Division of Hematology and Medical Oncology, Center for Lymphoma and Myeloma, Weill Medical College of Cornell University and New York Presbyterian Hospital, New York, NY, USA
Division of Hematology and Medical Oncology, Center for Lymphoma and Myeloma, Weill Medical College of Cornell University and New York Presbyterian Hospital, 525 East 68th Street, New York, NY 10021, USA
Non-Hodgkin's lymphoma (NHL) has become an increasingly common problem in the US and other westernized countries in recent years, and its incidence has approximately doubled in the past 2 decades.1 It is now the fifth most common cancer in the US, with more than 54,000 new cases arising each year.2 The estimated incidence in Europe is approximately 74,000 cases.3
By pathological and other criteria, NHL is divided into a range of subtypes with differing clinical features and outcomes. These subtypes can frequently follow an aggressive or an indolent course. Of the indolent subtypes, follicular lymphoma (FL) is the most common, accounting for approximately 22% of all cases of NHL.4, 5 FL remains a generally incurable disease with a median survival of approximately 10 years,6 during which patients typically receive multiple successive chemotherapy regimens. Current chemotherapeutic regimens for FL include cyclophosphamide, vincristine and prednisone (CVP); CVP with rituximab (R-CVP); cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP); CHOP with rituximab (R-CHOP); fludarabine (with or without mitoxantrone); chlorambucil and prednisone; single-agent rituximab; rituximab with fludarabine; radioimmunotherapy with radiolabeled monoclonal antibodies such as 131I-tositumomab; ibritumomab tiuxetan; and various other combinations. Initial overall response rates (ORRs) of 50–100% have been obtained, but until recently, no regimen had demonstrated a significant advantage (vs. another) in terms of overall survival.5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 However, results from a phase 3 study involving 201 patients with FL indicate that immunochemotherapy with rituximab plus mitoxantrone, chlorambucil and prednisolone (R-MCP) is significantly superior to chemotherapy with MCP and offers a significant survival advantage.16
Mantle cell lymphoma (MCL) is an aggressive subtype of NHL. It accounts for approximately 5% of all cases of NHL in North America and Europe, and has a relatively unfavorable prognosis, with a median survival of 3–4 years.17, 18, 19, 20, 21 As a lymphoma that is both rapidly progressive and generally incurable, MCL exhibits some of the adverse features of indolent and aggressive lymphomas. First-line chemotherapeutic regimens for relapsed MCL include hyper-fractionated cyclophosphamide, vincristine, doxorubicin and dexamethasone alternating with methotrexate and cytarabine; CHOP with or without rituximab; dexamethasone, high-dose cytarabine and cisplatin (DHAP); CHOP–DHAP; fludarabine and combinations; and rituximab and combinations. Initial ORRs of 30–95% have been achieved, but remission durations are short and overall survival remains limited.14, 17, 18, 19, 20, 21, 22 New agents or regimens for first-line treatment of MCL must be assessed against strategies involving combination chemotherapy with or without rituximab and stem-cell transplantation, which result in ORRs in excess of 90%.21, 23, 24 However, it should be noted that most patients with MCL are unsuitable for allogeneic stem cell transplantation owing to their age, and the majority who undergo autologous stem cell transplantation relapse.25 By contrast to MCL, diffuse large B-cell lymphoma (DLBCL), which is the most common NHL subtype, accounting for 40% of all cases,26 may be cured by conventional chemotherapy (CHOP or R-CHOP) in roughly half of cases.22, 27
Waldenström's macroglobulinemia (WM) is a low-grade B-cell lymphoma that accounts for approximately 1–2% of hematologic cancers.28 WM straddles the transition from lymphoma to multiple myeloma (MM) and, as a result, shares several important similarities with MM. The malignant cells in WM are found as small B lymphocytes, plasmacytoid lymphocytes and plasma cells. Additionally, in WM, there is a high rate of immunoglobulin synthesis and secretion. Commonly employed treatments for WM include chlorambucil, melphalan, cyclophosphamide, purine analogs and rituximab both as a single agent and in combination therapy, which produce ORRs of 30–100%. Median survival for patients with WM has been reported as 5–6 years. However, the disease can also have an indolent course and at least 20% of patients have been observed to survive longer than 10 years, with many dying of unrelated causes.28
Although the introduction of new treatment agents and regimens for NHL has resulted in improved complete response (CR) rates and ORRs in some settings, the lack of any significant improvement in overall survival in many of the subtypes indicates a clear need for further novel drugs and interventions, particularly for MCL.8, 17, 22
Proteasome inhibition is one such therapeutic approach that could provide clinical benefit in NHL. Bortezomib (VELCADE®, Millennium Pharmaceuticals Inc. and Johnson & Johnson Pharmaceuticals Research & Development, L.L.C.), a dipeptide boronic acid analog, is the first proteasome inhibitor to be evaluated in human studies.29, 30 Bortezomib is approved for second-line treatment of MM in the US and in Europe, and its activity is being investigated in NHL. The US FDA has granted bortezomib fast-track designation for relapsed and refractory MCL on the basis of data from 4 phase 2 trials of the agent in this treatment setting. This manuscript reviews preclinical and clinical data on the activity of bortezomib in various NHL subtypes.
