Preferential Response of Basal-Like Head and Neck Squamous Cell Carcinoma Cell Lines to EGFR-Targeted Therapy Depending on EREG-Driven Oncogenic Addiction.

The management of locally advanced head and neck squamous cell carcinoma (HNSCC) with Cetuximab, a monoclonal antibody targeting the epidermal growth factor receptor (EGFR), achieves only moderate response rates, and clinical trials that evaluated EGFR-blockade with tyrosine kinase inhibitors (TKI) yielded disappointing results. Inter-tumor heterogeneity may hinder the therapeutic efficiency of anti-EGFR treatments. HNSCC heterogeneity was addressed in several studies, which all converged towards the definition of molecular subgroups. They include the basal subgroup, defined by the deregulated expression of factors involved in the EGFR signaling pathway, including the epiregulin EGFR ligand encoded by the EREG gene. These observations indicate that basal tumors could be more sensitive to anti-EGFR treatments. To test this hypothesis, we performed a screen of a representative collection of basal versus non-basal HNSCC cell lines for their sensitivity to several anti-EGFR drugs (Cetuximab, Afatinib, and Gefitinib), tested as monotherapy or in combination with drugs that target closely-linked pathways [Mitogen-activated protein kinase kinase/ extracellular signal-regulated kinases (MEK), mammalian Target of Rapamycine (mTOR) or Human Epidermal growth factor Receptor 2 (HER2)]. Basal-like cell lines were found to be more sensitive to EGFR blockade alone or in combination with treatments that target MEK, mTOR, or HER2. Strikingly, the basal-like status was found to be a better predictor of cell response to EGFR blockade than clinically relevant mutations [e.g., cyclin-dependent kinase Inhibitor 2A (CDKN2A)]. Interestingly, we show that EGFR blockade inhibits EREG expression, and that EREG knock-down decreases basal cell clonogenic survival, suggesting that EREG expression could be a predictive functional marker of sensitivity to EGFR blockade in basal-like HNSCC.

Novel molecular diagnostic approaches for X-linked centronuclear (myotubular) myopathy reveal intronic mutations.

X-linked centronuclear myopathy (XLMTM), also called myotubular myopathy, is a severe congenital myopathy characterized by generalized hypotonia and weakness at birth and the typical histological finding of centralization of myo-nuclei. It is caused by mutations in the MTM1 gene encoding the 3-phosphoinositides phosphatase myotubularin. Mutations in dynamin 2 and amphiphysin 2 genes lead to autosomal forms of centronuclear myopathy (CNM). While XLMTM is the most frequent and severe form of CNM, no mutations are found in about 30% of patients by sequencing all MTM1 exons. Moreover, the impact of MTM1 sequence variants is sometimes difficult to assess. It is thus important to devise a complete molecular diagnostic strategy that includes analysis of the myotubularin transcript and protein expression. We therefore developed novel antibodies against human myotubularin and showed that they are able to detect the endogenous protein by direct Western blot from muscle samples and from cultured cells. In conjunction with RT-PCR analysis we validated the consequences of missense and splice mutations on transcript integrity and protein level. We also detected and characterized a novel deep intronic mutation consisting of a single nucleotide change that induces exonisation of a conserved intronic sequence. Patients with centronuclear myopathy and no molecular diagnosis should be investigated for MTM1 defects at the cDNA and protein level.