ELTOP

The Breast 40 (2018) 67e75

Contents lists available at ScienceDirect

The Breast journal homepage: www.elsevier.com/brst

A randomized phase II trial of trastuzumab plus capecitabine versus lapatinib plus capecitabine in patients with HER2-positive metastatic breast cancer previously treated with trastuzumab and taxanes: WJOG6110B/ELTOP Toshimi Takano a, *, Junji Tsurutani b, Masato Takahashi c, Takeharu Yamanaka d, Kazuko Sakai e, Yoshinori Ito f, Junya Fukuoka g, Hideharu Kimura h, Hidetaka Kawabata i, Kenji Tamura j, Koji Matsumoto k, Kenjiro Aogi l, Kazuhiko Sato m, Kazuto Nishio e, Kazuhiko Nakagawa b, Toshiaki Saeki n a

Department of Medical Oncology, Toranomon Hospital, Tokyo, Japan Department of Medical Oncology, Kindai University, Osaka, Japan Department of Breast Surgery, NHO Hokkaido Cancer Center, Hokkaido, Japan d Department of Biostatistics, Yokohama City University, Kanagawa, Japan e Department of Genome Biology, Kindai University, Osaka, Japan f Department of Breast Medical Oncology, Cancer Institute Hospital, Tokyo, Japan g Department of Pathology, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan h Hematology/Respiratory Medicine, Kanazawa University Faculty of Medicine, Institute of Medical, Pharmaceutical and Health Sciences, Ishikawa, Japan i Department of Breast and Endocrine Surgery, Toranomon Hospital, Tokyo, Japan j Department of Breast and Medical Oncology, National Cancer Center Hospital, Tokyo, Japan k Department of Medical Oncology, Hyogo Cancer Center, Hyogo, Japan l Department of Breast Oncology, Shikoku Cancer Center, Ehime, Japan m Department of Breast Oncology, Tokyo-West Tokushukai Hospital, Tokyo, Japan n Department of Breast Oncology, Saitama Medical University International Medical Center, Saitama, Japan b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 December 2017 Received in revised form 15 March 2018 Available online 23 April 2018

Background: For human epidermal growth factor receptor 2 (HER2)-positive metastatic breast cancer (MBC) with progression on trastuzumab-based therapy, continuing trastuzumab beyond progression and switching to lapatinib combined with chemotherapy are both valid options. We conducted an open-label, randomized phase II trial to compare the efficacy of these strategies. Patients and methods: Women with HER2-positive MBC previously treated with trastuzumab and taxanes were randomly assigned to receive trastuzumab plus capecitabine (HX) or lapatinib plus capecitabine (LX). The primary endpoint was progression-free survival (PFS) and the secondary endpoints included overall survival (OS) and the objective response rate (ORR). To explore the predictive value of the differential benefit of anti-HER2 drugs, PIK3CA mutations were assessed using circulating tumor DNA. Results: Eighty-six patients (43 in each arm) were enrolled. The median PFS was 6.1 months in the HX arm and 7.1 months in the LX arm (hazard ratio, 0.81; 90% CI, 0.55e1.21; p ¼ 0.39); the median OS was 31.0 months in the HX arm and was not reached in the LX arm (hazard ratio, 0.58; 95% CI, 0.26e1.31; p ¼ 0.18). The ORR was 40% in the HX arm and 41% in the LX arm. PIK3CA mutations were detected in 23% of the 35 analyzed patients, and in patients without PIK3CA mutations, LX yielded relatively longer PFS and OS than HX. Conclusion: In women with HER2-positive MBC previously treated with trastuzumab and taxanes, no significant differences in PFS and OS were observed between patients treated with LX and HX. Trial registration number: UMIN000005219. © 2018 Elsevier Ltd. All rights reserved.

The results of this study were presented at San Antonio Breast Cancer Symposium; December 9, 2016; San Antonio, Texas, USA. Keywords: Lapatinib Trastuzumab Beyond progression PIK3CA Brain metastasis Circulating tumor DNA

* Corresponding author. Department of Medical Oncology, Toranomon Hospital, 2-2-2 Toranomon, Minato-ku, Tokyo 105-8470, Japan. E-mail address: [email protected] (T. Takano). https://doi.org/10.1016/j.breast.2018.04.010 0960-9776/© 2018 Elsevier Ltd. All rights reserved.

