Oncology Reviews

, Volume 3, Issue 3, pp 137–148

Epidermal growth factor receptor pathway as therapeutic development in head and neck cancers: present and future

Authors

  • Cesar A. Perez
    • Internal Medicine/Department of MedicineCleveland Clinic Florida
  • Chancellor E. Donald
    • Hematology/Medical OncologyTulane University Health Sciences Center, Tulane Cancer Center
  • Luis E. Raez
    • Hematology/Oncology Division, Thoracic Oncology Group, Sylvester Comprehensive Cancer CenterUniversity of Miami Leonard M. Miller School of Medicine
    • Hematology/Oncology Division, Thoracic Oncology Group, Sylvester Comprehensive Cancer CenterUniversity of Miami Leonard M. Miller School of Medicine
Review

DOI: 10.1007/s12156-009-0022-7

Cite this article as:
Perez, C.A., Donald, C.E., Raez, L.E. et al. Oncol Rev (2009) 3: 137. doi:10.1007/s12156-009-0022-7
  • 26 Views

Abstract

The understanding of the epidermal growth factor pathway in terms of intracellular signaling and its role in proliferation and cell survival has impacted the therapeutic management of many solid tumor malignancies in which this pathway has been dysregulated. Once the receptor is targeted at its cellular membrane level or tyrosine kinase domain, its blockage induces downregulation of oncogenic and tumorigenesis mechanisms which were in place, and thus inhibits proliferation and induces apoptosis of the malignant cell. Nowadays, we have several monoclonal antibodies as well as small molecules which target the receptor of epidermal growth factor. Although several receptors have been described within the human epidermal receptor family, our discussion will be focused on the impact of human epidermal receptor-1 as a therapeutic option for locally advanced squamous cell carcinoma of the head and neck.

Keywords

CetuximabEpidermal growth factor receptor (EGFR)ErlotinibGefitinibHead and neck cancerLapatinibPanitumumabSquamous cell carcinomaTargeted therapyTyrosine kinase inhibitor

Introduction

In 2009, an estimated 48,010 people (35,160 men and 12,850 women) will develop head and neck cancer, and an estimated 11,260 deaths (8,140 men and 3,120 women) will occur [1]. As a group (oral cavity, oropharynx, larynx), SCCHN represents 2% of all new cancers and 1% of all cancer deaths [2]. Until recently, the therapy of SCCHN has relied on combined “conventional” cytotoxic chemotherapy either alone or in combination with radiation therapy (RT). However, the introduction of epidermal growth factor receptor (EGFR) antagonists or inhibitors has increased the response rate, event-free survival (EFS), and overall survival (OS) in patients with SCCHN when combined with RT or chemotherapy [3, 4]. The impact of these novel agents in the metastatic setting has encouraged researchers to introduce their use earlier in the natural history of this disease either as neoadjuvant or induction therapy or concomitant treatment. Thus far, the treatment of SCCHN still focuses in controlling the disease locally with surgical resection of the primary tumor (plus radical node dissection if indicated) alone, followed by either RT as single modality or concurrent chemoradiotherapy (combined modality). A caveat of this approach has been the distant disease rate which is the main reason for poor outcomes in SCCHN. In a way to decrease the rate of distant metastasis without affecting locoregional control, the use of induction therapy followed by concurrent chemoradiation is gaining more popularity due to its efficacy described in three-phase III randomized clinical trials [57]. Concomitantly with these developments in conventional cytotoxic treatment for SCCHN, biological agents have also entered in the treatment algorithm of this disease.

The novel biological agents target specific mechanism associated with tumor proliferation and represent an addition to the current therapy approaches due to their antitumor efficacy without adding major toxicities. To date, we have not identified the best combination for these agents, or the best therapy schedule, and their ultimate role in treating these diseases. Certainly, encouraging results such as the EXTREME (Erbitux in First Line Treatment of Recurrent or Metastatic Head and Neck Cancer) trial recently published by Vermorken et al. [3] confirmed the significant role of EGFR inhibition in this disease, and also represents a hope that this combination can reproduce its results in an early stage. Gefitinib and erlotinib, both tyrosine kinase inhibitors (TKIs), proved to be beneficial in patients with non-small cell lung cancer (NSCLC) [810]. Following their approval by the US Food and Drug Administration (FDA) for metastatic NSCLC, researchers started to investigate their activity in SCCHN. As single agents, both compounds have a lower activity, but combined with chemotherapy, they can induce acceptable clinical responses, and hence, several clinical trials are studying them with different combination [1115]. However, the anti-EGFR agent which has encountered more success in SCCHN management has been cetuximab, a chimeric monoclonal antibody (MoAb) against the extracellular domain of the EGFR. This compound has fueled the enthusiasm for investigating its role as a targeted agent against tumorigenesis in SCCHN driven by EGFR pathway and to combine it with different agents in different clinical settings (“neoadjuvant” or induction treatment, adjuvant, and metastatic setting). Furthermore, the success of this compound has pushed researchers to look into alternative mechanisms against the tumor biology of SCCHN, once cetuximab has proved that these tumors are not exempted to this kind of therapy manipulation. Noteworthy, cetuximab has shown not only clinical activity but also survival advantage in the relapsed and/or metastatic disease [3, 4]. Cetuximab was initially approved by the US FDA to be used in locally advanced SCCHN in combination with RT [4]. Further studies have shown that this chimeric MoAb improved overall response rate (ORR) over cisplatin alone or when combined with this platinum salt in recurrent/metastatic SCCHN. Cetuximab has also shown efficacy in platinum-failure cases [16, 17]. Recently, cetuximab showed to improve response rate, progression-free survival (PFS), and OS when combined with cisplatin/5-fluoruracil (5-FU) in the metastatic setting [3].