Overview of proteasome inhibition as anticancer therapy
The 26S proteasome is a major component of the ubiquitin–proteasome pathway, which hydrolyzes unwanted or damaged proteins and is critical for regulated degradation of proteins involved in many aspects of the cell cycle, including apoptosis. Inhibition of this pathway is therefore a logical target for anticancer therapies.29, 31, 32, 33, 34, 35, 36 The proteasome is a multi-catalytic unit consisting of a 19S cap on both ends, which recognizes ubiquitin-tagged proteins that are marked for degradation, and a 20S core, which contains 6 proteolytically active sites of 3 different types, namely chymotrypsin-like, caspase-like and trypsin-like sites.37, 38, 39 Cell cycle regulatory proteins are tagged with ubiquitin and then degraded by the 26S proteasome, the activity of which has been comprehensively reviewed elsewhere (see Fig. 1).32, 34, 37, 38, 40, 41
Among the proteins degraded by the ubiquitin–proteasome pathway are cyclins, cyclin-dependent kinase inhibitors such as p21 and p27, the tumor suppressor p53 and the inhibitory protein IκB, which inhibits nuclear factor kappa-B (NF-κB).34, 40, 41 Inhibition of the 26S proteasome results in the accumulation of these substrates and therefore causes cell cycle disruption and promotes cell death via multiple pathways. Table I lists some of the cell cycle proteins degraded by the 26S proteasome, along with their functions. Significantly for anticancer therapies, malignant cells are much more sensitive to the proapoptotic effects of proteasome inhibition than normal cells. This phenomenon possibly relates to higher cellular replication rates but may be due to the different effects of proteasome inhibition on the turnover of cell cycle regulatory proteins in malignant cells.29, 30, 32, 35, 42 Nonetheless, it is believed that dosing of a proteasome inhibitor must be carefully controlled to minimize any adverse impact on normal, healthy cells.
Table I. Proteins Involved in the Cell Cycle that are Degraded by the Proteasome (Adapted from Adams et al29, 40, 41)
Class of protein
Cyclins A, B, D and E
Cyclin-dependent kinase inhibitors
Regulation of cyclin activity
c-fos/c-jun, c-myc, N-myc
Inhibitor of NF-κB
CDK1/cyclin B phosphatase
Tyrosine amino transferase
Topoisomerase I and IIα
Relaxation of DNA supercoiling
Pro- and antiapoptotic factors
Inhibitor of apoptosis
The NF-κB pathway appears to be important for the anticancer activity of proteasome inhibitors in some tumor types. In quiescent cells, NF-κB is bound in the cytoplasm and inhibited by IκB. In response to a wide variety of external signals, such as stress induced by radiotherapy or chemotherapy, IkB is phosphorylated, ubiquitinated and consequently degraded by the proteasome. NF-κB then translocates to the nucleus, where it activates antiapoptotic and cell-growth genes, causing increased synthesis of growth factors, cell adhesion molecules, angiogenesis factors and antiapoptotic factors.31, 32 The stabilization of IkB through proteasome inhibition therefore prevents NF-κB activation, making cells more susceptible to stressors.29, 32 This is potentially significant for the use of proteasome inhibitors in combination with other anticancer therapies, because proteasome inhibitors may sensitize resistant cells to the effects of standard chemotherapy agents or radiotherapy.
