Safety and efficacy of subcutaneous formulation of bortezomib versus the conventional intravenous formulation in multiple myeloma

Abstract

The discovery of the ubiquitin–proteasome pathway first, and the proteasome inhibitors thereafter were not made in the hope of improving the treatment of malignant diseases. However, bortezomib, the first in class proteasome inhibitor introduced in the clinical practice has contributed to improve the outcome of patients with multiple myeloma, at relapse or disease progression as well as upfront. The results observed in a large randomized trial (APEX) comparing bortezomib and high-dose dexamethasone demonstrated a significant benefit for bortezomib in terms of response rate, progression-free and overall survival. These results led to bortezomib being approved for use in relapsed and/or refractory myeloma patients. Subsequent studies demonstrated that its activity could be enhanced in combination with other drugs; and the next step was to move to the newly diagnosed patient population; in fact, bortezomib–melphalan–prednisone (VMP) is approved as a standard of care for newly diagnosed elderly patients. However, toxicity, especially peripheral neuropathy, as well as the intravenous route required for its administration are the two most significant bortezomib-related issues. To try to reduce the peripheral neuropathy, new guidelines for its management and the introduction of weekly schedules of administration have contributed to significantly decrease its incidence and the subcutaneous administration has been recently introduce to avoid the intravenous (IV) route. Results obtained in phase I/II and III studies have confirmed that subcutaneous administration is feasible and represents an additional step towards the optimization of bortezomib use, resulting in a probably more convenient method than the IV route that is at least as effective.

Introduction

The outcome of multiple myeloma patients (MM) has significantly improved over recent years, mainly due to the discovery of novel antimyeloma agents together with a better knowledge of the biology of the disease [Kumar et al. 2008]. One of these novel agents is bortezomib, the first in the class proteasome inhibitors introduced in the clinical practice approximately one decade ago.

Bortezomib represents an excellent example of a novel drug that has quickly moved from the bench to the bedside and now, after a decade of experience in clinical practice, the investigation continues in order to optimize its use not only in terms of efficacy but safety and tolerability as well.

However, it is interesting to note that the original idea to synthesize proteasome inhibitors for possible therapeutic applications was not made in the hope of progressing the treatment of malignant diseases, but to clarify the mechanisms of muscle atrophy occurring after denervation. The decision for its use as a therapeutic agent was not directly related to MM either. However, one of the most interesting pieces of evidence was that bortezomib could reduce the protein degradation and these findings were closely related with the discovery of the ubiquitin–proteasome pathway, in which protein substrates are linked to ubiquitin chains, the 26S proteasome [Mitch and Goldberg, 1996]. Finally, other findings, such as the knowledge of its role in the activation of NF-κB suggested anti-inflammatory and antineoplastic actions and led to clinical trials against various cancers [Palombella et al. 1994].

Bortezomib is a boronic acid dipeptide that directly binds to enzymatic complex to block chimotrypsin-like activity of proteasome. It blocks degradation of multiproteins, including regulators of the cell cycle, anti-apoptosis, inflammation, as well as immune surveillance [Adams et al. 1999]. In MM cells, bortezomib directly induces a cell stress response followed by activation of several proteins inducing cell cycle arrest, followed by caspase-dependent apoptosis. In addition, it also modulates the bone marrow environment, including stromal cells and osteoblasts [Hideshima et al. 2003].

In the first phase I trial testing bortezomib against all cancers, surprisingly, one patient showed a dramatic, complete remission (CR). That patient had late-stage MM [Orlowski et al. 2002]. Some additional responses were observed in other patients with myeloma and, consequently, phase II trials focused on myeloma patients were activated, for which there was no adequate therapy.

The results obtained in the different trials provided further support for proteasome inhibition as a valid rational therapeutic approach and bortezomib (Velcade®) was approved by the Food and Drug Administration in 2003. This initial approval, which was for MM patients who had received at least two prior therapies and whose disease had demonstrated progression on the last of these, also made bortezomib the first new agent approved for myeloma in over a decade. Since that time, further studies of bortezomib, both alone and as part of different combinations, have shown the versatility of this agent and made it a universally accepted backbone for the treatment of myeloma patients.

In this article, we review the most significant results obtained with bortezomib in the treatment of myeloma patients, first through intravenous (IV) administration followed by subcutaneous administration.