Herein, we discuss the latest clinical trials targeting EGFR pathway as well as some of the novel dual inhibitors tested in SCCHN.

Epidermal growth factor receptor (EGFR) pathway

The EGFR is abnormally activated in head and neck cancer [18]. The EGFR is a member of the erbB family of four related cell membrane receptors. This includes EGFR (Her-1 or erbB1), erbB2 (Her-2), erbB3 (Her-3), and erbB4 (Her-4). All are transmembrane glycoprotein receptors, each of which has an extracellular ligand-binding domain. For the past two decades, the role of the EGFR (ErbB1/HER1) in the initiation and progression of several epithelial malignancies has been widely studied [19]. As part of the ErbB family of receptor TKs, the EGFR consists in an extracellular ligand-binding domain, a hydrophobic transmembrane component and an intracellular domain with TK activity [19, 20]. When activated, a ligand-induced dimerization in the ErbB monomer leads to trans-autophosphorylation of adjacent receptors monomers, resulting in recruitment and phosphorylation of several intracellular substrates, stimulating signaling pathways that drive transcriptional programs controlling cell proliferation, survival, differentiation, and DNA repair [20, 21]. Alterations in these signaling pathways can be secondary to several mechanisms, including overexpression of EGFR ligand and/or the receptor, gene amplification, activating mutations, and loss of negative regulatory mechanisms [20]. Somatic mutations in the tyrosine kinase domain of the EGFR gene described in NSCLCs have been associated with increased sensitivity to EGFR-specific TKIs. However, these mutations are rare in head and neck cancers. In gliomas, a truncation mutation, the EGFR variant III (EGFRvIII) has been commonly described and associated with poor prognosis. In SCCHN, EGFRvIII expression was detected in 42% of SCCHN tumors where EGFRvIII was always found in conjunction with wild-type EGFR [22]. Interestingly, EGFRvIII positive SCCHN cells have shown growth inhibition after treatment with cetuximab [22]. On the other hand, EGFR overexpression has been found in up to 90% of SCCHN and is associated with radioresistance and a decrease in both disease-free (DFS) and cause-specific survival [2326].

EGFR inhibition by TKIs

Several therapeutic approaches against the EGFR have been used, but only the MoAbs directed against the extracellular domain of EGFR, and the small-molecule EGFR TKIs which target the intracellular domain have demonstrated clinical activity and are currently in clinical use. The TKIs such as erlotinib and gefitinib cross the plasma membrane and compete reversibly with ATP to bind to the intracellular catalytic domain of EGFR TK, thereby interfering with the receptor enzymatic activity and downstream signaling [27]. Because Her-2 is the preferred dimerization partner for EGFR and the EGFR/Her-2 heterodimers can potentiate receptor signaling and resistance to EGFR inhibitors, lapatinib, a TKI of both EGFR (ErbB1/Her-1) and Her-2 has also been studied in SCCHN [27, 28].

Erlotinib

Erlotinib as single agent

Erlotinib (Tarceva®, Genentech, South San Francisco, California, USA and OSI Pharmaceuticals, Melville, NY, USA) is a reversible small-molecule TKI of EGFR currently approved as monotherapy for the treatment of patients with locally advanced or metastatic NSCLC after failure of at least one prior chemotherapy regimen. The therapeutic potential of erlotinib in SCCHN was initially described in a phase I study conducted by Hidalgo et al. [29] where erlotinib produced stable disease (SD), lasting 15 months in one patient with SCCHN. A phase II trial was later developed by Soulieres et al. [13] using erlotinib as a single agent in patients with locally recurrent and/or metastatic SCCHN with an initial dose of 150 mg daily. In the 115 patients enrolled, the ORR reported by intention-to-treat analysis was 4.3% and disease stabilization was achieved in 38.3% for a median duration of 16.1 weeks [13]. The most common drug-related toxicities were rash and diarrhea and there was a significant better OS in patients who developed skin rashes than those who did not, conferring this clinical observation the potential to serve as a predictor marker [11, 3033].

As neoadjuvant treatment, the safety and efficacy of erlotinib was evaluated in a clinical trial conducted by Thomas et al. [34]. Herein, 35 (31 evaluable) patients with locally advanced non-metastatic SCCHN and pending surgical management were enrolled. Patients were treated with erlotinib at 150 mg/day for a short period (18–29 days) and even though the results could not be confirmed because the patients were undergoing surgery, 9 of the 31 evaluable patients had a decrease in tumor size of at least of 25%. Interestingly, in a retrospective analysis, baseline p21waf expression in the basal-like cell layer, but not EGFR gene copy number, was statistically positively correlated with clinical response to treatment [34].

