Lapatinib in combination with capecitabine versus continued use of trastuzumab in breast cancer patients with trastuzumab-resistance: a real-world study

Background: The efficacy and safety of lapatinib plus capecitabine (LC or LX) versus trastuzumab plus chemotherapy in patients with HER-positive metastatic breast cancer who were resistant to trastuzumab is largely unknown. Methods: we retrospectively analyzed breast cancer patients who began the treatment with regimen of lapatinib plus capecitabine (LC or LX) or trastuzumab beyond progression (TBP) at eight hospitals between May 2010 and October 2017. Results: Among 554 patients who had developed resistance to trastuzumab, the median PFS was 6.77 months in the LX group compared with 5.6 months in the TBP group (hazard ratio 0.804; 95% CI, 0.67 to 0.96; P=0.019). The central nervous system progression rate during treatment was 5.9% in LX group and 12.5% in TBP group, respectively (P=0.018). PIK3CA and TP53 are the most common mutations in resistant patients, but do not significantly differ between primary resistant cancers and secondary resistant cancers. Conclusion: The combination of lapatinib and capecitabine has shown a prolonged PFS compared with TBP in patients who had progressed on trastuzumab.

. Clinically, HER2-positive tumors are characterized by aggressive clinical course and poor overall prognosis [4]. The introduction of the anti-HER2 monoclonal antibody trastuzumab has dramatically improved the poor prognosis of this population of patients [5][6][7]. Trastuzumab binds to the extracellular domain of the HER2 receptor, prevents receptor homo-and hetero-dimerization, therefore inhibiting the activation of downstream oncogenic signaling [8]. Trastuzumab-containing treatment is the standard approach for HER-2 positive metastatic breast cancer. Despite its overall clinical efficacy, de novo and acquired resistance to trastuzumab administration have been observed [9]. The development of distant metastasis to liver, bone, lung and brain has become a major challenge in management of patients with HER-2 positive breast cancer, probably due to longer life expectancy and acquired trastuzumab resistance [10].
Therefore, it is urgently needed to develop new strategy of salvage therapy for patients who have developed resistance to trastuzumab.
However, the consensus guidelines on targeted treatment for resistance in HER2-positive breast cancer are not available [11,12]. Combination of anti-HER2 agents with chemotherapy, anti-HER2/HER3 dimerization, or inhibitors of the downstream signaling might improve prognosis [13]. Fujimoto-Ouchi demonstrated that trastuzumab in combination with taxanes or capecitabine showed antitumor activity in trastuzumab resistant model [14]. The GBG 26/BIG 3-05 enrolled patients with HER2-positive metastatic breast cancer (Stage IV) that progresses during treatment with trastuzumab. Among these patients, 78 patients were randomly assigned to receive capecitabine, and 78 patients to capecitabine plus trastuzumab. The results showed that median TTP were 5.6 months vs 8.2 months, P = 0.0338 [15]. In a similar study, patients who received trastuzumab treatment beyond progression (TBP) had longer median OS than those who terminated (21.3 months VS 4.6 months (P 0.0001)) [16]. Taken together, these studies suggest that a clinical benefit has been observed for treatment with trastuzumab beyond progression.
Lapatinib, an orally active, small molecule, tyrosine kinase inhibitor, has shown non-cross resistance with trastuzumab. It binds reversibly to the cytoplasmic domains of both EGFR and HER2 that in turn blocks the activating signaling cascades in the MAPK and PI3K pathways [17]. Given its unique mechanistic function, lapatinib might be a suitable option in HER2-positive MBC that have become resistant to suppression by trastuzumab. Studies have also shown that the phosphorylation of p95 HER2 (a truncated version lacking the extracellular domain) and the formation of heterodimers between HER2 and other members of the HER family might be inhibited by lapatinib, but not trastuzumab [18,19].
In EGF100151 trial, lapatinib plus capecitabine reduced the hazard for time-to-disease progression (hazard ratio 0.49; 95%CI 0.34-0.71; P <0.001) in HER2-positive breast cancer that progressed on anthracycline, a taxane and trastuzumab [11,20]. In 2010, the US FDA approved the use of lapatinib in combination with capecitabine for the treatment of patients with HER2-positive MBC. In addition, lapatinib in combination with capecitabine shows excellent activity against cerebral nervous system (CNS) disease. An study suggested that patients with brain metastasis achieved significant longer overall survival in the lapatinib group compared with trastuzumab-based therapy (19.1 vs. 12 months, P = 0.039) [21]. FoundationOne next-generation sequencing (NGS). The methods was described before.
Four were primary and four were secondary resistance.

