First report of clinical response to venetoclax combination with pentostatin in T-cell-prolymphocytic leukemia (T-PLL)

T-cell prolymphocytic leukemia is a postthymic mature lymphoid leukemia known to be associated with poor outcome. The median age at diagnosis is 61 years, and the clinical picture includes high white blood cell count (WBC), lymphocytosis, prolymphocytes in peripheral blood, splenomegaly, lymphadenopathy and skin rash. Only a minority of patients (around 6%) do not require treatment and are considered “indolent” at the time of presentation [1]. The immunophenotype is usually CD2⁺, CD3⁺, CD5⁺, CD7⁺, CD1a⁻, TdT^-, CD25^± with variable CD4/CD8 [1]. Since its recognition in 1970’s, several markers were identified to be associated with this disease and its aggressive nature [2]. Chromosome 14 abnormalities, which involve the proto-oncogene, TCL1 (T-cell leukemia 1) were the most common (overexpressed in 70% of patients) [1]. Others include chromosome X abnormalities involving MTCP1 (mature T-cell proliferation 1); chromosome 8 abnormalities, particularly isochromosome 8 where it leads to overexpression of MYC; deletion of chromosome 11q22.3 where it involves ATM gene; and complex karyotype [3–8]. Using whole-genome and whole-exome sequencing, gain-of-function mutations in IL-2RG, JAK-1/JAK-3 or STAT5B were observed in approximately three-quarters of patients [9–11]. Despite expanding knowledge into the molecular underpinnings of this disease, it continues to be therapeutically challenging. Historically, patients were treated with alkylating agents, such as chlorambucil, or combinations such as CHOP, with only a minority (30%) achieving short-lived responses (3 months) with median overall survival (OS) of 7 months [12]. Pentostatin in a previously treated population yielded a response rate of 45%, a median progression free survival (PFS) of 6 months and median OS of 9 months [13]. In this relatively chemo-insensitive leukemia, the anti-CD52 monoclonal antibody alemtuzumab, given intravenously (IV), was the first therapy to produce significantly improved responses. In a treatment-naïve population, the ORR was 91% with a PFS of 67% at one year and an OS of 37% at 4 years [14]. In the salvage setting, alemtuzumab is associated with an ORR of 50–76% with median PFS 4.5–7 months and median OS of 7.5–10 months [15–17]. Combining Pentostatin with alemtuzumab (IV) in treatment naïve patients produced 82% ORR with a median OS 10.4 months and a median PFS 4.3 months. However in salvage setting, ORR is 75% with median OS 2.6 months and median PFS of 2.6 months (only 5 patients) [1]. Stem cell transplant (SCT) is the main consolidative therapy in T-PLL, but incremental improvement in outcomes after SCT have been modest. The EBMT and Royal Marsden Consortium registry analyzed 41 T-PLL patients who received Allo-SCT in a salvage-like setting. The 3 years relapse-free and OS were 19% and 21%, respectively with a median follow-up of 36 months [18]. In prospective observational study of Allo-SCT in patients aged 65 years or younger, the 4-year PFS and OS were 30% and 42%, respectively. By univariate analysis, total body irradiation of 6 Gy or higher in the conditioning was the only significant predictor for a low relapse risk [19]. More recently, the ex vivo activity of the BCL2 inhibitor venetoclax was identified using functional drug screening [20]. This was confirmed in vivo with documented short-lived partial responses in 2 patients with relapsed/refractory disease.

Herein, we present a case of a novel combination of venetoclax and pentostatin producing a durable complete remission in a patient with multiply treated, relapsed and refractory T-PLL.

The patient is a 73 year old male who initially presented to an outside facility with severe abdominal pain, WBC of 350,000 and splenomegaly with spontaneous splenic rupture. Several weeks prior to his acute presentation, he noted increased fatigue, as well as dark purple spots under his left armpit, both hands, and inside his mouth. He underwent emergent leukapheresis and was started on hydroxyurea. Once the diagnosis of T-PLL was confirmed by peripheral blood flow cytometry, he was started on intravenous (IV) alemtuzumab. He received a standard dose of 30 milligrams IV 3 times a week for total 6 weeks. He responded with resolution of his symptoms, normalization of his counts and disappearance of splenomegaly and lymphadenopathy. By the time he was referred to our center (April-2016) for consideration for allogeneic SCT, his CBC had normalized and his symptoms had resolved. Bone marrow at our facility showed complete morphological remission, with positive MRD at a level of less than 0.01%. PET scan showed no foci of fluorodeoxyglucose (FDG)-avidity to indicate a metabolically active disease. Given his age and comorbidities, the higher risk of transplant related toxicity, and the continued risk of relapse after transplant, he was not considered to be a good candidate for SCT. The patient returned home and continued to follow with his local oncologist.

