CADTH Reimbursement Review

Pertuzumab (Perjeta)

Sponsor: Hoffmann-La Roche Ltd.

Therapeutic area: Early-stage breast cancer

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Clinical Review

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Introduction

Breast cancer will strike 1 in 8 Canadian women during their lifetime and 1 in 33 will die of the disease. Due to advances in and widespread use of screening programs, breast cancer is typically diagnosed at an early stage, and approximately 95% of cases are diagnosed at stages I to III. Prior to the advent of anti–human epidermal growth factor receptor 2 (HER2) therapies, patients who had HER2-positive breast cancer had a poorer prognosis than those without HER2 overexpression. Patients undergoing neoadjuvant therapy can be subclassified into those with locally advanced breast cancers that are not operable, those with locally advanced cancers that are operable, and those with primary operable breast cancers. In locally advanced cancers, 1 purpose of neoadjuvant therapy is to convert the tumour from an inoperable to an operable state. In all cancers, the purpose of neoadjuvant therapy is to downstage the tumour to potentially avoid mastectomy in favour of breast-conserving surgery, to assess response to systemic therapy, to potentially escalate or de-escalate subsequent adjuvant therapy based on the response to neoadjuvant therapy, and to initiate systemic therapy early to try to limit systemic spread. Neoadjuvant therapy is the standard of care according to international and local guidelines for patients who are at stage II or III, and some patients with stage I disease. By treating with chemotherapy early, the risk of systemic recurrence is also decreased. A small number of patients with small tumours may have surgery first followed by adjuvant therapy, and this approach would be the norm for stage I disease. The standard regimen, according to the clinical experts consulted by CADTH for this review, for stages II and III HER2-positive breast cancer, would be doxorubicin, cyclophosphamide, and paclitaxel plus trastuzumab or docetaxel and carboplatin plus trastuzumab. The American Society of Clinical Oncology (ASCO) guidelines for neoadjuvant management of HER2-positive, node-positive, or high-risk node-negative breast cancer recommend an anthracycline and taxane or a non-anthracycline-based regimen in combination with trastuzumab; they also note that pertuzumab may be added to this regimen. In many regions, including the US, Europe, and the UK, pertuzumab would be added to this regimen.¹ The European Society for Medical Oncology recommends a dual blockade with pertuzumab and trastuzumab in high-risk patients (with node-positive or estrogen receptor–negative disease), starting before or after surgery; for neoadjuvant therapy, the German Gynecological Oncology Group recommends adding pertuzumab when treating patients with node-positive disease.^2,3 The goal of neoadjuvant treatment is curative.

Pertuzumab is an HER2 inhibitor. It is administered by IV infusion in combination with trastuzumab and chemotherapy. After an initial 840 mg loading dose, the maintenance dose is 420 mg every 3 weeks for 3 to 6 cycles in the neoadjuvant setting. It is indicated, in combination with trastuzumab and chemotherapy, for the neoadjuvant treatment of patients with HER2-positive locally advanced, inflammatory, or early-stage breast cancer (either > 2 cm in diameter or node-positive). This drug was previously reviewed by CADTH for off-label use under the same indication.

The objective of this report is to perform a systematic review of the beneficial and harmful effects of pertuzumab by IV infusion in combination with trastuzumab and chemotherapy for the neoadjuvant treatment of patients with HER2-positive locally advanced, inflammatory, or early-stage breast cancer (either > 2 cm in diameter or node-positive).

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient groups that responded to CADTH’s call for patient input and from the clinical experts consulted by CADTH for the purpose of this review.

Patient Input

• Two patient groups provided input, the Canadian Breast Cancer Network (CBCN) and Rethink Breast Cancer. Information was gathered through phone interviews (11 interviewees) and 2 surveys, 1 distributed by CBCN (52 respondents, all Canadian) and 1 by Rethink Breast Cancer (62 respondents, 60% Canadian).

• Patients described the emotional distress associated with being diagnosed with a type of breast cancer that is known to have a poor prognosis in the absence of HER2-directed therapy. Patients also noted the adverse effects associated with the disease and the treatments (cardiotoxicity, fever, cough, muscle pain, fatigue, diarrhea, and nausea) and noted that fatigue, pain, and nausea are most likely to impact their daily lives. Patients also noted the financial burden associated with lost income and treatment costs, with 17% of respondents in 1 survey reporting a very large financial impact and 38% reporting some financial impact.

• The most important outcomes for patients were the elimination of cancer cells, prevention of recurrence, and preventing metastases. Maintaining quality of life was also rated by the majority of patients as very important or important, as was managing adverse effects. Patients were clear that they were very willing to tolerate new adverse effects from drugs to extend life expectancy.

Clinician Input

Input From Clinical Experts Consulted by CADTH

• Per indication, pertuzumab in the neoadjuvant setting would be used in combination with trastuzumab and chemotherapy. The shift in the treatment paradigm would simply be the addition of pertuzumab to the standard therapies already being used.

• According to the clinical experts, the patients most likely to respond to the addition of pertuzumab would be those who have HER2-positive (or HER2-overexpressing) breast cancer. According to the clinical experts, all patients who are HER2-positive and are candidates for neoadjuvant therapy would be eligible for the addition of pertuzumab to their regimen, and those who are not candidates for either chemotherapy (due to being too ill) or neoadjuvant therapy (those with small stage I cancer) would not be eligible for pertuzumab. It was noted that it is very rare for a patient to be too ill to receive chemotherapy.

• The clinical experts noted that, ultimately, response in the neoadjuvant setting is determined at the time of surgery, when assessment of pathologic complete response (pCR) is performed. Prior to surgery, patients would most likely be assessed every 2 to 3 weeks at the time they come in to receive their chemotherapy, typically by a physical exam, although sometimes this may be supplemented by imaging of the breast (ultrasound or MRI). If, during therapy, the tumour is growing or not responding, the chemotherapy protocol may be modified or the patient may be sent for surgery earlier than planned. A clinically meaningful response is a shrinkage of the tumour to facilitate surgical removal.

• One of the clinical experts consulted by CADTH said they believe that increasing pCR rates would result in a reduced risk of relapse in this population.

• With respect to deciding when to discontinue treatment, the clinical experts noted this may occur if the tumour is growing, in which case surgery may be performed earlier than planned or, in some cases, other chemotherapy protocols may be instituted. Patients with clear disease progression after receiving 1 to 2 cycles of optimized taxane-based chemotherapy should be considered for discontinuation.

• One clinical expert noted that the addition of pertuzumab to the current treatment paradigm is important, given that this is a curable disease that often occurs in younger patients. The other clinical expert noted the importance of increased rates of tumour downstaging and pCR in reducing longer-term treatment-related morbidity.

Clinician Group Input

• Two clinician groups provided input, the BC Cancer Breast Tumour Group (BCC-BTG) and the Ontario Health (Cancer Care Ontario) Breast Cancer Drug Advisory Committee (OH-CCO BCDAC).

• One clinician group noted that the greatest need for pertuzumab is in patients with inflammatory breast cancer and inoperable stage IIIC breast cancers to downstage to get to primary surgery.

• The groups did not specifically refer to their experiences with pertuzumab; however, 1 clinician group noted that combining pertuzumab with trastuzumab is the international standard of care in stage II to III HER2-positive breast cancer.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. The Provincial Advisory Group (PAG) noted that in most provinces, the current standard of care for the neoadjuvant treatment of HER2-positive breast cancer is trastuzumab plus chemotherapy. Pertuzumab, being an IV drug, would be administered in an outpatient chemotherapy centre for appropriate administration and monitoring of infusion-related reactions. PAG highlighted several enablers of the implementation of pertuzumab in the neoadjuvant setting, including that the dose and frequency of pertuzumab in the neoadjuvant setting are the same as in the metastatic setting, that it is used as an add-on drug to existing treatment, and that drug wastage is not a concern since pertuzumab vials contain the amount of the fixed dose. PAG also identified barriers to implementation that include the high cost of pertuzumab and the additional preparation time and chair time needed for the infusion. Pertuzumab is administered for 4 to 6 cycles before surgery and PAG noted that given the high cost of pertuzumab, there is a significant difference in cost between 4 cycles and 6 cycles.

Clinical experts were consulted by CADTH for questions related to the implementation of pertuzumab in current provincial drug plans. Overall, most implementation questions related to the dosing schedule and administration, the eligible patient population, pCR as an end point, and re-treatment with pertuzumab in subsequent lines of treatment.

Clinical Evidence

Pivotal Studies and Protocol Selected Studies

Description of Studies

• Four trials, all identified as pivotal by the sponsor, were included in the CADTH review. NEOSPHERE (N = 417, randomized 1:1:1:1 across 4 groups) was an open-label randomized controlled trial (RCT) that had an arm that contained pertuzumab and trastuzumab plus docetaxel and an arm that contained trastuzumab plus docetaxel. PEONY (N = 329, randomized 2:1 across 2 treatment arms) was a double-blind RCT that randomized patients to either pertuzumab and trastuzumab plus docetaxel or trastuzumab plus docetaxel. TRYPHAENA (N = 225, randomized 1:1:1 across 3 treatment arms) and BERENICE (N = 400, distributed 1:1 across 2 cohorts, non-RCT) were designed to compare different background regimens of chemotherapy combined with pertuzumab and trastuzumab. The primary focus of this review was the NEOSPHERE and PEONY trials, with TRYPHAENA and BERENICE providing supportive evidence, where available. All trials included patients with early breast cancer that was HER2-positive.

• All trials featured a neoadjuvant treatment phase followed by surgery and then an adjuvant treatment phase.

  • In NEOSPHERE and PEONY, the neoadjuvant phase lasted 4 cycles and consisted of the treatments described previously. In the adjuvant phase of NEOSPHERE, treatment arms received 3 cycles of 5-fluorouracil and epirubicin plus cyclophosphamide (FEC) and trastuzumab for up to 1 year. The adjuvant phase of PEONY included 3 cycles of FEC followed by pertuzumab and trastuzumab for cycles 8 to 17 in the arm that received pertuzumab and trastuzumab plus docetaxel in the neoadjuvant phase, and placebo plus trastuzumab for cycles 8 to 17 in the arm that received trastuzumab plus docetaxel in the neoadjuvant phase.

  • In the neoadjuvant phase of TRYPHAENA, patients in arm A received pertuzumab and trastuzumab plus FEC for 3 cycles followed by pertuzumab and trastuzumab plus docetaxel for 3 cycles, patients in arm B received FEC for 3 cycles then pertuzumab and trastuzumab plus docetaxel for 3 cycles, and patients in arm C received pertuzumab plus docetaxel and carboplatin plus trastuzumab for 6 cycles. In the adjuvant phase, patients received trastuzumab from cycle 7 onward, up to 1 year.

  • In BERENICE, patients in arm A received doxorubicin plus cyclophosphamide for cycles 1 to 4, pertuzumab and trastuzumab plus paclitaxel for cycles 5 to 8, and patients in arm B received FEC for cycles 1 to 4, followed by pertuzumab and trastuzumab plus docetaxel for cycles 5 to 8. For the adjuvant phase, patients in both arms received pertuzumab and trastuzumab.

• The primary outcome of NEOSPHERE was the pCR rate at the conclusion of the neoadjuvant treatment period, and the primary outcome of PEONY was the total pathologic complete response (tpCR) rate, also at the conclusion of the neoadjuvant treatment period. PEONY also reported on the pathologic complete response in the breast (bpCR) rate at the conclusion of the neoadjuvant period. Both trials were designed to report on various longer-term outcomes such as overall survival (OS), progression-free survival (PFS), event-free survival (EFS), and disease-free survival (DFS); however, these outcomes were assessed during or after the adjuvant treatment period. The primary objective of TRYPHAENA and BERENICE was to assess safety and tolerability. The primary safety outcomes in TRYPHAENA were incidence of symptomatic cardiac events and clinically significant left ventricular ejection fraction (LVEF) decline, and the primary safety outcomes in BERENICE were incidence of New York Heart Association (NYHA) class III and IV heart failure and incidence of LVEF decline.

