Phase 1 trial of bortezomib plus R-CHOP in previously untreated patients with aggressive non-Hodgkin lymphoma




Bortezomib has preclinical and clinical in B-cell lymphomas, both alone and in combination with other agents. A phase 1 evaluation was conducted of bortezomib with rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) in patients with untreated diffuse large B-cell lymphoma (DLBCL) or mantle cell lymphoma (MCL).


Twenty patients (16 with DLBCL and 4 with MCL) with a median age of 66 years (range, 29-84 years) were enrolled. Eleven subjects (55%) had an elevated lactate dehydrogenase level, and 10 patients (50%) had International Prognostic Index scores of 3 to 5. Standard R-CHOP was administered on a 21-day cycle for 6 cycles, with 1 of 3 dose levels of bortezomib (0.7 mg/m2 [n = 4 patients], 1.0 mg/m2 [n = 9 patients], or 1.3 mg/m2 [n = 7 patients]) administered on Days 1 and 4 of each cycle.


The maximum tolerated dose of bortezomib with R-CHOP was not reached, and the 1.3-mg/m2 dose level had acceptable tolerability. A dose-limiting toxicity (pulmonary) was only observed in 1 patient receiving 1.0 mg/m2 of bortezomib. Neuropathy occurred in 13 patients (65%), but was mostly grade 1 (45%) and reached grade 3 in only 1 patient (all toxicities were graded using the Common Terminology Criteria for Adverse Events, version 3.0). Grade 4 hematologic toxicity occurred in 7 patients (35%). Of 19 evaluable patients, all responded, with 18 (95%) cases of complete response/complete response unconfirmed achieved and 1 (5%) partial response reported. At a median follow-up of 56 months, overall survival at 4 years was 75% and progression-free survival was 58%.


Bortezomib at a dose of 1.3 mg/m2 twice per cycle can be added to R-CHOP chemotherapy with acceptable toxicity. Multi-institutional and cooperative group follow-up studies of this regimen are currently ongoing. Cancer 2010. © 2010 American Cancer Society.

Non-Hodgkin lymphoma (NHL) is comprised of a diverse group of malignancies with different biologic characteristics and clinical outcomes to therapy.1, 2 Diffuse large B-cell lymphoma (DLBCL), the most common lymphoma subtype, is an aggressive disease that is currently curable in over half of patients.3 However, patients with resistant or recurrent disease often have unfavorable outcomes. Mantle cell lymphoma (MCL) is a heterogeneous disease, with indolent and aggressive variants, and is generally considered to be incurable with standard cytotoxic chemotherapy.4 The addition of rituximab (R) to the regimen of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) resulted in improved outcomes in patients with DLBCL and the adoption of chemoimmunotherapy as the current standard of care.5-9 Although initial therapy for newly diagnosed MCL is less standardized, treatment based on R-CHOP is commonly used for initial treatment.10, 11 Although benefits have been noted with the addition of R to CHOP, resistance is common and more effective strategies are necessary to improve the outcomes of patients with DLBCL and MCL. Incorporation of new agents capable of targeting relevant biological pathways may lead to improved outcomes.

The proteasome inhibitor bortezomib impacts on many cellular processes relevant to the pathogenesis and treatment of DLBCL and MCL, including accumulation of cell cycle regulatory proteins (eg, cyclin-dependent kinase inhibitors), alteration in the balance of bcl-2 family members, accumulation of p53, and inhibition of nuclear factor-kappa B (NF-κB).12, 13 The NF-κB pathway is believed to play a key role in the pathogenesis of the activated B-cell–like (ABC) subtype of DLBCL, which carries a poor prognosis compared with germinal center B-cell like (GCB), thereby providing rationale for the use of bortezomib in these patients.14-16 In MCL cell lines and patient specimens, bortezomib induced apoptosis by down-regulating Bcl-XL and bfl/A1 and inducing degradation bcl-2 and inhibition of cyclin D1 expression.16-19