Endpoint
The primary endpoint was PFS, defined as time from the initiation of TBP or LX until the earliest date of disease progression or death. Secondary included ORR (the ratio of patients who have complete or partial tumor remission), CBR (clinical benefit rate), defined as the ratio of patients who have complete or partial tumor remission or stable disease for more than 6 months.

Statistical analysis
Statistical analyses were performed using SPSS version 24.0 (SPSS Inc, Chicago, IL, USA).

Patient characteristics
A total of 554 patents were identified and the median follow-up time was 15 months. The demographic characteristics between the two groups were showed in Table 1 (Table 2).

Efficacy
The median PFS was 6.77 months in the LX group compared with 5.6 months in the TBP group (hazard ratio 0.7955; 95% CI, 0.6632 to 0.9542; log-rank P =0.014; Figure 1A). In the primary resistant patients, the median was significantly increased from 4.3 months for TBP to 6.8 months for LX (P <0.001; Figure 1B). In the secondary resistant patients, no significant difference was observed (median PFS: 6.6 months for LX VS 6.3 months for TBP, P =0.8827; Figure 1C). The best overall response to treatment was not evaluable in 64 patients. We observed no significant difference in the ORR and CBR between two groups (P =0.822; P =0.224; eTable 1 in Supplement 1).

First line treatment
In  Figure 2). 15 patients were not evaluable in best response to treatment. The ORR was significantly increased from 8.3% for TBP to 27.6% for LX (P =0.04). The CBR was significantly improved too (36.1% to 69%, P =0.008; eTable 2 in Supplement 1).

Second and third line treatment
After developing resistance to the trastuzumab-containing treatment, 218 patients received LX, and 256 patients continued using trastuzumab in the later lines. The result showed that median PFS was 6.6 months for LX group compared with 5.9 months for TBP group (hazard ratio 0.8605; 95% CI, 0.7068 to 1.048; log-rank P =0.135; Figure 3A). No improvement in median PFS was observed. Median PFS in the primary resistant population increased from 4.3 months for TBP to 6.6 months for LX group (hazard ratio 0.5057; 95% CI, 0.335 to 0.7633; log-rank P =0.001; Figure 3B). The best response to treatment was missing in 22 patients in the second line setting. The ORR and CBR between the two groups have no statistical difference (eTable 3 in Supplement 1). In the third line setting, 27 patients were not evaluable in best response to treatment. We found no statistical difference in ORR or CBR (eTable 4 in Supplement 1).

Multivariate analysis
We carried out a multivariate analysis to investigate whether the anti-HER2 therapy effect was different according to baseline characters. The model included treatment after resistance to trastuzumab, age, hormone receptor status, metastatic sites, treatment line.
We noted that secondary or primary resistance had a differential prognostic effect in trastuzumab treated patients, the HR for PFS favoured patients who were secondary resistance. Findings from our analysis suggest that the effect of lapatinib on PFS may be explained by excellent effect in primary resistant patients ( Figure 4).  Table 4).

Safety
The most common adverse events were neutropenia, thrombocytopenia and hand-foot syndrome. 42 (17.8%) patients in the LX group and 61 (20.6%) patients in the TBP group experienced grade 3 or 4 toxicities (P =0.415). The most frequent grade III-IV AEs were diarrhoea (5.1%), hand-foot syndrome (10.2%) in LX group, and ALT/AST increased (9.1%) and neutropenia (6.4%) in TBP group. Treatment-related LVEF decline was observed in 2 patients in the trastuzumab group but was moderate in severity (Table 5). This study was retrospective by nature, the adverse events may be underestimated.