• Patients in the included trials were about 50 years old at baseline and the majority (approximately 70% to 80%) were White, except for PEONY, where all patients were Asian. Most patients (nearly 90%) had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 and the rest had a status of 1. Approximately half (47% in NEOSPHERE, 51% in PEONY) of patients were either estrogen or progesterone receptor–positive, except for BERENICE, in which approximately 2-thirds of patients were estrogen or progesterone receptor–positive. In terms of baseline disease category, the majority of patients in NEOSPHERE and TRYPHAENA were stage T2N0M0 (NEOSPHERE, |;|;|; TRYPHAENA, 31%) or stage T2N1M0 (NEOSPHERE, |;|; TRYPHAENA, 33%). In PEONY, most patients were stage T2 (67%), followed by T3 (22%), and had lymph node–positive disease (76%). In BERENICE, most patients were stage T2 (67%) followed by T3 (20%); 47% were N1, 8% were N2, and 2% were N3; 100% were M0.

Efficacy Results

The median overall time on study in NEOSPHERE was |||||||||||||||||||||||||||||||||||||||||| in the pertuzumab and trastuzumab plus docetaxel arm and |||||||||||||||||||||||||||||||||||||||||||| in the trastuzumab plus docetaxel arm. In PEONY, the median time on study was |||||||||||||||||||||||||||||||||||||||||||| in the pertuzumab and trastuzumab plus chemotherapy arm and |||||||||||||||||||||||||||||||||||||| in the trastuzumab plus chemotherapy arm. In TRYPHAENA, the median time on study ranged from ||||||||||||||||||||||||||||||||||||||||||||||||||||||||| in the 3 treatment arms. In BERENICE, the median time on study was |||||||||||||||||||||||||||||||||||||||||||||| in cohort A and |||||||||||||||||||||||||||||||||||||||||||||| in cohort B.

Assessment of longer-term outcomes, such as OS, EFS, DFS, and PFS, included treatment regimens received in both the neoadjuvant and adjuvant phases of treatment. OS was not assessed in NEOSPHERE; OS data are not yet mature from PEONY, according to the sponsor, and there are no comparative OS data available from TRYPHAENA or BERENICE. Data for invasive DFS or EFS were not available from the included trials, either because it was not assessed or because the data were reported as not yet mature by the sponsor. With respect to PFS, in NEOSPHERE, progression events occurred in 15.9% of patients in the pertuzumab and trastuzumab plus docetaxel arm and in 17.8% of patients in the trastuzumab plus docetaxel arm, for a hazard ratio (HR) of 0.69 (95% confidence interval [CI], 0.34 to 1.40). These results were consistent with the PFS data reported in TRYPHAENA, where the PFS event rates were 13.7% in arm A, 14.7% in arm B, and 18.2% in arm C. Data on PFS were not yet mature in PEONY, according to the sponsor, and PFS was not assessed in BERENICE. DFS events occurred in 14.9% of patients in the pertuzumab and trastuzumab plus docetaxel arm in NEOSPHERE, and 17.5% of patients in the trastuzumab plus docetaxel arm, and these results were consistent with those reported in TRYPHAENA, where the PFS events were 14.5% in arm A, 11.9% in arm B, and 15.3% in arm C. The DFS data were not yet mature in PEONY, according to the sponsor, and DFS was not assessed in BERENICE. None of the trials were powered to assess between-group differences in these longer-term outcomes.

In NEOSPHERE, a pCR was achieved by 45.8% of patients in the pertuzumab and trastuzumab plus docetaxel arm, and 29.0% of patients in the trastuzumab plus docetaxel arm, for a difference in response rates between groups of 16.8% (95% CI, 3.5 to 30.1; P = 0.0094). In PEONY, the tpCR rate assessed by an independent review committee (IRC) was 39.3% in the pertuzumab and trastuzumab plus docetaxel arm and 21.8% in the trastuzumab plus docetaxel arm, for a difference in response rates of 17.45% (95% CI, 6.89 to 28.01; P = 0.0014). The difference in pCR rates between the 2 trials may reflect the different definitions of pCR used, as NEOSPHERE used only breast tissue to assess pCR, while PEONY used breast and nodes. Additionally, PEONY reported the bpCR rate as a secondary outcome, and the IRC-assessed bpCR rate was consistent with that of the tpCR rate (42.0% versus 23.6%), for a between-group difference of 18.37% (95% CI, 7.60 to 29.15). The pCR rates ranged from 57.3% to 66.2% across the 3 arms in TRYPHAENA and were 60.7% and 61.8% in the 2 cohorts in BERENICE.

In NEOSPHERE, a complete response (CR) was observed in 18.9% of patients in the pertuzumab and trastuzumab plus docetaxel arm and 18.3% of patients in the trastuzumab plus docetaxel arm, and a partial response (PR) was observed in 49.1% of patients and 49.3% of patients, respectively, when assessed by X-ray or mammography. When assessed by clinical exam, a CR was observed in 25.0% versus 21.6% of patients, respectively, and a PR was observed in 63.0% versus 59.8% of patients, respectively. In PEONY, clinical response was assessed as a secondary outcome, and an objective response, defined as obtaining either a CR or PR) during cycles 1 to 4 occurred in 88.6% of patients in the pertuzumab and trastuzumab plus docetaxel arm and 78.2% of patients in the trastuzumab plus docetaxel arm, for a difference in objective response rates between groups of 10.4% (95% CI, 1.12 to 19.69). A CR was observed in 11.0% versus 10.0% of patients, and a PR was observed in 77.6% versus 68.2% of patients, respectively.

Duration of response, health-related quality of life, and symptoms were not assessed in the included studies. Among patients in whom mastectomy was initially planned, breast-conserving surgery was achieved in 23.2% of patients in the pertuzumab and trastuzumab plus docetaxel arm and in 22.6% of patients in the trastuzumab plus docetaxel arm in NEOSPHERE. This outcome was not assessed in PEONY. In TRYPHAENA, the percentage of patients undergoing breast-conserving surgery was consistent with that of NEOSPHERE, ranging between 16.7% and 27.0% of patients across arms in the subgroup of patients in whom mastectomy was initially planned. In the BERENICE study, 44.4% and 42.9% of patients with T2 or T3 tumours in the 2 cohorts had breast-conserving surgery.

Harms Results

The percentage of patients experiencing adverse events was similar between pertuzumab and trastuzumab plus docetaxel and trastuzumab plus docetaxel, occurring in 96% to 98% of patients across treatment arms in NEOSPHERE and PEONY. The most common adverse events in the trials for pertuzumab and trastuzumab plus docetaxel versus trastuzumab plus docetaxel, were alopecia (63.6% versus 65.4% in NEOSPHERE; 49.1% in PEONY for both treatments), neutropenia (50.5% versus 62.6% in NEOSPHERE; 48.2% versus 44.5% in PEONY), and diarrhea (45.8% versus 33.6% in NEOSPHERE; 38.5% versus 16.4% in PEONY). The most common grade 3 or greater adverse event was neutropenia (44.9% versus 57.0% in NEOSPHERE; 38.1% versus 32.7% in PEONY). Similar results were seen in TRYPHAENA and BERENICE, where approximately 99% of patients experienced an adverse event at some time during the study, and neutropenia was the most common grade 3 or greater adverse event.

Serious adverse events occurred in 10.3% of patients in the pertuzumab and trastuzumab plus docetaxel group and 16.8% of patients in the trastuzumab plus docetaxel group in NEOSPHERE, and in 10.1% versus 8.2% of patients, respectively, in PEONY. Febrile neutropenia was the most common serious adverse event in both NEOSPHERE groups, occurring in 5.6% of patients treated with pertuzumab and trastuzumab plus docetaxel and 6.5% of patients treated with trastuzumab plus docetaxel. In PEONY, febrile neutropenia occurred in 1.8% of patients treated with pertuzumab and trastuzumab plus docetaxel and in no patients treated with trastuzumab plus docetaxel. In TRYPHAENA, 28% of patients experienced a serious adverse event across the treatment arms, and, in BERENICE, 24% of patients experienced a serious adverse event. Febrile neutropenia was the most common serious adverse event in both studies, occurring in about 10% of patients.

Few patients across the trials stopped treatment due to an adverse event: 0.9% of patients in the pertuzumab and trastuzumab plus docetaxel arm versus no patients in the trastuzumab plus docetaxel arm in NEOSPHERE, and 0.5% of patients in the pertuzumab and trastuzumab plus docetaxel arm and no patients in the trastuzumab plus docetaxel arm in PEONY. The number of patients withdrawing due to an adverse event was 7% across arms in TRYPHAENA and 3.5% across cohorts in BERENICE.

One patient died in each of the pertuzumab and trastuzumab plus docetaxel and trastuzumab plus docetaxel arms in NEOSPHERE; both deaths were considered to be due to complications of breast cancer. One patient died in the pertuzumab and trastuzumab plus docetaxel arm in PEONY, due to a suicide, and there were no deaths in the trastuzumab plus docetaxel arm.

Among notable harms, cardiac dysfunction occurred in 2.8% of patients in the pertuzumab and trastuzumab plus docetaxel arm, and 0.9% of patients in the trastuzumab plus docetaxel arm in NEOSPHERE; no patients in PEONY had an LVEF decline to less than 40% or a primary or secondary cardiac event. Events of drug hypersensitivity or anaphylaxis occurred in NEOSPHERE in |||| of patients in the pertuzumab and trastuzumab plus docetaxel arm and in |||| of patients in the trastuzumab plus docetaxel arm, and in |||| versus |||| of patients in PEONY, respectively.

Critical Appraisal

• NEOSPHERE was an open-label study, and no centralized blinded review of pathology was conducted when assessing pCR responses. Although pathology findings are unlikely to be biased by knowledge of treatment assignment, a blinded review of pathology is recommended by regulatory bodies. With respect to the primary outcome, pCR was defined differently between NEOSPHERE and PEONY. In NEOSPHERE, the primary outcome of pCR included only breast tissue (commonly described as bpCR) while, in PEONY, assessment of pCR for the primary outcome included breast and nodes, referred to as tpCR; the latter is the method recommended by the FDA. TRYPHAENA and BERENICE provide only limited supportive information regarding efficacy; as neither trial had a comparator, neither was designed to test hypotheses with respect to efficacy outcomes, and BERENICE was not a randomized trial. The alpha in NEOSPHERE was set at 0.2 instead of the traditional 0.05, and this might have increased the risk of finding a statistically significant difference in pCR rates between arms where none existed.

• OS was not assessed as an efficacy outcome in NEOSPHERE and the OS data for PEONY were not yet mature, according to the sponsor; therefore, there is no information to determine whether the addition of pertuzumab to neoadjuvant treatment with trastuzumab and docetaxel improves this important outcome. Health-related quality of life and symptoms were also not assessed and, although these outcomes may not be as important in early breast cancer and in the neoadjuvant setting, assessment of health-related quality of life would help in assessing what impact the addition of pertuzumab has on adverse effects.

Indirect Treatment Comparisons

No indirect treatment comparisons (ITCs) were submitted by the sponsor, and none were found in the literature that would inform this review.

Other Relevant Evidence

There were no other studies that were found that would be relevant to this review.

Conclusions

Four trials that were identified as pivotal by the sponsor were included in this review. Rates of pCR were improved when pertuzumab was added to standard neoadjuvant regimens with trastuzumab plus chemotherapy in the 2 trials that featured a comparator, NEOSPHERE and PEONY. It is unclear whether these improvements in pCR translate into improved OS, as this outcome was not studied in NEOSPHERE, and the survival data from PEONY were not available at the time of review. The combination of pertuzumab with trastuzumab plus chemotherapy did not appear to improve invasive DFS, PFS, EFS, or DFS, either because these outcomes were not studied, the data were not yet available, or there was a lack of statistical significance when they were assessed. Health-related quality of life and symptoms were not assessed in any of the included studies. Based on the included studies, the addition of pertuzumab to trastuzumab plus chemotherapy did not appear to introduce significant safety or tolerability issues.

Introduction

Disease Background

It is estimated that 1 in 8 Canadian women will be diagnosed with breast cancer during their lifetime, and that 1 in 33 will die of the disease.⁸ Due to advances in and widespread use of screening programs, breast cancer is typically diagnosed at an early stage, and approximately 95% of cases are diagnosed at stages I to III. Generally speaking, primary operable breast cancers are categorized as stage I or II, and locally advanced breast cancers are categorized as stage III. The overall 5-year survival rate from breast cancer is 88%, though this will vary based on the stage and subtype of breast cancer.⁸ Prior to the use of anti-HER2 therapy, patients who had HER2-positive breast cancer tended to have a poorer prognosis than patients without HER2 overexpression.

