Randomized Phase II Study of BR96-Doxorubicin Conjugate
`in Patients With Metastatic Breast Cancer
`
`By Anthony W. Tolcher, Steven Sugarman, Karen A. Gelmon, Roger Cohen, Mansoor Saleh, Claudine Isaacs,
`Leslie Young, Diane Healey, Nicole Onetto, and William Slichenmyer
`
`Purpose: BMS-182248-1 (BR96-doxorubicin immu-
`noconjugate) is a chimeric human/mouse monoclonal
`antibody linked to approximately eight doxorubicin
`molecules. The antibody is directed against the Lewis-Y
`antigen, which is expressed on 75% of all breast can-
`cers but is limited in expression on normal tissues.
`Preclinical xenograft models demonstrated significant
`antitumor activity, including cures. A randomized phase
`II design was chosen to estimate the activity of the
`BR96-doxorubicin conjugate in metastatic breast cancer
`in a study population with confirmed sensitivity to
`single-agent doxorubicin.
`PatientsandMethods: Patients with measurable meta-
`static breast cancer and immunohistochemical evidence
`of Lewis-Y expression on their tumor received either
`BR96-doxorubicin conjugate 700 mg/m2 IV over 24
`hours or doxorubicin 60 mg/m2 every 3 weeks. Pa-
`tients were stratified on the basis of prior doxorubicin
`exposure, visceral disease, and institution. Cross-over
`to the opposite treatment arm was allowed with pro-
`gressive or persistently stable disease.
`
`Results: Twenty-three patients who had received a
`median of one prior chemotherapy regimen were as-
`sessable. There was one partial response (7%) in 14
`patients receiving the BR96-doxorubicin conjugate and
`one complete response and three partial responses
`(44%) in nine assessable patients receiving doxorubi-
`cin. No patient experienced a clinically significant hyper-
`sensitivity reaction. The toxicities were significantly dif-
`ferent between the two treatment groups, with the
`BR96-doxorubicin conjugate group having limited he-
`matologic toxicity, whereas gastrointestinal toxicities,
`including marked serum amylase and lipase elevations,
`nausea, and vomiting with gastritis, were prominent.
`Conclusion: The BR96-doxorubicin immunoconju-
`gate has limited clinical antitumor activity in metastatic
`breast cancer. The gastrointestinal toxicities likely repre-
`sent binding of the agent to normal tissues expressing
`the target antigen and may have compromised the
`delivery of the immunoconjugate to the tumor sites.
`JClinOncol17:478-484.r 1999byAmericanSociety
`ofClinicalOncology.
`
`BMS 182248-01 (BR96-doxorubicin immunoconjugate)
`
`is an immunoconjugate developed to deliver doxorubi-
`cin specifically to tumor cells that express the tumor-
`associated surface antigen Lewis-Y while sparing normal
`tissues. The BR96-doxorubicin conjugate is a part human,
`part mouse chimeric immunoglobulin (Ig) conjugated to
`approximately eight doxorubicin molecules (Fig 1). In
`preclinical studies, after BR96-doxorubicin conjugate binds
`to the Lewis-Y surface antigen, it is rapidly internalized, the
`doxorubicin molecules are released from the immunoglobu-
`lin by hydrolysis of the acid-labile linkage, and cytotoxicity
`results.1,2 Lewis-Y is abundantly expressed on the surface of
`
`From the British Columbia Cancer Agency, Vancouver, British
`Columbia V5Z 4E6, Canada; Stony Brook Health Science Center, State
`University of New York (SUNY), Stony Brook, NY; University of Virginia
`Health Sciences Center, Charlottesville, VA; University of Alabama,
`Birmingham, AL; Georgetown University Medical Center, Washington,
`DC; and Bristol-Myers Squibb Pharmaceutical Research Institute,
`Wallingford, CT.
`Submitted May 1, 1998; accepted October 22, 1998.
`Supported by Bristol-Myers Squibb Pharmaceutical Research Insti-
`tute, Wallingford, CT 06492-7660.
`Address reprint requests to Anthony W. Tolcher, MD, Institute for
`Drug Development, Cancer Therapy and Research Center, Suite 250,
`8122 Datapoint Dr, San Antonio, TX 78229.
`r 1999 by American Society of Clinical Oncology.
