Window opportunity studies in head and neck cancer to target the insulin growth factor-1 and the epidermal growth factor receptors : from bed to bench

Despite progress in the management of patients with squamous cell carcinoma of the head and neck (SCCHN), the mortality rate of patients with advanced disease remains high. Monoclonal antibodies targeting the epidermal growth factor receptor (EGFR) have shown clinical benefits in the palliative and curative settings. However, only a minority of patients with recurrent or metastatic (R/M) SCCHN will have meaningful tumour regression with these agents, and virtually all will develop acquired tumour resistance after a few months of treatment. Therefore, research and development into new compounds that target activated pathways in SCCHN is of interest. Moreover, clinical study designs that allow the collection of pre- and post-treatment tumour samples are important to better understand the molecular mechanisms involved in treatment response or resistance. The first part of this work investigated the antitumour activity of figitumumab, a fully human monoclonal antibody IgG2 subtype which targets the insulin growth factor-1 receptor (IGF-1R) overexpressed in SCCHN, and which is associated with poor prognosis. In a phase II trial, patients with palliative SCCHN were treated with figitumumab IV 20 mg/kg, every three weeks. Despite encouraging preclinical results, figitumumab did not show any relevant clinical benefit and the study had to be closed prematurely after only 17 patients were included. In an exploratory analysis, we performed immunohistochemical analysis of the IGF-1R pathway on paired biopsies. Interestingly, we noticed that Akt was upregulated and further investigations supported cross-talk between the IGF-1R and the EGFR. Microarray analysis, confirmed by qRT-PCR, showed upregulation of cIAP2, an inhibitor of apoptosis upregulated by the Akt signaling pathway. Unfortunately this kind of study design is unlikely to fully exploit the anti-tumour potential of targeted agents because patients with palliative disease have received multiple treatment regimens (RT/CT/CRT/Surgery) leading to multifactorial resistance. Bearing in mind the lesson learned from the figitumumab phase II study, we modified the study design in the second part of this work where we investigated the clinical activity and the molecular response to cetuximab, a monoclonal antibody that targets the EGFR. We investigated cetuximab in the pre-operative window period in treatment-naïve SCCHN patients selected for primary curative surgery. The antibody was administrated for two weeks prior to surgery. 18FDG-PET, MRI, tumour biopsies and plasma samples were performed or collected before and after cetuximab. We observed 18FDG-PET partial response (EORTC guidelines) in 90% of the patients included in the cetuximab group versus 0% in the control group. Delta maximal standardized uptake values (ΔSUVmax) correlated with tumour cellularity on the surgical specimens (p=0.0004). For patients with ΔSUVmax < -25% or < -50%, Ki67 was significantly decreased upon cetuximab (p=0.01 and 0.003). Pathological examination of the surgical tumour specimens showed isolated tumour islands surrounded by a dense fibrotic matrix and inflammatory cells. Decrease of the tumour cell density and increase of the extracellular matrix were also recorded by diffusion MRI acquisitions and perfusion CT. Analysis of microarray data confirmed modifications in tumour composition with increased expression of matrikines and genes associated with cancer-associated fibroblasts (CAFs) and epithelial to mesenchymal transition (EMT). Whereas decorin seems to be related to tumour growth inhibition through direct blockage of multiple growth receptors and ligands, other upregulated genes associated with CAFs and even EMT, like CXCL12, are implicated in tumour progression and potential tumor re-growth. Independently from metabolic modifications, we identified two clusters of patients with different expressions of EMT markers, indicating that phenotypical tumour modifications are not correlated to initial clinical response and that one potential acquired resistance mechanism to cetuximab could be related to a mesenchymal cell phenotype. Whether EMT is related to the acquisition of a mesenchymal phenotype or the selection of such a phenotype is unknown. We therefore hypothesize that the association between cetuximab and medications that block mesenchymal cell types and CAFs could overcome resistance. Further, we investigated the inflammatory process by examining tumour infiltration by immune cells after cetuximab. We observed increased infiltration of the tumour by T cells and macrophages. In addition, increased expression of genes implicated in complement-derived cytotoxicity was also observed. This lead to the hypothesis that NK derived ADCC is probably not the only immunologic mechanism of cetuximab in SCCHN that should be investigated. These observations confirmed that besides a direct cytotoxic effect, multiple immunologic mechanisms may play a role in the activity of the medication.