Bortezomib, alone and in combination with R or dexamethasone, has demonstrated efficacy in patients with MCL20-26 and is approved in the United States as a single agent for the treatment of MCL in patients who have received at least 1 prior therapy. It is interesting to note that the activity of bortezomib in MCL is independent of responsiveness to prior chemotherapy.27 Although bortezomib demonstrated minimal efficacy as a single agent in patients with recurrent and refractory DLBCL, when combined with dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin (DA-EPOCH), no significant increase in toxicities were noted, suggesting that it could be safely used in combination with chemotherapy.28

We hypothesized that the addition of bortezomib to R-CHOP would be well tolerated and effective in patients with untreated DLBCL and MCL.16 Because of the potentially overlapping toxicities of individual agents used in this combination therapy, particularly neuropathy and cytopenias, we first conducted a phase 1 study using standard R-CHOP with a dose escalation of bortezomib in patients with untreated DLBCL and MCL.



Patients with histologically confirmed CD20+ DLBCL or MCL who had not received prior therapy were eligible for enrollment. Other inclusion criteria included age >18 years; measurable stage II (inappropriate for radiotherapy), III, or IV disease; Karnofsky performance status ≥50%; absolute granulocyte count ≥1000 cells/mm3; platelet count ≥50,000 cells/mm3; and creatinine and total bilirubin ≤2 times the upper limit of normal. Patients with known central nervous system involvement by lymphoma or pre-existing peripheral neuropathy > grade 2 were excluded. The trial was conducted in accordance with International Conference on Harmonization Good Clinical Practice Guidelines and the Declaration of Helsinki after approval by the sites' Institutional Review Boards. All patients provided written informed consent.

Study Design and Drug Administration

The primary objective was to determine the maximum tolerated dose (MTD) and toxicity profile of bortezomib administered in combination with R-CHOP to patients with untreated DBLCL or MCL. Secondary objectives included assessment of the overall response (OR) and complete response (CR) rates, progression-free survival (PFS), and overall survival (OS). Patients received standard, full-dose CHOP chemotherapy comprised of cyclophosphamide at a dose of 750 mg/m2 intravenously (iv) on Day 1, doxorubicin at a dose of 50 mg/m2 iv on Day 1, vincristine at a dose of 1.4 mg/m2 iv (not to exceed 2.0 mg total) on Day 1, and prednisone at a dose of 100 mg orally on Days 1 through 5 every 21 days for 6 cycles. R at a dose of 375 mg/m2 with standard premedication (acetaminophen and diphenhydramine) was administered before CHOP on Day 1 of each cycle. One of 3 dose levels of bortezomib (0.7 mg/m2, 1.0 mg/m2, and 1.3 mg/m2) was administered on Days 1 and 4 of each cycle. Day 1 bortezomib was administered before the R-CHOP. This schedule was chosen to allow for proteosome inhibition at the time of chemotherapy administration. Dose escalation occurred according to the time-to-event continual reassessment method (TITE-CRM) as patients entered the study.29, 30 When a new patient was registered on to study, a vital summary for the currently enrolled patients was reported to the statistician who estimated the MTD based on currently available observations. The newly enrolled patient was then treated at the updated MTD.

The MTD was defined as the dose that caused a dose-limiting toxicity (DLT) to occur in 25% of patients during the first 6 cycles of treatment. A DLT was defined as a grade 3 motor neuropathy, grade 3 neuropathic pain, or other grade 4 neurologic toxicity; platelet count <10,000/mm3 (unless due to underlying bone marrow involvement by lymphoma); grade 3 or greater nonhematologic or non-neurologic toxicity; or requirement of >2 dose reductions in doxorubicin, cyclophosphamide, or bortezomib, or treatment delays beyond 4 weeks.