`0732-183X/99/1702-0478$3.00/0
`
`many human carcinomas, including breast, lung, colon, and
`ovary carcinomas.3 Most normal tissues do not express the
`Lewis-Y antigen, except for epithelial cells of the esopha-
`gus, stomach, and proximal small intestine and some acinar
`cells of the pancreas.3
`The BR96-doxorubicin conjugate demonstrated a broad
`spectrum of antitumor activity in xenograft tumor models,
`including cures in established L2987 lung, RCA colon, and
`MCF-7 breast tumors.1,4 This activity was in marked con-
`trast to single-agent doxorubicin, which did not produce
`cures at its optimal dose and schedule in the same xenograft
`models.1 Furthermore, equivalent antitumor activity was
`achieved with the BR96-doxorubicin conjugate at 13% of
`the comparable free doxorubicin dose. This finding suggests
`preferential delivery of the doxorubicin to the Lewis-Y–
`expressing tumor by the immunoconjugate.1 In rodent
`toxicology studies, the BR96-doxorubicin conjugate was
`significantly less toxic, including cardiotoxicity, than equiva-
`lent doses of doxorubicin.5 An acute enteropathy was
`observed in dogs that was not observed in rodents and that
`differed from the toxicity normally associated with doxoru-
`bicin.6 The acute enteropathy was believed to be related to
`the binding of the antibody to Lewis-Y–related antigens
`expressed by gastrointestinal epithelial cells.
`Clinical
`trials were initiated shortly after the broad
`spectrum of antitumor activity was demonstrated preclini-
`
`478
`
`JournalofClinicalOncology,Vol 17, No 2 (February), 1999: pp 478-484
`
`Downloaded from jco.ascopubs.org on May 23, 2014. For personal use only. No other uses without permission.
`Copyright © 1999 American Society of Clinical Oncology. All rights reserved.
`
`IMMUNOGEN 2010, pg. 1
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`BR96-DOXORUBICIN IN METASTATIC BREAST CANCER
`
`479
`
`PATIENTS AND METHODS
`
`Eligibility
`
`Patients aged 18 years or older with histologically confirmed,
`measurable metastatic breast cancer were eligible for this study. Patients
`were permitted to have had one prior chemotherapy regimen in either
`the adjuvant or metastatic setting. Prior chemotherapy for metastatic
`disease could not contain anthracyclines. However, prior adjuvant
`chemotherapy could include an anthracycline if the patient had relapsed
`more than 12 months after the completion of the chemotherapy and the
`cumulative dose was equal to or less than 300 mg/m2 doxorubicin.
`Patients who received nonanthracycline-containing adjuvant chemother-
`apy must have relapsed more than 3 months after treatment completion.
`Patients were permitted prior hormone therapy for adjuvant and
`metastatic disease; however, patients who had previously responded to
`hormone therapy had to have discontinued it at least 4 weeks before
`randomization and tumor evaluation was repeated after a 4-week
`washout interval. Prior radiotherapy was acceptable if it was completed
`4 weeks before study entry.
`Before randomization, all patients were required to have a breast
`cancer specimen positive for Lewis-Y antigen by immunohistochemis-
`try, a normal left ventricular ejection fraction, an Eastern Cooperative
`Oncology Group performance status of 0 or 1, or a Karnofsky
`performance status of more than 80%. Laboratory requirements in-
`cluded the following: a hemoglobin count of more than 10 g/dL; a
`granulocyte count of 1,500 cells/µL or higher; blood urea nitrogen and
`serum creatinine levels within 1.25 times the upper limit of normal; a
`bilirubin level of less than 1.5 mg/dL; AST and ALT levels within 2.5
`times the institutional upper limit of normal; and serum lipase or
`amylase levels that were less than twice the institutional upper limit.
`Alkaline phosphatase elevations of more than 2.5 times the upper limit
`of normal were permitted if bone metastases were documented. Women
`of child-bearing potential were required to have a negative serum or
`urine pregnancy test result within 72 hours before the start of study
`medication and to practice contraception.
`Patients were excluded if they had been exposed to therapeutic or
`diagnostic murine, murine/human chimeric, or humanized monoclonal
`antibodies within 6 months before randomization. A history of neo-
`plasm other than breast cancer within 5 years of study entry was a
`reason for exclusion, with the exception of nonmelanoma skin cancer or
`curatively treated carcinoma-in-situ of the cervix. Patients were not
`permitted to have CNS metastases that required active treatment or
`receive concomitant cytotoxic, hormone, radiation, or investigational
`therapy while participating in this study.
`The study was approved by each participating institution’s institu-
`tional review board, and written informed consent according to the
`guidelines of the participating institutions was obtained from patients
`before study entry.
