Proteasome Inhibitors in Waldenstro¨m Macroglobulinemia
INTRODUCTION
Waldenstro¨ m macroglobulinemia (WM) is an incurable B-cell lymphoproliferative dis- order that is characterized by the infiltration of the bone marrow by clonal lymphoplas- macytic cells and the production of monoclonal immunoglobulin M (IgM) by these cells.1–4 WM may have a long course, even in symptomatic patients, and during the course of the disease, multiple regimens may be used to control the disease and its symptoms.2,5,6 Currently, therapy is considered only for patients with symptomatic disease, and primary options include combinations based on anti-CD20 monoclonal antibodies, mainly rituximab.
In recent years, proteasome inhibitors (PIs) have become a mainstay of therapy in plasma cell malignancies but also in some specific lymphomas.7 Regarding the treat- ment landscape in WM, PIs have also become part of primary and salvage options for patients with WM,2,5,6 based on the results of several phase 2 studies, mostly with bortezomib, the first in the class of PI. In addition, new PIs have become available and may also find their way into WM therapy, such as carfilzomib, ixazomib, and oprozomib.In this review, the authors focus on the clinical results of PI-based therapy and the challenges phased in the changing landscape of available therapies for WM.
MECHANISMS OF ACTION
PIs block degradation of ubiquitinated proteins by the proteasome, an organelle found in all cells.8,9 Blocking proteasome activity results in the accumulation of ubiq- uitinated proteins and leads to dysregulation of multiple pathways within the cells but also to an increase of endoplasmatic reticulum (ER) stress. Cells that are sensitive cannot cope with the increased ER stress load, and this leads to the activation of apoptotic pathways. Although this schema may be oversimplified, cells that are most sensitive seem to be those that are producing higher amounts of protein, such as plasma cells and other B cells.10 Different PIs have differences in their affinity for the various proteasome subunits, reversibility or nonreversibility of interaction, and pharmacokinetics. Bortezomib is the first in the class of PI and is a slowly revers- ible boronated inhibitor of the 26S proteasome, mostly of the chymotryptic unit of the proteasome.11 Carfilzomib is an irreversible tetrapeptide epoxyketone–based PI (analogue of epoxomicin).12 Ixazomib is an orally available, boronated, PI, which is metabolized to its active form,13,14 whereas oprozomib is an oral analogue of carfilzomib.
Preclinical studies have elucidated multiple mechanisms of action for PIs in WM, and bortezomib is the PI mostly studied.16,17 Bortezomib blockade of the ubiquitin- proteasome degradation pathway affects signaling pathways that also include NF-kB,16 which has critical function in WM cells’ survival and immunoglobulin pro- duction. The induction of ER stress has also been implicated as a mechanism for bortezomib activity leading to disruption of the unfolded protein response that prompts WM cell apoptosis,17,18 which is active in WM cell lines and in primary tu- mor cells. PIs may also impact the supportive bone marrow microenvironment in WM as has also been implicated for its activity in multiple myeloma.17,18 Bortezomib has also demonstrated synergistic and/or additive preclinical activity in combination with other agents, including steroids, rituximab, and signaling inhibitors in WM cells.
CLINICAL DATA ON THE ACTIVITY OF PROTEASOME INHIBITORS
PIs have undergone extensive clinical investigation in WM, either as a single agent or as part of combination therapies (Table 1). The first PI studied in the clinic was borte- zomib, initially in patients with relapsed or refractory WM, and was given as a single agent and through an intravenous (IV) route.
Dimopoulos and colleagues21 investigated the activity of single-agent bortezomib (1.3 mg/m2 as IV push on days 1, 4, 8, and 11 of a 21-day cycle for up to 6 cycles) in 10 previously treated patients. A major response, that is, at least a partial response (PR), was observed in 6/10 patients (60%), whereas 2 (20%) additional patients attained a minor response for an overall response rate of 80%. The median time to a major response was 16 weeks, and the time to progression (TTP) was expected to exceed 11 months for the responders.
In a multicenter study by the Waldenstro¨ m’s Macroglobulinemia Clinical Trials Group, patients with previously treated WM received single-agent bortezomib (1.3 mg/m2 IV push on days 1, 4, 8, and 11 of a 21-day cycle).22 Twenty-seven patients received up to 8 cycles of bortezomib, and the overall and major response rates were 85% and 48%, respectively. Importantly, responses were rapid in this trial, with a me- dian time to first and best response of 1.4 months and of 4.1 months, respectively. The median TTP for all patients was 7 months, and there was a trend for a longer median TTP in patients achieving a major response (9 months).
