V O L U M E 2 3 䡠 N U M B E R 4 䡠 F E B R U A R Y 1 2 0 0 5
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`O R I G I N A L R E P O R T
`
`Phase II Clinical Experience With the Novel Proteasome
`Inhibitor Bortezomib in Patients With Indolent
`Non-Hodgkin’s Lymphoma and Mantle Cell Lymphoma
`Owen A. O’Connor, John Wright, Craig Moskowitz, Jamie Muzzy, Barbara MacGregor-Cortelli,
`Michael Stubblefield, David Straus, Carol Portlock, Paul Hamlin, Elizabeth Choi, Otila Dumetrescu,
`Dixie Esseltine, Elizabeth Trehu, Julian Adams, David Schenkein, and Andrew D. Zelenetz
`
`A
`
`B
`
`S
`
`T
`
`R
`
`A
`
`C
`
`T
`
`Purpose
`To determine the antitumor activity of the novel proteasome inhibitor bortezomib in patients
`with indolent and mantle-cell lymphoma (MCL).
`Patients and Methods
`Patients with indolent and MCL were eligible. Bortezomib was given at a dose of 1.5 mg/m2
`on days 1, 4, 8, and 11. Patients were required to have received no more than three prior
`chemotherapy regimens, with at least 1 month since the prior treatment, 3 months from
`prior rituximab, and 7 days from prior corticosteroids; absolute neutrophil count more
`than 1,500/␮L (500/␮L if documented bone marrow involvement); and platelet count
`more than 50,000/␮L.
`Results
`Twenty-six patients were registered, of whom 24 were assessable. Ten patients had
`follicular lymphoma, 11 had MCL, three had small lymphocytic lymphoma (SLL) or chronic
`lymphocytic leukemia (CLL), and two had marginal zone lymphoma. The overall response
`rate was 58%, with one complete remission (CR), one unconfirmed CR (CRu), and four
`partial remissions (PR) among patients with follicular non-Hodgkin’s lymphoma (NHL). All
`responses were durable, lasting from 3 to 24⫹ months. One patient with MCL achieved a
`CRu, four achieved a PR, and four had stable disease. One patient with MCL maintained his
`remission for 19 months. Both patients with marginal zone lymphoma achieved PR lasting
`8⫹ and 11⫹ months, respectively. Patients with SLL or CLL have yet to respond. Overall, the
`drug was well tolerated, with only one grade 4 toxicity (hyponatremia). The most common
`grade 3 toxicities were lymphopenia (n ⫽ 14) and thrombocytopenia (n ⫽ 7).
`Conclusion
`These data suggest that bortezomib was well tolerated and has significant single-agent
`activity in patients with certain subtypes of NHL.
`
`J Clin Oncol 23:676-684.
`
`INTRODUCTION
`
`The ubiquitin-proteasome pathway plays an
`essential role in the degradation of most
`short- and long-lived intracellular proteins
`in eukaryotic cells.1 At the heart of this deg-
`radative pathway is the 26S proteasome, an
`adenosine triphosphate– dependent, multi-
`catalytic protease. Some of the proteins
`
`degraded by the ubiquitin-proteasome
`pathway include p53, p21, p27, nuclear fac-
`tor kappa B (NF-␬B), and bcl-2.1-4 Preclin-
`ical observations have
`suggested that
`inhibitors of this pathway act through mul-
`tiple mechanisms to arrest tumor growth,
`induce cell death, and inhibit tumor metas-
`tasis and angiogenesis.5-8 Phase I trials have
`confirmed that bortezomib is well tolerated
`
`From the Memorial Sloan-Kettering
`Cancer Center, Department of
`Medicine, Division of Hematologic
`Oncology, Lymphoma and Develop-
`mental Chemotherapy Services,
`Departments of Rehabilitation Medicine
`and Radiology, New York, NY;
`Millennium Pharmaceutical, Cambridge,
`MA; and Drug Development Branch,
`National Cancer Institute, Bethesda, MD.
`
`Submitted February 6, 2004; accepted
`September 10, 2004.
`
`Supported under an NCI Phase II
`grant (UO1 CA 69913). O.A.O. is the
`recipient of the Leukemia and
`Lymphoma Society Scholar in Research
`Award and is supported generously by
`the Werner and Elaine Dannheisser
`Fund for Research on the Biology of
`Aging of the Lymphoma Foundation.
`
`O.A.O. is on the Speakers Bureau for
`Millenium Pharmaceuticals.
