VOLUME 30 䡠 NUMBER 26 䡠 SEPTEMBER 10 2012
`
`JOURNAL OF CLINICAL ONCOLOGY
`
`O R I G I N A L R E P O R T
`
`A Phase II Study of Trastuzumab Emtansine in Patients
`With Human Epidermal Growth Factor Receptor
`2–Positive Metastatic Breast Cancer Who Were Previously
`Treated With Trastuzumab, Lapatinib, an Anthracycline, a
`Taxane, and Capecitabine
`Ian E. Krop, Patricia LoRusso, Kathy D. Miller, Shanu Modi, Denise Yardley, Gladys Rodriguez,
`Ellie Guardino, Michael Lu, Maoxia Zheng, Sandhya Girish, Lukas Amler, Eric P. Winer, and Hope S. Rugo
`
`A
`
`B
`
`S
`
`T
`
`R
`
`A
`
`C
`
`T
`
`Purpose
`To determine whether the antibody-drug conjugate trastuzumab emtansine (T-DM1), which
`combines human epidermal growth factor receptor 2 (HER2) –targeted delivery of the potent
`antimicrotubule agent DM1 with the antitumor activity of trastuzumab, is effective in patients with
`HER2-positive metastatic breast cancer (MBC) who have previously received all standard
`HER2-directed therapies.
`Patients and Methods
`In this single-arm phase II study, T-DM1 3.6 mg/kg was administered intravenously every 3 weeks
`to patients with HER2-positive MBC who had prior treatment with trastuzumab, lapatinib, an
`anthracycline, a taxane, and capecitabine. The primary objectives were overall response rate (ORR)
`by independent review and safety.
`Results
`Among 110 pretreated patients (median, seven prior agents for MBC; median follow-up, 17.4
`months), the ORR was 34.5% (95% CI, 26.1% to 43.9%), clinical benefit rate was 48.2% (95%
`CI, 38.8% to 57.9%), median progression-free survival (PFS) was 6.9 months (95% CI, 4.2 to 8.4
`months), and median duration of response was 7.2 months (95% CI, 4.6 months to not estimable).
`In patients with confirmed HER2-positive tumors (n ⫽ 80 by retrospective central testing), the
`response rate was 41.3% (95% CI, 30.4% to 52.8%), and median PFS was 7.3 months (95% CI,
`4.6 to 12.3 months). Most adverse events were grades 1 to 2; the most frequent grade ⱖ 3 events
`were thrombocytopenia (9.1%), fatigue (4.5%), and cellulitis (3.6%).
`Conclusion
`T-DM1 is well tolerated and has single-agent activity in patients with HER2-positive MBC who
`have previously received both approved HER2-directed therapies and multiple chemotherapy
`agents. T-DM1 may be an effective new treatment for this patient population.
`
`J Clin Oncol 30:3234-3241. © 2012 by American Society of Clinical Oncology
`
`INTRODUCTION
`
`Amplification of the human epidermal growth fac-
`tor receptor 2 (HER2) gene occurs in approximately
`20% to 25% of primary breast cancers.1,2 Such
`tumors are considered HER2-positive and are asso-
`ciated with aggressive growth and poor clinical out-
`comes.1 Therapies that target the HER2 pathway
`have greatly improved outcomes for patients with
`HER2-positive breast cancer.3-8 Trastuzumab
`(Herceptin; Genentech, South San Francisco,
`CA), a humanized monoclonal antibody against
`the extracellular domain of HER2, is associated
`
`with single-agent activity7 and improved survival
`when added to chemotherapy5 in patients with
`HER2-positive metastatic breast cancer (MBC).
`The addition of the HER1/HER2 tyrosine kinase
`inhibitor lapatinib to capecitabine improves time
`to disease progression.8 This combination is indi-
`cated for patients with HER2-positive advanced
`cancer or MBC after prior trastuzumab, anthra-
`cycline, and taxane therapy. Despite these treat-
`ments, disease progression eventually occurs. In
`clinical practice, anti-HER2 therapies are continued
`through multiple lines of treatment,9 given the evi-
`dence that HER2 overexpression persists after
`
`Ian E. Krop and Eric P. Winer, Dana-
`Farber Cancer Institute, Boston, MA;
`Patricia LoRusso, Karmanos Cancer
`Institute, Detroit, MI; Kathy D. Miller,
`Indiana University Melvin and Bren
`Simon Cancer Center, Indianapolis, IN;
`Shanu Modi, Memorial Sloan-Kettering
`Cancer Center, New York, NY; Denise
`Yardley, Sarah Cannon Research Insti-
`tute; Tennessee Oncology, Nashville,
`TN; Gladys Rodriguez, South Texas
`Oncology/Hematology, San Antonio, TX;
`Ellie Guardino, Michael Lu, Maoxia
`Zheng, Sandhya Girish, and Lukas
`Amler, Genentech, South San Fran-
`cisco, CA; and Hope S. Rugo, Univer-
`sity of California, San Francisco
`Comprehensive Cancer Center, San
`Francisco, CA.