In addition, lymphocytes, along with monocytes and neurons, demonstrate constitutive activation of NF-κB. In these cells, NF-κB is activated even in the absence of external signals. Inhibition of NF-κB activity is potentially one of the mechanisms of action in the lymphocidal effect of glucocorticoids.43, 44 Chronic lymphocytic leukemia (CLL) is a lymphoproliferative disease in which NF-κB has been shown to be of significance. A raised level of constitutive NF-κB activity is seen in CLL cells,45 and this might be responsible for mediating resistance to fludarabine.46 CLL cells cultured with an agonistic anti-CD40 monoclonal antibody showed a decreased sensitivity to fludarabine. This sensitivity could be restored by concurrently treating the cells with decoy oligonucleotides, which interfered with NF-κB induction of gene transcription. In WM, it might be predicted that proteasome inhibition would be an efficacious therapeutic option, given the similarities between WM and MM. However, preliminary data suggest that NF-κB might not be as important in WM as it is in MM.47 Therefore, WM might not be responsive to proteasome inhibition if inhibition of NF-κB is the primary mechanism of action, whereas clinical activity of bortezomib in WM would suggest a mechanism of action other than inhibition of NF-κB.
Several viral-associated lymphomas demonstrate a dependency on NF-κB, most notably adult T-cell leukemia/lymphoma (ATLL), post-transplant lymphoproliferative disorders (PTLDs) and primary effusion lymphomas (PELs). These lymphomas might therefore be responsive to proteasome inhibition therapy. In ATLL, the HTLV-1 tax protein binds and constitutively activates the IKK complex, leading to constitutive NF-κB activity.48, 49 In PTLDs, the EBV latent membrane protein 1 (LMP1) binds to and activates signaling through TRAFs 1 and 3, mimicking signaling through TNF-receptor family members and leading to constitutive NF-κB activation.50 Human herpes virus-8, the etiologic agent for Kaposi's sarcoma, PELs and multicentric Castleman's disease, constitutively activates NF-κB through its vFLIP protein binding of the IKK complex. Inhibition of NF-κB leads to apoptosis of the PEL cells.51, 52
A wide range of natural and synthetic compounds have been found to inhibit the proteasome, including lactacystin and derivatives, peptide vinyl sulfones, peptide epoxyketones, peptide aldehydes and dipeptidyl boronic acids.34, 38, 41, 53 However, many of these compounds are nonspecific to the proteasome and/or bind irreversibly to the 20S core, making them relatively unsuitable as therapeutic agents. Bortezomib, by contrast, has proved to be a potent, more selective and reversible inhibitor of the proteasome, with demonstrated clinical efficacy.
In vitro and in vivo studies have shown single-agent bortezomib to be active against a range of different tumor types, including a number of NHL subtypes. In addition, bortezomib has been shown to potentiate the activity of various agents used to treat NHL in preclinical models. The results of these preclinical studies have provided the rationale for the subsequent investigation of bortezomib in clinical trials.
