Combining chemotherapy with epidermal growth factor receptor inhibition in advanced non-small cell lung cancer

Abstract

Treatment of advanced stage lung cancer is changing rapidly. With the new found knowledge on molecular targets such as the epidermal growth factor receptor (EGFR), effective therapy is now available in a selected population with the target mutation. Single-agent epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI) is a standard first-line therapy for patients with activating-EGFR mutation such as base-pair deletion in exon 19 or point mutation at exon 21. At the same time, this class of drugs may be combined with chemotherapy. Studies on the concurrent combination of chemotherapy and EGFR-TKI confirmed a lack of efficacy. A phase II study on sequential intercalated combination has demonstrated an improvement in progression-free survival (PFS), but this needs to be validated by the ongoing phase III study. The third approach is to combine EGFR-TKI as maintenance therapy after tumour response or stable disease to cytotoxic chemotherapy. Two phase III studies have shown improvement in PFS, but the use of biomarkers for the selection of maintenance therapy remains debatable. Cetuximab is a monoclonal antibody against EGFR and its combination with chemotherapy was shown to improve overall survival in an unselected population. A new biomarker using the H-score will help to select patients for this combination.

Introduction

Advanced stage non-small cell lung cancer (NSCLC) continues to be one of the most fatal malignancies. Median overall survival (OS) is persistently lower than 12 months despite optimal platinum-based doublet chemotherapy [Schiller et al. 2002; NSCLC-Meta-Analyses-Collaborative-Group, 2008; Azzoli et al. 2009]. Much work was dedicated to improving the treatment outcome by combining novel drugs to standard platinum-based doublets, but most were not successful. An example that showed superior outcome was the addition of bevacizumab, a monoclonal antibody targeting the vascular endothelial growth factor, to standard chemotherapy [Reck et al. 2009; Sandler et al. 2006]. OS was only modestly increased from 10.3 to 12.3 months (p = 0.003) [Sandler et al. 2006]. The other promising approach is the addition of an epidermal growth factor receptor (EGFR) inhibitor to chemotherapy. In this article we explore the various ways of combining different types of EGFR inhibitors, including tyrosine kinase inhibitors (TKIs) and monoclonal antibodies (Mabs), with chemotherapy in advanced NSCLC. The EGFR inhibitors could be given either concurrently, sequentially in an intercalated approach, or selectively as maintenance therapy. Based on existing data, we also discuss future development and application.

EGFR inhibition

The EGFR, with its related ErbB-family receptors and their signal transduction network, is the most extensively studied molecular target in lung cancer. It is expressed in approximately 80% of NSCLCs and plays a role in the development and progression of the disease [Meert et al. 2002; Mendelsohn and Baselga, 2006]. Two different anti-EGFR strategies are currently available in the therapeutic armamentarium: Mabs that prevent EGFR-ligand binding (e.g. cetuximab) and TKIs that block phosphorylation of the intracellular tyrosine kinase component of the EGFR (e.g. gefitinib and erlotinib). Both of these strategies dampen signal transduction through some of the downstream pathways such as the RAS/RAF/mitogen-activated protein kinase and phosphoinositide 3-kinase-AKT cascades, thus limiting cell growth, proliferation, invasion, angiogenesis and metastasis [Mendelsohn and Baselga, 2006; Normanno et al. 2006]. Mabs may also elicit a host immune response through antibody-dependent cell-mediated cytotoxicity, where natural killer cells, monocytes and eosinophils are recruited to lyse the targeted cells [Loong et al. 2009].