Patients who are undergoing neoadjuvant therapy can be subclassified into those with locally advanced breast cancers that are not operable, those with locally advanced cancers that are operable, and those with primary operable breast cancers. In locally advanced cancers, 1 purpose of neoadjuvant therapy is to convert the tumour from an inoperable to an operable state. In all cancers, the purpose of neoadjuvant therapy is to downstage the tumour to potentially avoid mastectomy in favour of breast-conserving surgery, to assess response to systemic therapy, to potentially escalate or de-escalate subsequent adjuvant therapy based on the response to neoadjuvant therapy, and to initiate systemic therapy early to try to limit systemic spread.

Standards of Therapy

For the majority of patients with HER2-positive breast cancer who are either stage II or III, as well as some patients who are stage I, neoadjuvant systemic treatment is the current standard of care in Canada. According to the clinical experts consulted by CADTH, this represents current accepted international guidelines and is increasingly the standard therapy used in Canada, with some exceptions. By treating with chemotherapy early, the risk of systemic recurrence is also decreased. There are a small number of patients who may have surgery first and then have adjuvant therapy, particularly if they have small cancers, and this would be the norm for stage I disease.

According to the clinical experts, standard neoadjuvant chemotherapy is doxorubicin, cyclophosphamide, and paclitaxel plus trastuzumab or docetaxel and carboplatin plus trastuzumab. A small number of patients might get 5-fluoruracil, epirubicin, and docetaxel plus trastuzumab. These are all standard protocols used internationally, although pertuzumab is often given with trastuzumab in other countries, according to the clinical experts, and there are some patients who gain access to pertuzumab through private insurance. The ASCO guidelines for the neoadjuvant management of HER2-positive, node-positive, or high-risk node-negative breast cancer recommend an anthracycline and taxane or non-anthracycline-based regimen in combination with trastuzumab, and they also suggest that pertuzumab can be used with trastuzumab in the neoadjuvant setting.¹ The European Society for Medical Oncology recommends a dual blockade with pertuzumab and trastuzumab in high-risk patients (node-positive or estrogen receptor–negative), starting before or after surgery and for neoadjuvant therapy; the German Gynecological Oncology Group recommends adding pertuzumab when treating node-positive disease.^2,3 According to 1 clinical expert, patients who achieve a pCR after neoadjuvant treatment tend to have a better prognosis and are recommended to receive adjuvant trastuzumab for up to 14 cycles, while patients with residual invasive disease are recommended to receive trastuzumab emtansine (T-DM1) in the adjuvant setting. Patients with hormone receptor–positive and HER2-positive breast cancer are also recommended for adjuvant endocrine therapy. The goal of treatment in the neoadjuvant setting is curative, as this means that otherwise healthy patients can go on to live a normal life span and maintain employment. This also reduces caregiver burden and is generally beneficial to society.

Drug

Pertuzumab is a monoclonal antibody that targets the extracellular dimerization domain of the HER2 receptor protein and thus blocks ligand-dependent heterodimerization of HER2 with other members of the human epidermal growth factor receptor family. Pertuzumab therefore inhibits ligand-initiated intracellular signalling through 2 pathways, the mitogen-activated protein kinase pathway and the phosphoinositide-3 kinase pathway, causing cell growth arrest and apoptosis. Additionally, pertuzumab mediates antibody-dependent cell-mediated cytotoxicity.

On February 25, 2021, Health Canada issued a Notice of Compliance for the use of pertuzumab in the neoadjuvant setting.⁹ Pertuzumab is indicated, in combination with trastuzumab and chemotherapy, for the neoadjuvant treatment of patients with HER2-positive locally advanced, inflammatory, or early-stage breast cancer (either > 2 cm in diameter or node-positive).⁹ Pertuzumab was previously reviewed by CADTH for off-label use under the same indication in 2015 and the CADTH pan-Canadian Oncology Drug Review (pCODR) Expert Review Committee (pERC) issued a negative recommendation for reimbursement. Pertuzumab in combination with trastuzumab is also indicated as adjuvant treatment in HER2-positive, early breast cancer with lymph node–positive and/or hormone receptor–negative disease, and with docetaxel for patients with HER2-positive metastatic breast cancer who have not received prior anti-HER2 therapy or chemotherapy for metastatic disease. CADTH reviewed pertuzumab for both indications and pertuzumab received a negative recommendation for reimbursement for the adjuvant indication and a recommendation for reimbursement for the metastatic breast cancer indication.

Pertuzumab is administered by IV infusion. In the neoadjuvant setting, after an initial 840 mg loading dose, the maintenance dose is 420 mg every 3 weeks for 3 to 6 cycles. Pertuzumab is given in combination with trastuzumab, as part of 1 of the following regimens in early-stage breast cancer:

• Four pre-operative cycles of pertuzumab in combination with trastuzumab and docetaxel (75 mg/m² with the option to escalate to 100 mg/m² at physician discretion if the initial dose is well tolerated) every 3 weeks, followed by 3 post-operative cycles of FEC (5-fluorouracil: 600 mg/m²; epirubicin: 90 mg/m²; cyclophosphamide: 600 mg/m²) every 3 weeks, as given in the NEOSPHERE trial.

• Three or 4 pre-operative cycles of FEC (5-fluorouracil 500 mg/m²; epirubicin: 100 mg/m²; cyclophosphamide: 600 mg/m²) alone, every 3 weeks, followed by 3 or 4 pre-operative cycles of pertuzumab in combination with docetaxel (75 mg/m² with the option to escalate to 100 mg/m² at physician discretion if the initial dose is well tolerated) and trastuzumab every 3 weeks, as given in the TRYPHAENA and BERENICE trials, respectively.

• Six pre-operative cycles of pertuzumab in combination with docetaxel and carboplatin plus trastuzumab: 75 mg/m² docetaxel (escalation of docetaxel above 75 mg/m² is not recommended), carboplatin (area under the plasma concentration versus time curve [AUC] 6) and trastuzumab every 3 weeks, as given in the TRYPHAENA trial.

• Four pre-operative cycles of dose-dense doxorubicin and cyclophosphamide (doxorubicin: 60 mg/m²; cyclophosphamide: 600 mg/m²) alone every 2 weeks, followed by 4 pre-operative cycles of pertuzumab in combination with trastuzumab every 3 weeks, and paclitaxel (80 mg/m²) every week for 12 weeks, as given in the BERENICE trial.

The sponsor’s reimbursement request is the same as the Health Canada indication, in combination with trastuzumab and chemotherapy, for the neoadjuvant treatment of patients with HER2-positive locally advanced, inflammatory, or early-stage breast cancer (either > 2 cm in diameter or node-positive). The sponsor also added that patients should receive neoadjuvant treatment with pertuzumab in combination with trastuzumab and chemotherapy for 3 to 6 cycles, depending on the regimen chosen. Patients who start pertuzumab in combination with trastuzumab and chemotherapy in the neoadjuvant setting and who do not have residual disease following surgery should continue to receive adjuvant trastuzumab to complete 1 year of HER2-directed therapy.

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups.

About the Patient Groups and Information Gathered

Two patient groups provided input for this submission, the CBCN and Rethink Breast Cancer. The CBCN is a patient-directed, national health charity committed to ensuring the best quality care for all Canadians affected by breast cancer through the promotion of information, education, and advocacy activities. Rethink Breast Cancer’s mission is to empower young people who are affected by or concerned about breast cancer through education, support, and advocacy. It represents the voice of young women to ensure their needs and values are considered in all aspects of breast cancer treatment and care.

For the preparation of this submission, the CBCN noted that Roche Canada, the sponsor, connected them with patients who had experience with pertuzumab. Rethink Breast Cancer requested information from Roche and its scientific advisory committee on the characteristics of the drug and its benefits and they contracted a freelance health technology assessment writer to help prepare the submission and to develop a survey and analyze the responses received.

Information for the submission was obtained from CBCN’s 2017 Lived Experience Breast Cancer Patient Survey, from which CBCN summarized the responses from 52 Canadians with early-stage, HER2-positive breast cancer (stage I to III). All respondents identified as female and from various regions: Alberta (4% of respondents), Ontario (23%), Saskatchewan (12%), Quebec (8%), British Columbia (8%), Manitoba (4%), and the Atlantic provinces (17%); 25% did not specify their location. The age at which respondents were diagnosed with breast cancer was categorized as follows: 30 to 39 years of age (19%), 40 to 49 years (40%), 50 to 59 years (29%), and 60 to 69 years (6%); the remaining 6% of respondents did not disclose their age. In addition, 4 patients who had direct experience with pertuzumab participated in a phone interview. The CBCN also conducted a literature review to identify issues and experiences shared among women with breast cancer.

Rethink Breast Cancer conducted an online survey from March 23 to April 19, 2021 that was circulated through the organization’s mailing list, partner organizations, and social media. Of the 62 respondents, 37 (60%) were from Canada (Alberta, British Columbia, New Brunswick, Ontario, Quebec, and Saskatchewan), 22 were from the US (35%), and 3 were from other countries or did not disclose their location. Seven respondents agreed to participate in telephone interviews with staff. All 62 respondents were diagnosed with HER2-positive breast cancer in stage I, II, or III. Among the respondents, 41 (66%) had treatment experience with pertuzumab including 35 (56%) who match the full indication for this review.

Disease Experience

A diagnosis of early-stage, HER2-positive breast cancer has a significant impact on the day-to-day life of the patient, particularly as this cancer subtype is traditionally associated with more aggressive cancer with a poor prognosis in the absence of HER2-directed therapy. It is also associated with a higher risk of recurrence or metastases. Both the diagnosis as well as the treatments that are used impact the emotional and physical well-being of a patient. Some of the adverse effects of HER2-positive breast cancer and the therapies used to manage this disease include cardiac toxicity, fever, cough, muscle pain, fatigue, diarrhea, and nausea. Many of these symptoms have the ability to impact daily life, primarily, fatigue, pain, and nausea.

The financial burden associated with living with breast cancer extends far beyond any loss of income during a temporary or permanent absence from employment. In addition to the loss of income during illness, breast cancer patients can incur substantial costs associated with treatment and disease management. In the CBCN survey, 17% of respondents had a very large financial impact, and 38% had some financial impact from their diagnosis. One patient stated:

“Very hard on my family...had to return to work still not feeling strong enough...very hard ...when you’re sick you don’t need this stress on top of everything else.”

Experience With Treatment

In the Rethink Breast Cancer survey, 56 of 62 patients had received trastuzumab and 12 had received T-DM1. The most commonly reported chemotherapy drugs received included carboplatin, docetaxel, paclitaxel, doxorubicin (Adriamycin), and cyclophosphamide. Fatigue was the most commonly reported adverse effect of these treatments (80%), followed by diarrhea (64%), nausea (44%), and insomnia (39%). Fatigue was most frequently cited as the hardest-to-tolerate adverse effect of these treatments. Diarrhea, nausea, neuropathy, and taste changes were also cited by at least 10% of respondents. Most respondents (73%) did not report any problems accessing treatment.

In the CBCN 2017 survey, most of the HER2-positive, early-stage breast cancer patients had been or were currently being treated with a combination of surgery, radiation, chemotherapy, and the HER2-directed therapy, trastuzumab. Most patients had undergone surgery (44 out of the 52 respondents), radiation therapy (33 respondents), and chemotherapy (35 respondents) as part of their overall breast cancer treatment.

In the CBCN survey, respondents reported barriers to treatment including lack of access to private insurance coverage and support medications. While 40 of the 52 patients surveyed reported having private insurance coverage, several (6 respondents) also reported challenges accessing medications not publicly reimbursed. Respondents stated they had to use private insurance (17 respondents) or pay out of pocket (11 respondents) to access medications they had been prescribed. One patient said:

“When I found out how expensive my treatment is, I was absolutely flabbergasted. I had to leave the pharmacy empty handed because a one month supply was over $1400 and I didn't have the money or amount available on credit. I was told about a form to fill out if my income was below a certain amount but I didn't qualify. So, we paid for it out of pocket and did get some reimbursed from work insurance plan.”

Rethink Breast Cancer reported the experience of 35 patients who received neoadjuvant pertuzumab in combination with trastuzumab and chemotherapy for locally advanced, inflammatory, or early-stage breast cancer. Twenty-one patients achieved a pCR within 1 year of their surgery. Of these 21 patients, 1 patient had since had a recurrence; the other 20 remained free of cancer. Most of the respondents felt that pertuzumab had improved their quality of life in every listed area, including activities such as the ability to work, sleep, drive, care for children, and perform household chores. Said 1 patient respondent:

“If you have a scan or you have another ultrasound and you see the reduction in the tumor so quickly, it had an impact on anxiety, on positivity, on quality of life.”