Erlotinib in combination with conventional chemotherapy

Based on the results of erlotinib as a single agent in patients with advanced head and neck cancer and the fact that preclinical data demonstrated possible additive anti-tumor effects of erlotinib in combination with docetaxel, the idea of using erlotinib with conventional chemotherapy was feasible [35]. Thus, a phase I/II trial was conducted by Siu et al. [15] to determine the phase II dose and objective response rate of erlotinib in combination with cisplatin in 51 patients with recurrent or metastatic SCCHN with no prior chemotherapy. This study reported one complete response (CR) and eight partial responses (PR) with an ORR of the 21% and disease stabilization in 49% of the 43 evaluable patients [15]. In the same group of patients, Agulnik et al. [36] studied the pre-treatment and on treatment tumor and skin biopsies; their results suggested that a high copy of EGFR gene in tumor specimens may predict response to therapy and a decrease in phosphorylated EGFR with therapy correlated with increased OS. However, these biomarkers were examined on already heavily pretreated patients and with the combination of cisplatin plus erlotinib regimen, not erlotinib alone.

In the 2007 American Society of Clinical Oncology (ASCO) annual meeting, Kim et al. [37] presented the final results of a phase II study using erlotinib, docetaxel, and cisplatin in 50 patients with advanced SCCHN. Treatment included administration of docetaxel at 75 mg/m2 and cisplatin at 75 mg/m2 intravenously every 3 weeks plus erlotinib at 150 mg by mouth daily. The results of the trial were very encouraging with an ORR of 67%, disease control rate (DCR) of 95% and with survival benefits superior than most of the trials reported in patients with EGFR TKIs in combination with chemotherapy (see Table 1). Grade 3–4 toxicity included febrile neutropenia, dehydration, diarrhea, and gastrointestinal bleeding in 4–12% of the patients.
Table 1

Phase II clinical trials using erlotinib and gefitinib as single agent or in combination with chemotherapy in head and neck cancer

Study

Disease state

Treatment

No. of patients in study

ORR (%)

Median PFS (months)

OS (months)

Wheeler et al. [45]

Metastatic or recurrent

Gefitinib

32

9.3

3

6

Cohen et al. [11]

Metastatic or recurrent

Gefitinib

52 Enrolled (47 pts evaluable)

10.6

3.4

8.1

Kirby et al. [12]

Metastatic or recurrent

Gefitinib

47

8 (DCR: 36)

NR

3.4

Soulieres et al. [13]

Metastatic or recurrent

Erlotinib

150

4.3

2.4

6.0

Siu et al. [15]

Metastatic or recurrent

Erlotinib + cisplatin

51

21

3.3

7.9

Kim et al. [37]

Metastatic or recurrent

Erlotinib + docetaxel + cisplatin

50 Enrolled (43 pts evaluable)

67

6.01

11

Belon et al. [14]

Metastatic or recurrent

Gefitinib + docetaxel + cisplatin

17 Enrolled (16 eligible for response)

62.5

5.1

NR

ORR overall response rate, PFS progression-free survival, OS overall survival, DCR disease control rate (complete remission + partial remission + stable disease), NR not reported

Erlotinib plus combined modality

EGFR overexpression is associated with radioresistance and preclinical data indicates synergism between erlotinib and both conventional chemotherapy and RT [35, 38]. Savvides et al. [39] studied erlotinib in combination with docetaxel and RT in a phase I trial that enrolled 23 chemonaive patients with stage III-IVB SCCHN. Three dose-limiting toxicities (DLTs) were reported including a death within 30 days from last treatment in one patient and grade 3–4 mucositis; the study reported CR post-concurrent chemoradiation in 15 patients [39]. In a phase II trial which included 25 patients with stages III and IV SCCHN, Herchenhorn et al. [40] combined erlotinib with cisplatin and RT. Although 84% of the patients had complete pathological response (pCR), grade 3–4 in-field dermatitis was reported in 56%, and 84% of the patients required enteral feeding. The results of these trials have shown promising clinical efficacy of erlotinib in combination with chemoradiation albeit increased local toxicity of the regimen; however, the clinical benefit warrants larger cohort trials.

Erlotinib plus other novel targeted agents

Although EGFR TKIs have shown clinical activity in SCCHN, this has been quite modest. In the search for an explanation of these results, the use of concomitant targeted therapy agents has also been explored. Dual targeted inhibition using cetuximab and either erlotinib or gefitinib has shown to increase EGFR signaling inhibition, and hence, it caused more profound tumor regression than either treatment alone [41]. In addition, bevacizumab (a MoAb against the vascular endothelial growth factor ligand) has been combined with erlotinib and irradiation; this has also shown synergistic effects in SCCHN cells [42]. In a phase I/II trial by Cohen et al. [43] using bevacizumab and erlotinib in 51 patients with advanced SCCHN with up to 1 prior chemotherapy regimen for recurrent disease, no DLT toxicities were observed during the phase I of the study. A total of 46 patients were treated at the phase II with erlotinib at 150 mg oral daily and bevacizumab at 15 mg/kg every 3 weeks. The authors reported four patients with CR and three with PR. This novel combination was well tolerated, and despite having modest response rates, it was encouraging.