Intravenous formulation: efficacy and safety results

The initial phase I trial with bortezomib as monotherapy in patients with relapsed and/or refractory hematologic malignancies showed a clinical benefit in the nine patients included in the trial with plasma cell dyscrasias, including one with a durable CR, who also was the first myeloma patient ever treated with bortezomib [Orlowski et al. 2002]. This activity against relapsed/refractory myeloma was then further evaluated in two multicenter phase II trials, the Study of Uncontrolled Multiple Myeloma Managed with Proteasome Inhibition Therapy (SUMMIT) and the Clinical Response and Efficacy Study of Bortezomib in the Treatment of Relapsing Multiple Myeloma (CREST) [Jagannath et al. 2004; Richardson et al. 2003]. Both administered bortezomib as an IV push on days 1, 4, 8 and 11 of every 3-week cycle, which continues to be until today the schedule most commonly used. Patients on the SUMMIT trial received an initial dose of 1.3 mg/m², while the smaller CREST study also explored a lower dose of 1.0 mg/m². On the SUMMIT trial, where most of the patients included had disease refractory to the last line of treatment, a partial response (PR) or better was seen in 27% of the 193 evaluable patients. In addition, 10% of patients achieved CR or near-CR, with a median time to progression (TTP) of 7 months, approximately twice as many months as compared with their previous line of therapy [Richardson et al. 2003]. The patients included in the CREST trial had relapsed or refractory disease after frontline therapy and received bortezomib either alone at 1.3 mg/m², or with the addition of dexamethasone and the response rate was 50%; on the other hand, the cohort of patients who received bortezomib at 1.0 mg/m² achieved an overall response rate of 38% and, considering these results, most of the following trials have used bortezomib as an IV push at 1.3 mg/m². However, it is interesting to note that the efficacy observed with 1.0 mg/m² was balanced with a lower likelihood of developing some adverse events and this finding has been considered on the management of bortezomib for dose reduction schedules [Jagannath et al. 2004].

The data observed in the SUMMIT trial formed the rationale for an accelerated approval of bortezomib for relapsed/refractory myeloma patients, and led to a randomized phase III trial, the Assessment of Proteasome Inhibition for Extending Remissions, or APEX study [Richardson et al. 2005]. This trial included patients who had relapsed after no more than three prior lines of therapy, and were randomized to receive bortezomib as monotherapy at 1.3 mg/m² as an IV push on days 1, 4, 8 and 11 followed by a 10-day rest period or dexamethasone at high dose as monotherapy. At the first report of the results, a significant benefit for bortezomib arm was already observed, with a ≥PR rate of 38%, including 9% CR, compared with 18% and less than 1% for dexamethasone arm. Continued therapy led to an improvement in the responses rate on bortezomib arm up to 43%, while no significant benefit was observed in patients on the dexamethasone arm. The median TTP was 6.22 months with bortezomib and 3.49 months with dexamethasone, and this benefit was also translated to overall survival, with a median of 29.8 and 23.7 months for bortezomib and dexamethasone, respectively [Richardson et al. 2007]. These data supported full approval of bortezomib for patients with relapsed and/or refractory myeloma who had received at least one prior therapy and it was registered as an antineoplastic agent for IV use only at a dose of 1.3 mg/m² given as a 3–5 seconds bolus IV injection via a peripheral or central IV catheter, followed by a standard saline flush; in addition, it is indicated to maintain at least a 72-hour rest period between doses in order to allow a restoration of the proteasome function towards baseline.

These three trials clearly support the notion that the modulation of the proteasome function was an attractive therapeutic option; moreover, preclinical studies showed that bortezomib could enhance the sensitivity to other agents and, in many cases, even overcome the drug resistance [Rajkumar et al. 2005]. The first preclinical data showed a synergistic and/or additive effect between bortezomib and corticosteroids and, accordingly, in the SUMMIT and CREST studies patients with suboptimal response to bortezomib as a single agent received dexamethasone at 20 mg on the day on and after of each dose of bortezomib. With this approach, improvements in the quality of response were seen in up to one third of these patients, and other trials have shown that with the addition of corticosteroids, response rates improve to 60% or more, without increases in toxicity [Jagannath et al. 2004; Richardson et al. 2003].

