Articles

Lenalidomide in combination with rituximab for patients

with relapsed or refractory mantle-cell lymphoma:
a phase 1/2 clinical trial
Michael Wang, Luis Fayad, Nicolaus Wagner-Bartak, Liang Zhang, Fredrick Hagemeister, Sattva S Neelapu, Felipe Samaniego, Peter McLaughlin,
Michelle Fanale, Anas Younes, Fernando Cabanillas, Nathan Fowler, Kate J Newberry, Luhong Sun, Ken H Young, Richard Champlin, Larry Kwak,
Lei Feng, Maria Badillo, Maria Bejarano, Kimberly Hartig, Wendy Chen, Yiming Chen, Catriona Byrne, Neda Bell, Jerome Zeldis, Jorge Romaguera

Summary
Lancet Oncol 2012; 13: 71623 Background The combination of rituximab and lenalidomide has shown promise for the treatment of mantle-cell
Published Online lymphoma (MCL) in preclinical studies. We aimed to identify the maximum tolerated dose (MTD) of lenalidomide
June 6, 2012 when combined with rituximab in a phase 1 trial and to assess the ecacy and safety of this combination in a phase 2
DOI:10.1016/S1470-
trial in patients with relapsed or refractory MCL.
2045(12)70200-0
See Comment page 657
Methods Patients with relapsed or refractory MCL who had received one to four previous lines of treatment were
Department of Lymphoma and
Myeloma (M Wang MD,
enrolled in this single-arm, open-label, phase 1/2 trial at MD Anderson Cancer Center. In phase 1, to identify the
L Fayad MD, L Zhang MD, MTD of lenalidomide, four patient cohorts received escalating doses (10, 15, 20, and 25 mg) of daily oral lenalidomide
Prof F Hagemeister MD, on days 121 of each 28-day cycle. 375 mg/m intravenous rituximab was also administered in four weekly doses
S S Neelapu MD, during cycle 1 only. In phase 2, patients received rituximab plus the MTD of lenalidomide, following the same
F Samaniego MD,
P McLaughlin MD, M Fanale MD,
cycles as for phase 1. Treatment in both phases continued until disease progression, stem-cell transplantation, or
Prof A Younes MD, N Fowler MD, severe toxicity. The primary ecacy endpoint was overall response (complete or partial response). The secondary
K J Newberry PhD, L Sun PhD, ecacy endpoint was survival. We used the Kaplan-Meier method to estimate response duration, progression-free
Prof L Kwak MD, M Badillo BS,
survival, and overall survival. Analysis was by intention to treat. This study is registered with ClinicalTrials.gov,
M Bejarano MD, K Hartig,
W Chen PA, Y Chen MD, number NCT00294632.
Prof J Romaguera MD),
Department of Diagnostic Findings 52 patients were enrolled between Feb 10, 2006 and July 30, 2009, 14 in phase 1 and 44 (including six patients
Radiology
who received the MTD of lenalidomide in the phase 1 portion) in phase 2. The MTD was 20 mg lenalidomide. One
(N Wagner-Bartak MD),
Department of patient who was treated with 25 mg lenalidomide developed a grade 4 non-neutropenic infection and died. In the
Hematopathology phase 2 portion of the study, grade 34 haematological toxicities included neutropenia (29 patients), lymphopenia
(K H Young MD), Department of (16 patients), leucopenia (13 patients), and thrombocytopenia (ten patients). There were only two episodes of febrile
Stem Cell Transplantation and
neutropenia. Among 44 patients in phase 2, 25 (57%) had an overall response: 16 (36%) had a complete response and
Cellular Therapy
(Prof R Champlin MD), and nine (20%) had a partial response. The median response duration was 189 months (95% CI 170 months to not
Department of Biostatistics reached [NR]). The median progression-free survival was 111 months (95% CI 83 to 249 months), and the median
(L Feng MS), The University of overall survival was 243 months (198 months to NR). Five of 14 patients who had received bortezomib treatment
Texas MD Anderson Cancer
before enrolment achieved an overall response.
Center, Houston, TX, USA;
Hospital Auxilio Mutuo,
San Juan, Puerto Rico Interpretation Oral lenalidomide plus rituximab is well tolerated and eective for patients with relapsed or refractory MCL.
(F Cabanillas MD); and Celgene,
Summit, NJ, USA (C Byrne RN,
N Bell BS, J Zeldis MD)
Funding Celgene.
Correspondence to:
Dr Michael Wang, Department of Introduction cell lines and in primary MCL cells to a greater extent
Lymphoma and Myeloma, Mantle-cell lymphoma (MCL) is a distinct subset of B-cell than either drug alone.11 Furthermore, this combination
Unit 429, The University of Texas non-Hodgkin lymphoma characterised by t(11;14) prolonged the survival of MCL-bearing mice in vivo.11
MD Anderson Cancer Center,
chromosomal translocation, which results in over- Lenalidomide increased the antibody-dependent cellular
1515 Holcombe Boulevard,
Houston, TX 77030, USA expression of cyclin D1 and dysregulation of the cell cytotoxicity of rituximab both in vitro and in vivo via a
[email protected] cycle.13 MCL is incurable with the current front-line natural-killer-cell-mediated mechanism.12,13
treatments, and innovative approaches are needed.46 On the basis of these preclinical data in vitro and
The anti-CD20 monoclonal antibody rituximab has in vivo, we hypothesised that the combination of
minimal toxicity and modest monotherapy activity in rituximab and lenalidomide might have more anti-MCL
patients with MCL.7,8 Lenalidomide is an immuno- activity than either drug alone and might prove to be an
modulatory drug that is eective against lymphoma.9,10 eective therapeutic regimen for patients with relapsed
In our laboratory at The University of Texas MD or refractory MCL. To test this hypothesis, we initiated a
Anderson Cancer Center (Houston, TX, USA), single-arm, open-label, phase 1/2 clinical trial to identify
lenalidomide in combination with rituximab inhibited the maximum tolerated dose (MTD) of lenalidomide
cell growth and induced apoptosis in vitro in four MCL when combined with rituximab and to assess the

