Mutations in the epidermal growth factor receptor (EGFR) gene in triple negative breast cancer: possible implications for targeted therapy

Triple negative breast cancers, defined by the lack of estrogen receptor (ER), progesterone receptor (PR) and epidermal growth factor receptor 2 (Her2/cerbB2/EGFR2) expression, account for 10 to 20% of all breast carcinomas in Asian and Western populations [1–7], but occur at much higher frequencies in individuals of African descent [1–3, 8]. These tumours are usually of higher histological grade (Grade 3) [1, 3, 4, 6, 9, 10] and are associated with distinctive metastatic patterns [9, 11], shorter time to recurrence and earlier mortality [9, 11, 12]. Recent focus on this breast cancer subtype relates to resistance to endocrine and anti-HER2 directed therapy, phenotypic similarity to breast cancers in BRCA1/2 mutation carriers and the development of polyADP-ribose polymerase (PARP) inhibitors which have demonstrated promising activity in this disease. Despite this breakthrough, sustained complete remissions in advanced triple negative breast cancer are rare and additional therapies directed against appropriate molecular targets are needed.

EGFR is a receptor tyrosine kinase important in transducing extracellular signals from the cell surface to the cell interior, mediating crucial processes such as cell proliferation, differentiation, migration and apoptosis. Dysregulated expression of these receptors can lead to aberration of homeostatic cellular processes, resulting in malignant transformation of cells. Activating EGFR mutations have been reported in cancers such as non-small cell lung cancer (NSCLC) and head and neck cancers and are predictive of response to gefitinib or erlotinib therapy [13–15]. EGFR protein is expressed in 30% to 52% of triple negative breast cancers [7, 16, 17] and up to 60% of the closely related basal-like breast cancers and is associated with poor prognosis [18–21]. These observations are the basis for a number of ongoing clinical trials which are exploring the role of monoclonal antibodies against EGFR such as cetuximab and EGFR tyrosine kinase inhibitors such as erlotinib in triple negative breast cancer.

Many mutations in the EGFR gene have been reported in NSCLC but only a few have been validated, either from in vitro studies or from tumour responses in NSCLC patients, to be associated with responses to EGFR tyrosine kinase drugs [13, 14]. These mutations are usually found in exons 18, 19, 20 and 21, and include missense substitutions such as G719A/S and L858R and deletions like E746 to A750 (removal of amino acids Glucine-Leucine-Arginine-Glucine-Alanine (ELREA)) which are associated with sensitivity to tyrosine kinase inhibitors [13, 14]. Mutations associated with resistance to EGFR tyrosine kinase inhibitors are D761Y [22] and T790M [23, 24]. We sought to determine whether such mutations exist in triple negative breast cancers, the results of which may help to select patients suitable for inclusion in clinical trials evaluating the role of anti-EGFR directed therapies in this condition.

In this study, we report that 8 of 70 samples (11.4%) of triple negative breast cancers harbor EGFR mutations, including exon 19 deletions, inversions and exon 21 missense substitutions, which may predict sensitivity to EGFR tyrosine kinase drugs, thus suggesting a rationale for the clinical applicability of detecting EGFR mutations in these tumours, and potential use of EGFR tyrosine kinase inhibitor therapy.

In this study, we report the presence of EGFR mutations, notably exon 19 deletions and exon 21 missense (L858R) mutations, in 11.8% of triple negative breast cancers evaluated. All samples showed heterozygous deletions, suggesting they are likely dominant and play a role in tumourigenesis [13]. EGFR mutations have been found to occur in 13% to 64% of all non-small cell lung cancers [27] and exon 19 deletions and L858R mutations account for >80% of all EGFR mutations detected in NSCLC [27]. More significantly, there is a plethora of studies demonstrating that patients with these exon 19 deletions and L858R mutations respond very well to EGFR inhibitors in NSCLC, as summarized in several reviews [27–29]. Exon 19 deletions and L858R mutations cluster around the ATP-binding pocket of the kinase domain and such mutations confer ligand-independent activation and increased activation duration compared to wild-type receptors [13, 14, 30, 31]. In vitro studies showed that exon 19 deleted mutants and L858R mutant receptors appear to be more sensitive to gefitinib inhibition as compared to wild type receptors [13, 14]. This could explain why NSCLC patients harboring such mutations respond better to EGFR tyrosine kinase inhibitors (TKIs) than patients without such mutations. Patients with exon 19 harboring deletions were found to have longer survival following treatment with gefitinib or erlotinib compared with those having L858R mutations in NSCLC [32, 33]; however, Marks et al. reported no difference in survival between exon 19 deletions and L858R mutations in the absence of EGFR targeted therapy [34].

Several variant types of exon 19 deletions have been reported in NSCLC where the majority (65% to 75%) have EGFR exon 19 deletions of the 15 bp type (ELREA) and only a minority (1 to 1.5%) have 24 bp exon 19 deletions (SPKANKEI) [35]. Conversely, in our triple negative breast cancers, we discovered that both exon 19 type deletions (del E746 to A750 and del S752 to I759) are seen at similar frequencies. Correlative studies where exon 19 deletions were found to predict good response to EGFR TKIs mainly include the 15 bp deletion type, while 24 bp deletion types were rarely included (probably due to its low frequency) or at best, analysed as a group with the 15 bp deletion mutant. To our best knowledge, there appear to be no reports addressing differing sensitivity of different EGFR exon 19 deletions to gefitinib or erlotinib therapy. Future studies can be done to determine whether there is a variation in sensitivity to EGFR TKIs between the 15 bp deletion type and 24 bp deletion type.

