Head and neck squamous cell carcinoma (HNSCC) has an incidence of more than 600,000 new cases worldwide per year [
1]. The majority of HNSCC patients are diagnosed in the later stages of the disease, with more than half of patients having locoregionally advanced (LA) HNSCC at the time of diagnosis and approximately 10% of patients having metastatic disease [
2]. Prognosis in LA patients is poor, with around 50% of unresectable patients relapsing 5 years after receiving definitive chemoradiotherapy (CRT) [
3,
4], a standard treatment in this setting. Resectable patients receiving adjuvant CRT following surgery have a 5-year recurrence rate of 20% [
5]. Furthermore, recurrent and/or metastatic (R/M) HNSCC patients receiving first-line chemotherapy only have a median overall survival (OS) of approximately 7 months [
6]. Therefore, the R/M setting represents a group of patients who require novel treatment approaches.
Non-targeted treatments for R/M HNSCC
The most common non-targeted treatment approach in R/M HNSCC is a platinum-containing agent combined with either a taxane or 5-fluorouracil (5-FU) [
7]. Response rates and OS in R/M HNSCC following platinum-based doublet chemotherapy in the first-line setting are low. Results from four large randomised studies in this setting comparing cisplatin plus 5-FU with other single-agent chemotherapy agents demonstrated that combination regimens elicited response rates of around 20–30% [
8‐
11]. In three further studies assessing cisplatin in combination with paclitaxel, median OS was reported to be between 6.5 and 8.0 months [
12‐
14]. Moreover, following failure on a platinum-containing regimen there is no defined standard of care and second-line treatment options for R/M patients are limited, thus highlighting the need for alternative treatments that can improve outcomes in these patients.
Methotrexate is commonly used in R/M HNSCC [
7] and continues to be used as a standard comparator in some Phase III trials, mainly after platinum failure or in patients judged unfit for platinum therapy [
9,
15‐
17]. Taxanes have also been used in this setting. However, no study has been able to show that these agents improve OS. The ability of methotrexate to increase OS in HNSCC patients has not been formerly demonstrated in Phase III trials. This agent produces a response of short duration (approximately 3–6 months) in around 4–24% of cases and only rarely elicits complete responses (CRs) [
15,
16,
18‐
20].
Targeted treatment approaches
A recent approach to new cancer therapies has been to develop targeted agents that inhibit particular signalling pathways implicated in tumourigenesis. Epidermal growth factor receptor (EGFR; ErbB1) is a member of the ErbB family of receptor tyrosine kinases that plays an integral role in the oncogenesis of several ErbB-driven cancers, including HNSCC [
21]. Overexpression of EGFR provides tumour cells with growth and survival advantages, and this process is thought to substantially contribute to the aggressive nature of cancer cell proliferation. Approximately 90% of patients with HNSCC overexpress EGFR and prognosis for these patients can be lower than for patients without high levels of EGFR expression, with increased EGFR expression correlating with a reduction in recurrence-free survival or OS rates [
21]. One study has shown that in patients with laryngeal squamous cell carcinoma, those with low EGFR expression levels have a 5-year OS rate of 81% compared with 25% for patients with high levels of EGFR expression [
22].
Cetuximab is an EGFR-targeting monoclonal antibody and is the only targeted treatment approved in the US and Europe for the treatment of HNSCC in combination with radiotherapy for LA disease and in combination with platinum-based chemotherapy for R/M disease [
23,
24]. It is also approved in the US as monotherapy in R/M HNSCC following progression on platinum-based chemotherapy [
23]. In a Phase III trial in R/M patients, combination treatment with cetuximab and cisplatin led to an objective response rate (ORR) of 26% versus 10% with cisplatin plus placebo (p = 0.03) [
25]. However, owing to this trial being underpowered, no significant difference was observed for progression-free survival (PFS) or OS in both arms. In the larger confirmatory EXTREME study, cetuximab in combination with platinum-based chemotherapy elicited an OS benefit in untreated R/M HNSCC patients versus chemotherapy alone [
6]. The median OS was prolonged from 7.4 months in patients receiving chemotherapy alone to 10.1 months in the cetuximab plus chemotherapy arm. Median PFS was also increased from 3.3 months in the chemotherapy alone group to 5.6 months in the combination group.
In platinum-refractory R/M HNSCC patients with disease progression, three studies have been performed assessing the efficacy of cetuximab either alone or in combination with platinum-based chemotherapy. In 2005, two trials evaluated cetuximab in combination with either cisplatin or carboplatin in this setting. Herbst
et al. reported an ORR of 10% and OS of 5.2 months in patients receiving cetuximab plus cisplatin [
26] and similar results were observed by Baselga
et al. who determined an ORR of 10% and OS of 6 months following treatment with cetuximab plus cisplatin or carboplatin [
27]. Cetuximab has also been investigated as a monotherapy in the R/M population in patients who have failed platinum-based chemotherapy, with a best overall response rate of 13% and OS of 5.9 months observed [
28]. This trial suggests that single-agent cetuximab offers similar efficacy to combination treatment with platinum-based chemotherapy in R/M HNSCC patients refractory to platinum-containing therapy. A pooled analysis of these three trials was performed in 2008, which compared them to a retrospective trial by Leon
et al.[
29]. Leon
et al. assessed the outcomes of platinum-refractory R/M HNSCC patients treated between 1990 and 2000 with best supportive care or various second-line therapies. This indirect comparison indicated that median OS may be increased by approximately 2 months when cetuximab is administered following platinum failure, with OS ranging between 5.2 and 6.1 months in the cetuximab studies versus 3.4 and 3.6 months in Leon
et al.’s retrospective analysis [
30].