716 www.thelancet.com/oncology Vol 13 July 2012

 Articles

ecacy and safety of this combination in patients with events were assessed weekly during the rst month of
relapsed or refractory MCL. treatment and twice a month thereafter, on the basis of
the National Cancer Institute common terminology
Methods criteria for adverse events (version 3.0).14
Patients The primary ecacy endpoint was the overall response
Patients with relapsed or refractory MCL who met the (OR) rate (complete response [CR] or partial response
following criteria were eligible for this study: conrmed [PR]). Response to treatment was assessed according to
tissue diagnosis of MCL with CD20 and cyclin D1 guidelines developed by an international workshop on
positivity; had undergone one to four previous lines of lymphoma response criteria,15 with restaging after every
treatment; had received no treatment in the month two cycles by CT scan and bone marrow biopsies. Both a
before study entry; and had an Eastern Cooperative regular radiological assessment and subsequent
Oncology Group performance status score of 2 or less. reassessment by a designated radiologist (NW-B) were
Additional inclusion criteria were a serum creatinine done. The secondary ecacy endpoint was survival.
concentration of 18 mol/L or less, aspartate amino-
transferase and alanine aminotransferase concentrations Statistical analysis
of less than two times the upper limit of normal, a In phase 1, the MTD of lenalidomide when combined
platelet count of at least 5010 cells per L, and an with rituximab was identied by a 3+3 algorithm. In
absolute neutrophil count of at least 110 cells per L. phase 2, the ecacy of the regimen was assessed by
Previous treatment with rituximab or bortezomib was Simons optimal two-stage design.16
permitted, regardless of sensitivity. The OR rate of rituximab-naive relapsed or refractory
This study was approved by the Institutional Review MCL patients treated with rituximab alone ranges from
Board at MD Anderson Cancer Center (Houston, TX, 27% to 53%.7,8,17,18 However, patients enrolled on this study
USA) and was compliant with institutional guidelines and had previously received rituximab. Because there are no
the Declaration of Helsinki. Informed written consent
was obtained from all participants before enrolment.
Phase 1 Phase 2 patients Phase 2 plus phase 1
(n=14) only (n=38) patients (n=44)*
Procedures
Age (years) 72 (5684) 64 (4685) 66 (4685)
We did a single-arm, open-label, phase 1/2 clinical trial to
Men 12 (86%) 35 (92%) 40 (91%)
assess the safety and ecacy of rituximab and
Women 2 (14%) 3 (8%) 4 (9%)
lenalidomide for the treatment of patients with relapsed
Time from diagnosis (months) 32 (1487) 27 (296) 27 (395)
or refractory MCL. In phase 1, patients were sequentially
enrolled in four cohorts to receive escalating daily doses Stage 4 at diagnosis 14 (100%) 38 (100%) 44 (100%)

of 10, 15, 20, and 25 mg of oral lenalidomide by a Bone marrow involvement at study entry 8 (57%) 17 (45%) 22 (50%)

3+3 algorithm. There were no intra-patient dose Duration of last remission (months) 23 (443) 14 (149) 16 (149)
escalations. Patient cohorts received lenalidomide daily Previous lines of therapy 2 (14) 2 (14) 2 (14)
on days 121 of each 28-day cycle; they also received Number of previous therapies
375 mg/m of intravenous rituximab once per week for 1 5 (36%) 15 (39%) 17 (39%)
4 weeks only during cycle 1, beginning on day 1. 2 2 (14%) 13 (34%) 15 (34%)
Patients treated at the MTD of lenalidomide in phase 1 3 4 (29%) 5 (13%) 6 (14%)
were counted as the rst patients enrolled in the phase 2 4 3 (21%) 5 (13%) 6 (14%)
study. In phase 2, patients received daily lenalidomide at Type of previous therapy
the MTD on days 121 of each 28-day cycle and R-HyperCVAD/R-MTX+Ara-C 10 (71%) 34 (89%) 39 (85%)
375 mg/m of intravenous rituximab once per week R-CHOPradiotherapy 7 (50%) 10 (26%) 17 (39%)
for 4 weeks only during cycle 1, beginning on day 1. Rituximab maintenance 6 (43%) 4 (11%) 9 (20%)
Anticoagulants, antithrombotic drugs, and prophylaxis Bortezomib 3 (21%) 11 (29%) 12 (27%)
for tumour lysis syndrome with allopurinol were not Zevalin/17AAG/RICE/IL-6 Ab 4 (29%) 5 (13%) 5 (11%)
required by the protocol. Patients in both phases Autologous BMT 0 (0%) 5 (13%) 5 (11%)
were treated until disease progression, stem-cell trans- Mantle-cell international prognostic index 3 (26) 2 (07) 2 (07)
plantation (SCT), or withdrawal for toxicity. (range)
For dose modications during both phases, in accord- Data are median (range) or number (%). Some percentages do not sum to 100 because of rounding. R-HyperCVAD/
ance with the study design, lenalidomide dose was reduced R-MTX+Ara-C=rituximab-hyper-cyclophosphamide, vincristine, doxorubicin, dexamethasone alternating with
from 20 mg to 15 mg, 10 mg, and 5 mg in a de-escalating rituximab, methotrexate, and cytarabine. R-CHOP=rituximab plus cyclophosphamide, doxorubicin, vincristine, and
prednisone; Zevalin/17AAG/RICE/IL-6 Ab=90Y-ibritumomab tiuxetan, 17-allylaminogeldanamycin, interleukin-6
fashion if there were grade 3 or 4 non-haematological
antibody, and rituximab plus ifosfamide plus carboplatin plus etoposide. BMT=bone marrow transplantation. *Phase 2
toxicities or grade 4 haematological toxicities. included six patients from phase 1 who received 20 mg lenalidomide. Six patients (one in phase 1 and ve in phase 2)
Dose-limiting toxicity (DLT) was dened as a grade 3 or were not included because their disease was refractory to previous treatments.
higher non-haematological event or a grade 4 haema-
Table 1: Demographics and baseline characteristics of patients in phase 1 and phase 2
tological adverse event during the rst cycle. Adverse