68

T. Takano et al. / The Breast 40 (2018) 67e75

1Introduction Since the benefit of adding trastuzumab, an anti-human epidermal growth factor receptor 2 (HER2) monoclonal antibody, to first-line chemotherapy was shown [1], HER2-targeted therapy has been widely used in HER2-positive metastatic breast cancer (MBC). At present, trastuzumab, pertuzumab, and taxane for firstline treatment [2] and trastuzumab emtansine (T-DM1) for second-line treatment [3] are strongly recommended in American Society of Clinical Oncology (ASCO) Clinical Practice Guideline [4] and European School of Oncology (ESO) and European Society of Medical Oncology (ESMO) International Consensus Guidelines for Advanced Breast Cancer [5,6]. The ASCO Guideline also states that there are limited data on the first-line treatment with trastuzumab, pertuzumab, and taxane in patients who have received adjuvant or neoadjuvant trastuzumab [4]. In patients whose disease has progressed on trastuzumab-based therapy, continuing trastuzumab beyond progression and switching to lapatinib, a HER2 tyrosine kinase inhibitor, in combination with chemotherapy are both valid options. The GBG 26/BIG 03-05 trial compared trastuzumab plus capecitabine (HX) and capecitabine alone (X) after first-line trastuzumab-based chemotherapy and showed that HX was superior to X in terms of time to progression (TTP) [7]. The EGF100151 trial compared lapatinib plus capecitabine (LX) and X in patients previously treated with anthracyclines, taxanes, and trastuzumab and showed that LX was superior to X in terms of TTP [8]. We have 2 strategies, which are trastuzumab beyond progression or switching to lapatinib; however, it is unclear which strategy is more effective and how we can select the correct strategy for each patient. Another concern is which systemic treatment is useful for the treatment and prevention of brain metastases. In HER2-positive MBC, brain metastases often develop even when systemic diseases other than brain metastases are controlled by trastuzumabbased treatment. Some studies suggested that lapatinib monotherapy or LX were effective for brain metastases in patients with HER2-positive MBC [9,10]. PIK3CA mutations, which activate the PI3K/AKT/mTOR pathway downstream of HER2, are major mechanisms of resistance to antiHER2 drugs [11] and are candidate predictors of the differential benefit from anti-HER2 drugs especially after progression on trastuzumab. To determine the gene profiles, analyses of circulating tumor DNA (ctDNA) in the peripheral blood, so-called liquid biopsy, are useful. We conducted a randomized phase II trial of HX versus LX in women with HER2-positive MBC whose disease progressed on trastuzumab, including those with brain metastases. We also evaluated the mutational status of PIK3CA using archival tumor tissues and ctDNA. 2. Patients and methods 2.1. Study design and treatment The West Japan Oncology Group (WJOG) 6110B/Early switch to Lapatinib versus Trastuzumab beyond Progression (ELTOP) study is an open-label, multicenter, randomized phase II trial to comparatively evaluate the efficacy and safety of HX or LX in women with HER2-positive MBC who were previously treated with taxanes and progressed on trastuzumab-containing regimens. Randomization was stratified by institution, hormone receptor status, number of previous chemotherapy regimens used for metastatic disease (0 or 1 versus 2), and the presence of brain metastasis. Patients received trastuzumab (4 mg/kg loading then 2 mg/kg weekly or 8 mg/kg loading then 6 mg/kg every 3 weeks) and

capecitabine (2500 mg/m2/day on days 1e14 every 3 weeks) in the HX arm and lapatinib (1250 mg/day) and capecitabine (2000 mg/ m2/day on days 1e14 every 3 weeks) in the LX arm until progression or intolerable toxicity. The study protocol was approved by the Institutional Review Board of each institution. Written informed consent was provided by all participants prior to inclusion in the trial. 2.2. Patients Eligible patients were women aged 20 years or older with HER2positive MBC or unresectable locally advanced breast cancer who were previously treated with taxanes, with progression on trastuzumab-containing regimens. HER2 positivity was defined as 3 þ staining by immunohistochemistry or HER2 gene amplification (HER2:CEP17 signal ratio of 2.0 or more) by in situ hybridization. Patients treated with more than 2 chemotherapy regimens for MBC were excluded. Eligible patients had an Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 0e2 and adequate bone marrow, cardiac, hepatic, and renal function. Patients with brain metastases were included if they were asymptomatic. 2.3. Endpoints The primary endpoint was progression-free survival (PFS) and the secondary endpoints included overall survival (OS), the objective response rate (ORR), the disease control rate (DCR), the proportion of patients with brain metastases as the site of first progression, and safety. Tumor response and progression were assessed using Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. Chest/abdomen CT was performed at baseline and every 6 weeks. Brain MRI or CT was performed at baseline and every 6 weeks in patients with brain metastases and every 12 weeks in patients without brain metastases. 2.4. Analyses of PIK3CA mutations Archival tumor tissues of primary lesions or metastases and plasma samples at enrollment were collected from all patients who gave their consent. DNA/RNA extraction from the formalin-fixed paraffin-embedded (FFPE) tumor tissues was performed using an Allprep DNA/RNA FFPE kit (Qiagen, Valencia, CA), according to the manufacturer's instructions. ctDNA was purified using a QIAamp Circulating Nucleic Acid Kit (Qiagen) in accordance with the manufacturer's instructions. PIK3CA mutations were measured using the QX100 Droplet Digital PCR System in accordance with the manufacturer's instructions (Bio-Rad, Hercules, CA). The primers and probes for detecting the PIK3CA mutations E542K, E545K, and H1047R were purchased from Bio-Rad. Polymerase chain reaction (PCR) was performed using the following cycling conditions: 95  C for 10 min, 40 cycles of 94  C for 30 s and 55  C for 60 s, and enzyme deactivation at 98  C for 10 min. The digital PCR data were analyzed using the QuantaSoft analytical software package (Bio-Rad). 2.5. Statistical analyses This randomized phase II study was designed to compare the efficacy of LX as the experimental arm with that of HX as the control arm using PFS as the primary endpoint. The planned sample size was initially 170 randomized patients, to achieve 151 events based on the expected hazard ratio (HR) of 0.667, a one-sided significance level (a) of 0.05, and power of 0.80. However, because of slow accrual, the protocol was amended in February 2014 to decrease the planned sample size to 110, to achieve 93 events based on one-