Gefitinib

Gefitinib as single agent

Gefitinib (Iressa®, ZD1839, AstraZeneca, UK) was the first oral active selective TKI of EGFR that was approved by the US FDA as a third line monotherapy for patients with locally advanced or metastatic NSCLC after failure of both platinum-based or docetaxel chemotherapy. Gefitinib, as erlotinib, has also shown antiproliferative effects and induced apoptosis in different cancer cell lines as well as human tumors [44]. In the clinical scenario, Cohen et al. conducted two phase II trials of gefitinib as a single agent with different dosages. The first trial, using 500 mg daily reported a response rate of 10.6% and DCR in 56% of the 52 patients with recurrent or metastatic SCCHN enrolled into the study [11]. In the second trial, using a dose of 250 mg daily in 70 patients, the response rate decreased to 1.4% postulating a probable dose–response relationship [30]. Diarrhea (grade 3) was the main adverse effect, and in both trials skin toxicity was a clinical predictor of better outcome. Similar responses were reported in another phase II trial in which there was a subjective symptom improvement in 34% of the patients treated, although sometimes with worse OS (see Table 1) [12, 45].

Gefitinib in combination with chemotherapy

In the preclinical setting, gefitinib has shown to have synergistic cytotoxic effects when combined with either docetaxel or cisplatin in head and neck cancer cell lines [46, 47]. A combination of docetaxel, gefitinib, and celecoxib (a Cox-2 inhibitor) showed in preclinical studies to be active, and the dual action of gefitinib with celecoxib seemed to have similar synergistic effects in the clinical setting [46, 48]. Docetaxel and cisplatin were used in conjunction with gefitinib in a phase II trial by Belón et al. [14]. DCR was 75% (95% CI 47.6–92.7) and median PFS was 5.1 months (95% CI 2.0–7.1). All 17 patients were assessed for safety. Grade 3–4 adverse effects were neutropenia (41.2%; febrile neutropenia: 23.5%), anemia (17.7%), asthenia, diarrhea, vomiting, anorexia, and leucopenia. Thus, this study reported a high response rate and encouraging PFS in this setting [14]. Likewise, a phase III trial was conducted by Steward et al. [49] in which 486 patients with advanced SCCHN were randomized to receive either gefitinib versus methotrexate; the trial showed no survival advantage in either arm and, although the gefitinib arm had the highest response rate, this was not statistically significant.

Gefitinib plus combined modality

Similar to erlotinib, when gefitinib is combined with RT, it has shown additive synergistic growth inhibition [50]. However, the only clinical study published so far combining gefitinib with RT alone was disappointing in terms of response, survival, and treatment-related toxicity [51]. When gefitinib was used with combined modality, the results were more encouraging. In this regard, Rueda et al. [52] developed a phase II study which enrolled 46 patients with locally advanced unresectable SCCHN using gefitinib plus concomitant boost-accelerated RT and concurrent weekly cisplatin. Fifty-two percent of the patients attained CR and 11% PR. The 2-year PFS and OS were 47 and 56%, respectively. Another phase I/II trial conducted by Doss et al. [53] added gefitinib to an induction regimen (carboplatin, paclitaxel, and 5-fluoruracil) followed by concurrent carboplatin/paclitaxel/RT. The study enrolled 45 patients with no previous therapy; ORR was 85% and PFS and OS at 1 year were 68 and 86%, respectively [53].

Perhaps, the most promising results using gefitinib was presented by Ahmed et al. [54] at the 2007 ASCO meeting. Herein, the regimen consisted of induction chemotherapy (carboplatin/paclitaxel) followed by concurrent chemorradiation (gefitinib, 5-fluorouracil, hydroxyurea, and RT), followed by gefitinib for 2 years after the start of combined modality. In 56 evaluable subjects, the observed CR was 91% and the estimated OS was 83% at 2 years and 73% at 3 years. Grade 3 and 4 toxicities were similar to other studies and included mucositis, dermatitis, rash, and diarrhea. Survival rates were slightly lower, but similar findings were also reported by Chen et al. [55]. Other approaches such as double EGFR blockage using gefitinib and cetuximab have shown a superior inhibition of EGFR-dependent signaling and induction of apoptosis in preclinical studies [56].

The impact of EGFR mutations in SCCHN

With the development of targeted agents in the past decade, one of the main goals in the therapeutic approach of SCCHN is to find predictive markers that may help clinicians to properly identify which subgroups of patients are likely to benefit from these agents. In the case of NSCLC, activating somatic mutations in the EGFR TK domain, usually in-frame deletions within exon 19 or the L858R mutant within exon 20, have been found to be tightly associated with sensitivity to the EGFR TK inhibition [57]. However, these mutations are rare in patients with SCCHN, being reported in only 0–7% [5860]. Even in gefitinib-responsive SCCHN patients, a mutation of ErbB2 rather than EGFR may be linked to treatment response [61].

Overexpression of EGFR, a more common scenario in SCCHN, is associated with poor outcome; however, it has not consistently been found to correlate with the antitumor activity of EGFR inhibitors [58, 62]. A preclinical study by Erjala et al. [63] however, suggested that EGFR gene amplification may predict sensitivity whereas high protein levels of other ErbB family members and activation status of ErbB2 are associated with resistance of SCCHN cells to gefitinib. Therefore, the incorporation of both EGFR copy number and activating mutations as predictive markers in the primary analysis of clinical trials is now imperative.

EGFR inhibiton by monoclonal antibody

MoAbs against EGFR bind to the extracellular domain of EGFR with high affinity and while on its inactive configuration, compete with ligand binding and block ligand-induced EGFR TK [20, 27]. Thus far, there are two anti-EGFR MoAbs approved by the US FDA: cetuximab (ERBITUX®, ImClone Systems Incorporated, Branchburg, NJ, USA) and panitumumab (Vectibix®, Amgen, Thousand Oaks, CA, USA). These two MoAbs compete for receptor binding by occluding the ligand-binding region and thereby, blocks ligand-induced TK activation [64, 65]. Cetuximab has shown to inhibit the proliferation of a variety of cultured human tumor cell lines that overexpress EGFR and to enhance the antitumor activity of several chemotherapeutic agents in xenograft models [6669]. Furthermore, it has also shown to be a radiosensitizer on EGFR-expressing A431 tumor cells in both in vitro and in vivo studies [70, 71].

