Anna Korycka-Wolowiec, Dariusz Wolowiec, Aleksandra Kubiak-Mlonka & Tadeusz Robak
Abstract
Introduction: Venetoclax, an antagonist of BCL-2 protein plays an important role in the pathogenesis of chronic lymphocytic leukemia (CLL). It has been approved by the FDA for the treatment of relapsed/refractory CLL with del17p,and by the EMA for patients with del17p/TP53 mutation who have failed aBCR inhibitor, or in patients without those aberrations who have failed previous therapy, regardless of their genetic/molecular profile. Venetoclax in combination with rituximab has been also approved for the treatment of CLL after at least 1 prior therapy, regardless of del17p.Areas covered:This article reviews the chemical structure, mechanisms of action, pharmacokinetic and the clinical applications of venetoclax in monotherapy and in combined treatment of CLL. Publications dated 2010 through March 2019 were obtained from the MEDLINE database. The proceedings of the American Society of Hematology held during the last five years were also included.Expert opinion: Venetoclax shows high efficacy, a favorable toxicity profile and a high rate of minimal residual disease negativity, which is thought to have an impact on overall survival. It is efficient inpatients with del17p/TP53 mutations, the incidence of which increases during clonal CLL evolution, and after failure of BCR pathway inhibitors.
Keywords: adverse events, chronic lymphocytic leukemia, clinical efficacy,pharmacokinetics,toxicity, venetoclax
1.Introduction
Chronic lymphocytic leukemia (CLL) is a lymphoproliferative disorder which typically occurs in elderly patients with a median age at diagnosis of 71 years [1]. The disease leads to an uncontrolled, clonal accumulation of morphologically mature B cells, caused by the inhibition of either extrinsic apoptosis, triggered via death receptor and extracellular ligands, or intrinsic apoptosis in which the intracellular signaling pathway plays the most important role. The activation of this pathway induces the release of cytochrome c from mitochondria and results in effector caspase activation [2-4].The key regulators of the intrinsic apoptotic pathway are BCL-2 family proteins, the role of which is well documented in the pathogenesis of some hematological disorders [5]. The overexpression of the anti-apoptotic BCL-2 protein, typical for CLL and some other B-cell malignancies, allows malignant cells to evade apoptosis by sequestering pro-apoptotic proteins [5, 6]. However, BCL-2 has been implicated not only in tumorigenesis, but also in increased chemotherapy resistance [5].
Due to its role in both malignant cell survival and chemoresistance, BCL-2 has emerged as an attractive therapeutic target. Several early attempts have been made to target BCL-2 for therapeutic purposes, and a number of small molecules binding to BCL-2 family anti-apoptotic factors have been the subject of preclinical and clinical studies [7-9]. Oblimersen was the first antisense oligonucleotide designed to target BCL-2 mRNA; however, it showed only limited efficacy in mono- and polychemotherapy. Other agents, gossypol-polyphenolic (AT-101) and obatoclax (GX15-070), have shown major toxicity with limited efficacy in relapsed/refractory (R/R) CLL patients [7-10].
During the last few years, a greater understanding of the molecular pathogenesis of hematological malignancies has led to the breakthrough in the development of a new class of BCL-2 family inhibitors able to mimic the function of pro-apoptotic BH3-only proteins (BH3s). They block the hydrophobic BH3-binding domain of BCL-XL and seem to represent a promising therapeutic strategy [11-12]. ABT-737 was the first compound synthetized as a BH3-only mimetic. It has poor aqueous solubility, it is not orally bioavailable and induces thrombocytopenia [13]. In contrast, navitoclax (ABT-263) is orally bioavailable with a binding profile similar to ABT-737. It was the first generation of NOXA-like BH3 mimetics to be developed for apoptosis-directed therapy of CLL [14]. Although navitoclax showed promising activity in R/R CLL patients, with an overall response rate (ORR) of 33%, the compound induced thrombocytopenia and demonstrated extensive toxicity in combination with other neoplastic drugs [15].
Recent preclinical and clinical studies have shown that of the BH3-only mimetics, venetoclax (ABT-199, GDC-0199, VEN) is the most suitable agent for clinical use. It is a unique, small and high selective, orally-bioavailable BCL-2 inhibitor that directly induces apoptosis independently of TP53, with an acceptable safety profile [16-21]. Venetoclax reduces the lymphocyte counts in a dose-dependent manner and when used as a single agent, allows a high ORR to be achieved in R/R CLL patients, together with a favorable complete remission rate (CR), progression free survival (PFS) and overall survival (OS). The clinical response was independent of del17p, TP53 mutation and TP53 function [17]. It is important to note that VEN allows to achieve durable response with a high rate of undetectable minimal residual disease (MRD), and the MRD negativity is associated with improved outcome regardless of therapy [22]. Results of the MURANO clinical trial revealed that VEN can be also used in combination with rituximab (RTX) for CLL patients after at least 1 prior therapy. Significant PFS benefit and MRD eradication was demonstrated for fixed-duration VEN-RTX treatment regimen as compared to VEN-bendamustine (BENDA) [23,24].