www.thelancet.com/oncology Vol 13 July 2012 717

 Articles

published data on OR rates in patients retreated with censored for all response assessments at the time of SCT.
rituximab, we estimated the response rate to be 25% on For the response duration calculation, responders who
the basis of the experience of the faculty at the Department went o study for any reason other than disease
of Lymphoma and Myeloma, The University of Texas MD progression or death were censored on the last CT date,
Anderson Cancer Center. A 20% increase in the OR rate to either before going o study or within 2 weeks of being
45% with the addition of lenalidomide would justify o study. Responders who did not experience disease
assessment of this regimen in larger, more denitive progression but died of another reason were censored at
trials. For the sample size calculation, we used a one-sided the date of death. Responders who were still actively on
type 1 error rate of 5% and needed at least 80% power to study were censored at the survival date (ie, date of last
conclude that this regimen elicits an OR rate signicantly follow-up) or the last CT date, whichever was later.
greater than 25% if the true response rate of this regimen We calculated the median time to event in months and
is 45% or higher. Based on these assumptions, if ve or corresponding 95% CIs. OS was dened as the time from
fewer ORs were reported in the rst 17 patients, we would study entry until death due to any cause, or until the last
conclude that the regimen is insuciently eective to survival date if still alive.15 PFS was dened as the time
warrant further study. Otherwise, accrual would continue from the start of treatment until lymphoma progression
to a total of 41 patients. If 15 or more ORs were noted in or death regardless of cause.15 For the PFS calculation,
41 patients, we would conclude that this regimen is patients who underwent SCT were censored on the last
suciently eective to warrant further study. We added CT scan date, either before going o study or within
the six patients treated at the MTD in phase 1 to the 2 weeks of being o study. Patients who were still actively
38 patients enrolled in phase 2 for a total of 44 patients, on the study were censored at the survival date or the last
increasing the statistical power of our study. CT date, whichever was later. All statistical calculations
We used the Kaplan-Meier method to estimate were done with SAS (version 9.1.3, SAS, Cary, NC, USA)
response duration, progression-free survival (PFS), and or S-Plus 8.0 (TIBCO Software, Palo Alto, CA, USA).
overall survival (OS). Analyses were by intention to treat. This trial is registered with ClinicalTrials.gov, number
Response duration was calculated from the response NCT00294632.
date to the date of relapse, disease progression, or death.
Eligible patients who achieved PR or better and who Role of the funding source
See Online for appendix requested subsequent consolidation with SCT were Celgene provided lenalidomide and nancial support. The
sponsor of the study had no role in data collection, data
analysis, or data interpretation. This investigator-initiated
Grade 1 Grade 2 Grade 3 Grade 4 Grade 5
study was designed by the corresponding author who
Haematological events
worked with collaborators at the Department of
Anaemia 11 (79%) 4 (29%) 0 (0%) 0 (0%) 0 (0%)
Lymphoma and Myeloma at The University of Texas
Neutropenia 5 (36%) 8 (57%) 7 (50%) 5 (36%) 0 (0%) MD Anderson Cancer Center. We communicated and
Febrile neutropenia 1 (7%) 1 (7%) 1 (7%) 0 (0%) 0 (0%) discussed the study design with the sponsor, but this did
Thrombocytopenia 8 (57%) 3 (21%) 4 (29%) 1 (7%) 0 (0%) not lead to alteration of the study design. The sponsor
Leucopenia 5 (36%) 3 (21%) 4 (29%) 1 (7%) 0 (0%) reviewed the manuscript before submission and provided
Lymphopenia 8 (57%) 7 (50%) 3 (21%) 0 (0%) 0 (0%) grammatical revisions. The corresponding author had full
Non-haematological events access to all the data in the study and had nal
Pruritus 10 (71%) 3 (21%) 0 (0%) 0 (0%) 0 (0%) responsibility for the decision to submit for publication.
Fatigue 11 (79%) 5 (36%) 2 (14%) 0 (0%) 0 (0%)
Limb pain 4 (29%) 0 (0%) 1 (7%) 0 (0%) 0 (0%) Results
Ataxia 0 (0%) 0 (0%) 1 (7%) 0 (0%) 0 (0%) From Feb 10, to Dec 12, 2006, 14 patients were enrolled in
Cough 8 (57%) 1 (7%) 1 (7%) 0 (0%) 0 (0%) the phase 1 portion of the study, 86% of whom were men
Increased liver function 9 (64%) 2 (14%) 0 (0%) 0 (0%) 0 (0%) (table 1). The median age was 72 years (range 5684) and
Syncope 0 (0%) 0 (0%) 1 (7%) 0 (0%) 0 (0%) the median time from diagnosis to initiation of study
Non-neutropenic infections 8 (57%) 4 (29%) 1 (7%) 1 (7%)* 1 (7%)* treatment was 32 months (range 1487). The median
Hyperglycaemia 12 (86%) 6 (43%) 0 (0%) 0 (0%) 0 (0%) number of previous treatment regimens was two (range
Myalgia 3 (21%) 2 (14%) 1 (7%) 0 (0%) 0 (0%) one to four). All patients had received previous rituximab-
Hyperuricaemia 5 (36%) 0 (0%) 1 (7%) 1 (7%) 0 (0%) containing treatments, including eight who had received
Hypercalcaemia 1 (7%) 0 (0%) 1 (7%) DLT 0 (0%) 0 (0%) rituximab as part of their last treatment before study
entry. No patients had received previous treatment with
Data are number (%). All grade 3 or 4 adverse events and grade 1 and 2 adverse events experienced by at least ten
thalidomide.
patients are shown. DLT=dose-limiting toxicity. *One patient had grade 4 non-neutropenic fever. This patient chose to
be admitted to a hospice and died and so this event is also listed as a grade 5 toxicity. Three patients were enrolled in cohort 1 at the 10 mg dose
and three were enrolled in cohort 2 at the 15 mg dose
Table 2: Common adverse events in patients in phase 1 (n=14) after 107 cycles of lenalidomide plus
(appendix). Three patients were initially enrolled at the
rituximab
20 mg dose in cohort 3 but, in accordance with the