Cetuximab in SCCHN

Cetuximab in combination with radiotherapy

Based on the results of a phase I trial conducted by Robert et al. [72] in which cetuximab was combined with RT in patients with advanced SCCHN, cetuximab was immediately tested in the first-line treatment in a multinational randomized phase III trial. Bonner et al. [4] randomized 424 patients with locoregional advanced SCCHN and who had not received chemotherapy in the previous 3 years, were radiation-naïve, and had not undergone through surgical resection. Arm 1 consisted of cetuximab plus RT (n = 211) and arm 2 was RT alone (n = 213). Patients on arm 1 received daily RT and weekly cetuximab until end of RT. The primary endpoint was duration of control of locoregional disease. OS, PFS, RR, and safety were all secondary endpoints [4]. All endpoints were statistically significant in favor of cetuximab/RT combination (locoregional control: 24.4 vs. 14.9 months, P = 0.005; OS: 49 vs. 29.3 months, P = 0.03; PFS: 17.1 vs. 12.4 months; P = 0.006). In addition, the combination arm did not increase common side effects associated with RT in this setting [4].

Cetuximab plus chemotherapy

Cetuximab has been combined with several conventional chemotherapy agents including platinum salts (e.g., oxaliplatin, cisplatin), taxanes, vinca alkaloid (e.g., vinorelbine), and other biologic agents (e.g., bevacizumab), and shown to improve its tumor toxicity. Cetuximab in combination with cisplatin was studied in patients with refractory metastatic or recurrent SCCHN or to platinum-refractory patients [73, 74]. The combination was shown to be active and most importantly, it did not exacerbate cisplatin toxicity; however, an association of skin rash and occasional serious allergic reactions were noted. In combination with taxanes, Knoedler et al. [75] presented in 2008 ASCO meeting a phase II clinical trial combining cetuximab plus docetaxel in patients who had stage III/IV recurrent and/or metastatic SCCHN and had failed a platinum-based therapy. Patients received weekly cetuximab at 250 mg/m2 until PD, and docetaxel given weekly at 35 mg/m2 on days 1, 8, 15 of a 4-week cycle for a maximum of 6 cycles. From the 45 patients (out of 47) evaluable for response, a clinical benefit of 47% was seen (20% PR and 27% SD). Treatment was well tolerated. Major grade 3–4 adverse events included gastric perforation (n = 1), pneumonia (n = 5), mucositis (n = 4), and skin toxicities (n = 9) [75].

In two phase III randomized trials, the addition of cetuximab proved to be superior to chemotherapy alone. In the first one, cisplatin plus cetuximab was compared against cisplatin plus placebo in patients with recurrent/metastatic SCCHN [16]. A total of 117 patients were analyzed for PFS which was the primary endpoint. The results of the study were as follows: median PFS (4.2 vs. 2.7 months); median OS: 9.2 versus 8.0 months (P = 0.21); ORR 26 versus 10% (P = 0.03). Thus, the addition of cetuximab albeit improving response rate, did not improve PFS and OS in this study (see Table 2) [16]. This trial, however, was underpowered for its survival endpoints. Also, the patients in the control arm that progressed were allowed to crossover to the cetuximab arm, factor that may have been a confounding factor in the evaluation of the OS. The second phase III randomized trial (EXTREME study) added cetuximab to platinum salt (cisplatin or carboplatin) and 5-FU [3]. Herein, the triplet was tested against platinum/5-FU-based chemotherapy in the frontline treatment of recurrent or metastatic SCCHN. A total of 442 patients who had stage III/IV recurrent and/or metastatic SCCHN, not suitable for local therapy, were randomized to receive either cetuximab/platinum agent (cisplatin or carboplatin)/5-FU for a maximum of six cycles every 3 weeks (n = 222) or platinum (cisplatin or carboplatin)/5-FU at the same dosing (n = 220). Cetuximab was administrated until PD or unacceptable toxicity. This landmark trial reached all its primary and secondary endpoints: median OS, 7.4 versus 10.1 months (HR for death, 0.80; 95% CI, 0.64–0.99; P = 0.04); PFS, 3.3 versus 5.6 months (HR for progression, 0.54; P < 0.001); and response rate, 20 versus 36% (P < 0.001) [3]. No differences among grade 3–4 hematologic adverse events were seen within the two groups. However, sepsis occurred in nine patients in the cetuximab group and in one patient in the chemotherapy-alone group (P = 0.02). There were no cetuximab-related deaths.
Table 2

Landmark phase III clinical trials using cetuximab with radiation therapy or systemic conventional chemotherapy

Study

Clinical setting

Treatment

No. of pts

ORR

PFS (months)

OS (months)

Burtness et al. [16]

Metastatic or recurrent, chemotherapy naïve

Cisplatin vs. ABX/cisplatin

57 vs. 60

10 vs. 26

P = 0.03

2.7 vs. 4.2

P = 0.09

8.0 vs. 9.2

P = 0.21

Bonner et al. [4]