`
`Authors’ disclosures of potential con-
`flicts of interest are found at the end of
`this article.
`
`Address reprint requests to Owen A.
`O’Connor, MD, PhD, Memorial
`Sloan-Kettering Cancer Center,
`Department of Medicine, Box 329,
`Lymphoma and Developmental
`Chemotherapy Services, 1275 York
`Ave, New York, NY 10021; e-mail:
`oconnoro@mskcc.org.
`
`0732-183X/05/2304-676/$20.00
`
`DOI: 10.1200/JCO.2005.02.050
`
`676
`
`Downloaded from ascopubs.org by 12.40.227.4 on March 2, 2018 from 012.040.227.004
`
`Copyright © 2018 American Society of Clinical Oncology. All rights reserved.
`
`IPR2018-00685
`Celgene Ex. 2038, Page 1
`
`

`

`Bortezomib in Indolent NHL and MCL
`
`and may have potential clinical activity in patients with
`select lymphomas.9,10
`The ubiquitin-proteasome pathway plays a critical role
`in regulating cell cycle control.4 For example, cyclins,
`cyclin-dependent kinases (cdk), and cdk inhibitors (p21,
`p27kip1) are temporally degraded during the cell cycle by the
`ubiquitin-proteasome pathway. Both p21 and p27 can in-
`duce cell cycle arrest by inhibiting cdk.11 The ordered deg-
`radation of these proteins is required for progression
`through cell cycle and mitosis. Another target of the
`ubiquitin-proteasome pathway is the tumor suppressor
`p53, which acts as a negative regulator of cell growth. p53 is
`required for the transcription of a number of genes involved
`in cell cycle control and DNA synthesis, and also plays an
`important function in apoptosis induced by cellular dam-
`age, including ionizing radiation.11
`In addition to the regulation of cell cycle control, the
`ubiquitin-proteasome pathway plays an important role in
`modulating the important transcription factor NF-␬B.
`NF-␬B is responsible for the activation of several genes that
`contribute to the malignant phenotype, including genes
`that promote cell proliferation, cytokine release, antiapop-
`tosis, and changes in cell surface adhesion molecules. The
`activity of NF-␬B is tightly regulated by the ubiquitin-
`proteasome pathway through the accumulation or degrada-
`tion of I␬B, which binds to and inactivates NF-␬B.4,12 Cell
`adhesion molecules (CAMs), such as E-selectin, ICAM-1,
`and VCAM-1, are proteins regulated by NF-␬B and are
`involved in tumor metastasis and angiogenesis in vivo.13 As
`such, tumor cell metastasis may well be prevented through
`the downregulation of NF-␬B– dependent CAM expres-
`sion. NF-␬B also controls cell viability by regulating both
`anti- and proapoptotic proteins in the mitochondrial mem-
`brane.14 Several lines of preclinical evidence now suggest
`that inhibiting NF-␬B activation by stabilizing the I␬B pro-
`tein can render cells more sensitive to environmental stress
`and cytotoxic agents, ultimately leading to programmed
`cell death. In addition, many lines of experimental inves-
`tigation have shown that inhibition of the proteasome,
`and perhaps even NF-␬B, sensitizes cells to a host of
`conventional cytotoxic therapies.1,5,7
`Although dysregulation of NF-␬B is common to many
`malignancies, select lymphoproliferative malignancies are
`often characterized by pathognomonic molecular lesions,
`which may render them especially vulnerable to inhibitors
`of this pathway. Specific lesions include the following ex-
`amples. First, the constitutive overexpression of cyclin D1
`(bcl-1, PRAD1) in mantle-cell lymphoma (MCL) is due to
`the t(11;14)(q13;q32) translocation, which may also be aug-
`mented through the constitutive activation of NF-␬B (and
`AP-1), which has been shown in cell lines of MCL.15 Sec-
`ond, the constitutive overexpression of the antiapoptotic
`protein bcl-2 in follicular lymphoma due to the t(14;
`18)(q32;q21) translocation can be mitigated through the
`
`inhibition of the 26S proteasome,16,17 which may be attrib-
`uted in part to inactivated NF-␬B. Third, the constitutive
`overexpression of NF-␬B is noted in gene array studies of
`chemotherapy-refractory diffuse large B-cell neoplasms (ie,
`ABC or activated B-cell lymphomas).18,19 Fourth, the t(1:
`14)(p22:q32) translocation leading to expression of the
`bcl-10 gene in mucosa-associated lymphoid tissue lympho-
`mas is believed to involve a caspase recruitment domain–
`containing protein that activates NF-␬B.20,21
`Bortezomib (Velcade, formerly known as PS-341; Mil-
`lenium Pharmaceuticals, Cambridge, MA) is a dipeptidyl
`boronic acid inhibitor with high specificity for the 26S
`proteasome.22 It is the first member of this new class of
`antitumor agents to be studied in human clinical trials,
`leading recently to its approval by the US Food and Drug
`Administration for the treatment of relapsed or refractory
`multiple myeloma. Phase I and II clinical studies have dem-
`onstrated that bortezomib is a well-tolerated agent with
`minimal hematologic toxicity. In addition, it has been
`shown that bortezomib is capable of producing a dose-
`related effect on proteasome inhibition when analyzed 1
`hour postinfusion, with little interpatient variability.9,10
`We present here the first clinical experience of bort-
`ezomib in patients with indolent and mantle cell non-
`Hodgkin’s lymphoma (NHL).
