Efficacy and safety of trastuzumab, lapatinib, and paclitaxel neoadjuvant treatment with or without prolonged exposure to anti-HER2 therapy, and with or without hormone therapy for HER2-positive primary breast cancer: a randomised, five-arm, multicentre, open-label phase II trial

Trastuzumab has been found to prolong survival and reduce relapse in human epidermal growth factor receptor 2 (HER2)+ patients with advanced and early stage breast cancer when used in combination with other chemotherapy agents and in a neoadjuvant and adjuvant setting [1‐4]. In the sequential neoadjuvant and adjuvant settings, the NOAH study compared the treatment with and without trastuzumab in women with HER2+ breast cancer treated with a neoadjuvant chemotherapy regimen [4]. The 3-year event-free survival was higher in the trastuzumab group than the non-trastuzumab group. When comparing the trastuzumab group with the HER2+ chemotherapy alone group, a 41% risk reduction of recurrence, progression, or death by trastuzumab treatment was shown. Similarly, the achievement of higher pathological complete response (pCR) rates with chemotherapy plus trastuzumab has improved survival outcomes in the neoadjuvant setting [5].

Lapatinib, a dual, reversible HER2/epidermal growth factor receptor (HER1) tyrosine kinase inhibitor, both alone and in combination with chemotherapy, has significantly improved progression-free survival in metastatic HER2+ breast cancer patients [6]. Synergistic interactions have been recognised between lapatinib and trastuzumab, owing to their different mechanisms of action and receptor site activity. Studies have shown that combined treatment with lapatinib and trastuzumab was more effective in terms of pCR than trastuzumab alone in patients with HER2+ breast cancer [7‐9].

Results of preclinical studies suggest that cross talk between HER2 and oestrogen receptor (ER) signalling pathways in breast cancer contributes to resistance to hormonal therapy [10, 11]. Therefore, it was hypothesised that inhibiting both pathways could be more beneficial than ER or HER2 inhibition alone. In the TAnDEM study, postmenopausal women with HER2/hormone receptor-copositive metastatic breast cancer were treated with anastrozole combined with trastuzumab, without chemotherapy. This combination was found to significantly improve progression-free survival, time to progression, clinical benefit rate, and overall response rate compared with hormone therapy alone [12]. Similarly, in the EGF30008 trial, women with metastatic breast cancer that coexpressed ER and ErbB2 were treated with a combination of lapatinib and letrozole, and achieved significantly increased progression-free survival and clinical benefit rates [13]. In the TBCRC 006 trial, 66 women with locally advanced HER2+ and ER+ breast cancer were treated with neoadjuvant trastuzumab, lapatinib, and letrozole. This approach of targeted therapy without chemotherapy resulted in an overall pCR rate of 27%, and of 21% in ER+ patients [14].

The aim of the present study (JBCRG-16 [Neo-LaTH]) was to evaluate the efficacy and safety of lapatinib and trastuzumab therapy followed by lapatinib and trastuzumab plus weekly paclitaxel in Japanese patients with primary HER2+ breast cancer in a neoadjuvant setting. Additionally, we examined the effects of different periods of lapatinib and trastuzumab therapy (6 vs. 18 weeks) and the effects of add-on endocrine therapy in ER+ patients.

The characteristics of the patients are summarised in Table 1. Patients had a median age of 53.0 years (range, 26–70 years); 65% were classified as T2 and 55% as N0. Among the 117 N0 patients, 35 (29.9%) underwent pre-systemic therapy SLN biopsy. As a result, 28 patients (80.0%) were SLN-negative. Following successful surgery, 10 patients underwent SLN biopsy again. No axillary metastasis was found in any of these patients.