718 www.thelancet.com/oncology Vol 13 July 2012

 Articles

3+3 design, another three were enrolled, to make a total of patients. For phase 2, an additional 38 patients were
six patients at the 20 mg dose in cohort 3. Two patients enrolled from Jan 19, 2007, to July 30, 2009, for a total of
were enrolled in cohort 4 at the 25 mg dose, but because 44 patients treated at the MTD. 91% of these 44 patients
the rst two patients at the 25 mg dose had potentially
Grade 1 Grade 2 Grade 3 Grade 4
study-related adverse events, a third patient was not
enrolled according to the 3+3 design. One of the rst two Haematological events
patients enrolled in cohort 4 (25 mg of lenalidomide) Anaemia 31 (70%) 6 (14%) 1 (2%) 0 (0%)
experienced a DLT. This patient developed grade 3 Neutropenia 20 (45%) 22 (50%) 16 (36%) 13 (30%)
hypercalcaemia, hyperuricaemia, and raised creatinine Febrile neutropenia 1 (2%) 7 (16%) 2 (5%) 0 (0%)
and needed admission to hospital and intravenous Thrombocytopenia 23 (52%) 9 (20%) 8 (18%) 2 (5%)
hydration for 3 days. The cause remained unknown after Leucopenia 26 (59%) 14 (32%) 10 (23%) 3 (7%)
extensive assessments, and all abnormalities resolved Lymphopenia 27 (61%) 21 (48%) 12 (27%) 4 (9%)
within 2 weeks. This patient resumed treatment with Non-haematological events
lenalidomide at a lower dose of 20 mg daily with no Pruritus 19 (43%) 3 (7%) 0 (0%) 0 (0%)
relapses of hypercalcaemia, hyperuricaemia, or renal Fatigue 39 (89%) 17 (39%) 2 (5%) 0 (0%)
insuciency. The patient eventually received 14 cycles of Constipation 27 (61%) 3 (7%) 0 (0%) 0 (0%)
treatment, during which the best response was a PR. The Neuropathy 27 (61%) 5 (11%) 1 (2%) 0 (0%)
second patient enrolled at 25 mg, who had a history of Cough 17 (39%) 1 (2%) 1 (2%) 0 (0%)
coronary artery disease, developed grade 4 non- Nausea 15 (34%) 5 (11%) 0 (0%) 0 (0%)
neutropenic fever, hypotension, and sepsis, and was Vomiting 11 (25%) 4 (9%) 0 (0%) 0 (0%)
admitted to a hospice. At the patients request all Memory impairment 11 (25%) 2 (5%) 0 (0%) 0 (0%)
aggressive measures were stopped, and the patient died Mood alteration 11 (25%) 1 (2%) 0 (0%) 0 (0%)
after receiving palliative care. An autopsy was declined. Ataxia 1 (2%) 0 (0%) 1 (2%) 0 (0%)
Because we could not rule out the possibility of a relation Dizziness 14 (32%) 4 (9%) 0 (0%) 0 (0%)
between lenalidomide and the non-neutropenic fever, we Blurred vision 13 (30%) 2 (5%) 0 (0%) 0 (0%)
deemed this event to be a potential toxic event. Although Oral cavity pain 14 (32%) 1 (2%) 0 (0%) 0 (0%)
this event occurred in the second cycle, which is beyond
Abdominal pain 7 (16%) 2 (5%) 2 (5%) 0 (0%)
the protocol denition for a DLT, the study investigators
Limb pain 6 (14%) 0 (0%) 1 (2%) 0 (0%)
collectively decided that, because of this possible study-
Dyspnoea 19 (43%) 4 (9%) 0 (0%) 0 (0%)
drug-related death and the DLT in the other patient treated
Pleural eusion 2 (5%) 2 (5%) 2 (5%) 0 (0%)
at 25 mg, a dose of 25 mg was too toxic to be used in this
Non-neutropenic infection 16 (36%) 11 (25%) 1 (2%) 0 (0%)
cohort of patients. Thus, no additional patients were
Diarrhoea 22 (50%) 7 (16%) 0 (0%) 0 (0%)
enrolled at 25 mg and the MTD was identied as 20 mg.
Rash 21 (48%) 6 (14%) 2 (5%) 0 (0%)
Patients in phase 1 received a total of 107 cycles
Myalgia 20 (45%) 8 (18%) 2 (5%) 0 (0%)
(median 3, range 129) of treatment. The most common
grade 12 haematological adverse events (10 events) in Hyperuricaemia 12 (27%) 0 (0%) 1 (2%) 1 (2%)

phase 1 were lymphopenia, anaemia, neutropenia, and Hypercalcaemia 2 (5%) 0 (0%) 1 (2%) 0 (0%)