Initial in vitro screening using the National Cancer Institute's tumor cell line panels showed single-agent bortezomib to be uniquely active against a wide range of tumor types,29, 31, 41 and subsequent studies have elucidated this activity in MM, NHL and nonsmall cell lung cancer (NSCLC) cell lines, among others. Bortezomib inhibits growth, induces apoptosis and overcomes drug resistance in human MM cell lines and primary patient MM cells through a range of molecular mechanisms, including upregulation of p53 expression and activation of caspase pathways.54, 55 It also inhibits human MM cell growth in vivo and was shown to prolong survival in an SCID mouse model.56 Analysis of tumors from treated mice indicates that bortezomib induces apoptosis and reduces angiogenesis in vivo. In NSCLC, single-agent bortezomib induces growth inhibition in A549, H520, H460, H358 and H322 cell lines.57, 58, 59 Bortezomib also upregulates p21 and p27 in NSCLC, thereby leading to cell cycle arrest.60 This finding is potentially important given that p27 loss is associated with a poor prognosis in NSCLC.60
NHL cells also appear to be affected by bortezomib. The agent has been shown to cause cell cycle arrest and rapid induction of apoptosis in MCL and PEL cells. As NF-κB is constitutively active in both these tumor types, inhibition of the NF-κB pathway by bortezomib may contribute to the apoptotic activity.61, 62 In addition, bortezomib has been shown to induce apoptosis through the upregulation of p21, p27 and p53 in PEL cell lines.63 The effectiveness of bortezomib in PEL cells is particularly significant, because these tumors, a rare subtype of NHL, are relatively resistant to standard cytotoxic chemotherapy.62 Activity has also been demonstrated in cell samples derived from patients with MCL and patients with FL. In 1 study, using a primary culture system, bortezomib was found to be effective in cells from both NHL subtypes; MCL cells were significantly more sensitive to bortezomib than FL cells, with median EC50 values of 209 nM (range 122–989 nM, n = 5) and 1,311 nM (range 153–2,211 nM, n = 8), respectively.64 The mechanisms of bortezomib activity in MCL have been investigated in a study using 4 MCL cell lines. MCL is characterized by the constitutive overexpression of cyclin D1, with loss of p27 through proteasomal degradation identifying MCL patients with poor clinical outcomes.65 Expression of cell cycle regulators has been identified in bortezomib studies as an early effect of proteasome inhibition; this is followed by downregulation of cyclin D1 expression and subsequent cell cycle arrest.66 Bortezomib has also demonstrated in vivo activity in MCL, suppressing transcription factors and cell growth, and preventing the development of tumors in human MCL-xenografted mice.29, 30, 67 Additionally, bortezomib has demonstrated potent inhibition of ATLL cells in in vitro and xenograft murine models.68
Bortezomib in combination
Combinations of bortezomib and other agents have shown additive or synergistic activity in vitro and in vivo. Bortezomib potently sensitizes MM cell lines and patient cells to anthracyclines such as doxorubicin, to mitoxantrone and to alkylating agents such as melphalan, and this sensitization even occurs in cells resistant to these drugs.69, 70 Similarly, bortezomib in combination with dexamethasone or immunomodulatory derivatives of thalidomide (IMiDs) has demonstrated enhanced antitumor activity in MM cells.56 Numerous combinations have been shown to be additive or synergistic in NSCLC, including bortezomib plus docetaxel, gemcitabine, gemcitabine/carboplatin and histone deacetylase inhibitors.59, 71, 72, 73, 74
In NHL, co-treatment of MCL cell lines with bortezomib plus the chemotherapeutic agents doxorubicin, vincristine and 4-hydroperoxycyclophosphamide has been shown to produce a synergistic effect, which is greater if cells are sequentially treated with doxorubicin or vincristine and then bortezomib.75 The order of treatments in a combination regimen was similarly shown to be important in another study, where synergistic interactions occurred when PEL cell lines were pretreated with bortezomib prior to doxorubicin or paclitaxel.62 Interactions were additive or even antagonistic only with simultaneous treatment or chemotherapy pretreatment. Bortezomib in combination with rituximab has been shown to offer at least additive activity in in vitro and in vivo DLBCL models, and may be active in other lymphomas that are sensitive to rituximab, such as FL.12, 76 The same combination has also shown additive in vitro activity in B-cell CLL (B-CLL) cells.77 Similarly, bortezomib plus fludarabine or 2-chlorodeoxyadenosine (2-CDA) results in increased cytotoxicity when compared to the single agents in CLL cells in vitro.78
Bortezomib sensitizes a number of NHL cell lines to apoptosis induced by antibodies to tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) receptors TRAIL-R1 and TRAIL-R2.79 It also overcomes resistance to other agents in PEL cell lines. In combination with another novel agent, antisense to Bcl-2, bortezomib sensitizes B-cell lymphomas and MM cells to cyclophosphamide. The optimal schedule appears to be antisense to Bcl-2 followed by cyclophosphamide preceding all doses of bortezomib.80 Finally, as well as sensitizing cells to the effects of chemotherapy agents, bortezomib has also been shown to enhance the radiosensitivity of tumor cells.81
Clinical efficacy in NHL
The efficacy of bortezomib in several subtypes of NHL has been demonstrated in phase 2 clinical trials. A number of studies of bortezomib, both as a single agent and in combinations, are ongoing or planned. Interim results indicate that bortezomib is well tolerated and active in NHL, with some durable responses and ORRs of 18–60% in FL and 39–56% in MCL. There is also preliminary evidence of activity in other subtypes, including marginal zone lymphoma (MZL) and WM. Table II summarizes the response data from ongoing clinical trials.
