditions, cells were lysed and the freed ATP was detected by luciferase and luciferin in the media, producing light proportional to the level of ATP which was measured using Polarstar Optima microplate reader. There is synergistic depression of cell metabolism and growth for murine cells treated with both tofacitinib and ruxolitinib.

Audience members at New York Lymphoma Rounds response suggestions: (1) repeat a second allogeneic transplant. (2) repeat alemtuzumab. (3) purine analog.

Management and outcomes

Considering the limited available options and based on the in vitro results, treatment with tofacitinib was offered. The patient was initially treated with tofacitinib (starting at 5 mg PO BID, increasing to 10mg PO BID after 15 days of treatment), with adequate tolerance and a short clinical response. After one month of treatment, ruxolitinib was added. Within the first 2 months of starting the dual treatment, there was an improvement of fatigue, resolution of low-grade fevers and night sweats, and there was regression of leukemia cutis lesions. His lymphocyte count stabilized.

Main toxicities were grade 1 edema (just after starting tofacitinib), development of anemia (from ~11 g/ dL to nadir of ~7 g/dL) and thrombocytopenia (~70 × 109/L to ~20 × 109/L). Absolute neutrophil count (ANC) remained stable. There were no significant liver function test abnormalities. The patient did not require admissions and no CMV reactivation occurred. After 10 months on combination therapy, the patient developed rapidly increasing lymphocytosis, fevers, and night sweats. Tofacitinib and ruxolitinib were discontinued and treatment with alemtuzumab was restarted. After an initial response to treatment, the patient developed renal failure requiring prompt discontinuation of therapy. He died within 3 months due to the progression of disease.

Discussion and conclusions

T-cell prolymphocytic leukemia (T-PLL) is a rare, aggressive, mature T-cell hematologic malignancy, with poor response to treatment and poor prognosis [3,4]. Historically, survival rates were less than a year with a relapse rate of nearly 100% [5]. Alemtuzumab, a humanized monoclonal IgG1 antibody against CD52, has significant activity in T-PLL, but responses are transient and disease progression is inevitable [5–9]. Allogeneic stem cell transplant is the preferred modality to consolidate therapy and improve progression-free survival [10]. With optimal therapy, 5-year survival from diagnosis approaches 20% and some allogeneic transplant recipients achieve durable remissions [5,11,12]. The available options for relapsed/refractory patients are limited and their prognosis is poor.

Recent studies showed that all cases of T-PLL show recurrent chromosomal aberrations and/or molecular mutations that could elucidate novel pathways for therapeutic interventions. The most frequent mutations detected by FISH and array comparative genomic hybridization (CGH) analyses include mutations involving the TCRA/D locus (86%), TCL1 rearrangement due to inversion of chromosome 14 (80%), deletions at 11q23 [the locus for ATM (69%)], abnormalities of chromosome 8 [gain of 8q (77%) and i(8)(q10) (61%)] leading to overexpression of MYC, and deletions in 17p resulting in TP53 loss (31%) [13]. Sanger sequencing and next-generation sequencing studies with both whole genome and whole exome sequencing (WGS and WES) have revealed novel gain-of-function mutations on IL2RG (2%), JAK1 (6–8%), JAK3 (20–40%), and STAT5B (36%), implicating a frequent constitutive activation of the IL2RG-JAK1-JAK3-STAT5B signaling pathway in more than 75% of the patients [2,13–17]. These findings suggest that JAK inhibition might represent an effective therapeutic strategy in these patients.

Herein we report a case report of sequential treatment with tofacitinib, a pan-JAK inhibitor, followed by combination with ruxolitinib (a JAK1/2 inhibitor), in a heavily treated patient with JAK3 mutated, relapsed T-PLL based on synergistic activity in in vitro modeling. The patient had exhausted all prior therapies including allogeneic transplantation. Lymphocytosis improved, leukemia cutis lesions resolved as did constitutional symptoms for a sustained period of ten months.

The successful implementation of JAK1/2 inhibitors in several myeloproliferative neoplasms led to the screening for JAK mutations in other hematologic disorders [18]. The JAK-STAT pathway constitutes a signal transduction system through which a large spectrum of extracellular cytokines converge towards an intracellular transduction system using the JAK family of protein-tyrosine kinases (JAK1, JAK2, JAK3, and TYK2 [Tyrosine kinase 2]) and seven STAT factors [19]. JAK1/2 and TYK2 are ubiquitously expressed whereas JAK3 is confined to hematopoietic, myeloid, and lymphoid cells [20]. The Janus kinases play an important role in normal hematopoiesis; JAK2 plays a role in the function and maintenance of hematopoietic stem cells as well as various stages of myelopoiesis [21], whereas JAK1 cooperates with JAK3 for lymphopoiesis [22]. Accordingly, it is hypothesized that when the pathway is activated, targeting JAK/STAT with inhibitors might have therapeutic relevance.

The rate of JAK3 mutations in T-PLL has been reported to be between 20 and 42% in different cohorts [2,13,14]. There are several mutations identified but the most common, the hotspot mutation JAK3^M511I accounts for 20–60% of the cases. This mutation has been reported to be the most efficient oncokinase with highest transforming properties [23]. Notably, in a mouse model of lymphoid transformation, JAK3 has been proven to lead to transformation through the constitutive activation of JAK1 and there was in-vitro proliferation inhibition by both tofacitinib or ruxolitinib [24]. Bellanger identified that JAK1 and JAK3 mutations can coexist in half the patients and increase the rate of resistance to JAK inhibitors [2,25,26].

We decided to use tofacitinib, a pan-JAK inhibitor (JAK1/2/3 and TYK2), based on the available trial data for rheumatoid arthritis with adequate safety and dosing data in order to target the JAK3 mutation in our patient. Our preclinical studies showed that tofacitinib had synergy in combination with ruxolitinib. This enhanced potency has been documented in-vitro in other disease models aiming to inhibit the IL2RG-JAK1-JAK3-STAT5B signaling pathway [24,25,27]. The findings of this single patient experience are exploratory in nature. However, they suggest the potential benefit of synergistic JAK inhibitors in relapsed/refractory T-PLL with tolerable side effects that warrant further exploration of the effect on more diverse patients and with a larger sample. Further studies to understand the mechanisms of synergy and resistance are also needed to confirm preliminary data of safety and efficacy in this rare, poor prognostic disease.