Adjuvant Targeted Therapy in Early Breast Cancer

Transcription

1 Adjuvant Targeted Therapy in Early Breast Cancer John Mackey, MD 1, Deanna McLeod, BSc 2, Joseph Ragaz, MD 3, Karen Gelmon, MD 4, Sunil Verma, MD 5, Kathleen Pritchard, MD 5, Kara Laing, MD 6, Louise Provencher, MD 7, and Lauren F. Charbonneau, RPh, BSc(Pharm) 5 For this review, the authors appraised the evidence for adjuvant trastuzumab therapy in early breast cancer. There was level 1 evidence to support the routine use of 1 year of adjuvant trastuzumab in conjunction with chemotherapy for women with human epidermal growth factor receptor 2 (HER-2)-positive early breast cancer. The relative benefits of concurrent versus sequential administration remained unclear; however concurrent administration permitted the earliest possible intervention with trastuzumab with possible superiority. There was evidence to support the use of trastuzumab in both lymph node-positive and highrisk lymph node-negative patients, and preliminary data suggested that all patient subgroups that were eligible for the trials benefit equally from trastuzumab. Adjuvant trastuzumab was associated with a risk of cardiotoxicity, the long-term impact of which remains largely unknown. Routine cardiac risk assessment considering left ventricular ejection fraction, age, and prior history of cardiac events is recommended along with the selection of trastuzumab-based regimens that minimize cardiotoxicity. Trastuzumab acquisition costs for 1 year of therapy were the largest component of treatment costs. Cancer 2009;115: VC 2009 American Cancer Society. KEY WORDS: breast neoplasms, drug therapy, trastuzumab, therapeutic use, monoclonal antibodies, adjuvant chemotherapy, erbb-2 receptor, antineoplastic agents, humans. Increased knowledge of the biologic diversity of breast cancer has been accompanied by increased efforts to individualize breast cancer treatment based on the underlying molecular features of each tumor. Approximately 15% to 20% of early breast cancers have amplification of the human epidermal receptor-2 (HER-2) gene, with resultant overexpression of the HER-2 protein. 1-3 HER-2 belongs to a family of 4 transmembrane receptor tyrosine kinases that mediate the growth, differentiation, and survival of cells. 4-6 Tumors that over express this protein are imbued with more aggressive qualities, including enhanced growth and proliferation, increased invasive and metastatic capability, and stimulation of angiogenesis. 4,6 Individual trials, population-based studies, and retrospective analyses also have shown that HER-2 overexpression is associated with poorly differentiated, high-grade tumors and lymph node involvement. 4,6 Until Corresponding author: John Mackey, MD, Cross Cancer Centre, University Avenue, Edmonton, Alberta, T6G 1Z2, Canada; Fax: (780) ; johnmack@cancerboard.ab.ca 1 Cross Cancer Center, Edmonton. Alberta, Canada; 2 Kaleidoscope Strategic, Toronto, Ontario, Canada; 3 McGill University, Montreal, Quebec, Canada; 4 British Columbia Cancer Agency, British Columbia, Canada; 5 Sunnybrook Odette Cancer Center, Toronto, Ontario, Canada; 6 Cancer Care Program, Dr. H. Bliss Murphy Cancer Center, St. John s, Newfoundland, Canada; 7 Deschênes-Fabia Center for Breast Disease, Quebec, Canada We thank Kaleidoscope Strategic Inc., for the information management services required for developing this review. Received: June 24, 2008; Revised: September 12, 2008; Accepted: September 19, 2008 Published online: January 21, 2009, VC 2009 American Cancer Society DOI: /cncr.24114, Cancer March 15, 2009

2 Adjuvant Trastuzumab for Breast Cancer/Mackey et al recently, patients who had breast cancer with HER-2 overexpression (HER-2-positive breast cancer) faced a markedly poorer prognosis than patients who had breast cancer without HER-2 overexpression (HER-2 negative breast cancer). 4,6 A promising recent addition to the adjuvant treatment arsenal is trastuzumab (Herceptin; Genentech, South San Francisco, Calif), a new agent biologically targeted against the HER-2 protein that has demonstrated the ability to mitigate the prognostic disadvantage of HER-2 expression. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the HER-2 protein. Trastuzumab activates antibody-dependent, cellmediated cytotoxicity and disrupts the signal transduction process through interference with receptor dimerization, downstream effectors, and receptor internalization and degradation. 5 Trastuzumab was tested formally in combination with chemotherapy for the first-line treatment of metastatic breast cancer and demonstrated both improved response rate and improved survival compared with chemotherapy alone in patients with HER-2-positive breast cancer. 7 Although the risk of cardiotoxicity increased with the addition of trastuzumab to chemotherapy, otherwise, trastuzumab was well tolerated. These results triggered several prospective clinical trials to test the efficacy of adjuvant trastuzumab in early breast cancer. In the current review, we have appraised the evidence for adjuvant trastuzumab therapy in early breast cancer, considered treatment costs, and reviewed select qualityof-life factors as a means of navigating the landscape of available and emerging treatment options. Search Strategy and Study Overview Published clinical trials of for early breast cancer were identified in a search of the PubMed database through to April 2007 using the terms adjuvant or postoperative, trastuzumab, and breast cancer. In addition, conference proceedings from the American Society of Clinical Oncology (ASCO) ( ), the San Antonio Breast Cancer Symposium ( ), and the European Society of Clinical Oncology ( ) were searched manually for studies that met the eligibility criteria. Results presented at symposia were considered an important source of current clinical information, and, whenever available, data from presentations were used to supplement published abstract information. Studies were eligible for inclusion if they reported efficacy results from prospective, randomized controlled trials; if they were published in the English language, and if they investigated the use of adjuvant trastuzumab therapy in patients with early breast cancer. Six prospective clinical trials met the eligibility criteria of this review. Study Findings Efficacy The largest of the adjuvant trastuzumab trials is the Herceptin Adjuvant (HERA) trial, 8,9 which an international randomized study investigating the use of trastuzumab in HER-2-positive patients who have completed locoregional therapy and a minimum of 4 courses of chemotherapy. In total, 5102 women with HER-2-positive breast cancer were assigned randomly within 6 weeks of completing primary therapy to 1 of 3 groups: 1) women who underwent observation alone; 2) women who were treated with adjuvant trastuzumab (8 mg/kg load followed by 6 mg/kg every 3 weeks) for 1 year; and 3) women who were treated with adjuvant trastuzumab at the same dose and schedule for 2 years. The baseline characteristics of the patients in the HERA trial are summarized in Table 1. It is noteworthy that just over half (57%) of the women in this trial had lymph node-positive disease, less than one-third (26%) had received prior adjuvant anthracycline and taxane therapy, and an additional 11% received neoadjuvant chemotherapy. At a median follow-up of 23.5 months, 1 year of trastuzumab therapy resulted in significant reductions in the risk of recurrence (hazards ratio [HR], 0.64; 95% confidence interval [95% CI], [P <.0001]) and, despite a 51% crossover rate, the risk of death was significantly reduced (HR, 0.66; 95% CI, [P ¼.0115]) (Table 2) compared with observation. 9 When considering the clinical application of sequentially administering trastuzumab after chemotherapy as in the HERA trial (chemo-trastuzumab sequences), it is important to consider the following: 1) a potential further benefit may exist from the concurrent administration of chemotherapy and trastuzumab that was not explored in the trial; 2) significant losses in patient follow-up (5.7% experimental arm and 3.4% observation arm) and a high rate of patient Cancer March 15,

3 Table 1. Patient Characteristics, Tumor Characteristics, and Prior Chemotherapy HERA: Piccart- Gebhart 2005, 8 Smith N9831/B-31 Analysis: Perez Study Group BCIRG 006: Slamon PACS-04 (Trastuzumab Subgroup): Spielmann FinHer (Trastuzumab Subgroup): Joenssu Variable CTfiTr for 1y,n51703 CT, n51698 ACfiP/Tr, n51989 ACfiP, n51979 ACfiD/Tr, n51074 CB/D/Tr, n51075 ACfiD, n51073 FEC or EDfiTr, n5260 FEC or ED, n5268 D or V/TrfiFEC, n5116 DorVfiFEC, n5116 Adjuvant AT therapy, % Hormone therapy: ER/PgR-positive Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Age <50 y, % (Med) 49 (Med) 51 (Med) 50 (Med) Hormone-receptor positive, % (ERþ)/ 44 (ERþ)/ 39 (PgRþ) 29 (PgRþ) Lymph node status, % Neoadjuvant CT NA NA NA NA NA NA NA NA NA High-risk lymph node negative No. of positive lymph nodes Tumor size, % 2 cm (<2 cm) 35 (<2 cm) cm (2 cm) 65 (2 cm) 59 (>2 cm) 71 (>2 cm) >5 cm Trastuzumab initiated based on normal cardiac function At completion of RT and CT At completion of AC At completion of AC for AC!D and after surgery for CB/D/Tr At completion of RT and CT After surgery HERA indicates the Herceptin Adjuvant trial; N9831, North Central Cancer Treatment Group trial N9831; B-31, National Surgical Adjuvant Breast and Bowel Project trial B-31; BCIRG 006, Breast Cancer International Research Group trial 006; PACS-04, Protocole Adjuvant dans le Cancer du Sein; FinHer, Finland Herceptin study; CT, chemotherapy; Tr, trastuzumab; AC, doxorubicin and cyclophosphamide; P, paclitaxel; D, docetaxel; CB, carboplatin; FEC, fluorouracil, epirubicin, and cyclophosphamide; ED, epirubicin and docetaxel; V, vinorelbine; AT, anthracycline-taxane; ER, estrogen receptor; PgR, progesterone receptor; Med, median; þ, positive; NA, not applicable; RT, radiotherapy Cancer March 15, 2009

4 Adjuvant Trastuzumab for Breast Cancer/Mackey et al Table 2. Disease-free Survival and Overall Survival Study Group HERA: Smith N9831/B-31 Analysis: BCIRG 006: Slamon PACS-04 (Trastuzumab Perez Subgroup): Spielmann FinHer (Trastuzumab Subgroup): Joenssu Variable CTfiTr for 1 y, n51703 CT, n51698 ACfiP/Tr, n51989 ACfiP, n51979 ACfiD/Tr, n51074 CB/D/Tr, n51075 ACfiD, n51073 FEC or EDfiTr, n5260 FEC or ED, n5268 D or V/TrfiFEC, n5116 DorVfiFEC, n5116 Median follow-up, y Disease-free survival, % 3-Y *,y 77.6y 4-Y NA NA NAy NAy HR [95% CI] 0.64 [ ] 0.48 [ ] AC!D/Tr vs AC!D, 0.86 [ ] 0.42 [ ]y 0.61 [ ]/CB/D/Tr vs AC!D, 0.67 [ ] P <.0001z < <.0001/ y Overall survival, % 3-Y Y NA NA NR NR HR [95% CI] 0.66 [ ] 0.65 [ ] AC!D/Tr vs AC!D, 1.27 [ ] 0.41 [ ] 0.59 [ ]/CB/D/Tr vs AC!D, 0.66 [ ] P.0115 k /<.017 NR.07 HERA indicates the Herceptin Adjuvant trial; N9831, North Central Cancer Treatment Group trial N9831; B-31, National Surgical Adjuvant Breast and Bowel Project trial B-31;BCIRG 006, Breast Cancer International Research Group trial 006; PACS-04, ; FinHer, Finland Herceptin study; CT, chemotherapy; Tr, trastuzumab; AC, doxorubicin and cyclophosphamide; P, paclitaxel; D, docetaxel; CB, carboplatin; FEC, fluorouracil, epirubicin, and cyclophosphamide; ED, epirubicin and docetaxel; V, vinorelbine; NA, not applicable; HR, hazard ratio; 95% CI, 95% confidence interval; RFS, recurrence-free survival; NR, not reported. * P < y Recurrence-free survival. z Censored analysis: HR, 0.63; 95% CI, (P <.0001). The HR was adjusted for lymph nodes, receptor status, paclitaxel schedule, and protocol. Censored analysis: HR, 0.63; 95% CI, (P ¼.0051). Cancer March 15,