The second step was to combine bortezomib with other agents, such as pegylated liposomal doxorubicin, and a phase III randomized trial showed that this combination was able to improve the median TTP as compared with bortezomib alone by approximately 3 months (9.3 versus 6.5 months) and based on these results, bortezomib plus pegylated doxorubicin was approved for relapsed and/or refractory myeloma patients [Orlowski et al. 2007]. Bortezomib has been also combined with alkylating agents, including cyclophosphamide, melphalan and bendamustine. Melphalan-based combinations with bortezomib have been widely used, ranging from the doublet to four-drug programs, such as bortezomib, melphalan, prednisone and thalidomide [Mitsiades et al. 2003]. The next step was to combine bortezomib with the immunomodulator drugs, thalidomide and lenalidomide. These combinations have resulted in an overall response rate of up to 60–70% and, notably, they have shown activity even in patients who had previously relapsed or progressed through bortezomib plus dexamethasone, or lenalidomide plus dexamethasone [Jakubowiak et al. 2011; Richardson et al. 2009, 2010].

Based on the positive results obtained with bortezomib in relapsed and/or refractory myeloma patients, several groups moved to use it upfront, both in young patients as candidates for autologous stem-cell transplantation and in elderly patients. Four randomized trials have evaluated the role of bortezomib-based combinations as induction therapy in transplant candidate myeloma patients, revealing a high efficacy (>80% response rate, with 20–30% CR) that increased after autologous stem-cell transplantation, confirming the results of numerous previous pilot studies with bortezomib-based combinations [Cavo et al. 2010; Harousseau et al. 2010; Rosinol et al. 2010; Sonneveld et al. 2010]. In patients who are not candidates to autologous stem-cell transplantation, bortezomib in combination with melphalan and prednisone has also proved to be superior to conventional therapy, with high overall and CR rates (81% and 30%, respectively), and a significantly longer TTP (24 months) and overall survival (60% at 3 years) as compared with conventional schemes. This combination, bortezomib plus melphalan and prednisone, was the last approval obtained for bortezomib in this case for untreated MM patients not eligible for stem-cell transplantation [San Miguel et al. 2008].

As far as tolerability is concerned, all of these trials contributed also to establishing the pattern of adverse events (AEs) of bortezomib and their management. In the SUMMIT trial, the most significant AEs reported were thrombocytopenia, fatigue, peripheral neuropathy and neutropenia [Richardson et al. 2003]. It should be noted that in the CREST trial, the cohort of patients receiving bortezomib at a dose of 1 mg/m² experienced a lower likelihood of developing some adverse events such as diarrhea, vomiting and neuropathy [Jagannath et al. 2004]. The APEX trial, including more than 600 patients, was optimal in defining the toxicity profile and the most significant side effects of all grades were diarrhea (57%), nausea (57%), fatigue (42%), constipation (42%), neuropathy (36%), vomiting (35%), anorexia (35%) and thrombocytopenia (35%). Although the majority of these AEs were grade 1 or 2, thrombocytopenia, gastrointestinal toxicity and peripheral neuropathy have been more extensively studied because they are probably the most significant, especially peripheral neuropathy. Thus, the frequency of grade 3–4 thrombocytopenia and gastrointestinal symptoms are approximately of 25% and 20%, respectively [Richardson et al. 2005]. Concerning bortezomib-related peripheral neuropathy, its incidence of grade 3–4 ranged from 8% to 15% and no significant differences in incidence, severity and outcome have been reported between newly diagnosed and relapsed and/or refractory patients. However, patients who received bortezomib-containing therapy as initial induction did experience less neuropathic pain and symptoms, which resolved or improved more quickly than in those with relapsed disease. Moreover, the new combination schemes using weekly doses of bortezomib have shown a significant reduction in the incidence of peripheral neuropathy, which is now between 5% and 8% [Mateos et al. 2010; Palumbo et al. 2010].

Subcutaneous formulation: efficacy and safety results

One different option to optimize the use of bortezomib is the change in the route of administration, and the next approach was to move to the subcutaneous administration of bortezomib, commonly abbreviated as subcu or subQ. The scientific rationale for this step was that in cynomolgus monkeys, the bioavailability, exposure variability, and extent and duration of whole blood proteasome inhibition after subQ bortezomib administration was comparable with IV administration, and the antitumor activity demonstrated in human xenograft studies after subQ administration was also similar to that observed after IV route.