`
`Study Design
`
`the eligibility criteria were randomized after
`Patients who met
`stratification to receive either BR96-doxorubicin conjugate 700 mg/m2
`intravenously (IV) or single-agent doxorubicin 60 mg/m2 IV every 3
`weeks.
`BR96-doxorubicin conjugate (BMS-182248-01; Bristol-Myers
`Squibb, Wallingford, CT) was administered by continuous IV infusion
`over 24 hours once every 21 days. The full calculated dose was divided
`into four equal doses and prepared in 250 ml of normal saline or 5%
`dextrose water; each of the four doses was administered over 6 hours.
`Premedication consisted of oral dexamethasone 20 mg daily starting 2
`days before the start of infusion and continued for 2 days after therapy.
`
`Fig 1. Schematic of BR96-doxorubicin immunoconjugate (BMS-182248-
`01). Abbreviations: HC, immunoglobulin heavy chain; LC, immunoglobulin
`light chain; Dox-linker, acid labile doxorubicin linkage.
`
`cally. In the phase I studies, vomiting with hematemesis was
`dose-limiting, and an exudative gastritis was found at
`endoscopy.6,7 Premedication with corticosteroids, 5-hydroxy-
`tryptamine-3 antagonists, and a 24-hour infusion schedule
`led to the recommended phase II dosage of 700 mg/m2
`BR96-doxorubicin every 3 weeks (21 mg/m2 conjugated
`doxorubicin).8 In addition to the reversible gastritis, toxici-
`ties included asymptomatic alterations in pancreatic en-
`zymes, most prominently serum lipases, mild hypersensitiv-
`ity reactions characterized by fever or
`rash without
`anaphylactic reactions, modest hematologic toxicity of ane-
`mia, and rare cases of neutropenia. No alopecia, mucositis,
`or cardiomyopathy was reported. During the phase I study of
`this schedule,
`two partial responses were observed in
`patients with breast and gastric carcinoma.8
`On the basis of provocative preclinical data and the
`encouraging phase I study, we initiated a randomized phase
`II study to determine the antitumor activity of BR96-
`doxorubicin conjugate in metastatic breast cancer. The
`randomized phase II design using a single-agent doxorubi-
`cin reference arm was chosen to estimate the activity of the
`BR96-doxorubicin conjugate in metastatic breast cancer and
`confirm the sensitivity of the enrolled population to single-
`agent doxorubicin. Moreover, this design would permit a
`rapid determination of the future development of this drug in
`breast cancer.
`
`Downloaded from jco.ascopubs.org on May 23, 2014. For personal use only. No other uses without permission.
`Copyright © 1999 American Society of Clinical Oncology. All rights reserved.
`
`IMMUNOGEN 2010, pg. 2
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`480
`
`TOLCHER ET AL
`
`Antiemetics, including 5-hydroxytryptamine-3 antagonists, were admin-
`istered according to institutional policy. Lorazepam 1 to 2 mg orally
`(PO) or IV was administered at least 30 minutes before the start of
`infusion. Vital signs were monitored in all patients receiving the
`BR96-doxorubicin conjugate immediately before, 15 minutes and 1
`hour after the start, and at the completion of the infusion.
`Doxorubicin was reconstituted with sodium chloride injection USP
`(0.9% normal saline) or sterile water for injection. Doxorubicin was
`given as a bolus injection IV once every 3 weeks. Vitals signs were
`monitored before treatment and at completion of the infusion.
`
`lesions determined by two observations not less than 4 weeks apart. No
`simultaneous increase in the size of any lesion or appearance of new
`lesions could occur. Nonmeasurable lesions were required to remain
`stable or regress to meet the PR criterion. Stable disease was defined as
`a decrease of less than 50% in total tumor size up to an increase of less
`than 25% in the total tumor size. To meet the stable disease criterion, no
`new lesions could develop. Progressive disease was the unequivocal
`increase of at least 25% from the best response or baseline in one or
`more measurable lesions. The appearance of new lesions constituted
`progressive disease.
`
`Toxicity Evaluation and Dose Modifications
`
`Toxicity was evaluated and graded according to the National Cancer
`Institute common toxicity criteria, with the exception that serum lipase
`was graded using the same scale as serum amylase.