5 crossover (51%) may affect outcomes, particularly overall survival; and 3) the prior chemotherapy in this trial may be considered suboptimal. In addition, women were randomized after they completed chemotherapy, which may have resulted in a more favorable cohort, because those with early recurrences were omitted and were not eligible for enrollment. The 5 subsequent adjuvant trastuzumab trials investigated the administration of trastuzumab in conjunction with prescribed chemotherapy, whereby trastuzumab was administered either as a chemo-trastuzumab sequence or concurrently with the taxane component of the prescribed regimen (concurrent trastuzumab-taxane regimens) in women with HER-2-positive tumors. The North Central Cancer Treatment Group trial (N9831) 10,13,15 and the National Surgical Adjuvant Breast and Bowel Project trial (B-31) 10 are multicenter, open-label, randomized controlled trials that were designed to explore the benefits of adding trastuzumab to an adjuvant combination of doxorubicin and cyclophosphamide (AC) followed by paclitaxel. In the N9831 trial, 10 women with lymph node-positive and high-risk lymph node-negative breast cancer were assigned randomly to 1 of 3 arms: 1) women who received AC then weekly paclitaxel (AC!P), 2) women who received the same AC!P with weekly trastuzumab (4 mg/kg load followed by 2 mg/kg per week) administered concurrently beginning with the first dose of paclitaxel and continuing for 1 year (AC!P/Tr), 3) and women who received the same AC!P followed sequentially, after the completion of paclitaxel, by 1 year of weekly trastuzumab therapy at the same dose and schedule as the other 2 arms (AC!P!Tr). In the B-31 trial, 10 only lymph node-positive patients were eligible. Enrolled women were assigned to 1 of 2 arms: 1) AC followed by paclitaxel administered every 3 weeks (AC!P), and 2) the same AC!P with weekly trastuzumab administered concurrently beginning with the first cycle of paclitaxel and continuing for 1 year (AC!P/Tr). Because the N9831 and B-31 trials had somewhat similar control and treatment arms, a joint analysis (N9831/B-31 Analysis) comparing the concurrent trastuzumab-taxane arms (AC!P/Tr) with the control arms (AC!P) of the 2 trials was conducted. The primary endpoint of the N9831/B-31 Analysis was disease-free survival (DFS). A subset of 3968 women qualified for inclusion in the joint analysis, of whom 93% had lymph node-positive disease (Table 1) and approximately 57% had received weekly paclitaxel. 13 At a median follow-up of 2.9 years, despite a 21% crossover rate, 13 AC!P/Tr produced significant reductions in the risk of both recurrence (HR, 0.48; 95% CI, [P <.00001]) and death (HR, 0.65; 95% CI, [P ¼.0007]) compared with AC!P (Table 2). 13 These data confirm the findings of the HERA trial. When considering these findings, it is important to remember the following: 1) the joint analysis of 2 distinct trials with different treatment and statistical approaches is unorthodox; 2) the standards for validation and source verification used in these trials were less rigorous than those used in the HERA trial and the Breast Cancer International Research Group 006 (BCIRG 006) trial; 3) a small percentage of patients who were treated on the study were identified as HER-2 negative at central review 13,16 ; and 4) AC!P may be considered suboptimal adjuvant therapy. 3,17 The BCIRG 006 trial is an international, multicenter, open-label, randomized controlled study 12,18 that was designed to assess the benefits of adding trastuzumab to an anthracycline/docetaxel-based or a nonanthracycline, docetaxel, and carboplatin-based adjuvant chemotherapy regimen in patients with HER-2-positive breast cancer. In total, 3222 women were assigned randomly to 1 of 3 treatment arms: 1) women who received a combination of AC followed by docetaxel administered every 3 weeks (AC!D); 2) women who received AC!D with concurrent, weekly trastuzumab beginning with docetaxel (4 mg/ kg load followed by 2 mg/kg weekly) for the duration of chemotherapy followed by trastuzumab administered every 3 weeks for a total therapy duration of 1 year (AC!D/Tr); and 3) women who received a combination of carboplatin, docetaxel, and weekly trastuzumab followed by trastuzumab administered every 3 weeks for a total therapy duration of 1 year (CB/D/Tr). Patient characteristics were well balanced across study arms, and approximately 71% of patients had lymph node-positive disease (Table 1). 12 At a median follow-up of 36 months, both the AC!D/Tr regimen and the CB/D/Tr regimen, compared with AC!D, produced significant reductions in the risk of both recurrence (HR, 0.61; 95% CI, [P <.0001]; and HR, 0.67; 95% CI, [P ¼.0003], respectively) and death (HR, 0.59; 95% CI, [P ¼.004]; and HR, 0.66; 1158 Cancer March 15, 2009

