Report from the FDA

Bortezomib for the Treatment of Mantle Cell Lymphoma

Robert C. Kane, Ramzi Dagher, Ann Farrell, Chia-Wen Ko, Rajeshwari Sridhara,
Robert Justice, and Richard Pazdur

Abstract Purpose: To describe the Food and Drug Administration review and marketing approval
considerations for bortezomib (Velcade) for the treatment of patients with mantle cell lymphoma.
Experimental Design: Food and Drug Administration reviewed a multicenter study of
bortezomib in 155 patients with progressive mantle cell lymphoma after at least one prior therapy.
Results: Seventy-seven percent were stage IV, and 75% had one or more extranodal sites
of disease. Prior therapy included an anthracycline or mitoxantrone, cyclophosphamide, and
rituximab. Median age was 65 years. All received bortezomib 1.3 mg/m2 i.v. on days 1, 4, 8, and
11 of each 3-week cycle. The primary end point was response. Response and progression were
determined by independent review of serial computed tomography scans using International
Lymphoma Workshop Response Criteria. The overall response rate was 31%, including complete
response (CR) plus CR unconfirmed (CRu) plus partial response; median response duration
was 9.3 months. The CR plus CRu response rate was 8% with a median duration of 15.4 months.
Adverse events were similar to those observed previously for bortezomib. The most commonly
reported treatment-emergent adverse events were asthenia (72%), peripheral neuropathies
(55%), constipation (50%), diarrhea (47%), nausea (44%), and anorexia (39%). The most
common adverse event leading to discontinuation was neuropathy.
Conclusions: Bortezomib received regular approval for the treatment of patients with mantle cell
lymphoma in relapse after prior therapy.

On December 8, 2006, bortezomib (Velcade for injection) proteolysis, which then may alter multiple signaling cascades
received marketing approval by the U.S. Food and Drug within the cell.
Administration (FDA) for the treatment of patients with mantle Bortezomib had received accelerated approval in 2003 for
cell lymphoma who have received at least one prior therapy multiple myeloma after two prior therapies and in 2005
for their disease. This drug approval is the first for the specific received regular approval for the treatment of multiple
indication of mantle cell lymphoma, a subtype of non- myeloma after one prior therapy (2). The sponsor, Millennium
Hodgkin’s lymphoma characterized by unique pathologic, Pharmaceuticals, Inc., in conjunction with Johnson and
cytogenetic, and clinical features (1). Johnson Research and Pharmaceutical Development, met with
Bortezomib, a modified dipeptidyl boronic acid derived from FDA during the design of this registration study and during the
leucine and phenylalanine, is a reversible inhibitor of the conduct and analysis phases to assure agreement on the study
chymotrypsin-like activity of the 26S proteasome, a large design, end points, adjudication plan for assessing the end
protein complex that degrades most intracellular proteins. The points, and the analysis plan intended to show efficacy and
ubiquitin-proteasome pathway plays an essential role in safety necessary for drug approval. The regular marketing
regulating the intracellular concentration of specific proteins. approval for mantle cell lymphoma (after prior therapy) was
Inhibition of the 26S proteasome prevents this targeted based on the results of the FDA review of the sponsor’s single,
multicenter study. This report summarizes the FDA analysis and
basis for approval.

Authors’ Affiliation: Division of Drug Oncology Products, Office of Oncology Materials and Methods
Drug Products, Center for Drug Evaluation and Research, U.S. Food and Drug
Administration, Silver Spring, Maryland The sponsor conducted a single-arm, single-agent prospective study
Received 4/12/07; accepted 6/29/07. of bortezomib in 155 patients with relapsed, progressive mantle cell
Note: This work done under U.S. government auspices is not subject to copyright. lymphoma following one or two prior therapies among 35 centers in
The publisher or recipient acknowledges right of the U.S. government to retain a
North America and Europe. The initial proposal was for a single-arm
nonexclusive, royalty-free license in and to any copyright covering the article.
Requests for reprints: Robert C. Kane, Division of Drug Oncology Products,
phase 2 study to assess time-to-progression, overall response rate, and
Center for Drug Evaluation and Research, U.S. Food and Drug Administration, response duration for previously treated mantle cell lymphoma
Room 2109, Building 22, 10903 New Hampshire Avenue, Silver Spring, MD patients. The sample size initially was chosen to provide 80% power,
20993-0002. Phone: 301-796-1384; E-mail: robert.kane@ fda.hhs.gov. using a two-sided a of 0.05, to show a 50% improvement in time to
doi:10.1158/1078-0432.CCR-07-0871 progression when compared with a historical control group. The