`The drug dose was reduced 25% when febrile neutropenia occurred
`or a platelet transfusion was required. Failure to recover the absolute
`neutrophil count to 1,000/µL or higher or the platelet count to 100,000
`cells/µL or higher required a delay of treatment until recovery and a
`25% dose reduction. Grade 3 or 4 nausea and vomiting for more than 48
`hours’ duration required a 25% dose reduction. If symptoms persisted at
`the planned time of re-treatment, patient participation was delayed until
`recovery. Gastrointestinal hemorrhage up to grade 1 was permitted as
`long as patients had recovered by the day of re-treatment. Grade 2 or
`greater gastrointestinal hemorrhage required discontinuation of therapy.
`Pancreatic toxicity was defined as elevations of serum amylase or lipase
`with or without symptoms consistent with pancreatic inflammation.
`Any grade 4 pancreatic toxicity, or the presence of symptoms regardless
`of grade, required a 25% dose reduction and a resolution of the
`biochemical toxicity to grade 0 to 3 with resolution of symptoms before
`re-treatment.
`All other grade 3 or 4 nonhematologic toxicities, with the exception
`of alopecia, warranted a 25% dose modification with resolution of
`toxicity to grade 0 or 1 by the date of re-treatment. Grade 1
`cardiovascular toxicities were permitted provided that patients recov-
`ered by the time of re-treatment. Any grade 2 or greater cardiovascular
`toxicity required discontinuation from further treatment.
`
`Treatment Duration and Cross-Over Criteria
`
`Patients were treated with either BR96-doxorubicin conjugate 700
`mg/m2 IV over 24 hours or doxorubicin 60 mg/m2 every 21 days for a
`maximum of six cycles. Tumor response was assessed every two cycles.
`Any patient receiving a lifetime cumulative dose of unconjugated
`doxorubicin of 550 mg/m2 was discontinued from treatment. Patients
`were permitted to receive additional cycles of therapy in the absence of
`disease progression if it was considered in their best interest by the
`treating physician.
`Patients were permitted to cross-over to the alternate treatment arm if
`they demonstrated either disease progression or stable disease after four
`cycles and if it was considered in the patient’s best interest by the
`treating physician. Cross-over was permitted only if the patient met the
`original eligibility criteria.
`All patients who received one dose of the study drug were assessable
`for toxicity. All patients who received at least two cycles of treatment
`were considered assessable for response.
`A complete response (CR) was defined as disappearance of all
`clinically evident tumor, including normalization of any tumor markers
`determined by two observations not less than 4 weeks apart. A partial
`response (PR) was defined as a 50% or greater decrease in the sum of
`the products of the bidimensional measurements of the measured
`
`Anti-Immunoconjugate Antibody Assessment
`(Human Anti-Mouse Antibody)
`
`Serum samples were collected from patients for human anti-mouse
`antibody (HAMA) analysis before they received BR96-doxorubicin,
`after the first cycle of therapy (day 22), and 3 to 6 weeks after
`completion of treatment. Serum samples were assessed for immunoge-
`nicity using an enzyme-linked immunoadsorbent assay specific for
`F(ab8)2 fragments of BMS-182248, BR96 F(ab8)2 Dox. Briefly, Immu-
`lon II enzyme-linked immunoadsorbent assay plates (Dynex, Chantilly,
`VA) were coated overnight with BR96 F(ab8)2 Dox 100 µL at 2 µg/mL
`in phosphate-buffered saline [PBS]). Plates were blocked with 100 µL
`of PBS containing 5% goat sera. The plates were washed (PBS/0.5%
`goat sera Tween 20, pH 7.4), and two-fold serial dilutions of the test sera
`in PBS/5% goat sera were added to the appropriate wells and incubated
`for 2 to 4 hours at 37oC. After washing, bound human antibodies were
`detected using alkaline phosphatase–conjugated, Fc-specific goat anti-
`human IgG and IgM antibodies (Sigma, St. Louis, MO) and IgA
`antibody (Biosource, Camarillo, CA) (100 µL at 1:10,000, 1:7,500, and
`1:5,000, respectively) incubated for 1 to 2 hours at 37oC. Subsequently,
`the plates were washed and substrate (1 mg/mL p-nitrophenyl phos-
`phate in diethanolamine buffer) was added to each well. After incuba-
`tion for 30 minutes at 25oC, 50 µL of 3N NaOH stop reagent was added
`and the absorbance was read (dual wavelength, 405 and 550 nm).