6 Adjuvant Trastuzumab for Breast Cancer/Mackey et al 95% CI, [P ¼.017], respectively) (Table 2). 12 When considering these study results, it is important to note that: 1) the results of the BCIRG 006 trial have not yet been published or peer reviewed, and 2) the nonanthracycline CB/D regimen is not an established adjuvant regimen for breast cancer. The Protocole Adjuvant dans le Cancer du Sein (PACS) 04 study is a multicenter, open-label, randomized controlled trial. 19 It is the first study designed to evaluate the benefits of a chemo-trastuzumab sequence following a prescribed escalated-dose epirubicin or epirubicin-taxane regimen. In total, 3010 women with axillary lymph node-positive breast cancer were assigned randomly to 1 of 2 arms: 1) a combination of epirubicin and docetaxel (ED), or 2) a combination of escalateddose epirubicin, fluorouracil and cyclophosphamide (FEC 100 ). Once their HER-2 status was determined, women with HER-2-positive tumors (n ¼ 528) immediately underwent a second randomization to 1 year of trastuzumab administered every 3 weeks (8 mg/kg load followed by 6 mg/kg every 3 weeks) or observation. As in the HERA trial, trastuzumab therapy was initiated after the completion of both chemotherapy and radiotherapy. Baseline characteristics of the patients with HER- 2-positive disease were well balanced (Table 1). 19 At a median follow-up of 4 years, the PACS-04 study has the longest follow-up of trastuzumab trials reported to date. Results from the first randomization comparing concomitant ED with FEC 100 have yet to be reported; however results of the second randomization failed to demonstrate a significant reduction in the risk of recurrence (HR, 0.86; 95% CI, [P ¼.41]) or death (HR, 1.27; 95% CI, [P value not reported]) for chemo-trastuzumab sequences compared with observation (Table 2). When considering these findings, it is important to note that: 1) the study may have been underpowered because of the small sample size; 2) as in the HERA trial, the sequential approach may have precluded a potential benefit from the concurrent administration of chemotherapy and trastuzumab ; 3) in contrast to the HERA trial population, a large proportion of patients in the PACS-04 trial received optimal chemotherapy, which may explain the reduced incremental benefit of trastuzumab observed in the PACS-04 relative to the HERA trial; and 4) the PACS-04 results have not been published or peer reviewed. The Finland Herceptin (FinHer) study is a multicenter, open-label, randomized controlled trial 11 that was designed to assess the benefits of docetaxel administered every 3 weeks (D) or weekly vinorelbine (V) followed by fluorouracil, epirubicin, and cyclophosphamide (FEC) in 1010 women with axillary lymph node-positive or highrisk, lymph node-negative breast cancer. The study also assigned 232 women with HER-2-positive tumors to receive 9 cycles of weekly trastuzumab (4 mg/kg load followed by 2 mg/kg weekly) administered concurrently with either docetaxel, or vinorelbine, or no further treatment. Baseline characteristics of the patients were balanced with the exception of axillary lymph node metastases, which were somewhat more frequent in the trastuzumab group than in the observation group (89% vs 78%, respectively) (Table 1). 11 At a median follow-up of 3 years, as in the 4 larger trastuzumab trials, fewer events of recurrence were evident for women who received trastuzumab (HR, 0.42; 95% CI, [P ¼.01]) compared with women who did not (Table 2), although overall survival reached only borderline significance, most likely because of the small sample size (HR, 0.41; 95% CI, [P ¼.07]) (Table 2). 11 Although the short 9-week course of trastuzumab therapy (short-course trastuzumab therapy) proved beneficial, caution should be used when interpreting these findings, because: 1) the sample size of the HER-2-positive patient population was small; 2) the study prescribed lower than standard doses of both anthracyclines and taxanes; and 3) although there was a trend toward improved overall survival with the addition of trastuzumab, the findings were not statistically significant. Safety Noncardiac adverse events Trastuzumab is associated with few grade 3 or 4 adverse effects. Small but significant increases in severe adverse events were reported with trastuzumab compared with observation (11% vs 6%; P <.0001) in the HERA trial. 9 However, the addition of trastuzumab to chemotherapy did not result in clinically significant increases in the incidence of adverse events compared with controls in the FinHer trial, 11 the N9831/B-31 Analysis, 10 or the AC!D/Tr arm of the BCIRG 006 trial. 12 The CB/D/Tr regimen was the only treatment that produced less severe toxicity compared with controls. 12 Cancer March 15,

7 Cardiotoxicity Because both anthracyclines and trastuzumab are cardiotoxic, significant measures were undertaken to mitigate cardiac risk in all trials. These include appropriate eligibility criteria, cardiac screening and monitoring, and strict criteria for the initiation and discontinuation of trastuzumab therapy. 8-12,19 Chemo-trastuzumab sequences were explored in 3 of the 6 trials: the HERA trial, the sequential arm of the N9831 trial, and the PACS-04 trial. The reported cardiac data captured differing components of therapy across trials. Because patients in the HERA and PACS- 04 trials were randomized after the completion of chemotherapy, the reported cardiotoxicity data from these trials reflects trastuzumab-induced cardiotoxicity exclusively, because patients with compromised cardiac capacity after chemotherapy were excluded from the trastuzumab portion of the trial. In the N9831 trial, although patients were randomized before they received chemotherapy, the trial reports cumulative post-ac cardiotoxicity, which may not fully capture the anthracycline-induced cardiotoxicity of the AC portion of chemotherapy. Furthermore, the rigor of cardiac screening in the 3 trials differed considerably: Patients in the HERA trial had a more favorable cardiac profile (left ventricular ejection fraction [LVEF] >55% after chemotherapy and locoregional therapy) relative to patients in both the PACS-04 trial (LVEF 55% or between 50% and 55% with cardiologist approval) and the N9831 trial (LVEF >50%). Chemo-trastuzumab sequences were associated with increases in cardiotoxicity in all 3 trials. Patients in the HERA trial experienced a small but significantly greater degree of cardiotoxicity with trastuzumab compared with observation, including a confirmed, significant decrease in LVEF (3.04% vs 0.53%; P <.0001), symptomatic congestive heart failure (CHF) (2.15% vs 0.12%; P <.0001), and severe CHF (0.60% vs 0%; P <.0001); thus, 4.3% of patients were unable to complete trastuzumab therapy for cardiac reasons. 