Venetoclax is well tolerated in patients with high-risk prognostic factors, but does not induce such high levels of thrombocytopenia as typified by earlier BCL-2 inhibitors. The most frequent adverse events (AEs) include grade 3 to 4 neutropenia, and diarrhea, nausea, and fatigue usually grades 1 to 2. Serious AEs (SAEs) such as febrile neutropenia, pneumonia, thrombocytopenia or upper respiratory tract infection were rarely observed [18]. Although VEN is efficacious in monotherapy, the CR rate was often low, mainly due to residual lymphadenopathy; hence, later studies explored the potential of combined VEN therapy with other active agents such as anti-CD20 monoclonal antibodies or B-cell receptor (BCR) pathway inhibitors [17, 18, 25-27].The efficacy of VEN is not limited to CLL. The drug is also active in other hematological disorders such as multiple myeloma, elderly acute myeloid leukemia and non- Hodgkin’s lymphoma, as well as in solid tumors [28, 29].
2. BCL-2 proteins and their role in mitochondrial apoptosis
BCL-2 proteins regulate mitochondrial apoptosis and play a crucial role in the immune response, tissue homeostasis and cancer development. Based on differences in structure and function, vertebrate proteins related to BCL-2 family bear from one to four BCL-2 homology domains (BH) and can be categorized into three main functional groups. The first group of four tandem BH domains contains anti-apoptotic, pro-survival proteins (BCL-2, MCL1, BCL-XL, BCL-W, BFL1), the main function of which is to inhibit pro-apoptotic proteins. The second group of three BH domains contains the effector proteins (BAX, BAK) required to carry out the pro-apoptotic effects mediated by other proteins within this family. Finally, the third group of pro-apoptotic BH3s with a single BH domain (BIM, tBID, BAD, PUMA, NOXA, BIK, BMF and HRK) shares a sequence homology only in their amphipatic 15-amino-acid “-helical BH3 region, which is required for their death activity [9, 11, 13, 30, 31]. Similarly to all multidomain family members, the BH3s have a C-terminal transmembrane (TM) domain for anchoring to organelles, most notably the mitochondrial outer membrane (MOM), and execute their function by the binding of their BH3 domains into the hydrophobic binding groove of multidomain pro-apoptotic or anti-apoptotic members [11, 32, 33]. They can act as activators (BIM, BID/tBID, PUMA) or sensitizers (BAD-like domains proteins, NOXA) [28].
In normal cells, pro-survival proteins prevent BAX and BAK activity by binding the BH3-domain of any destabilized BAX or BAK monomers and prevent MOM permeabilization. It is controversial whether pro-apoptotic BH3s proteins promote apoptosis through direct protein-protein interaction with each other, mediated by the BH domain or indirectly (displacement model) by inhibition of the anti-apoptotic BCL-2 family members [12].Therefore, two main models of BAX and BAK activation have been proposed [9, 33- 35]. In the direct activation modelpro-survival proteins sequester activator BH3s and prevent BAX/BAK activation,whereas sensitizers bind directly to these pro-survival proteins and allow the release of the activators, which acquire the capacity to activate BAX/BAK proteins.BAX/BAK permeabilize MOM after their homo-oligomerization with subsequent loss of mitochondrial membrane potential, activate the caspase cascade and promote apoptosis [11, 31, 32, 34, 35].In contrast, indirect activation model (displacement model) postulates that BAX and BAK are constitutively active and they must be sequestered by anti-apoptotic proteins to allow cell survival [33, 35]. In response to an apoptotic signal, the activator BH3s displace BAX and BAK from their binding with anti-apoptotic BCL-2 family members and enable active BAX and BAK to trigger the intrinsic apoptotic pathway. It is also possible that these two models co- exist leading to a unified model which postulated that anti-apoptotic proteins sequester both BH3s and activated BAX/BAK [36]. It has been also suggested that NOXA strongly and directly activates BAX/BAK independently of BID, BIM and PUMA, suggesting that NOXA also plays an important role in BAX/BAK activation. Additionally, autoactivation of BAX and BAK can occur independently of activator BH3s through downregulation of BCL-2, BCL-XL and MCL-1 [33].Determining the dependence of CLL cells on particular anti-apoptotic BCL-2 family proteins has become critical for deciding a personalized therapy in this disease.