`
`PATIENTS AND METHODS
`
`Patient Selection
`Patients were required to have histologically confirmed lym-
`phoma according to the WHO/Revised European-American Lym-
`phoma classification, including chronic lymphocytic leukemia
`(CLL); B-cell small lymphocytic lymphoma (SLL); marginal zone
`lymphoma; follicular lymphoma, grades 1, 2, or 3; MCL; and
`Waldenström’s macroglobulinemia. Prior history of transformed
`lymphoma was permitted as long as recent biopsies revealed no
`evidence of aggressive lymphoma. Patients had to meet the follow-
`ing eligibility requirements for enrollment onto the study. Patients
`had to have measurable disease (defined as ⱖ 1 cm with spiral
`computed tomography scan); for patients with leukemic forms of
`NHL, including CLL, patients had to have an absolute lymphocy-
`tosis more than 5 ⫻ 109/L with a B-cell phenotype (CD19⫹,
`CD20⫹, or CD23⫹), and more than 30% bone marrow lympho-
`cytes. Patients had to have received no more than three prior lines
`of conventional cytotoxic therapy, and were required to have
`stopped receiving cytotoxic chemotherapy for at least 4 weeks
`before study enrollment (6 weeks for bischloroethylnitrosourea or
`mitomycin, and at least 7 days must have elapsed since the use of
`corticosteroids). There had to be a period of at least 3 months since
`the last administration of any monoclonal antibody. Patients had
`to be 18 years of age or older, have a life expectancy of 3 months or
`greater, and have a Karnofsky performance status more than 60%.
`Patients could have no signs of congestive heart failure according
`the New York Heart Failure Guidelines Class III/IV. Patients were
`allowed febrile episodes up to 38.5°C as long as there was no
`evidence of active infection. Patients were also eligible only if they
`had at baseline a grade 1 or less sensory neuropathy. The protocol
`
`www.jco.org
`
`677
`
`Downloaded from ascopubs.org by 12.40.227.4 on March 2, 2018 from 012.040.227.004
`
`Copyright © 2018 American Society of Clinical Oncology. All rights reserved.
`
`IPR2018-00685
`Celgene Ex. 2038, Page 2
`
`

`

`O’Connor et al
`
`was approved by the Memorial Sloan-Kettering Cancer Center
`Institutional Review Board, and all patients were required to sign
`an informed consent approved by the Institutional Review Board.
`In addition, patients were required to meet the following
`criteria within 2 days of study drug administration: an absolute
`neutrophil count ⱖ 1,500/␮L (if known lymphomatous involve-
`ment of the bone marrow, then absolute neutrophil count
`⬎ 500/mL); a platelet count of ⱖ 50,000/␮L for the first dose of
`every cycle, and more than 30,000/␮l for doses delivered on days 4,
`8, and 11; a total bilirubin ⱕ 1.5⫻ upper institutional limit of
`normal (ULN); an AST and ALT ⱕ 2.5⫻ ULN (4⫻ ULN if the
`patient had liver involvement); and a creatinine ⱕ 1.5⫻ ULN.
`Initially, patients were required to have a platelet count ⱖ 100,000/
`␮L, but these criteria were modified downward given the ab-
`sence of bleeding complications and the realization that many
`patients were unnecessarily missing doses secondary to the high
`cutoff value. Patients were excluded if they were pregnant; had
`evidence of intracranial disease; had major surgery within 4
`weeks of study drug administration; had uncontrolled illness
`including active infection, symptomatic congestive heart fail-
`ure, uncontrolled hypertension, unstable angina pectoris, car-
`diac arrhythmia, a myocardial infarction, or cerebrovascular
`accident within 6 months of study enrollment; had known HIV
`disease; or had a psychiatric illness that would limit compliance
`with study requirements.