`Results are reported as units per milliliter based on the hyperimmunized
`monkey anti-murine BR96 reference standard. The plate background
`absorbance was subtracted, and unit-per-milliliter values were calcu-
`lated for each dilution of each patient’s sample by comparing it to the
`standard curve. The unit-per-milliliter value for each patient’s sample
`was then determined by calculating the mean of three consecutive
`dilutions with the lowest coefficient of variability. Plate background was
`determined as the absorbance measurement recorded in the absence of
`serum. A patient was considered to have seroconverted when the
`individual’s response was greater than the mean plus three standard
`deviations of all the patients’ predose values (predose values were
`estimated from the means in the three ongoing clinical studies, n 5 32)
`and also greater than two times the pretreatment value for that
`individual patient.
`
`Statistical Analysis
`
`The study was designed as a randomized phase II study assessing
`antitumor activity and the safety of BR96-doxorubicin conjugate every
`21 days and doxorubicin every 21 days. Eligible patients were stratified
`according to prior doxorubicin exposure, the dominant site of disease
`(visceral v other), and institution. Randomization occurred at a central
`office (Wallingford, CT). A Gehan two-stage design was used to
`optimize the number of patients entered onto this phase II study.9 The
`purpose of the first stage was to determine whether the given dose and
`schedule of the BR96-doxorubicin conjugate were worth further study
`in metastatic breast cancer. Before study initiation, an acceptable level
`
`Downloaded from jco.ascopubs.org on May 23, 2014. For personal use only. No other uses without permission.
`Copyright © 1999 American Society of Clinical Oncology. All rights reserved.
`
`IMMUNOGEN 2010, pg. 3
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`BR96-DOXORUBICIN IN METASTATIC BREAST CANCER
`
`of antitumor activity was defined as more than or equal to 30% for
`continued accrual beyond 15 patients to occur. After the entry of nine
`patients to each arm, the investigators met to decide whether accrual
`should continue based on the response rate observed in the experimental
`arm compared with the reference arm. Analysis at that time suggested
`that if the drug had a 30% response rate, one or more successes should
`have been observed in the first nine patients entered into the experimen-
`tal arm (rejection error, # 5%, beta).9 Therefore, it was decided that the
`protocol should be amended, further randomization to the doxorubicin
`arm should cease, and further accrual should continue only to the
`BR96-doxorubicin arm to determine whether the experimental agent
`met the conventional phase II level of 20% antitumor activity.
`The toxicities (hematologic and biochemical) experienced by the two
`treatment groups were compared on the first cycle, and worst toxicity
`(both hematologic and nonhematologic) was compared over all cycles,
`using a two-sided Monte Carlo simulation, a permutation of a Wilcoxon
`rank sum statistic. The Bonferroni method was used because a large
`number of variables were compared, and the significance level of each
`variable was set at .002 to ensure a 0.05 experiment-wise error rate.
`
`RESULTS
`A total of 25 patients were entered onto this study: 10
`patients were randomized to receive the BR96-doxorubicin
`conjugate, and 10 patients were randomized to single-agent
`doxorubicin. After the protocol was amended, five addi-
`tional patients received the BR96-doxorubicin conjugate
`without randomization. Two randomized patients (one pa-
`tient in each treatment arm) were never treated. One of these
`patients refused their assigned single-agent doxorubicin
`treatment, and one patient rapidly deteriorated from disease
`before BR96-doxorubicin conjugate treatment. The median
`age of patients entered was 52 years (range, 25 to 65 years),
`and they had a median of two sites of metastatic disease.
`Thirteen patients had received prior chemotherapy (eight
`patients in the BR96-doxorubicin conjugate arm and five in
`the doxorubicin arm). Two patients treated with BR96-
`doxorubicin conjugate and one patient randomized to the
`doxorubicin arm had received prior adjuvant doxorubicin.
`All patients treated were assessable for response and toxic-
`ity. Patient characteristics are listed in Table 1.