Table II. Overall Response Rates (ORRs) in Single-Agent Bortezomib Studies in NHL
In an NCI-sponsored, multi-institutional phase 2 study, bortezomib (1.5 mg/m2) was given to patients with relapsed indolent NHL on the conventional schedule of twice weekly for 2 out of 3 weeks. The ORR in 19 patients with FL was 60%, including 1 CR and 1 unconfirmed CR (CRu). Responses were durable, but patients generally required at least 3 cycles of bortezomib before a response occurred. In addition, 4/4 patients with MZL and 1/5 patients with small lymphocytic lymphoma (SLL) achieved a partial response (PR).82 In another phase 2 study, in patients with relapsed or refractory B-cell NHL, responses were seen in several indolent subtypes, with 1 CRu in 5 FL patients, 1 PR in 2 WM patients and 1 CR in 4 SLL patients.83 In a third phase 2 study, in patients with relapsed or refractory NHL who received bortezomib (1.3 mg/m2) on the conventional schedule, 2/11 evaluable FL patients achieved a PR, for an ORR of 18% 3 months after the end of treatment.85 This latter study discontinued therapy in nonresponders (even without progression). The seemingly lower response rates (relative to the earlier study, which employed treatment until progression) suggest that the time to response in FL may be longer than in other subtypes and that more extended treatment in FL may be necessary. An alternative explanation is that the modest difference in dose (1.3 mg/m2vs. 1.5 mg/m2) may be relevant.
Very encouraging activity has been observed with bortezomib in MCL. In the NCI-sponsored phase 2 study of bortezomib (1.5 mg/ m2) in relapsed indolent NHL, the ORR among evaluable MCL patients was 56%, with responses typically being achieved by the second cycle of therapy. Responses were durable in some cases, with a range of 6–19 months.82 In a second study of bortezomib at 1.5 mg/m2, in relapsed/refractory B-cell NHL, the ORR in patients with MCL was 41%, and responses were similarly durable (Fig. 2).83 Comparable response rates have been reported in studies using bortezomib (1.3 mg/m2) on the conventional schedule, with ORRs of 46% (1 CRu and 12 PRs) and 39% (1 CR/CRu and 6 PRs) reported in patients with advanced stage MCL and relapsed/refractory MCL, respectively.85, 86 Interestingly, in the multi-center study of patients with advanced stage MCL, the ORR was comparable in previously untreated patients and in relapsed/refractory patients.86 In addition, interim results from the PINNACLE phase 2 study of bortezomib (1.3 mg/m2) in patients with relapsed or refractory MCL show a response rate of 42% at the second stage of analysis of 48 patients. This includes 2 CR, 2 CRu and 16 PR.84
Bortezomib is also active in WM. In a dedicated phase 2 study of bortezomib (1.3 mg/m2) on the conventional schedule in patients with previously untreated or treated WM, an ORR of 21.1% was achieved. Of 19 evaluable patients, 4 experienced a PR and 14 had stable disease.87 Evidence of activity was also seen in another phase 2 study of bortezomib (1.3 mg/m2), in which 2/5 (40%) WM patients achieved a PR on the basis of paraprotein reduction.85 By contrast, no responses were reported in a phase 2 study of bortezomib in patients with fludarabine-refractory B-CLL.88 However, 6/19 evaluable patients experienced a ≥50% reduction of lymphadenopathy, suggesting that use in combination with other chemotherapeutic agents warrants evaluation.88
Bortezomib in combination
Owing to the synergistic and additive effects seen in cell-line and preclinical studies, a number of clinical trials are underway to investigate the toxicity and efficacy of bortezomib in combination with a range of chemotherapeutic agents. Table III highlights some of the combination regimens currently being investigated in NHL.