`
`Submitted November 10, 2011;
`accepted April 9, 2012; published online
`ahead of print at www.jco.org on May
`29, 2012.
`
`Supported by Genentech.
`
`Presented at the 2010 European Soci-
`ety for Medical Oncology Congress,
`Milan, Italy, October 8-12, 2010, and
`the 2009 San Antonio Breast Cancer
`Symposium, San Antonio, TX, Decem-
`ber 9-13, 2009.
`
`Authors’ disclosures of potential con-
`flicts of interest and author contribu-
`tions are found at the end of this
`article.
`
`Clinical Trials repository link available on
`JCO.org.
`
`Corresponding author: Ian E. Krop, MD,
`PhD, Dana-Farber Cancer Institute, 450
`Brookline Ave, Boston MA 02215;
`e-mail: ikrop@partners.org.
`
`© 2012 by American Society of Clinical
`Oncology
`
`0732-183X/12/3026-3234/$20.00
`
`DOI: 10.1200/JCO.2011.40.5902
`
`3234
`
`© 2012 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on May 20, 2014. For personal use only. No other uses without permission.
`Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
`
`IMMUNOGEN 2120, pg. 1
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Phase II Study of T-DM1 in Pretreated HER2-Positive MBC
`
`progression.10-12 However, there is no standard HER2-directed regi-
`men approved for these heavily pretreated patients,9 and additional
`HER2-directed therapies are needed.
`Trastuzumab emtansine (T-DM1) is an antibody-drug conju-
`gate (ADC) designed for treatment of HER2-positive cancer.13
`T-DM1 combines the selective intracellular delivery of DM1, a potent
`derivative of the antimicrotubule agent maytansine, with the antitu-
`mor properties of trastuzumab.13-17 In T-DM1, trastuzumab and
`DM1 are covalently linked via a stable thioether linker (N-
`maleimidomethyl)cyclohexane-1-carboxylate, which is thought to
`limit the exposure of normal tissue to DM1.18,19 A prior proof-of-
`concept phase II study (TDM4258g) of single-agent T-DM1 (3.6
`mg/kg given every 3 weeks) in 112 patients with HER2-positive MBC
`who had progressed while receiving HER2-directed therapy, showed
`antitumor activity.19 The objective response rate (ORR) by indepen-
`dent assessment was 25.9% (95% CI, 18.4% to 34.4%), and median
`progression-free survival (PFS) was 4.6 months (95% CI, 3.9 to 8.6
`months). Median PFS was higher (8.2 months) among patients with
`tumors that were HER2-positive by retrospective central testing
`(n ⫽ 74) compared with those found to be HER2-normal (n ⫽ 21; 2.6
`months). T-DM1 was well tolerated with no dose-limiting cardiotox-
`icity. Of note, 60% of patients had received prior lapatinib in addition
`to trastuzumab. The ORR (23.9%) in this subgroup was similar to that
`in patients who had received trastuzumab alone, suggesting that
`T-DM1 is effective even in patients who had previously received both
`of the approved HER2-targeted therapies.
`The study described here, TDM4374g, was conducted to confirm
`that T-DM1 is active in tumors refractory to current therapies. In
`contrast to TDM4258g, which required only that patients had received
`and progressed on at least one prior HER2-directed agent and received
`at least one chemotherapy in the metastatic setting, this study required
`that all patients had received trastuzumab, lapatinib, a taxane, an
`anthracycline, and capecitabine. In addition, patients had to have at
`least two HER2-directed therapy regimens in the metastatic or locally
`advanced setting and progression on the most recent regimen. Explor-
`atory subgroup analyses are also presented that assess efficacy based on
`HER2 expression levels and on phosphatidyl inositol-3-kinase (PI3K)
`mutations, since alterations in the PI3K pathway have been implicated
`in tumor insensitivity to trastuzumab in HER2-positive MBC.20-24
`
`PATIENTS AND METHODS
`
`Patients
`Key eligibility criteria included age ⱖ 18 years, documented HER2-
`positive MBC (immunohistochemistry [IHC] 3⫹ or fluorescent in situ hy-
`bridization [FISH] positive by local laboratory criteria), measurable disease
`per Response Evaluation Criteria in Solid Tumors (RECIST) v.1.0,25 Eastern
`Cooperative Oncology Group (ECOG) performance status 0 to 2, and prior
`treatment with trastuzumab, lapatinib, an anthracycline, a taxane, and cape-
`citabine in the neoadjuvant, adjuvant, locally advanced, or metastatic setting.
`Patients also must have had at least two HER2-directed therapy regimens in
`the metastatic or locally advanced setting and progression on the most recent
`regimen (see Appendix [online only] for more detail on methods).