thrombocytopenia (table 2). Grade 34 haematological Hypocalcaemia 8 (18%) 2 (5%) 0 (0%) 0 (0%)
adverse events were neutropenia, thrombocytopenia, Hypophosphataemia 2 (5%) 5 (11%) 1 (2%) 0 (0%)
lymphopenia, leucopenia, and febrile neutropenia. The Hyperglycaemia 25 (57%) 14 (32%) 2 (5%) 0 (0%)
most common grade 12 non-haematological adverse Hyperkalaemia 1 (2%) 1 (2%) 1 (2%) 0 (0%)
events (10 events) were hyperglycaemia, fatigue, Hypoalbuminaemia 10 (23%) 4 (9%) 0 (0%) 0 (0%)
pruritus, non-neutropenic infections, and increased liver Hypomagnesaemia 15 (34%) 0 (0%) 0 (0%) 0 (0%)
function. Grade 34 non-haematological adverse events Raised BUN 10 (23%) 0 (0%) 0 (0%) 0 (0%)
included fatigue (two events), non-neutropenic infections Raised B2M 13 (30%) 0 (0%) 0 (0%) 0 (0%)
(two events, one grade 5), cough (one event), ataxia (one Raised LDH 12 (27%) 1 (2%) 0 (0%) 0 (0%)
event), limb pain (one event), myalgia (one event), and Elevated liver function test 15 (34%) 3 (7%) 0 (0%) 0 (0%)
syncope (one event). Additionally, one patient experienced Facial oedema 3 (7%) 0 (0%) 1 (2%) 0 (0%)
hypercalcaemia (one event, grade 3 DLT) with Oedema limb 18 (41%) 2 (5%) 0 (0%) 0 (0%)
hyperuricaemia (two events, one grade 3 [DLT] and one Rhinorrhoea 17 (39%) 2 (5%) 0 (0%) 0 (0%)
grade 4). Infusion-related reactions to rituximab, such as Red eyes 14 (32%) 2 (5%) 0 (0%) 0 (0%)
fever, rigors and chills, developed in seven patients Thrombosis or thromboembolism 0 (0%) 0 (0%) 2 (5%) 1 (2%)
(50%), but these were limited to grade 1 or 2 and occurred
Data are number of patients (%). Data are for the six patients from phase 1 who were treated with 20 mg lenalidomide
mainly during the rst administration of rituximab. No plus an additional 38 patients in phase 2. All grade 3 or 4 adverse events and grade 1 and 2 adverse events experienced
grade 34 anaemia events were reported in phase 1. by at least ten patients are shown. BUN=blood urea nitrogen. B2M=2 microglobulin. LDH=lactate dehydrogenase.
Six patients from phase 1 who received 20 mg lena-
Table 3: Common adverse events in phase 2 (n=44) after 379 cycles of lenalidomide plus rituximab
lidomide (the established MTD) were counted as phase 2

www.thelancet.com/oncology Vol 13 July 2012 719

 Articles

were men (table 1). The median age was 66 years All 44 patients in phase 2 were assessed for response to
(range 4685) and the median time from diagnosis was treatment: 36% achieved a CR, 20% a PR, 23% had stable
27 months (395). The median number of previous lines disease, and 20% had progressive disease (table 4). Nine
of treatment was two (range one to four). For 18 of of the 25 patients who achieved an OR (PR or CR) had
44 patients with assessment of the Ki-67 index at study treatment breaks longer than 1 week (median 2 weeks,
entry, the median value was 28% (range 580%). range 15 weeks), and all nine remained in remission
Although we aimed to measure Ki-67 expression for all during the breaks.
patients, we were not able to obtain specimens for all The median follow-up was 231 months (range
cases or in some cases there was not enough tissue for 156542 months) and the median time to rst
diagnostic or prognostic analyses. All patients in phase 2 response was 2 months (range 28 months; table 4).
had received previous rituximab-containing treatments. Only one of 44 patients had a CR after six cycles of
12 patients had received previous bortezomib therapy, treatment. The median response duration for the
nine of whom were refractory to bortezomib. 25 patients with an OR was 189 months (95% CI 170
379 cycles of study treatment were delivered to the months to not reached [NR]), the median PFS was
44 patients in phase 2, and the median dose applied was 111 months (95% CI 83 to 249 months), and the
15 mg (range 520 mg). Common (10 events) grade 12 median OS was 243 months (198 months to NR;
haematological adverse events included lymphopenia,
neutropenia, leucopenia, anaemia, thrombocytopenia, and A Response duration
febrile neutropenia (table 3; appendix). Grade 34 haem- 10
atological adverse events were neutropenia, lymphopenia,
08
leucopenia, anaemia, thrombocytopenia, and febrile neu-

Probability
tropenia. Common grade 12 non-haematological adverse 06
events (experienced by >50% of patients) were fatigue, 04
constipation, neuropathy, hyperglycaemia, and diarrhoea.
Grade 34 non-haematological events were fatigue, 02
neuropathy, cough, ataxia, abdominal pain, limb pain, 0
pleural eusion, non-neutropenic infection, rash, myalgia, 0 6 12 18 24 30
hyperuricaemia, hypercalcaemia, hypophosphataemia, Time from response date (months)
hyperglycaemia, hyperkalaemia, facial oedema, and throm- Number at risk 25 14 10 6 3
bosis or thromboembolism. Infusion-related reactions to B Progression-free survival
rituximab developed in seven patients and occurred mainly 10
during the rst administration of rituximab.
08
23 of 44 patients (52%) in phase 2 needed at least one
Probability

dose reduction or interruption because of adverse events. 06

These were due to neutropenia in 23 patients (52%), and
04
thrombocytopenia in four patients (9%). Neutropenia
coexisted with thrombocytopenia in ve patients. Patients 02
received a median of two cycles (range one to 26 cycles) 0
at the MTD. 0 6 12 18 24 30
Time (months)
Phase 2 (n=44)* Number at risk 44 23 14 8 4 2

Complete response 16 (36%) C Overall survival

Partial response 9 (20%) 10

Overall response 25 (57%) 08

Stable disease 10 (23%)
Probability

06
Progressive disease 9 (20%)
Response duration (months) 189 (170NR) 04
Progression-free survival (months) 111 (83249)
02
Overall survival (months) 243 (198NR)
Time to rst response (months) 2 (28) 0
0 6 12 18 24 30 36 42 48 54
Time to best response (months) 2 (212)
Time (months)
Follow-up time (months) 231 (156542)
Number at risk 44 41 38 26 13 8 4 4 2 2
Data are number (%) or median (range). NR=not reached. *Includes six patients
Figure: Response duration, progression-free survival, and overall survival
from phase 1 who were treated with 20 mg lenalidomide.
(A) Response duration for 25 patients who achieved an overall response,
Table 4: Response rates at the maximum tolerated dose in phase 2 (B) progression-free survival, and (C) overall survival of 44 patients enrolled
in phase 2.