Locally advanced

ABX/RT vs. RT

211 vs. 213

74 vs. 64

P < 0.001

17.1 vs. 12.4

P < 0.006

49.0 vs. 29.3

P < 0.03

Vermorken et al. [3]

Metastatic or chemotherapy naive

ABX/platinum/5-FU vs. platinum/5-FU

222 vs. 220

30.6 vs. 19.5

P = 0.0001

5.6 vs. 3.4

P < 0.0001

10.1 vs. 7.4

P = 0.036

ABX panitumumab, pts patients, ORR overall response rate, PFS progression-free survival, OS overall survival

For the past 2 years, the addition of a taxane to cisplatin/5-FU-based chemotherapy has become the standard triplet as induction or “neoadjuvant” treatment for SCCHN based on three large randomized clinical trials (TAX 323, TAX 324, and GORTEC) which have shown to improve OS [57]. Hence, the addition of cetuximab to an efficacious “triplet” induction therapy is an engaging and provocative concept due to safety, tolerability, and efficacy shown by MoAbs in general. In this regard, Kuperman et al. [76] conducted a retrospective analysis in which cetuximab was used in combination with induction chemotherapy for locally advanced SCCHN. Twenty-three locally advanced SCCHN patients were treated with TPF plus cetuximab (TPF-C) for 1–3 cycles as induction regimen (docetaxel and cisplatin both at 75 mg/m2 on day 1 and 5-FU 750 mg/m2/day for 3 days) given at 3 week intervals [76]. Twenty-one out of 23 patients were evaluable for tumor response. ORR was 71% and grade 3–4 adverse events during induction TPF-C were febrile neutropenia (4), sepsis (1), and infusion reactions (4). There were no deaths reported in this phase. Those patients with favorable tumor response (CR or PR) to induction TPF-C were treated with definitive RT-based therapy, whereas others were treated with surgery first. Later, TPF-C novel combination was formally studied in a phase I trial by Tishler et al. [77] in which the primary goal was to determine MTD of 5-FU as part of this combination. A standard dose escalation scheme of 5-FU was used (three patients enrolled on each cohort). 5-FU tested doses were 700, 850, and 1,000 mg/m2/day for 4 days via continuous infusion. The doses for cetuximab, docetaxel, and cisplatin were as follows: 400 mg/m2 initial loading dose followed by 250 mg/m2 (weekly), 75 mg/m2 (every 21 days), and 100 mg/m2 (every 21 days), respectively. After induction treatment, all patients received concurrent chemoradiotherapy. Nineteen chemonaïve patients were enrolled [77]. No DLT was encountered on dose levels 1 and 2. At dose level 3 (1,000 mg/m2), one DLT was encountered (febrile neutropenia) and three more patients were enrolled with no further DLT seen. In the expansion cohort at the MTD, two patients had grade 4 GI toxicity consisting of abdominal enteritis requiring hospitalization and blood transfusion. Due to these adverse events, 5-FU at level 2 was selected. Fifteen patients have completed all three cycles of TPF-C at the time of presentation. The response rate found after completion of the entire treatment plan was 11 patients attained CR, and 1 PD. Thus, TPF-C regimen is a feasible, safe and an active combination, supporting the prior results from Kuperman et al. in their retrospective analysis. By reducing 5-FU in TPF to 850 mg/m2, it also reduces GI toxicity. This study continues at the declared MTD dose level 2 of 5-FU (850 mg/m2) [77].

Cetuximab plus concurrent chemoradiation in SCCHN

Two large randomized clinical trials (RTOG 9501 and EORTC 22932) have established the use of high-dose cisplatin in combination with radiation as the “preferred” concurrent chemoradiation modality for patients with SCCHN [78, 79]. Thus, it looks very compelling and promising to study the addition of cetuximab into this regimen. In this sense, Tsoutsou et al. [80] reported the initial results of 19 chemonaïve patients who had locally advanced SCCHN and were treated with cisplatin, cetuximab, and RT. Patients received accelerated concomitant boost RT (2.7 Gy/day to a total dose of 59.4 Gy in 4.5 weeks, biological equivalent dose to 70 Gy) combined with weekly cisplatin at 30 mg/m2 and cetuximab at conventional delivery weekly thereafter. Patients also received cytoprotection with amifostine at 1,000 mg/day subcutaneously. Mucositis was the cause of RT delay for 1 week in 6/19 (32%) patients. Noteworthy, no severe late RT sequelae have been observed. Out of 14 patients with measurable disease, 12 (86%) had CR, 1 PR (7%), and 1 (7%) minimal response. In this cohort, one patient had progression of disease locally, another one at a distant organ, and one patient died from intercurrent disease after a 12-month follow-up [80].

Another trial using cisplatin (100 mg/m2 intravenously weeks 1 and 4), cetuximab (400 mg/m2 intravenously week 1, followed by 250 mg/m2 weeks 2 to 10) combined with concomitant boost RT (70 Gy) included 22 stage III or IV patients with SCCHN [81]. The study, however, was closed for significant adverse events, including two deaths (one pneumonia and one unknown cause), one myocardial infarction, one bacteremia, and one atrial fibrillation. In terms of survival, the 3-year OS and PFS rate were 76 and 56%, respectively; the 3-year locoregional control rate was 71%. In summary, although the combination of cisplatin/cetuximab/RT has encouraging response rates, its safety profile still needs further investigation.