3.The structure and mechanisms of action of venetoclax
Venetoclax,chemically(4-(4-(2-(4-chlorophenyl)-4,4-dimetylocyclohex-1-en-1-yl) methyl)piperazin-1-yl)-N-(3-nitro-4-[(tetrahydro-2H-piran-4-ylomethyl)amino]phnyl sulfonyl)-2-(1H-pirrolo [2,3-b) pirydin-5-yloxy)benzamide is a highly selective, bioavailable BCL2 inhibitor belonging to the BH3-only mimetic group. In comparison to its predecessor, navitoclax, venetoclax lacks a thiophenyl group located at the P4 hotspot which re-engineers the BH3-binding domain [13]. Its empirical formula is C45H50ClN7O7S, with a molecular mass of 868.4g/mol [13, 37] (Figure 1). VEN, as a BH3-only mimetic, binds directly to anti-apoptotic BCL-2 and displaces activator BH3s, BIM, from its link with this prosurvival protein; this uncoupling subsequently allows BIM to activate BAX/BAK leading to their homo-oligomerization. The BAX/BAK permeabilize the MOM by forming proteinaceous pores, promoting release of cytochrome c and Smac/Diablo from the mitochondria to the cytosol and inducing caspase-mediated apoptosis (Figure 2) [11, 31, 38].Importantly,VEN induces the intrinsic apoptotic pathway via a TP53-independent mechanism. It directly antagonizes BCL-2 function downstream of TP53 and may circumvent the block to apoptosis associated with loss of TP53 function.Therefore VEN can restore apoptosis also in TP53-disrupted cells and may overcome the negative prognostic impact of TP53 disruption [19,20].
4. Pharmacokinetics and metabolism
Venetoclax demonstrates a strong affinity only for BCL-2 (Ki <0.01nM), with more than 100-fold less affinity for other anti-apoptotic proteins such BCL-XL (Ki=48nM), BCL- W (Ki=245nM) and MCL1 (Ki >444nM) [20].
Roberts et al.[18] in a first-in-human study of VEN monotherapy in 116 relapsed/refractory CLL/small lymphocyte leukemia(SLL)patients, including a del(17p) cohort, estimated the pharmacokinetic profile of VEN and reported that a peak VEN level was attained six to eight hours after the first dose and the terminal half-life (T1/2) was approximately 19 hours after a single 50-mg dose. No maximum tolerated dose was identified. The assessment of blood samples of 155 patients for VEN pharmacokinetics tests collected during the ramp-up period of the drug administration and on day 1 of weeks 8, 12, 16 and 24,and every 12 weeks thereafter, revealed that the mean post-dose (eight hours) plasma concentrations increased with the weekly increase in VEN dose during the ramp-up period to reach 1.89 μg/ml on week 5 day 1, at the first 400 mg dose [39]. The mean predose concentration at the dose of 400mg ranged between 0.69 and 0.99 μg/ml across visits between weeks 8 and 120 [39]. A phase 1b study by Seymour et al. [25] determined the maximum tolerated dose of VEN and the recommended phase 2 dose of the drug, and assessed the pharmacokinetic profile when VEN was given in combination with RTX. For patients receiving the recommended VEN dose of 400 mg, the median time from dosing to maximum serum concentration (Tmax) was six hours, and the maximum plasma concentration (Cmax) was 1.93± 0.69 μg/mL. The log-transformed dose-normalised mean Cmax and area under the plasma concentration–time curve (AUC) for VEN were not affected by RTX.
Venetoclax bounds to the plasma proteins in more than 99%, with a T1/2 of 16-19 hours [39]. Peak concentration was observed after four to five hours and it was delayed by approximately two hours when the drug was taken with a meal [36]. Low-fat meals increased VEN exposure by 3.4-fold, whereas high-fat meals increased it by ~5.2-fold. However, Salem et al. [39] demonstrated that inpatients with del17p, no specific recommendation in terms of fat content in the meal is needed for the intake of VEN.
The drug is metabolized by CYP3A4 and CYP3A5, and it is substrate of the P- glycoprotein efflux pump [36, 37]. It is important to remember that concomitant administration with preimplantation genetic diagnosis strong CYP3A inhibitors was identified as the main factor affecting VEN bioavailability and clearance [40]. For this reason, concomitant administration of VEN should be avoided with protease inhibitors, some macrolide antibiotics, azole antifungals or P-glycoprotein inhibitors such as amiodarone,clarithromycin,ciclosporin,colchicine,diltiazem,erythromycin, felodipine, lansoprazole, omeprazole and other proton-pump inhibitors,nifedipine, paroxetine, sertraline, quinidine, tamoxifen and verapamil. If the administration of above agents is necessary, the venetoclax dose should be reduced [36, 41]. Moderate and strong CYP3A inhibitors decreased venetoclax apparent clearance by 19% and 84%, respectively, whereas weak CYP3A inhibitors and inducers did not affect drug clearance [40]. Co-administration of digoxin and VEN increases digoxin Cmax by 35% and AUC by 9%, therefore digoxin should be administered six hours prior to VEN [42, 43].Jones et al. [40] analyzed 7483 VEN plasma concentrations obtained from 505 subjects receiving VEN, and demonstrated that food increased bioavailability by 2.99- to 4.25-fold when compared to the fasting state. It is also important to note that fruits, especially grapefruit fruit and juice are potent inhibitors of CYP3A4, and thus can affect the metabolism of a variety of drugs by increasing their bioavailability [44]. On the other hand, co-administration of gastric acid-reducing agents had no impact on the rate or extent of VEN absorption [40].
Additionally,female subjects had 32% lower central volume of distribution when compared to male subjects. However,sex and subject population (healthy vs. oncology subjects) had only a minimal impact on VEN pharmacokinetics. The pharmacokinetic profile of the drug in CLL patients with del17p is comparable to that seen in the overall CLL population, as well as in non-Hodgkin’s lymphoma patients [39].