www.aacrjournals.org 5291 Clin Cancer Res 2007;13(18) September 15, 2007

Report from the FDA

primary analysis population was all patients who received at least one transplant. Baseline patient and disease characteristics are
dose of drug. FDA cautioned that interpreting time-to-progression in a summarized in Tables 1 and 2.
single-arm study would be problematic due to variables such as patient FDA reviewed the sponsor’s data and independently analyzed
selection, natural history, and evolving concurrent therapies and response and progression. The response rate was determined
supportive measures. The sponsor agreed prospectively that the primary
using the entire enrolled population (155 patients) rather than
end point for the supplemental New Drug Application submission was
overall response rate, as defined by the 1999 International Lymphoma
some other response-evaluable subgroup as the denominator.
Workshop Response Criteria (IWRC; ref. 3) to include the proportion of The results were calculated using the algorithm-determined
patients achieving complete response (CR) plus CR unconfirmed (CRu) response and duration of response in all patients except for one
plus partial response (PR). In accord with these 1999 criteria, no patient in whom only clinical tumor measurements (palpable
positron emission tomography or isotope scans were used for disease neck lymph nodes) were available. Protocol deviations and
assessment. Duration of response was calculated from the date of initial violations were judged as minor and not likely to have altered
documentation of first response to the date of progressive disease. the results. Several study sites were audited for comparison of
Tumor assessments for the primary end point were determined by the source documents with the data recorded in the case report
serial tumor measurements from computed tomography (CT) scans. forms. No substantive deficiencies were found.
The sponsor selected an independent radiology review contractor and The efficacy findings are summarized in Table 3. The overall
developed a protocol for the process of blinded scan review. The
response rate was 31% (48 of 155) and the 95% confidence
sponsor also developed an algorithm to apply the IWRC consistently.
interval (95% CI) was 24%, 39%. The median duration of
Response, as defined by the IWRC, is a composite requiring assessment
of all of the following: CT scan results, disease-related symptom response was 9.3 months (95% CI, 5.4, 13.8). For the CR plus
assessment, physical examination of nonscanned areas, disease-related CRu responder group, the response rate was 8% (12 of 155;
biochemistry (e.g., lactate dehydrogenase), and bone marrow evaluation 95% CI, 4%, 13%); the median duration of response was
(if positive at baseline). CR requires not only the complete disappearance 15.4 months (95% CI, 13.4, 15.4). [The 95% CI upper bound is
of disease but also disappearance of all disease-related symptoms. equal to the median duration of response for the CR plus CRu
CRu denotes patients who fulfill the CR criteria except for a residual responders (n = 12) because of the high percentage of censoring
lymph node mass (and >75% shrinkage) or indeterminate bone marrow. (8 of the 12 responders did not experience disease progression
PR required a z50% decrease in the sum of the perpendicular diameters or death)]. Only four events of progression or death were
of the six largest dominant nodes or nodal masses. observed in CR plus CRu responders at 45, 143, 409, and
Protocol eligibility required a pathologic diagnosis of mantle cell
470 days since their first response.
lymphoma, including expression of cyclin D1 or evidence of t(11;14)
Adverse events were generally similar to those previously
by cytogenetics, fluorescence in situ hybridization, or PCR; an
independent pathology review was also planned to verify the pathology described in myeloma studies and are described in the product
findings. In addition to adequate performance status (Karnofsky score, label. Adverse events resulting in drug discontinuation occurred
z50%), patients were required to have measurable (or evaluable) in 30% of the patients and were consistent with prior Velcade
disease with documented relapse or progression following first- or experience. The most common adverse event associated with
second-line treatment including an anthracycline or mitoxantrone, drug discontinuation was neuropathy.
cyclophosphamide, and rituximab. No prior bortezomib therapy,
recent radiation, or recent radioisotope therapy was allowed. Discussion
All patients received bortezomib 1.3 mg/m2 i.v. bolus on days 1, 4, 8,
and 11 every 21 days for up to 12 months. Prophylactic use of leukocyte The study population seems representative of a population
growth factors was proscribed. Tumor assessments by CT and clinical of patients with mantle cell lymphoma treated with optimal
examinations were done every 6 weeks through week 18 and then every
12 weeks. Dose adjustments and interruptions followed the schedule in
the FDA-approved label.1 Table 1. Baseline demographic characteristics in
155 mantle cell lymphoma patients
Results N = 155