`Study Design and Objectives
`T-DM1 was administered intravenously at 3.6 mg/kg every 3 weeks. The
`primary objectives were to assess ORR by independent radiologic facility (IRF)
`review and to evaluate safety and tolerability. Secondary objectives included
`clinical benefit rate (complete response plus partial response plus stable dis-
`ease ⱖ 6 months), duration of objective response (DOR), PFS, and pharma-
`
`cokinetic (PK) profile. Correlation of efficacy with biomarkers was an
`exploratory objective.
`All patients provided written informed consent. The study was reviewed
`and approved by the institutional review board at each site, according to
`local guidelines.
`
`Assessments and Data Collection
`Tumor assessments were conducted every other cycle by investigator and
`retrospectively by double-reader IRF as needed. Echocardiogram or multi-
`gated acquisition scans were obtained at screening and every 6 weeks thereafter
`until study termination.
`
`Diagnostic and Biomarker Assays
`Archival tumor tissue was collected and evaluated by validated assays for
`HER2 expression (by using FISH, IHC, and quantitative reverse transcriptase
`polymerase chain reaction [qRT-PCR]) and for mutations in the PI3K cata-
`lytic subunit (PIK3CA).
`
`Statistics
`The planned sample size of 100 patients, calculated by using the exact
`method for a single proportion, was chosen to ensure ⱖ 80% power to reject a
`null hypothesis of a response rate of ⱕ 14% against an alternative of ⱖ 25%.
`Statistical analyses were conducted approximately 15 months after the last
`patient was enrolled.
`Efficacy and safety end points were analyzed in patients who received at
`least one dose of T-DM1. Objective response was defined as complete or
`partial response on two consecutive tumor assessments ⱖ 4 weeks apart; PFS
`was defined as time from first day of study treatment to first documented
`disease progression or death on study (ie, death from any cause within 30 days
`from last dose of study drug). Median PFS and DOR were calculated by using
`Kaplan-Meier methods.
`
`RESULTS
`
`Demographics and Patient Characteristics
`In all, 110 patients enrolled onto this study and received at least
`one T-DM1 dose (Table 1). All patients had received prior trastu-
`zumab, lapatinib, an anthracycline, a taxane, and capecitabine except
`for one patient who had received prior ixabepilone instead of a taxane.
`Patients received a median of 8.5 (range, 5 to 19) prior anticancer
`agents (excluding hormonal therapies) in all settings, and a median of 7.0
`(range, 3 to 17) prior agents (excluding hormonal therapies) for MBC.
`
`Efficacy
`Among treated patients, 38 had an objective tumor response (all
`partial responses) by IRF assessment, corresponding to an ORR of
`34.5% (95% CI, 26.1% to 43.9%; Table 2). Median PFS was 6.9
`months (95% CI, 4.2 to 8.4 months) by IRF (Table 2 and Fig 1A). ORR
`in patients with documented progression on prior trastuzumab, lapa-
`tinib, and chemotherapy was similar to that of the overall patient
`population (Table 2). Median DOR was 7.2 months (95% CI, 4.6
`months to not estimable [N/E]; Fig 1B). By investigator assessment,
`the ORR was 32.7% (95% CI, 24.1% to 42.1%), median DOR was 9.7
`months (95% CI, 7.1 month to N/E), and median PFS was 5.5 months
`(95% CI, 4.2 to 7.9 months).
`
`Correlation of Efficacy With Biomarker Expression
`HER2 status was centrally reassessed on archival primary tumors
`in 95 patients; 80 (84.2%) had confirmed HER2-positive disease and
`15 (15.8%) had HER2-normal disease, defined as HER2 FISH ratio
`less than 2.0 and IHC ⱕ 2⫹ (Table 3). By IRF assessment, ORR was
`41.3% (95% CI, 30.4% to 52.8%) in patients with confirmed HER2-
`
`www.jco.org
`
`© 2012 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on May 20, 2014. For personal use only. No other uses without permission.
`Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
`
`3235
`
`IMMUNOGEN 2120, pg. 2
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Krop et al
`
`Table 1. Demographics and Baseline Characteristics (N ⫽ 110)
`
`Table 2. Responses and Efficacy Assessments by IRF (N ⫽ 110)
`
`Characteristic
`
`Age, years
`Median
`Range
`Sex
`Male
`Female
`ECOG performance status
`0
`1
`2
`Baseline LVEF, %
`Median
`Range
`Time from metastatic diagnosis to first study treatment, months
`Median
`Range
`No. of distinct sites of metastases
`1
`2
`3⫹
`Sites of metastases
`Locoregional
`Lung
`Bone
`Liver
`CNS
`ER and PR status
`ER-positive and/or PR-positive
`ER-negative and PR-negative
`Unknown
`Prior therapies
`Trastuzumab
`Lapatinib
`Taxane
`Capecitabine
`Anthracycline
`Radiotherapy
`Hormonal therapy
`Other targeted therapy (excluding trastuzumab, lapatinib, and
`hormonal therapy)
`No. of prior anticancer agents in all settings
`Median
`Range
`No. of prior anticancer agents for metastatic disease
`Median
`Range
`Total duration of prior trastuzumab treatment in the metastatic
`setting, months
`Median
`Range
`Total duration of prior lapatinib treatment in the metastatic
`setting, months
`Median
`Range
`
`No. %
`
`52.5
`34-77
`
`1.8
`2
`108 98.2
`
`54 49.1
`53 48.2
`3
`2.7
`
`60
`50-77
`
`42.8
`4.6-148.9
`
`4.5
`5
`24 21.8
`81 73.6
`
`70 63.6
`69 62.7
`57 51.8
`49 44.5
`19 17.3
`
`55 50.0
`51 46.4
`4
`3.6
`
`110 100
`110 100
`109 99.1
`110 100
`110 100
`95 86.4
`53 48.2
`
`37 33.6
`
`8.5
`5-19
`
`7.0
`3-17
`
`19.7
`1.8-115.8
`
`6.8
`0.2-23.3
`
`Abbreviations: ECOG, Eastern Cooperative Oncology Group; ER, estrogen
`receptor; LVEF, left ventricular ejection fraction; PR, progesterone receptor.
`
`positive tumors and 20.0% (95% CI, 5.7% to 44.9%) in patients with
`HER2-normal tumors; median PFS was 7.3 months (95% CI, 4.6 to
`12.3 months) and 2.8 months (95% CI, 1.3 months to N/E), respec-
`tively (Fig 1C).
`
`Assessments
`
`Patients with objective response
`Best objective response
`Complete response
`Partial response
`Clinical benefit rateⴱ
`Duration of objective response, months
`Median
`Progression-free survival, months
`Median
`Patients who had progressed on
`trastuzumab plus chemotherapy
`and lapatinib plus chemotherapy
`ORR
`
`No. of
`Patients
`
`38
`
`0
`38
`
`7.2
`
`6.9
`
`25/73†
`
`%
`
`95% CI
`
`34.5
`
`26.1 to 43.9
`
`0.0
`34.5
`48.2
`
`38.8 to 57.9
`
`4.6 to N/E
`
`4.2 to 8.4
`
`34.2
`
`23.5 to 45.4
`
`Abbreviations: IRF, independent radiologic facility; N/E, not estimable; ORR,
`objective response rate.
`ⴱClinical benefit is defined as objective response plus stable disease at
`6 months.
`†No. of responders/No. of patients in subgroup.
`
`Patients with retrospectively confirmed HER2-positive disease
`and available qRT-PCR data (n ⫽ 69) were grouped according to
`qRT-PCR–determined HER2 expression levels (equal to, greater
`than, or less than median; Appendix Figure A1, online only). ORR per
`IRF in patients with at least median HER2 expression (n ⫽ 35) was
`42.9% (95% CI, 26.3% to 60.6%); median PFS was 8.0 months (95%
`CI, 5.4 months to N/E; Fig 1D). ORR in patients with less than median
`HER2 expression (n ⫽ 34) was 38.2% (95% CI, 22.2% to 56.4%; Table
`3), and median PFS was 6.2 months (95% CI, 3.9 to 12.3 months; Fig
`1D). Exploratory analysis was also performed according to HER2
`amplification (assessed by FISH): ORR per IRF in patients with at least
`median HER2/CEP17 [chromosome centromere 17] ratio (n ⫽ 36)
`was 44.4%, and ORR per IRF in patients with less than median HER2/
`CEP17 ratio (n ⫽ 36) was 38.9% (Table 3).
`Among patients with centrally reassessed HER2-positive tumors,
`the ORRs were similar among 49 patients with wild-type PI3KCA and
`11 patients with mutant PI3KCA (ORRs, 36.7% and 36.4%, respec-
`tively; Table 3).
`
`T-DM1 Exposure and Patient Disposition
`As of the data cutoff date (June 21, 2010), patients had received a
`median of 7.0 doses (range, 1 to 30 doses) of T-DM1 or 19.3 weeks
`(range, 0.1 to 88.0 weeks) of treatment (Appendix Table A1, online
`only); 18 patients (16.4%) were still receiving treatment, and 92
`(83.6%) had been discontinued from study (Appendix Table A1).
`Treatment discontinuations due to adverse events (AEs) oc-
`curred in seven patients (Appendix Table A1). Dose reductions were
`reported for 18 patients, 11 patients reduced their doses to 3.0 mg/kg,
`and seven reduced their doses to 2.4 mg/kg. The most common
`reasons for dose reduction were serum transaminase abnormalities
`and thrombocytopenia.