Table III. Ongoing Studies of Bortezomib in Combination with other Agents
Blum and coworkers
Bortezomib + 17-AAG
Relapsed/refractory hematologic malignancies
Grant and coworkers
Bortezomib + flavopiridol
Recurrent/refractory indolent B-cell neoplasms
Hillmen and coworkers
Bortezomib + cyclophosphamide + prednisone
Relapsed/refractory indolent lymphoma
Koc and coworkers
Bortezomib + fludarabine
Relapsed/refractory indolent NHL, CLL
Leonard and coworkers
Bortezomib + R-CHOP
Untreated DLBCL, MCL
Lister and coworkers
Bortezomib + rituximab
Recurrent FL, MCL, WM
Wilson and coworkers
Bortezomib + EPOCH
Bortezomib is being investigated in combination with rituximab in a phase 2 study of patients with indolent NHL (FL or MZL), using 2 different administration schedules. Patients in Arm A receive bortezomib (1.3 mg/m2) on the conventional schedule, whereas in Arm B, the dose is 1.6 mg/m2, given on days 1, 8, 15 and 22 of a 35-day cycle. All patients receive the standard dose of rituximab, of 375 mg/m2 once weekly for 4 weeks. Preliminary results from the planned interim analysis of 34 evaluable patients show response rates of 35% (1 CR, 1 Cru and 4 PR) and 41% (1 CR and 6 PR) in Arm A and Arm B, respectively. Notably, 3/5 patients with MZL achieved a PR.89
Bortezomib plus pegylated liposomal doxorubicin has demonstrated promising antitumor activity in a phase 1 trial in patients with advanced hematologic malignancies, including relapsed/refractory T-cell NHL, B-cell NHL and MM. Eight of 22 evaluable MM patients achieved a CR or near-CR and a further 8 had PRs, while 1 patient with relapsed/refractory T-cell NHL achieved a CR and 2 patients with B-cell NHL had PRs.90 Bortezomib has also been combined with dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide and doxorubicin (EPOCH) chemotherapy in a phase 1/2 study in patients with relapsed or refractory DLBCL. Preliminary results indicate modest activity in this resistant-disease population; bortezomib alone produced 1 PR (n = 17), but bortezomib in combination with dose-adjusted EPOCH produced 4 CRs and 3 PRs (n = 28), for an ORR of 25%.91
Bortezomib, adriamycin and dexamethasone combination therapy has been observed to be an effective regimen in previously untreated MM patients. A 95% ORR was seen in a phase 1/2 study of 21 patients, including 24% with CR and an additional 5% near CR, and the use of the therapy did not adversely affect the subsequent harvesting of peripheral blood stem cells for transplantation.92 In addition, bortezomib plus dexamethasone has been found to be an effective induction therapy prior to autologous stem cell transplantation in patients with newly diagnosed MM. An ORR of 80% has been obtained in 1 phase 2 study of 35 patients, including 30% with CR plus a very good rate of PR.93 Similar regimens may prove applicable in lymphoma.
Safety and tolerability
The standard regimen of bortezomib—twice weekly for 2 of 3 weeks—has generally been found to be well tolerated in phase 1, 2 and 3 trials, with a predictable and manageable toxicity profile both as a single agent and in combination with other chemotherapeutic regimens. In addition, additive toxicity does not appear to arise with bortezomib in preliminary data from combination regimens. This attribute may prove beneficial to its use alongside many standard chemotherapy agents and regimens, which themselves are associated with various levels of toxicity.