Diarrhea and fatigue were the most commonly reported adverse effects of pertuzumab (84% and 81%, respectively), followed by alopecia (38%), neutropenia (25%), and nausea (22%). However, respondents overwhelmingly described these adverse effects as tolerable. Moreover, respondents explicitly said they were willing to tolerate the adverse effects of pertuzumab for its medical benefits. A few respondents noted they had difficulties with the loading dose of pertuzumab but found the subsequent doses to be tolerable. Others were not able to distinguish the adverse effects of pertuzumab from the other drugs they were receiving concurrently.

The CBCN reported the experience of 4 patients who had received pertuzumab in addition to trastuzumab, chemotherapy, and surgery. It is not clear if pertuzumab was administered in the adjuvant or neoadjuvant setting. All patients were hopeful that this addition to their treatment regimen would help improve their survival and reduce the risk of recurrence. The patients had difficulty determining if the adverse effects they experienced were related to pertuzumab or other therapies, but all rated their quality of life as mid-range to high on a 10-point scale. All patients expressed concerns about the lack of access to new treatments and the potential financial burden of paying out of pocket. One patient respondent said:

“Having just that additional little bit of peace of mind that I’m doing everything that I can. I’m pretty young. I’ve got a young family. I’ve got a three-year-old. So I need to be able to say that I’ve done everything that I possibly can to beat it. So having that peace of mind that I’m getting the same care that others are getting elsewhere in the world, so I don’t have to look at going somewhere else and all the costs and finances involved. If there was a breakthrough treatment that was working in the U.S. but not available in Canada, having to somehow try to finance going there to go get that treatment.”

Improved Outcomes

Among respondents to the Rethink Breast Cancer survey, eliminating cancer, preventing recurrence, and preventing metastases were overwhelmingly rated as the most important outcomes for their breast cancer treatment with greater than 96% of respondents rating these goals as “very important.” Maintaining quality of life was also rated as “very important” or “important” for 63% and 18% of respondents, respectively, while managing adverse effects was “very important” for 46% and “important” for 23% of respondents. Respondents also indicated they were highly willing to tolerate new adverse effects from new drugs to extend life expectancy:

“…I just finished chemo and feel like there’s no way I’d ever do it again. Period. But at the same time how do you not do *whatever* it takes to stay alive? I’d undergo near death side effects in order to avoid death…”

“I will tolerate whatever symptoms I have to so that I can survive and take care of my children.”

Respondents of CBNC’s 2017 survey indicated that the following key factors influenced their decision-making around treatments:

• Effectiveness of the treatment: How well the treatment stabilized their disease and delayed the progression of their cancer. Effectiveness was ranked as “very important” by 40 of 52 respondents (77%), with 73% of respondents stating effectiveness was the single most important factor in their treatment decisions. The majority of respondents indicated that reducing the risk of recurrence was “important” or “very important.”

“I was willing to do whatever was best to rid myself of the cancer. I could deal with the side effects and disruption in my life for the long term good.”

“I just wanted to make sure they did everything to get rid of the cancer.”

• Reducing the risk of recurrence without sacrificing quality of life: Being able to maintain productive, active lives with minimal disruption to daily routines and avoiding relapse of their cancer. Approximately 2-thirds of respondents indicated maintaining quality of life and mobility was “important” or “very important,” while maintaining productivity was an important concern for 35% of respondents. Six respondents raised concerns regarding their ability to provide childcare, and this was an important factor in their treatment decisions.

“I am a mother to 3 children. I wanted to be aggressive in order to increase my chances of survival.”

“I only wanted to reduce my risk of recurrence as much as possible. Everything else was secondary.”

“My quality of life during and after treatment was the biggest issue for me.”

“If I had to do it over again, I would opt out of chemo.”

• Adverse effect management: Minimizing risk while stabilizing their disease. Minimal adverse effects of treatment was an “important” or “very important” concern for half the respondents.

“Les effets de la chimio qui nous sont inconnus et qui fait peur car on en attends parler tellement négativement.” [Translation: The effects of chemo that are unknown and scary because we hear people talk about it so negatively.]

• Cost and accessibility of treatments: Affordability and ease of accessing treatments. Having access to new treatments was important to patients.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise regarding the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 2 clinical specialists with expertise in the diagnosis and management of breast cancer.

Unmet Needs

One of the clinical experts consulted by CADTH said they believe that, with respect to pCR rates, there is room for improvement with current treatments, as lower rates of pCR correspond to higher relapse rates. The other clinical expert noted that increasing the rate of tumour downstaging and pCR should improve cosmesis and decrease survivorship morbidity through less invasive local-regional management approaches and post-operative T-DM1, a drug associated with a higher risk of toxicity than trastuzumab. These toxicities include peripheral neuropathy, cytopenia, abnormal liver enzymes, and, rarely, pneumonitis and hepatic nodular regenerative hyperplasia.

Place in Therapy

Per indication, pertuzumab would be used in combination with trastuzumab and taxane-based chemotherapy in the neoadjuvant setting. The clinical expert consulted by CADTH noted that, if pertuzumab were used in the neoadjuvant setting, this would likely reduce the number of relapses and thus the number of patients who would need to be treated for advanced disease. The shift in the treatment paradigm would simply be the addition of pertuzumab to the standard therapies already being used in the neoadjuvant setting.

Patient Population

The patients most likely to respond to the addition of pertuzumab would be those who are HER2-positive (or overexpressing HER2) with T2 or more advanced breast tumours and/or lymph node involvement, according to the clinical experts consulted by CADTH. One of the clinical experts noted that pCR occurs in patients with either estrogen receptor–positive or estrogen receptor–negative disease; however, the pCR rate is consistently higher in those who are estrogen receptor–negative. According to the clinical experts, all patients who are HER2-positive and considered candidates for neoadjuvant therapy would be eligible for the addition of pertuzumab to their neoadjuvant regimen. Patients who are not candidates for either chemotherapy (due to being too ill) or for neoadjuvant therapy (because they have a tiny stage I cancerous tumour) would not be eligible for neoadjuvant pertuzumab. It was also noted that it is very rare for a patient to be too ill to receive chemotherapy.

Assessing Response to Treatment

One clinical expert noted that, ultimately, response in the neoadjuvant setting is determined at the time of surgery, when assessment of pCR is performed. Prior to surgery, patients would most likely be assessed every 2 to 3 weeks depending on the regimen at the time they come in to receive their chemotherapy. Assessment is typically done by a physical exam, although sometimes this may be supplemented by imaging of the breast (ultrasound or MRI). A clinically meaningful response is a shrinkage of the tumour to facilitate surgical removal.

Discontinuing Treatment

With respect to deciding when to discontinue treatment, the clinical experts noted this may occur if the tumour is growing, in which case surgery may be performed earlier than planned or, in some cases, other chemotherapy protocols may be instituted. Patients with clear disease progression after receiving 1 to 2 cycles of optimized taxane-based chemotherapy should be considered for discontinuation. One of the clinical experts noted that hypersensitivity to pertuzumab may be a reason to discontinue therapy.

Prescribing Conditions

Pertuzumab would be administered wherever chemotherapy is given; in Canada, most chemotherapy is administered in cancer centres, hospitals with medical oncologists and chemotherapy units, hospitals with regional chemotherapy units or, occasionally, in hospitals with general practitioner oncologists who are being directed by medical oncologists at cancer centres. Chemotherapy in Canada is very protocol-oriented and is also funded on a provincial level, which requires that standard protocols be followed in specialized units. The other clinical expert noted that if there were no signs of infusion reactions, subsequent infusions could be administered in centres that are not accredited to give an advanced drug if it is appropriate to do so for the patient’s convenience.

Additional Considerations

One clinical expert noted that the addition of pertuzumab to the current treatment paradigm is important, given that this is a curable disease that often occurs in younger patients. The other clinical expert noted the importance of increased rates of tumour downstaging and pCR in reducing longer-term treatment-related morbidity.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups.

Input from 2 clinician groups was received on the reimbursement review of pertuzumab as neoadjuvant treatment: 1 from the BCC-BTG and the other from the OH-CCO BCDAC.

The BCC-BTG is a multidisciplinary group that operates within BC Cancer and sets the standards of care, treatment options, and pathways across the spectrum of breast cancer care in British Columbia. Information from a previous review for neoadjuvant pertuzumab in HER2-positive breast cancer was also used to inform this submission.

The OH-CCO BCDAC provides timely evidence-based clinical and health system guidance on drug-related issues in support of OH-CCO’s mandate, including the Provincial Drug Reimbursement Programs and the Systemic Treatment Program. Information for this submission was discussed jointly through email at a BCDAC meeting.

Unmet Needs

The BCC-BTG noted the intent of treatment is complete eradication of disease (both gross disease in breast and nodes and potential microscopic metastases) to improve cure rates. The group added that pertuzumab concurrent with taxane and trastuzumab is not publicly funded as a standard neoadjuvant regimen in eligible stage II to III HER2-positive breast cancer in Canada, though there are patient assistance programs. For example, Roche’s OnCare program provides assistance in insurance navigation and co-payments and coordination of infusion care. The program does not include compassionate or free pertuzumab; therefore, at present, there is significant inequity for access to pertuzumab in Canada.

With respect to the most important goals that an ideal treatment would address, the BCC-BTG noted that inducing a tumour response for enhanced breast cancer surgery (i.e., tumour downstaging for breast-conservation surgery) and pathological response (hopefully pCR) such that adjuvant treatment is not escalated to T-DM1) are important goals for treatment. The OH-CCO BCDAC noted that improving survival, preventing recurrence, acceptable toxicities, and improvement in health-related quality of life are the most important treatment goals for therapy.

The OH-CCO BCDAC noted that despite optimal systemic and local therapy, approximately 25% to 30% of HER2-positive patients still experience disease recurrence; thus, better and improved treatments are needed. They added that the patients who have the greatest unmet need for an intervention like pertuzumab are patients diagnosed at a higher stage of disease, since they have the greatest risk of disease recurrence

The BCC-BTG noted that the international standard of care for stage II to III HER2-positive breast cancer is neoadjuvant pertuzumab and trastuzumab concurrent with a taxane. The clinician group noted that the treatment gap in Canada is a significantly lower pCR rate (approximately 50% lower) without pertuzumab, which not only translates into a potentially lower long-term clinical outcome, but also exposes more patients to more toxic adjuvant treatments like T-DM1 rather than adjuvant trastuzumab alone. The clinicians from the BCC-BTG added that the greatest need for pertuzumab is in the inflammatory breast cancer population and for inoperable stage IIIC breast cancers to downstage the tumour for primary surgery.

Place in Therapy

The BCC-BTG noted that for HER2-positive stage II to III breast cancer, the current treatment pathway in Canada includes neoadjuvant chemotherapy (taxane ± anthracycline) concurrent with neoadjuvant trastuzumab (for the taxane component) for 6 to 8 cycles before surgery. The BCC-BTG added that the 2 main clinical reasons for this neoadjuvant approach is to downstage the primary tumour and axillary nodes to improve the surgical approach (lumpectomy rather than mastectomy and/or sentinel lymph node biopsy versus axillary lymph node dissection), and to use pathological response determination (pCR versus no pCR) to decide on the course of adjuvant treatment, where trastuzumab is given to patients who achieve a pCR and T-DM1 is given to patients who do not achieve a pCR.

The OH-CCO BCDAC noted that the current standard of care for T1c or greater (if N0) or any node-positive disease is neoadjuvant trastuzumab and chemotherapy. The group added that if there is residual disease post surgery, patients would be offered adjuvant T-DM1.

Both clinician groups agreed that pertuzumab would be added to neoadjuvant trastuzumab. The BCC-BTG noted that pertuzumab would fit into the current treatment pathways of neoadjuvant taxane and trastuzumab plus or minus anthracycline. Both clinician groups agreed it would not be appropriate for patients to try other treatment options before pertuzumab, as it is a neoadjuvant therapy with curative intent. The BCC-BTG clinicians noted that clinical and radiological imaging are not accurate in predicting pCR; thus, as long as the treatment is tolerated, the goal is to deliver chemotherapy and anti-HER2 therapy upfront before surgery.

Both clinician groups agreed that the use of pertuzumab in the neoadjuvant setting may impact the use of pertuzumab in the metastatic setting, as pertuzumab and trastuzumab are funded in the first-line metastatic setting. The BCC-BTG noted that it would be reasonable to consider pertuzumab, trastuzumab, and a taxane as first-line treatment (the current standard of care) for metastatic HER2-positive breast cancer if 12 months or longer had elapsed since the last dose of neoadjuvant pertuzumab.