Several other targeted agents are currently being investigated for HNSCC, including the monoclonal antibody panitumumab, the small-molecule tyrosine kinase inhibitors dacomitinib and lapatinib, and the oncolytic virus reolysin. The monoclonal antibody nimotuzumab is already approved in numerous countries, including Brazil, India, China, Argentina and Indonesia. However, this is still under investigation for HNSCC in the US and Europe.
Acquired or primary resistance to targeted therapies is common, with several mechanisms being implicated in this process. These postulated or hypothetical mechanisms include receptor-independent activation of downstream signalling cascades, cross-talk with other receptor tyrosine kinases, and environmental factors, such as viral infections and inflammatory agents [
31]. A novel approach to overcome treatment resistance is inhibition of multiple ErbB family members simultaneously or binding multiple ErbB family members irreversibly [
32]. By blocking all ErbB family members, greater efficacy may be achieved as all ErbB-driven oncogenic pathways are compromised. Furthermore, irreversible inhibition, mediated by covalent binding to specific residues of the target, may lead to sustained suppression of tumour growth as prolonged cellular activity is inhibited.
Afatinib, an irreversible ErbB family blocker
Afatinib is an oral ErbB family blocker that completely and irreversibly blocks signalling by all relevant ErbB family members, including EGFR, human epidermal growth factor receptor-2 (ErbB2) and ErbB4, and also blocks transphosphorylation of ErbB3 [
33,
34]. It is approved in the US for the first-line treatment of EGFR mutation-positive metastatic NSCLC and it is also being developed for the treatment of a number of other ErbB-driven tumours, including breast cancer and HNSCC.
In the LUX-Lung clinical trial programme, afatinib has been investigated for the treatment of EGFR mutation-positive NSCLC either in the first-line setting [
35‐
37] or in patients with no more than one prior chemotherapy [
38]. It has also been assessed following chemotherapy and/or EGFR tyrosine kinase inhibitor therapy [
39‐
43]. In the proof-of-concept LUX-Lung 2 trial, afatinib monotherapy elicited an ORR of 61% in NSCLC patients [
38] and in LUX-Lung 3, to our knowledge being the largest, prospective, randomised trial in EGFR mutation-positive NSCLC patients, the primary endpoint of PFS was met, with a median PFS of 11.1 months observed for afatinib-treated patients versus 6.9 months in chemotherapy-treated patients [
35]. Afatinib has also demonstrated a manageable safety profile, with recent pooled data analyses in patients with solid tumours showing that gastrointestinal and dermatological adverse events in particular can be effectively managed in this patient population [
44,
45].
In HNSCC, afatinib has demonstrated preclinical activity in both
in vitro and
in vivo models [
46,
47] and clinical activity in a proof-of-concept Phase II study [
48,
49]. In the human HNSCC FaDu cell line, afatinib inhibited tumour cell proliferation in the low nanomolar concentration range, with additive growth inhibitory effects demonstrated when combined with standard chemotherapies versus single-agent treatment [
47]. The Phase II proof-of-concept study showed comparable activity between afatinib and cetuximab in R/M HNSCC patients following failure of platinum-based chemotherapy. In Stage I, ORRs were 8.1% in afatinib-treated patients and 9.7% in cetuximab-treated patients (independent central review) [
48]. Furthermore, in Stage II of the study, after crossover to the opposite treatment arm, afatinib elicited a disease control rate of 33% in patients who received cetuximab in Stage I (vs. 19% in cetuximab-treated patients after crossover from afatinib) and demonstrated PFS of 9.3 weeks in the afatinib group versus 5.7 weeks in the cetuximab group, suggesting sequential therapy with afatinib may be efficacious in patients pretreated with an EGFR-targeted therapy [
49]. Therefore, these data warrant further investigation of this compound for the treatment of R/M HNSCC.
In afatinib monotherapy trials, the maximum tolerated dose was determined to be continuous daily afatinib at either 40 mg or 50 mg [
50,
51]. Afatinib 50 mg/day was the starting dose used in the proof-of-concept LUX-Lung 2 trial in EGFR mutation-positive patients who had received no more than one previous chemotherapy [
38]. However, the dose was reduced to 40 mg/day to improve the safety profile of afatinib and, as there was no difference in efficacy in patients receiving both doses, 40 mg/day afatinib was the starting dose used in the subsequent LUX-Lung 3 and 6 trials [
35,
36]. In the HNSCC proof-of-concept trial, a starting dose of 50 mg/day afatinib was used [
48]; however, afatinib demonstrated a more manageable safety profile at 40 mg/day in this study and so this is the chosen starting dose of afatinib in LUX-Head & Neck 1, with individual dosing allowed depending on how well patients tolerate treatment. It has been established that using a dose-reduction scheme in the administration of afatinib is an effective approach to minimising the consequences of adverse events and discontinuation of afatinib. Therefore, this is the approach being adopted in LUX-Head & Neck 1.
The LUX-Head & Neck 1 study (NCT01345682) has been initiated to assess the efficacy and safety of afatinib versus methotrexate in the second-line treatment of R/M HNSCC patients following failure of first-line platinum-based chemotherapy. Given methotrexate is a standard treatment in R/M HNSCC in many countries, and is used as a standard comparator in other Phase III trials, this agent is considered an appropriate comparator in this study. In particular, this second-line trial is powered to detect superiority of afatinib over methotrexate in terms of a PFS and OS benefit. There are currently no approved predictive tumour- or serum-derived biomarkers guiding treatment with ErbB-directed therapies in HNSCC. Therefore, this study also includes a biomarker assessment part.