Background
In patients with advanced Non-Small Cell Lung Cancer (NSCLC) harbouring Epidermal Growth Factor Receptor (EGFR) activating mutations, disease progression occurs after a median of 10–12 months of treatment with EGFR-Tyrosine Kinase Inhibitors (TKIs), such as gefitinib, erlotinib, and afatinib. The acquisition of the EGFR-T790M secondary mutation is the main mechanism of acquired resistance, occurring in 50–60% of the cases [
1,
2].
Third-generation EGFR-TKIs, such as osimertinib, are active against both EGFR sensitizing mutant and T790M resistant tumors [
3]. Osimertinib represents the new standard of care in the treatment of T790M-positive NSCLC patients resistant to previous generation TKIs [
4]. However, also for this agent acquired resistance was expected and some mechanisms have been recently identified [
5‐
7]; in particular, a new C797S mutation in exon 20 of
EGFR gene was found as sufficient to promote resistance to osimertinib [
8]. Other mechanisms include EGFR L718Q mutation [
5], MET amplification [
9], BRAF V600E mutation [
10], and transformation to small-cell carcinoma [
11]. Initial data showed that T790M mutation occurs in up to 80% of previously untreated EGFR mutated NSCLCs, suggesting that the presence of
de novo resistant clones may be more common than previously appreciated [
12], and providing evidence to support osimertinib as a potentially better first-line option compared to the currently approved EGFR-TKIs. Emerging preclinical and clinical evidence supports a scenario whereby the selective pressure imposed by the chronic exposure to targeted agents leads to the expansion of pre-existing cell clones carrying specific genetic alterations, which may ultimately become dominant. Previous studies have mainly focused on the development of strategies to overcome specific mechanisms of resistance emerging after treatment with EGFR-TKI monotherapy. However, implementing this approach clinically may be difficult given the wide variety of the identified resistance mechanisms, also for osimertinib. A more feasible,and potentially successful strategy may be to identify a combination treatment approach that prevents the occurrence of more than one resistance mechanism [
6]. The good safety profile of osimertinib would permit combinations with other drugs in a tolerable fashion in clinic. Our previous data provide a rationale for combining first-generation EGFR-TKIs with monoclonal antibodies or chemotherapy [
13‐
15]. Despite significant advances in the knowledge about osimertinib therapy, the effect of this drug in combination with monoclonal antibodies targeting HER family has not been investigated yet.
T-DM1, trastuzumab emtansine, is an antibody-drug conjugate composed by the microtubule polymerization inhibitor DM1 (derivative of maytansine) linked with a stable thioether linker to trastuzumab targeting HER-2 receptors [
16]. T-DM1 was approved by FDA as a treatment for HER-2 positive pre-treated breast cancers and has been also evaluated as first-line therapy [
17]. HER-2 represents a relatively new therapeutic target for NSCLC and T-DM1 is currently being tested in a phase II study in NSCLC patients with HER-2 overexpression (NCT02289833) [
18]. Moreover, HER-2 amplification has been identified as a mechanism of osimertinib acquired resistance in a patient treated with osimertinib [
9], providing evidence for targeting HER-2 in this clinical setting. Interestingly, HER-2 amplification occurred together with the loss of EGFR T790M mutation. Similar findings were observed by Oxnard et al. in 2 of 40 patients treated with osimertinib [
19]. In addition, in vitro models provide functional evidence that HER-2 amplification may also induce innate resistance to osimertinib, confirming the clinical observations [
20].
In this study we explored the potential of combining osimertinib with T-DM1 in order to improve the efficacy of third-generation EGFR-TKI and to delay or overcome resistance in NSCLC cell lines carrying EGFR activating mutation and T790M mutation or HER-2 amplification.
Discussion
In this study we demonstrated the efficacy of T-DM1 when combined with osimertinib in delaying and overcoming resistance to the EGFR TKI in T790M-positive models (PC9-T790M and H1975 cells) and in PC9/HER2 cell lines, respectively.
Osimertinib is an effective treatment in NSCLC patients with T790M EGFR mutation, progressed after first-line therapy with first- or second-generation EGFR-TKIs [
4]. Moreover, osimertinib was also evaluated in the first-line setting in patients with EGFR activating mutations compared with gefitinib or erlotinib (FLAURA Trial NCT02296125). In this trial, recently presented at ESMO meeting [
26], osimertinib showed a statistically significant PFS benefit vs. gefitinib/erlotinib [median PFS 18.9 months in the osimertinib group (95% CI, 15.2, 21.4) and 10.2 months in the gefitinib/erlotinib group (95% CI, 9.6, 11.1); HR 0.46 (95% CI, 0.37, 0.57),
p-value <0.0001]. However, also for this agent, resistance is expected and new therapeutic approaches are needed to prolong its benefit [
7].
HER-2 was shown to be overexpressed (IHC 3+) in 2–6% of NSCLC, with gene amplification found in 2–4%, mainly in the adenocarcinoma histotype [
27]. In particular, HER-2 amplification has been identified as a potential mechanism of osimertinib acquired resistance [
9,
20].