T. Takano et al. / The Breast 40 (2018) 67e75

sided a of 0.10 and power of 0.70. Ultimately, enrollment was terminated with the enrollment of 86 patients in December 2014. Statistical analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC).

3. Results 3.1. Patients Between May 2011 and December 2014, 86 patients (43 in the HX arm and 43 in the LX arm) from 28 institutions were enrolled in this study. All patients received the study treatment and were included in the efficacy and safety analyses (Fig. 1). Patient characteristics were balanced between the 2 arms, as shown in Table 1.

3.2. Efficacy The median follow-up time was 44.6 months. The median PFS was 6.1 months in the HX arm and 7.1 months in the LX arm (stratified HR, 0.81; 90% confidence interval [CI], 0.55e1.21; p ¼ 0.39; Fig. 2A). The median OS was 31.0 months in the HX arm and was not reached in the LX arm (stratified HR, 0.58; 95% CI, 0.26e1.31; p ¼ 0.18; Fig. 2B). The ORR and DCR were evaluated in 77 patients (90%) with measurable lesions; the ORR was 40% (16/40) in the HX arm and 41% (15/37) in the LX arm (p ¼ 1.00), and the DCR was 73% (29/40) in the HX arm and 92% (34/37) in the LX arm (p ¼ 0.038), The proportion of patients with brain metastases as the site of first progression was 5% (2/43) in the HX arm and 5% (2/43) in the LX arm. The subgroup analysis of PFS according to the baseline clinical characteristics showed similar results across all subgroups, except for the duration of prior systemic treatment for MBC (Fig. 3). The PFS benefit in the LX arm compared with the HX arm was significantly larger if the duration was less than 1 year (interaction p ¼ 0.007; Fig. 4A and B). This result indicated that patients whose disease had progressed on trastuzumab-based therapy within one year benefited more from LX than from HX.

69

Table 1 Patient characteristics. HX n ¼ 43 Age (years), median (range) 57 (34e81) ECOG performance status, n (%) 0 23 (53%) 1 18 (42%) 2 2 (5%) Hormone receptor, n (%) a Positive 27 (63%) Negative 16 (37%) Stage, n (%) Recurrence after primary surgery 26 (60%) Stage IV 16 (37%) Stage IIIB/IIIC 1 (2%) (Neo-)adjuvant chemotherapy, n (%) Yes 22 (51%) No 21 (49%) (Neo-)adjuvant trastuzumab, n (%) Yes 14 (33%) No 29 (67%) Prior chemotherapy regimens for MBC, n (%) 0 2 (5%) 1 31 (72%) 2 10 (23%) Duration of prior systemic treatment for MBC, n (%) <1 year 13 (30%) 1 year 29 (67%) Unknown 1 (2%) Prior treatment with pertuzumab, n (%) Yes 2 (5%) No 41 (95%) Prior treatment with trastuzumab emtansine, n (%) Yes 1 (2%) No 42 (98%) Brain metastasis, n (%) Yes 6 (14%) No 37 (86%)

Allocated to HX ITT population (n = 43)

Allocated to LX ITT population (n = 43)

Received HX Safety population (n = 43)

Received LX Safety population (n = 43)

(n = 42) (n = 36) (n = 5) (n = 1) (n = 1)

31 (72%) 12 (28%) 0 (0%) 27 (63%) 16 (37%) 22 (51%) 20 (47%) 1 (2%) 17 (40%) 26 (60%) 6 (14%) 37 (86%) 3 (7%) 30 (70%) 10 (23%) 13 (30%) 30 (70%) 0 2 (5%) 41 (95%) 1 (2%) 42 (98%) 7 (16%) 36 (84%)

AbbreviationsHX, trastuzumab plus capecitabine; LX, lapatinib plus capecitabine; ECOG, Eastern Cooperative Oncology Group; MBC, metastatic breast cancer. a Hormone receptor positivity was defined as expression of estrogen receptor and/or progesterone receptor in 1% of tumor cells.