Regarding the primary study endpoint, CpCR was achieved in 101 (47.9%) patients. This was calculated based on 211 patients in the full analysis set who underwent surgery (one patient elected not to undergo surgery because her lesion had disappeared in response to treatment). CpCR was significantly higher in ER− patients than in ER+ patients (Group A vs. Group C, P = 0.0034, Table 2 and Supplementary Material S4). In groups A, B, C, D, and E, CpCR was achieved by 65.9, 60.4, 34.1, 33.3, and 41.0% of patients, respectively. No significant difference was observed in CpCR among the groups with different durations (6 vs. 18 weeks) of lapatinib plus trastuzumab (A vs. B, P = 0.59). CpCR in ER+ patients who received add-on endocrine therapy was not significantly greater than that in ER+ patients without endocrine therapy (C vs. D, P = 0.94).

The relative dose intensities of lapatinib during the lapatinib and trastuzumab period were over 90%, and during the lapatinib and trastuzumab plus weekly paclitaxel period were around 80% (Supplementary Material S5). Grade ≥ 3 adverse events were observed in 42.3% of patients (Supplementary Material S6); the most common were neutropenia (19%), diarrhoea (12%), skin and subcutaneous disorders, elevated alanine transaminase (5% each), and paronychia (3%). No deaths were reported. There were no significant changes in the mean left ventricular ejection fraction from baseline in any regimen.

CpCR and CpCR + ypN0 were not improved by the extended lapatinib plus trastuzumab or add-on endocrine therapy (Table 2). ORRs evaluated by magnetic resonance imaging (MRI) or computed tomography (CT) were 81.8, 81.3, 85.4, 97.4, and 92.5%; and BCRs were 63.6, 55.3, 70.7, 53.8, and 68.4% (Supplementary Material S4). However, prolongation of lapatinib plus trastuzumab resulted in improved clinical efficacy (Table 2), as shown by clinical response in the lapatinib plus trastuzumab phase evaluated by MRI/CT (ORR and clinical CR). Supplementary Material S7 shows the subpopulation treatment effect pattern plot analysis. The ORR increased linearly with increasing lapatinib dose in the lapatinib plus trastuzumab period especially in the ER+ cohort.

A similar tendency was observed in quasi-pCR [QpCR (CpCR + near pCR)]. For ER+ patients, quasi-pCR tended to improve with increasing lapatinib dose and add-on ET (Supplementary Material S8). Clinical efficacy at the last time point before surgery (i.e., at the end of weekly paclitaxel add-on period following lapatinib and trastuzumab period) is shown in Table 2. There was no significant difference among the five groups in terms of clinical efficacy. In groups B and E, the benefit from the add-on weekly paclitaxel tended to be small.

From the experimental analyses regarding the percent change in maximum tumour size from pre-treatment to end of treatment measured by MRI/CT during the time course and relationship between pathological response (Fig. 2), among ER− patients, the prolongation of the lapatinib plus trastuzumab phase seemed to be important for achieving pCR. Conversely, among ER+ patients, pCR was mainly influenced by the reduction effect through the addition of paclitaxel. This tendency was observed especially in groups C and D. In Group B, at 18 weeks of lapatinib plus trastuzumab therapy, the percent change in maximum tumour size was 25.1% (95% CI 13.5–36.8%) in patients with pCR and 64.5% (95% CI 45.7–83.4%) in patients without pCR, indicating a significant reduction in the patients with pCR. In the hormone receptor-positive groups C and D, the percent change in maximum tumour size at the end of the lapatinib plus trastuzumab plus weekly paclitaxel therapy (at the end of all the cycles) was significantly different between patients with and without pCR. In Group C, the percent change in maximum tumour size was 17.4% (95% CI 4.4–30.5%) in patients with pCR and 49.0% (95% CI 37.1–60.9%) in patients without pCR. In Group D, the ratio was 15.7% (95% CI 2.2–29.3%) in patients with pCR and 34.9% (95% CI 25.2–44.6%) in patients without pCR.

Supplementary Material S9 shows the changes in tumour diameter from baseline to the end of treatment. Among ER− patients, those who responded to treatment early in the lapatinib plus trastuzumab phase tended to achieve pCR. Among ER+ patients, there was no significant correlation between clinical efficacy and pathological efficacy.