For example, the R-CHOP regimen is commonly associated with adverse events including neutropenia, alopecia, nausea, fever and leukopenia, (including grade 3–4 cytopenias in some cases). Most of these adverse events are attributed to CHOP, with grade 1–2 fever and chills being the most common adverse events attributed to rituximab.11, 15, 21, 94 Similarly, R-CVP can be associated with adverse events including fatigue, neutropenia (24% grade 3–4) and back pain; in a recent study, 97% of patients receiving R-CVP and 95% of patients receiving CVP reported at least 1 adverse event.10 Fludarabine plus mitoxantrone with or without rituximab has been shown to have hematologic toxicity similar to that of CHOP with or without rituximab, but significantly lower rates of grade 3–4 nausea, alopecia and peripheral neurologic toxicity,9 while rituximab plus fludarabine is associated with substantial rates of grade 3–4 neutropenia and lower rates of grade 1–2 anemia and thrombocytopenia.12 Single-agent rituximab is associated with low rates of initial grade 3–4 toxicity and no cumulative or additional toxicities with maintenance therapy.95 Data with bortezomib suggest that toxicities associated with single-agent use are comparable (if not better) than those commonly observed with standard NHL regimens, and that combination regimens may ultimately be feasible.
In a phase 1 study of bortezomib in patients with refractory hematologic malignancies, in which bortezomib was given twice weekly for 4 out of 6 weeks, dose-limiting toxicities (DLTs) at doses greater than the well-tolerated 1.04 mg/m2 maximum tolerated dose (MTD) included thrombocytopenia, hyponatremia, hypokalemia, fatigue and malaise.96 A further phase 1 study in patients with advanced solid tumors, who received bortezomib twice weekly for 2 out of 3 weeks, found DLTs of diarrhea and sensory neurotoxicity, and suggested an MTD of 1.56 mg/m2 on this schedule.97 The toxicity profiles observed in these phase 1 studies led to the development of the conventional schedule used in phase 2 and 3 studies: bortezomib twice weekly for 2 out of 3 weeks.36, 96 A recent phase 1 study in patients with advanced solid tumors reported DLTs of diarrhea and hypotension at 2.0 mg/m2 when bortezomib was administered weekly for 4 out of 5 weeks; an MTD of 1.6 mg/m2 was recommended for this less-intense schedule.96 Such schedules continue to be investigated.
Two phase 2 studies in relapsed/refractory MM of bortezomib (1.3 mg/m2) on the conventional schedule have indicated that the most common grade 3 adverse event associated with the agent is thrombocytopenia. In 1 study, grade 3 events included thrombocytopenia (28% of patients), fatigue (12%), peripheral neuropathy (12%) and neutropenia (11%), and grade 4 events occurred in 14% of patients.98 In the other study, the most common grade 3 adverse events were thrombocytopenia (24%), neutropenia (17%), lymphopenia (11%) and peripheral neuropathy (9%), with 9% of patients experiencing a grade 4 event.99 In both studies, adverse events were generally manageable, and thrombocytopenia was transient, resolving during the 10-day rest period. A multinational, randomized, phase 3 study has compared the efficacy of bortezomib with high-dose dexamethasone in patients with relapsed MM who had received 1–3 prior treatments.100 The dexamethasone arm was closed after a preplanned interim analysis demonstrated statistically significant improvements in time-to-progression and survival for bortezomib. No major unexpected safety differences were observed between the 2 treatment groups.