In November 2017, after being in remission for 18 months, he was noted to have a steadily rising WBC count with peripheral lymphocytosis. He was confirmed to have relapse of T–PLL and was retreated with single-agent IV Alemtuzumab. He had a partial response in his WBC count, but developed progressive and bulky bilateral cervical/axillary lymphadenopathy with constitutional symptoms while on therapy. He was referred back to our center for disease progression despite ongoing treatment with alemtuzumab. A PET scan showed metabolically active lymph nodes with maximum SUV of 6.3 (Range 3–6.3) (Figure 1(A,B)). Lymph node biopsy was morphologically consistent with T-PLL (Figure 2(A)). A panel of immunohistochemical stains supported the diagnosis (Figure 2(B,D,E)). Anti-BCL-2 antibody demonstrated uniformly strong BCL-2 expression (Figure 2(C)). Fluorescent in-situ hybridization (FISH) for TCL-1 dual-color break-apart probe was positive for TCL-1 rearrangement (Figure 2(F)). The concurrent bone marrow showed no morphologic involvement, but flow cytometry was positive for minimal involvement by T-PLL (0.4%). It is important to note that while PET scan imaging is not standard in T-PLL and not widely reported, t seems that T-PLL is a metabolically active disease and could be useful in diagnostic dilemmas or identifying involved areas amenable for biopsy.

Figure 1.

PET scan showed metabolically active lymph nodes above and below the diaphragm with maximum SUV of 6.3 (Range 3–6.3 with median SUV of 6) at the time of first relapse (A&B). After 2 cycles of treatment with Venetoclax + Pentostatin, PET scan showed complete resolution of the previously noted lymphadenopathy, consistent with complete metabolic response to therapy (median SUV of 2) (C&D). Panel E shows TLS labs, Day 1 is the day prior to starting venetoclax/pentostatin. As the treatment was given, a nice response in WBC was seen by day 5–6 of initiation. There was no evidence of TLS with just mild increase in TLS labs from baseline.

Figure 2.

H&E section of right axillary lymph node biopsy shows effacement of lymph node architecture with monotonous population of small to medium sized T-cells with clumped chromatin and inconspicuous nucleoli (panel A, 50X) which were positive for CD3 (panel B, 50X), BCl-2 (panel C, 50X), CD4, CD8 (panel D), CD7 bright and CD26 bright (panel E). Fluorescent in-situ hybridization (FISH) for TCL-1 break-apart probe was positive for TCL-1 rearrangement (panel F, 50X).

The patient was started on salvage treatment with cladribine 5 mg/m2 IV x 5 days in combination with alemtuzumab 30 milligrams 3 times a week. He had subjective transient reduction in lymph nodes that was quickly followed by progression (rising WBC and worsening lymphadenopathy) during the third week of treatment. Given the progressive disease and limited standard of care options available, we discussed the recent data of venetoclax in T-PLL and the rationale of combining pentostatin with venetoclax. The patient consented to the approach and was started on therapy. Pentostatin was administered at a dose of 4 milligrams/m2 weekly, while venetoclax started on 100 mg daily on Day 1–3, 200 mg on day 4–6 and 400 mg on day 7 onward. One cycle was considered 4 weeks. Rapid response in WBC was noted, with no evidence of Tumor lysis syndrome (Figure 1(E)). Along with resolution of constitutional symptoms, PET scan after cycle 2 (60 days from treatment start date) showed resolution of the previously present FDG avid lymphadenopathy, in addition to normal liver and spleen, consistent with complete metabolic response (CMR). Although PET is not an established tool in T-PLL, it did show dramatic response. Bone marrow showed morphologic remission with MRD positive only by flow cytometry (0.7%) (Figure 1(C,D)). The patient recovered his blood counts after the first cycle, a response that is consistent and complete remission. With subsequent treatment cycles, the patient experienced myelosuppression as expected and delayed count recovery; during cycle 3, the patient’s course was complicated by adenovirus viremia and nephritis, with BK virus detected in urine. This was treated with cidofovir weekly for five weeks with resolution of his symptoms and conversion to undetectable adenovirus PCR. After completing 3 cycles of combination treatment, he continued on single agent venetoclax; however, mild cytopenias persisted necessitating venetoclax dose reduction. The dose was reduced to 300 mg/day, and the patient was kept on levofloxacin, valacyclovir, and bactrim for infectious prophylaxis. The patient showed possible signs of relapse disease while on single agent venetoclax, with 1 cm palpable cervical lymph node four months after achieving CR. Since the patient’s infectious issues were resolved, he was restarted on treatment with pentostatin and regular dose venetoclax (400 mg/day). The patient had stable disease, at which the dose of venetoclax was increased to 500 mg/day and then to 600 mg/day in pursuit of consolidating remission, however the patient disease progressed one year after initiation of treatment and 10 months after achieving CR, yielding 2 months as the time to best response, 10 months as duration of response. This compared favorably to reported single agent pentostatin or alemtuzumab salvage treatment, where PFS is reported to be 4.5 to 7 months and OS of 7.5 to 10 months [13,16,17].