In the Neo-LaTH study, the efficacy and safety of lapatinib and trastuzumab therapy followed by lapatinib and trastuzumab plus weekly paclitaxel in primary HER2+ breast cancer patients were equivalent to that in two previous studies [7, 8]. Regarding the effect of the extension of lapatinib plus trastuzumab treatment period from 6 to 18 weeks and addition of endocrine therapy, the pCR rates (primary endpoint) were not improved by the extended treatment. Although the primary endpoint did not match the expected result, the secondary clinical efficacy indicates that in ER− patients finally achieving pCR, combined HER2 blockade therapy alone, which was given before the initiation of the weekly paclitaxel treatment, tended to induce a greater reduction due to the prolongation of the therapy from 6 to 18 weeks. This suggests that in ER− patients who respond to combined HER2 blockade therapy, the prolongation of the treatment period can lead to the avoidance of the future use of cytotoxic anti-cancer drugs. The comparison of Groups D and E with combined hormone therapy showed that among ER+ patients, the proportion of patients with complete response or partial response tended to be greater when the treatment period was prolonged to 18 weeks. However, the difference was not as great as that observed in ER− patients. Our findings are in agreement with previous chemotherapy-free regimens (the NeoSPHERE [17] and the NeoALTTO [7] studies). Thus, the results of the chemotherapy-free phase of Neo-LaTH support the concept that dual HER2 blockage can lead to decreased length of chemotherapy regimens, and that a proportion of HER2+ tumours could potentially be eradicated with chemotherapy-free regimens.

Changes in tumour diameter tended to be larger after add-on endocrine therapy and extended treatment with lapatinib and trastuzumab in ER+ patients in the present study. Similarly, 40% of patients in the TBCRC 006 study achieved pCR by 12 weeks of dual HER2 blockage with lapatinib plus trastuzumab despite having large initial tumour size [14]. Together, these results suggest that the changes in tumour diameter increase in proportion to the extended period of molecularly targeted drug therapy. Overall clinical response rates were also significantly higher in the 18-week group rather than the 6-week group. Our findings are useful for the future exploration of chemotherapy-free anti-HER2 regimens for women who cannot receive cytotoxic drugs.

We hypothesised that the addition of endocrine therapy may increase the pCR rate in ER+ luminal HER2 subtype. Regimen D showed high overall clinical response rate (97.4%) with 31.6% of clinical CR. In the regimen-E group, it was also as high as 92.5% with 37.5% of clinical CR, indicating that the endocrine therapy combination would be able to provide high levels of clinical response. However, pCR in ER+ patients who received add-on endocrine therapy was not significantly greater than that in ER+ patients without endocrine therapy. These results were unexpected since improved pCR rates had been achieved previously in the TBCRC 006 study of women with locally advanced HER2+ and ER+ breast cancer who were treated with neoadjuvant trastuzumab, lapatinib, and letrozole [14]. To increase the pCR rate against luminal HER2 disease, further strategies need to be developed.

With regard to toxicity, the frequencies of grade 3/4 adverse events in the Neo-LaTH study were similar to those observed in other studies. The relative dose intensity of lapatinib was also similar to that in previous results. From the STEPP analysis, a relationship was revealed between the administration dose of lapatinib during the anti-HER2 therapy phase and the clinical response.

The strength of the present study is the novelty of the study design, which allowed the prolonged duration of dual HER2 blockage with lapatinib plus trastuzumab in the neoadjuvant setting. This study had several potential limitations, in particular those inherent to phase II open-label studies. However, current outcomes may be informative for future studies, and correlative translational research is ongoing to analyse molecular mechanisms of tumour responses and therapeutic resistance. Another limitation is the small sample size; however, this was determined on a statistical basis, and it is important to verify a hypothesis in a small sample size study.

In conclusion, the findings of the present study confirmed the efficacy and safety of lapatinib and trastuzumab followed by lapatinib plus trastuzumab plus weekly paclitaxel for the treatment of HER2+ invasive breast cancer in Japanese patients. These results are relevant for clinicians considering the optimal duration of preoperative anti-HER2 treatment for HER2+ disease and the combination with endocrine therapy for luminal HER2 subtype. Further analysis of disease-free survival/overall survival in conjunction with biomarker studies is necessary to explore the therapeutic impact.