In a similar trial conducted by the National Cancer Institute of Canada, single-agent bortezomib was given (IV push of 1.3 mg/m2 with the abovementioned schedule).23 The trial enrolled 27 untreated or previously treated patients for an overall and major response rate of 78% and 44%, respectively. Again, the responses were rapid, with a median time to first response of 1.5 months, and the median progression-free sur- vival (PFS) for all patients was 16 months. However, in this study, nodal responses were also assessed, and it was observed that these were slower than IgM responses, although occurred in most patients with a median time to response of 12 weeks. Such discordance of nodal to IgM responses has also been observed in other trials with single-agent bortezomib, including discordance between serum IgM and bone marrow burden reductions, but the clinical significance of this phenomenon is not clear.22,24
In a recently published prospective study from France, a different approach to bor- tezomib therapy was used, in which therapy started with bortezomib alone and then dexamethasone was added if the response was inadequate.25 In this phase 2 trial, 34 patients with relapsed/refractory WM were enrolled. Bortezomib was given at 1.3 mg/m2 IV on days 1, 4, 8, and 11 every 21 days for 6 cycles. In nonresponding pa- tients, dexamethasone (20 mg daily for 2 days) was added to each infusion after the second cycle. Using a Bayesian statistical approach, after 2 cycles, the Bayes esti- mated overall response rate was 43.2 (with 95% credible interval 28.0%–59.1%) using the “informative prior”. The 2-year survival rate was 84.0%, and the median PFS was 15.3 months without a difference between patients treated with or without dexameth- asone. Based on this approach, the investigators concluded that dexamethasone can improve the efficacy of bortezomib and should be associated with bortezomib-based regimens.
Based on the above data, it is clear that bortezomib alone is an active agent in WM, whereas because of its mechanism of action, the available preclinical data, and also based on the extensive experience from myeloma, it is an ideal drug to combine with other WM-acting drugs. Thus, further development of bortezomib in WM followed the pathway of combinations with rituximab.
COMBINATION THERAPIES THAT INCLUDE PROTEASOME INHIBITORS
Bortezomib Combinations
To improve depth of response as well as PFS, combination therapies with bortezomib have been extensively evaluated.The combination of bortezomib, dexamethasone, and rituximab (BDR) was investi- gated by the WMCTG as a primary therapy in 23 WM patients.26 Patients received bor- tezomib IV at a dose of 1.3 mg/m2 along with dexamethasone 40 mg on days 1, 4, 8, and 11, whereas rituximab was given on day 11 of each 21-day cycle. Four cycles of induction therapy were given. Twelve weeks after induction therapy completion, one cycle of therapy was given every 12 weeks for a total of 4 cycles as a maintenance therapy. The overall response rate was 96%, with 83% of patients achieving a major response. The median time to response was 1.1 months. An “IgM flare” was observed in only 9% of patients; however, those with high levels of IgM received preemptive plasmapheresis. After a long follow-up of the study, the median TTP was 52 months with this regimen.27 However, bortezomib-related neuropathy was common and led to bortezomib discontinuation in 60% of patients, although with the longer follow-up a near complete resolution or partial resolution of treatment-related neuropathy to grade 1 was observed in most patients. Other toxicities associated with combination included dexamethasone-related hyperglycemia, myopathy, and gastritis that also prompted omission or dose reduction of steroids in many patients. Furthermore, the addition of steroids to bortezomib was deemed responsible for herpes zoster that occurred in 4 of the first 7 patients entered on this trial, and who did not receive herpes zoster prophylaxis, leading to institution of prophylaxis while on active therapy plus 6 additional months following end of therapy.
In an effort to ameliorate steroid-related toxicity and also to reduce bortezomib- associated neuropathy, Ghobrial and colleagues30,31 and Agathocleous and col- leagues28 examined bortezomib in combination with rituximab (VR) in untreated as well as previously treated patients with WM, using a different schedule.
In the Agathocleous and colleagues28 trial, VR was given in previously treated WM patients, in a study that also included other lymphoma histologies, and randomized between weekly and twice weekly IV bortezomib along with rituximab. Nine of 10 (90%) WM patients in this study had a major response, and 4 of these patients remained free of progression after 2 years. However, the risk of neuropathy was not impacted by treatment arm (weekly vs biweekly bortezomib administration), although only a small number of WM patients were included in each arm.