`Six patients developed isolated brain metastases on study and,
`after receiving local therapy for these lesions, continued T-DM1 treat-
`ment per protocol. The median number of cycles received before and
`after CNS progression were 7.5 (range, 4 to 16) and 4.5 (range, 2 to 8),
`respectively. Five patients discontinued because of subsequent disease
`
`3236
`
`JOURNAL OF CLINICAL ONCOLOGY
`© 2012 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on May 20, 2014. For personal use only. No other uses without permission.
`Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
`
`IMMUNOGEN 2120, pg. 3
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`T-DM1 (n = 38)
`Median: 7.2 (95% CI, 4.2 to N/E)
`
`Phase II Study of T-DM1 in Pretreated HER2-Positive MBC
`
`1.0
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`Proportion Alive Without
`
`Progression
`
`B
`
`T-DM1 (n = 110)
`Median: 6.9 (95% CI, 4.2 to 8.4)
`
`1.0
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`A
`
`Progression-Free Survival
`
`(rate)
`
`0
`
`2
`
`4
`
`6
`
`8
`10
`12
`Time (months)
`
`41
`
`16
`
`18
`
`20
`
`0
`
`2
`
`4
`
`6
`
`8
`10
`12
`Time (months)
`
`14
`
`16
`
`18
`
`20
`
`No. at risk
`T-DM1
`C
`
`110
`
`83
`
`59
`
`41
`
`33
`
`22
`
`19
`
`14
`
`8
`
`3
`
`0
`
`1.0
`
`No. at risk
`T-DM1
`D
`
`1.0
`
`38
`
`35
`
`27
`
`16
`
`14
`
`11
`
`10
`
`6
`
`2
`
`1
`
`0
`
`HER2+, median (n = 35), median: 8.0 (95% CI, 5.4 to N/E)
`HER2+, < median (n = 34), median: 6.2 (95% CI, 3.9 to 12.3)
`HER2 normal (n = 15), median: 2.8 (95% CI, 1.3 to N/E)
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`Progression-Free Survival
`
`(rate)
`
`HER2 normal (n = 15), median: 2.8 (95% CI, 1.3 to N/E)
`HER2+ (n = 80), median: 7.3 (95% CI, 4.6 to 12.3)
`
`0.8
`
`0.6
`
`0.4
`
`0.2
`
`Progression-Free Survival
`
`(rate)
`
`0
`
`2
`
`4
`
`6
`
`8
`10
`12
`41
`Time (months)
`
`16
`
`18
`
`20
`
`22
`
`0
`
`2
`
`4
`
`6
`
`8
`10
`12
`Time (months)
`
`14
`
`16
`
`18
`
`20
`
`No. at risk
`HER2 normal
`HER2+
`
`15
`80
`
`7
`65
`
`3
`48
`
`1
`36
`
`0
`29
`
`0
`20
`
`0
`17
`
`0
`13
`
`0
`8
`
`0
`3
`
`0
`0
`
`0
`0
`
`No. at risk
`HER2+, < median 34
`HER2+, median 35
`HER2 normal
`15
`
`27
`28
`7
`
`18
`23
`3
`
`14
`18
`1
`
`12
`13
`0
`
`8
`9
`0
`
`6
`8
`0
`
`4
`7
`0
`
`3
`4
`0
`
`1
`1
`0
`
`0
`0
`0
`
`Fig 1. Kaplan-Meier plots. (A) Progression-free survival (PFS) in all treated patients and (B) duration of response. (C) PFS by retrospectively assessed human epidermal
`growth factor receptor 2 (HER2) status and (D) PFS by quantitative reverse transcriptase polymerase chain reaction HER2 levels in patients with retrospectively
`confirmed HER2-positive (HER2⫹) status. NE, not estimable; T-DM1, trastuzumab emtansine.
`
`progression, and one patient discontinued because of investigator
`decision (patient no longer deriving benefit).
`
`Safety
`The most common AEs of any grade were fatigue (61.8%),
`nausea (37.3%), and thrombocytopenia (38.2%; Table 4). The
`most common grade ⱖ 3 AEs were thrombocytopenia (9.1%),
`fatigue (4.5%), and cellulitis (3.6%). Fifty-two patients (47.3%)
`experienced at least one grade 3 AE; six patients experienced at least
`one grade 4 AE (two with thrombocytopenia, one with spinal cord
`compression, one with spinal cord compression and abdominal
`pain, one with sepsis, and one with cellulitis). Three patients expe-
`rienced grade 5 AEs: one death from pneumonia (in the patient
`with grade 4 sepsis), one from interstitial lung disease, and one due
`to abnormal hepatic function in the context of coexisting nonalco-
`holic fatty liver disease and multiple comorbidities, including
`contrast-induced renal dysfunction.