720 www.thelancet.com/oncology Vol 13 July 2012

 Articles

gure). Of the 44 patients in phase 2, 16 declined follow-

up bone-marrow assessment. Of the 28 patients with Panel: Research in context
follow-up bone marrow biopsies, 12 were positive for Systematic review
bone marrow involvement. We undertook a review of the literature before we designed this clinical trial. We searched
Two patients had bulky tumour masses that were over Medline between 1995 and 2005 for English-language articles using the search terms
5 cm in the longest dimension, and both achieved a CR. lenalidomide, thalidomide, lymphoma and rituximab. We also searched the abstracts
In one patient with a bulky mediastinal mass and an from meetings of the American Society for Clinical Oncology, European Hematology
abdominal mass (>10 cm), a marked tumour response Association, and American Society of Hematology using the same search criteria. The
was noted after two cycles of treatment (appendix). The reference lists of relevant papers and our own les were also searched. The novel oral
patients masses in the mediastinum and abdomen immunomodulatory drug lenalidomide was eective as a single drug in relapsed or
completely resolved after four and six cycles, respectively. refractory non-Hodgkin lymphoma.19 The monoclonal antibody rituximab was also
12 patients, two from phase 1 and ten from phase 2, eective as monotherapy in patients with MCL.7,8 Because these drugs have a
went o study after achieving stable disease or better (four non-overlapping toxicity prole, we hypothesised that the combination of lenalidomide
had stable disease, four a PR, and four a CR) and qualifying and rituximab could be eective in relapsed or refractory MCL. We reported that the
for intensive therapy with autologous SCT. None dropped lenalidomide-rituximab combination had synergistic eects in preclinical models.11
out because of intolerance to treatment. Although these Although in a large-scale randomised study,20 maintenance treatment with rituximab
patients were censored at the time of transplant, all increased progression-free survival compared with observation alone in patients with
12 patients achieved a CR after autologous SCT. follicular lymphoma, when we designed this trial in 2006, maintenance treatment with
14 patients had received bortezomib treatment before rituximab had not yet been validated. Therefore, we designed the study to treat patients
enrolment, three in phase 1 and 11 in phase 2. Five of with rituximab weekly for 4 weeks in cycle 1 only. The data published at the time of our
14 patients achieved an OR. Among the 44 patients study design and our preclinical evidence provided a strong rationale for designing this
treated at the MTD, 11 were refractory and 33 were trial to assess the ecacy of the lenalidomide-rituximab combination in MCL.
sensitive to rituximab. Five of eleven rituximab-refractory
patients and 22 of 33 rituximab-sensitive patients had a Interpretation
response (PR or CR). In this phase 1/2 study, the combination of lenalidomide and rituximab was eective in
relapsed or refractory MCL, with a favourable toxicity prole. These ndings are consistent
Discussion with studies that showed that lenalidomide has single-drug activity in relapsed or
In this phase 1/2 study, we identied the MTD of refractory MCL.2123 We believe our data will eventually prove to be benecial for a dened
lenalidomide to be 20 mg when combined with rituximab. cohort of patients with MCL. A phase 3 trial of this combination will be necessary to assess
The combination was eective, with an OR rate of whether clinical practice should be changed.
57% and minimal toxicity in patients with relapsed or
refractory MCL (panel). Our data suggest that 20 mg study. In two other studies, lenalidomide was given at
lenalidomide plus rituximab was associated with less 25 mg for relapsed MCL,24,25 and more patients had
severe toxicities than 25 mg lenalidomide plus rituximab toxicities, with more grade 34 adverse events than those
and, therefore, allowed longer drug exposure. reported in the present study. In the study by Witzig and
The rate of severe neuropathy was low; only one patient colleagues,24 adverse events led to discontinuation from
had grade 3 neuropathy. This result compares favourably study treatment in 49 (23%) of 217 patients, whereas
with bortezomib in the setting of relapsed or refractory none of our patients discontinued the study treatment.
MCL.21,22 In the phase 1 portion of this study, there were no The median time to dose reduction or interruption was
thromboembolic events, and in the phase 2 portion, there two cycles (range one to nine cycles) and the median
were three thromboembolic events. Additionally, there time to response was 2 months (range 28 months).
was no concurrent use of steroids such as dexamethasone. Therefore, on the basis of present data and because the
Our data are consistent with ndings from an international 20 mg dose of lenalidomide was eective in our study, we
phase 2 trial of lenalidomide monotherapy in 217 patients believe the optimal dose in relapsed MCL is 20 mg.
who had relapsed or refractory aggressive B-cell non- The combination of lenalidomide and rituximab was
Hodgkin lymphoma, which showed a low rate of deep eective, with 57% (25/44) of patients achieving an OR,
vein thrombosis (2%).24 Based on the results from these and the responses were good quality, with 36% of patients
trials, no anticoagulation or antithrombotic prophylaxis achieving a CR. Most patients who achieved a CR had
has been recommended. Although lenalidomide caused received one previous treatment (ten of 16), although CRs
neutropenia in our study, it was rarely associated with were also noted in patients who had received up to four
febrile neutropenia. The rate of grade 34 thrombo- previous treatments and in those who had received high-
cytopenia was very low. Therefore, bone marrow dose or dose-intensive chemotherapy. Furthermore,
suppression was mild to moderate and reversible. lenalidomide in combination with rituximab was eective
Because one of the patients treated at 25 mg developed in patients who were unable to tolerate bortezomib or
a DLT and the other developed a grade 4 non-neutropenic whose disease was refractory to bortezomib.
infection and died in cycle 2, we used a starting dose of The responses induced by this combination were
20 mg instead of 25 mg for the phase 2 portion of the durable, with a median duration of 189 months and a