The IFM group (French Francophone Myeloma Intergroup) was the first to conduct a phase I trial to compare pharmacokinetics and pharmacodynamics, and to assess the safety and efficacy of IV versus subQ administration of bortezomib in patients with relapsed and/or refractory MM [Moreau et al. 2008]. Twenty four patients were included and randomized 1:1 to receive the IV or subQ formulation. In this first trial, the final concentration at injections for administration was identical, 1 mg/ml in both IV and subQ routes of administration. The scheme consisted on 1.3 mg/m² IV or subQ on days 1, 4, 8 and 11 for up to eight 21-day cycles. Blood samples for pharmacokinetic/pharmacodynamic analysis were collected on days 1 and 11, with cycle 1 administered 30 minutes before bortezomib administration, and at 2, 5, 15, 30 and 60 minutes, and 2, 4, 6, 10, 24, 48 and 72 hours postdosing. Analyses were performed using a whole-blood 20S proteasome specific activity inhibition assay.

The most significant finding of this preliminary study was that subQ administration of bortezomib was comparable with the IV route, in terms of overall systemic availability and pharmacodynamic activity, similar toxicity profiles, and similar response rates. Concerning biological studies, it was well known that after IV bortezomib administration the volume of distribution was high, and this occurred in a similar way for the subQ route, indicating that both routes result into an extensive distribution into peripheral tissues. As expected, the time required for subQ administration to be absorbed is longer than for the IV route and accordingly the mean plasma concentration values were significantly lower for subQ administration, and the median time to reach the final plasma concentration was also longer. Nevertheless, the 20S proteasome inhibition was similar in both routes of administration, and the longer time required for subQ formulation to be absorbed resulted in a less-pronounced initial spike in proteasome inhibition has previously been mentioned, although cumulative pharmacodynamic activity was comparable with IV administration. The median number of cycles given was identical in both arms, six cycles, and this translated into a similar median total dose received of bortezomib, 28.55 mg/m² and 24.55 mg/m² for the IV and subQ arm, respectively. Although efficacy and safety were not the primary objectives of this study, preliminary results were also reported. Safety profile for subQ administration appeared to be similar to that for IV route. The frequency of bortezomib side effects of all grades was similar in both arms: thrombocytopenia (25% for IV and 33% for subQ), vomiting (25% for IV and 33% for subQ), nausea (42% for IV and 33% for subQ) and peripheral neuropathy (58% in both arms). Considering the adverse events of grade 3–4, 17% of patients developed neuropathy and 25% thrombocytopenia, in both arms. Only two patients, both in the IV route, experienced grade 4 adverse events. No deaths occurred during treatment, and due to the fact that this phase I trials was evaluating a novel route of administration, cardiac safety was evaluated by performing serial electrocardiograms (ECGs) to analyze the possible QTc interval prolongation. Minimal changes were reported postdosing for both arms, and none of the patients had QTc interval prolongations of more than 30 ms as compared with the QTc baseline. Reactions at the site injection, mainly erythema, were the most significant side effect reported in the subQ arm, occurring in approximately half of the administrations (51%), in 11 out of the 12 patients, but they did not require local or systemic therapy. No severe local reactions, such as ulceration or necrosis, were reported. Finally, as far as efficacy is concerned, although the sample size was small, the overall ≥PR rate was similar in both arms and comparable to that previously reported in the large trials conducted in relapsed patients, 42% and 58% for the IV and subQ routes, respectively, including one CR patient in each arm. In summary, the results of this phase I trial demonstrated that subQ administration of bortezomib is possible because pharmacokinetic and pharmacodynamic analyses comparing both routes were similar, preliminary safety data indicates that subQ formulation is manageable and, moreover, preliminary efficacy results showed not inferiority for this new route.