`Study Design
`This was a single-center, single-agent phase II study of bort-
`ezomib in patients with relapsed, refractory, or untreated indolent
`NHL and MCL. The major objectives of the study were to deter-
`mine the frequency and duration of complete and partial response
`for patients with indolent lymphoproliferative disorders treated
`with bortezomib. The study used a Simon two-stage design. Ini-
`tially, 18 patients were enrolled onto the first stage. If no more than
`two patients responded, the trial would have been terminated. If at
`least three patients responded, then up to 35 patients could be
`enrolled. These statistics were predicated on the assumption that
`should 20% of patients respond, the drug will be declared as
`having promising activity.
`Drug Administration
`Bortezomib (N-pyrazinecarbonyl-L-phenylalanine-L-leucine
`boronic acid; CAS No. 179324-69-7) was supplied to investigators
`by the Division of Cancer Treatment and Diagnosis, National
`Cancer Institute. Bortezomib for injection was supplied as a ly-
`ophilized powder for reconstitution. Each vial contained 3.5 mg of
`bortezomib and 35 mg mannitol United States Pharmacopeia.
`Each vial was reconstituted with 3.5 mL normal (0.9%) saline,
`such that the reconstituted solution contained bortezomib at a
`concentration of 1 mg/mL. The pH of the reconstituted solution
`was between 5 and 6. The drug was injected during 3 to 5
`seconds into a side arm of a running intravenous infusion of
`normal saline at 100 mL/h. At the end of the drug infusion, 10
`mL of normal saline was infused to flush the line. There was no
`upper limit on planned therapy, and patients could continue to
`receive drug as long as there was evidence of clinical benefit
`without excess toxicity.
`Dose Modification
`Patients were treated at a dose of 1.5 mg/m2 twice weekly for
`2 weeks (days 1, 4, 8, and 11) followed by a 1-week rest period (one
`cycle). Treatment was delayed for patients whose peripheral blood
`counts failed to meet the eligibility criteria for re-treatment, and
`
`they were re-treated once their counts recovered. Use of antiemet-
`ics, erythropoietin, and filgrastim was allowed if deemed necessary
`by the treating physician. Their use was dictated by standard
`institutional guidelines. Although rare, in most patients in
`whom an antiemetic was required, a single oral dose of granis-
`etron was administered.
`In patients who developed a grade 3 or 4 nonhematologic
`toxicity of any sort or grade 4 neutropenic fever or thrombocyto-
`penia, the dose was reduced to 1.3 mg/m2, and then to 1.1 mg/m2
`for a repeat episode of toxicity. Patients who experienced persis-
`tent nonhematologic toxicity despite this dose reduction were
`removed from the study. Treatment was delayed until these toxic-
`ities resolved to baseline. Dose reductions were also allowed for
`patients who developed asthenia, anorexia, or neuropathy of any
`grade, which in the judgment of the treating physician, was be-
`lieved to be clinically significant and detrimental to the patients’
`continued involvement on study.
`Response Criteria
`Response criteria for patients enrolled onto the study fol-
`lowed the guidelines previously reported by Cheson et al.23 All
`responses were characterized as either complete remission (CR),
`unconfirmed complete remission (CRu), partial remission (PR),
`stable disease (SD), or progression of disease (POD). Response
`criteria for patients with leukemic forms of NHL, including CLL,
`followed the National Cancer Institute guidelines previously re-
`ported by Cheson et al.24 Response was routinely assessed after
`every two cycles. For patients removed from the study, a 1-month
`confirmatory scan was performed, and patients were then restaged
`every 3 months. There was no upper limit on the amount of
`bortezomib any patient could receive. In general, patients contin-
`ued participating in the study until one of the following criteria
`were met: the patient withdrew consent or the patient experienced
`unacceptable toxicity. If the patient achieved a PR, therapy con-
`tinued until maximal benefit; if the patient achieved a CR, the
`patient continued for two cycles beyond CR.
`Nerve Conduction Studies
`Electrodiagnostic evaluations, including nerve conduction
`studies and needle electromyography, were completed on select
`patients. Motor and sensory nerve conduction and needle electro-
`myographic studies were performed and responses recorded using
`standardized equipment and techniques. Needle electromyogra-
`phy was performed on selected muscles of the upper and or lower
`extremity and their corresponding paraspinal muscles. Interpre-
`tation of the completed electrodiagnostic studies was done by a
`board-certified electromyographer.