We tested T-DM1 in preclinical experiments suggesting its activity against HER-2-driven lung cancer cell lines in vitro [
13]. Only recently this agent was evaluated in patients with HER2-driven NSCLC in two trials presented at ASCO meeting in 2017 [
28,
29]. Stinchcombe et al. [
28] reported primary results from an ongoing phase 2 study (NCT02289833) of 49 patients with previously treated HER2-overexpressing advanced NSCLC who received single-agent T-DM1, showing objective tumor responses only in 4 patients with elevated (3 +) HER2 immunohistochemical expression. In the Li et al. study (NCT02675829) [
29] patients with HER2-mutant lung cancers were enrolled into a cohort of the basket trial of TDM1 in HER2-amplified or mutant cancers. This study has met its primary endpoint, evidencing a tumor response in 8 out of 18 (44%) patients, across different mutation subtypes.
As previously reported by Scaltriti et al. for lapatinib in breast cancer cells [
30] and by our group for erlotinib in EGFR wild-type NSCLC cells [
14], we demonstrated, in this study, that also osimertinib increased the cell surface expression of HER-2 in EGFR-mutated PC9 in PC9-T790M and in H1975 cell lines. The efficacy of dual inhibition of EGFR and HER-2 has been documented in EGFR wild-type NSCLC cell lines combining gefitinib or erlotinib with trastuzumab or pertuzumab [
14,
31,
32], and in EGFR T790M mutant cell lines combining anti-EGFR, anti HER-2 and anti HER-3 antibodies [
33]. This combination strategy has not been evaluated yet in NSCLC clinical trials, probably considering the availability of agents able to block EGFR and HER-2 simultaneously, such as afatinib and dacomitinib [
18].
Here we demonstrated that, differently from trastuzumab, T-DM1 showed anti-proliferative effects in PC9, PC9-T790M and H1975 cells and exerted an additive effect when combined with osimertinib in terms of inhibition of proliferation, cell death and ADCC induction. Because osimertinib is highly effective in mutated NSCLC cell lines, the addition of T-DM1 caused a modest benefit when evaluated in short-term growth assay, by contrast the combination significantly delayed the acquired resistance to osimertinib in PC9-T790M and in the less osimertinib-sensitive model H1975. The combination was effective either when the cells were treated continuously with osimertinib and a simultaneous treatment with TDM1 was cyclically performed (see Fig.
4 and Fig.
5, schedule C) or when the cells were treated with osimertinib plus TDM1 intercalated with osimertinib alone (schedule E). It is worthy of note that the application of an intermittent treatment, in which osimertinib was switched with T-DM1 (schedule D), demonstrated a negative interaction between osimertinib and T-DM1 with a strong reduction in sensitivity to T-DM1 in both PC9 and PC9-T790M cells. In addition, an up-regulation of the anti-apoptotic Mcl1 protein and a down-regulation of the pro-apoptotic Bim protein were documented in cells treated with osimertinib and then exposed to T-DM1 compared to cells treated with T-DM1 alone. Similarly, a drug-antagonism has been also demonstrated when a EGFR-TKI was administered before pemetrexed [
15,
34].
At present, to our knowledge, this is the first study demonstrating that osimertinib reduces the efficacy of an antibody-chemotherapeutic drug conjugate given as second treatment; moreover, the novelty of this study is related to the effect of different combinatory schedules in the acquisition of osimertinib resistance. In PC9-T790M cells, the discontinuous protocol (osimertinib→T-DM1) anticipated the appearance of resistance to osimertinib compared to cells continuously treated with osimertinib alone, by contrast in the protocols in which cells were simultaneously exposed to both drugs (osimertinib→osimertinib + T-DM1 or osimertinib + T-DM1→osimertinib), no resistant colonies were detected up to 50 days of treatment. In H1975 cells, all schedules delayed the acquisition of resistance; however, also in this model the less effective was the intermittent schedule (osimertinib→T-DM1) both in terms of percentage of resistant wells and of median time to resistance.
Overall, these data may support the use of a combination of osimertinib with T-DM1 using a specific schedule to delay or prevent the acquisition of resistance to the EGFR-TKI, similarly to other combinations of third-generation EGFR-TKIs (with MEK or MET inhibitors) tested in clinical trials currently recruiting patients [
7].
In the last part of this study we investigated the role of HER-2 amplification in the responsiveness to osimertinib and the efficacy of T-DM1 in vitro and in vivo to overcome osimertinib resistance. HER-2 amplification, indeed, has been reported as a mechanism of acquired resistance to EGFR-TKIs with a frequency of 12–13% [
2,
35].
PC9/HER2c1 cells were exposed to osimertinib in vitro and in vivo in xenotransplanted mice. These cells were significantly more resistant to osimertinib than the parental cells in vitro and tumors grew as the controls indicating intrinsic resistance to osimertinib. T-DM1 significantly reduced tumor growth but the combination of osimertinib with T-DM1 showed a complete suppression of tumor growth, suggesting an important role of T-DM1 in overcoming osimertinib resistance. These data confirm the role of HER-2 amplification as a potential cause of intrinsic resistance to osimertinib, as demonstrated by Ortiz-Cuaran S et al. [
20], and indicate the combination with TDM1 as a potential strategy to overcome such resistance. Although in the absence of a cell line model with EGFR-activating/T790M developing HER-2 amplification as a resistance mechanism, these results are particularly relevant considering the potential future use of osimertinib in first-line therapy for EGFR-mutated NSCLC patients.