9,23 Patients in the PACS-04 trial also experienced increased cardiotoxicity with the addition of trastuzumab, resulting in increases in severe CHF (1.7% vs 0.4%) and a higher rate of discontinuation due to cardiotoxicity (17.9%) (Table 3). 19 Finally, when sequential trastuzumab was added to the AC!P arm of the N9831 trial, increases in cardiac events once again were apparent (2.8% vs 0.3%). 24 Concurrent trastuzumab-taxane regimens were explored in 3 of the 6 trials. 10,12,15 In all 3 studies, only patients who had an LVEF >50% after the completion of AC therapy were eligible for trastuzumab therapy, resulting in a small proportion of enrolled patients in all 3 trials being denied trastuzumab therapy (N9831/B31 Analysis, 6.7% of patients; BCIRG 006 trial, 2% of patients). Cardiotoxicity was evident in all 3 trials. The addition of trastuzumab to paclitaxel after the completion of AC resulted in an increase in cumulative cardiac events (either events from cardiac causes or New York Heart Association [NYHA] class III or IV CHF) compared with controls in both the N9831 and B31 trials (3-year cumulative cardiac events [N9831], 3.3% vs 0.3%; 5-year cumulative cardiac events [B31], 3.8% vs 0.9%). 24,25 Furthermore, 15% of patients in the N9831 trial and 19% of patients in the B31 trial were unable to complete trastuzumab therapy for cardiac reasons. 26,27 Increased cardiotoxicity also was apparent in the AC!D/Tr arm of the BCIRG 006 compared with controls (>10% relative LVEF decline; 18.1% vs 10.1%, respectively; P <.0001; NYHA grade 3 or 4 CHF, 1.9% vs 0.4%, respectively) 12 ; and, similar to the other concurrent trastuzumab-taxane trials, 8.3% of patients who received AC!D/Tr were unable to complete therapy because of trastuzumab-mediated cardiotoxicity (unpublished data). The nonanthracycline-containing, concurrent trastuzumab-taxane regimen (CB/D/Tr) of the BCIRG 006 trial and the short-course trastuzumab regimen of the FinHer trial were the least cardiotoxic. At 3 years, no increase in cardiotoxicity was apparent for CB/D/Tr compared with controls (>10% relative LVEF decline, 8.6% vs 10.1%; grade 3/4 CHF, 0.4% vs 0.4%) in the BCIRG 006 trial (Table 3), 12 and none of the 4 patients who experienced cardiac infarction or failure in the FinHer trial had received trastuzumab (Table 3). 11 DISCUSSION There is overwhelming evidence to support the overall benefits of adding adjuvant trastuzumab to chemotherapy in women with HER-2-positive cancer despite the risk of developing clinically significant cardiac toxicity. Although the benefit of adjuvant trastuzumab is clear, many 1160 Cancer March 15, 2009

8 Table 3. Cardiac Safety Study Group N9831/B-31 Analysis: Perez 2008, 24 Rastogi 2007, 25 Perez 2005, 26 Tan-Chiu HERA: Suter ACfiP/Tr ACfiP BCIRG 006: Slamon * PACS 04: FinHer: Joensuu Spielmann Variable CTfiTr CTfi Observation N9831 B-31 N9831 B-31 ACfiD/Tr CB/ D/Tr Control FEC or EDfiTr FEC or ED Dor V/TrfiFEC Dor VfiFEC Sample size, no. of patients y z Trastuzumab discontinued because of cardiac problems Percentage of patients 4.3 NA NA NA 8.3* 3.6* NA 17.9 NA NR NA Median follow-up, y 2 3 Decreases in LVEF 15% points from baseline Percentage of patients NR NR NR NR 3.5# 6.0 zz P < Median follow-up, y Severe CHF Percentage of patients (no.) ** 3.8 (n¼35) 25 yy ** 0.9 (n¼6) 25 yy 1.9 (n¼20)zz (n¼4)zz (n¼4)zz (n¼4) 0.4 (n¼1) 0.0 (n¼0) NR Median follow-up, y Cardiac death Percentage of patients (no.) 0.0 (n¼0) 0.06 (n¼1) 0.2 (n¼1) (n¼0) (n¼1) (n¼1) (n¼0) (n¼0) (n¼0) (n¼0) 0.0 (n¼0) 0.0 (n¼0) 0.0 (n¼0) Median follow-up, y HERA indicates the Herceptin Adjuvant trial; N9831, North Central Cancer Treatment Group trial N9831; B-31, National Surgical Adjuvant Breast and Bowel Project trial B-31; BCIRG 006, Breast Cancer International Research Group trial 006; PACS-04, Protocole Adjuvant dans le Cancer du Sein; FinHer, Finland Herceptin study; CT, chemotherapy; Tr, trastuzumab; AC, doxorubicin and cyclophosphamide; P, paclitaxel; D, docetaxel; CB, carboplatin; FEC, fluorouracil, epirubicin, and cyclophosphamide; ED, epirubicin and docetaxel; V, vinorelbine; NA, not applicable; LVEF, left ventricular ejection fraction; NR, not reported; CHF, congestive heart failure. * Unpublished data. yat 3.75 years, n¼570. At 3.75 years, n¼664. A decrease in LVEF 10% from baseline to an LVEF <50% at any time. The greatest change in LVEF 15% compared with the pre-ac level at any time point in the study. Defined as a >10% relative LVEF decline. # Defined as 1 measurement(s) of LVEF <15% of the baseline value. **The 3-year cumulative incidence of cardiac events. A cardiac event was defined as death from cardiac causes or New York Heart Association class III or IV CHF. yythe 5-year cumulative incidence of cardiac events. A cardiac event was defined as death from cardiac causes, or National Heart Association class III or IV CHF, and an absolute drop in LVEF on a mutigated acquisition scan or echocardiogram of >10% points to below 55% or >5% and below the lower limit of normal in subpopulations of patients eligible for trastuzumab therapy. zzgrade 3 or 4 CHF, excluding arrhythmias. Defined as cardiac failure or cardiac infarction.