In the 2 trials led by Ghobrial and colleagues,30,31 bortezomib was administered IV, as in the previous ones, but was given weekly at a dose of 1.6 mg/m2 on days 1, 8, 15, every 28 days (for up to 6 cycles). Rituximab was given at the standard 375 mg/m2 dose weekly during cycles 1 and 4, following an “extended rituximab schedule.”29 The overall response rate for the 26 untreated patients who received VR was 88%, with a major response of 65%.30 The median TTP at the time of reporting the study was not reached, with an estimated 1-year event-free rate of 79%. Common grade 3 and 4 therapy-related adverse events included reversible neutropenia in 12%, ane- mia in 8%, and thrombocytopenia in 8% of patients. Regarding treatment-related neu- ropathy, this was observed in 15% of patients at grade 2, and in none of the patients at the grade 3 or 4 level. Among the 37 previously treated patients who received VR, the overall response rate was 81%, including 51% of patients who achieved a major response.31 The median TTP in this study was 16.4 months. The most common grade 3 and 4 therapy-related adverse events included reversible neutropenia in 16%, ane- mia in 11%, and thrombocytopenia in 14%. Again, mild peripheral neuropathy (PN) grade 1 to 2 was 41% and grade 3 treatment-related PN was low and occurred in 2 patients (5%). In the above studies with VR, rituximab-related IgM flare phenomenon occurred in 20% to 30%, indicating that a different strategy should be used to amelio- rate this complication by exploiting bortezomib’s activity.
In a prospective, multicenter clinical trial conducted by the European Myeloma Network, a novel schedule of administration for BDR was evaluated in 59 untreated WM patients, most of which were of advanced age and with adverse prognostic fac- tors.32 Patients received twice weekly IV bortezomib (1.3 mg/m2 days 1, 4, 8, and 11 of a single 21-day cycle) for the first cycle, in order to rapidly reduce IgM and avoid IgM flare when rituximab was given. Afterward, they received weekly bortezomib (at a dose of 1.6 mg/m2 on days 1, 8, 15, and 22 of 35-day cycles) for cycles 2 to 5, along with dexa- methasone (40 mg). Rituximab (at a standard dose of 375 mg/m2) was given weekly in cycles 2 and 5. However, and unlike the WMCTG trial, no maintenance therapy was given after induction completion. An overall response rate of 85% was attained, including a major response in 68% of patients. The median PFS in this study was 43 months, and patients with very good partial response (VGPR) or better had significantly longer PFS. Treatment-related PN occurred at grade 2 in 17% and was grade 3 in 7% of patients. Importantly, rituximab-related IgM flare was observed in 11% of patients, but there was no need for plasmapheresis. An update of this study after a minimum follow-up of 6 years showed that this alkylator-free regimen was associated with a median PFS of 43 months, whereas the median duration of response for patients with at least PR was 64.5 months and the overall survival at 7 years was 66%. Importantly, no patient had developed secondary myelodysplasia, whereas transformation to high-grade lym- phoma occurred in 3 patients who had received chemoimmunotherapy after BDR.
The combination of the mTOR inhibitor everolimus with bortezomib and rituximab (RVR) has also been examined in a prospective trial in patients with previously treated WM.34 Thirty-six patients received 6 cycles of RVR. Each 28-day cycle consisted of everolimus 10 mg by mouth daily, bortezomib 1.6 mg/m2 given IV weekly on days 1, 8, 15, and for cycles 1 and 4 only, rituximab 375 mg/m2 was given IV on days 1, 8, 15, and 22. Following 6 cycles of therapy, patients continued on everolimus alone as maintenance therapy until progression. Most patients on RVR had received prior rituximab and/or bortezomib. An overall response rate of 89%, with a major response rate of 53%, was observed in this study. The median PFS was 21 months. No grade 3 PN was observed in this study, although cytopenias were commonly observed that included grade 3 anemia (30%), neutropenia (13%), and thrombocytopenia (13%). However, bortezomib has been clearly associated with certain toxicities, the most prominent of which is neurotoxicity.21–23 Bortezomib-associated neuropathy is challenging to manage if unnoticeable and left without appropriate action.35 Actually, bortezomib-related neuropathy is the most common reason for dose reductions and early discontinuation of therapy,26,32 limiting the duration of treatment with this agent. The use of subcutaneous instead of IV bortezomib has been associated with similar efficacy and less neuropathy in patients with myeloma,36,37 but no such direct com- parisons exist in WM. Early dose adjustments, when neuropathy is still mild, and recognition of patients at risk (such as diabetics, those with prior exposure to neuro- toxic drugs) are all important. Weekly versus twice-per-week dosing is also associated with lower risk of neuropathy and probably with better relative dose intensity of ther- apy with bortezomib. Infectious complications are not uncommon, and a risk of herpes zoster reactivation is high if no prophylaxis is given.26 Gastrointestinal toxicity (constipation or diarrhea), conjunctivitis, and lung toxicity have also been reported with various but rather low frequencies. Generally, bortezomib has not been consid- ered cardiotoxic, although some data may indicate some association with arrhythmias and left ventricular ejection fraction reduction.