Bortezomib has been similarly well tolerated in phase 2 studies in NHL. In 1 study in relapsed or refractory B-cell lymphoma, grade 3 toxicities included thrombocytopenia (47%), gastrointestinal side effects (20%), fatigue (13%), neutropenia (10%) and peripheral neuropathy (5%), and grade 4 toxicities occurred in 15% of patients.83 In a similar study in untreated, relapsed or refractory NHL, the most common grade 3 toxicities were lymphopenia and thrombocytopenia; again, these were generally short-lived and resolved during the week-off therapy.101 Another phase 2 study of bortezomib in relapsed/refractory NHL and Hodgkin's disease found the most common grade 3–4 toxicities to be thrombocytopenia (43%), anemia (15%), fatigue (13%) and peripheral neuropathy (5%).85
Serious adverse events recorded during bortezomib studies include necrotizing vasculitis and edema. A maculopapular rash was reported in 6/59 patients in a phase 2 trial in relapsed/refractory B-cell NHL, of which 2 cases were confirmed by biopsy as grade 3 necrotizing vasculitis.83 Another phase 2 study in indolent NHL and MCL reported several incidences of rash, with 3 cases linked to grade 1/2 vasculitis; the rash resolved during treatment rest weeks and was not associated with any severe complications.101 In a further phase 2 trial in MCL patients, there were 3 cases of peripheral edema and 2 deaths related to grade 4 acute edema and to progressive MCL with severe edema.102 These 5 patients had baseline dyspnea and/or edema, leading investigators to exclude patients with these conditions; once these patients were excluded, this toxicity was no longer seen.86
Bortezomib plus pegylated liposomal doxorubicin has been administered safely, with no significant drug–drug interactions, in patients with advanced hematologic malignancies.90 In patients with DLBCL, bortezomib is tolerated in combination with dose-adjusted EPOCH containing vincristine; toxicities are similar to those associated with EPOCH alone.91 Phase 1/2 studies for bortezomib in combination with R-CHOP are ongoing. This lack of additive toxicity reflects the predictable and manageable toxicities seen in studies of combination regimens in MM.93, 98, 99, 103, 104, 105 Nevertheless, it will be important to consider the potential for additive toxicity in terms of neurotoxicity and thrombocytopenia, owing to the association of these toxicities with bortezomib monotherapy.
Discussion and conclusions
With the increase in incidence of NHL in recent years, there is a clear clinical need for novel agents to offer new options in resistant disease and potentially to improve outcomes, even in curative settings. Proteasome inhibition has emerged as a therapeutically active approach, and the first proteasome inhibitor to enter clinical trials, bortezomib, is approved for use in second-line MM and shows promise in NHL therapy, albeit at an early stage in its clinical development. As one of a new generation of targeted agents, bortezomib has demonstrated antitumor activity in preclinical studies. Results and preliminary data from 9 phase 2 studies involving limited numbers of patients have shown bortezomib to have activity in a number of NHL subtypes, including FL, MZL, MCL and WM. The biological differences between these subtypes may account for the variations in activity observed with bortezomib; for example, patients with FL appear to respond to bortezomib later in treatment than patients with MCL, possibly as a consequence of the indolent course of FL when compared with the aggressive nature of MCL. It will be important to bear such differences in mind in the future development of the agent and in the analysis of its activity in clinical trials. Furthermore, new techniques, such as gene-expression profiling, continue to elucidate the molecular mechanisms involved in NHL and could clarify the varying effectiveness of bortezomib in different NHL subtypes.
Bortezomib sensitizes NHL cell lines to other therapeutic agents and, given the potential lack of significant additive toxicity effects, may therefore be appropriate as part of combination therapies throughout the NHL treatment algorithm. It will be important to investigate, in clinical studies, the extent to which these preclinical findings are reflected in the clinical setting. It will also be important to explore the effects of different doses and differing dosing schedules to obtain optimum synergy between bortezomib and other agents. Indeed, optimum scheduling may vary between combination partner agents, and among histologic subtypes. Further elucidation of the mechanism of action of bortezomib, both as a single agent and in combination, will aid understanding of these effects and assist in maximizing the efficacy of bortezomib-based combination regimens for different NHL subtypes.
Ongoing studies are examining the effectiveness of bortezomib alone or in combination with other agents as either first-line or salvage therapy. Phase 1 and phase 2 studies are exploring the integration of bortezomib into regimens including R-CHOP, R-EPOCH, rituximab–cyclophosphamide–prednisone and fludarabine. The rationale for these trials is that bortezomib may enhance the activity of such standard regimens, potentially even by overcoming chemoresistance. In addition, combinations with novel, biologic targeted agents represent other possibilities.
Preclinical and clinical data available so far suggest that proteosome inhibition with bortezomib represents a promising new direction for lymphoma therapy. Confirmation of early results in follow-up studies will potentially lead to its use in a variety of commonly encountered NHL treatment settings. The principal challenge facing investigators and clinicians, given the heterogeneity of the disease and the large number of other active agents, will be the development of optimal dosing regimens and rational combinations that will ultimately result in the most clinical benefits for lymphoma patients.