5. Clinical efficacy
A phase 1 study (M12-175) by Roberts et al. [18] evaluating the efficacy of VEN after oral administration of eight dose levels ranging from 150 to 1200 mg per day found the drug to be active in all 56 enrolled patients in the dose escalation cohort. In the expansion cohort, 60 additional patients were treated with a weekly stepwise ramp-up in doses as high as 400 mg per day. The ORR did not vary according to age, number of previous therapies or any of the risk factors typically associated with a poor outcome for chemoimmunotherapy-based treatment. Among the whole cohort of 116 patients enrolled into the study, the pooled ORR was 79%, with 20% displaying CR or CR with incomplete recovery of blood counts (CRi). MRD negativity in the bone marrow (BM) was achieved in 5% of all study patients aged below 86 years, but in 35% of those subjects who were MRD tested (Table 1). Similarly, the ORR was 79% inpatients with fludarabine-resistant disease and 71% in those patients with del17p in whom loss of TP53 function was detected. CR or CRi inpatients with del17p was 16%. During the follow-up period, disease progression occurred in 35% subjects,with Richter’s transformation in 16%; progression was more commonly observed among patients with del17p. The median PFS was 25 months when data was pooled across dose-escalation cohorts, and the two-year overall survival (OS) estimate for all the patients was 84%. The median PFS was 16 months for patients with del17p CLL across all doses, whereas 71% of patients without this aberration remained progression-free at 15 months.
A multicenter, open-label, single arm, phase 2 study (M13-982) conducted by Stilgenbauer et al. [17] assessed the activity and safety of VEN as monotherapy in 107 R/R patients with del17p. Venetoclax was given at a dose of 400 mg daily (after initial ramp-up starting at 20 mg) and continued until disease progression or appearance of another reason for discontinuation. The ORR was 79.4%, including CR/CRi in 8% of patients, partial remission (PR) in 69% and nodular partial remission in 3%. At a median follow-up of 12.1 months, median time to first response by independent review committee was 0.8 months, the median time to CR or CRi was 8.2 months. Among 45 patients assessed for MRD in peripheral blood, 18 were found to be negative (40%), with a median time to achieving peripheral blood (PB) MRD negativity of 8.8 months. Ten out of these 18 patients were tested for MRD in BM, and six (60%) were negative (Table 1).A recent study published by Stilgenbauer et al. [45] performed on 158 treatment-naïve or R/R del17p CLL patients who received VEN at 400 mg per day after an initial dose ramp-up revealed an ORR of 77% with 20% of CR and estimated PFS of 54% at 24 months. Additionally, in this high-risk population, a high rate of blood MRD < 10-4 was achieved (30%). Coutre et al. [46] assessed the efficacy of VEN inpatients who progressed during or after idelalisib (IDE) treatment. The initial dose of VEN was 20 mg daily, followed by interpatient ramp-up to 400mg daily.The ORR was 67%.
The median PFS has not been reached but the estimated 12 months PFS was 79%. In a phase 2 study, Jones et al.[47] assessed the VEN activity in 54 CLL patients refractory to or who had relapsed after BCR inhibitors: ibrutinib (IBR) (43 patients) or IDE (21 patients). The ORR was 70% among patients refractory to IBR and 62% in those refractory to IDE. After 11.8 months of follow-up, median duration of response, PFS and OS have not been reached. The estimated 12-month PFS was 80%, and 14 out of 42 (33%) patients were MRD-negative in the PB. In the interim analysis of a multicentre, non-randomized, open-label, phase 2 trial (M14-32),which assessed VEN in R/R CLL patients progressing after IBR, 84 patients were enrolled to the main cohort, and 83 others were recruited to the expansion. The efficacy and safety analysis included 43 and 48 patients, respectively. The median follow-up was 19 months for the main cohort, 12 months for the expansion cohort and 14 months for all 91 patients. The ORR was achieved in 70%, 60% and 65% of patients, respectively [48].Recent studies have examined the efficacy of VEN in CLL patients who progressed after IBR or IDE as the last BCR inhibitor before enrollment; their findings indicate similar ORR for R/R subjects after IBR and IDE (70 vs 67% after IBR vs IDE in main cohort, and 60 vs 67% after IBR vs IDE in expansion cohort, respectively) [46, 48]. The median PFS for patients treated with IBR followed by VEN was 24.7 months, with an estimated 12-month PFS of 75%, whereas the median PFS has not yet been reached inpatients progressing after IDE [46, 48].