Patients randomly assigned (n = 86)

Discontinued study treatment Progressive disease Adverse event Patient’s request Continued study treatment

LX n ¼ 43 59 (37e78)

Discontinued study treatment Progressive disease Adverse event Patient’s request

(n = 43) (n = 29) (n = 12) (n = 2)

Fig. 1. CONSORT diagram. Abbreviation: HX, trastuzumab plus capecitabine; LX, lapatinib plus capecitabine; ITT, intent-to-treat.

T. Takano et al. / The Breast 40 (2018) 67e75

Progression-Free Survival (%)

70

100

Median PFS (months) 80

60

LX

7.1

HX

6.1

Hazard ratio 0.81 (90% CI, 0.55–1.21) Log-rank p = 0.39

40

20

0 0

Number at risk LX 43 HX 43

3

6

9

12

15

18

21

24

27

33

36

39

42

45

1 3

1 2

0 2

0 2

0 2

0 0

Time (months) 38 31

25 22

14 10

8 6

7 4

7 3

3 3

2 3

1 3

Median OS (months)

100

Overall Survival (%)

30

80

LX

Not reached

HX

31.0

60

40

Hazard ratio 0.58 (95% CI, 0.26–1.31) Log-rank p = 0.18

20

0 0 Number at risk LX 43 HX 43

6

12

18

24

30

36

42

48

54

8 7

5 5

1 1

0 0

Time (months) 42 40

42 34

33 27

26 18

18 11

Fig. 2. A. Progression-free survival in the intention-to-treat population; Abbreviation: PFS, progression-free survival; LX, lapatinib plus capecitabine; HX, trastuzumab plus capecitabine. B. Overall survival in the intention-to-treat population; Abbreviation: OS, overall survival; LX, lapatinib plus capecitabine; HX, trastuzumab plus capecitabine.

3.3. Treatment exposure and safety The median duration of the study treatment was 5.3 months in the HX arm and 6.2 months in the LX arm. The relative dose intensity during the first 12 weeks of study treatment was 96.3% for trastuzumab and 80.1% for capecitabine in the HX arm and 89.0% for lapatinib and 84.1% for capecitabine in the LX arm. Adverse events are listed in Table 2. Palmar-plantar erythrodysesthesia syndrome was the most common grade 3 adverse events in both arms. Diarrhea, rash, paronychia, and increased blood bilirubin were observed more in the LX arm. Five patients (12%) in the HX arm and 12 patients (28%) in the LX arm discontinued the study treatment because of adverse events. No

treatment-related deaths were observed. With regard to cardiac events, grade 3 left ventricular systolic dysfunction resulting from disease-related malignant pericardial effusion was observed in 1 patient in the HX arm, and grade 4 myocardial infarction and subsequent grade 4 heart failure was observed in 1 patient in the LX arm. These 2 patients recovered after appropriate treatment. No other cardiac adverse events were observed. 3.4. Post-study therapies Most patients received post-study anticancer therapies, including anti-HER2 drugs, cytotoxic chemotherapy, and endocrine

T. Takano et al. / The Breast 40 (2018) 67e75

No. of patients All patients Age

86 38 48

0

54 32

Hormone receptor Positive Negative Stage

54 32 48 38

Stage IIIB/IIIC/IV (NeoYes No (NeoYes No

39 47 20 66 MBC

0 2

1

66 20

<1 1

26 59

Yes No

13 73

71

HR (95% CI) 0.83 (0.54–1.29) p = 0.58 0.73 (0.37–1.42) 0.97 (0.54–1.75) p = 0.32 1.02 (0.58–1.78) 0.65 (0.30–1.41) p = 0.74 0.80 (0.46–1.37) 0.92 (0.43–1.93) p = 0.56 0.93 (0.52–1.66) 0.66 (0.33–1.31) p = 0.88 0.81 (0.43–1.55) 0.82 (0.45–1.51) p = 0.96 0.86 (0.32–2.30) 0.82 (0.50–1.36) p = 0.46 0.78 (0.47–1.29) 1.01 (0.39–2.60) p = 0.007 0.24 (0.087–0.65) 1.11 (0.65–1.88) p = 0.59 0.62 (0.18–2.17) 0.87 (0.54–1.40)

0.0

0.1

0.5

HR

1.0

2.0

Fig. 3. Subgroup analysis of progression-free survival; Abbreviation: HR, hazard ratio; LX, lapatinib plus capecitabine; HX, trastuzumab plus capecitabine; MBC, metastatic breast cancer. P values are for interaction.

therapy (Table 3). Patients in the LX arm received more lines of treatment after the study than those in the HX arm. In particular, more patients in the LX arm received T-DM1 after the study (42% in the HX arm and 73% in the LX arm).

yielded relatively longer PFS (p ¼ 0.035) and OS (p ¼ 0.025) than HX.