Discovery of the EGFR mutation has changed the management of lung cancer [Lynch et al. 2004; Paez et al. 2004]. Activating-EGFR mutation is usually referred to base-pair deletion in exon 19 or point mutation at exon 21, which are more sensitive to EGFR-TKI. The Iressa Pan-Asia Study was the first randomized phase III study that confirmed the role of EGFR-TKI monotherapy as a first-line therapy for patients with a known EGFR mutation [Mok et al. 2009a]. The objective response rate for EGFR mutation-positive patients treated with gefitinib was 71.2%, compared with 47.3% in patients treated with chemotherapy (p < 0.001). In the EGFR mutation-negative group, only around 1.1% of patients responded to gefitinib, compared with a response rate of 23.5% to chemotherapy (p = 0.001). Progression-free survival (PFS) was prolonged in the gefitinib group (hazard ratio [HR] 0.48; p < 0.0001). However, due to the significant portion of crossover to the TKI group, OS was similar. To date, a total of five randomized studies from both Western and Asian countries have demonstrated similar results (Table 1) [Lee et al. 2009; Maemondo et al. 2010; Mitsudomi et al. 2010; Rosell et al. 2011; Zhou et al. 2011].

Table 1.

Randomized phase III studies of first-line epidermal growth factor receptor tyrosine kinase inhibitors in advanced or metastatic non-small cell lung cancer.

Study	Study name	Sample size (EGFR-mutation positive) TKI/chemotherapy	Tumour response rate (TKI versus chemotherapy)	Median progression-free survival (months)
Mok et al. [2009]	IPASS	132/129	71.2% versus 47.3%, p < 0.001	9.8 versus 6.4, p < 0.001
Lee et al. [2009]	First-SIGNAL	26/16	53.5% versus 42%, p = 0.1533	8.4 versus 6.7, p = 0.084
Mitsudomi et al. [2010]	WJTOG 3405	86/86	62.1% versus 32.2%, p < 0.0001	9.2 versus 6.3, p < 0.0001
Maemondo et al. [2010]	NEJ02	114/114	73.7% versus 30.7%, p < 0.001	10.8 versus 5.4, p < 0.001
Zhou et al. [2011]	OPTIMAL	83/82	83% versus 36%, p < 0.0001	13.1 versus 4.6, p < 0.0001

EGFR, epidermal growth factor receptor; First-SIGNAL, first-line single-agent Iressa versus Gemcitabine and Cisplatin Trial in Never-Smokers with Adenocarcinoma of the Lung; IPASS, Iressa Pan-Asia Study; NEJ, North East Japan; OPTIMAL, erlotinib versus chemotherapy as first-line treatment for patients with advanced EGFR mutation-positive, non-small cell lung cancer; TKI, tyrosine kinase inhibitor; WJTOG, West Japan Thoracic Oncology Group.

Chemotherapy and EGFR-TKI: concurrent combination

EGFR-TKIs were combined with first-line platinum-based chemotherapy. Four randomized studies comparing the concurrent combination of EGFR-TKIs and platinum-based doublet chemotherapy with chemotherapy alone were reported nearly a decade ago before the discovery of the EGFR mutation [Giaccone et al. 2004; Herbst et al. 2004, 2005; Gatzemeier et al. 2007]. Table 2 summarizes the results, and evidence is clear that the combination had failed to improve tumour response or survival. A recent meta-analysis that included these studies and a phase II study on the combination of chemotherapy and vandetanib reinforced this observation [Heymach et al. 2007]. In this meta-analysis, 3918 patients treated with EGFR-TKIs plus the platinum-based doublet chemotherapy combination or chemotherapy alone were identified and there was no statistical difference in median OS (10.6 versus 11.0 months, HR 1.04; 95% confidence interval [CI] 0.96− 1.13; p = 0.348) [Chen et al. 2011]. However, the PFS was marginally prolonged from 5.4 to 5.6 months (p = 0.03). Overall, the current data suggest that concurrent administration of chemotherapy and EGFR-TKI in an unselected population is not a standard therapy for lung cancer.

Table 2.

Randomized phase III studies of concurrent combination of chemotherapy and an epidermal growth factor receptor tyrosine kinase inhibitor.