`Thrombocytopenia, previously identified as a T-DM1 dose-
`limiting toxicity, was observed in 42 patients (38.2%), including 10
`
`(9.1%) with grade 3 or 4 AEs. Platelet count nadirs typically occurred
`around day 8 of each treatment cycle, with recovery to grade 1 or lower
`before the next cycle. Thrombocytopenia infrequently led to T-DM1
`discontinuation or dose reduction. Platelet transfusions were infre-
`quent (four patients). Hemorrhagic AEs were generally mild, al-
`though one patient (0.9%) experienced grade 3 epistaxis with
`concurrent grade 2 thrombocytopenia. No patients discontinued
`T-DM1 because of a hemorrhagic event.
`Although overall hepatic toxicities were generally mild, there
`appeared to be a temporal relationship between T-DM1 dosing and
`increases in serum transaminases. Common liver function test abnor-
`malities reported as AEs of any grade were increases in serum AST
`(26.4% of patients) and ALT (13.6%), with increases in total bilirubin
`(2.7%) observed less frequently. Nine patients (8.2%) experienced
`some type of grade ⱖ 3 hepatic AE (seven patients had grade 3
`laboratory abnormalities, one patient had grade 3 hepatotoxicity, and
`one patient had a grade 5 hepatic event). No left ventricular ejection
`fraction decline to ⱕ 45% or symptomatic congestive heart failure was
`
`www.jco.org
`
`© 2012 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on May 20, 2014. For personal use only. No other uses without permission.
`Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
`
`3237
`
`IMMUNOGEN 2120, pg. 4
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Krop et al
`
`Table 3. Objective Responses by Retrospectively Assessed HER2 Status
`in Treated Patients With Central HER2 Data (biomarker subgroups, exploratory
`analyses; n ⫽ 95)
`
`Response
`
`No. of Patients
`
`HER2-positive status
`ORR
`HER2-normal status
`ORR
`HER2 qRT-PCR levelⴱ†
`At least median
`ORR
`Less than median
`ORR
`HER2/CEP17 ratio†
`At least median
`ORR
`Less than median
`ORR
`PIK3CA mutation status†‡
`Wild-type
`ORR
`Mutant§
`ORR
`
`80
`
`15
`
`35
`
`34
`
`36
`
`36
`
`49
`
`11
`
`%
`
`84.2
`41.3
`15.8
`20.0
`
`95% CI
`
`30.4 to 52.8
`
`5.7 to 44.9
`
`42.9
`
`26.3 to 60.6
`
`38.2
`
`22.2 to 56.4
`
`44.4
`
`27.9 to 61.9
`
`38.9
`
`24.2 to 56.5
`
`36.7
`
`23.6 to 51.6
`
`36.4
`
`13.5 to 66.7
`
`Abbreviations: CEP17, chromosome centromere 17; HER2, human epider-
`mal growth factor receptor 2; ORR, objective response rate; PIK3CA, phos-
`phatidyl
`inositol-3-kinase catalytic subunit; qRT-PCR, quantitative reverse
`transcriptase polymerase chain reaction.
`ⴱSee Appendix Figure A1 for box plot showing distribution of qRT-PCR values
`around the median.
`†In patients with fluorescent in situ hybridization–positive and/or immuno-
`chemistry 3⫹ tumors and who had measurable disease at baseline deter-
`mined by an independent radiologic facility.
`‡E545K/D, and H1047R hotspot mutations in PIK3CA were detected by
`using the DxS PIK3CA mutation test kit (DxS, Manchester, United Kingdom).
`§Four H1047R, two E542K, and five E545K/D mutations in patients with
`confirmed HER2-positive disease.
`
`observed, and no patients discontinued treatment because of cardio-
`toxicity. Most other reported AEs were mild and manageable, result-
`ing in few drug discontinuations, dose modifications, or dose delays
`(Appendix Table A1).
`
`PK
`
`The T-DM1 half-life was 3.96 days, with an average maximum
`concentration (Cmax) of 79.5 ␮g/mL and no significant accumulation
`over the treatment cycles (Appendix Fig A2, online only). Average
`Cmax of DM1 was 5.36 ng/mL in cycle 1 and 5.97 ng/mL in cycle 4,
`suggesting no DM1 accumulation. This PK profile of T-DM1 is simi-
`lar to that determined in the prior phase I18 and II19 studies. Of the 108
`evaluable patients, six (5.6%) had an antitherapeutic antibody (ATA)
`response, and one of these six patients also had an ATA response in
`their baseline sample before study treatment. There was no observed
`effect of ATA response on T-DM1 PK.