Interesting results have been reported by Mato et al. [49] in a multicenter, retrospective analysis of 683 CLL patients treated with either IBR, or Selleckchem AZD7545 IDE or VEN. The aim of the study was to determine the optimal sequence of these drugs. For the patients who discontinued IBR, the results found IDE or VEN to demonstrate superior performance to chemoimmunotherapy. The ORR scores of IBR and IDE as the initial therapy were 69% and 81%, respectively; however, the ORR was 79% for VEN and 46% for IDE among patients who previously discontinued IBR treatment due to toxicity or progression, According to ESMO Clinical Practice Guidelines for diagnosis, treatment and follow- up of CLL, patients with del(17p) or TP53 mutation who are unsuitable for BCR inhibitor therapy may be treated with the BCL2 inhibitor -venetoclax [17,18,50].Current studies tend to focus on the efficacy and safety of combining venetoclax with other drugs; these can be anti-CD20 antibodies or other small molecules, particularly BCR pathway inhibitors, targeted against other constitutionally active pathways. Combinations of VEN with chemoimmunotherapy are also being explored.
A study examining the combination of navitoclax, another BH3 mimetic hepatopulmonary syndrome BCL-2 inhibitor, with RTX for treatment-naïve CLL patients opened the door to further extensive research on BCL-inhibitors combined with other neoplastic drugs [51]. Following promising results of preclinical studies as well as VEN monotherapy in R/R CLL patients, clinical trials were performed evaluating the efficacy and safety of the combination of VEN with monoclonal antibodies or BCR pathway inhibitors [26, 51-56].
A phase 1b (M13-365) dose-escalation study assessing the efficacy of VEN plus RTX was performed in 49 R/R CLL patients by Seymour et al. [25]. Oral VEN was given daily using a stepwise weekly escalation schedule to designated cohort doses ranging from 200 mg to 600 mg (400 mg for the expansion cohort). The median time to first response was 2.9 months and CR were achieved after a median of 9.2 months. Response was observed in 86% patients, with CR in 51%. The two-year estimate for PFS was 82%, and ongoing response was 89%. Among 42 patients in whom MRD in BM was evaluated, 28 (67%) had no MRD, including 20 out of 25 (80%) with CR. In 22 out of 28 patients (79%), negative MRD was achieved within seven months of treatment.Based on the data from the phase 1b study, a phase 3 MURANO study was performed to compare the efficacy of combined VEN with RTX (VR) in 194 R/R patients with that of combined BENDA with RTX (BR) in 195 patients [23]; better results were obtained for the VR regimen. The median two-year rate of PFS was 84.9% in the VR cohort but only 36.3% for the BR group. The two-year PFS was higher after VR than after BR therapy, both for patients with del17p (81.5% versus 27.8%) and for those without (85.9% versus 41.0%).
Moreover, higher MRD negativity rates were observed for VR (62.4%) than BR (13.3%) (Table 2) [23]. Recently, post-treatment follow-up of MURANO study with a median follow- up of 36 months was published. PFS and OS remained superior in VR than in the BR arm. At the end of the combined therapy patients in the VR arm had a higher rate of MRD negativity in the peripheral blood compared to the BR group. Undetectable MRD at the end of the combined therapy was predictive for longer PFS. In patients without eradication of MRD, low MRD burden predicted longer PFS as compared to patients with high level of MRD [24]. Based on the results of MURANO trial the European Medicines Agency (EMA) approved VEN plus RTX with a 24-month fixed duration for the treatment of patients with R/R CLL, following at least one prior therapy. The drug is given in combination with RTX for 6 cycles and then in monotherapy to complete the 24-month treatment regimen [57].The studies have also assessed the efficacy of VEN with obinutuzumab (Gazyvaro, GA) for treatment-naïve CLL and R/R patients [58, 59]. Cramer et al. [58] reported a primary endpoint analysis of VEN and GA (VG) in patients previously treated with BENDA in a phase 2 study. Among 63 patients analyzed, 34 were treatment-naïve and the others were R/R with a median of two previous therapies: BR, fludarabine /cyclophosphamide /rituximab (FCR), IBR or IBR plus IDE. All patients had completed induction treatment. An ORR was 100% in treatment-naïve and 90% in R/R patients. CR and CRi values were 40% for all patients, 50% for treatment-naïve and 28% for R/R. MRD negativity (<10-4) in PB was documented in 87% of patients, including 91% treatment-naïve and 83% R/R, whereas MRD negativity was obtained in the BM in 13% of patients (12% untreated and 14% R/R ).Flinn et al. [26, 59] in a phase 1b study (GP28331) indicated that the combination of VEN with GA may represent a potentially important treatment option, not only for R/R but also for treatment-naïve CLL patients . Combined therapy with VG in 32 previously untreated CLL patients returned an ORR of 100% with 56.3% CR. MRD negativity in PB was noted in all patients, whereas MRD in BM was undetectable in 62.5%. The PFS rate at 12 months was 100%. First-line VEN studies are underway on a range of subjects including patients who are unfit or elderly. The tolerability of VEN and
GA in patients with co-existing conditions was examined during a safety run-in phase in the CLL14 trial. The results are promising, and suggest venetoclax may one day be used as a targeted agent in treatment-naïve CLL patients
[50, 60].
