Phase II Trial of Weekly Vinorelbine and Trastuzumab as First-Line Therapy in Patients with HER2⁺ Metastatic Breast Cancer

Abstract

Learning Objectives

After completing this course, the reader will be able to:

1. Summarize the epidemiology and natural history of HER2⁺ breast cancer.

2. Differentiate the two primary methods for measuring HER2 in breast cancer patients.

3. Appreciate the rationale for investigating vinorelbine in combination with trastuzumab.

4. Discuss the relative efficacy and safety of trastuzumab and vinorelbine in HER2⁺ breast cancer.

Access and take the CME test online and receive one hour of AMA PRA category 1 credit at CME.TheOncologist.com

Purpose. Human epidermal growth factor receptor 2 (HER2) overexpression is associated with a more aggressive form of breast cancer that responds well to trastuzumab therapy. Trastuzumab-based combination regimens have shown greater antitumor activity than chemotherapy alone. These findings, coupled with the favorable antitumor activity and tolerability profile of vinorelbine in breast cancer, provided the rationale for investigating the novel combination of vinorelbine and trastuzumab.

Patients and Methods. A phase II, open-label trial of intravenous vinorelbine (30 mg/m² on day 1, then weekly) and trastuzumab (4 mg/kg on day 0, then 2 mg/kg weekly) was conducted in previously untreated HER2⁺ metastatic breast cancer patients. Vinorelbine dose was adjusted for grade 3/4 neutropenia; patients remained on combination therapy until disease progression or patient withdrawal due to adverse events.

Results. Of 40 enrolled patients (median age 51 years, range 30-82), 37 were evaluable for response. Overall response rate was 78% (29/37, 95% confidence interval [CI] 62%-90%), including four (11%, 95% CI 3%-25%) complete and 25 (68%) partial responses. Objective tumor response correlated with degree of HER2 positivity: immunohistochemistry (IHC) 3⁺ = 82% (18/22) response and IHC 2⁺ = 58% (7/12) response. Median time to progression was 72 weeks (95% CI 37-138 weeks); median survival has not been reached. Grade 3/4 neutropenia was the most frequent serious toxicity and cause of dose reductions (9% of courses) and omissions (10% of courses). No patient experienced serious cardiac toxicity.

Conclusions. Weekly vinorelbine/trastuzumab offers a high therapeutic index as initial therapy in patients with HER2⁺ metastatic breast cancer. Further investigation of this novel regimen is planned.

Introduction

Breast cancer is the second leading cause of cancer-related death (after lung cancer) in women in the U.S. and will comprise approximately 31% of all new cancer cases diagnosed in 2002 [1]. An estimated 203,500 new cases of breast cancer will be diagnosed in 2002, with 39,600 deaths expected [1]. Despite significant improvements in the early diagnosis and treatment of breast cancer and accompanying incremental gains in 5-year survival, a significant number of women will relapse and ultimately die of metastatic disease. Moreover, in the subgroup of breast cancer patients who overexpress human epidermal growth factor receptor 2 (HER2), which comprises 25%-30% of all human breast cancers [2], the disease is particularly virulent and is associated with a greater risk for disease progression and death [2–5]. In this patient subgroup, therapy with alkylating agents or tamoxifen may be proven suboptimal [6, 7], thereby providing the rationale for pursuing novel treatment approaches for the HER2-overexpressing tumor phenotype.

Breast cancers that overexpress the HER2 protein have a unique natural history characterized by high-grade histology, preponderance of estrogen receptor negativity, a greater likelihood of recurrence, and shortened survival. In metastatic breast cancer patients, HER2 overexpression identifies women who are candidates for treatment with the humanized antibody against the extracellular domain of HER2, trastuzumab (Herceptin^®; Genentech, Inc.; South San Francisco, CA). The binding of trastuzumab to HER2 in the presence of peripheral blood mononuclear cells induces antibody-dependent cellular toxicity [8]. Additionally, downregulation of HER2 inhibits tumor cell proliferation, and the role of trastuzumab in causing G₁ arrest may explain, in part, the synergistic antitumor activity observed when administered in combination with chemotherapy.

In patients treated with trastuzumab monotherapy, when HER2 expression was measured by immunohistochemical assays, objective responses were seen in 26% of chemotherapy-naïve patients and 15% of previously treated patients [9–11]. Additionally, trastuzumab administered with either paclitaxel or the combination of doxorubicin and cyclophosphamide significantly improved the objective response rate and prolonged the survival of women with metastatic HER2⁺ breast cancers [12–14]. Unfortunately, the addition of trastuzumab to an anthracycline-based regimen resulted in a more than fourfold increase in cardiac toxicities and one toxic death, attributed to a fatal arrhythmia. Because cardiac toxicity was a concern with some combination regimens, investigators have been interested in combining agents with more favorable safety profiles, among those with demonstrated activity in metastatic breast cancer.