`
`DISCUSSION
`
`This study assessed whether T-DM1 was efficacious in patients with
`HER2-positive MBC who had previously received both trastuzumab-
`and lapatinib-based therapy, in addition to anthracyclines, taxanes,
`and capecitabine. Eligible patients had received at least two HER2-
`
`directed regimens in the advanced disease setting and had progressed
`on their most recent regimen. The purpose of these stringent eligibility
`criteria was to select a homogeneous, treatment-refractory population
`that would best reflect patients with the greatest medical need. To the
`best of our knowledge, this study is the first to evaluate a new treat-
`ment in this carefully defined subset of patients.
`Although all patients were heavily pretreated (median of seven
`prior therapeutic agents for MBC), T-DM1 demonstrated clinical
`efficacy, with an independently assessed ORR of 34.5% and clinical
`benefit rate of 48.2%. The ORR was associated with a PFS of 6.9
`months and DOR of 7.2 months. Notably, a response rate of 20%
`(three of 15 patients) was observed in patients with MBC designated as
`HER2-normal on central laboratory retesting. This observation sug-
`gests that HER2 expression below the currently used clinical threshold
`may be sufficient to confer sensitivity to T-DM1. It is also possible that
`the HER2 status of the tumor changed over time (ie, the tumor HER2
`status at the time of treatment was different from that of the archived
`tumor sample). Because of the small number of patients in this subset,
`no definitive conclusions can be drawn, and further study of this issue
`is needed. Exploratory analyses of other biomarkers to potentially
`predict responsiveness to T-DM1, including HER2 mRNA level, FISH
`copy number, and PIK3CA mutation status, were also performed in
`this study. No clear relationship between these markers and clinical
`outcome was observed, and more work needs to be done to identify
`patients most likely to respond to this agent.
`Several other HER2-directed agents in development have shown
`promising activity in previously treated patients with HER2-positive
`MBC. Neratinib,26 pertuzumab,27,28 and the HSP-90 inhibitor tane-
`spimycin29 (the latter two in combination with trastuzumab) have
`demonstrated ORRs of approximately 25% in phase II studies in pa-
`tients with HER2-positive MBC who had progressed on trastuzumab-
`based therapy. However, none of these trials specifically evaluated
`patients who had previously received both trastuzumab- and
`lapatinib-based therapy, making it difficult to compare these agents’
`activities in this clinically relevant population.
`The safety profile of T-DM1 in this study was encouraging. Most
`AEs were grades 1 or 2, and study drug discontinuations were uncom-
`mon. No dose-limiting cardiotoxicities were reported; however, all
`patients had extensive prior anti-HER2 therapy and left ventricular
`ejection fraction of ⱖ 50% at entry, so this study cannot fully assess the
`potential impact of T-DM1 on cardiac function. Thrombocytopenia
`was the only grade ⱖ 3 AE present in more than 5% of patients and
`was associated with significant hemorrhage in only one patient (grade
`3 epistaxis). In vitro mechanistic studies with hematopoietic stem
`cells, megakaryocytes, and platelets suggest that T-DM1 does not have
`a direct effect on platelet function but can impair megakaryocyte
`maturation and platelet production.30 Although further research is
`underway, these data support the hypothesis that the thrombocytope-
`nia observed in patients following T-DM1 treatment may be ex-
`plained, at least in part, by impaired platelet production from
`megakaryocytes in the bone marrow. Alopecia was not observed with
`T-DM1 therapy, and other toxicities typically associated with chemo-
`therapy (eg, nausea, neuropathy, and neutropenia) were infrequent.
`Seven patients (6.4%) developed grade 3 increases in serum transami-
`nases, which were successfully managed by holding and/or reducing
`the dose of T-DM1. A patient who died from hepatic dysfunction had
`preexisting liver disease and had developed contrast-induced renal
`failure before manifesting hepatic dysfunction.
`
`3238
`
`JOURNAL OF CLINICAL ONCOLOGY
`© 2012 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on May 20, 2014. For personal use only. No other uses without permission.
`Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
`
`IMMUNOGEN 2120, pg. 5
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Phase II Study of T-DM1 in Pretreated HER2-Positive MBC
`
`Table 4. AEs (all grades) in ⬎ 10% of Patients (N ⫽ 110) and Grade ⱖ 3 AEs Occurring in at Least Two Patients
`
`Grade (%)
`
`AE
`AEs (any grade) occurring in ⬎ 10% of patients
`Fatigue
`Thrombocytopenia
`Nausea
`Increased AST
`Constipation
`Pyrexia
`Epistaxis
`Headache
`Decreased appetite
`Hypokalemia
`Dry mouth
`Anemia
`Back pain
`Cough
`Peripheral neuropathy
`Dyspnea
`Vomiting
`Arthralgia
`Infusion-related reaction
`Pain in extremity
`Increased ALT
`Myalgia
`Diarrhea
`Increased alkaline phosphatase
`Muscle spasms
`Decreased Weight
`Depression
`Additional AEs, for which grade ⱖ 3 events
`occurred in at least two patients
`Abdominal pain
`Hyperglycemia
`Cellulitis
`Pneumonia
`Abnormal liver function test
`Spinal cord compression
`Basal cell carcinoma
`
`Abbreviation: AE, adverse event.