Among the 155 mantle cell lymphoma patients who were Age (y)
enrolled and received at least one dose of bortezomib, 89% Mean (SD) 64.9 (9.3)
Median 65.0
were confirmed as fulfilling diagnostic criteria for mantle cell
Minimum, maximum 42, 89
lymphoma by independent pathology review. This high level of Sex, n (%)
pathology concordance was expected and is deemed satisfacto- Male 125 (81)
ry. The median time from original diagnosis of mantle cell Female 30 (19)
lymphoma to the start of bortezomib was 2.3 years, the median Race, n (%)
White 142 (92)
patient age was 65 years, 77% were stage IV, and 55% had Black 6 (4)
positive bone marrows at entry. Hispanic 4 (3)
With regard to prior therapy, all patients had received at least Asian or Pacific Islander 3 (2)
one prior regimen, and 91% had received all three agents: KPS, n (%)
<50 0
anthracycline, cyclophosphamide, and rituximab. In addition,
50-60 7 (5)
37% had received prior high-intensity chemotherapy including 70-80 37 (24)
hyper-CVAD, ICE, ESHAP, or DHAP with or without stem cell 90-100 109 (71)
Missing 2

Abbreviation: KPS, Karnofsky performance score.

1
http://www.fda.gov/cder/foi/label/2006/021991lbl.pdf

Clin Cancer Res 2007;13(18) September 15, 2007 5292 www.aacrjournals.org

 Bortezomib: for Mantle Cell Lymphoma

Minor modifications to the IWRC definitions were made. The

Table 2. Baseline disease characteristics in 155 lymph node dimensions had to exceed 1.0 cm to be considered
mantle cell lymphoma patients measurable by CT, and the minimum increment in dimension
had to be at least 0.5 cm to qualify for assessment of change.
Prognostic factor N = 155
For PR, response assessment criteria were changed to allow
Time since initial diagnosis to first dose (y) inclusion of sites of dominant extranodal disease (still requiring
Mean (SD) 2.7 (1.9)
Median 2.3
a 50% or greater decrease in the sum of the perpendicular
Minimum, maximum 0.2, 11.2 diameters of measurable sites of disease) because extranodal
Diagnosed <3 y before first dose, n (%) 103 (66) disease is common in mantle cell lymphoma. The sponsor
Diagnosed z3 y before first dose, n (%) 52 (34) presented these findings, and FDA concurred with the changes.
MCL stage at screening, n (%)/n (%)
These changes may be of interest to others using the IWRC
Stage I/II 5 (3)/7 (5)
Stage III/IV 24 (15)/119 (77) measurement procedures for response and progression.
IPI score For this group, single-agent bortezomib therapy resulted in a
0-1 34 (23) 31% overall response rate with a median duration of response
2 48 (33) of 9.2 months. For the 12 CR plus CRu patients, the response
3 48 (33)
4-5 17 (12)
rate was 8%, the median duration of response for this group
Missing 8 was 15.4 months, and the Kaplan-Meier estimate of 1-year
LDH, n (%)/n (%) survival was 100%. No unexpected frequency or severity of
Normal/elevated 95 (64)/54 (36) adverse events was observed.
Missing 6
This experience in a different population, patients with
No. involved extranodal sites, n (%)/n (%)
0/1 38 (25)/64 (41) mantle cell lymphoma, provided further safety information for
2/z3 32 (21)/21 (14) bortezomib beyond the myeloma experience. Interstitial lung
Bone marrow evaluation (biopsy and/or aspirate) disease, which has been reported rarely in myeloma patients
Positive results 84 (55) (4, 5), was not observed during this study. Fluid retention
Negative/indeterminate 70 (45)
No. prior lines of therapy