Dose modifications for toxicity were allowed. For neurotoxicity, bortezomib doses were adjusted according to recommendations in the prescribing information. For vincristine, doses were reduced by 25% for grade 2 motor and grade 3 sensory neuropathy. For grade 4 thrombocytopenia (but not <10,000 cells/mm3, which would be DLT), bortezomib doses were reduced by 25% (to a minimum of 0.7 mg/m2), and cyclophosphamide and doxorubicin doses were reduced by 20%. Platelet counts were required to return to ≥50,000 cells/mm3 before the next cycle of therapy. Patients unable to be treated by Day 49 (a 4-week delay) were removed from the study and considered to have a DLT. In patients with febrile neutropenia or an absolute neutrophil count of <500 cells/mm3 for >7 days, the dose of bortezomib was reduced by 25%, and cyclophosphamide and doxorubicin doses were reduced by 20%. Supportive care, including hematopoietic growth factors, antibiotics, and other medications, could be used at the discretion of the treating physician.

Pretreatment and Safety Assessments

Tumor tissue was collected to confirm the diagnosis and to test for CD20 expression and other characteristic phenotypes. Pretreatment assessments included patient history, physical examination with vital signs and performance status assessment, serum chemistries including lactate dehydrogenase, and complete blood counts (CBC) with differential and platelets. These, along with toxicity assessments, were repeated on Day 1 of each cycle and at the completion of treatment. CBC with differential and platelets were repeated on Days 4 and 10 or 11 of each cycle. All toxicities were graded using the Common Terminology Criteria for Adverse Events (CTCAE, version 3.0).31 Computed tomography (CT) scans of the neck, chest, abdomen, and pelvis and a bone marrow aspirate and core needle biopsy were performed at baseline. CT scan was repeated after the third and sixth cycles of chemotherapy. The bone marrow aspirate and core needle biopsy were repeated after Cycles 3 or 6 if the patient was otherwise in a CR, and were not repeated once negative. Patients were evaluated with physical examination, laboratory assessments, and CT scan every 3 months for the first year and then every 6 months until disease progression or death until Year 5 after treatment.

Response Criteria

Patients were monitored for response using criteria modified from those recommended by the National Cancer Institute (NCI)-sponsored international working group.32 Briefly, CR was defined as the complete disappearance of all detectable clinical and radiographic evidence of disease and related symptoms: all lymph nodes and lymph node masses regressed to normal (<1.5 cm) or a >75% decrease in the sum of the products of their greatest transverse diameters (SPD) if between 1.1 and 1.5 cm at baseline, normalization of serum chemistries, and normal bone marrow evaluation. CR/unconfirmed (CRu) was defined as meeting the criteria of CR except for a residual lymph node measuring >1.5 cm in greatest transverse diameter that regressed by >75% in the SPD or individual lymph nodes that were previously confluent and regressed by >75% in their SPD, or indeterminate bone marrow status. A partial response (PR) was defined as a decrease of >50% in the SPD of the 6 largest lymph node masses; normal or a decrease in the size of other lymph nodes, liver, or spleen; and no new sites of disease. Progressive disease (PD) was defined as a >50% increase in the SPD from nadir of any previously identified abnormal lymph node or the appearance of a new lesion. Stable disease was defined as a response insufficient to qualify as a PR nor sufficient disease progression to qualify as PD.

Statistical Analyses

For the phase 1 dose escalation of bortezomib, the TITE-CRM with an empirical dose-toxicity model33 and a linear weight function29 was used with a sample size of 18. This model is calibrated to eventually select a dose that results in DLTs in 20% to 30% of patients.30 The TITE-CRM makes use of each patient's time-to-toxicity as partial information even before a complete follow-up has occurred. The sample size of 18 ensures that estimates of any binary variable will have a 95% confidence interval of width <0.24.


Patients and Dose Escalation

Twenty patients were enrolled in the phase 1 portion of the study. Patient and disease characteristics are listed in Table 1. Four patients received the 0.7-mg/m2 bortezomib dose, 9 patients received the 1.0-mg/m2 bortezomib dose, and 7 patients received the 1.3-mg/m2 bortezomib dose on Days 1 and 4 of each cycle.

Table 1. Patient and Disease Characteristics (n = 20)
ParameterNo. %
  1. LDH indicates lactate dehydrogenase.