Vinorelbine tartrate (Navelbine^®; GlaxoSmithKline; Philadelphia, PA) is a semisynthetic vinca alkaloid that inhibits tubulin polymerization, thereby impairing the function of mitotic spindles. Vinorelbine is associated with response rates of 35%-50% in first-line treatment of breast cancer and 16%-30% when used in second-line or salvage treatment [15–26]. The primary toxicities associated with vinorelbine are hematologic (i.e., neutropenia) and neuropathic. The reported incidence of cardiac toxicity and alopecia associated with vinorelbine is low, and neurotoxicity is generally mild. Data from in vitro studies have suggested that vinorelbine and trastuzumab are synergistic [27]. Additionally, the results of recent clinical studies have suggested that vinorelbine and trastuzumab may be synergistic [28, 29]. In a single-center feasibility study, Burstein et al. [29] reported an overall response (OR) rate of 75% (95% confidence interval [CI] 57%-89%, with a subset of front-line patients showing an 84% response rate). High antitumor response rates were seen not only in the first-line but also in the second- and third-line settings. Additionally, the toxicity profile of weekly vinorelbine and trastuzumab was favorable and conducive to long-term treatment. The generally mild toxicity profile of vinorelbine and trastuzumab and their activity in metastatic breast cancer patients provided the rationale for this study.

Methods

Eligibility

Female patients ≥18 years of age with microscopically confirmed carcinoma of the breast and HER2 overexpression were enrolled into the study. HER2 overexpression was determined in all patients by immunohistochemistry (IHC), and later, in a subgroup of patients by fluorescence in situ hybridization (FISH).

Patients were required to have bidimensionally measurable disease by physical exam or roentgenographic imaging, Karnofsky performance status of 70% or greater, and a life expectancy of >16 weeks. Patients were ineligible if bone was the only site of measurable disease. Patients could not have received cytotoxic therapy for metastatic disease. However, women who had received adjuvant chemotherapy were considered eligible provided that therapy was completed more than 12 months before study enrollment. Patients who had received radiation therapy and surgery were also eligible provided the interventions had been completed 3 and 2 weeks, respectively, before treatment with weekly vinorelbine and trastuzumab. Patients were required to have a cardiac ejection fraction >50% as measured by multigated radionuclide angiography or echocardiography. Eligible patients demonstrated adequate bone marrow, renal, and hepatic function.

Patients with a medical history of other malignancy within the previous 5 years that might affect the diagnosis of breast cancer, history (<6 months) of myocardial infarction, central nervous system metastases, and clinically significant (grade 2 or worse) peripheral neuropathy were excluded. Patients were required to practice adequate contraception and could not be pregnant or lactating. The Institutional Review Board of participating centers approved the study and all patients provided written informed consent.

Study Design and Treatment Plan

Patients received once-weekly vinorelbine and trastuzumab by intravenous bolus. We chose 28 days (4 weekly doses) as a cycle. Vinorelbine, 30 mg/m², was diluted to a concentration of 1.5 to 3.0 mg/ml in 0.9% saline or 5% dextrose injection and administered into a peripheral or central vein over 6 to 10 minutes. Vinorelbine was administered on days 1, 8, 15, and 22 of each 28-day course. Trastuzumab was given in a loading dose (4.0 mg/kg) on day 0, and subsequently, weekly at 2.0 mg/kg (days 8, 15, and 22). The infusion interval for trastuzumab was set at 90 minutes on day 0, and reduced over subsequent weeks to 30 minutes provided no infusion syndrome was evident. Trastuzumab was administered prior to vinorelbine for all weeks following the initial dose.

Adjustments in the dose of vinorelbine (i.e., 30 mg/m² to 20 mg/m²) were made in patients who experienced grade 3/4 hematologic toxicity, febrile neutropenia, grade 4 neutropenia persisting for 7 days, or grade 4 thrombocytopenia. Hematologic support with G-CSF (Neupogen^®; Amgen Inc.; Thousand Oaks, CA) and recombinant human erythropoietin (PROCRIT^®; Ortho Biotech Products, LP; Raritan, NJ) was applied at the investigator's discretion. Delays in vinorelbine dosing of 1 week were incurred for grade 3 neutropenia and for grade 2 or higher thrombocytopenia or neurologic toxicity. No adjustments in the dose of trastuzumab were permitted. If serious toxicity persisted for 4 weeks despite dose adjustments or omissions, the patient was taken off the study. Chemotherapy was administered until documented disease progression or unacceptable toxicity.