Study	Study name	Sample size (n)	Tyrosine kinase inhibitor	Chemotherapy	Response rate	Median overall survival (weeks)	Hazard ratio (95% CI)
Giaccone et al. [2004]	INTACT 1	1093	Gefitinib 250 mg, 500 mg, placebo	Cisplatin/ gemcitabine	50.3% versus 49.7% versus 44.8%	39.6 versus 39.6 versus 43.6	NS
Herbst et al. [2004]	INTACT 2	1037	Gefitinib 250 mg, 500 mg, placebo	Carboplatin/ paclitaxel	30.4% versus 30% versus 28.7%	39.2 versus 34.8 versus 39.6	NS
Herbst et al. [2005]	TRIBUTE	1059	Erlotinib 150 mg	Carboplatin/ paclitaxel	21.5% versus 19.3%	42.4 versus 42	0.88 (0.86 − 1.16)
Gatzemeier et al. [2007]	TALENT	1172	Erlotinib 150mg	Cisplatin/ gemcitabine	31.5% versus 29.9%	43.0 versus 44.1	1.06 (0.9 − 1.23)

CI, confidence interval; INTACT, Iressa NSCLC Trial Assessing Combination Treatment; NS, not significant; TALENT, Tarceva Lung Cancer Investigation; TRIBUTE, Tarceva responses in conjunction with paclitaxel and carboplatin.

A more specific question is whether the combination would improve outcome in patients with activating-EGFR mutation given that single-agent EGFR-TKI is highly efficacious in this group. Janne and colleagues studied the combination of paclitaxel/carboplatin and erlotinib versus erlotinib alone in an enriched population for the EGFR mutation [Janne et al. 2010]. In the subgroup of 66 patients with the EGFR mutation who received either single-agent erlotinib or concurrent combination of chemotherapy and erlotinib, PFS was 15.7 and 17.2 months; OS was 31.3 and 39 months, respectively. The addition of chemotherapy to erlotinib did not appear to improve treatment outcomes in patients with the EGFR mutation, but this study is underpowered because of its sample size.

Combination of chemotherapy and EGFR-TKI: intercalated combination

Concurrent administration may not work for the reason of TKI-induced, G1-phase cell-cycle arrest [Piperdi et al. 2004]. During the arrest, cell-cycle phase-dependent chemotherapeutic agents will not be effective [Davies et al. 2006; Li et al. 2007]. By giving EGFR and chemotherapy sequentially and thus achieving pharmacodynamic separation of the two agents, the mutual inhibitory effect could be avoided [Mahaffey et al. 2007]. A phase I study on sequential administration of pemetrexed followed by erlotinib provided support for the hypothesis [Davies et al. 2009].

Mok and colleagues further tested this concept in a randomized phase II study the First-line Asian Sequential Tarceva and Chemotherapy Trial (FASTACT) [Mok et al. 2009b]. A total of 154 treatment-naïve Asian patients were randomized to receive gemcitabine plus cisplatin or carboplatin at day 1 and gemcitabine at day 8 followed by erlotinib from days 15 to 28 or similar chemotherapy followed by placebo on a 4-weekly cycle. The primary endpoint of nonprogression rate at 8 weeks was similar. However, PFS was significantly longer in the combination arm of chemotherapy and erlotinib compared with chemotherapy and placebo (HR 0.47; p = 0.018). Tumour response rate was also higher (35.5% versus 24.4%; p = 0.12), favouring the sequential combination. OS was similar as the placebo control arm was unblinded at progression and offered the option of erlotinib as a poststudy treatment. No additional toxicity was reported with the sequential combination. A similar intercalated regimen combining paclitaxel/carboplatin with erlotinib was reported by Hirsch and colleagues [Hirsch et al. 2008]. Erlotinib was given from days 2 to 15, while paclitaxel/carboplatin was given at day 1 of a 3-weekly cycle. Only patients with activating-EGFR mutations demonstrated a significantly higher 6-month PFS rate with erlotinib treatment compared with patients without these mutations (89% versus 23%; p = 0.003).

The encouraging data from FASTACT is being validated in a randomized phase III trial (FASTACT II). The study design is similar to FASTACT and the primary endpoint is PFS. To date, a total accrual target of 450 patients has been met and over 60% of patients have tissue samples available for translational biomarker research. This study will help to define the role of an intercalated combination of chemotherapy and EGFR-TKI in either EGFR-mutation positive or negative patients.