Patient Population

The BCC-BTG noted that patients with stage II to III HER2-positive breast cancer who are medically fit to receive neoadjuvant chemotherapy and dual anti-HER2-directed therapies would be best suited to receive treatment with pertuzumab. The OH-CCO BCDAC noted that the sponsor’s funding request is for pertuzumab in combination with trastuzumab and chemotherapy for the neoadjuvant treatment of patients with HER2-positive locally advanced inflammatory or early-stage breast cancer (either 2 cm in diameter or node-positive). The OH-CCO BCDAC noted that when selecting patients who may be potential candidates for adjuvant treatment with T-DM1, patents with tumours 1 cm in diameter or node-positive disease were eligible in the KATHERINE trial. The group commented that in Ontario’s Evidence Building Program, approximately 100 patients per year are diagnosed with T1a to T1b N0 HER2-positive disease.

Both clinician groups agreed that testing for HER2 status is standard for all patients newly diagnosed with invasive breast cancer, as per the ASCO and College of American Pathologists (ASCO/CAP) guidelines. The BCC-BTG noted that most pathology laboratories participate in a quality-assurance program and patients are identified by the surgeon for referral to a medical oncologist for consideration of neoadjuvant systemic therapies. The BCC-BTG commented that the future direction is for greater neoadjuvant treatment in primary operable breast cancer and that breast cancer surgeons across the country are very aware and familiar with this approach and treatment strategy.

Both clinician groups agreed that patients who are not suited for neoadjuvant chemotherapy or HER2-directed therapy due to the presence of comorbidities or poor LVEF would be least suitable for treatment with pertuzumab.

With respect to patients who are the most likely to exhibit a response to neoadjuvant treatment with pertuzumab in practice, the BCC-BTG noted there is currently no predictive clinical factor or biomarker that has been validated as predictive of clinical benefit to neoadjuvant anti-HER2-directed therapy other than HER2 positivity (immunohistochemistry [IHC] score of 3 or greater and/or positive for fluorescent in situ hybridization [FISH], as per ASCO/CAP guidelines). The clinicians from the BCC-BTG added there are emerging data that stromal tumour-infiltrating lymphocytes are predictive of a higher pCR to both chemotherapy and to HER2-directed antibodies (pertuzumab and trastuzumab), though this biomarker is not standardly reported or used in clinical decision-making today. The OH-CCO BCDAC noted that the drug under review will be used only in breast cancer patients who are HER2-positive.

Assessing Response to Treatment

The BCC-BTG noted that a pCR, defined as no invasive disease in the breast and lymph nodes, where residual DCIS is permissible in the definition, can be used to determine whether a patient is responding to treatment in clinical practice. Both clinical groups agreed that a clinically meaningful response to treatment includes achievement of a pCR, which is associated with improved long-term clinical outcomes, including DFS and OS. The BCC-BTG added that enhanced breast cancer surgeries, such as downstaging from mastectomy to breast-conserving surgery or from axillary node dissection to sentinel node biopsy, are also clinically meaningful responses. The clinicians from BCC-BTG also added that achieving a pCR would abrogate the need for adjuvant T-DM1.

According to the clinician groups, treatment response should be assessed pathologically and patients should be monitored routinely during neoadjuvant therapy to evaluate response to treatment.

Discontinuing Treatment

As per the BCC-BTG input, factors to consider when deciding to discontinue treatment include clinical progression of disease (as documented by physical exam and/or radiological imaging) or significant adverse events (such as clinically significant cardiotoxicity or diarrhea). The OH-CCO BCDAC added disease progression on treatment (which is rare) and toxicities are important factors to consider when deciding whether to discontinue treatment.

Prescribing Conditions

Both clinician groups agreed that outpatient clinics are the most appropriate setting for treatment with pertuzumab. The BCC-BTG added that delivery is concurrent with taxane and trastuzumab; thus treatment should be in the outpatient chemotherapy unit.

Additional Considerations

The OH-CCO BCDAC noted that the treatment strategy proposed has not been specifically evaluated in any clinical trial.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. PAG noted that in most provinces, the current standard of care for the neoadjuvant treatment of HER2-positive breast cancer is trastuzumab plus chemotherapy. Pertuzumab, being an IV drug, would be administered in an outpatient chemotherapy centre for appropriate administration and monitoring of infusion-related reactions. PAG highlighted several enablers to the implementation of pertuzumab in the neoadjuvant setting: the dose and frequency of pertuzumab in the neoadjuvant setting is the same as in the metastatic setting, it is an add-on to existing treatment, and drug wastage is not a concern since pertuzumab vials contain the amount of the fixed dose. PAG also identified barriers to implementation that include: the high cost of pertuzumab and the additional preparation time and chair time needed for the infusion. Pertuzumab is administered for 4 to 6 cycles before surgery and PAG noted that given the high cost of pertuzumab, there is a significant difference in cost between 4 cycles and 6 cycles. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 4.

Clinical Evidence

The clinical evidence included in the review of pertuzumab is presented in 3 sections. The first section, the systematic review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada as well as those studies that were selected according to an a priori protocol. The second section normally includes indirect evidence from the sponsor and indirect evidence selected from the literature that met the selection criteria specified in the review; however, no indirect evidence was submitted and none was found in the literature. The third section includes sponsor-submitted, long-term extension studies and additional relevant studies that were considered to address important gaps in the evidence included in the systematic review; however, none were considered relevant for this review.

Systematic Review (Pivotal and Protocol Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of pertuzumab by IV infusion in combination with trastuzumab and chemotherapy for the neoadjuvant treatment of patients with HER2-positive locally advanced, inflammatory, or early-stage breast cancer (either > 2 cm in diameter or node-positive).

Methods

Studies selected for inclusion in the systematic review included pivotal studies provided in the sponsor’s submission to CADTH and Health Canada, as well as those meeting the selection criteria presented in Table 5.

Outcomes included in the CADTH review protocol reflect outcomes considered to be important to patients, clinicians, and drug plans.

The literature search for clinical studies was performed by an information specialist using a peer-reviewed search strategy according to the PRESS Peer Review of Electronic Search Strategies checklist.

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946—) through Ovid and Embase (1974—) through Ovid. All Ovid searches were run simultaneously as a multi-file search. Duplicates were removed using Ovid deduplication for multi-file searches, followed by manual deduplication in Endnote. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concepts were Perjeta (pertuzumab) and Herceptin (trastuzumab). Clinical trials registries were searched: the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform (ICTRP) search portal, Health Canada’s Clinical Trials Database, and the European Union Clinical Trials Register.

Methodological filters were applied to limit retrieval to RCTs or controlled clinical trials. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. See Appendix 1 for the detailed search strategies.

The initial search was completed on May 12, 2021. Regular alerts updated the search until the meeting of the CADTH pan-Canadian Oncology Drug Review Expert Committee (pERC) on Sep 8, 2021.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool for Searching Health-Related Grey Literature checklist. Included in this search were the websites of regulatory agencies (FDA and European Medicines Agency [EMA]). Google was used to search for additional internet-based materials. See Appendix 1 for more information on the grey literature search strategy.

These searches were supplemented by reviewing bibliographies of key papers and through contacts with appropriate experts. In addition, the sponsor of the drug was contacted for information regarding unpublished studies.

Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the predetermined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer were acquired. Reviewers independently made the final selection of studies to be included in the review, and differences were resolved through discussion.

Findings From the Literature

A total of 4 studies were identified from the literature for inclusion in the systematic review (Figure 1). The included studies are summarized in Table 6 through to Table 9. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Description of Studies

NEOSPHERE was an open-label multinational RCT (5 Canadian sites) whose objective was to make a preliminary assessment of the efficacy of neoadjuvant treatment with trastuzumab plus docetaxel (N = 107) compared with pertuzumab and trastuzumab plus docetaxel (N = 107), or compared with pertuzumab and trastuzumab (N = 107), and to compare pertuzumab plus docetaxel (N = 96) with pertuzumab and trastuzumab plus docetaxel in patients with T2 to T4d HER2-positive breast cancer, based on pCR (Figure 2). The treatment comparison relevant to this review is pertuzumab and trastuzumab plus docetaxel compared with trastuzumab plus docetaxel.

In NEOSPHERE, randomization was conducted 1:1:1:1 by an interactive web and voice response system (IxRS) and was stratified by breast cancer type (operable, locally advanced, or inflammatory) and by hormone receptor status. Arm D was added as a protocol amendment; therefore, there were fewer patients enrolled into this treatment group. The study consisted of a 12-week, 4-cycle neoadjuvant phase followed by surgery, and then an adjuvant phase where patients received FEC for 3 cycles and trastuzumab for 1 year. The treatment group that did not receive docetaxel in the neoadjuvant phase (Arm C) received docetaxel for 4 cycles, then 3 cycles of FEC, in addition to trastuzumab for 1 year. The primary outcome of the trial was evaluated when all patients had received 4 cycles of neoadjuvant treatment and had undergone surgery or had withdrawn from the study.

Figure 2: Design of NEOSPHERE

FEC = 5-fluorouracil and epirubicin plus cyclophosphamide; q3w = every 3 weeks.

Source: Clinical Study Report for NEOSPHERE.⁴

PEONY was a double-blind RCT conducted entirely in Asia whose primary objective was to evaluate the efficacy of pertuzumab and trastuzumab plus docetaxel (N = 219) compared with placebo plus trastuzumab plus docetaxel (N = 110), in a neoadjuvant setting before surgery, in patients with early-stage or locally advanced HER2-positive breast cancer. Other efficacy outcomes were to be assessed in the final analysis of the study 5 years after randomization of the last patient. Randomization was carried out 2:1 using IxRS, stratified by disease category (early-stage or locally advanced breast cancer) and hormone receptor status (estrogen receptor–positive and/or progesterone receptor–positive or negative for both). Patients had their pathologic response evaluated after completing 4 cycles of neoadjuvant therapy and surgery. Post surgery, patients received FEC for 3 cycles and then continued on HER2 therapy every 3 weeks until disease recurrence or unacceptable toxicity for up to 1 additional year.

Figure 3: Design of PEONY

FEC = 5-fluorouracil and epirubicin plus cyclophosphamide; HER2 = human epidermal growth factor receptor 2.

Source: Clinical Study Report for PEONY.⁵

TRYPHAENA was an open-label, multinational study with 2 Canadian sites. TRYPHAENA (N = 225) and BERENICE (N = 400) were multinational studies with Canadian sites (3 sites in TRYPHAENA, 5 sites in BERENICE). In TRYPHAENA, randomization of 225 patients was carried out 1:1:1 by IxRS and was stratified by breast cancer type (operable, locally advanced, or inflammatory) and by hormone receptor status (hormone receptor–positive or –negative). In TRYPHAENA, patients received 6 cycles of neoadjuvant therapy with pertuzumab, trastuzumab, and an anthracycline or carboplatin-based chemotherapy, followed by surgery then trastuzumab for 1 year. TRYPHAENA did not have a comparator group, and the primary objective was to assess safety and tolerability of pertuzumab in combination with various regimens (Figure 4). After 5 years had elapsed since the randomization of the last patient, the study was completed on January 25, 2016.

BERENICE was an open-label, non-randomized study where patients were allocated 1:1 to 1 of 2 cohorts. The choice of neoadjuvant treatment was made by the investigator; however, only 1 cohort at a time was opened to enrolment at any given site; thus, investigators could not enrol patients into both cohorts simultaneously. In BERENICE, patients received 4 cycles (12 weeks) of neoadjuvant therapy, followed by surgery, then 13 cycles of adjuvant therapy with pertuzumab and trastuzumab, for a total of 17 cycles. BERENICE did not have a comparator group, and the primary objective was to assess the safety and tolerability of pertuzumab in combination with various regimens (Figure 5). BERENICE is ongoing, and the study is planned to end 5 years after enrolment of the last patient, or when all patients have died or the trial is terminated by the sponsor, whichever comes earlier. It is expected that the study will last approximately 6.5 years, including time on treatment of approximately 1 year and follow-up for cardiac safety and efficacy for an additional 4 years.

Figure 4: Design of TRYPHAENA

5FU = 5-fluorouracil; FEC = 5-fluorouracil and epirubicin plus cyclophosphamide; TCH = docetaxel and carboplatin plus trastuzumab.