3.5. Analyses of PIK3CA mutations

We conducted this study to compare the efficacy of “continuing trastuzumab beyond progression” and “switching to lapatinib” in patients with HER2-positive MBC previously treated with trastuzumab and taxanes. This study was started in 2011, before pertuzumab and T-DM1 became available in clinical practice. Since then, standard first-line and second-line treatments for HER2-positive MBC have drastically changed. However, even after first-line treatment with trastuzumab, pertuzumab, and taxane and second-line treatment with T-DM1, the strategy to select either trastuzumab or lapatinib remains unclear. The ASCO Clinical Practice Guideline also states that options include LX and trastuzumab plus chemotherapy in this setting, and that there is insufficient evidence to recommend one regimen over another [4]. This study showed no significant differences in PFS and OS between patients treated with LX and HX. Although the HRs for PFS and OS favored the LX arm (0.81 and 0.58, respectively), these results were inconclusive due to small sample size. Because DCR was higher in the LX arm (92% vs. 73%; p ¼ 0.038) while ORR was not

Among the 86 patients in this study, 50 participated in the accompanying biomarker study. FFPE tissue samples and plasma samples were collected from 41 to 35 patients, respectively, and were evaluable for PIK3CA mutational status in 39 and 35 patients, respectively. PIK3CA mutations were detected in 33% (13/39 patients; H1047R, 10; E542K, 2; E545K, 1) of the tissue samples and in 23% (8/35; H1047R, 8) of the plasma-derived ctDNA samples. Both tissue and ctDNA samples were evaluable in 26 patients, and the concordance of PIK3CA mutations was 85%, with a sensitivity of 60% and specificity of 100%. Discordance was seen in 4 patients, in whom PIK3CA mutations (H1047R, 3; E545K, 1) were detected only in tissue samples. PFS and OS according to the treatment arms and the PIK3CA mutational status evaluated in ctDNA samples are shown in Fig. 5A and B, respectively. PFS and OS were relatively shorter in patients with PIK3CA mutations, irrespective of the treatment arm; however, in patients without PIK3CA mutations, LX

4. Discussion

T. Takano et al. / The Breast 40 (2018) 67e75

Progression-Free Survival (%)

72

100

Median PFS (months) 80

60

LX

8.1

HX

3.7

Hazard ratio 0.24 (95% CI, 0.087–0.65)

40

20

0 0

3

Number at risk LX 13 HX 13

6

9

12

15

18

21

24

27

30

33

36

39

42

45

0 0

0 0

0 0

0 0

0 0

0 0

Time (months) 11 8

8 2

5 0

2 0

2 0

2 0

0 0

0 0

0 0

Progression-Free Survival (%)

100

Median PFS (months) 80

60

LX

7.0

HX

7.2

Hazard ratio 1.11 (95% CI, 0.65–1.88)

40

20

0 0

Number at risk LX 30 HX 29

3

6

9

12

15

18

21

24

27

30

33

36

39

42

45

1 3

1 2

0 2

0 2

0 2

0 0

Time (months) 27 22

17 20

9 10

6 6

5 4

5 3

3 3

2 3

1 3

Fig. 4. A. Progression-free survival in patients treated with prior systemic treatment for metastatic breast cancer for less than 1 year. Abbreviation: PFS, progression-free survival; LX, lapatinib plus capecitabine; HX, trastuzumab plus capecitabine. B. Progression-free survival in patients treated with prior systemic treatment for metastatic breast cancer for 1 year or more. Abbreviation: PFS, progression-free survival; LX, lapatinib plus capecitabine; HX, trastuzumab plus capecitabine.

different (40% vs. 41%; p ¼ 1.00), LX may be more effective in terms of disease stabilization rather than tumor shrinkage. As to toxicities, both treatments had manageable safety profiles, while diarrhea, rash, paronychia, and increased blood bilirubin were observed more in the LX arm. Although both strategies, continuing trastuzumab beyond progression and switching to lapatinib, remain standard options, we should choose between the 2 strategies in consideration of such toxicities. The combination of trastuzumab plus chemotherapy is generally effective for HER2-positive MBC; however, most diseases will progress at some point because of intrinsic or acquired resistance to the treatment. A combined regimen of continued trastuzumab with another chemotherapy or switch to lapatinib are options to

overcome the resistance, and understanding the mechanisms of resistance can enable selection of more effective options. Subgroup analysis in this study indicated that disease progression on prior trastuzumab-based therapy for MBC within one year predicted a PFS benefit in the LX arm compared with the HX arm. This suggests that continuing trastuzumab is not effective and switching to lapatinib is a better option in patients with intrinsic or early acquired resistance to trastuzumab. We analyzed PIK3CA mutations in this study. High concordance (85%) was observed in the PIK3CA mutational status between archival tissue samples, mainly obtained at breast surgery, and ctDNA samples obtained at enrollment in this study. This indicates the reliability of liquid biopsy and the persistent existence of