Combination of chemotherapy and EGFR-TKI: maintenance therapy

Maintenance therapy is the continuation of similar or different therapy immediately after the initial response or stable disease to platinum-based doublets. The previous standard was first-line chemotherapy followed by observation and, only upon disease progression, was second-line therapy given. One of the new approaches is to give non-cross-resistant therapy [Azzoli et al. 2009; Mok and Ramalingam 2009]. Maintenance therapy with single-agent chemotherapy such as gemcitabine and docetaxel prolonged PFS but not OS. Only maintenance pemetrexed is associated with improvement in both PFS and OS [Ciuleanu et al. 2009]. Alternatively, adding an EGFR-TKI after standard cytotoxic chemotherapy is a way to combine the two agents. In a phase III study, Sequential Tarceva In Unresectable NSCLC (SATURN), 889 patients who did not progress after four cycles of first-line platinum-based chemotherapy were randomized to receive either maintenance erlotinib 150 mg daily or placebo [Cappuzzo et al. 2010]. This study has met the primary endpoint with median PFS and OS significantly longer with erlotinib than placebo in the overall population. The primary endpoint was a small increment of 1 week in PFS (12.3 versus 11.1 weeks; HR 0.71; p < 0.0001). OS benefit was demonstrated (12 versus 11 months; HR 0.81; p = 0.0088), but was limited by the fact that about 30% of patients in the placebo group did not receive any subsequent treatments and only around 20% of patients in the placebo group received erlotinib after they progressed. In a biomarker analysis, EGFR mutation was confirmed to be a predictive biomarker for maintenance erlotinib [Brugger et al. 2011]. Improvement in PFS in patients (n = 49) with the EGFR mutation was dramatic, while less but significant improvement was also observed in patients without the mutation. Median PFS was significantly prolonged when comparing maintenance erlotinib with placebo (44.6 versus 13 weeks; HR 0.1; p < 0.0001). Lack of benefit in OS is explained by the high degree of censoring and the extensive crossover of placebo patients to EGFR-TKI. Interestingly, for patients without the EGFR mutation, there were benefits in both PFS (HR 0.77) and OS (HR 0.78). The interaction test for PFS was positive, thus the EGFR mutation is considered a predictive biomarker for maintenance therapy. Moreover, EGFR mutation is a predictive biomarker for first-line EGFR-TKI, the targeted drug is only indicated in mutation-positive patients in the first-line setting. The confusion is explained by the difference in the control arm with standard doublet therapy in first line and only placebo in the maintenance situation. Currently, the European Commission has approved erlotinib as maintenance therapy only in patients with stable disease after first-line chemotherapy as they seemed to derive a greater OS benefit with maintenance erlotinib than those who had partial or complete response in this study. To explore further the role of maintenance erlotinib, another phase III study [ClinicalTrials.gov identifier: NCT01328951] will compare maintenance erlotinib with the same treatment at the time of disease progression.

Another study, ATLAS, a study comparing bevacizumab with or without erlotinib for first-line treatment of NSCLC, investigated the combination of maintenance erlotinib and bevacizumab [Kabbinavar et al. 2010]. The study design was similar to the SATURN study except that bevacizumab was added to first-line chemotherapy and maintenance erlotinib or placebo. The study reported a PFS of 4.8 versus 3.7 months (HR 0.72; p = 0.0012), but there was no significant improvement in OS. Retrospective analysis of certain subsets like Asian/Pacific Islander patients, never-smokers and those with EGFR-sensitive mutations appeared to have a survival benefit when receiving the combination of maintenance bevacizumab plus erlotinib.