OTHER PROTEASOME INHIBITORS
Carfilzomib
Carfilzomib is a second-generation PI, and extensive experience in myeloma patients has shown that it is associated with low neuropathy risk, although it has been associ- ated with a potential of cardiotoxicity.WM was evaluated in combination with rituximab and dexamethasone (CaRD) in WM patients that were naive to both bortezomib and rituximab.39 CaRD therapy con- sisted of IV carfilzomib given at 20 mg/m2 (on cycle 1), and on 36 mg/m2 thereafter (cycles 2–6), on days 1, 2 and 8, 9 with dexamethasone 20 mg on days 1, 2, 8, 9 and rituximab 375 mg/m2 on days 2, 9 every 21 days. Maintenance therapy followed 8 weeks after cycle 6 with IV carfilzomib 36 mg/m2 and dexamethasone 20 mg on days 1, 2 and rituximab 375 mg/m2 on day 2 every 8 weeks for 8 cycles. The overall response rate in this study was 87%, and 68% of the patients achieved a major response. In this study, MYD88 and CXCR4 tumor mutational status were examined and did not appear to impact response attainment, although the numbers are small. With a median follow-up of 15.4 months, 20 patients remained progression-free at the time of reporting. Grade 2 PN occurred in only 1 patient with underlying disease-related PN, and no grade 3 or higher treatment-related neuropathy events were recorded. Other grade 2 toxicities included asymptomatic hyperlipasemia (41.9%), reversible neutropenia (12.9%), and cardiomyopathy in one patient (3.2%) with multiple risk factors. Declines in serum levels of IgA and IgG were common and contributed to recurring sinobronchial infections and IV immunoglobulin use in a few patients.
In a recent case series report, Vesole and colleagues40 reported a single-center experience of carfilzomib treatment in relapsed WM treated in a phase 1b/phase 2 program. Patients received carfilzomib at 56 or 70 mg/m2 after 20 mg/m2 on the first 2 doses (day 1 and 2 of the first cycle), whereas dexamethasone 8 mg was adminis- tered on each day of carfilzomib therapy during the first cycle and was optional for subsequent cycles. If patients achieved less than a PR after 4 cycles, then rituximab 375 mg/m2 was added on day 16 of each cycle, and the carfilzomib dose was decreased to 27 mg/m2. Patients were treated to maximal response plus 2 additional cycles (for a maximum 12 cycles). Seven patients received carfilzomib, and no patient received rituximab. Among the 7 patients, 4 received carfilzomib at 70 mg/m2 and 3 received carfilzomib at 56 mg/m2. No dose-limiting toxicities occurred, although 6 patients reported at least 1 grade 3 adverse event and one patient discontinued treatment due to an adverse event. Two patients had neuropathy events (grade 3), whose relationship to study drug was considered probable. Six patients had a prior exposure to bortezomib, and 2 were refractory. All patients achieved a minor response (MR) or better, including 1 stringent complete response (sCR) (overall 1 sCR, 3 PRs, 2 VGPRs, and 1 MR). The 2 patients that were bortezomib refractory achieved a PR. In this small series of patients, 6/7 patients did not have a PD at a follow-up time of 13 to 27 months.
Oral Proteasome Inhibitors
Ixazomib is an oral boronated PI that has now been approved for use in patients with myeloma in combination with lenalidomide and dexamethasone,41 whereas an exten- sive clinical program is running and several clinical studies recruit patients in different clinical settings.
Castillo and colleagues42 recently presented the results of a prospective phase 2 study in which oral ixazomib was given in combination with dexamethasone and ritux- imab (IDR) in 26 symptomatic, previously untreated patients with WM. Ixazomib was given at 4 mg with dexamethasone at 20 mg, on days 1, 8, and 15 every 28 days for induction cycles 1 and 2 (as an induction), and then in combination with rituximab 375 mg/m2 IV, on day 1 of cycles 3 to 6. Maintenance therapy followed 8 weeks later with IDR given every 8 weeks for 6 cycles. All patients were evaluated for MYD88 and CXCR4 mutational status and were MYD88 L265P mutated, whereas 15 patients (58%) had a CXCR4 mutation. The median time to response was 8 weeks, which was longer however (at 12 weeks) in WM patients with CXCR4 mutations (log-rank P 5 .03). The overall response rate was 96% (including VGPR 15%, PR 62%, MR 19%) with a major response rate (VGPR 1 PR) of 77%, but it was not impacted by CXCR4 mutations. With a median follow-up of 18 months, the median PFS was not reached, and the 18-month PFS rate was 90% (95% confidence interval 65%– 97%). Grade 2 adverse events included infusion-related reactions (19%), rash (8%), and insomnia (8%). However, ixazomib was not associated with clinically signif- icant risk of neuropathy.