Study Analysis and Data Handling

Patients who received at least two courses of vinorelbine and trastuzumab were evaluable for efficacy, and patients who received at least one dose were evaluable for toxicity. The antitumor activity of the combination was assessed every 8 weeks by radiologic evaluation or by medical imaging (magnetic resonance imaging, computed tomography scan). Lesions followed by chest x-rays or physical examination were evaluated every 4 weeks. World Health Organization criteria were used to assess antitumor response. Tumor responses were determined by each investigator and subsequently verified at the Boca Raton Cancer Center (there was no central review of radiographic data). Time to progression was defined as the period of time from the first day of treatment to the time when disease progression or relapse was clearly documented. Survival was defined as the interval from the first day of treatment to the date of patient death.

The safety and tolerability of the combination were evaluated by changes in hematologic parameters and other clinical laboratory values, by physical examination, and by the frequency and severity of adverse events. Toxicity was graded according to the revised Cancer Therapy Evaluation Program Common Toxicity Criteria (version 2.0). Serious adverse events included grade 3 hematologic and nonhematologic toxicities. Potential trastuzumab-induced cardiac toxicity was monitored at 3-month intervals or more frequently in patients who experienced an uncomplicated 15% reduction in ejection fraction while receiving trastuzumab therapy.

All data are presented as the mean or median. The primary end point was overall tumor response, which was composed of complete and partial responses. The 95% CIs were determined for response rates and time to progression. Time to progression and overall survival were estimated using the Kaplan-Meier method.

Results

Patient Characteristics and Disposition

Forty women were enrolled in the trial between March 1999 and May 2001. The characteristics of this patient population are summarized in Table 1. The median age was 51 years (range 30 to 82), and 95% of patients had Karnofsky performance status ≥80 at study entry. HER2 protein overexpression was determined by IHC in 39 women and by FISH alone in one. Tumor specimens from 24 patients were retrospectively subjected to FISH. Seventeen (44%) patients had tumors that overexpressed HER2 weakly (2⁺), and in 22 (56%), the overexpression was 3⁺. In 25 patients where FISH analysis was available, 12 of the tumors had 2⁺ overexpression by IHC and only one was positive by FISH. All but one of the 3⁺ staining tumors by IHC were positive by FISH. The majority of patients (72%) had received adjuvant therapy, with patients evenly divided among those who received chemotherapy and hormone therapy. Most patients presented with two sites of metastases.

A total of 313 courses of vinorelbine and trastuzumab were administered (range 1 to 33). The mean weekly dose of vinorelbine was 24 mg/m². There were 244 adjustments to the vinorelbine dose; the primary reasons included neutropenia (9%), neuropathy (5%), and unknown (3%). Additionally, there were 179 dose omissions, the majority (71%) of which were attributed to neutropenia. Per protocol, there were no adjustments in the dose of trastuzumab.

Antitumor Response

Thirty-seven patients were evaluable for tumor response, and 40 were evaluable for toxicity. Of the three unevaluable patients, one was diagnosed with tuberculosis during the first course, one died of sepsis (unrelated to neutropenia) during the first course, and one was taken off study after the first course after it was found that she was FISH⁻. The antitumor response to weekly vinorelbine and trastuzumab is summarized in Table 2. The intent-to-treat OR rate was 72%. The OR rate was 78% (95% CI 62%-90%), including four (11%, 95% CI 3%-25%) complete responses (CRs) and 25 (68%) partial responses (PRs). Four (11%) patients exhibited stable disease, the durations of which were 4, 12, 21, and 122 weeks. Finally, four (11%) patients progressed.

Tumor response by HER2-overexpression status is shown in Table 3. The OR rate for 22 patients with IHC 3⁺ was 82%, compared with 58% of 12 patients with IHC 2⁺. Likewise, 83% of patients with FISH⁺ tumors achieved an OR, compared with 54% of patients who were FISH⁻. For 24 patients, HER2 status was assessed by both detection methods to determine the concordance between IHC and FISH. Ten of 11 patients with IHC 3⁺ tested HER2⁺ for gene amplification via FISH. Only 1 of 11 patients with weak HER2 expression (IHC 2⁺) tested positive for HER2 gene amplification (Table 4).

Time to progression and overall survival estimates are shown in Figure 1. The median time to progression was 72 weeks (17 months, 95% CI 37 to 138 weeks), while the median overall survival has not been reached.