Source: Clinical Study Report for TRYPHAENA.⁶

Figure 5: Design of BERENICE

Source: Clinical Study Report for BERENICE.⁷

Populations

Inclusion and Exclusion Criteria

NEOSPHERE included females with locally advanced, inflammatory, or early-stage unilateral invasive breast cancer. The primary tumour had to be greater than 2 cm in diameter and confirmed to be HER2-positive, and patients were to have an ECOG performance status of 0 or 1. PEONY did not require patients to be female (but all were); otherwise, PEONY and NEOSPHERE had similar inclusion criteria, stipulating that patients had to have early-stage or locally advanced HER2-positive breast cancer with an ECOG performance status of 0 or 1. Both studies required an LVEF of 55% or greater at baseline.

Both studies excluded patients with stage IV metastatic disease, patients who had received prior anti-cancer therapy or radiotherapy for any malignancy, and patients with other malignancies (excluding carcinoma in situ of the cervix and basal cell carcinoma). PEONY excluded patients with inflammatory breast cancer, but NEOSPHERE did not. Patients with serious cardiac illness were also excluded from both studies.

The inclusion and exclusion criteria for TRYPHAENA and BERENICE did not differ markedly from that of NEOSPHERE and PEONY. BERENICE was the only study of the 4 included studies that included patients with a primary tumour that was less than 2 cm in diameter as long as it was greater than 5 mm and the cancer was node-positive.

Baseline Characteristics

Patients in NEOSPHERE and PEONY were about 50 years old at baseline (in NEOSPHERE, the mean age was 49.8 years, standard deviation [SD] 10.0 years; in PEONY, it was 48.8 years, SD 9.5 years). The majority of patients (approximately 70%) were White in NEOSPHERE, and all patients were Asian in PEONY. The mean age was similar in TRYPHAENA (50.2 years, SD 10.9) and BERENICE (49.6 years, SD 11.6), and a higher percentage of patients were White (approximately 75% to 85%) compared with NEOSPHERE. The majority of patients (nearly 90%) in NEOSPHERE, PEONY, and TRYPHAENA had an ECOG performance status of 0; the rest had a status of 1. The ECOG performance status of patients at baseline was not reported in BERENICE. Approximately half (47% in NEOSPHERE, 51% in PEONY, 51% in TRYPHAENA) of patients were either estrogen or progesterone receptor–positive while, in BERENICE, about 2-thirds of patients were estrogen or progesterone receptor–positive. The majority of patients in NEOSPHERE had breast cancer types that were either locally advanced (32%) or operable (61%) and the remainder had inflammatory breast cancer while, in PEONY, 70% were considered early stage and the remainder were considered locally advanced. Similar to NEOSPHERE, greater than 90% of patients in TRYPHAENA had breast cancer that was considered locally advanced or operable, and the remainder had inflammatory breast cancer. In BERENICE, baseline disease was not categorized as locally advanced and so forth; instead, disease category was reported using tumour, nodes, and metastases (TNM) staging. In BERENICE, the majority of patients had T2 (67%) tumours followed by T3 (20%); 47% had N1 disease, 8% had N2, 2% had N3, and 100% had disease classified as M0. The majority of patients in NEOSPHERE and TRYPHAENA had disease classified as T2N0M0 (NEOSPHERE: ||||||, TRYPHAENA: 31%) or T2N1M0 (NEOSPHERE: ||||||, TRYPHAENA: 33%). In PEONY, most patients had disease classified as T2 (67%), followed by T3 (22%), and most had lymph node–positive disease (76%).

In NEOSPHERE, there were some imbalances in baseline characteristics across treatment arms; the percentage of patients who were White in the pertuzumab plus docetaxel arm was 63.5% compared with 72.0% to 74.8% in the other 3 arms. ECOG performance status at baseline also appeared to be imbalanced across treatment arms in NEOSPHERE, as was histologic tumour grade (between 26.0% and 34.6% of patients had tumours that were classified as moderately differentiated). In PEONY, 70.8% of patients in the pertuzumab and trastuzumab plus docetaxel arm and 64.5% in the trastuzumab plus docetaxel arm had T2 tumours, and most had positive lymph node status (73.1% of patients in the pertuzumab and trastuzumab plus docetaxel arm and 80.9% of patients in the trastuzumab plus docetaxel arm). There were also imbalances between arms in TRYPHAENA for race (76.4% of patients were White in arm A versus 69.3% in arm B versus 84.2% in arm C, |||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||, hormone receptor status (estrogen and/or progesterone receptor–positive: 53.4% versus 46.7% versus 51.9%, respectively), and disease category (20.5% had locally advanced disease versus 22.7% versus 31.2%, respectively). In BERENICE, there were imbalances between groups in hormone receptor status (44.2% of the patients in 1 group had disease that was estrogen and progesterone receptor–positive versus 36.8% in the other group).

Interventions

In NEOSPHERE and the other studies, pertuzumab was administered on day 1 of cycle 1 as an IV infusion at a loading dose of 840 mg. On day 22 (3 weeks after the first dose) and every 3 weeks thereafter, pertuzumab was administered as an IV infusion at a dose of 420 mg. The initial dose of pertuzumab was administered over 60 (± 10) minutes and patients were observed for a further 60 minutes. The infusion was slowed or interrupted if the patient experienced fever, chills, or other infusion-associated symptoms. If the infusion was well tolerated, subsequent doses were administered over 30 (± 10) minutes and patients observed for a further 60 minutes. All infusion-associated symptoms had to be resolved either before chemotherapy was given or the patient was discharged. Trastuzumab was administered on day 1 of cycle 1 as an IV infusion at a loading dose of 8 mg/kg. On day 22 (3 weeks after the first dose), and every 3 weeks thereafter, trastuzumab was administered as an IV infusion at 6 mg/kg. The initial dose of trastuzumab was administered over 90 (± 10) minutes and patients were observed for at least 30 minutes from the end of the infusion for infusion-associated symptoms. Interruption or slowing of the infusion was used to control symptoms, which could be resumed if symptoms abated. If the infusion was well tolerated, subsequent infusions were administered over 30 (± 10) minutes and patients were observed for a further 30 minutes. All infusion-associated symptoms had to be resolved either before further study treatment was given or the patient was discharged. Docetaxel was administered at 75 mg/m² as an IV infusion over 60 (± 10) minutes, after the trastuzumab or pertuzumab infusion observation period. From day 22 onward (3 weeks after the first dose), docetaxel was escalated in the subsequent cycle(s) up to 100 mg/m² if no limiting toxicity was experienced. In the adjuvant phase, patients exposed to docetaxel in the neoadjuvant phase received trastuzumab 6 mg/kg IV followed by FEC on day 1 and every 3 weeks thereafter for cycles 5 to 7, with each cycle lasting 21 days beginning 2 weeks after surgery. Trastuzumab monotherapy was administered for cycles 8 to 17, at the 6 mg/kg IV dose, every 3 weeks. In the arm that did not receive docetaxel in the neoadjuvant phase, trastuzumab 6 mg/kg IV was administered, followed by docetaxel 75 mg/m² IV at cycle 5, and docetaxel was escalated to 100 mg/m² for cycles 6 to 8 (as long as there were no dose-limiting toxicities). For cycles 9 to 11, 6 mg/kg IV of trastuzumab followed by FEC was administered on day 1 and every 3 weeks thereafter, and trastuzumab 6 mg/kg IV monotherapy was continued every 3 weeks for cycles 12 to 17.

In PEONY, the study treatment was administered in 3-week cycles. Patients were treated with trastuzumab, pertuzumab, and docetaxel (arm A) or trastuzumab, placebo, and docetaxel (arm B) for 4 cycles in the neoadjuvant setting. Matched placebo contained the formulation but not the antibody itself. After surgery, patients received 3 cycles of FEC chemotherapy. Patients then continued HER2-targeted therapy with trastuzumab and pertuzumab (arm A) or trastuzumab and placebo (arm B) for up to 1 year in total (17 cycles, including 4 cycles in the neoadjuvant setting) or until disease recurrence, as assessed by the investigator, or unacceptable toxicity.

For a summary of regimens used in TRYPHAENA, see Figure 4. This study also featured a neoadjuvant phase followed by surgery and then an adjuvant phase, and combined pertuzumab with various chemotherapy regimens at the doses described previously. In the neoadjuvant phase of TRYPHAENA, pertuzumab and trastuzumab were given either concomitantly or sequentially with anthracycline-based chemotherapy or concomitantly with non-anthracycline-based chemotherapy. Neoadjuvant treatment was for 6 cycles in TRYPHAENA, which was longer than the 4 cycles in NEOSPHERE and PEONY.

For a summary of regimens used in BERENICE, see Figure 5. In BERENICE, the patients in arm A received doxorubicin for 4 cycles before receiving 4 cycles of pertuzumab plus trastuzumab combined with paclitaxel, while patients in arm B received 4 cycles of FEC followed by 4 cycles of pertuzumab plus trastuzumab combined with docetaxel. Adjuvant therapy in TRYPHAENA consisted of trastuzumab 6 mg/kg IV every 3 weeks and continued for a maximum of 1 year, while adjuvant therapy in BERENICE consisted of 13 cycles of pertuzumab plus trastuzumab. Additional radiotherapy, hormonal therapy, and chemotherapy post surgery and during adjuvant trastuzumab was allowed at the discretion of the investigator in TRYPHAENA while, in BERENICE, additional radiotherapy and adjuvant hormonal therapy could be given as clinically indicated, according to the guidelines provided per protocol.

As for concomitant medications, in general, across the trials, all medications taken by the patient for concomitant diseases were allowed to continue during the study treatment period and were recorded on the electronic case report form.

Outcomes

A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trials included in this review is provided in Table 14. These end points are further summarized subsequently. A detailed discussion and critical appraisal of pCR as an outcome measure is provided in Appendix 4.

Overall Survival

OS was not assessed as an efficacy outcome in NEOSPHERE. In PEONY, patients were to be followed for survival every 3 months for the first year and every 6 months thereafter until disease progression or recurrence, or until 5 years after randomization of the last patient, whichever came first. OS was an exploratory outcome in TRYPHAENA and BERENICE.

Invasive Disease–Free Survival

Invasive DFS was not assessed in NEOSPHERE, PEONY, or TRYPHAENA; these data were not yet mature in BERENICE, according to the sponsor, and were therefore not reported.

Disease-Free Survival

DFS was a secondary outcome. It was defined in NEOSPHERE as the time from the first date of no disease (i.e., date of surgery) to the first documentation of progressive disease (PD) or death, while, in PEONY, it was until disease recurrence or death. In NEOSPHERE, PD was identified by the investigator in response to a question on the electronic case report form and was not based solely on Response Evaluation Criteria in Solid Tumors (RECIST) criteria, and additional clinical information could be considered in the assessment of PD. Any evidence of contralateral disease in situ was not considered PD. Patients who had surgery but did not achieve a pCR were censored at the date of surgery. Patients who withdrew from the study without documented progression and for whom there was evidence that evaluations were made were censored at the date of the last assessment when the patient was known to be disease-free. DFS was an exploratory outcome in TRYPHAENA and was defined as it was in NEOSPHERE and was not assessed in BERENICE. Across trials, DFS included the adjuvant phase of treatment.

Progression-Free Survival

PFS was a secondary outcome in NEOSPHERE and PEONY and was defined as the time from the date of randomization to the first documentation of PD or death. Any evidence of contralateral disease in situ was not considered as PD, although invasive contralateral disease was counted as PD in PEONY. Otherwise, PD in PEONY was defined as described previously. Patients who withdrew from the study without documented progression and for whom there was evidence that evaluations were made were censored at the date of the last assessment when the patient was known to be free from PD. Patients without post-baseline assessments but known to be alive were censored at the time of randomization. PFS was an exploratory outcome in TRYPHAENA, defined as the time from randomization to first documentation of PD or death, and PFS was not studied in BERENICE. Across the trials, PFS included the adjuvant phase of treatment.

Event-Free Survival

EFS was a secondary outcome in PEONY and was defined as the time from randomization to the first documentation of 1 of the following: death from any cause, disease recurrence, or disease progression (before surgery). PD was determined by the investigator using RECIST v1.1. Any evidence of contralateral disease in situ was not counted as PD.

Patients who did not have an event at the time of the analysis were censored as of the date they were last known to be alive and event-free. EFS was not studied in NEOSPHERE and was not assessed in TRYPHAENA. EFS was exploratory in BERENICE, defined as the time from enrolment to first occurrence of PD, relapse, or death. Across trials, EFS included the adjuvant phase of treatment.