Panitumumab in SCCHN

Panitumumab (Vectibix, Amgen, Inc., Thousand Oaks, CA) is a recombinant, fully human IgG2 kappa MoAb that binds specifically to the human EGFR. Panitumumab, formerly called ABX-EGF, was initially developed by Abgenix using the Xenomouse transgenic technology. This is based on inactivating the mouse Ig genes that are replaced by a megabase gene containing the human heavy and kappa chains. The result is a creation of fully humanized antibodies that do not contain murine portions of the IgG molecule as chimeric antibodies do (e.g., cetuximab, rituximab). Less frequent hypersensitivity reactions and a longer half-life is achieved by avoiding the formation of human anti-mouse antibodies (HAMA) [82].

Panitumumab binds specifically and selectively to EGFR on both normal and tumor cells, and competitively inhibits the binding of ligands for EGFR. In in vitro models, this action reduces EGFR signaling and causes cell cycle arrest [65]. The mechanism of in vivo antitumor activity may involve downregulation of EGFR expression by triggering receptor internalization, induction of apoptosis triggered by blocking EGFR signaling pathways and induction of cell cycle arrest, and inhibition of angiogenesis [8286]. Given the fact that panitumumab is a fully humanized MoAb which essentially lacks effector functions, complement-dependent (CDC) and antibody-dependent cell-mediated cytotoxicities (ADCC) likely do not account for the in vivo antitumor activity [65].

The initial studies with panitumumab have focused on metastatic colorectal cancer. Van Cutsem et al. [87] conducted a randomized phase III clinical trial in which panitumumab was compared against the best supportive care in patients who had previously been treated with fluoropyrimidine, oxaliplatin, and irinotecan. A 10% response rate was reported with a significant reduction in the risk of tumor progression; however, no difference in OS was observed. This led to the FDA approval of panitumumab as monotherapy for third-line treatment of colorectal cancer, refractory to fluoropyrimidines, oxaliplatin or irinotecan. From there, several studies were launched to study panitumumab in other tumors including metastatic or recurrent SCCHN (see Table 3). Wirth et al. [88] conducted a phase I study of panitumumab in addition to carboplatin, paclitaxel, and intensity-modulated RT (IMRT) in 19 treatment-naïve patients with stage III/IV SCCHN. Paclitaxel was given at two dose levels (15 and 30 mg/m2 weekly for 7 weeks) with fixed-dose carboplatin and panitumumab (AUC of 1.5 and 2.5 mg/kg weekly, respectively) plus IMRT (70 Gy). Three patients were treated in dose level 1, and 16 in dose level 2. A grade 4 febrile neutropenia was the DLT in dose level 2. Other Grade 3/4 adverse events include dysphagia, mucositis, acneiform rash, radiation dermatitis, and nausea. Neither dose-reduction nor RT breaks >3 days were necessary. There were no grade 3/4 chronic toxicities. Of 15 patients evaluable for response, 13 (87%) had CR [88]. This combination seems safe and effective and, therefore, needs to be evaluated in a phase II trial.
Table 3

Current studies evaluating the role of panitumumab in head and neck cancer [100]

Study type

Disease setting

Treatment

Number of patients

Sponsor entity

Phase II (randomized)

Unresected, locally advanced head and neck cancer

Cisplatin/RT vs. ABX/RT

150

Amgen

Phase I

Locally advanced head and neck cancer

ABX/chemo-RT vs. ABX/induction chemo followed by chemo-RT

44

Dana-Farber Cancer Institute

Phase II (randomized)

Unresected, locally advanced head and neck cancer

Cisplatin/RT vs. cisplatin/ABX/RT

150

Amgen

Phase II

Postoperative setting

Cisplatin/ABX/RT

43

University of Pittsburgh

Phase II (PRISM)

Metastatic or recurrent head and neck cancer (second-line therapy)

ABX monotherapy

50

Amgen

Phase III

Locally advanced stage III/IV SCCHN

Cisplatin/RT vs. ABX/RT

320

NCI Canada

Phase III (SPECTRUM)

Metastatic or recurrent head and neck cancer

Cisplatin/5-FU vs. cisplatin/5-FU/ABX

650

Amgen

ABX panitumumab, RT radiation therapy, chemo-RT chemoradiation, PRISM panitumumab regimen in second-line monotherapy of head and neck cancer, SPECTRUM study of panitumumab efficacy in patients with recurrent and/or metastatic head and neck cancer

Dual inhibition of the EGFR in head and neck cancers

Lapatinib, a dual TKI (EGFR and Her-2/neu receptor), has also been explored in the treatment of SCCHN. As a single agent, lapatinib has not shown activity in recurrent and/or metastatic SCCHN [89]. Further studies combining lapatinib with RT has shown provocative results. In a phase I trial, escalating lapatinib doses (500, 1,000, and 1,500 mg/day) were combined with cisplatin (100 mg/m2 on days 1, 22, and 43) plus standard fractionation of RT [90]. Thirty-one patients with locally advanced SCCHN were enrolled, reporting as DLT a perforated ulcer in one patient in the 500-mg cohort and transient elevation of liver enzymes in one patient in the 1,000-mg cohort. The regimen induced an 81% ORR (16 CR and 9 PR). Due to these encouraging results, a phase II trial as well as a phase III randomized, double blind, placebo-controlled, multicenter trials are ongoing. In the phase III trial, lapatinib is being tested in the post-operative setting in combination with concurrent chemoradation for high-risk SCCHN. Lapatinib or placebo maintenance treatment will continue for 1 year and will determine if this approach is effective in reducing recurrence rate in this high-risk group of patients [91].