The CLARITY phase 2 study performed by Hillmen at al. [55] assessed the combination of IBR with VEN in 50 R/R CLL patients. This treatment was well tolerated, with only a single case of laboratory tumor lysis syndrome (TLS). After six months of combined IBR+VEN therapy, all patients displayed an objective response and 18/38 (47%) were in CR or CRi, with MRD negativity in BM stated in 32% of patients. Afterwards, the depth of response was improved after 12 months of the combined therapy. In 87% of patients no morphological evidence of CLL in the bone marrow biopsy was detected. In 94% of them leukemic cells accounted for less than 1% of all nucleated cells in the BM aspirate. MRD negativity (less than one CLL cell in 104 leukocytes) was achieved in 58% patients in the peripheral blood and in 41% in the bone marrow [56].An interim analysis of the phase 1b study (GO28440) by Stilgenbauer et al. [27] evaluated the combined treatment of both VEN+BENDA+RIT (VBR) and VEN+BENDA+GA (VBG) in untreated or R/R CLL patients. In the VBR group, response was achieved in 26/27 (96%) R/R and in all 14 (100%) treatment-naïve patients, with CR/CRi in 26% and 43% patients, respectively; whereas in the VBG group, the ORR was also 100% while CR/CRi was 43% for first-line treatment. In VBR group MRD negativity was reported in 76% R/R and 67% untreated patients, while in VBG patients MRD was undetectable in 50% of treatment-naïve subjects (Table 2).
Other clinical trials are evaluating the efficacy and safety profiles of VEN with GA and VEN with IBR for R/R, or treatment-naïve CLL patients [59, 61, 62]. Initial results by Rogers et al.[62] indicate an ORR of 92% and a CR or CRi of 42%. By the end of the assessment, all patients had achieved MRD negativity in either PB or BM, and in 50% of the patients, MRD was eradicated in both PM and BM. Those results justified the undertaking a novel phase 2 study on the efficacy and toxicity of this regimen in both untreated and R/R patients. A phase 3 trial with this regimen are now being initiated by US Cooperative Group.Although VEN demonstrates the high efficacy in CLL treatment, many patients do not respond or respond but subsequently progress. Therefore, for therapy to be optimized, it is crucial to understand the mechanism by which VEN resistance is acquired.In mouse lymphoma cells, the mutations F101C and F101L are believed to prevent VEN binding to the BH3 domain suppressing mitochondrial apoptosis. In resistant human lymphoma cells, a missense mutation in the C-terminal transmembrane BAX domain (G179E) prevents the anchoring of BAX to mitochondria and blocks venetoclax–induced apoptosis both in vitro and in vivo. This mutation was also found to induce partial cross-resistance to other antineoplastic drugs [63]. It is also possible that VEN monotherapy induces changes in the BCL-XL and BRF-1 proteins. Stimulation by CD40 and IL-4 in vitro may also increase BFL-1, BCL-XL and MCL-1 expression in CLL cells and may be reversed by antiCD-20 monoclonal antibodies and BCR pathway inhibitors. Therefore, combined therapy with VEN and BCR inhibitors or anti-CD20 monoclonal antibodies may help to overcome resistance to venetoclax monotherapy [28, 64, 65]. In a recent study, Blombery et al. [66] investigated the mechanisms of secondary resistance and documented a single heterozygous nucleotide variant in BCL-2 (NM_000633.2:c.302G>T, p.(Gly101Val), in 7 out of 15 samples obtained from relapsed, heavily pre-treated CLL patients. Gly101Val mutation was first detected at low variant allele fraction after 19-42 months of VEN therapy, up to 25 months earlier than when standard disease progression criteria were met. The Gly101Val mutation was not detected prior to venetoclax treatment in this cohort and was not detected in a series of samples from patients who had not received VEN.The findings of Deng et al. [53] that Bruton’s tyrosine kinase inhibition increases BCL-2 dependence and enhances sensitivity to VEN in CLL patients suggest that these combinations merit further clinical study. Recently opened or ongoing clinical trials in CLL involving the IBR/VEN combination hold promise for the design of a well-tolerated, time- limited therapy for CLL that can allow deep and durable remissions to be achieved [54, 61]. A list of selective, active, ongoing clinical trials is presented in Table 3.
6.Adverse events
Several studies have demonstrated that VEN is relatively safe when used in monotherapy, as well as in combination with other agents [17, 18, 47, 48]. The most important AEs associated with VEN monotherapy reported in a clinical trial of Roberts et al. [18] was TLS, which was observed in 18% patients of the dose escalation cohort. This AE occurred either after the first dose of 200, 100 or 50 mg of VEN or immediately after ramp-up of the dose to 150, 800 or 1200 mg. It was avoided in the expansion cohort by extending the stepwise ramp-up starting at 20 mg and providing prophylactic treatment with intravenous fluids and allopurinol (orrasburicase in severe TLS), as well as strict monitoring of laboratory signs of TLS [18].Among the types of hematological toxicity associated with VEN monotherapy observed by Jones et al. [48], neutropenia of any grade was the most common, including grade 3-4 neutropenia in 31-51% subjects. The other hematological AEs were grade 3-4 anemia (29- 36%) and thrombocytopenia (16-29%). A comprehensive safety analysis of VEN monotherapy in R/R CLL patients with del(17p) and/or in patients who had received prior BCR inhibitors by Davids et al. [67] found cytopenias (neutopenia 40%, anemia 31% and thrombocytopenia 21%) to be the most common AEs of any grade.