`
`RESULTS
`
`Demographics
`This study reports on our initial experience of 26 con-
`secutive patients enrolled between June 2001 and December
`of 2003. Table 1 presents the demographic and clinical
`features of all study patients. Twenty-six patients were reg-
`istered for treatment, of whom 24 were assessable for re-
`sponse (ie, they completed at least two cycles of therapy). All
`patients were assessable for toxicity. One of the two inas-
`sessable patients had a history of follicular lymphoma and
`received two dose of bortezomib, which was followed by a
`
`678
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`Downloaded from ascopubs.org by 12.40.227.4 on March 2, 2018 from 012.040.227.004
`
`Copyright © 2018 American Society of Clinical Oncology. All rights reserved.
`
`IPR2018-00685
`Celgene Ex. 2038, Page 3
`
`

`

`Bortezomib in Indolent NHL and MCL
`
`Table 1. Patient Demographics
`
`Characteristic
`
`No. of
`Patients
`
`No. of patients
`Male
`Female
`White non-Hispanic
`African American or Hispanic
`Age, years
`Median
`Range
`Disease
`Follicular lymphoma, grade
`I
`II
`III
`Mantle cell lymphoma
`Small lymphocytic
`Lymphoma or CLL
`Marginal zone lymphoma
`Prior treatment
`Median number of prior therapiesⴱ (all)
`Median number of prior cytotoxic therapies
`Alkylator-based therapy
`CHOP
`R-CHOP
`CVP
`R-ICE
`ICE
`EPOCH
`ESHAP
`BACOP or CHOEP
`Oral CTX
`Chlorambucil
`HDC ⫹ ASCT
`Purine analog–based therapy
`Fludarabine
`CyFlu
`2-Chlorodeoxyadenosine
`Biological- or experimental-based therapy
`Rituximab alone
`Zevalin
`Bexxar
`PEG interferon/ribavarin
`Anti-B1 MAB
`PDX
`Radiation therapy
`No prior treatment
`
`63
`44–78
`
`3
`3
`
`26
`14
`12
`24
`2
`
`10
`4
`4
`2
`11
`3
`
`2
`
`12
`5
`5
`2
`1
`1
`1
`1
`1
`1
`3
`
`1
`2
`1
`
`21
`1
`1
`1
`1
`1
`9
`1
`
`%
`
`54
`46
`92
`8
`
`38
`15
`15
`8
`42
`12
`
`8
`
`48
`19
`20
`8
`4
`4
`4
`4
`4
`4
`12
`
`4
`10
`4
`
`84
`4
`4
`4
`4
`4
`36
`4
`
`Abbreviations: CLL, chronic lymphocytic leukemia; CHOP, cyclophos-
`phamide, doxorubicin, vincristine, prednisone; R-CHOP, CHOP plus ritux-
`imab; CVP, cyclophosphamide, vincristine, prednisone; R-ICE, ICE plus
`rituximab; ICE, ifosfamide, carboplatin, etoposide; EPOCH, etoposide,
`prednisone, vincristine; ESHAP, etoposide, methylprednisone, high-dose
`cytarabine, cisplatin; BACOP, bleomycin, doxorubicin, cyclophospha-
`mide, vincristine, prednisone; CHOEP, CHOP plus etoposide; CTX,
`cytoxan; HDC, high-dose chemotherapy; ASCT, autologous stem-cell
`transplantation; PEG, polyethylene glycol; MAB, monoclonal antibody;
`PDX, 10-propargyl-10-deazaaminopterin (novel antifolate).
`ⴱCytotoxic therapies include all chemotherapy-based treatment pro-
`grams, radiation, and radioimmunotherapies, and excludes all biologically
`based treatments including rituximab- and interferon-based treatments.
`
`thrombocytopenia with platelet counts ranging between
`60,000 and 95,000/mL. This toxicity persisted for 2 weeks,
`which per the earlier versions of the study rendered him
`
`ineligible to be re-treated. As discussed, the platelet require-
`ments were downgraded in subsequent amendments to the
`study. The second ineligible patient had a history of MCL
`with extensive gastrointestinal involvement. After several
`doses of bortezomib, he developed a grade 3 diarrhea with a
`documented Clostridium difficile infection that did not re-
`solve during a 2-week period of observation and treatment,
`rendering him ineligible for treatment on study. A
`follow-up computed tomography scan of the abdomen and
`pelvis revealed marked progression of his disease.