Biomarkers for EGFR treatment

EGFR-amplification or overexpression has been found to be a poor prognostic marker in SCCHN. The activation of EGFR leads to a phosphorylation cascade mediated via TK, and signaling downstream through the MAPK, Akt, ERK, and Jak/STAT pathways which are involved into proliferation, apoptosis, invasion, angiogenesis, and metastasis processes. Furthermore, studies have also shown that EGFR may serve as a predictive marker for response when an anti-EGFR agent, either a small inhibitor or a MoAb, is chosen as therapy [92].

Revisiting the issue of prognosis, EGFR overexpression (detected by IHC) has not been the only factor studied and associated with survival, but also the EGFR copy number amplification and mutations of the gene [93, 94]. However, the incidence of EGFR mutation in SCCHN is low when compared with other tumor types such as lung cancer, and hence, its practical use in clinic may be more difficult [95, 96].

In regard to cetuximab, studies done in colorectal cancer (CRC) have shown that K-ras status is important to stratify which CRC patients will benefit from anti-EGFR therapy by cetuximab [97]. The analyses from a phase III trial (CRYSTAL) showed a significant improvement in PFS and best ORR when cetuximab was added to 5-FU/leucovorin/irinotecan (FOLFIRI) regimen in the first-line therapy for metastatic CRC in patients who carry wild-type K-ras (PFS: P = 0.0167, HR estimate 0.68, CI: 0.051–0.934; best ORR: 59.3 vs. 43.2%, P = 0.0025) [97]. This benefit was not observed in those patients whose tumors carried the mutated K-ras gene. Thus, K-ras mutation status is a predictive marker for treatment with cetuximab in metastatic CRC patients. The question is what about other tumor types such as SCCHN where cetuximab has become an important component of the treatment armamentarium? In head and neck cancer, studies on K-ras mutation status have shown inconsistent results in terms of its impact on response to treatment (prediction) and prognosis. Bissada et al. presented interesting data at the 2008 ASCO meeting regarding the prevalence of K-ras codon 12 mutations in locally advanced SCCHN patients treated with chemoradiation ± surgery; the impact of K-ras in terms of locoregional control, OS, DFS, and distant metastasis free survival was also evaluated. The prevalence of K-ras codon 12 presence from 197 cases in which DNA extraction was successful was low (3.5%). The results for mutated versus non-mutated cases were as follows: CR to chemoradiation (71 vs. 73%, P = 0.32), locoregional control (83 vs. 32%, P = 0.03), DFS (68 vs. 27%, P = 0.12), distant metastasis-free survival (81 vs. 100%, P = 0.30), and OS (65 vs. 57%, P = 0.14), all them at 3-year mark. The authors concluded that K-ras mutational status is not associated with response; however, it may be associated with failure pattern and aggressiveness in general which may affect the selection of treatment for these patients [98].

Furthermore, tumor biology has described an interaction between EGFR and K-ras pathways. Recently, Stoehlmacher et al. [99] showed that EGFR polymorphisms (HER-497, EGFR-216) may be associated with PFS in metastatic CRC treated with erlotinib single agent. This report is the first one showing a relationship between EGFR polymorphisms and erlotinib in this disease. Certainly, it would be interesting to see if similar results are reproducible in SCCHN patients, and what is the role of EGFR polymorphisms and K-ras mutations for these small molecules of TKIs.

Conclusion

The introduction of targeted agents against the EGFR pathway by either small molecule inhibitors such as TKIs (e.g., erlotinib, gefitinib, lapatinib) or MoAbs like cetuximab, panitumumab, and others represents a progress in the management of SCCHN beyond the use of RT alone or conventional cytotoxic agents. In general, these novel agents have a better toxicity profile which allow investigators to combine them not only with RT modalities but also to explore the issue of maintenance therapy as it was explored in the EXTREME trial and adjuvant setting as is being investigated with lapatinib in an ongoing clinical trial.

Tumor biology continues providing data of the crucial role that EGFR pathway plays in proliferation, survival, anti-apoptosis, and metastatic potential. EGFR is part of a complex network with other critical tumorigenesis pathways such as K-ras and VEGF. Studies are ongoing nowadays to discover these relationships that may help us to either identify novel targets or to understand better the intrinsic signaling among them. The presence or not of gene mutation in the EGFR domain or K-ras partially explains our clinical observations. However, polymorphisms found on these genes may play a crucial role in the development and progression of tumors as well as their sensitivity to our novel treatments. These and other important questions are waiting for well-designed clinical trials to explore these carcinogenesis-driven forces.

Conflict of interest statement

Luis E. Raez received research funding from Bristol-Myers-Squibb, Genentech, Imclone Systems Incorporated, and Sanofi-Aventis and serves as Speaker's Bureau for Genentech, Imclone System Incorporated, OSI Pharmaceuticals, and Sanofi-Aventis. Edgardo S. Santos has served as Speaker's Bureau for Genentech, GlaxoSmithKline, Sanofi-Aventis, and Amgen. Cesar A. Perez and Chancellor E. Donald have nothing to disclose.

Copyright information

© Springer-Verlag 2009