Responses to subsequent anti-HER2 therapy after treatment with trastuzumab-DM1 in women with HER2-positive metastatic breast cancer

Abstract

Background: Women with human epidermal growth factor receptor 2 (HER2)-positive metastatic breast cancer (MBC) can respond to multiple lines of anti-HER2 therapy. It is unknown whether these patients will derive further clinical benefit following treatment with trastuzumab-MCC-DM1 (T-DM1).

Patients and methods: We retrospectively identified HER2-positive MBC patients treated with T-DM1 and characterized outcomes during subsequent lines of anti-HER2 therapy. Response was determined by a blinded radiology review. Time-dependent analyses were carried out using Kaplan–Meier estimates.

Results: We identified 23 patients treated with single-agent T-DM1 and report on the 20 patients who discontinued protocol therapy. All patients received trastuzumab-based metastatic therapy before initiation of T-DM1 [median 7 regimens (range 3–14)]. Of these 20 patients, 75% (15 of 20) received further therapy with or without anti-HER2 agents after discontinuing T-DM1. Partial response to either first- or second-subsequent line(s) of therapy was seen in 5 of 15 (33%) treated patients, including 33% (4 of 12) who received a regimen containing trastuzumab and/or lapatinib. Median durations of therapy to first- and second-subsequent regimens after T-DM1 were 5.5 and 6.4 months, respectively.

Conclusions: In heavily pretreated HER2-positive MBC patients, prior exposure to T-DM1 does not exhaust the potential benefit of ongoing anti-HER2 therapy with trastuzumab- and/or lapatinib-based regimens.

introduction

Human epidermal growth factor receptor 2 (HER2/neu, erbB2) is a transmembrane tyrosine kinase that when overexpressed in breast carcinomas is associated with a worse clinical prognosis. Trastuzumab, a monoclonal antibody to the extracellular domain of HER2, has significantly improved both systemic control and overall survival in patients with HER2-positive metastatic breast cancer (MBC) [1]. However, even in patients who initially respond to trastuzumab-based therapy, acquired resistance invariably develops. Multiple HER2-targeted agents are in development to overcome resistance pathways and create durable responses in patients with presumed trastuzumab-refractory disease [2, 3].

Trastuzumab-MCC-DM1 (T-DM1) is a novel antibody–drug conjugate designed to target the HER2-positive cancer cell by conjugating the maytansinoid antimicrotubule cytotoxic agent, DM1, to trastuzumab via a linker molecule [4]. Two phase II studies have shown efficacy of T-DM1 in heavily pretreated HER2-positive patients with evidence of progression on trastuzumab. In a single-arm phase II trial, T-DM1 demonstrated a confirmed overall response rate (ORR) by independent assessment of 25.9% in 112 patients who progressed on at least one prior HER2-directed therapy [5]. Similar results were demonstrated in a second single-arm phase II study of 110 patients who received prior chemotherapy, trastuzumab and lapatinib and found a confirmed ORR of 34.5% [6].

Although T-DM1 exhibits significant activity in trastuzumab-refractory breast cancer, it is unknown whether these patients will derive clinical benefit from further HER2-directed therapy. Clinical practice over the past decade has moved to adopt empiric continuation of trastuzumab beyond progression, and this approach has been validated in two prospective clinical trials that demonstrate an improvement in progression-free and overall survival [7, 8]. The combination of lapatinib and capecitabine has shown a lengthening of progression-free but not overall survival compared with capecitabine alone in patients with prior anthracycline, taxane, and trastuzumab exposure; and a number of other HER2-targeted drugs are in phase I–III clinical trials [2, 9, 10]. As an increasing number of HER2-directed agents enter clinical trials and clinical practice, the optimal sequence and duration of anti-HER2 treatment remains unclear. The purpose of this study was to characterize and describe the clinical outcomes of patients with HER2-positive MBC who received further treatment after exposure to T-DM1.