Biomarker analyse
To determine the transcriptional differences between primary and secondary resistant tumors, we analysed four pairs trastuzumab-resiatant patient metastatic samples.
The trial was terminated earlier. However, although afatinib is a second-generation, broader inhibitor of the ErbB family of proteins [29], no randomized trials to compare the efficacy of afatinib with lapatinib in women progressed during trastuzumab treatment.
Furthermore, a major difference between MA.31 trial and our study was that a large part of patients were newly diagnosed advanced breast cancer and were trastuzumab-naïve.
This might affect the survival outcome.
Lapatinib has a different mechanism of inhibition on HER2 and EGFR signaling compared with trastuzumab. Preclinical evidence suggests non-cross resistance of trastuzumab and lapatinib. PTEN abrogates phosphatidyl inositol-3-kinase (PI3K), which results in inhibition of Akt signaling. Non-existent or limited expression of PTEN (phosphatase and tensin homologue deleted on chromosome 10), might be a marker of resistance to trastuzumab [30]. Previous studies have confirmed PTEN expression has no correlation with response to lapatinib [31]. IGF-1R (insulin-like growth factor receptor) is important for cell proliferation and survival [32]. It has been reported that overexpression of IGF-1R predicted resistance to trastuzumab in breast cancer cells [32][33][34]. IGF-1R belongs to the tyrosine kinase receptor family and breast cancer cells that express IGF-1R may still be sensitive to lapatinib [35].
We try to identify subsets of patients who would derive the greatest benefit from further would preventtrastuzumab binding and are associated with poor prognosis in patients.
Lapatinib inhibits p95HER2 phosphorylation while trastuzumab doesn't [36]. That may explain why switching to lapatinib predicts for extended PFS in the primary resistant group.
Unlike primary resistant patients, a clinical benefit was observed for treatment with trastuzumab-containing regimen in patients acquired resistance to anti-HER-2 therapy.
We also found patients in the second line treatment had a higher proportion of trastuzumab beyond progression therapy than those in the third-line setting. The predominant HER-2 targeted therapy in the second line setting was trastuzumab instead of lapatinib. A plausible reason for the disparities concerns the assumption that patient refractory to prior chemotherapy agent but not to trastuzumab itself. Secondly, anti-HER2 therapy is expensive and time consuming, and varying medical insurance policies may contribute to the continued use of trastuzumab.
Breast cancer patients with HER2 overexpression have greater risk for developing brain metastasis, and trastuzumab treatment has emerged as a factor contributing to this risk [39]. Previous studies support the hypothesis that brain is a 'sanctuary' site for the development of metastases due to the limited ability of trastuzumab in penetrating the Blood Brain Barrier (BBB) [40]. Lapatinib is a small dual tyrosine-kinase inhibitor of HER1 and HER2 with a hypothetical ability to cross the BBB [41]. The combination of lapatinib with capecitabine has central nervous system (CNS) activity for the treatment of patients with HER2-positive brain metastatic breast cancer. Clinical evidence supports that patients with HER2-positive brain metastasis achieved a significant clinical benefit from lapatinib and capecitabine both as single agents and in combination in patients with brain metastases [42][43][44]. In the present study, the percentage of patients with central nervous system progression was higher in the TBP group. In addition, the comparison of CNS progression rate indicates that lapatinib is more effective in brain metastases than trastuzumab. These findings are consistent with the result of a randomized clinical trial which evaluate the effect of neratinib compared with trastuzumab in previously untreated metastatic ERBB2-Positive Breast Cancer. Neratinib, another kind of oral irreversible ERBB family blocker, was associated with fewer central nervous system recurrences (relative risk, 0.48; 95%CI, 0.29-0.79; P =0 .002) and also delayed the time to CNS relapses compared with trastuzumab (HR, 0.45; 95%CI, 0.26-0.78; P = 0.004) [45]. In EMILIA trial, modest activity of lapatinib plus capecitabine upon CNS recurrences, 2.0% (9/450) in the T-DM1 group and 0.7% (3/446) in the LX group developed new brain metastases [22,46]. It appears that switching patients with treated brain metastases to lapatinib-containing treatment more effectively prevent brain progression.
