Combined treatment of Nimotuzumab and rapamycin is effective against temozolomide-resistant human gliomas regardless of the EGFR mutation status

The EGFR signaling system is an attractive target for therapeutic intervention. EGFR gene amplification and overexpression account for approximately 40%-60% of GBMs [1,2]. In a report by Heimberger and colleagues, 42.6% of GBM patients failed to express EGFR, 25.9% had an overexpression of wtEGFR and 31.5% expressed a specific EGFR mutant (EGFRvIII, also known as EGFR type III, de2-7, ΔEGFR) [2]. In a subsequent study, 46% of GBMs lack EGFR [3]. Of note, EGFRvIII expression in GBM is frequently associated with amplification and co-expression of the wtEGFR [4]. Elevated levels of EGFR or EGFRvIII expression confer enhanced cell proliferation and invasion [5]. Given the high frequency of EGFR dysregulation, inhibiting EGFR signaling pathway appears to be a promising and rational therapeutic strategy for attenuating GBM growth.

Nimotuzumab is a humanized monoclonal antibody that targets the extracellular domain of EGFR and inhibits the binding of EGF ligands [6]. It has been extensively used in various solid tumors such as esophageal carcinoma [7], pancreatic cancer [8] and glioma [9,10]. Nimotuzumab has been shown to increase the sensitivity of radioresistant cancer stem cells when used in combination with radiation [11] and prolonged the survival of patients with concurrent radiochemotherapy treatment [12]. Hyperactivation of downstream phosphatidylinositol 3-kinase (PI3K)/AKT/mammalian target of rapamycin (mTOR) pathway is another common occurrence of human glioma [13,14]. mTOR-dependent processes plays a critical role in controlling mRNA translation, ribosome biogenesis, autophagy and metabolism [15]. Rapamycin, also known as Sirolimus, inhibits the highly conserved mTOR by forming a complex with its intracellular receptor, the FK506-binding protein [16]. The latter binds directly to mTOR complex 1 but not mTOR complex 2. Growth factors are known to stimulate mTORC1 through the PI3K/AKT pathway [17]. Rapamycin has been shown to be effective against intracerebral glioma xenografts and has a cytostatic effect against gliomas [18]. It is safe when co-administered with another EGFR tyrosine kinase inhibitor (Gefitinib) in several clinical trials [19,20] and could enhance cell death in combination with Nimotuzumab in epidermoid carcinoma cell line A431 [21].

In a recent study, wild-type epidermal growth factor receptor (wtEGFR) has been shown to phosphorylate a truncated EGFR (known as EGFRvIII), leading to the phosphorylation of STAT proteins and progression in gliomagenesis [22]. This finding may explain earlier report of why EGFRvIII expression rarely occurs without EGFR gene amplification [4]. Despite the fact that combination of EGFR targeting drugs and rapamycin has been used before, the effect of mono-treatment of Nimotuzumab, rapamycin and combination therapy in human glioma expressing different types of EGFR is not well-studied. Herein, we evaluated the efficacy of dual blockage using monoclonal antibody against EGFR (Nimotuzumab) and an mTOR inhibitor (rapamycin) in Caucasian patient-derived human glioma cell lines, Asian patient-derived human glioma cell lines, primary glioma cells derived from the Mayo GBM xenografts, and primary short-term glioma culture derived from high-grade glioma patients. Collectively, our results showed that Nimotuzumab, in combination with rapamycin, significantly enhanced the therapeutic efficacy in human glioma cells compared to single treatment. The combined treatment is effective regardless of the EGFR status in human glioma.

With the increasing ability to dissect cancer genome in clinical tumor samples, instead of having therapies that are based on non-specific one-size-fits-all strategies, moving towards developing new approaches that would allow patient-specific molecular targeted therapies are warranted.

The advances in understanding the molecular basis of cancer formation and progression have unraveled the challenges in treating tumor heterogeneity. It is becoming clear that combination or sequential treatment modalities that target multiple pathways can lead to better control of aberrant cell proliferation. Stupp and colleagues have reported that survival of patients who received the combination therapy of TMZ and radiation exceeded that of radiation alone [32]. Unfortunately, there is no significant benefit of using similar approach in our patient cohort [33]. This may be due to the small number of patient cohort, or perhaps, the genetic aberrations of Asian glioma patients differ from those of Caucasian glioma patients. The astrocytic gliomas derived from Caucasians have been reported to exhibit a different spectrum of genetic abnormalities when compared to non-Caucasian patients [34]. The Cancer Genome Atlas (TCGA) network described a robust gene expression based molecular classification of GBMs that divided them into proneural, neural, classical and mesenchymal subtypes [35]. However, the classical gene signature was not observed in a study conducted by Yan et al. on 225 Chinese glioma patients selected for whole genome gene expression profiling, highlighting differences between Asian gliomas and Caucasian gliomas [36]. Thus, there is a need to identify effective treatment option for Asian glioma patients, and to correlate the outcome to clinical parameters and biomarkers for advancing our understanding of the disease.