methods

Approval for this study was obtained from the Institutional Review Board of the Dana-Farber Cancer Institute (DFCI)/Harvard Cancer Center. We conducted a retrospective study of all 23 patients with metastatic HER2-positive breast cancer, who were enrolled in clinical trials using T-DM1 as a single agent at DFCI between December 2006 and February 2009. Follow-up information was available through 30 April 2010. Pathology records were reviewed to determine HER2 status of primary and/or metastatic biopsies. A positive result for HER2 was defined as 3+ by immunohistochemistry or amplification of HER2 by fluorescent in situ hybridization (FISH, defined as a ratio of HER2/CEP17 ≥2). Medical records for each patient were reviewed for the following information: date and stage of initial breast cancer diagnosis (American Joint Committee on Cancer, seventh edition), site of initial and subsequent disease recurrence, date of diagnosis of metastatic disease, type and duration of metastatic regimens, type and duration of subsequent therapy after T-DM1 discontinuation, vital status, clinical response to post-T-DM1 regimens, and date of death (or last follow-up).

Response was determined by an independent and blinded radiology review using modified RECIST 1.1 criteria without confirmatory scans; patients without radiological assessment were considered nonresponders. For patients without radiographic assessments, clinical response or stable disease was evaluated according to the interpretation of the primary oncologist from findings on physical examination, laboratory markers, and imaging studies. Duration of therapy was defined from initiation of therapy until treatment discontinuation, and for patients continuing on therapy, times were censored at date of last visit; analysis was carried out using Kaplan–Meier methodology.

results

description of study population

We identified all 23 patients treated on protocol-based therapy with single-agent T-DM1 at DFCI and report on the 20 patients who discontinued protocol, and hence T-DM1 therapy. Patient and initial tumor characteristics are shown in Table 1. Mean age at diagnosis was 44 years. The majority of patients were white (95%). At the time of initial presentation, 85% of patients had stage I–III disease, and 15% of patients had distant disease. The predominant histological subtype was grade 3 invasive ductal carcinoma. Of the patients without metastatic disease at initial presentation, 14 (82%) received adjuvant chemotherapy, 5 (29%) neoadjuvant therapy, and 10 (59%) adjuvant hormonal therapy and 5 (29%) were exposed to trastuzumab in either the neoadjuvant or adjuvant setting. Only 2 (12%) patients completed 1 year of adjuvant trastuzumab. The majority of patients (65%) were treated by a medical oncologist at DFCI before their diagnosis of metastatic disease.

Table 1.

Baseline patient and tumor characteristics

Characteristic	n (%)
N	20
Median age at initial diagnosis in years (range)	42 (22–66)
Female	20 (100)
White	19 (95)
Stage at diagnosis
I–III	17 (85)
IV	3 (15)
ER positive and/or PR positive	10 (50)
Types of adjuvant and/or neoadjuvant chemotherapy exposure
None	5 (25)
Chemotherapy alone	10 (50)
Trastuzumab with chemotherapy	5 (25)
Adjuvant radiation therapy	16 (80)
Adjuvant hormonal therapy	10 (50)

ER, estrogen receptor; PR, progesterone receptor.

The median length of follow-up was 111.0 months (range 55.2–267.8 months) from the time of initial breast cancer diagnosis. Ten patients were still alive at the time of the data cut-off. Median time from initial breast cancer diagnosis to diagnosis of distant metastatic disease was 37.5 months (range 0.2–195.8 months). Five (25%) patients were diagnosed and treated for local/regional recurrence before distant disease diagnosis. The most common locations of involvement at time of initial diagnosis of recurrent/metastatic disease are listed in Table 2. Breast/chest wall, lung/pleura, and liver were the most common sites of recurrent disease at initial diagnosis. The most common sites of metastatic disease throughout a patient’s disease course were lung/pleura (85%) and liver (80%). A substantial proportion of patients (65%) developed central nervous system (CNS) metastases, with only one patient presenting with CNS disease at time of metastatic diagnosis.

Table 2.

Characteristics of recurrent disease and metastatic therapy before T-DM1

Characteristic	n (%)
N	20
Stage IV at first recurrence	15 (75)
Site(s) of initial recurrence/metastasis
Breast/chest wall	8 (40)
Lung/pleura	7 (35)
Liver	7 (35)
Lymph nodes	6 (30)
Bone	4 (20)
CNS	1 (5)
Sites of initial and subsequent recurrence/metastasis
Lung/pleura	17 (85)
Liver	16 (80)
CNS	13 (65)
Bone	12 (60)
Lymph nodes	12 (60)
Breast/chest wall	8 (40)
Median number of metastatic regimens before T-DM1 (range)	7 (3–14)
Median number of metastatic trastuzumab-based regimens before T-DM1 (range)	4 (2–11)
Exposure to lapatinib-based metastatic regimen before T-DM1	8 (40)

T-DM1, trastuzumab-MCC-DM1; CNS, central nervous system.