Bhargava et al. reported no EGFR mutations in 11 EGFR-amplified sporadic breast tumours examined, out of which eight tumours were triple negative [36]. In a group of 47 metaplastic breast carcinomas which belonged to a subset of basal-like breast cancers, no EGFR tyrosine kinase mutations were identified as well; however, the actual ER, PR and cerbB2 status of these tumours analysed were not explicitly clarified [21]. Generali et al. also reported no EGFR mutations in 42 sporadic breast tumours (no selection for triple negative breast tumours) [37]. In a study of 58 triple negative breast tumours from Japanese patients, using Taqman genotyping assays against 14 known EGFR mutations including those for exon 19 deletions and L858R missense, EGFR mutations were also not found [38]. However, Weber et al. detected a higher rate of EGFR missense mutations in BRCA1/2 positive tumours (45.8%) compared with sporadic breast cancers (14.6%) [39]. BRCA1-linked breast tumours typically exhibit triple negative expression (ER-/PR-/Her2-) [12], express basal cytokeratins 5 and 6, and EGFR [40–42] and show similar histopathological features to basal-like/triple negative breast cancers [42]. However, EGFR missense mutations identified in the BRCA1/2 tumours were different from those encountered in our triple negative breast cancer study, possibly due to different patient selection where the BRCA1/2 linkage of our study cohort is unknown. The apparent differing findings of the above reports with ours could suggest heterogenous EGFR genomic instability in different breast cancer groups and raises the need for selection of specific breast cancers for EGFR mutation analysis. Different patient ethnicity and backgrounds may also be a contributing factor to the contradictory results seen in our predominantly Chinese cohort with those of the Japanese study group.

The development of anti-EGFR directed therapy in triple negative breast cancer has been supported by the availability of some preclinical data. In vitro studies on effects of EGFR inhibition in triple negative breast cancer cell lines revealed that gefitinib inhibited EGFR phosphorylation, which led to reduced signaling by the mitogen activated protein kinase (MAPK) and Akt pathway and causing cell cycle arrest at G1 phase [43]. In addition, gefitinib enhanced chemotherapeutic response to both carboplatin and docetaxel in these cells [43]. In a Phase II trial of erlotinib in patients with advanced breast cancer, 2 of 69 patients had partial responses, one of which had triple-negative histology [44]. Clinical studies are currently underway to evaluate the efficacy of EGFR tyrosine kinase inhibitors in patients with triple negative breast cancers.

As observed from our study, there is incongruity between EGFR positive immunostaining and the presence of EGFR mutations. It appears that positive EGFR protein expression does not predict the presence of mutations in triple negative breast tumours; conversely, mutations have been found in negative EGFR immunostained breast tumours. Monoclonal antibodies recognize extracellular epitopes of the EGFR and mutations in the EGFR gene could confer structural conformational changes that render them unrecognizable by the antibodies; besides, there are also possibilities of mutations in other exons of the EGFR that were not probed in our study. Pinter et al. have described that EGFR mutations in lung adenocarcinomas are not consistently accompanied by EGFR protein positivity by standard immunohistochemistry [45]. They reported that some of the EGFR mutant patients with major clinical response to gefitinib or erlotinib showed negative EGFR immunostaining [45]. Other studies have also noted such disagreement between EGFR immunoexpression and presence of mutations detected via molecular diagnostic methods [37, 46]. These results suggest the importance of molecular diagnostic methods to identify EGFR mutations and subsequently those who will benefit from EGFR inhibitor therapy. Specific antibodies that can detect EGFR mutants have been developed by Brevet et al., and could be advantageous in early identification of candidates for EGFR mutational analysis via molecular methods in the near future [35].

The relatively small sample size used in this study renders it difficult to make significant statistical judgments. Nevertheless, the identification of important EGFR mutations in our randomly selected cases from the larger series of triple negative breast cancers highlights the need to understand the frequency and type of EGFR mutations in a larger cohort, in order to promote a deeper understanding of the prospective utility of screening for EGFR mutations in relation to the therapeutic use of EGFR tyrosine kinase inhibitors for triple negative breast cancer in a clinical trial setting.

This study demonstrated existing EGFR mutations in a relatively small study size of 70 triple negative breast cancers in a Singapore population (8 of 70 samples). Mutations observed included EGFR exon 19 deletions (4 of 70 samples ) and EGFR exon 21 substitutions (3 of 70 samples), both of which are commonly found in NSCLC and are good predictors of sensitivity to tyrosine kinase inhibition therapy. This advocates the promising application of gefitinib or erlotinib therapy in triple negative breast cancer, where the need to find new tailored treatment is critical. These findings also encourage larger scale prospective trials to evaluate the need for EGFR mutation screening for anti-EGFR treatment in triple negative breast cancers. It is also apparent that EGFR positivity by standard immunohistochemistry is not necessarily accompanied by EGFR mutations, suggesting that molecular diagnostic methods appear to be more important for selection of potential prospective patients with triple negative breast cancers who may benefit from EGFR inhibitor therapy.