`ⴱOne patient had a grade 5 hepatic dysfunction event.
`
`1
`
`30.0
`12.7
`26.4
`11.8
`20.0
`13.6
`19.1
`18.2
`12.7
`19.1
`17.3
`7.3
`11.8
`12.7
`12.7
`10.0
`10.9
`10.0
`8.2
`10.0
`8.2
`12.7
`9.1
`8.2
`8.2
`7.3
`4.5
`
`5.5
`4.5
`0
`0.9
`0
`0
`0
`
`2
`
`27.3
`16.4
`10.0
`11.8
`2.7
`8.2
`2.7
`3.6
`7.3
`0.9
`2.7
`10.9
`3.6
`5.5
`5.5
`3.6
`5.5
`2.7
`5.5
`1.8
`2.7
`0.9
`3.6
`2.7
`2.7
`2.7
`2.7
`
`1.8
`0
`2.7
`0
`0.9
`0
`0
`
`3
`
`4.5
`7.3
`0.9
`2.7
`0.9
`0.9
`0.9
`0
`0.9
`0.9
`0
`1.8
`2.7
`0
`0
`2.7
`0
`1.8
`0
`1.8
`2.7
`0
`0
`0
`0
`0.9
`2.7
`
`0.9
`2.7
`2.7
`1.8
`1.8
`0
`1.8
`
`4
`
`0
`1.8
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`
`0.9
`0
`0.9
`0
`0
`1.8
`0
`
`5
`
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`0
`
`0
`0
`0
`0.9
`0ⴱ
`0
`0
`
`All
`
`61.8
`38.2
`37.3
`26.4
`23.6
`22.7
`22.7
`21.8
`20.9
`20.9
`20.0
`20.0
`18.2
`18.2
`18.2
`16.4
`16.4
`14.5
`13.6
`13.6
`13.6
`13.6
`12.7
`10.9
`10.9
`10.9
`10.0
`
`9.1
`7.3
`6.4
`3.6
`2.7
`1.8
`1.8
`
`Six patients developed isolated brain metastases as site of first
`progression and, after treatment of their CNS lesions, continued on
`T-DM1 for a median of 4.5 cycles. Although no definitive conclusions
`can be made about the safety of continuing T-DM1 in patients with
`stable visceral disease who develop isolated brain metastases, these
`preliminary data suggest it is worthy of further investigation.
`To date, ADCs have had limited clinical success, particularly
`in the treatment of epithelial cancers.31 However, T-DM1 has
`consistently demonstrated encouraging clinical activity and ap-
`pears to fulfill the initial promise of ADCs by significantly improv-
`ing the therapeutic index of the cytotoxic agent (DM1). HER2 may
`be an ideal ADC target for several reasons. HER2 is highly overex-
`pressed on HER2-positive cancer cells compared with normal tis-
`sue and is rapidly internalized without downregulation of
`expression. Moreover, HER2 expression appears to be retained in
`tumor cells that have acquired resistance to HER2-directed
`agents,12 making it unlikely that selection for HER2 nonexpressing
`
`(and therefore nontargetable) variants can occur. In addition, the
`trastuzumab component of T-DM1 retains its ability to inhibit
`HER2 signaling and activate antibody-dependent cellular cytotox-
`icity,17 complementing the cytotoxic effects of DM1.
`T-DM1 is the first ADC to use a thioether linker which, in
`preclinical testing, was more stable than other commonly used link-
`ers.13 This stability is reflected in the extremely low levels of free DM1
`detected in patient plasma and is likely to play a role in T-DM1’s
`favorable safety profile.
`In summary, this is the first study to evaluate a novel therapy in
`patients with HER2-positive MBC who were previously treated with
`all standard HER2-directed therapies in addition to multiple chemo-
`therapies. These patients currently have limited treatment options.
`The results presented here confirm the effectiveness and tolerability of
`T-DM1 in this patient population. As such, T-DM1 could prove to be
`a new standard for these patients. Because there is currently no stan-
`dard treatment for these patients, the results of this study have
`
`www.jco.org
`
`© 2012 by American Society of Clinical Oncology
`Downloaded from jco.ascopubs.org on May 20, 2014. For personal use only. No other uses without permission.
`Copyright © 2012 American Society of Clinical Oncology. All rights reserved.
`
`3239
`
`IMMUNOGEN 2120, pg. 6
`Phigenix v. Immunogen
`IPR2014-00676
`
`

`

`Krop et al
`
`prompted the initiation of a phase III trial in which patients will be
`randomly assigned to single-agent T-DM1 or treatment of the physi-
`cian’s choice. Moreover, the favorable benefit-risk profile of T-DM1
`suggests that