Gefitinib is another option for maintenance therapy. In the West Japan Thoracic Oncology Group 0203 trial, 604 patients with advanced NSCLC were randomized to maintenance gefitinib after three cycles of platinum-based chemotherapy versus six cycles of platinum-based chemotherapy [Takeda et al. 2010]. The trial failed to meet its primary endpoint of OS. Survival benefit was seen in the adenocarcinoma subgroup with maintenance gefitinib (HR, 0.79; 95% CI, 0.65− 0.98; p = 0.03), but no biomarker result was available. A European phase III study (EORTC 08021-ILCP 01/03) with a similar design to the SATURN study but using gefitinib instead was closed prematurely because of low accrual [Gaafar et al. 2011]. With 173 patients enrolled, there was significant prolongation of PFS (4.1 versus 2.9 months; HR 0.61; p = 0.0015), but the primary endpoint of OS was not met. The phase III trial from China, INFORM, also compared maintenance gefitinib with placebo. After a median follow up of 16.8 months, PFS was significantly longer with maintenance gefitinib than placebo (4.8 versus 2.6 months; HR = 0.42; p < 0.0001), while there was no difference in OS [Zhang et al. 2011].

Combination of chemotherapy and monoclonal antibody against EGFR-TKI

Cetuximab is an anti-EGFR Mab against EGFR. Single-agent activity in patients with advanced NSCLC is low with a partial response rate at 4.5% [Hanna et al. 2006]. Combination studies of cetuximab and platinum-based doublets have since been completed. In a recently published meta-analysis of randomized trials comparing anti-EGFR therapy and platinum-based doublet chemotherapy versus platinum-based doublet chemotherapy alone, Chen and colleagues [Chen et al. 2011] performed a meta-analysis on four clinical trials (Table 3) comparing cetuximab plus platinum-based doublet chemotherapy with platinum-based doublet chemotherapy alone [Butts et al. 2007; Rosell et al. 2008; Pirker et al. 2009; Lynch et al. 2010]. Median OS was 10.9 months in combination groups and 9.8 months in platinum-based doublet alone groups, with the significant difference favouring the cetuximab combination (HR = 0.87; 95% CI 0.78− 0.96; p = 0.004). There was marginal difference in PFS (4.7 versus 4.4 months) favouring the combination. Tumour response rate was also higher (32% versus 24%, respectively). Another meta-analysis of the same studies also reached the same conclusions [Ibrahim et al. 2011]. A more detailed look at adverse events in the latter meta-analysis showed that the cetuximab-combination groups developed a significantly higher incidence of adverse events associated with anti-EGFR therapy, including diarrhoea, asthenia, infusion-related reaction, hypokalaemia and skin toxicity. Current data suggested a small OS benefit associated with the combination of chemotherapy and cetuximab in an unselected population. Limited by the small benefit, the drug has not been approved for usage in lung cancer in either the USA or Europe.

Table 3.

Clinical trials of cetuximab and chemotherapy combination versus chemotherapy alone.

Study	Trial	Phase	Trial size (n)	Chemotherapy	Primary endpoint	Median PFS (month)	HR (95% CI)	Median OS (month)	HR (95% CI)
Butts et al. [2007]	–	II	65	Cisplatin or carboplatin/gemcitabine	RR	5.09 versus 4.21	–	12.0 versus 9.3	–
Rosell et al. [2008]	–	II	86	Cisplatin/ vinorelbine	RR, safety, PK	5.0 versus 4.6	–	8.3 versus 7.3	–
Pirker et al. [2009]	FLEX	III	1125	Cisplatin/ vinorelbine	OS	PFS reported as NS	–	11.3 versus 10.1	0.87 (0.762 − 0.996); p = 0.044
Lynch et al. [2010]	BMS099	III	676	Carboplatin/ taxane (paclitaxel or docetaxel)	PFS	4.40 versus 4.24	0.9 (0.76 − 1.07); p = 0.236 (NS)	9.7 versus 8.4	0.89 (0.75 − 1.05); p = 0.17 (NS)

BMS099, Bristol-Myers Squibb 099; CI, confidence interval; FLEX, First-Line Erbitux in Lung Cancer; HR, hazard ratio; NS, not significant; OS, overall survival; PFS, progression-free survival; PK, pharmacokinetics; RR, response rate.