`Forty-five cycles were administered to the 14 patients
`treated initially with BR96-doxorubicin. In addition, 25
`cycles were administered to four patients who crossed over
`to receive BR96-doxorubicin conjugate after progression or
`persistently stable disease during single-agent doxorubicin
`treatment. Forty-one cycles of single-agent doxorubicin
`were administered to the nine randomized patients, whereas
`two cycles of doxorubicin were given to one patient who
`crossed over after disease progressed during BR96-
`doxorubicin treatment. Patients treated with the BR96-
`doxorubicin conjugate as initial therapy received a median
`of three cycles of therapy (range,
`two to five cycles),
`whereas patients treated with single-agent doxorubicin ini-
`tially received a median of four cycles (range, two to six
`
`481
`
`Table 1. Patient Characteristics
`
`BR96-
`doxorubicin
`
`Doxorubicin
`
`No. of patients entered
`No. of patients treated
`Age, years
`Median
`Range
`No. of patients with prior radiotherapy
`No. of patients with prior chemotherapy
`Adjuvant chemotherapy alone
`Metastatic chemotherapy alone
`Both
`No. of patients with prior doxorubicin adjuvant
`No. of patients with prior hormone therapy
`Median no. of hormone therapies
`Range
`Sites of disease
`Soft tissue
`Lymph node
`Lung
`Hepatic
`Breast
`
`15
`14
`
`52
`32-65
`9
`8
`7
`1
`1
`2
`13
`1
`0-3
`
`3
`5
`4
`9
`3
`
`10
`9
`
`51
`25-63
`5
`5
`5
`0
`0
`1
`4
`0
`0-3
`
`2
`5
`3
`4
`1
`
`cycles). In the four patients who crossed over to receive
`BR96-doxorubicin conjugate after disease progression dur-
`ing doxorubicin treatment, the median number of courses
`was five (range, one to 14 courses).
`
`Toxicity Data
`
`There were significant differences between the toxicities
`experienced by patients on the two randomized treatment
`arms. Hematologic toxicity, particularly neutropenia and
`leukopenia, was significantly more frequent with single-
`agent doxorubicin compared with BR96-doxorubicin conju-
`gate treatment (P # .0004). Abnormalities of serum amylase
`and lipase occurred frequently among patients treated with
`BR96-doxorubicin conjugate, whereas this toxicity was not
`observed in patients who received doxorubicin (P 5 .002
`and P 5 .002, respectively). Nonhematologic toxicities
`associated with BR96-doxorubicin included marked nausea,
`vomiting, symptoms of gastritis, and hematemesis in two
`patients. Vomiting occurred in 93% of patients treated with
`BR96-doxorubicin, and 43% of patients experienced grade 3
`or 4 vomiting. The doxorubicin arm had significantly less
`severe nausea and less frequent vomiting (P 5 .0028). No
`episodes of hematemesis were noted in the doxorubicin arm.
`In addition, a nonspecific, poorly localized abdominal pain
`without sequelae was observed in five patients who received
`BR96-doxorubicin. There was no evidence of cardiac toxic-
`ity nor clinically significant (. grade 1) hepatic or renal
`toxicity in either of the two treatment arms.
`No clinically significant hypersensitivity reactions oc-
`curred in patients randomized initially to the BR96-
`
`Downloaded from jco.ascopubs.org on May 23, 2014. For personal use only. No other uses without permission.
`Copyright © 1999 American Society of Clinical Oncology. All rights reserved.
`
`IMMUNOGEN 2010, pg. 4
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`482
`
`doxorubicin conjugate. Fever was noted in three patients and
`wheezing was observed in two patients treated with the
`BR96-doxorubicin conjugate. One patient who crossed over
`and received the BR96-doxorubicin conjugate experienced
`four hypersensitivity reactions on successive cycles, with
`symptoms characterized by cough and wheezing. This
`patient could be re-treated safely after temporary interrup-
`tion of the infusion. The toxicities are summarized in Tables
`2 and 3.
`
`Treatment Modification and Delay
`
`A dose reduction was required in eight of 45 cycles of
`BR96-doxorubicin conjugate and one of 41 cycles of
`single-agent doxorubicin. The reasons for dose reduction in
`the BR96-doxorubicin conjugate arm were all nonhemato-
`logic toxicities: grade 3 or 4 nausea and vomiting (two
`cycles), nausea, vomiting, gastritis, and abdominal pain (one
`cycle), gastritis/pancreatitis symptoms (one cycle), and
`marked elevation in serum lipase more than or equal to five
`times the upper limit of normal (four cycles). None of the
`patients who received the BR96-doxorubicin conjugate after
`cross-over required a dose reduction. The one dose reduction
`and delay in the doxorubicin arm was due to febrile
`neutropenia with documented bacteremia. There were eight
`treatment delays in three patients receiving the BR96-
`doxorubicin conjugate; however, none of them were due to
`toxicity.