Age, y   
 Median 66 
 Range 29–84 
 Male/female12/8 60/40
 Diffuse large B-cell lymphoma16 80
 Mantle cell lymphoma4 20
Stage of disease   
 II2 10
 III4 20
 IV14 70
 LDH >normal11 55
International Prognostic Index   
 0-13 15
 27 35
 37 35
 4-53 15

Treatment-Emergent Toxicities

Seventeen patients (85%) completed all 6 cycles of treatment according to the protocol. Major adverse events are listed in Table 2. One DLT was reported in a patient with MCL receiving bortezomib at a dose of 1.0 mg/m2. The patient was admitted to hospital with dyspnea and found to have significant bilateral pleural effusions and bilateral ground glass opacities on CT imaging. Upon further investigation, the effusions were found to be chylous. The patient was taken off study and was treated with alternate therapy with subsequent resolution of the effusions and radiologic findings. No other DLTs were observed. Two patients were not fully evaluable for DLT, although toxicity is reported here. One patient receiving the 0.7-mg/m2 bortezomib dose withdrew from the study after 3 cycles of treatment while in response due to patient preference. This patient received no further lymphoma therapy and died of unknown causes 15 months after study entry. One patient receiving the 1.0-mg/m2 bortezomib dose was considered inevaluable due to protocol deviation. After the fourth cycle of treatment, he developed low-grade fevers. The fifth cycle was administered with a 50% dose reduction of bortezomib, cyclophosphamide, and doxorubicin. One week later, he was admitted to the hospital with worsening fever and headache. Lumbar puncture revealed an elevated white blood cell count but no evidence of lymphoma. Molecular studies performed on the cerebral spinal fluid were positive for Borreliaburgdorferi, and antibiotics were initiated with rapid resolution of the symptoms. This infection and related symptoms, including neuropathy (given that the patient lived in an endemic area) were not judged to be bortezomib therapy. He subsequently received a sixth cycle of treatment with full doses of cytotoxic chemotherapy but no bortezomib. Although the protocol specified 6 cycles of therapy, he received 2 additional cycles due to his treatment modifications. The seventh cycle was administered with a 25% dose reduction of bortezomib and vincristine due to grade 2 motor and sensory neuropathy, after which his sensory neuropathy increased to grade 3. He received the eighth cycle of treatment with no vincristine or bortezomib. One other patient had a dose reduction of cyclophosphamide, doxorubicin, and bortezomib according to protocol for thrombocytopenia. The prospectively planned highest dose level to be assessed in the study was bortezomib at a dose of 1.3 mg/m2 on Days 1 and 4 of each cycle, which was filled without a DLT being observed.

Table 2. Number of Patients Experiencing Grade 3 to 4 Nonhematologic Toxicitya
ToxicityBortezomib Dose, mg/m2
0.7 (n = 4)1.0 (n = 9)1.3 (n = 7)
  • CHF indicates congestive heart failure; DVT, deep vein thrombosis.

  • a

    All toxicities were graded using the Common Terminology Criteria for Adverse Events (CTCAE; version 3.0).

Pleural effusion010
Febrile neutropenia110
Secondary malignancy010

Myelosuppression was the most common grade 4 toxicity, occurring in 7 (35%) patients, and included thrombocytopenia (n = 3 patients) and neutropenia (n = 4 patients) (Table 3). Two episodes of febrile neutropenia were reported. Thirteen patients (65%) reported neuropathy of any grade. Of these, 9 (45%) had grade 1, 3 (15%) had grade 2, and 1 (5%) had grade 3 neuropathy. The grade 3 neuropathy occurred at the 1.0-mg/m2 bortezomib dose level. There were no incidents of grade 4 neuropathy reported (Table 4). Neuropathy tended to occur by the third cycle.

Table 3. Number of Patients Experiencing Grade 3 to 4 Hematologic Toxicitya
ToxicityBortezomib Dose, mg/m2
0.7 (n = 4)1.0 (n = 9)1.3 (n = 7)
  • a

    All toxicities were graded using the Common Terminology Criteria for Adverse Events (CTCAE, version 3.0).