Preplanned subgroup analysis was performed in the tumour samples from the First-Line Erbitux in Lung Cancer study, a randomized study that compared a combination of chemotherapy and cetuximab with chemotherapy alone [Pirker et al. 2009]. Overall 1100 patients with treatment-naïve advanced NSCLC with immunohistochemical evidence of EGFR expression were randomized and patients in the combination group attained slightly longer OS (11.3 versus 10.1 months, p = 0.044) A new immunohistological scoring system (H-score) was adopted to define EGFR-protein expression as a predictive biomarker. About one-third of all enrolled patients had an H-score of 200 or above, which was considered to be ‘high’ for EGFR-protein expression [O’Byrne et al. 2010]. Tumour response rate in the high-expression group was 44.4% and 28.1% for chemotherapy/cetuximab and chemotherapy alone groups, respectively (p = 0.002). In contrast, there was no difference in response rate in the low-expression group (p = 0.36). The survival data were presented in a separate meeting [Pirker et al. 2011]. OS was 12.0 months in patients with high EGFR expression who received a combination of chemotherapy and cetuximab, compared with 9.6 months in those who received chemotherapy only (p = 0.011). These data have been re-submitted to the US Food and Drug Administration and European Medicines Agency forapproval of cetuximab in patients with high EGFR-protein expression.

Other Mabs against EGFR are being investigated. Panitumumab is a fully humanized anti-EGFR IgG2 Mab that failed to demonstrate clinical benefit in untreated advanced NSCLC patients when used in combination with paclitaxel/carboplatin [Socinski, 2007]. This may possibly be attributed to the lack of antibody-dependent cellular cytotoxicity in a fully humanized antibody. Necitumumab (IMC-11F8) is a recombinant human anti-EGFR Mab and similar in structure to cetuximab [Vanhoefer et al. 2004]. However, due to the absence of murine structures, hypersensitivity reactions are anticipated to be less frequent with necitumumab compared to cetuximab. Two phase III trials investigating necitumumab in patients with advanced NSCLC were initiated. One of these trials is evaluating cisplatin plus pemetrexed with and without necitumumab in patients with nonsquamous NSCLC (INSPIRE [ClinicalTrials.gov identifier: NCT00982111]), whereas the second trial is evaluating cisplatin plus gemcitabine with and without necitumumab in patients with squamous cell NSCLC (SQUIRE [ClinicalTrials.gov identifier: NCT00981058]). The primary endpoint of both trials is survival. Unfortunately, INSPIRE was closed prematurely due to concerns about the increased risk of thromboembolic events in the experimental arm. Accrual of SQUIRE is still ongoing.

Matuzumab, a humanized anti-EGFR monoclonal IgG1 antibody with a prolonged half-life, was also evaluated in patients with cancer [Kollmannsberger et al. 2006]. The maximum-tolerated doses of matuzumab were 1600 mg/week as a single agent and 800 mg/week when combined with paclitaxel. Results of a randomized phase II trial that compared pemetrexed plus matuzumab (either 800 mg weekly or 1600 mg every 3 weeks) with pemetrexed alone as second-line therapy in 148 patients with advanced NSCLC demonstrated trends for improvement in response rate and OS for weekly matuzumab plus matuzumab then pemetrexed alone [Schiller et al. 2010]. Further clinical trials are warranted for confirmation.

Summary

Personalized medicine according to EGFR mutation status is now feasible in patients with advanced lung cancer. Single-agent EGFR-TKI is a standard first-line therapy for patients with activating-EGFR mutation. Concurrent combination of chemotherapy and EGFR-TKI does not improve treatment outcome in either mutation-positive or negative patients. A sequential intercalated combination is now under investigation and the phase III result will be available in the near future. Using EGFR-TKI as maintenance therapy after achieving stable disease or tumour response to cytotoxic chemotherapy is feasible and is associated with significant improvement in PFS. Use of a biomarker for the selection of maintenance therapy remains debatable. The small improvement in OS from a combination of cetuximab and chemotherapy did not gain approval from authorities, but a new biomarker may improve the situation.