In order to confirm the safety and efficacy of this new route, an open-label, phase III trial was conducted at 53 centers in 10 countries in Europe, Asia and South America and recruited 222 patients. All of them were patients with MM who had received one to three previous lines of therapy, with adequate hematological, hepatic and renal function, without previous exposure to bortezomib and without peripheral neuropathy of grade 2 or more [Moreau et al. 2011]. Randomization was done in a ratio 2:1 to receive up to eight 21-day cycles of bortezomib at dose of 1.3 mg/m² on days 1, 4, 8 and 11 given by subQ or IV route. Patients with suboptimal response at the end of cycle 4 could additionally receive dexamethasone 20 mg orally on the day of and day after each bortezomib dose. The guidelines for IV preparation of bortezomib were the standard recommendation at a concentration of 1 mg/ml, but for subQ route, the concentration was higher in order to reduce the volume of injection, and bortezomib was prepared at a final concentration of 2.5 mg/ml. Accordingly, 3.5 mg of bortezomib (1 vial) was diluted in 1.4 ml of saline for subQ formulation instead of the 3.5 ml of saline used for the IV administration. The sites for subQ injection were the thighs or the abdomen, and were rotated for successive injections, alternating between right and left abdomen, upper and lower quadrant, or right and left thigh, and proximal and distal sites.

In both groups, patients received a median of eight cycles, and the median cumulative dose was also similar in both arms, 31.46 mg/m² and 33.76 mg/m² for IV and subQ, respectively. The ≥PR rate after four cycles of single-agent bortezomib was 42% in both the IV and subQ groups, including 14% and 12% CRs, respectively. The efficacy was re-evaluated after eight cycles, with 50% of patients in both groups having added dexamethasone to bortezomib single-agent treatment. The ≥PR rate was 52% in both groups, including 22% and 20% CRs in the IV and subQ arms, respectively. The median TTP was also similar in both groups, 9.4 months for IV route and 10.4 months for subQ administration. No differences so far were observed in terms of overall survival. The evaluation of the safety profile showed that grade 3 or higher adverse events were reported in 70% of patients in the IV group and 57% in the subQ arm. Thrombocytopenia of all grades was similar in both arms (35% and 36%), as well as the overall rates of gastrointestinal disorders, pneumonia, asthenia and fatigue were similar for all grades and grade 3–4. However, concerning peripheral neuropathy, the rates of any grade (38% and 53%), and of grade 2 (24% and 41%) and 3 (6% and 16%) severity were significantly lower with subQ than with IV route. Risk factors for development of peripheral neuropathy were well balanced in both groups, including the proportion of patients included in the trial with pre-existing peripheral neuropathy of grade 1 (23% and 28% in IV and subQ arm), diabetes at baseline or previous exposure to neurotoxic agents. Although the subQ route was associated with a lower incidence of peripheral neuropathy, the resolution or improvement was similar in both arms (67% and 62% for IV and subQ arms, respectively), in a median time of 1.5 months for subQ route and 2.8 for IV arm. In this trial, only 6% of patients developed one or more subcutaneous infection-site reactions reported as an adverse event, with resulted in a bortezomib dose modification in two (1%) patients. Two patients in the subQ group had severe injection-site reactions, but all of them resolved completely in a median of 6 days. All of these results confirm that subQ formulation of bortezomib is not inferior to IV route, with even an improved safety profile.

Discussion

There is already abundant information about the efficacy of bortezomib as a single agent and in combination with other agents in relapsed and/or refractory as well as in newly diagnosed myeloma patients, and all data have contributed to confirm bortezomib as one of the key drugs of the backbone treatment of myeloma patients. However, there are also some important drawbacks to be considered. The first is its toxicity, especially peripheral neuropathy, and the second is the route and schedule of administration, IV and twice a week requiring frequent visits at the hospital with an available vein access for its administration.

During this decade of experience with bortezomib treatment, some groups tried to optimize its use, with a dual objective: to reduce toxicity and to maintain efficacy. What approaches have been investigated to optimize the use of bortezomib?