In this study, we demonstrate proof of concept that Nimotuzumab and rapamycin is effective as a combination therapy in glioma cell lines derived from Caucasian and Asian glioma patients. The enhanced efficacy of the combination therapy compared to mono-treatment is demonstrated in various human glioma cell lines (Figures 1F, 2B-D, and 3D and E), primary glioma cells derived the Mayo GBM xenografts (Figure 4B and C), and primary short-term glioma culture derived from high-grade Singapore GBM patients (Figure 4D and E). Of note, we are not certain why immortalized normal human astrocytes are not killed by concentrations of TMZ greater than the stated IC₅₀ (Figure 1A), it is possible that they have shifted to become more tumor-like cells. These are based on the facts that these cells overexpress viral oncoproteins required for immortalization, and can become tumorigenic in the presence of a Forkhead box transcription factor [37]. For future study, normal primary human astrocytes should be included as additional control. Regardless, it is encouraging that the combination treatment of Nimotuzumab with rapamycin is consistently more effective than the current standard of care therapy, TMZ. One major advantage of Nimotuzumab is that it does not induce skin [38], renal, and gastrointestinal mucosa-related toxicities [39]. Nimotuzumab requires bivalent binding for stable attachment on cell surface receptor while other monoclonal antibody such as Cetuximab can bind in a monovalent manner [40]. Thus, in the skin, where cell surface EGFR density is low, Cetuximab is expected to be more active than Nimotuzumab. It has also been suggested that Nimotuzumab only interferes with ligand-dependent EGFR activation; thus, the basal level of EGFR signalling which is required for the survival of normal epithelial cells is not affected. As a consequence, non-specific toxicity is reduced [6].

The binding affinity and kinetics of Nimotuzumab has been shown to be similar between wtEGFR and EGFRvIII [41]. In non-small cell lung cancer cell lines, the inhibitory effect of Nimotuzumab on EGFR signaling was found to be dependent on the cell surface expression of EGFR but not the status of EGFR mutation [42]. Herein, we showed that Nimotuzumab was effective in Asian patient-derived human glioma cell lines which expressed wild-type EGFR (Figure 2A-D), and Caucasian patient-derived human glioma cell lines expressing either wild-type EGFR (Figure 3B) or mutant EGFRvIII (Figure 3C), indicating that the effect of Nimotuzumab was indeed independent of the endogenous EGFR mutation status [22]. Interestingly, EGFR-null glioma cells (Gli36; Figure 1F) and parental U87MG (data not shown) are also responsive to Nimotuzumab treatment. In Gli36 cells, Nimotuzumab treatment resulted in activation of AKT, thus, the observed reduction in cell viability must be mediated through an AKT-independent pathway (Figure 1G). In U87MG.wtEGFR and U87MG.EGFRvIII cells, Nimotuzumab treatment inhibited activation of AKT, consistent to the reduced pAKT levels in wtEGFR-expressing A431 tumors treated with Nimotuzumab [21]. Although the results indicate that EGFR-null glioma cells are responsive to Nimotuzumab, the mechanism of action is unclear and we cannot exclude the possibility that EGFR-null cells are sensitive to the reagent used to prepare Nimotuzumab, in this case, it is a buffer solution containing polysorbate 80 of unrevealed concentration which has been associated with cytotoxic effect on the cells [43]; the slight reduction in cell viability observed may be attributed to this effect. Alternatively, the findings may represent a new mechanism of action. In normal cells, ligand-stimulated activation of EGFR is followed by subsequent internalization, ubiquitination and degradation in lysosomes. In GBM, mutant EGFRvIII is always present at the cell membrane due to its defective internalization properties [44]. As a result, there is a decrease in associating with Cbl proteins and degradation. The enhanced half-life and the constitutive phosphorylation of EGFRvIII are known to contribute to gliomagenesis. The finding that EGFR-null is also responsive to Nimotuzumab treatment could be due to possible receptor dimerization and crosstalk activities. Receptor dimerization has also been reported between urokinase-type plasminogen activator receptor and EGFRvIII that supports the survival and growth of GBM [45]. Given that receptor dimerization and crosstalk contribute to advantage in cell growth, it is therefore possible that when Nimotuzumab binds to the 3A epitope of the extracellular domain of EGFRvIII, it may be affecting oncogenic receptor dimerization events that lead to reduced cell viability.

The use of rapamycin alone in cancer therapy has shown modest success, perhaps due to the re-assembly of mTOR in the mTORC2/Rictor complex, leading to phosphorylation and reactivation of AKT. James and colleagues have shown that the treatment of A-431 cells with Nimotuzumab is effective in preventing feedback activation of pAKT by rapamycin in vivo [21]. Further, tumors derived from combination treatment were compared with mono-therapies using microarray analysis. Combination treatment resulted in the downregulation of genes beyond the typical pathways associated with Nimotuzumab and rapamycin. These pathways include metabolic, ECM-receptor interactions, tight junctions, biosynthesis of unsaturated fatty acids, ubiquitin mediated proteolysis pathways etc. Although this study differs from ours in many ways including experimental objectives, concentration of drugs and presence of EGF ligands and different cancer types, it is nevertheless encouraging that the combination treatment is effective given different cancer model. This is especially relevant in GBM because it highlights the plausibility of targeting TMZ resistant and EGFR-null glioma cells with alternative combination drugs such as Nimotuzumab and rapamycin. Furthermore, Nimotuzumab has recently been shown to enhance cancer radiosensitivity by inhibiting DNA-PKcs activation via the blockage of the PI3K/AKT pathway [46]. Although we have yet to determine whether the radiosensitizing effect of Nimotuzumab may be further enhanced with rapamycin, our results have nevertheless indicated that the combination of Nimotuzumab and rapamycin is more efficacious compared to TMZ and single treatment although it warrants further studies to delineate the underlying mechanism of action given different EGFR receptor status and possible crosstalk interaction.

The present study showed that the combination of Nimotuzumab and rapamycin could enhance glioma cell death, in an EGFR independent manner. Moreover, the results showed that combination treatment was effective in TMZ-resistant glioma cells, suggesting that Nimotuzumab and rapamycin may potentially be of clinical relevance for future treatment of human gliomas.