Safety and Tolerability

Weekly vinorelbine and trastuzumab treatment was well tolerated. The most common serious toxicity was hematologic, with 44 (14%) grade 4 neutropenic events observed over 313 courses. Neutropenia was usually brief and noncumulative, was not associated with infections, and was manageable with dose adjustments. There were three hospitalizations attributed to febrile neutropenia; however, all cases resolved following vinorelbine dose reductions and G-CSF support. Low-grade anemia was common; however, grade 3 anemia was limited to one patient, and patients received recombinant human erythropoietin at the discretion of the treating physician. Additionally, there was an absence of thrombocytopenia.

The nonhematologic safety profile of weekly vinorelbine and trastuzumab was mild and favorable (Table 5). No patient was hospitalized as a result of this toxicity. There were three cases of grade 3 neuropathy that led to vinorelbine dose reductions. Neuropathy rarely (n = 2 events) led to dose omissions. No other nonhematologic toxicity was associated with dose reductions or omissions. All but one infusion-site reactions were mild, and all were resolved following intravenous access device placement. No patient experienced a serious infusion syndrome related to trastuzumab treatment. One grade 4 nonhematologic toxicity (fatigue) was observed. There was one case of grade 2 cardiac toxicity evidenced by a decreased ejection fraction of >20% but <24%; the patient was asymptomatic and remained on study until progression of disease. There was no grade 3 or 4 cardiac toxicity, and no patient discontinued trastuzumab therapy because of cardiac toxicity. Finally, no patient died as a result of complications related to treatment with vinorelbine and trastuzumab.

Discussion

It is always a challenge to find effective yet well-tolerated anticancer treatments. In the present study, patients received weekly vinorelbine (30 mg/m²) and trastuzumab (2 mg/kg) in a convenient, outpatient setting. Although the optimum doses of each agent and dosing duration are unknown, patients received the combination at or near full-dose levels for up to 33 courses without major untoward toxicity. The mild toxicity profiles of vinorelbine and trastuzumab suggest that the treatment can be administered over the long term as clinically warranted. In addition to a favorable toxicity profile, the results of our multicenter study confirmed the high antitumor response rates reported previously in a single institution trial of weekly vinorelbine and trastuzumab [29].

The addition of vinorelbine to trastuzumab therapy appears to improve the OR rates up to threefold compared with either one alone in HER2-overexpressing metastatic breast cancer patients [29]. The OR rate observed in this trial (78%, 95% CI 62%-92%) was at least comparable with that reported by Burstein et al. from their multicenter trial with vinorelbine/trastuzumab (64%, 95% CI 49%-77%) [28] despite the large number of patients accrued in our trial who were HER2⁺/FISH⁻. In contrast, patients enrolled in the multicenter Burstein et al. trial were required to have HER2⁺ tumors by FISH⁺ or IHC 3⁺.

Other trastuzumab-based regimens have been of interest to investigators, including novel combinations with taxanes. Although a direct comparison of response rates among regimens used in different trials is problematic due to differences in selected patients, number and type of prior treatments, and HER2 testing methodologies, the response rates achieved with weekly vinorelbine and trastuzumab are comparable with the best responses observed with other more toxic combination regimens [12, 30]. In a recent study by Esteva et al. [30], the OR rate in patients receiving weekly docetaxel and trastuzumab therapy was 63% (19/30). In that trial, the majority (24/28) had HER2⁺ tumors by FISH and had received prior chemotherapy. Similarly, the activity of trastuzumab and paclitaxel has been established in previously treated metastatic breast cancer patients [12]. The efficacy and tolerability of trastuzumab-based therapy with docetaxel and carboplatin have also been reported [31]. In that study, among 14 FISH⁺ patients, there were two CRs and seven PRs for an OR rate of 64%. A similar study of trastuzumab and docetaxel with cisplatin administered in first-line metastatic breast cancer patients reported an OR rate of 76% (26/34) [32]. That study enrolled patients who were IHC 2⁺ or 3⁺. Finally, trastuzumab-based combination therapy with gemcitabine has been evaluated in previously treated HER2⁺ metastatic breast cancer patients, with early success (PR rate of 32%) [33].

In addition to improving OR rates, vinorelbine and trastuzumab extended the time to disease progression compared with historical data of trastuzumab used as single-agent therapy. Similar to our tumor response findings, the time to progression of patients receiving vinorelbine and trastuzumab compared favorably with the best durations reported with other metastatic breast cancer treatments.