The first approach was to modify the conventional regimen of administration: this was first done in the APEX trial, where patients were initially treated with eight cycles twice a week, and could then continue therapy with days 1, 8, 15 and 22 of every 35-day cycle for three cycles [Richardson et al. 2005]. In the VISTA trial, after four 6-week cycles of induction with the conventional biweekly dose, patients also received five 5-week cycles of consolidation consisting on the weekly administration of bortezomib. In both cases, the weekly schedules resulted in a reduction of AEs, but it was not possible to assess the efficacy due to the trial design [San Miguel et al. 2008]. The next approach was to use weekly bortezomib, but a higher dose, 1.5 mg/m² or 1.6 mg/m², and more recently, at a conventional dose of 1.3 mg/m², particularly when bortezomib is used in combination with other drugs [Mateos et al. 2010; Palumbo et al. 2010]. The reduced frequency of administration of bortezomib was generally associated with an improved toxicity profile, including reductions in cytopenias and peripheral neuropathy. Thus, the Italian and Spanish groups have recently reported that moving from the twice per week schedule to a once per week schedule results in a significant reduction in the incidence of peripheral neuropathy (from 14% to 3% of grade 3–4 in the Italian study and 5% in the Spanish VMP combination), and this strategy was especially useful for the elderly, a fragile patient population. Moreover, this reduction in the frequency of administration did not compromise the treatment efficacy, and in fact a longer TTP was observed in both the Italian and Spanish trials, as compared with the VISTA which could be attributed to the reduction in toxicity that allowed patients to remain on treatment for a longer period of time. The third approach has been to change the route of administration, from the IV to the subQ route, and the first objective planned for this new route was that subQ formulation should not be inferior to the conventional IV route in efficacy. This third approach has been validated in a large, randomized trial, showing no significant differences between IV and subQ routes in terms of efficacy; in addition, in patients with a suboptimal response after four cycles, the addition of dexamethasone resulted in a similar improvement of efficacy in both groups with no differences in TTP and overall survival between arms. Moreover, the frequency of peripheral neuropathy, the most important bortezomib-related adverse event, appears to be significantly reduced with the subQ route.

If we consider all of these results, it can be concluded that subQ administration of bortezomib is feasible, and could contribute to optimizing the management of bortezomib in the treatment of myeloma patients. SubQ formulation facilitates the treatment with bortezomib, especially for those with poor venous access. SubQ administration can also require less time at the outpatient facilities on each bortezomib dose and in the future could allow home administration. The incidence of peripheral neuropathy is significantly lower with subQ formulation and it is one additional advantage for its use in clinical practice. However, it is important to keep in mind that bortezomib can be associated with some serious adverse events, and the control by nurses and doctors is essential. The treatment with bortezomib still requires a blood count and biochemistry to be performed at least once per cycle (usually on day 1 of each cycle), and some patients develop bortezomib-related hypotension requiring IV hydration after bortezomib administration, and this cannot be ignored because if we relax on the management of bortezomib side effects, we could end up with the paradoxical result that these approaches for the optimization have finally translated into a step backwards.

An attractive possibility to be explored is the use of subQ route in other settings, especially because the efficacy and safety results resulted from a phase III randomized trial, but conducted in a selected group of patients, relapse and/or refractory and bortezomib-naïve patients. It would be attractive to explore its efficacy and safety in weekly dose schedule, particularly for bortezomib combinations such as VMP, VCD, PAD, VTD or VRD, since we should expect even lower incidence of peripheral neuropathy and the possibility of more prolonged treatment, including maintenance approaches.

In addition, it is important to remark that the preparation of IV and subQ formulations is different and the concentration of the subQ formulation is higher in order to reduce the volume of administration, so it would be an important mistake to administer a subQ formulation via IV route.

Some questions remain opened and under investigation, such as a better understanding of the mechanisms inducing the reduction of the incidence of neuropathy, or how the subQ formulation is distributed in the subcutaneous fat, and whether the amount of subcutaneous fat should be taken into account when prescribing this new formulation. In this way, one caveat is the role of the efficacy of the subQ formulation in extramedullar disease, due to the fact that the first studies in tumor models suggested that the IV administration was optimal to optimize the antitumor effect due to the short time required in order to reach the final plasma concentration. Other aspects should also be considered, such as the concomitant medications especially the use of antiplatelets drugs or anticoagulants that could induce subcutaneous bleeding after subQ administration of bortezomib resulting in an uncontrolled distribution of the drug.

In conclusion, the subQ formulation of bortezomib represents an additional step towards the optimization of bortezomib use, resulting in a more convenient route that is at least as effective as the IV route. However, there is no official approval of the subQ route of administration yet and it would be convenient to wait for the authorities to approve the new route of administration before we widely begin to give subQ bortezomib to all patients.