`There was a roughly even distribution in sex (54% male
`patients), with an unintended bias toward a largely white
`non-Hispanic population (92%). The median age of 63
`years approximates the median age for patients with indo-
`lent lymphomas in general. The indolent lymphomas rep-
`resent a vast diversity of biology and different disease
`subtypes, most of which are represented in this population.
`Approximately one third of all patients had follicular lym-
`phoma (38%), including grades 1 to 3. Eleven patients
`(42%) had MCL, whereas three patients (12%) had SLL and
`two patients (8%) had marginal zone lymphoma. The me-
`dian number of all prior therapies was three (range, 0 to 5),
`whereas the median number of prior cytotoxic chemother-
`apy regimens was also three (range, 0 to 3). All patients
`(with the exception of the one patient with no history of
`prior treatment) had been treated with at least one form of
`an alkylator-based treatment program, whereas only 15%
`received a purine analog– based regimen. Three patients
`had undergone prior peripheral-blood stem cell transplan-
`tation, and two had received prior radioimmunotherapy
`(tositumomab and iodine-131 tositumomab or yttrium-90
`ibritumomab tiuxetan). These patients did not seem to
`exhibit any more toxicity than that of other patients partic-
`ipating in the study, and in fact, two of the five attained
`major remissions (CRu and PR). Fifteen of the 26 patients
`received prior rituximab as a single agent, with five of those
`patients receiving two or more courses of antibody therapy.
`Nine (35%) patients received prior radiotherapy. Only one
`patient was registered to the study who had received no
`prior therapy of any sort.
`
`Dose Modifications
`In general, the treatment was well tolerated. For the 26
`registered patients, a total of 103 cycles of bortezomib were
`administered, which included 378 doses of the drug. The
`average number of cycles and doses received per patient was
`approximately four and 14.5, respectively. There was no
`difference in the amount of therapy administered to re-
`sponders or nonresponders (4.2 v 4.3 cycles/patient and
`16.3 v 15 doses per patient for responders and nonre-
`sponders, respectively). Thirteen patients (50%) missed at
`least one dose of drug. Among these patients, the median
`number of doses missed was four. The most common rea-
`son for a missed dose was thrombocytopenia, and the bulk
`
`www.jco.org
`
`679
`
`Downloaded from ascopubs.org by 12.40.227.4 on March 2, 2018 from 012.040.227.004
`
`Copyright © 2018 American Society of Clinical Oncology. All rights reserved.
`
`IPR2018-00685
`Celgene Ex. 2038, Page 4
`
`

`

`O’Connor et al
`
`of these missed doses occurred early in the study, when the
`more stringent platelet count requirements were in place
`(ie, ⱖ 100,000/␮L). After the modification to platelet count
`requirements, there have been no missed doses for throm-
`bocytopenia. Missed doses were also attributed to neuro-
`toxicity in four patients (grade 2 to 3 sensory neuropathy),
`gastrointestinal toxicity in one patient who had bulky intra-
`abdominal disease that was shrinking on treatment, rash in
`two patients, and asthenia in three patients. Dose reduc-
`tions from 1.5 to 1.3 mg/m2 were noted in 14 patients
`(54%), largely as the result of thrombocytopenia before the
`protocol revision. Five patients (19%) had additional dose
`reductions from 1.3 to 1.1 mg/m2. Twelve of the 25 patients
`(ie, 46%) tolerated the 1.5 mg/m2 dose without significant
`toxicity or need for dose reduction.
`Toxicity
`A summary of the toxicity observed during this study is
`presented in Table 2. Other than thrombocytopenia, there
`were no other dose-limiting hematologic toxicities. Of note,
`grade 3 lymphopenia was seen in 14 patients (60%), sug-
`
`Table 2. Major Hematologic and Nonhematologic Toxicities for
`Patients Receiving Bortezomib
`
`Toxicity
`
`Grade 1 Grade 2 Grade 3 Grade 4
`
`AST
`ALT
`Alkaline phosphatase
`Bilirubin
`Creatinine
`Hemoglobin
`Hyperglycemia
`Hyperkalemia
`Hypernatremia
`Hypoalbuminemia
`Hypocalcemia
`Hypoglycemia
`Hypokalemia
`Hypomagnesemia
`Hyponatremia
`Hypophosphatemia
`Infection without neutropenia
`Leukocytes
`Lymphopenia
`Nausea
`Neuropathic pain
`Neuropathy, sensory
`Neutrophils
`PT
`PTT
`Thrombocytopenia
`Vasculitis
`Anorexia
`Constipation
`Diarrhea
`Fatigue
`
`14
`6
`8
`4
`5
`17
`20
`3
`7
`10
`6
`6
`3
`3
`6
`0
`0
`14
`0
`11
`13
`15
`10
`3
`6
`21
`2
`2
`8
`11
`14
`
`1
`3
`1
`3
`1
`5
`0
`0
`0
`2
`5
`0
`0
`0
`0
`3
`0
`9
`0
`1
`2
`2
`7
`2
`1
`10
`1
`1
`2
`1
`3
`
`0
`1
`1
`0
`0
`1
`0
`1
`0
`0
`0
`0
`2
`0
`2
`0
`2
`1
`14
`1
`0
`2
`1
`2
`0
`7
`0
`1
`1
`0
`1
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`1
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`
`Abbreviations: PT, prothrombin time; PTT, partial thromboplastin time.
