New life for an old therapy: ELTD1 as a downstream target of angiogenesis

Glioblastoma (GBM) is frequently diagnosed—and the most malignant—brain tumor in adults. The current standard of care for GBM is maximal surgical resection followed by radiation and temozolomide.¹ However, drug resistance is rapid and tumor recurrence is inevitable. Other therapeutic avenues are being actively studied as even with our advancements in tumor sequencing and genetic profiling, the median overall survival of patients diagnosed with GBM remains at ~14-16 months.² One avenue that has attracted attention for therapeutic potential is to improve the efficacy of targeting the vascular endothelial growth factor (VEGF) pathway, where currently a monoclonal antibody against VEGF—Bevacizumab—is FDA approved for GBM patients.³ As tumors grow more rapidly than their surrounding normal tissue, they need to form new blood vessels to obtain the nutrients they need from the body to grow.⁴ However, it is commonly observed that these quickly created blood vessels are leaky and improperly formed, which can prevent proper drug profusion.⁴ Nevertheless, while Bevacizumab is FDA approved, GBM in most of the patients progressed after their initial response to VEGF therapy,⁵ and VEGF is needed for normal angiogenesis and vessel homeostasis.⁴ Therefore, more research is needed to find tumor-specific, non-homeostatic, angiogenesis interventions.

In this issue of Neuro-Oncology, Huang et al⁶ delve into the details of determining if the epidermal growth factor, latrophilin, and seven transmembrane domain-containing protein 1 (ELTD1) could be used as a target, in conjunction with immunotherapy, in GBM mouse models. ELTD1 has previously been investigated as a target in GBM models because of its increased expression. Here, the authors expand upon their previous findings of ELTD1 being upregulated post-VEGF-A and TGFβ2 treatment⁷ and show by a tissue microarray analysis that ELTD1 protein is expressed predominately in brain vessels and increases in RNA expression as glioma tumor grade increases. Moving into a null ELTD1^−/− mouse model, they observed no difference in vasculature formation in post-natal day 6 retina samples, or ELTD1^−/− mouse liver, kidney, or lung, suggesting that ELTD1 is not necessary for proper normal vasculature development. Next, they tested the effect of an ELTD1^−/− microenvironment on orthotopic glioma tumor growth and survival in their animal models. Testing both an early timepoint of symptom onset and late timepoint of 23-day post-injection, no differences were noted in mouse overall survival or tumor sizes, demonstrating that ELTD1 in the tumor microenvironment (TME) does not affect murine GBM tumor progression. However, differences were observed in vasculature function between WT and ELTD1^−/− mice where a decrease in ELTD1 increased lectin perfusion and the number of Lef1 positive vessels, decreased the leakage of fibrinogen, and reduced tumor hypoxia. Taken together, these data show that loss of ELTD1 in the TME can normalize tumor vasculature and strengthen the blood-brain barrier function.

Finally, while others have depicted ELTD1 as a target in GBM,⁸ the authors went one step further to determine other pathways that may be affected with ELTD1 changes and if these could be used as targets in combination therapy with an anti-ELTD1 antibody. Upon performing RNA-Seq of ELTD1^−/− endothelial cells and downstream GSEA (gene set enrichment analysis), the authors found a decrease in Myc and E2F pathways, suggesting a more quiescent state, but also an increase in inflammatory and interferon alpha and gamma response pathways. To further investigate these immune response pathways, they performed immunohistochemistry (IHC) in mouse GL261 GBM tumors from both WT and ELTD1^−/− mice and found an increase in endothelin 1 (EDN1) and ICAM1 in the ELTD1^−/− background. GO (gene ontology) term analysis largely corroborated with GSEA findings where GO term enrichment revealed pathways associated with antigen processing and presentation. These intriguing results led the authors testing the efficacy of PD-1 checkpoint blockade in the ELTD1^−/− mice with GL261 orthotopic tumors. While an anti-PD-1 antibody alone was beneficial in the WT mouse background (35 days vs 30 days by IgG) an increase in median overall survival was observed in the ELTD1^−/− mice (47 days vs 33 days by IgG). Furthermore, CD3⁺ total T cells and CD8⁺ cytotoxic T cells were significantly increased in the ELTD1^−/− background, though total CD45⁺ cells were similar in all treatment groups. Overall, Huang et al have built upon the previous work which showed ELTD1 as a promising target in GBM and have added a potential synergistic pathway of increased immune response which can be targeted with PD-1 checkpoint blockade.

Previous studies focused on targeting ELTD1 in the glioma cells, where a monoclonal antibody was shown to improve animal survival in vivo.⁸ However, in this study, ELTD1 was not observed to be expressed in the glioma cells, only the tumor vasculature, suggesting a possible antibody-dependent cellular cytotoxicity in the antibody-mediated glioma cell response.⁸ For this reason, this study modulates the microenvironment and mouse background, rather than the tumor cells, to better determine the function of ELTD1 in vasculature formation and maintenance.⁶ However, several outstanding questions still remain. What is the exact role of ELTD1 in vasculature integrity? Will the anti-ELTD1 antibody plus anti-PD-1 antibody therapy show an improved overall survival? Will ELTD1 expression in the tumor vasculature alone be a sufficient predictive marker for accessing this dual antibody treatment?

Although targeting VEGF for cancer treatment was proposed 25 years ago (~1995-1998)⁴ and a substantial amount of work has been completed to understand the role of VEGF in tumors, VEGF targeting has been largely disappointing in improving patient survival across the board.⁹ Previous angiogenesis modulators have focused on targeting soluble VEGF via antibodies or downstream VEGF signaling by inhibiting VEGF receptors with either targeted antibodies or receptor tyrosine kinase inhibitors.³ However, even though angiogenesis is a hallmark of cancer, VEGF inhibition has not had the expected therapeutic impact, clinically.¹⁰ With the more recent findings of VEGF being immunosuppressive, combination trials of VEGF modulation and immune checkpoint blockade have shown more promising results.⁹ Therefore, combining more tumor-specific angiogenesis modulators—like ELTD1—with immune checkpoint blockade may decrease toxicity, reinvigorate the angiogenesis field, and increase patients’ overall survival.

Conflict of interest statement. The text is the sole product of the authors and that no third party had input or gave support to its writing. The authors have no financial or competing interests to declare.

Funding

This work was supported by a United States National Institutes of Health grant NS115403 and Lou and Jean Malnati Brain Tumor Institute at Northwestern Medicine (S.Y.C.).