Oprozomib is an oral epoxyketone proteasome inhibitor that is an analogue of carfil- zomib. The clinical development of oprozomib has been delayed because of gastroin- testinal toxicity issues. The schedule of administration of oprozomib has not been fully optimized, and 2 different schedules have been evaluated in phase 1 and 2 studies. Two schedules of oprozomib given once daily either on days 1 to 5 (5/14 schedule) or on days 1, 2, 8, and 9 (2/7 schedule) of a 14-day treatment cycle have been evaluated. An overall response rate of 59% with a major response rate of 29% was observed in a phase 2 study in 17 previously treated WM patients treated with single-agent oprozo- mib.43 However, gastrointestinal intolerance was prominent with oprozomib necessi- tating intensive antiemetic use. Although the activity of the drug seems to be high, it is still very early for oprozomib to be placed in the armamentarium of available PIs for WM.
WHERE DO PROTEASOME INHIBITORS FIT IN THE CURRENT TREATMENT RECOMMENDATIONS?
PIs have prominent activity in WM, in all disease settings, either as single agents or in combinations with other active drugs. The major advantages of PIs are that they are fast acting, reduce IgM levels quite rapidly, are non-myelotoxic, are not associated with a risk of secondary malignancies or myelodysplasia, and can be easily combined because they have nonoverlapping toxicities with most other commonly used drugs. The effect of MYD88 and CXCR4 mutational status on PI activity is unclear and less prominent than on ibrutinib efficacy. Bortezomib is the main PIs for which there is a significant amount of data and long-term experience, whereas carfilzomib use has less published data. At the moment, a randomized phase 3 study evaluates the role of addition of bortezomib to standard DRC regimen (dexamethasone, rituximab, and cyclophosphamide) in previously untreated symptomatic patients with WM (NCT01788020). The results of this study will elucidate the role of adding a PI at the frontline over a standard effective WM regimen. Ixazomib is the first oral PI that has been approved for use in myeloma patients. Also, there are data from a phase 2 study in newly diagnosed WM, whereas another ongoing phase 2 study evaluates ixazomib- rituximab combination in relapsed or refractory WM. The toxicities of PIs are different with bortezomib being associated mainly with neurotoxicity, carfilzomib with a risk of cardiotoxicity, whereas ixazomib is less neurotoxic but may have some gastrointes- tinal toxicity.
Bortezomib-containing combinations can be considered an option for the primary therapy for patients with high levels and in need for rapid reduction of IgM, such as those with symptomatic hyperviscosity or at risk for clinically significant IgM flare.5 Bortezomib induction followed by bortezomib-rituximab has been used to avoid IgM flare.32 When myelotoxicity is undesirable, bortezomib with rituximab may also be a primary option.2,5,6,44 Carfilzomib is not widely available for WM but could be consid- ered a neuropathy-sparing PI. Ixazomib may also be a good alternative if a PI is needed, but still data are limited and are not available for WM outside clinical trials.
However, the major change in the therapy for MW has been the introduction of ibru- tinib and other BTK inhibitors.45,46 Ibrutinib has shown substantial activity and is rec- ommended at least for patients who relapse or have refractory disease, especially those with rituximab-refractory disease.5,44 There are still limited data on the manage- ment and outcomes of those patients failing ibrutinib, but it seems that PIs could be a good option for patients who fail ibrutinib therapy, as shown by a recent report of pa- tients who failed or discontinued ibrutinib due to toxicity.
In summary, PIs have shown significant activity in WM; the high rates of efficacy, particularly in combination therapy, as well as their overall tolerability, have resulted in their adoption as important mainstays of WM therapy. The development of strate- gies to mitigate neuropathy risk of PIs, including use of weekly subcutaneous admin- istration for bortezomib and exploration of novel neuropathy-sparing agents such carfilzomib and ixazomib will invariably lead to further advances in the use of this class of agents in the treatment of WM. Even in the era of BTK inhibitors, PIs will remain a major option for newly diagnosed and relapsed or refractory WM patients.