`
`gesting intrinsic sensitivity of the lymphoid lineage to the
`effects of proteasome inhibition. This was not associated
`with an increase in opportunistic infections, although two
`patients developed an outbreak of shingles on study. Seven
`patients developed grade 3 thrombocytopenia. In general,
`all cytopenias were short lived, and resolved to baseline
`during the week off therapy. Electrolyte abnormalities in-
`cluding hyponatremia, which was a dose-limiting toxicity
`in the original phase I study, were also commonly seen
`among these patients. Three patients experienced hypona-
`tremia (grade 3 or greater), two patients experienced hypo-
`kalemia (grade 3 toxicity), and one patient experienced
`hyperkalemia (grade 3 toxicity). In addition, several pa-
`tients experienced an unusual rash, which after biopsy re-
`vealed a small-vessel necrotizing vasculitis. The rash
`typically appeared during cycles 3 and 4, approximately
`during the third or fourth doses of those cycles. The rash
`resolved during the week of rest from treatment (week 3),
`and met National Cancer Institute common toxicity criteria
`for a grade 1 vasculitis. In all cases, the rash was not associated
`with any adverse sequelae, nor could we document an associ-
`ated presence of perinuclear antineutrophil cytoplasmic anti-
`body or cytoplasmic antineutrophil cytoplasmic antibody.
`Overall, two patients experienced grade 3 sensory neu-
`ropathy; one of these patients progressed into a grade 3
`sensorimotor neuropathy. A thorough work-up of her con-
`dition at the time, including multiple magnetic resonance
`imaging studies and lumbar punctures to rule out lepto-
`meningeal disease, and an electromyelogram (EMG), were
`all consistent with a likely pre-existing underlying neuro-
`logic disorder. The EMG analysis revealed complex repeti-
`tive discharges in both the paraspinal muscles and tongue.
`The electrophysiologic studies were interpreted as being
`most consistent with severe motor and sensory axonal poly-
`neuropathy, although this possibly was not attributable to
`bortezomib. Electrophysiologic evidence for a widespread
`denervating process affecting motor nerves at all spinal
`levels and the tongue was also noted. A sural nerve biopsy
`confirmed marked axonal loss and axonal degeneration.
`
`Response
`Table 3 lists the response data from the 26 registered
`patients; 24 of those patients were assessable for response.
`Figure 1 is a histogram of the response as a function of the
`disease subtype and overall best and worse response. Seven
`of the nine assessable patients with follicular lymphoma
`achieved objective remissions of their disease, with one CR
`and one CRu. In all patients, the remissions were confirmed
`at 1 month, and in all patients, the bulkiness of the disease
`did not seem to influence response. One patient in PR is still
`in active follow-up more than 20 months from the end of
`treatment with bortezomib. She achieved only a 7-month
`duration of remission from her prior therapy, which was
`rituximab. In all, 77% of the patients with follicular
`
`680
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`Downloaded from ascopubs.org by 12.40.227.4 on March 2, 2018 from 012.040.227.004
`
`Copyright © 2018 American Society of Clinical Oncology. All rights reserved.