syndromes, including pleural effusion, ascites, edema, conges-
1/2 84 (54)/ 65 (42) tive heart failure, and respiratory distress, have occurred at low
z3 6 (4) frequency and are described in the current product label.
Received prior regimen containing the following agents Ongoing clinical attention to hydration status is important for
Anthracycline/mitoxantrone 152 (98)
the safe use of bortezomib because adverse events have
Alkylating agents 150 (97)
Rituximab 149 (96) included both dehydration and fluid retention syndromes.
Received at least 2 of the above 3 155 (100) The sponsor’s study is the primary basis for the approval
Received all of the above 3 141 (91) action. During the conduct of this study, several other clinical
Received prior high-intensity therapy* 58 (37) investigations were reported (6 – 8). Although the additional
Received prior high-intensity therapy 47 (30)
as last prior regimen*
studies were not formally reviewed for this supplemental New
Drug Application, the findings are from multiple independent
sources and seem consistent with the Millennium results.
Abbreviations: MCL, mantle cell lymphoma; IPI, International
Prognostic Index; LDH, lactate dehydrogenase.
Mantle cell lymphoma is recognized as an uncommon and
*High-intensity prior regimen defined as hyper-CVAD, R-hyper- distinct category of lymphoma based on several unique
CVAD, ICE/ESHAP/DHAP with or without rituximab, and stem cell characteristics, including pathology, cytogenetics, biomarkers
transplant. (including immunohistochemistry), and clinical behavior.
Despite high initial response rates, most patients with mantle
cell lymphoma relapse and median survival from diagnosis is
front-line combination chemotherapy and without available 3 to 4 years. For a well-characterized population of relapsed
therapy of likely benefit. The study patients are probably or refractory mantle cell lymphoma patients such as those
healthier and slightly younger than a community-based patient enrolled here, FDA agreed that alternative therapy of general
sample; this is expected by the study design and referral center benefit is not available. A randomized study in a population
nature of the study sites. In this group, with median age 65 years,
77% stage IV, 75% with extranodal disease sites, and 37% with
prior ‘‘high-intensity’’ therapy, such as stem cell transplant or Table 3. FDA response analysis in 155 mantle cell
hyper-CVAD, historical experience indicates that few patients lymphoma patients
would be expected to obtain benefit from further therapies.
In designing this study, the sponsor sought to apply the Response analyses (N = 155) n (%) 95% CI
IWRC lymphoma response and progression consensus criteria
Overall response rate (CR + CRu + PR) 48 (31) (24, 39)
literally using an algorithm. During the initial prospective CR (CR + CRu) 12 (8) (4, 13)
validation of the algorithm, some criteria proved to be too CR 10 (6) (3, 12)
sensitive, and some modifications of the IWRC definitions were CRu 2 (1) (0, 5)
necessary to achieve concurrence with clinical findings. For PR 36 (23) (17, 31)
Duration of response, median (mo)
example, serial measurements of small lymph nodes (<1 cm)
CR + CRu + PR (n = 48) 9.3 (5.4, 13.8)
could vary sufficiently between CT assessments to suggest prog- CR + CRu (n = 12) 15.4 (13.4, 15.4)
ression mathematically (a 50% increment from nadir), although PR (n = 36) 6.1 (4.2, 9.3)
there was no evidence clinically that progression was occurring.