Pathologic Complete Response

The primary outcome of NEOSPHERE and PEONY was pCR. In NEOSPHERE, pCR was assessed specifically in the breast; in PEONY, tpCR was assessed as the primary outcome and bpCR was assessed as a secondary outcome. In both studies, pCR was assessed post surgery, after patients had received 4 cycles of neoadjuvant therapy or after study withdrawal. In NEOSPHERE, pCR was defined as the absence of invasive neoplastic cells on microscopic examination of the surgical specimen (residual in situ disease was allowed). In PEONY, tpCR was defined as the absence of invasive neoplastic cells in the surgically resected breast specimen and all sampled ipsilateral nodes. In NEOSPHERE, all tumour responses were assessed locally and were not independently reviewed while, in PEONY, tumour responses were assessed both locally and centrally by an IRC. In PEONY, IRC assessments were used for the primary outcome and were conducted in a blinded manner consistent with FDA recommendations. In TRYPHAENA, pCR was assessed in the breast as a main efficacy outcome and was evaluated post surgery; in BERENICE, pCR was assessed in breast and nodes post surgery and was also a main efficacy outcome.

Objective Response

Clinical response, defined as achieving a CR or PR, was a secondary outcome in NEOSPHERE and PEONY. In NEOSPHERE, tumour burden was assessed at baseline using mammography and clinical breast exam after completion of all pre-operative treatment cycles. Any additional conventional assessment methods used by local practice were also collected (i.e., ultrasound, CT, MRI, X-ray). Clinical breast exam was also conducted after each treatment cycle in the neoadjuvant phase. In PEONY, the clinical response rate was defined as the proportion of patients who achieved a clinical response during cycles 1 to 4 (pre-surgery). If PD was suspected, then an unscheduled assessment was performed and, if confirmed, the patient was removed from study treatment and offered local standard of care, such as a second-line cytotoxic regimen, radiation, or surgery. For assessment of response, modifications to RECIST were employed in NEOSPHERE. For the primary lesion in the breast, the RECIST criteria were applied in terms of percentages, but the sum of lesions was not used; instead, only the size of the primary breast lesion was used to determine response. For overall response, the sizes would be summed only if the method of assessment was the same for all lesions (breast and nodes). For example, if a patient had a breast lesion assessed by mammogram and lymph nodes by ultrasound, then each would be summed only within that method of assessment. ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| In PEONY, clinical response was required to be assessed by clinical breast exam, at each cycle between days 15 and 21, or on study day 1 of the next cycle, and by mammography at baseline and cycle 4. Clinical response was defined as CR, PR, stable disease, and PD, and was identified as per local practice based on RECIST criteria. Clinical response was an exploratory outcome in TRYPHAENA and BERENICE and was defined as patients achieving either a CR or PR at any time pre-surgery. In TRYPHAENA, clinical response was assessed as per local practice at each cycle, between days 15 and 21, or on study day 1 of the next cycle, and, in BERENICE, it was assessed before each new cycle by clinical exam, mammography, and/or other methods, as per local practice.

Breast-Conserving Surgery

The breast-conserving surgery rate was defined as the proportion of patients who achieved breast-conserving surgery out of the intention-to-treat (ITT) population without inflammatory breast cancer, as these patients received mastectomy irrespective of their response to neoadjuvant therapy. The breast-conserving surgery rate among patients in whom mastectomy was initially planned was a secondary outcome in NEOSPHERE and an exploratory outcome in TRYPHAENA. The breast-conserving surgery rate among all patients was an exploratory outcome in the BERENICE study and was not assessed in PEONY.

Symptoms and Health-Related Quality of Life

Symptoms and health-related quality of life were not assessed in any of the included studies.

Statistical Analysis

Primary Outcome(s) of the Studies

Power Calculation

With a sample size of 400 patients and a randomization ratio of 1:1:1:1, NEOSPHERE had 80% power to detect a 15% increase between each of the 3 primary comparisons between treatment arms at an overall alpha of 0.2. A pCR rate of 25% was expected in the trastuzumab plus docetaxel and the pertuzumab plus docetaxel arms, while a 40% pCR in the pertuzumab and trastuzumab plus docetaxel and pertuzumab and the trastuzumab arms was described as being “of clinical interest.” |||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||. PEONY assumed a tpCR of 20% in the trastuzumab plus docetaxel arm, an absolute increase of 15% in the pertuzumab and trastuzumab plus docetaxel group, and a 2-sided significance of 5% to arrive at 85% power for a sample of 328 patients and a 2:1 randomization ratio. The data source that informed the estimates for tpCR was not reported.

No formal hypothesis testing was planned in TRYPHAENA. The planned sample size was based on the primary safety end point (incidence of symptomatic cardiac events and clinically significant declines in LVEF during the neoadjuvant period), and approximately 75 patients were to be recruited into the study. The approximate expected pCR rates were 50% in arm A, 45% in arm B, and 40% in arm C. The data source informing the estimates for pCR was not reported. For the incidence of symptomatic left ventricular dysfunction, if the true underlying incidence was 3%, the probability of observing more than 5 events in a treatment arm was 0.025. No power calculation was provided in BERENICE.

Statistical Test or Model

The 3 comparisons between the treatment arms for the primary outcome in NEOSPHERE were considered to be of equal importance and were made using a Cochran-Mantel-Haenszel test, stratified by operable (T2 to T3, N0 to N1, M0), locally advanced (T2 to T3, N2 or N3, M0; T4a to T4c, any N, M0) and inflammatory (T4d, any N, M0) breast cancer and estrogen and/or progesterone positivity (either is positive versus both are negative) (Table 15). This primary analysis focused on the neoadjuvant phase. Secondary outcomes (best tumour response, clinical response rate, time to clinical response, proportion of patients with T2 to T3 tumours receiving breast-conserving surgery, PD) were calculated and summarized for descriptive purposes only.

In PEONY, for the primary outcome of tpCR rate (IRC assessed), a 2-sided Cochran-Mantel-Haenszel test was performed for assessment of tpCR, stratified by disease category (early stage, locally advanced) and hormone receptor status (estrogen and/or progesterone receptor–positive, or negative for both) (Table 15). An unadjusted Fisher exact test was also performed. Secondary outcomes were assessed in a manner similar to how the primary outcome was assessed; however, the P values were stated to be “descriptive only.”

All efficacy outcomes in TRYPHAENA and BERENICE were assessed descriptively; therefore, no formal statistical analyses were performed.

Data Imputation Methods

In both NEOSPHERE and PEONY, patients with missing or unevaluable pCR, tpCR, or bpCR assessments were considered nonresponders. Other outcomes were not assessed formally and, therefore, no sensitivity analyses appear to have been planned.

Subgroup Analyses

In NEOSPHERE, pre-specified subgroups of interest included age (< 65 versus ≥ 65), menopausal status (postmenopausal versus premenopausal), primary tumour stage (T2 versus T3 or larger), lymph node status (positive versus negative), histological subtype (ductal versus non-ductal, not including unknown), disease category at baseline (early stage versus locally advanced), hormone receptor status at baseline (positive for estrogen and/or progesterone receptor versus negative for both), HER2 subgroups defined as an IHC score of 3+ (regardless of FISH status), IHC score of 2+ and FISH-positive, or IHC score of 0 or 1+ and FISH-positive, and pathological tumour stage at surgery (T0 or tumour in situ [Tis] versus T1 or larger). Comparability between groups (tests for interaction) was not checked nor was multiplicity accounted for.

The pre-specified subgroups of interest in PEONY included age (< 65 versus ≥ 65), menopausal status, primary tumour stage at baseline (T2 versus T3 or larger), lymph node status at baseline (positive versus negative), histological subtype (ductal versus non-ductal, not including unknown values), disease stage (early stage versus locally advanced), hormone receptor status at baseline (estrogen and/or progesterone receptor–positive versus negative for both), HER2 subgroups defined as an IHC score of 3+ (regardless of FISH status), IHC score of 2+ and FISH-positive, and IHC score of 0 or 1+ and FISH-positive and pathologic tumour stage at surgery (T0 or Tis versus T1 or larger). In PEONY, subgroups were assessed for various biomarkers as a sensitivity analysis for the primary outcome and were tested using an unstratified instead of a stratified Cochran-Mantel-Haenszel test. No tests were performed for interactions, and no P values were reported.

Multiplicity

In NEOSPHERE, for the primary outcome, there were 3 between-group comparisons made during the study, and a Simes multiplicity adjustment was applied to the individual P values obtained at the end of the study to maintain the overall false-positive risk at 0.2. No multiplicity adjustment for the secondary outcomes assessed was described in PEONY, and the P values for secondary outcomes were noted as descriptive.

Sensitivity Analyses

In NEOSPHERE, a sensitivity analysis was performed where assessment of pCR was repeated, counting only patients who completed surgery and had a valid pCR assessment (Table 15). This was done to exclude early dropouts due to PD from the analysis. A sensitivity analysis was also performed to assess the impact of changes to the TNM codes that were made in protocol version C (see list of protocol amendments that follows). A cross tabulation of responses was performed for patients recruited under protocol B and protocol C, as well as a chi-square test to test whether the proportion of patients in each breast cancer type under protocol B was not significantly different under protocol B versus protocol C.

Protocol Amendments

There were 3 protocol amendments in NEOSPHERE:

• Amendment 1 (December 4, 2007) added arm D to evaluate the efficacy of pertuzumab without trastuzumab and to update the hypothesis testing and analyses to reflect this change. Additionally, the number of patients in the study was increased from 180 to 400 and the number of study centres from between 45 to 55, to 100. Insulin-dependent diabetes mellitus was added as an exclusion criterion and the offset dosing schedule was clarified.

• Amendment 2 (December 11, 2008) corrected the TNM categories used to classify patients’ disease (operable, locally advanced, or inflammatory cancer) for the stratification groups.

• Amendment 3 (June 27, 2009) updated the definition of postmenopausal women, the contraceptive requirements for women of childbearing potential (as recommended by the Medicines and Health care products Regulatory Agency in accordance with the International Conference on Harmonization M3 guideline), and updated the pregnancy testing schedule. This amendment also clarified the clinical response definition.

There were 3 protocol amendments in PEONY:

• Amendments 2 (August 5, 2014) and 3 (August 15, 2014) were made to provide clarity and consistency around protocol procedures, assessments, and analyses (e.g., clinical tumour assessments, safety reporting, contraception use).

• Amendment 4 (December 6, 2016) changed the timing of the primary efficacy analysis. Originally, the analysis was to occur when all patients completed the treatment or discontinued visits (i.e., after completion of all adjuvant and neoadjuvant treatments). This was changed to occur when all patients who were eligible for surgery had completed surgery and been assessed for pathologic response (i.e., after surgical treatment following neoadjuvant therapy). This change is consistent with the globally accepted scientific conduct of neoadjuvant studies, according to the sponsor, and consistent with other pertuzumab neoadjuvant studies.

The total number of protocol amendments in TRYPHAENA was not reported but included changes to the timing of baseline mammography, |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||.

|||||||||||||| protocol amendments in BERENICE, |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| partners of male patients are not in scope), and discrepancies in the pathology manual and the schedule of assessments were corrected.

Analysis Populations

In NEOSPHERE, PEONY, and TRYPHAENA, the ITT population included all randomized patients regardless whether they received study medication and, in BERENICE, the ITT included all enrolled patients regardless whether they received study medication. The analyses for all studies were conducted according to the randomized or assigned treatment group. The safety population in all studies included all patients who received at least 1 dose of the study drug and had at least 1 safety assessment performed at baseline. NEOSPHERE also identified a per-protocol population that excluded patients from the ITT population who had a major protocol violation before the adjuvant phase of the study. TRYPHAENA included a per-protocol population that had to fulfill all of the following criteria: received 3 cycles of study medication (neoadjuvant setting), received no other anti-cancer treatment (non-study medication or radiotherapy), and underwent surgery. This analysis occurred only in the per-protocol population and differed from the ITT population by 10% or greater.

Results

Patient Disposition

Study withdrawals were less than 10% across all treatment arms in each study. The percentage of patients in each study who withdrew from treatment was generally higher than those withdrawing from the study, except for PEONY. The percentage of screen failures was typically greater than 25%. Not all studies reported the reasons for screen failures; however, in the ones that did (TRYPHAENA and BERENICE), common reasons included HER2-negative disease, metastatic disease, and HER2 positivity not confirmed by central laboratory.