In this study, although involving small subsets of patients, there was a suggestion that HER2 status correlated with tumor response. Patients with tumors that were IHC 3⁺ exhibited a higher response rate (82%) than those with tumors that were weakly positive or IHC 2⁺ (58%). These data are consistent with the Burstein et al. study [29] in which 80% of HER2 3⁺ and 60% of HER2 2⁺ patients achieved tumor responses. Furthermore, all but one IHC 3⁺ overexpressing tumors tested positive for HER2 gene amplification by FISH (n = 11). Weaker IHC overexpression signals were not associated with HER2 gene amplification. Of the 12 patients exhibiting tumors that were IHC 2⁺, only one was FISH⁺. These findings are consistent with those reported by Seidman et al. [14], in which the antibody used in the IHC test may have had a lower specificity for HER2, thereby leading to a potential overestimation of HER2 positivity, or, alternatively, the reporting of a higher frequency of false positives. Additionally, because tumor response to trastuzumab therapy has been shown to correlate with HER2 positivity [14, 29], treatment of patients with nonamplified HER2 levels may under-report the true efficacy via the treatment of patients who are less likely to derive benefits. Nonetheless, greater overexpression or amplification of HER2 by either IHC or FISH was associated with higher response rates.

Another important aspect of treating patients with metastatic breast cancer is the selection of a regimen that is well tolerated and does not affect quality of life. Although quality of life was not assessed in the current study, weekly vinorelbine and trastuzumab treatment was generally well tolerated. The combination was associated with a favorable symptom profile that included mild (grade 1) neuropathy, gastrointestinal effects, and no significant alopecia. In contrast, 29% of patients receiving paclitaxel and trastuzumab experienced grade 3/4 neuropathy [14]. Additionally, the worst cardiac toxicity associated with weekly vinorelbine and trastuzumab involved one case of grade 2 cardiac toxicity. These data are consistent with those previously reported with vinorelbine and trastuzumab therapy [29]. Although patients should be monitored for cardiac toxicity while on trastuzumab therapy, vinorelbine does not appear to enhance trastuzumab-induced cardiac toxicity. The incidence of cardiac toxicity (<1% of courses) seen in our study was consistent with that expected with the use of trastuzumab as single-agent therapy in patients who have not received prior treatment with doxorubicin [34]. By contrast, the addition of paclitaxel and doxorubicin to trastuzumab has been associated with a higher incidence of serious cardiac toxicity, including life-threatening toxicity [12, 14, 34].

As expected with the use of vinorelbine, the most common serious toxicity was grade 3/4 neutropenia. Hematologic toxicity was managed by adjustments in the dose of vinorelbine and delays in administering subsequent doses when warranted. No adjustments in the dose of trastuzumab were required because of hematologic toxicity. Finally, no patient died of a complication related to hematologic toxicity.

In summary, the results of this study support the continued development of weekly vinorelbine and trastuzumab in patients with HER2⁺ metastatic breast cancer. The response rates observed in this trial were comparable with the previous single-center trial using this combination and continue to demonstrate the superiority of the combination over therapy with either agent alone. Additionally, the incorporation of vinorelbine in this combination regimen yielded a very favorable toxicity profile. There was a significant reduction in the incidence of alopecia and neurotoxicity relative to that observed with other combination regimens, such as paclitaxel and trastuzumab. The majority of patients were able to remain on therapy at or near the recommended doses in an outpatient setting, with little disruption in the administration of subsequent courses. Additionally, the favorable toxicity profile and high antitumor activity of weekly vinorelbine and trastuzumab may translate into improvements in the quality of life of patients with HER2⁺ metastatic breast cancer. In conclusion, weekly vinorelbine/trastuzumab offers a high therapeutic index as initial therapy in patients with HER2⁺ metastatic breast cancer. Further investigation of this novel regimen in comparison with docetaxel/ trastuzumab in a randomized multicenter study is ongoing.

Acknowledgments

The authors thank the many investigators and participating sites, including Alan B. Astrow, M.D., at The St. Vincent's Comprehensive Cancer Center, Generosa Grana, M.D., at Cooper Hospital/University Medical Center, Rafat Ansari, M.D., at Michigan Hematology Oncology PC, Carol J. Fabian, M.D., at The University of Kansas Medical Center, and Alexander E. Denes, M.D., at Washington University School of Medicine. The authors also thank the study nurse coordinators, including Beth Davis, C.R.A., Steve West, R.N., Pat McLaughin, R.N., Rebecca Clark-Snow, R.N., Mary Jean Wasielewski, R.N., Michelle Cordner, R.N., Tawnie Thomas, R.N., and Alethea Paradis, R.N. Finally, we thank the patients and administrative assistants for their participation and assistance, respectively.

References