9 questions remain related to the clinical application of these findings. Are There Patient Groups That Benefit More or Less From Trastuzumab Therapy? In light of the risks and costs associated with trastuzumab therapy, the accurate identification of patients with HER- 2-positive tumors who preferentially may benefit from trastuzumab therapy is imperative. Both fluorescent in situ hybridization (FISH) and chromogenic in situ hybridization, as well as immunohistochemistry testing with or without FISH confirmation, have been used in the adjuvant setting to assess HER-2 status Central confirmation of HER-2 status before enrolment was required in all studies with the exception of an initial cohort of the N9831 trial and all patients in the B-31 trials, which relied on testing in approved local laboratories. Differences between local and central HER-2 assessment and between immunohistochemistry and FISH testing have been reported in the N9831 and B-31 trials, revealing a proportion of patients in both trials with HER-2-negative tumors at central review (8.8% and 9.7, respectively). 13,28 Unexpectedly, in addition to the benefits of trastuzumab among HER-2-positive patients, exploratory subgroup analyses from both the N9831 trial and the B-31 trial revealed a potential benefit for trastuzumab among patients who tested negative for HER-2 at central review. 13,28 Discordance in HER-2 testing and the possibility of heterogeneity in relative treatment effect underscore the importance of continued research into optimal HER-2 testing and the use of standardized HER-2 testing in clinical practice. Adjuvant trastuzumab trials enrolled significant proportions of patients with lymph node-positive and highrisk, lymph node-negative disease. Although low numbers of very small T1a and T1b tumors were included in the high-risk, lymph node-negative cohorts, 9,12 nevertheless, it is reasonable to consider the use of the aforementioned trastuzumab regimens in all patients who have lymph node-negative, HER-2-positive disease with a high-risk of recurrence. None of the trials that we reviewed were powered to detect the effect of trastuzumab in patient subgroups at the time of their interim analyses. However, an early look at the forest plots of HRs for DFS in the HERA trial, 9 the BCIRG 006 trial, 12 and the N9831/B-31 Analysis 10,13 suggest that all eligible patient subgroups, regardless of age, lymph node status, menopausal status, or hormone status, benefited equally from trastuzumab. Both HER-2/neu amplification and topoisomerase IIa (Topo IIa) amplification are associated with a relative benefit from anthracycline-containing regimens compared with nonanthracycline-containing regimens It has been postulated that this benefit may relate to the physical proximity on chromosome 17q of the HER-2 and Topo IIa gene, because topoisomerase II is involved integrally in the mechanism of action of the anthracyclines. Data from the BCIRG 006 trial demonstrated that CB/D/Tr is equivalent to AC!D/Tr in patients overall; however, in the 35% of women with HER2-positive early breast cancer in whom the Topo IIa and HER-2 genes are coamplified, CB/D/Tr and AC!D/Tr are equivalent to AC!D. 12 These data suggest that anthracycline-containing chemotherapy may not be required in HER-2-overexpressing women and/or that trastuzumab may not be required in addition to an anthracycline-containing regimen in women with HER-2/Topo IIa coamplified tumors. A meta-analysis of studies comparing higher dose anthracycline-containing regimens with standard-dose anthracycline-containing regimens suggested that women with HER-2-positive tumors benefited from higher dose chemotherapy compared with standard-dose chemotherapy, whereas women with HER-2-negative tumors did not. 35 Similarly, a significant interaction between HER-2 status and taxane-containing versus nontaxane-containing regimens have been reported. 36,37 In Cancer and Leukemia Group B trial 9344 trial, both women with HER-2- positive tumors and women with HER-2-negative tumors benefited from the addition of taxanes; however, those with HER-2-positive tumors did so to a greater degree. An analysis of the PACS-04 and BCIRG 006 clinical trial populations designed to assess the benefits of HER-2 and Topo IIa as predictors of response to trastuzumab currently is underway. Should Trastuzumab Be Administered Concurrently or Sequentially With Chemotherapy? Annualized DFS hazards rates for patients with HER-2- positive early breast cancer who have not received 1162 Cancer March 15, 2009

10 Adjuvant Trastuzumab for Breast Cancer/Mackey et al trastuzumab indicate an increased risk of recurrence within 1 to 2 years, providing a rationale for the earliest possible intervention with trastuzumab. 9,13 Concurrent trastuzumab-taxane regimens minimize postsurgery trastuzuamb treatment delays (1 4 months), whereas chemo-trastuzumab sequences administered in the HERA trial (8 months) and the PACS-04 trial 9,19 tend to incur greater trastuzumab delays. At 3 years, all 4 concurrent trastuzumab-taxane regimens 10,12 resulted in improved DFS compared with controls, and 2-year results from the HERA trial demonstrated a similar improvement in DFS for chemo-trastuzumab sequences compared with observation. However, at 4 years, outcomes for the PACS-04 trial did not indicate a benefit for aggressive anthracycline-based regimens or anthracycline-taxane regimens compared with controls (HR, 0.86; 95% CI, [P ¼.41]) despite a trend toward improved DFS during the first 18 months (HR, 0.57; 95% CI, [P value not reached]). 19 This suggests a potential attenuation of the benefit of trastuzumab with delayed administration, which may indicate a cytostatic effect for sequential trastuzumab administration, whereas concurrent use may result in a cytotoxic effect. The N9831 trial is the only adjuvant trastuzumab trial that was designed to directly compare the sequential versus concurrent approaches. 15 Although a sufficient number of events are currently unavailable to complete this analysis, initial data suggest that sequential trastuzumab administration may result in inferior efficacy (DFS: HR, 0.64; 95% CI, [P ¼.0114]) compared with concurrent administration. 15,24 Until the final results of this analysis are available, the benefits of concurrent versus sequential administration remain unclear. What Is the Optimal Duration of Trastuzumab-based Therapy? The balance of available evidence supports the use of 1-year of adjuvant trastuzumab therapy. 