T. Takano et al. / The Breast 40 (2018) 67e75

73

Table 2 Adverse events of any grade observed in 20% of patients in either arm and/or grade 3 adverse events observed in 3 patients in total. HX (n ¼ 43)

Adverse Events, n (%)

Diarrhea Palmar-plantar erythrodysesthesia syndrome Rash acneiform Rash maculopapular Paronychia Malaise White blood cell decreased Neutropenia Anemia Hypoalbuminemia Aspartate aminotransferase increased Alanine aminotransferase increased Alkaline phosphatase increased Blood bilirubin increased Hypokalemia Hyponatremia

LX (n ¼ 43)

Any Grade

Grade  3

Any Grade

Grade  3

21 (49%) 34 (79%) 2 (5%) 6 (14%) 1 (2%) 3 (7%) 22 (51%) 25 (58%) 14 (33%) 19 (44%) 19 (44%) 16 (37%) 10 (23%) 17 (40%) 14 (33%) 11 (26%)

4 9 0 0 0 3 0 2 0 0 1 2 2 1 4 2

38 (88%) 37 (86%) 17 (40%) 11 (26%) 11 (26%) 3 (7%) 21 (49%) 21 (49%) 19 (44%) 18 (42%) 24 (56%) 21 (49%) 11 (26%) 31 (72%) 11 (26%) 9 (21%)

7 9 3 0 4 1 1 1 0 0 3 1 0 1 2 0

(9%) (21%)

(7%) (5%)

(2%) (5%) (5%) (2%) (9%) (5%)

(16%) (21%) (7%) (9%) (2%) (2%) (2%)

(7%) (2%) (2%) (5%)

AbbreviationsHX, trastuzumab plus capecitabine; LX, lapatinib plus capecitabine.

Table 3 Post-study therapies.

Anti-HER2 drugs, n (%) Trastuzumab Pertuzumab Trastuzumab emtansine Lapatinib Cytotoxic chemotherapy, n (%) Any Taxanes Anthracycline-based chemotherapy Oral fluoropyrimidines Eribulin Vinorelbine Endocrine therapy, n (%) Any

HX (n ¼ 43)

LX (n ¼ 43)

37 16 18 10

37 17 27 11

(86%) (37%) (42%) (23%)

(86%) (40%) (63%) (26%)

33 (77%) 12 (28%) 2 (5%) 13 (30%) 9 (21%) 14 (33%)

35 (81%) 14 (33%) 5 (12%) 16 (37%) 12 (28%) 12 (28%)

7 (16%)

10 (23%)

Abbreviations: HX, trastuzumab plus capecitabine; LX, lapatinib plus capecitabine.

PIK3CA mutations over time in a certain proportion of HER2positive MBC cases. Because the newly identified PIK3CA mutations were not detected in ctDNA samples in this study, it is unclear whether acquired mutations of PIK3CA caused the resistance to trastuzumab; however, we think that liquid biopsy is useful to determine the tumor activity and the resistance status in a timely manner because it is less invasive, can be performed repeatedly, and it may reflect the tumor burden and sub-clonal selection under the pressure of treatment. This study showed that the PIK3CA mutations detected in ctDNA samples were associated with shorter PFS and OS in both arms, while in patients without PIK3CA mutations, PFS and OS were relatively longer in the LX arm. This suggests that both trastuzumab and lapatinib cannot overcome the resistance caused by PIK3CA mutations; however, lapatinib may overcome the resistance caused by other mechanisms such as truncated p95HER2 isoform with constitutive kinase activity [12] and loss of PTEN [13]. The CEREBEL study also evaluated HX and LX and showed that PFS with LX was inferior to that with HX in trastuzumab-naïve patients, and that similar results were observed between LX and HX in patients previously treated with trastuzumab [14]. Together with the results of our study, it can be stated that both lapatinib- and trastuzumab-based therapies are standard options in patients previously treated with trastuzumab; however, only trastuzumabbased therapy is a standard option in trastuzumab-naïve patients. The CEREBEL study evaluated the incidence of central nervous