Exposure to Study Treatments (Neoadjuvant Period)

Treatment exposure data are summarized in Table 20 to Table 23.

Across studies, patients generally received 3.9 or 4.0 of the 4 planned cycles of neoadjuvant treatment, on average. The use of co-interventions was reported to varying extents across the studies. Dose delays, interruptions, and discontinuations with pertuzumab occurred in 7.4% of patients in NEOSPHERE, 3.2% of patients in PEONY, 10% of patients in TRYPHAENA, and 17% of patients in BERENICE.

The use of radiotherapy in NEOSPHERE was similar between the pertuzumab and trastuzumab plus docetaxel group and the trastuzumab plus docetaxel group. The most common co-interventions in PEONY were 5-hydroxytryptamine 3 (5-HT3) antagonists, corticosteroids, and colony-stimulating factors, and their use was similar between groups. The most common co-interventions in TRYPHAENA were anti-estrogens, anti-anemia drugs, and colony-stimulating factors; radiotherapy, 5-HT3 antagonists, and steroids were the most common co-interventions in BERENICE.

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported subsequently. See Appendix 3 for detailed efficacy data. Note that data for longer-term survival outcomes reflect the treatment received in both the neoadjuvant and adjuvant phases. The overall median time on study in NEOSPHERE was |||||||||||||||||||||||||||||||||||||||||| in the pertuzumab and trastuzumab plus docetaxel arm and |||||||||||||||||||||||||||||||||||||||||||| in the trastuzumab plus docetaxel arm. In PEONY, the median time on study was |||||||||||||||||||||||||||||||||||||||||||| in the pertuzumab and trastuzumab plus chemotherapy arm and |||||||||||||||||||||||||||||||||||||||| in the trastuzumab plus chemotherapy arm. In TRYPHAENA, the median follow-up time was ||||||| in arm A |||||||||||||||||||||||||||||||||||||||| in arm B |||||||||||| and |||||||||||||| in arm C |||||||||||||| In BERENICE, median time on study was |||||||||||||||||||||||||||||||||||||||||||||| in cohort A and |||||||||||||||||||||||||||||||||||||||||||||| in cohort B.

Overall Survival

OS was only reported for the adjuvant period, where reported.

In NEOSPHERE, OS was not a pre-specified outcome; in PEONY, the data on OS were considered not mature by the sponsor and were therefore not reported. In TRYPHAENA, OS events ranged between 6.8% and 13.0% across arms (Table 26); in BERENICE, the data were not yet mature, according to the sponsor, and were therefore not reported.

Invasive Disease–Free Survival

This outcome was not assessed in NEOSPHERE, PEONY, or TRYPHAENA. In BERENICE, the data were not yet mature, according to the sponsor, and were therefore not reported.

Event-Free Survival

EFS was not assessed in NEOSPHERE. The data were not yet mature in PEONY and in BERENICE, according to the sponsor, and were therefore not reported. In TRYPHAENA, EFS was not assessed.

Progression-Free Survival

Assessment of PFS included the adjuvant phase. In NEOSPHERE (Table 24), progression events occurred in 15.9% of patients in the pertuzumab and trastuzumab plus docetaxel arm and 17.8% of patients in the trastuzumab plus docetaxel arm, for an HR of 0.69 (95% CI, 0.34 to 1.40). These data were not yet mature in PEONY, according to the sponsor. In TRYPHAENA, the PFS event rates (Table 26) were 13.7% in arm A, 14.7% in arm B, and 18.2% in arm C. This outcome was not assessed in BERENICE.

Figure 6: Kaplan–Meier Curve of PFS in NEOSPHERE

ITT = intention to treat; PFS = progression-free survival; vs = versus.

Source: Clinical Study Report for NEOSPHERE.

Disease-Free Survival

Reporting of DFS events included the adjuvant treatment phase in NEOSPHERE. A DFS event was reported for 14.9% of patients in the pertuzumab and trastuzumab plus docetaxel arm and 17.5% of patients in the trastuzumab plus docetaxel arm (Table 24). The DFS data in PEONY were not yet mature, according to the sponsor, and were therefore not reported. Results for DFS in TRYPHAENA were consistent with that of NEOSPHERE, with DFS events occurring in 14.5% of patients in arm A, 11.9% of patients in arm B, and 15.3% of patients in arm C (Table 26). DFS was not assessed in BERENICE.

Pathologic Complete Response

Of the 417 patients randomized into NEOSPHERE (Table 24), 392 underwent surgery and all of these patients had a valid assessment of pathologic response. In NEOSPHERE, the pCR rate was 45.8% in the pertuzumab and trastuzumab plus docetaxel arm and 29.0% in the trastuzumab plus docetaxel arm, for a difference in response rates between groups of 16.8% (95% CI, 3.5 to 30.1; P = 0.0094).

In PEONY (Table 25), the IRC-assessed tpCR rate was 39.3% in the pertuzumab and trastuzumab plus docetaxel arm and 21.8% in the trastuzumab plus docetaxel arm, for a difference in response rates of 17.45% (95% CI, 6.89 to 28.01; P = 0.0014). The tpCR rates in PEONY assessed by the local pathologist were 39.3% versus 20.9%, respectively (Table 25). Additionally, PEONY reported the bpCR rate as a secondary outcome, and the IRC-assessed bpCR rate was 42.0% versus 23.6%, for a between-group difference of 18.37% (95% CI, 7.60 to 29.15; P = 0.0010). The pCR rates in TRYPHAENA (Table 26) were 61.6% in arm A, 57.3% in arm B, and 66.2% in arm C, and 60.7% and 61.8% in the 2 cohorts in BERENICE (Table 27).

Subgroups

Complete subgroup data reported for the primary outcome in NEOSPHERE and in PEONY can be found in Table 34 and Table 35, respectively. In NEOSPHERE, in the subgroup of patients with locally advanced breast cancer, the pCR rate was 43.8% in the pertuzumab and trastuzumab plus docetaxel arm and 41.7% in the trastuzumab plus docetaxel arm. In the subgroup of patients with operable breast cancer, the rates were 47.7% versus 23.4%, respectively and, in those with inflammatory breast cancer, the rates were 40.0% versus 14.3%, respectively. The difference between the pertuzumab and trastuzumab plus docetaxel arm and the trastuzumab plus docetaxel arm for those patients who were estrogen and progesterone receptor–positive was 26.0% versus 20.0%, respectively; in those who were estrogen and progesterone receptor–negative, the rates were 63.2% versus 36.8%, respectively. The treatment effect appeared consistent across the various biomarkers assessed. In PEONY, the difference in tpCR rates in patients who were estrogen and/or progesterone receptor–positive was 33.3% versus 25.0%, respectively; in those who were estrogen and progesterone receptor–negative, the rates were 46.1% versus 18.5%, respectively.

Objective Response

In NEOSPHERE, when assessed by X-ray or mammography, a CR was observed in 18.9% of patients in the pertuzumab and trastuzumab plus docetaxel arm and in 18.3% of patients in the trastuzumab plus docetaxel arm, and a PR was observed in 49.1% of patients and 49.3% of patients, respectively (Table 24). When assessed by clinical exam, the rates of CR were 25.0% versus 21.6%, respectively and, for PR, were 63.0% versus 59.8%, respectively. In PEONY, clinical response was assessed as a secondary outcome, and an objective response (defined as obtaining either a CR or PR) during cycles 1 to 4 occurred in 88.6% of patients in the pertuzumab and trastuzumab plus docetaxel arm and in 78.2% of patients in the trastuzumab plus docetaxel arm, for a difference in objective response rates between groups of 10.4% (95% CI, 1.12 to 19.69). A CR was observed in 11.0% versus 10.0% of patients, and a PR was observed in 77.6% versus 68.2% of patients, respectively (Table 25).

Duration of Response

The duration of response was not reported in the included studies.

Health-Related Quality of Life

This outcome was not assessed in the included studies.

Symptoms

This outcome was not assessed in the included studies.

Breast-Conserving Surgery

Among patients with T2 or T3 tumours for whom mastectomy was initially planned, breast-conserving surgery was achieved in 23.2% of patients in the pertuzumab and trastuzumab plus docetaxel arm and in 22.6% of patients in the trastuzumab plus docetaxel arm in NEOSPHERE (Table 24). This outcome was not assessed in PEONY. In TRYPHAENA, for the subgroup of patients with T2 or T3 tumours in whom mastectomy was initially planned, the percentage of patients undergoing breast-conserving surgery was 21.7% in arm A, 16.7% in arm B, and 27.0% in arm C (Table 26). In the BERENICE study, 44.4% and 42.9% of patients with T2 or T3 tumours in each of the groups received breast-conserving surgery (Table 27).

Harms

Only those harms identified in the review protocol are reported subsequently. See Table 28 for detailed harms data.

Adverse Events

Adverse events occurred in 98.1% of patients in the pertuzumab and trastuzumab plus docetaxel arm and in the trastuzumab plus docetaxel arm in NEOSPHERE, and in 97.7% of patients in the pertuzumab and trastuzumab plus docetaxel arm and 96.4% of patients in the trastuzumab plus docetaxel arm in PEONY. The most common adverse events in the trials for pertuzumab and trastuzumab plus docetaxel versus trastuzumab plus docetaxel, were alopecia (63.6% versus 65.4% in NEOSPHERE; 49.1% in each arm in PEONY), neutropenia (50.5% versus 62.6% in NEOSPHERE; 48.2% versus 44.5% in PEONY), and diarrhea (45.8% versus 33.6% in NEOSPHERE; 38.5% versus 16.4% in PEONY). The most common grade 3 or greater adverse event was neutropenia (44.9% versus 57.0% in NEOSPHERE; 38.1% versus 32.7% in PEONY). Similar results were seen in TRYPHAENA and BERENICE, where approximately 99% of patients experienced an adverse event at some time during the study, and neutropenia was the most common grade 3 or greater adverse event.

Serious Adverse Events

Serious adverse events occurred in 10.3% of patients in the pertuzumab and trastuzumab plus docetaxel arm and 16.8% of patients in the trastuzumab plus docetaxel arm in NEOSPHERE, and in 10.1% versus 8.2% of patients, respectively, in PEONY. Febrile neutropenia was the most common serious adverse event in both NEOSPHERE arms, occurring in 5.6% of patients treated with pertuzumab and trastuzumab plus docetaxel, and 6.5% of patients treated with trastuzumab plus docetaxel and, in PEONY, occurring in 1.8% of patients in the pertuzumab and trastuzumab plus docetaxel arm and in no patients in the trastuzumab plus docetaxel arm. In TRYPHAENA, 28% of patients experienced a serious adverse event across arms and, in BERENICE, 24% of patients experienced a serious adverse event. Febrile neutropenia was the most common serious adverse event in both studies, occurring in about 10% of patients.

Withdrawals Due to Adverse Events

Few patients across studies stopped treatment due to an adverse event, which occurred in 0.9% of patients in the pertuzumab and trastuzumab plus docetaxel arm versus in no patients in the trastuzumab plus docetaxel arm in NEOSPHERE, and in 0.5% of patients in the pertuzumab and trastuzumab plus docetaxel arm and in no patients in the trastuzumab plus docetaxel arm in PEONY. The number of patients withdrawing due to an adverse event was 7% across arms in TRYPHAENA and 3.5% across cohorts in BERENICE.

Mortality

One patient died in each of the pertuzumab and trastuzumab plus docetaxel and trastuzumab plus docetaxel arms in NEOSPHERE, and both deaths were considered to be due to complications of breast cancer. One patient died in the pertuzumab and trastuzumab plus docetaxel arm in PEONY, due to a suicide, and there were no deaths in the trastuzumab plus docetaxel arm. There were no deaths in TRYPHAENA and no deaths in BERENICE.

Notable Harms

Notable harms of diarrhea were reported previously. Cardiac dysfunction occurred in 2.8% of patients in the pertuzumab and trastuzumab plus docetaxel arm and 0.9% of patients in the trastuzumab plus docetaxel arm in NEOSPHERE, and no patients in PEONY had an LVEF decline to less than 40%, or a primary or secondary cardiac event. Events of drug hypersensitivity or anaphylaxis occurred in 5.6% of patients in the pertuzumab and trastuzumab plus docetaxel arm and in 1.9% of patients in the trastuzumab plus docetaxel arm in NEOSPHERE, and in 3.2% versus 1.8% of patients in PEONY, respectively.