PD-L1 and HIF-2α Upregulation in Head and Neck Paragangliomas after Embolization

Abstract

Hypoxia activates pathways associated with tumor progression, metastatic spread, and alterations in the immune microenvironment leading to an immunosuppressive phenotype. In particular, the upregulation of PD-L1, a target for therapy with checkpoint inhibitors, is well-studied in several tumors. However, the relationship between hypoxia and PD-L1 regulation in pheochromocytomas and paragangliomas (PPGL), and especially in paragangliomas treated with embolization, is still largely unexplored. We investigated the expression of the hypoxia-marker HIF-2α and of PD-L1 in a PPGL-cohort with and without embolization as potential biomarkers that may predict the response to treatment with HIF-2α and checkpoint inhibitors. A total of 29 tumor samples from 25 patients who were operated at a single center were included and analyzed utilizing immunohistochemistry (IHC) for PD-L1 and HIF-2α. Embolization prior to surgery was performed in seven (24%) tumors. PD-L1 expression in tumor cells of head and neck paragangliomas (HNPGLs) receiving prior embolization (median PD-L1 positivity: 15%) was significantly higher as compared to PD-L1 expression in HNPGLs without prior embolization (median PD-L1 positivity: 0%) (p = 0.008). Consistently, significantly more HNPGLs with prior embolization were positive for HIF-2α (median nuclear HIF-2α positivity: 40%) as compared to HNPGLs without prior embolization (median nuclear HIF-2α positivity: 0%) (p = 0.016). Our results support the hypothesis that embolization with subsequent hypoxia leads to the upregulation of both PD-L1 and HIF-2α in HNPGLs, and could thus facilitate targeted treatment with HIF-2α and checkpoint inhibitors in the case of inoperable, locally advanced, or metastatic disease.

Immunohistochemical Profiling of SSTR2 and HIF-2α with the Tumor Microenvironment in Pheochromocytoma and Paraganglioma

“…Hypoxia-related pathways, which enhance PD-L1 expression in cancer cells, contribute to immune tolerance in the TME [ 24 , 45 ]. Previous studies on the relationship between PD-L1 and HIF-α expression in PPGLs were insufficient as only HIF-1α was assessed [ 29 ], not distinguishing between nuclear and cytoplasmic HIF-2α staining [ 30 ] or only in head and neck paragangliomas with prior embolization [ 31 ]. Although further research, particularly transcriptome analysis, is necessary to validate the IHC findings of HIF-2α, the elevated levels of both PD-L1 and HIF-2α observed in our diverse cohorts indicate a potential approach for combination therapy using HIF-2α inhibitors and ICIs.…”

“…The immunoreactivity of SSTR2A was evaluated based on Volante scores [ 21 , 33 ], considering the subcellular localization and extent of the staining in four levels: 0 (no staining), 1 (pure cytoplasmic immunoreactivity, either focal or diffuse), 2 (membranous reactivity in less than 50% of tumor cells, irrespective of the presence of cytoplasmic staining), 3 (circumferential membranous reactivity in more than 50% of tumor cells, irrespective of the presence of cytoplasmic staining). HIF-2α was evaluated in four levels of staining: 0 (no staining), 1 (weak), 2 (medium), and 3 (strong) [ 34 ], each for the tumor cell nuclear (HIF-2α NUC ) and cytoplasm (HIF-2α CYT ) based on previous reports in PPGLs [ 31 , 35 ]. For statistical analysis, SSTR2A and HIF-2α scores of 0 and 1 were considered negative, while scores of 2 and 3 were considered positive.…”

“…Clinical trials of Belzutifan (MK-6482), a selective HIF-2α inhibitor approved for von Hippel–Lindau (VHL) disease-associated tumors (RCC, pancreatic neuroendocrine tumor (NET), and central nervous system hemangioblastoma) [ 28 ], have expanded to PPGLs (NCT04924075). Although some PPGLs have pathogenic mutations in genes associated with hypoxia signaling, there are limited reports on the association between the immune cell components and HIF expression [ 29 , 30 , 31 ].…”