One recent approach to developing new therapies in oncology consists of identifying a pathogenic driver mutations, and then selecting treatment to target those aberrant growth and survival mechanisms. In T-PLL, Boidol et al reported on venetoclax having the strongest differential response for T-PLL, with sensitivity that correlated with BCL-2 expression ex vivo. This activity of venetoclax against T-PLL was observed clinically with a recent report of partial remissions (albeit transient) in two relapsed/refractory patients treated with single agent venetoclax [20]. The relapse seen in the patient who was continued on treatment suggests an escape mechanism, and arguing for rational venetoclax-based combination therapy. One approach is to combine venetoclax with previously known active agents, as it was the case in our patient with pentostatin [13]. Another approach is to identify and target venetoclax resistance mechanisms. Although not specific to T-PLL, overexpression of antiapoptotic BCL-2 family proteins, such as BCL-XL and MCL-1 is a well described phenomena that may engender resistance [21]. In CLL, another recently reported mechanism includes mutations/deletions in BTG1, BRAF, CDKN2A/B and PD-L1 amplification [22]. In DLBCL/MCL, a resistance mechanism was described through suppressing PTEN expression, leading to enhanced AKT pathway activation and concomitant susceptibility to PI3K/AKT inhibition [23]. Combining venetoclax with CDK9 inhibitors that downregulate MCL-1 is synergistic [21]. Other potentially effective combination is based on a recent usage of drug combination prediction and testing (DCPT) platform, where synergistic activity of tacrolimus and temsirolimus with venetoclax in T-PLL was shown, a signal that can be further explored clinically [24]. In another comprehensive bioinformatics analysis, STAT3 and IRS1 were the core upregulated genes associated with the progression of T-PLL [25]. However, it is worth noting that despite this prevalence and dependence of JAK-STAT pathway mutations in T-PLL, ex vivo sensitivity to JAK-STAT inhibitors did not correlate with high clinical response [26]. In one report, it was suggested that T-PLL cells are unable to activate TP53 in response to double-strand break induced by alkylating agent, explaining in part the clinical chemo-refractory behavior of T-PLL. This dysfunctionality in TP53 is not related to TP53 mutation, rather to dysfunctional proximal ATM that renders TP53 inactive. As such, strategies targeting TP53 reactivation by MDM2 inhibitors (such as idasanutlin) as well as restoring DNA-repair with histone deacetylase inhibitor, in a tumor with recurrent mutations in DNA repair molecules and histone modifiers, is appealing and shown to be synergistic in murine models [27].

With the venetoclax/pentostatin combination, adverse effects were mainly myelosuppression and infection. This patient was noted to be neutropenic (Mild-Moderate) for two weeks during cycle 3, and granulocyte-colony stimulating factor was given during this period. It was also during this time that the patient developed adenovirus viremia with suspected adenovirus nephritis. His urine test for BK virus was also positive. It is unclear given the etiology of this infections, whether this is related to a combined treatment of venetoclax/pentostatin, or to previous repeated treatment with alemtuzumab. The patient had no other clinically significant infectious issues.

In conclusion, we report on the successful treatment of a patient with multiply treated relapsed and refractory T-PLL with venetoclax/pentostatin combination. The combination appeared to be well tolerated and led to a complete remission in a patient with refractory disease. Although a single-case case, this represents an important signal of response in a disease state where there are no effective options. Future evaluation of this strategy prospectively is warranted, and future trials examining novel targeted combinations is crucial.