The non-hematological toxic effects included self-limited diarrhea (40-52%), nausea (41-47%), upper respiratory tract infection (26-48%) and fatigue (31-40%).Additionally, cough of any grade (30%), pyrexia (26%), headache (24%), constipation (21%) vomiting (18- 23%), peripheral oedema (23%) were reported [18, 47, 48]. Davids et al. [67] observed mild gastrointestinal events (diarrhea and nausea ca. 40% each) and serious infections in 15 % of patients (Table 1). The most common AEs related to VEN treatment observed in more than 10% of patients were neutropenia (33%),nausea(25%),diarrhea (24%), fatigue (15%), anemia(12%) and hyperphosphatemia (10%). The most common grade 3-4 AEs were hematological ones, mainly neutropenia (30%). Among the 166 patients included in the study, low TLS was reported in 61 (37%), medium in 59 (36%) and high in 46 (28%) patients [67].Roberts et al. [18] report the occurrence of SAEs such as febrile neutropenia in 6%, pneumonia in 4%, upper respiratory tract infection and immune thrombocytopenia in 3% patients. Similarly, SAEs were also observed by Stilgenbauer et al.[17] in 55% of tested patients, the most common being pyrexia and autoimmune haemolytic anaemia (AIHA) (7% each), pneumonia (6%) and febrile neutropenia (5%), independent of VEN administration. Common SAEs documented by Jones et al. [47] included febrile neutropenia and pneumonia and additionally multi-organ failure, septic shock, increased blood potassium and, in one patient, laboratory TLS.
In the MURANO study, a higher percentage of all grade 3-4 events was reported in the VR group (67%) than the BR group (53.3%). The most common AE of any grade in both the VR and BR-treated groups was neutropenia (57.7% vs 38.8%). In patients with severe neutropenia the G-CSF administration is recommended or the dose of the drug should be reduced. On the other hand, the percentage of grade 3-4 febrile neutropenia and grade 3-4 infections was lower in VR patients (3.6 and 17.5%, respectively) than in the BR group (9.6 and 21.8%) [23]. Lower frequencies of incidence of grade 3-4 anemia and thrombocytopenia were observed in the VR than the BR group. Infusion related reactions (IRRs) were observed in 5.3% patients in BR and in 1.5 % in VR group, whereas the rate of grade 3-4 TLS was 2.1 % in the VR group and 0.5 % in the BR group [23]. Additionally, the two groups demonstrated similar occurrences of Richter’s transformation: six VR and five BR patients.In the study ofFlinn et al. [59] performed on patients treated with VG nausea (65.5%), diarrhea (50%), pyrexia (46.9%), fatigue (43.8%), headache (37.5%) and vomiting (32.4%) were the most common AEs. IRRs were observed in 56.3% patients, which was connected with GA administration. Grade 3-4 hematological AEs included neutropenia in 40% patients, thrombocytopenia in 12.5%, febrile neutropenia in 12.5% and anemia in 9.4%.The above studies suggest that venetoclax shows a safe toxicity profile and the adverse events, especially the hematological and gastrointestinal ones, are usually manageable. Gradual dose escalation appears to minimize the risk of the TLS, the major toxicity associated with venetoclax.[18].
7. Regulatory Affairs
In April 2016, venetoclax (Venclexta/Venclyxto, AbbVie/Genentech) in monotherapy was approved by the US Food and Drug Administration (FDA) for the treatment of R/R CLL with del17p. In June 2018, based on the results of the phase 3 MURANO clinical trial, the FDA expanded venetoclax approval in combination with RTX for CLL patients progressing after at least one previous treatment, regardless of the 17p status. The EMA expanded the approval for the treatment of CLL patients with del17p or TP53 mutation who are unsuitable for, or have failed, a BCR pathway inhibitor, as well as for patients who have failed both chemoimmunotherapy and BCR pathway inhibitors, regardless of their genetic/molecular profile[19].In monotherapy venetoclax can be administered until progression or unacceptable toxicity. In October 2018, based on the findings from the phase 3 MURANO trial, the EMA approved venetoclax plus RTX with a 24-month fixed duration for the treatment of patients with R/R CLL, following at least 1 prior therapy. The drug is given in combination with RTX for 6 cycles and then in monotherapy to complete the 24-month treatment regimen [23,24,57].It is important to note that venetoclax shows high efficacy not only in CLL, but also in the treatment of other hematological disorders and solid tumors. In November 2018 VEN in combination with azacitidine or decitabine or low-dose cytarabine was approved by the FDA for the treatment of newly-diagnosed acute myeloid leukemia (AML) in 75-year-old patients or older, or who have comorbidities that preclude the use of intensive induction chemotherapy[57].
