UC San Diego

Amplification and mutation of the epidermal growth factor receptor (EGFR) gene are common genetic hallmarks of glioblastoma (GBM). The most common mutation is an in-frame deletion of exons 2-7, resulting in a constitutively active variant of the receptor, EGFRvIII. Indeed, these molecules are proven drivers of gliomagenesis; yet, therapies that are directed against them, such as small molecule tyrosine kinase inhibitors (TKIs), have been ineffective due to both upfront and acquired drug resistance. In a genetic model of tetracycline-regulated EGFRvIII expression, it was confirmed that this receptor is essential for the maintenance of glioma growth in vivo. However, similar to a clinical situation of acquired drug resistance, some tumors eventually regained aggressive growth, and these breakthrough tumors persisted despite sustained suppression of EGFRvIII. In this study, we establish that tumor recurrence in the absence of EGFRvIII is in part, afforded by the ability to overcome oxidative DNA damage. From the approach of characterizing the phenotypes that are indicative of therapeutic sensitivity, we found that these phenotypes are typically overturned upon relapse. Specifically, we established for the first time that an increase in DNA damage burden reflects sensitivity to EGFRvIII inhibition in GBM, and that populations that overcome the need for EGFRvIII receptor function, recover from this DNA damage. Interestingly, blocking EGFRvIII alone was sufficient to induce significant DNA damage and we determined that an increase in endogenous reactive oxygen species (ROS) activity contributed to this phenotype. Likewise, we also observed that the resistant populations displayed significantly lower ROS levels.

These results suggested that an adaptive response mechanism might have been activated to neutralize ROS, therefore affording escape. A key genetic change that was common in multiple resistant models was up-regulation of KLHDC8A. This gene has already been shown to be required for the in vivo maintenance of some EGFRvIII-independent populations, but a functional role beyond this was unknown. Here, we show for the first time that loss of KLHDC8A restores sensitivity to EGFRvIII inhibition, by causing an increase in ROS activity above the toxic threshold. Consequently, this resulted in increased DNA oxidation and persistent DNA damage in the form of double strand breaks. Thus, we have established a novel link between KLHDC8A and ROS homeostasis. Furthermore, our work demonstrates that the characterization of sensitive phenotypes might uncover novel mechanisms that are essential for progressive escape from EGFRvIII blockade, and may also define novel therapeutic windows, whereby greater tumor response to standard of care may be achievable in GBM.