www.thelancet.com/oncology Vol 13 July 2012 721

 Articles

median PFS close to 1 year. However, this combination In conclusion, lenalidomide in combination with
compares favourably with bortezomib, the only drug rituximab was eective in patients with relapsed or refrac-
approved by the US Food and Drug Administration for tory MCL, with durable responses and minimal toxicity.
relapsed or refractory MCL. In a multicentre phase 2 Contributors
study, bortezomib achieved an overall response in MW, LZ, FC, and LS designed the study. MW, NW-B, FH, SSN, PM, AY,
47 (33%) of 141 patients, with 11 (8%) CRs, and a median MBa, MBe, KH, WC, and YC collected and assembled the data. MW, LFa,
NW-B, FH, SSN, FS, MF, AY, NF, KHY, RC, LK, LFe, and JR analysed and
response duration of 92 months (range 49135).21,22 In interpreted the data. MW, LFa, NW-B, LZ, FH, SSN, FS, PM, MF, AY, FC,
a recent multicentre international trial of lenalidomide NF, KJN, LS, KHY, RC, LK, LFe, MBa, MBe, WC, YC, CB, NB, JZ, and JR
monotherapy for relapsed or refractory B-cell non- wrote the manuscript. All authors gave nal approval for the manuscript.
Hodgkin lymphoma, 24 (42%) of a subpopulation of FH, FS, MF, RC, AY, and WC provided study materials and patients.
LS provided administrative support.
57 patients with relapsed or refractory MCL had an OR
and 12 (21%) had a CR.24 The median PFS for all patients Conicts of interest
MW, SSN, NF, and LK have received research support from Celgene for
was 57 months, and the median OR duration for undertaking the preclinical and clinical trials of lenalidomide. MW and
responders had not been reached after a follow-up of AY have received honoraria from Celgene. SSN, PM, AY, and NF have
71 months. The aforementioned studies are not directly consultancy or advisory relations with Celgene. NB and JZ are employed
comparable to the present study because they did not by Celgene and own Celgene stock. CB owns Celgene stock and is a
former employee of Celgene. JZ has provided expert testimony for and
have simultaneous randomised control groups and the received other remunerations from Celgene. LF, NW-B, LZ, FH, FS, MF,
response and survival rates were not adjusted for patient AY, FC, KJN, LS, KHY, RC, LF, MBa, MBe, KH, WC, YC, and JR declare
characteristics. However, the data from the present study that they have no conicts of interest.
of the combination of lenalidomide and rituximab seem References
to be more favourable than lenalidomide monotherapy, 1 Fernndez V, Hartmann E, Ott G, et al. Pathogenesis of mantle-cell
lymphoma: all oncogenic roads lead to dysregulation of cell cycle
especially with regard to the PFS, which was twice as and DNA damage response pathways. J Clin Oncol 2005;
long (111 months) as that reported for the single drug in 23: 636469.
two trials (57 months24 and 56 months25). A well- 2 Prez-Galn P, Dreyling M, Wiestner A. Mantle cell lymphoma:
biology, pathogenesis, and the molecular basis of treatment in the
designed phase 3 multicentre controlled trial is needed genomic era. Blood 2011; 117: 2638.
to provide denitive data. 3 Barista I, Romaguera JE, Cabanillas F. Mantle-cell lymphoma.
In 2005, Rummel and colleagues23 reported the results Lancet Oncol 2001; 2: 14148.
from a study of bendamustine plus rituximab in patients 4 Romaguera JE, Fayad L, Rodriguez MA, et al. High rate of durable
remissions after treatment of newly diagnosed aggressive
with relapsed or refractory MCL. Bendamustine is a mantle-cell lymphoma with rituximab plus hyper-CVAD alternating
potent alkylating drug that is eective for treatment of with rituximab plus high-dose methotrexate and cytarabine.
J Clin Oncol 2005; 23: 701323.
MCL, even in cases that are refractory to rituximab and
5 Romaguera JE, Fayad LE, Feng L, et al. Ten-year follow-up after
other alkylating drugs. 16 patients with relapsed or intense chemoimmunotherapy with rituximab-hyperCVAD
refractory MCL were enrolled in Rummel and alternating with rituximab-high dose methotrexate/cytarabine
(R-MA) and without stem cell transplantation in patients with
colleagues study23 and the combination displayed good untreated aggressive mantle cell lymphoma. Br J Haematol 2010;
ecacy (75% [12/16] ORs and 50% [8/16] CRs), with 150: 20008.
acceptable toxicity. Another study26 recruited 12 relapsed 6 Wang M, Fayad L, Cabanillas F, et al. Phase 2 trial of rituximab plus
patients with MCL who were treated with bendamustine hyper-CVAD alternating with rituximab plus methotrexate-
cytarabine for relapsed or refractory aggressive mantle cell
and rituximab; 92% ORs, 42% CRs, 17% unconrmed lymphoma. Cancer 2008; 113: 273441.
CRs, and 33% PRs were reported. The primary toxicity 7 Ghielmini M, Schmitz SF, Cogliatti S, et al. Eect of single-agent
was myelosuppression (36% grade 3 or 4 neutropenia; rituximab given at the standard schedule or as prolonged
treatment in patients with mantle cell lymphoma: a study of the
9% grade 3 or 4 thrombocytopenia). Although the eects Swiss Group for Clinical Cancer Research (SAKK). J Clin Oncol
of bendamustine and rituximab were promising in 2005; 23: 70511.
patients with MCL, the combination seemed to cause 8 Foran JM, Rohatiner AZ, Cunningham D, et al. European
phase II study of rituximab (chimeric anti-CD20 monoclonal
signicant myelosuppression and is cytoreductive in antibody) for patients with newly diagnosed mantle-cell
nature. Whether bendamustine can be combined safely lymphoma and previously treated mantle-cell lymphoma,
with lenalidomide remains unknown and further data immunocytoma, and small B-cell lymphocytic lymphoma.
J Clin Oncol 2000; 18: 31724.
are needed. 9 Rubenstein JL, Treseler PA, Stewart PJ. Regression of refractory
Recently, Ansell and colleagues27 reported ndings intraocular large B-cell lymphoma with lenalidomide monotherapy.
from a phase 2 study of rituximab combined with J Clin Oncol 2011; 29: e59597.
10 Zeldis JB, Knight R, Hussein M, et al. A review of the history,
temsirolimus, an mTOR inhibitor, in patients with properties, and use of the immunomodulatory compound
relapsed or refractory MCL. Of 69 assessable patients lenalidomide. Ann N Y Acad Sci 2011; 1222: 7682.
with MCL, 41 (59%) had an OR, with 13 (19%) CRs and 11 Zhang L, Qian Z, Cai Z, et al. Synergistic antitumor eects of
lenalidomide and rituximab on mantle cell lymphoma in vitro and
28 (41%) PRs. The OR for rituximab-sensitive patients in vivo. Am J Hematol 2009; 84: 55359.
was 63% (95% CI 4776) and for rituximab-refractory 12 Bartlett JB, Wu L, Adams M, et al. Lenalidomide and
patients it was 52% (3074). Although the combination of pomalidomide strongly enhance tumor cell killing in vitro during
antibody-dependent cellular cytotoxicity (ADCC) mediated by
temsirolimus and rituximab had good ecacy in patients trastuzumab, cetuximab and rituximab. Proc Am Soc Clin Oncol
with MCL, it had signicant toxicity. 2007; 25: 3023.