8. Conclusions
Venetoclax is active against lymphoproliferative disorders, particularly CLL, with an unique inhibitory activity towards the anti-apoptotic BCL2 protein. It belongs to a class of agents known as small molecules, which are bioavailable orally and suitable for long-term treatment. Importantly, it displays activity in both frontline and R/R patients, including those with del(17p)/ TP53 mutation. A very important feature is its efficacy after the failure of BCR pathway inhibitors, so that it may offer hope for patients whose ibrutinib treatment was interrupted. The combination of venetoclax with rituximab show higher two-year PFS, as well as higher MRD negativity rates than the treatment with bendamustine plus rituximab.The adverse events,especially the hematological and gastrointestinal ones, are usually manageable.The efficacy and safety of venetoclax combined with anti-CD20 antibodies, BCR inhibitors and classical chemotherapeutics is currently understudy.
9.Expert Opinion
The management of patients with CLL represents a challenge, especially in the elderly, or patients with clinically important comorbidities, or with aberrations of TP53. Purine analogue-based chemoimmunotherapy remains a gold standard for younger patients in good general condition and allows a high percentage of complete remissions, but relapses are frequent. Moreover, it does not overcome the poor prognostic significance of TP53 aberrations and is unsuitable for older and/or physically frail patients, who constitute a great part of treatment-requiring CLL patients, if not the majority. Therefore there is a need to elaborate not only efficacious therapeutic strategies but also those with an acceptable safety profile which are suitable for elderly/physically-frail patients and which may be active in cases with TP53 aberrations.Small molecules make up a new category of therapeutics used in CLL with relatively selective activity against different intracellular pathways involved in the pathogenesis of CLL. The BCR-signaling pathway has been explored particularly thoroughly for this purpose and ibrutinibis already widely used in a large setting of indications, including patients who are physically less fit or those with TP53 aberrations. However, a bad prognosis inpatients in whomibrutinib had to be discontinued is a matter of concern.
Venetoclax is a very promising small molecule targeted against the anti-apoptotic BCL-2 protein, the overexpression of which is one of the molecular hallmarks of CLL. It shows efficacy both in first-line treatment and relapsed/refractory patients, with a high percentage of overall responses and complete remissions, as well as a long PFS.It is noteworthy that its application allowed the eradication of MRD in a number of patients, which is postulated to have an impact on overall survival. A very important finding is its efficacy in patients with aberrations of TP53; this is particularly significant,as these aberrations, which confer resistance to purine analogs and alkylating agents, are infrequent at diagnosis but their incidence increases during clonal evolution of the disease. Another precious property of the drug is its efficacy after BCR pathway inhibitor failure,i.e. in the group of patients with the most frustrating prognosis. In particular, venetoclax probably works better than idelalisib in the patients failing on ibrutinib, but this remains to be formally proven.The drug seems then to meet the needs listed at the beginning of this section. Since CLL is a disease of complex pathogenicity involving the deregulation of various pathways, it is then legitimate to expect that venetoclax will be more active when given in association with a drug acting on another pathway.
Of the other regimens used against CLL, the anti-CD20 antibody seems the most natural partner for venetoclax; its efficacy has been confirmed in many of the above studies, especially in terms of the frequency and depth of response. Venetoclax in combination with other cytotoxic drugs seems to offer the advantage of a fixed- term therapeutic approach, mimicking in this respect chemoimmunotherapy. Based on the results of the phase 3 MURANO study venetoclax plus rituximab, with a 24-month fixed duration, has been recently approved for the treatment of patients with R/R CLL, following at least one prior therapy. It is important to note that the two-year PFS as well as MRD negativity rates were higher after venetoclax plus rituximab treatment regimen than after bendamustine plus rituximab therapy, both for patients with del17p and for those without this aberration. Specified treatment time (fixed time duration) is a significant change compared to the existing monotherapy regimen, in which the treatment could be continued until progression or intolerance of the therapy. The treatment with VEN both in monotherapy and in combination with other agents, especially with RTX in R/R CLL patients is relatively safe. Gradual dose escalation of the drug appears to minimize the risk of the TLS. The most important adverse events in monotherapy and in combination treatment, especially myelotoxicity and gastrointestinal symptoms, are usually manageable and do not lead to the interruption of the therapy. [
Promising results have also been obtained when venetoclax was associated with BCR pathway inhibitors thought to increase the dependence of the survival of CLL cells on the BCL-2 system, thus overcoming venetoclax resistance. Currently, trials intended to choose the best combination of venetoclax with other agents active in CLL are ongoing. In our opinion, the most important asset offered by venetoclax for current clinical practice is that it displays activity for patients with aberrations of TP53 after the failure of BCR pathway inhibitors. In addition, it also offers potential as frontline treatment for patients with those aberrations, and may also be used in future therapy for frail patients, independent of TP53 status, i.e. beyond current approval. Therefore, it is likely that similarly to ibrutinib, venetoclax will revolutionize the treatment of CLL. However, before a full consensus concerning its place in this setting is reached, there is a need for studies aimed at comparing venetoclax with other small molecules and anti-CD20 antibodies, both in monotherapy and in combination.