Recently, much effort has been made to develop newer anti-HER2 agents that could effectively overcome trastuzumab-resistance. Pyrotinib is an oral, irreversible tyrosine kinase inhibitor (TKI) against epidermal growth factor receptor (EGFR)/HER1, HER2, and HER4, has demonstrated an advantage over lapatinib [47,48] . In a phaseⅡstudy, the combination of pyrotinib and capecitabine was superior to LX regarding PFS with not reached vs. 7.1 months (P=0.0031) in HER2-positive metastatic breast cancer patients previously treated with trastuzumab. The multicenter, randomized phase III study is now ongoing in China. Continued attempts to develop new TKI will provide additional information on identifying the most appropriate treatment therapy in trastuzumabrefractory patients.
In this study, we focused on identifying mutations and copy number alterations that may render to response prediction to trastuzumab. Our study indicates PIK3CA and TP53 were the most requently mutated genes in both primary and primary trastuzumab-resistant patients. This consistency was probably due to the two genes were resistance biomarkers.
PIK3CA mutations were linked to significantly reduced pCR rate in patients receiving trastuzumab and lapatinib (16.7% vs 39.1%; P<0.001) [50]. Meta-analyses investigated correlation between PIK3CA mutation status and response towards HER2-targeted regimens in clinical settings. Although the power was too low, there was a trend of lower clinical benefit of anti-HER2 therapy in tumors with a PIK3CA mutation [51,52]. The controversial results of mutant PIK3CA roles in responses to anti-HER2 treatment was probably the result of the heterogeneity between specimens.
We also found most samples TMBs were low to intermediate, the two hormone negative tumors were both TMB high. Breast cancers , especially HER2 positive subtype, commonly considered to be a 'non-inflamed' cancer, seldomly respond to immune checkpoint blockade therapy. Recently, most of the focus has been on the triple negative breast cancer. It is known that TMB is a predictive factor for response of immune therapy and sex hormones involved in anticancer immunity. Men have overall larger treatment effect from immune checkpoint inhibitors than do women [53]. Sex hormones may be a main regulator. Further investigation with a larger sample size is required.
It should be noted that there were a few limitations in our study. First, it's retrospectively, potential imbalance in contributing factors of patient prognosis and patient heterogeneity in terms of treatment does exist. For example, women who switching to lapatinib were younger and more likely to achieve antitumor activity of new anti-HER2 regimen. Second, patients who received chemotherapy and trastuzumab sequentially or concomitantly may affect the prognosis. Third, some data could not be abstracted from medical records and missing.

Conclusions
In conclusion, these data confirm that after developing resistance to trastuzumab, patients can still derive benefit from HER-2 targeted therapy. The combination of lapatinib and capecitabine has shown a prolonged survival compared with TBP in patients with prior trastuzumab exposure.

Ethics approval and consent to participate
The study was approved by First Affiliated Hospital of Nanjing Medical University (Nanjing, China) and written informed consent from each patient was obtained. The use of samples was approved by the Ethics Committee of the First Affiliated Hospital of Nanjing Medical University.

Consent for publication
Written informed consent for publication of their clinical details was obtained from the patients.

Availability of data and material
The datasets during and analysed during the current study available from the corresponding author on reasonable request.

18
Final approval of manuscript: Yongmei Yin   Comprehensive annotation of genomic alterations in this study. The number of tumor mutation burdens and CRGAs in patients with Extracranial metastases compared with brain metastases Figure 8 The number of tumor mutation burdens and CRGAs in primary and secondary resistant patients

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download.