www.aacrjournals.org 5293 Clin Cancer Res 2007;13(18) September 15, 2007

Report from the FDA

receiving second-line therapy for mantle cell lymphoma is reflecting the treatment effect on the entire study population
challenging to conduct considering the low frequency of the (such as time-to-progression or progression-free survival). For
disease, extensive yet varied prior therapy, comorbidities in an this reason, such time-to-event end points are more informative
older age population, and usual rapid progression of recurrent than the evaluation of a responder subgroup, assuming that the
mantle cell lymphoma. When a disease process is known to criteria defining progression are explicit and amenable to
proceed rapidly in the absence of effective therapy, a response careful measurement.
end point of sufficient magnitude and duration for a single- FDA reviewed the patients’ on-study characteristics, the study
agent treatment may be clinically meaningful and sufficient conduct, the response and progression determinations by the
evidence for an approval. investigators and the independent reviewers, and the tolerance
In this setting, a single-arm, single-agent study showing a to bortezomib treatment. The bortezomib responses are
clinically meaningful response rate, with durability, with convincing for their durability and for their reduction of the
evidence of response in visceral sites, with adherence to a disease burden on the responding patients (e.g., adenopathy
generally recognized set of criteria defining the response and and hepatosplenomegaly), and they were verified by indepen-
progression criteria, and with verification by independent dent blinded radiologic review of serial CT scans. Toxicity is
blinded review of the tumor assessment end point, could show similar to that observed in the myeloma setting, is already well
substantial evidence of efficacy and safety for bortezomib. In described in the existing label, and is familiar to hematology-
addition, safety could consist of evidence that the adverse oncology physicians. FDA judged the results sufficient to
events, toxicity, and dose adjustments of bortezomib remained conclude that clinical benefit was shown in this advanced
broadly similar to the findings already established in myeloma. disease state. Quality clinical science with compelling evidence
Where alternative therapy is available and perceived to be of allows the FDA to act promptly. Full prescribing information,
benefit, an active control comparator arm should be chosen, including clinical trial information, safety, dosing, drug-drug
and the design should show superiority on an end point interactions, and contraindications, is available online.1

References
1. Harris NL, Jaffe ES, Stein H, et al. A revised European- ria for non-Hodgkin’s lymphomas. J Clin Oncol 1999; Velcade in mantle cell lymphoma [abstract 608].
American classification of lymphoid neoplasms: a pro- 17:1244 ^ 53. Blood 2004;104:175a.
posal from the International Lymphoma Study Group. 4. Miyakoshi S, Kami M,Yuji K, et al. Severe pulmonary 7. Goy A, Younes A, McLaughlin P, et al. Phase II study
Blood 1994;84:1361 ^ 92. complications in Japanese patients after bortezomib of proteasome inhibitor bortezomib in relapsed or re-
2. Kane RC, Farrell AT, Sridhara R, Pazdur R. United treatment for refractory multiple myeloma. Blood fractory B-cell non-Hodgkin’s lymphoma. J Clin Oncol
States Food and Drug Administration approval sum- 2006;107:3492 ^ 4. 2005;23:667 ^ 75.
mary: bortezomib for the treatment of progressive 5. Boyer JE, Batra R, Ascensao JL, Schechter GP. 8. O’Connor OA, Wright J, Moskowitz C, et al. Phase II
multiple myeloma after one prior therapy. Clin Cancer Severe pulmonary complication after bortezomib clinical experience with the novel proteasome inhibitor
Res 2006;12:2955 ^ 60. treatment for multiple myeloma. Blood 2006; bortezomib in patients with indolent non-Hodgkin’s
3. Cheson BD, Horning SJ, Coiffier B, et al. Report of an 108:1113. lymphoma and mantle cell lymphoma. J Clin Oncol
international workshop to standardize response crite- 6. Belch A, KouroukisT, Crump M, et al. Phase II trial of 2005;23:676 ^ 84.

Clin Cancer Res 2007;13(18) September 15, 2007 5294 www.aacrjournals.org