Patients received a median of 7 (range 3–14) metastatic regimens (hormonal-based, chemotherapy-based, anti-HER2-based, bevacizumab-based, investigational) and 4 (range 2–11) trastuzumab-based regimens before T-DM1 exposure. The majority of (17 of 20; 85%) patients were administered trastuzumab alone or in combination with chemotherapy as first-line metastatic therapy. The most common trastuzumab-based regimen in the metastatic setting was trastuzumab combined with vinorelbine (90%). Overall, 8 (40%) patients were exposed to lapatinib before T-DM1, with a median number of metastatic lapatinib-based regimens of 1 (range 0–2).

characteristics and outcomes after T-DM1 therapy

Figure 1 illustrates the breakdown of patients observed in this study. At the time of this report, 3 (15%) patients were alive and remained on a T-DM1-based regimen. Of the 20 patients who stopped T-DM1 treatment, 15 (75%) were taken off study secondary to progressive disease, whereas the remainder discontinued therapy for toxicity. The majority of these patients (15 of 20; 75%) received further therapy with or without anti-HER2 agents after concluding T-DM1; reasons for not receiving additional treatment after T-DM1 include death (three) and interruption of therapy due to physician or patient request (two).

Figure 1.

Flow chart of observed patients. T-DM1, trastuzumab-MCC-DM1.

Of the 15 patients who went on to further therapy, 4 (27%) patients were administered a drug combination that they were previously exposed to before enrollment on a T-DM1 trial (Table 3). The majority (12 of 15; 80%) of patients treated beyond T-DM1 received a regimen containing trastuzumab and/or lapatinib at some point during their course. Similarly, 10 (67%) patients were administered a trastuzumab- and/or lapatinib-based treatment as first post-T-DM1 therapy. Best response to either first- or second-subsequent line(s) of therapy after T-DM1 as evaluated by a blinded radiology review using revised RECIST 1.1 criteria is shown in Figure 2. Median interval between baseline scan and follow-up scan was 3.3 months (range 1.7–8.7 months). Five of 15 (33%) treated patients achieved a partial response to either first- or second-subsequent line(s) of therapy, including 33% (4 of 12) who received a regimen containing trastuzumab or lapatinib. Specifically, in these four patients the median time on T-DM1 was 10.8 months (range 5.4–17.6 months) where two patients came off study for progressive disease and the other two came off study for medication toxicity. Post-T-DM1 regimens in these patients included trastuzumab alone (one), trastuzumab with paclitaxel (one), trastuzumab with capecitabine (one), or lapatinib with capecitabine (one). Three patients did not have radiographic assessments. Of these three patients, all demonstrated clinically stable disease (as determined by review of clinical data) to first treatment after T-DM1; time until change to next line of therapy on first-subsequent regimen after T-DM1 was 13.8, 28.1, and 1.9 months, respectively. Median duration of therapy to first-subsequent regimen after T-DM1 was 5.5 months (Figure 3). Nine patients went on to receive a second-subsequent regimen, with a median duration of therapy to the second-subsequent treatment of 6.4 months.

Table 3.

Characteristics of the patients who received therapy after T-DM1 discontinuation

Characteristic	n (%)
N	15
First therapy after T-DM1 discontinuation
Trastuzumab alone	4 (27)
Chemotherapy with or without bevacizumab	3 (20)
Investigationala	3 (20)
Trastuzumab with chemotherapy	2 (13)
Trastuzumab with lapatinib	2 (13)
Lapatinib with chemotherapy	1 (7)
Therapy exposure at any time after T-DM1 discontinuation
Trastuzumab with chemotherapy	10 (67)
Chemotherapy with or without bevacizumab	6 (40)
Trastuzumab alone	4 (27)
Lapatinib with or without chemotherapy	4 (27)
Investigational	4 (27)
Trastuzumab with lapatinib	2 (13)

^aIncludes one patient who received trastuzumab combined with an investigational agent.

T-DM1, trastuzumab-MCC-DM1.

Figure 2.