system (CNS) metastases as first site of relapse as the primary endpoint; no differences for this endpoint were observed between LX and HX [14]. Our study included patients who had brain metastases at baseline and evaluated the proportion of patients with brain metastases as the site of first progression; the proportion was similar between LX and HX, and the PFS benefit was not different between patients with and without brain metastases at baseline. Thus, it is still unclear whether we should use lapatinib preferentially in patients with brain metastases. This study has several limitations. First, because this is a smallsized randomized phase II trial and was closed before full accrual, the results are not conclusive. Second, because this study was started before pertuzumab and T-DM1 became available in clinical practice, only a few patients were previously treated with pertuzumab and T-DM1. This is different from the current situation, and some caution is needed in the extrapolation of the results for patients previously treated with pertuzumab and T-DM1. Third, many patients were treated with pertuzumab and/or T-DM1 after this study, which may have affected OS. Though OS numerically favored LX in this study, this is possibly because T-DM1 was used as poststudy therapy more frequently in the LX arm.

5. Conclusions Both switching to lapatinib and continuing trastuzumab in combination with capecitabine are still standard options in women with HER2-positive MBC previously treated with trastuzumab and taxanes. This study suggested that LX might be more useful in patients whose disease had progressed on trastuzumab-based therapy within one year and in those without PIK3CA mutations detected using liquid biopsy.

Ethical approval The study protocol was approved by the Institutional Review Board of each institution. Written informed consent was provided by all participants prior to inclusion in the trial.

Funding This work was financially supported by Novartis Pharmaceutical Japan and GlaxoSmithKline.

74

T. Takano et al. / The Breast 40 (2018) 67e75

Progression-Free Survival (%)

100 median PFS (months) LX, PIK3CA wild type 8.2 HX, PIK3CA wild type 4.9 LX, PIK3CA mutant 4.1 HX, PIK3CA mutant 6.1

80

60

Hazard ratio (PIK3CA wild type) 0.38 (95% CI, 0.16–0.93; p = 0.035) Hazard ratio (PIK3CA mutant) 0.60 (95% CI, 0.11–3.13; p = 0.54)

40

20

0

0

3

Number at risk LX, wild type 13 13 HX, wild type 14 11 LX, mutant 3 3 HX, mutant 5 3

6

9

12 15 18 21 24 27 30 33 36 39 42 45

Time (months) 9 6 1 3

5 4 1 0

5 1 0 0

5 0 0 0

5 0 0 0

2 0 0 0

2 0 0 0

1 0 0 0

1 0 0 0

1 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

0 0 0 0

Overall Survival (%)

100

80

median OS (months) LX, PIK3CA wild type Not reached HX, PIK3CA wild type 18.7 LX, PIK3CA mutant 15.4 HX, PIK3CA mutant 15.7

60

40

Hazard ratio (PIK3CA wild type) 0.088 (95% CI, 0.011–0.73; p = 0.025) Hazard ratio (PIK3CA mutant) 1.20 (95% CI, 0.20–7.32; p = 0.84)

20

0

0

Number at risk LX, wild type 13 HX, wild type 14 3 LX, mutant 5 HX, mutant

6

12

18

24

30

36

42

48

2 1 0 0

1 0 0 0

0 0 0 0

Time (months) 13 12 3 4

13 9 3 4

12 6 1 2

11 4 0 1

6 2 0 0

Fig. 5. A. Progression-free survival by treatment arms and PIK3CA mutational status evaluated in circulating tumor DNA samples. Abbreviation: PFS, progression-free survival; LX, lapatinib plus capecitabine; HX, trastuzumab plus capecitabine. B. Overall survival by treatment arms and PIK3CA mutational status evaluated in circulating tumor DNA samples. Abbreviation: OS, overall survival; LX, lapatinib plus capecitabine; HX, trastuzumab plus capecitabine.

Conflicts of interest TT reports grants and personal fees from Novartis, Chugai, and Taiho, personal fees from Daiichi-Sankyo, Kyowa Hakko Kirin, and Eisai, grants from Takeda, Ono, MSD, Merck Serono, outside the submitted work; JT reports personal fees from Novartis, Kyowa

Hakko Kirin, Taiho, Chugai, Eisai, Asahi Kasei Pharma, and DaiichiSankyo, outside the submitted work; MT reports personal fees from Eisai, and AstraZeneca, outside the submitted work; TY reports personal fees from Takeda, Chugai, Taiho, and Boehringer Ingelheim, outside the submitted work; YI reports grants from Novartis, Chugai, Parexel, Eisai, Taiho, EPS, Daiichi-Sankyo, Sanofi, MSD, and