 Grade 3010
 Grade 4000
 Grade 3000
 Grade 4012
 Grade 3001
 Grade 4121
Table 4. Maximum Neuropathy Grade Experienced, Shown by Patient Cohorta
Neuropathy GradeaBortezomib Dose, mg/m2
0.7 (n = 4)1.0 (n = 9)1.3 (n = 7)
  • a

    All toxicities were graded using the Common Terminology Criteria for Adverse Events (CTCAE; version 3.0).

Grade 1135
Grade 2201
Grade 3010
Grade 4000


At a median follow-up of 56 months (range, 15 to 68+ months), 11 patients were alive and in remission. The overall response rate on an intent-to-treat basis was 95%, including 13 CRs (65%), 5 cases of CRu (25%), and 1 PR (5%), with 1 patient being inevaluable for response secondary to meningitis resulting from Lyme disease (Table 5). The 4-year OS and PFS rates for all patients were 75% and 58% respectively (Fig. 1). The 4-year OS rate for patients with DLBCL was 81%.

Figure 1.

Overall and progression-free survival are shown for all patients in the current study (N = 20).

Table 5. Response by Disease and Dose Level
 Bortezomib Dose, mg/m2
 0.7 (n = 4)1.0 (n = 9)1.3 (n = 7)
  • DLBCL indicates diffuse large B-cell lymphoma; CR, complete response; CRu, complete response unconfirmed; PR, partial response; MCL, mantle cell lymphoma.

  • a

    One patient with MCL was not evaluated for response.

DLBCL (n = 16)   
MCL (n = 4)a   


The results of this phase 1, dose escalation study demonstrated that bortezomib at doses up to 1.3 mg/m2 on Days 1 and 4 of a 21-day cycle may be combined with R-CHOP chemotherapy with acceptable toxicity in patients with untreated DLBCL or MCL. The combination of R-CHOP and bortezomib was well tolerated and adverse events were predictable and manageable. Most notably, there was no apparent increase in either the incidence or severity of neuropathy or thrombocytopenia, adverse events most likely given the overlapping toxicities of the agents used. There was only 1 DLT experienced, which was a grade 3 pleural effusion. Therefore, the data from the current study demonstrate that the regimen is appropriate for further study.

The 1.3-mg/m2 dose of bortezomib is predicted to achieve a maximum inhibition of the 26S proteasome of 65% with return of sufficient proteasome activity by 72 hours.34, 35 Theoretically, it is preferable to have the proteasome inhibited at the time of chemotherapy to overcome chemoresistance. However, prolonged proteasome inhibition is likely to result in increased toxicity. For these reasons, it is relevant that the 1.3-mg/m2 dose was reached using the Days 1 and 4 schedule without excessive toxicity.

Although this phase 1 experience included only 20 patients, R-CHOP plus bortezomib demonstrated encouraging efficacy, with an overall response rate of 95% at a median follow-up of 56 months, including 90% confirmed and unconfirmed CRs. Safety and efficacy results from this study support an ongoing phase 2 trial. Patients with untreated DLBCL or MCL will receive R-CHOP plus bortezomib at a dose of 1.3 mg/m2 on Days 1 and 4 of a 21-day cycle with PFS as the primary endpoint. The subtype of DLBCL (GCB vs non-GCB) will be evaluated by immunohistochemistry using CD10, bcl-6, and multiple myeloma oncogene-1(MUM1) expression.36 Additional prognostic markers, including FOXP1, cyclin D2, and bcl-2 will be assessed.15, 37 This information may help determine whether all patients with DLBCL, or only a subset (ie, the ABC subtype), might derive benefit from the addition of bortezomib to R-CHOP.