`
`HAMA Analysis
`
`Five of nine patients assessable for analysis met the
`criteria for a positive HAMA response. The ratio of postdose
`to predose values at day 22 and after treatment completion
`ranged from 0.60 to 12.64 and from 0.51 to 12.88, respec-
`tively. Only two patients had at either time point a more than
`
`Table 2. Hematologic Toxicities by Grade for Randomized Patients Cycle 1
`and All Cycles by Initial Treatment Arm
`
`BR96-doxorubicin
`(n 5 9)
`
`Doxorubicin
`(n 5 9)
`
`Toxicity
`
`Cycle 1
`ANC
`WBC
`Platelets
`Hemoglobin
`All cycles
`ANC
`WBC
`Platelets
`Hemoglobin
`
`1
`
`0
`0
`0
`3
`
`0
`1
`1
`4
`
`2
`
`0
`0
`0
`2
`
`0
`0
`0
`4
`
`3
`
`0
`0
`0
`0
`
`0
`0
`0
`0
`
`4
`
`0
`0
`0
`0
`
`0
`0
`0
`0
`
`1
`
`0
`1
`1
`2
`
`0
`1
`1
`3
`
`2
`
`1
`3
`0
`1
`
`2
`4
`0
`1
`
`3
`
`2
`3
`0
`0
`
`1
`2
`0
`1
`
`4
`
`P
`
`5,
`1,
`0
`0
`
`6#
`2#
`0
`0
`
`.0004
`.0004
`.999
`.5
`
`.0004
`.0004
`1.0
`.24
`
`Abbreviations: ANC, absolute neutrophil count; WBC, WBC count.
`
`TOLCHER ET AL
`
`Table 3. Worst Nonhematologic Toxicities by Grade in Randomized Patients
`for All Cycles by Initial Treatment Arm
`
`Toxicity
`
`Amylase
`Lipase
`Fever
`Wheezing
`Abdominal pain
`Back pain
`Nausea
`Vomiting
`Gastritis
`Hematemesis
`Pancreatitis
`Alopecia
`Bone pain
`Arthralgia
`Sepsis
`Cardiac
`
`1
`
`3
`2
`1
`1
`1
`1
`3
`1
`0
`2
`0
`0
`2
`2
`0
`0
`
`BR96-doxorubicin
`(n 5 9)
`
`Doxorubicin
`(n 5 9)
`
`2
`
`1
`1
`0
`0
`1
`1
`2
`3
`0
`0
`0
`0
`1
`1
`0
`0
`
`3
`
`3
`1
`0
`0
`1
`1
`4
`3
`1
`0
`0
`0
`0
`0
`0
`0
`
`4
`
`0
`3
`0
`0
`0
`0
`0
`2
`0
`0
`1
`
`0
`0
`0
`0
`
`1
`
`0
`0
`0
`0
`1
`1
`4
`5
`1
`0
`0
`0
`1
`1
`0
`0
`
`2
`
`0
`0
`0
`0
`0
`0
`2
`2
`0
`0
`0
`8
`0
`0
`0
`0
`
`3
`
`0
`0
`0
`0
`0
`0
`1
`0
`0
`0
`0
`1
`0
`0
`1
`0
`
`4
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`
`0
`0
`0
`0
`
`P*
`
`.002
`.002
`
`.003
`
`.00004
`
`*Pvalues are provided for those toxicities that met or approached the criteria
`for significance as defined in the Patients and Methods section.
`
`or equal to 10-fold rise in titer; therefore, the response in
`these patients would be considered mild to moderate. The
`results are summarized in Table 4.
`
`Response Data
`
`One partial response (7%; hepatic metastases) was ob-
`served in 14 patients (95% confidence interval, 0 to 34%)
`initially treated with BR96-doxorubicin conjugate. In con-
`trast, four patients responded (one CR and three PR [44%];
`95% confidence interval, 14% to 79%) among the nine
`patients who received single-agent doxorubicin. Two pa-
`tients who crossed over to the BR96-doxorubicin arm after
`persistently stable disease through four cycles of doxorubi-
`cin had a PR in hepatic metastases.
`The response duration was 3 months for the patient
`treated initially with BR96-doxorubicin conjugate and 6 and
`61 months for the patients who responded to BR96-
`doxorubicin after cross-over. The response durations for the
`four patients randomized to doxorubicin and who responded
`were 2, 4, 10, and 15 months.