8-10,12,13 The HERA trial, which was designed to evaluate duration of therapy, assessed the benefits of 2 years of therapy compared with 1 year. 9 Results from that trial, anticipated in the fall of 2008, will provide insight into the risk-benefit ratio for extended trastuzumab exposure; however, even longer follow-up will be required to confirm the safety of this treatment approach. The use of short-course trastuzumab therapy has clinical appeal because of the enhanced safety profile, reduced cost, and improved quality of life of this approach compared with 1 year of trastuzumab therapy. The significant reductions in the risk of recurrence observed with short-course trastuzumab therapy compared with controls in the FinHer trial 11 and the noninferiority of short-course trastuzumab therapy relative to 1 year of therapy in the randomized phase 2 E2198 pilot trial, 38 although not powered to assess therapy duration, support this approach. Nevertheless, evidence from larger randomized clinical trials, such as the Synergism or Long Duration (SOLD) study and the Protocol of Herceptin Adjuvant with Reduced Exposure (PHARE) trial, directly comparing short versus longer durations of trastuzumab therapy is needed to establish the noninferiority of this approach. Until mature study findings from these trials are available, short-course trastuzumab therapy should be reserved for instances in which institutional budgetary restraints or patient preference limit the administration of a full year of trastuzumab therapy. What Are the Implications of Trastuzumab-related Cardiotoxicity? Unlike anthracycline-mediated cardiotoxicity, some degree of cardiac recovery appears to be possible with trastuzumab-mediated cardiotoxicity. In the HERA trial, heart failure treatment was recommended for all patients with severe CHF. At a median follow-up of 1 year, of the 10 patients who had severe CHF at the last scheduled assessment, 8 patients (80%) were asymptomatic, and 6 patients (60%) had an LVEF 55%. 23 In the N9831 trial, at a median follow-up of 2 years, the majority of patients who developed CHF improved on medical treatment, and LVEFs >50% were observed in approximately 50% of patients who received sequential therapy and in 61% of patients who received concurrent therapy at 2 years. 26 Finally, in the B-31 trial, with 5 years follow-up, women who had confirmed CHF and at least 6 months of follow-up usually were asymptomatic (85%). Although a large proportion required medication (61%), approximately 66% of patients had an LVEF >50% on a followup multigated acquisition (MUGA) scan. 25 Cancer March 15,

11 Which Patients Are at a Greater Risk of Cardiotoxicity? The ability to reliably predict a patient s risk of developing CHF is of great clinical importance. The HERA, N9831, and B-31 trials conducted analyses to identify factors that were correlated with an increased risk of cardiotoxicity with adjuvant trastuzumab therapy. Lower screening LVEFs, lower post-ac LVEFs, 25 and age 24,25 were associated with significant increases in cardiotoxicity. Other factors that were associated with increased cardiotoxicity included higher doses of anthracyclines, 23 increased body mass index, 23 and the requirement for hypertensive medications. 24,25 Although the analyses of potential risk factors are exploratory and were based on small numbers, routine cardiac risk assessment is recommended for all trastuzumab candidates, especially those at an increased risk for cardiotoxicity. Should Trastuzumab-based Therapy Be Given Concurrently With Radiotherapy? Preclinical data indicate that trastuzumab may exert a radiosensitizing effect on HER-2-positive breast cancer cells, although it is uncertain whether it has a similar effect on normal cells. 39,40 In the adjuvant studies that we reviewed, trastuzumab was administered either concurrently with radiotherapy after the completion of chemotherapy 10 or after the completion of both chemotherapy and radiotherapy. 9,19 When trastuzumab was administered concurrently, left-sided tumors and radiation therapy were not identified as cardiac risk factors, 25 radiotherapy was not observed to increase cardiac events or radiation-related adverse events, and internal mammary radiotherapy appeared to be feasible when the cardiac dose was limited. 39 These study findings support the continued practice of concurrent trastuzumab and radiotherapy administration after the completion of chemotherapy in conjunction with continued cardiac follow-up. Should Trastuzumab-based Therapy Be Given Concurrently With Endocrine Therapy? There is a strong rationale for the concurrent, early administration of endocrine therapy and trastuzumab in HER-2-positive and hormone receptor-positive breast cancer because of preclinical evidence of cross talk between HER-2 and estrogen receptor signaling pathways. Recent evidence from the first-line metastatic breast cancer setting provides further support for this approach. 41 Women who received concurrent endocrine therapy had a benefit from trastuzumab similar to the benefit experienced by those who did not receive endocrine therapy with no increase in toxicity. 8,10-12,19 These findings support the continued practice of concurrent endocrine and trastuzumab administration after the completion of chemotherapy in HER-2-positive and hormone receptor-positive patients. How Do Trastuzumab-based Regimens Compare in Terms of Cost? The administration of adjuvant trastuzumab therapy is associated with considerable expense. In addition to the cost of diagnostic testing, significant treatment costs include acquisition costs, administration costs (nursing and pharmacy time), and the costs of clinical cardiac monitoring. Numerous cost-effectiveness studies suggest that costs are comparable to other accepted and standard uses of healthcare dollars ; nevertheless, factoring in treatment costs when selecting from among safe and effective treatment options improves the overall cost-benefit ratio of the intervention. Selected treatment costs for chemotrastuzumab sequences and concurrent trastuzumab-taxane regimens are summarized in Table 4. Trastuzumab acquisition costs for 1 year of therapy were the single greatest factor contributing to total treatment costs (in U.S. dollars, between $41,459 and $43,023), followed by cardiac monitoring by MUGA scan (between $3675 and $4410). The administration costs associated with 1 year of trastuzumab therapy administered every 3 weeks were significantly less than those associated with weekly administration ($510 vs $1473). The inclusion of taxanes increased total treatment costs. The short-course trastuzumab regimen from the FinHer trial ($16,747) was associated with the lowest total treatment cost. Which Trastuzumab-based Regimen Is Most Convenient for Patients? The patient-related convenience of a regimen can have a significant impact on a patient s quality of life, especially 1164 Cancer March 15, 2009