722 www.thelancet.com/oncology Vol 13 July 2012

 Articles

13 Wu L, Adams M, Carter T, et al. Lenalidomide enhances natural 21 Fisher RI, Bernstein SH, Kahl BS, et al. Multicenter phase II study
killer cell and monocyte-mediated antibody-dependent cellular of bortezomib in patients with relapsed or refractory mantle cell
cytotoxicity of rituximab-treated CD20+ tumor cells. Clin Cancer Res lymphoma. J Clin Oncol 2006; 24: 486774.
2008; 14: 465057. 22 Goy A, Bernstein SH, Kahl BS, et al. Bortezomib in patients with
14 NCI. Common terminology criteria for adverse events v3.0. relapsed or refractory mantle cell lymphoma: updated time-to-event
http://ctep.cancer.gov/protocolDevelopment/electronic_ analyses of the multicenter phase 2 PINNACLE study. Ann Oncol
applications/docs/ctcaev3.pdf (accessed May 23, 2012). 2009; 20: 52025.
15 Cheson BD, Horning SJ, Coier B, et al. Report of an international 23 Rummel MJ, Al-Batran SE, Kim SZ, et al. Bendamustine plus
workshop to standardize response criteria for non-Hodgkins rituximab is eective and has a favorable toxicity prole in the
lymphomas. NCI Sponsored International Working Group. treatment of mantle cell and low-grade non-Hodgkins lymphoma.
J Clin Oncol 1999; 17: 1244. J Clin Oncol 2005; 23: 338389.
16 Simon R. Optimal two-stage designs for phase II clinical trials. 24 Witzig TE, Vose JM, Zinzani PL, et al. An international phase II
Control Clin Trials 1989; 10: 110. trial of single-agent lenalidomide for relapsed or refractory
17 Coier B, Haioun C, Ketterer N, et al. Rituximab (anti-CD20 aggressive B-cell non-Hodgkins lymphoma. Ann Oncol 2011;
monoclonal antibody) for the treatment of patients with relapsing 22: 162227.
or refractory aggressive lymphoma: a multicenter phase II study. 25 Habermann TM, Lossos IS, Justice G, et al. Lenalidomide oral
Blood 1998; 92: 192732. monotherapy produces a high response rate in patients with
18 Nguyen DT, Nguyen DT, Amess JA, Doughty H, Hendry L, relapsed or refractory mantle cell lymphoma. Br J Haematol 2009;
Diamond LW. IDEC-C2B8 anti-CD20 (rituximab) immunotherapy 145: 34449.
in patients with low-grade non-Hodgkins lymphoma and 26 Robinson KS, Williams ME, van der Jaqt RH, et al. Phase II
lymphoproliferative disorders: evaluation of response on multicenter study of bendamustine plus rituximab in patients with
48 patients. Eur J Haematol 1999; 62: 7682. relapsed indolent B-cell and mantle cell non-Hodgkins lymphoma.
19 Wiernik PH, Lossos IS, Tuscano JM, et al. Lenalidomide J Clin Oncol 2008; 26: 447379.
monotherapy in relapsed or refractory aggressive non-Hodgkins 27 Ansell SM, Tang H, Kurtin PJ, et al. Temsirolimus and rituximab in
lymphoma. J Clin Oncol 2008; 26: 495257. patients with relapsed or refractory mantle cell lymphoma:
20 Salles G, Seymour JF, Oner F, et al. Rituximab maintenance for a phase 2 study. Lancet Oncol 2011; 12: 36168.
2 years in patients with high tumour burden follicular lymphoma
responding to rituximab plus chemotherapy (PRIMA): a phase 3,
randomised controlled trial. Lancet 2011; 377: 4251.

www.thelancet.com/oncology Vol 13 July 2012 723