`
`DISCUSSION
`The immunoconjugate BR96-doxorubicin was developed
`to selectively deliver doxorubicin to tumors that express the
`tumor-associated surface antigen Lewis-Y while avoiding
`doxorubicin’s systemic toxicity. We determined that the
`BR96-doxorubicin conjugate on this schedule and dose has
`modest antitumor activity in metastatic breast cancer. This is
`in contrast to the doxorubicin arm, which as part of this
`randomized study, demonstrated a 44% response rate, includ-
`
`Downloaded from jco.ascopubs.org on May 23, 2014. For personal use only. No other uses without permission.
`Copyright © 1999 American Society of Clinical Oncology. All rights reserved.
`
`IMMUNOGEN 2010, pg. 5
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`BR96-DOXORUBICIN IN METASTATIC BREAST CANCER
`
`483
`
`Table 4. HAMA Analysis of Nine Patients Who Received BR96-Doxorubicin Conjugate
`
`Sample
`Time
`
`5-001
`
`5-002*
`
`5-006
`
`6-007
`
`6-009*
`
`6-012
`
`6-016*
`
`6-018*
`
`6-021*
`
`Patient No.
`
`Antigen F(ab8)2 Dox
`
`Cycle 1, day 1
`Cycle 1, day 22
`
`21 days after completion
`of treatment
`
`Cycle 1, day 1
`Cycle 1, day 22
`
`21 days after completion
`of treatment
`
`Response (U/mL)
`
`25
`316 †
`
`185
`
`43
`26
`
`22
`
`103
`119
`
`67
`
`46
`116
`
`28
`
`Ratio of Postdose to Predose Values
`
`42
`167
`
`25
`
`7
`58
`
`20
`
`74
`55
`
`225
`
`17
`92
`
`219
`
`33
`125
`
`60
`
`1.00
`2.52
`
`0.61
`
`1.00
`12.64
`
`7.40
`
`1.00
`0.60
`
`0.51
`
`1.00
`1.16
`
`0.65
`
`1.00
`3.98
`
`0.60
`
`1.00
`8.29
`
`2.86
`
`1.00
`0.74
`
`3.04
`
`1.00
`5.41
`
`12.88
`
`1.00
`3.79
`
`1.82
`
`NOTE. A positive response is any response that is greater than the mean and three standard deviations (86) of the predose of all the patients tested and also greater
`than two times the predose level for that patient.
`*Patient had a positive response.
`†Boxed values are those that fit either of the criteria for a positive response.
`
`ing one CR. These results, therefore, do not confirm the
`preclinical data, which show that the immunoconjugate was
`superior to single-agent doxorubicin1.
`The failure of BR96-doxorubicin to produce a clinically
`significant number of responses, despite the strong preclini-
`cal evidence, deserves examination. The patient population
`entered should have been sensitive to doxorubicin because
`they were minimally pretreated and were not eligible if they
`were refractory to adjuvant doxorubicin. The response rate
`in the doxorubicin reference arm supports this contention;
`therefore, the failure of BR96-doxorubicin was not due to
`intrinsic doxorubicin resistance. We believe that the rela-
`tively low dose of doxorubicin contained in the immunocon-
`jugate treatment (approximately one fourth of the dose in the
`free doxorubicin arm) should not have limited the antitumor
`activity significantly. In preclinical models, equivalent anti-
`tumor activity was achieved with the BR96-doxorubicin
`immunoconjugate at one eighth of the dose of free doxorubi-
`cin, and superior activity was noted for
`the BR96-
`doxorubicin immunoconjugate at doses above this level. The
`HAMA response to the murine F(ab8) fragment was modest,
`as would be expected in a chimeric part human, part murine
`immunoglobulin. Data from the prior phase I clinical trial
`demonstrated bound antibody to tumor cells in tumor biopsy
`specimens and the presence of doxorubicin in tumor cell
`nuclei, with the absence of a significant HAMA response.
`This suggests that targeting, binding, and delivery of doxoru-
`bicin by the chimeric immunoconjugate to Lewis-Y–
`expressing target cells was possible.10
`The toxicities observed with the BR96-doxorubicin conju-
`gate in this study, however, provide indirect evidence for the
`limited clinical antitumor activity observed. The toxicity
`
`associated with the BR96-doxorubicin conjugate differed
`markedly from that of the single-agent doxorubicin refer-
`ence arm and was predominantly gastrointestinal (ie, nausea
`and vomiting, including hematemesis, abdominal pain, and
`pancreatic enzyme abnormalities). Although the BR96 immu-
`noglobulin targets the tumor-associated antigen Lewis-Y,
`this antigen is also expressed by a few normal tissues,
`including the gastric mucosa, sma