12 Adjuvant Trastuzumab for Breast Cancer/Mackey et al Table 4. Treatment Costs Chemotherapy/Trastuzumab Administration Chemotherapy Acquisition Costs (US$)* Trastuzumab Acquisition Costs (US$) Administration Costs (US$) Clinical Monitoring for Duration of Treatment (US$) Total (US$) Chemotherapy-trastuzumab sequences Anthracycline, nontaxane (68%) FEC , ,911 Anthracycline, taxane (26%) ED36 11,560 43, ,278 AC34fiP , ,569 FEC fiD , ,197 Concurrent trastuzumab-taxane regimens N9831 ACfiP/Tr (1 y) , ,252 B-31 ACfiP/Tr (1 y) , ,702 BCIRG 006 ACfiD/Tr (1 y) , ,760 CB/D/Tr (1 y) , ,191 FinHer D/TrfiFEC ,747 FEC 100 indicates escalated-dose fluorouracil, epirubicin (100 mg/m 2 /cycle), and cyclophosphamide; ED, epirubicin and docetaxel; AC, doxorubicin and cyclophosphamide; P, paclitaxel; D, docetaxel; N9831, North Central Cancer Treatment Group trial N9831; Tr, trastuzumab; B-31, National Surgical Adjuvant Breast and Bowel Project trial B-31; BCIRG 006, Breast Cancer International Research Group trial 006; CB, carboplatin; FinHer, Finland Herceptin study. * Assumptions for cost calculations: cost/mg Medbuy Buying Group, March 2007 (Tr, $6.14; fluorouracil, $0.01; doxorubicin, $0.57; epirubicin, $4.00; D, $11.42; P, $0.70; cyclophosphamide,$0.01; vincristine; $1.93; CB, $0.10); body surface area¼1.7 m 2, weight¼65 kg; an area under the serum concentrationtime cure of 6¼600 mg; Canadian to US $ conversion factor (May 2008)¼$0.98. y Assumptions for administration costs used nonnormalized complexity values from the Clinical Care Options Oncology website (available at: July 2007); complexity values for chemotherapy and trastuzumab additive: wage per minute and benefits for pharmacy and nursing¼$0.85; administration costs were calculated by multiplying the complexity value per component by wage per minute by the number of cycles. z We assumed that a mutigated acquisition scan was obtained either at baseline or at the initiation of trastuzumab, either after the anthracycline-containing component or at the initiation of trastuzumab, and every 3 months for the duration of trastuzumab therapy. when patients live at a distance or have limited support networks. Selected factors that affect the quality of life of patients receiving either chemo-trastuzumab sequences or concurrent trastuzumab-taxane regimens are summarized in Table 5. Of the regimens that administered 1 year of therapy, the concurrent trastuzumab-taxane regimens resulted in shorter total treatment durations compared with the chemo-trastuzumab sequences. The administration of both trastuzumab and paclitaxel every 3 weeks resulted in significantly fewer patient visits compared with weekly administration. Total clinic time was shortest for nontaxane trastuzumab regimens and longest for regimens that involved paclitaxel. The short-course trastuzumab regimen of the FinHer study was considerably more convenient overall than all other regimens. Ongoing Clinical Trials Although many advances have been made in the use of adjuvant trastuzumab, our journey into the new frontier of targeted therapy has only just begun. The HERA, N9831, PHARE, and SOLD trials promise further insights into the optimal use of trastuzumab in terms of sequencing, treatment duration and chemo-trastuzumab combinations. Large randomized trials exploring the benefits of lapatinib used in conjunction with chemotherapy or trastuzumab (the Adjuvant Lapatinib and/or Trastuzumab Treatment Optimization or ALTTO trial and the Tykerb Evaluation After Chemotherapy or TEACH trial) and the effects of combining bevacizumab in combination with adjuvant trastuzumab and chemotherapy (the Cancer March 15,

13 Table 5. Factors That Influence Patient Convenience Chemotherapy/Trastuzumab Administration Total Treatment Duration, Months* Total No. of Visits/ Treatments Total Clinic Time/ Treatment, Hours Chemotherapy-trastuzumab sequences Anthracycline, nontaxane (68%) FEC Anthracycline, taxane (26%) ED AC34fiP FEC fiD Concurrent trastuzumab-taxane regimens N9831 ACfiP/Tr B-31 ACfiP/Tr BCIRG 006 ACfiD/Tr CB/D/Tr FinHer D/TrfiFEC FEC 100 indicates escalated-dose fluorouracil, epirubicin (100 mg/m 2 /cycle), and cyclophosphamide; ED, epirubicin and docetaxel; AC, doxorubicin and cyclophosphamide; P, paclitaxel; D, docetaxel; N9831, North Central Cancer Treatment Group trial N9831; Tr, trastuzumab; B-31, National Surgical Adjuvant Breast and Bowel Project trial B-31; BCIRG 006, Breast Cancer International Research Group trial 006; CB, carboplatin; FinHer, Finland Herceptin study. * For this analysis, the assumption was that 1 month¼4.3 weeks. y Assumptions for calculating total clinic time used the highest quoted value per component. Bevacizumab and Trastuzumab Adjuvant Therapy in HER-2-Positive Breast Cancer or BETH study) will provide insight into the benefits of combining targeted therapies. In conclusion, the current review indicates that there is level 1 evidence to support the routine use of 1 year of adjuvant trastuzumab in conjunction with chemotherapy in women who have early-stage, HER-2-positive breast cancer. Because adjuvant trastuzumab is associated with a risk of cardiotoxicity in the form of severe CHF, compromised LVEF function, and potential for cardiovascular disease, every effort should be made to preserve cardiac function through routine cardiac risk assessment and the selection of minimally cardiotoxic regimens. Conflict of Interest Disclosures Information support services for this review were funded through an unrestricted educational grant administered through a third party from Sanofi-Aventis. Dr. Mackey has received honoraria for Continuing Medical Education speaking events from the following: Sanofi-Aventis, Roche Oncology, and GlaxoSmithKline Oncology. Dr. Verma has served as a consultant and as a member of the advisory board for GlaxoSmithKline, Roche Canada, and Sanofi- Aventis. Over the past 3 years, Dr. Pritchard has been a consultant with Sanofi-Avenits, AstraZeneca, Roche, Pfizer, Ortho-Biotech, YM Biosciences, Novartis, Abraxis, Amgen, and GlaxoSmithKline. She has received research funding either directly, or through per case funding for studies, or indirectly through the National Cancer Institute of Canada Clinical Trials Group, contracted with pharmaceutical companies including AstraZeneca, YM Biosciences, Bristol Myers-Squibb, Sanofi-Aventis, Amgen, Ortho-Biotech, Pfizer, Novartis, GlaxoSmithKline, and Ortho Biotech. She also has received honoraria or has been part of the Speaker s Bureau for Sanofi-Aventis, AstraZeneca, Pfizer, Roche, YM Biosciences, and Novartis and has provided paid expert testimony for Sanofi-Aventis, AstraZeneca, and GlaxoSmithKline. Dr. Pritchard also has been a member of the Advisory Committee for Sanofi-Aventis, AstraZeneca, Ortho-Biotech, Roche, Pfizer, Novartis, YM Biosciences, and GlaxoSmithKline. Dr. Laing has received honoraria from Sanofi-Aventis and Roche and also has received travel grants to attend ASCO 2007 from Sanofi-Aventis. Dr. Provencher is a member of the Canadian Advisory Boards for the pharmaceutical companies Roche and Sanofi-Aventis. References 1. Martin M, Rodriguez-Lescure A, Ruiz A, et al. Multicenter, randomized phase III study of adjuvant chemotherapy for node positive breast cancer comparing 6 cycles of FE90C versus 4 cycles of FE90C followed by 8 weekly paclitaxel 1166 Cancer March 15, 2009