Waterfall plot. Best response to either first- or second-subsequent line of therapy after T-DM1 by RECIST 1.1. Solid bars designate patients on trastuzumab- and/or lapatinib-based regimens. Striped bars designate patients who received non-trastuzumab- and non-lapatinib-based regimens only. Three patients did not have radiographic assessments; however, all three demonstrated clinically stable disease (as determined by review of clinical data) to first treatment after T-DM1. T-DM1, trastuzumab-MCC-DM1.

Figure 3.

Kaplan–Meier analysis of duration of therapy. Duration of therapy was defined from initiation of therapy until treatment discontinuation; for patients continuing on treatment, times were censored at date of last visit. Solid line designates the duration of therapy for the 15 patients who received a first-subsequent regimen after T-DM1. Dashed line designates the duration of therapy for the nine patients who went on to receive a second-subsequent regimen after T-DM1. T-DM1, trastuzumab-MCC-DM1.

discussion

In the first-line setting, the inclusion of HER2-targeted therapy improves response rates, progression-free survival, and overall survival in patients with HER2-positive MBC [1, 9]. Next-generation HER2-directed agents, such as T-DM1, show significant promise in patients with prior exposure to multiple lines of anti-HER2 treatment [5, 6]. To our knowledge, this is the first report of clinical outcomes to subsequent therapy after exposure to T-DM1. We describe a partial response rate of 33% to first- or second-subsequent HER2-directed regimen(s) after T-DM1. This response rate is similar to the ORR of 26%–35% that was reported in the two trials using T-DM1 as a single agent [5, 6]. Median duration of therapy in our experience was ∼6 months to first- or second-subsequent line(s) of therapy after T-DM1, which is also similar to other reports of durable responses to trastuzumab-based therapies beyond progression [7–9]. These results suggest that prior exposure to T-DM1 does not exhaust the potential of ongoing HER2-targeted therapy with trastuzumab and/or lapatinib-based therapy, even in heavily pretreated patients, who had a median of four prior trastuzumab-based regimens.

Continued HER2 therapy after progression on trastuzumab is a well-established practice [7–9]. However, to what extent heavily refractory patients can respond to further HER2-directed regimens after exposure to next-generation anti-HER2 therapy is unknown. Baselga et al. [11] reported evidence of a confirmed response in 3 of 15 (20%) patients who were given trastuzumab after exposure to the HER2 dimerization inhibitor pertuzumab. Similarly, Spazzapan et al. [12] retrospectively identified 21 patients who were retreated with trastuzumab after lapatinib therapy and observed clinical benefit (defined as complete response plus partial response plus stable disease) in 10 patients (48%), with a median time to progression of 4.5 months. In conjunction with those reports, our data suggest that administration of trastuzumab- and/or lapatinib-based regimen, after exposure to a next-generation anti-HER2 agent may be beneficial in a subpopulation of women with HER2-positive MBC.

Our case series has several limitations. First, because of the current investigational status of T-DM1, we were only able to describe outcomes among patients treated with T-DM1 at our institution. Secondly, patients selected for phase II trials of T-DM1 may differ from other patients with HER2-positive MBC and could have conceivably more indolent disease and/or tumors more sensitive to HER2-targeted therapies. As potential evidence for that point, the patients with response to subsequent HER2 therapy after T-DM1 remained on T-DM1 therapy for a median of 10.8 months (range 5.4–17.4 months). Larger experiences with multiple lines of therapy in heavily pretreated patients with HER2-positive MBC will be valuable in defining the benefits of ongoing anti-HER2 treatments among these patients. Meanwhile, our report suggests that continued HER2-directed therapy is a valuable option after T-DM1, and these preliminary results support the design of a prospective trial evaluating the efficacy of trastuzumab following conclusion of T-DM1 administration.

funding

Canadian Cancer Society; Ontario Cancer Research Network; Ligue Nationale Contre le Cancer; Sanofi-Aventis (unrestricted grants); USA National Cancer Institute (CA 31946 and 33601); Alberta Heritage Foundation for Medical Research; Alberta Cancer Foundation.

disclosure

This work was not supported by an outside source and thus there was no involvement from an outside source in any part of this study. EMO, NUL, IEK, and EPW receive research support for clinical trials from Genentech Inc. The other authors declare no conflict of interest.