T. Takano et al. / The Breast 40 (2018) 67e75

AstraZeneca, outside the submitted work; JF owns stock in Pathology Institute Corp; KT reports grants from Novartis and Chugai, outside the submitted work; KM reports grants from Novartis, AstraZeneca, AbbVie, Ono, MSD, and Boehringer Ingelheim, outside the submitted work; KA reports personal fees from AstraZeneca, Eisai, Ono, Otsuka, Sanofi, Daiichi-Sankyo, Taiho, and Chugai, outside the submitted work; K. Nakagawa reports grants and personal fees from Novartis, Astellas, AstraZeneca, Pfizer, Chugai, Eli Lilly, MSD, EPS, Ono, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi-Sankyo, Taiho, Kyowa Hakko Kirin, personal fees from Showa Yakuhin Kako, Kissei, Ayumi, and Sym Bio, grants from GlaxoSmithKline, Yakult, Parexel, Otsuka, AbbVie, AC medical, Merck Serono, Quintiles, Japan Clinical Research Operations, Eisai, PPD-SNBL, Takeda, and Covance, outside the submitted work; TS reports personal fees from Novartis and Taiho, outside the submitted work; All remaining authors have nothing to disclose. Acknowledgements We thank the data managers and other support staff of West Japan Oncology Group, especially Dr Shinichiro Nakamura, M.D. and Kaori Mori. References [1] Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer overexpresses HER2. N Engl J Med 2001;344:783e92.
s J, Kim SB, Im SA, Hegg R, Im YH, et al. Pertuzumab plus [2] Baselga J, Corte trastuzumab plus docetaxel for metastatic breast cancer. N Engl J Med 2012;366:109e19. [3] Verma S, Miles D, Gianni L, Krop IE, Welslau M, Baselga J, et al. Trastuzumab emtansine for HER2-positive advanced breast cancer. N Engl J Med 2012;367: 1783e91.

75

[4] Giordano SH, Temin S, Kirshner JJ, Chandarlapaty S, Crews JR, Davidson NE, et al. Systemic therapy for patients with advanced human epidermal growth factor receptor 2epositive breast cancer: American Society of Clinical Oncology Clinical Practice Guideline. J Clin Oncol 2014;32:2078e99.
F, Barrios CH, et al. 3rd [5] Cardoso F, Costa A, Senkus E, Aapro M, Andre ESOeESMO international consensus guidelines for Advanced Breast Cancer (ABC 3). Breast 2017;31:244e59.
F, Barrios CH, et al. 3rd [6] Cardoso F, Costa A, Senkus E, Aapro M, Andre ESOeESMO international consensus guidelines for Advanced Breast Cancer (ABC 3). Ann Oncol 2017;28:16e33. [7] von Minckwitz G, du Bois A, Schmidt M, Maass N, Cufer T, de Jongh FE, et al. Trastuzumab beyond progression in human epidermal growth factor receptor 2epositive advanced breast cancer: a German Breast Group 26/Breast International Group 03-05 Study. J Clin Oncol 2009;27:1999e2006. [8] Geyer CE, Forster J, Lindquist D, Chan Romieu CG, Pienkowski T, et al. Lapatinib plus capecitabine for HER2 positive advanced breast cancer. N Engl J Med 2006;355:2733e43.
ras V, Paul D, Lossignol D, Christodoulou C, Stemmler HJ, et al. [9] Lin NU, Die Multicenter phase II study of lapatinib in patients with brain metastases from HER2-positive breast cancer. Clin Cancer Res 2009;15:1452e9.
ras V, Cropet C, Dalenc F, et al. Lapa[10] Bachelot T, Romieu G, Campone M, Die tinib plus capecitabine in patients with previously untreated brain metastases from HER2-positive metastatic breast cancer (LANDSCAPE): a single-group phase 2 study. Lancet Oncol 2013;14:64e71. [11] Berns K, Horlings HM, Hennessy B, Madiredjo M, Hijmans EM, Beelen K, et al. A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast cancer. Canc Cell 2007;12: 395e402. ~ a A, Anido J, Guzman M, Cortes J, et al. Expression of [12] Scaltriti M, Rojo F, Ocan p95HER2, a truncated form of the HER2 receptor, and response to anti-HER2 therapies in breast cancer. J Natl Cancer Inst 2007;99:628e38. [13] Xia W, Husain I, Liu L, Bacus S, Saini S, Spohn J, et al. Lapatinib antitumor activity is not dependent upon phosphatase and tensin homologue deleted on chromosome 10 in ErbB2-overexpressing breast cancers. Cancer Res 2007;67: 1170e5. _ [14] Pivot X, Manikhas A, Zurawski B, Chmielowska E, Karaszewska B, Allerton R, et al. CEREBEL (EGF111438): a phase III, randomized, open-label study of lapatinib plus capecitabine versus trastuzumab plus capecitabine in patients with human epidermal growth factor receptor 2epositive metastatic breast cancer. J Clin Oncol 2015;33:1564e73.