Several studies using the combination of bortezomib plus chemotherapy have been undertaken or are ongoing. These trials have met variable degrees of success, with complications in some cases arising due to overlapping neurotoxicity from vincristine and bortezomib. In patients with previously treated DLBCL, bortezomib was administered on Days 1 and 4 in combination with DA-EPOCH.28 In the phase 1 portion of the study, the MTD of bortezomib was 1.7 mg/m2. However, after multiple episodes of neuropathy resulting in discontinuation of bortezomib, the remainder of the phase 2 study was conducted at a dose of 1.5 mg/m2. As hypothesized, the regimen benefited those patients with the ABC subtype to a greater degree than those patients with the GCB subtype. In 11 patients with untreated indolent B-cell lymphoma, bortezomib doses of up to 1.6 mg/m2 were added on Days 1 and 8 to R-CHOP with vincristine capped at a dose of 1.5 mg.38 Treatment was well tolerated, with only 1 episode of grade 3 neuropathy reported. The same regimen was administered to 13 patients with aggressive T-cell lymphoma.39 The MTD of bortezomib was not reached at a dose of 1.6 mg/m2, despite not capping the dose of vincristine. In a randomized phase 2 study from the French Adult Lymphoma Study Group (GELA), Ribrag et al tested 2 schedules of bortezomib that were different from this report in combination with R-CHOP in patients with B-cell lymphoma: Arm A (Days 1, 4, 8, and 11 at doses of 1.0 mg/m2 or 1.3 mg/m2) and Arm B (Days 1 and 8 at doses of 1.3 mg/m2 or 1.6 mg/m2).40 With the dose of vincristine capped at 2 mg, grade 3 neuropathy was reported in 5 of 7 patients treated in Arm A at a dose of 1.3 mg/m2, and 4 of 15 patients treated in Arm B at a dose of 1.6 mg/m2. The CR rate was slightly higher in the biweekely group (90% vs 79%). In a feasibility study of 30 patients with mantle cell lymphoma treated with bortezomib, rituximab, cyclophosphamide, doxorubicin, vincristine, and dexamethasone (VcR-CVAD).41 A high incidence of painful neuropathy resulted with 1.5 mg/m2 of bortezomib on Days 1 and 4 and up to 2 mg of vincristine. Once the doses of bortezomib were capped at 1.3 mg/m2 and vincristine at 1 mg, only 1 of 16 patients experienced grade 3 neuropathy. In a phase 2 study from Canada that was also presented as an abstract, 95 patients with newly diagnosed follicular lymphoma were treated with rituximab, cyclophosphamide, vincristine (capped at 2 mg), prednisone, and bortezomib (1.3 mg/m2 Days 1 and 8).42 Only 6 patients developed grade 3 neurotoxicity, with no patients developing grade 4 neurotoxicity. Taken in conjunction, it is challenging to determine from these data the optimal dose or schedule of bortezomib when combined with vincristine-containing regimens.

The Southwest Oncology Group (SWOG) is conducting an ongoing study of R-CHOP + bortezomib (1.3 mg/m2 on Days 1 and 4 of each cycle) with bortezomib maintenance in patients with untreated MCL ( identifier: NCT00376961), a regimen identical to that used in the current report. The Eastern Cooperative Oncology Group (ECOG) is evaluating the VcR-CVAD regimen in patients with untreated MCL ( Identifier: NCT00433537). Millennium Pharmaceuticals, the manufacturer of bortezomib, recently initiated 2 randomized phase 3 trials of R-CHOP chemotherapy compared with R-CHOP chemotherapy with bortezomib substituted for vincristine for patients with untreated MCL ( identifier: NCT00722137), and R-CHOP versus R-CHOP plus bortezomib (in the dose and schedule used in the current study) for patients with non-GCB DLBCL ( identifier: NCT00931918). These studies will more definitively assess the potential value of the combination of proteasome inhibition with chemoimmunotherapy in patients with lymphoma.


We thank Lynne Lederman, PhD, for assistance in preparation of the article


Supported in part by Millennium Pharmaceuticals, Inc, and grants from the Lymphoma Research Foundation (to J.P.L.) and the Lymphoma Foundation (to J.P.L.)