`
`IPR2018-00685
`Celgene Ex. 2038, Page 5
`
`

`

`Bortezomib in Indolent NHL and MCL
`
`Table 3. Response Data for Patients Receiving Bortezomib
`
`Disease (assessable)
`
`CR/CRu
`
`PR
`
`SD
`
`POD
`
`Response Rate (%)
`
`Follicular lymphoma (9 of 10 assessable)
`Mantle cell lymphoma (10 of 11 assessable)
`Small lymphocytic lymphoma or CLL (3 of 3 assessable)
`Marginal zone lymphoma (2 of 2 assessable)
`
`1/1
`0/1
`0
`0
`
`5
`4
`0
`2
`
`1
`4
`2
`0
`
`1
`1
`1
`0
`
`77
`50
`0
`100
`
`Duration of Responseⴱ
`(months)
`
`3⫹, 4, 6, 9, 10, 20⫹, LTF
`6, 6⫹, 7⫹, 9⫹, 19†
`NA
`8⫹, 11⫹
`
`Abbreviations: CR, complete remission; CRu, unconfirmed complete remission; LTF, lost to follow-up; PR, partial remission; SD, stable disease; POD,
`progression of disease; NA, not available; CLL, chronic lymphocytic leukemia.
`ⴱOverall response rate, 58% (14 of 24 assessable patients).
`†This patient was re-treated and achieved a second PR after having a 19-month remission following his first course of therapy. His second PR lasted
`approximately 7 months.
`
`lymphoma responded to therapy. Five of 10 assessable pa-
`tients with MCL also achieved major remissions, with four
`of those patients still in active follow-up. One of these
`patients achieved a PR lasting about 6 months with cyclo-
`phosphamide, doxorubicin, vincristine, and prednisone
`(CHOP) chemotherapy followed by rituximab. After four
`cycles of bortezomib, he achieved a major PR, with more
`than 80% shrinkage of his disease that lasted approximately
`19 months. After his response, the protocol was amended to
`allow patients to be re-treated who achieved a PR or better
`lasting at least 6 months. On re-treatment, he attained a
`second PR that lasted approximately 7 months, and has now
`begun his third course of therapy. At the time of this report, the
`patient has achieved more than 26 months of remission from
`single-agent bortezomib after achieving only 6 months of re-
`mission from standard CHOP followed by rituximab.
`Interestingly, one patient with MCL, who had a prom-
`inent leukemic component to his disease and who was
`scored as having achieved SD, had a dramatic reduction in
`his leukemic disease after receiving bortezomib. He re-
`ceived four cycles of therapy in total. His pretreatment bone
`
`marrow revealed more than 95% involvement with MCL.
`After treatment with four cycles of bortezomib, his marrow
`involvement was reduced to approximately 70%, and the
`WBC and absolute lymphocyte count returned to normal
`levels. He voluntarily removed himself from study to pur-
`sue a syngeneic transplant.
`To date, the collective experience in patients with SLL
`suggests that this particular subtype of NHL is not respon-
`sive to bortezomib. All three patients on our study failed to
`demonstrate any reduction in their total disease burden. In
`contrast, both patients with nodal marginal zone lym-
`phoma responded rapidly to bortezomib, and although
`both patients are still in active follow-up, the duration of
`their responses to date are approximately 8⫹ (in a previ-
`ously untreated patient) and 11⫹ months.
`
`DISCUSSION
`
`Numerous lines of evidence have clearly established that
`the proteasome is a valid target in the treatment of
`
`Fig 1. Histogram showing distribution of
`best and worst responses seen for patients
`participating in the study while receiving
`treatment. Three patients not represented
`include two patients who were ineligible for
`response evaluation and one patient with
`a predominantly leukemic phase of dis-
`ease without measurable adenopathy.
`CR, complete remission; CRu, uncon-
`firmed CR; FL, follicular lymphoma; SLL,
`B-cell small
`lymphocytic lymphoma; MZL,
`marginal zone lymphoma.
`
`www.jco.org
`
`681
`
`Downloaded from ascopubs.org by 12.40.227.4 on March 2, 2018 from 012.040.227.004
`
`Copyright © 2018 American Society of Clinical Oncology. All rights reserved.
`
`IPR2018-00685
`Celgene Ex. 2038, Page 6
`
`

`

`O’Connor et al
`
`hematologic malignancies. The recent approval of bort-
`ezomib for the treatment of multiple myeloma has estab-
`lished this class of molecules in the treatment of
`lymphoproliferative malignancies.25
`Although proteasome inhibition represents a novel
`target, it by no means is an approach that results in a
`singular biologic effect. In fact, inhibition of proteasome
`function produces a wide panoply of biologic effects that
`can lead to changes in the balance of pro- and antiapoptotic
`proteins in the mitochondrial membrane,26,27,28 NF-␬B–
`influenced gene transcription, CAM expression, cytokine
`production and signaling, and angiogenesis. Although inhi-
`bition of the proteasome clearly facilit