Over the last several decades, an improved understanding of von Hippel–Lindau disease and its underlying biology has informed the successful development of numerous anti-cancer agents, particularly for the treatment of advanced renal cell carcinoma.
Over a century ago, the identification of a familial pattern of retinal hemangioblastomas by ophthalmologist Eugen von Hippel, and the subsequent association of this disorder with an increased risk of both benign and malignant systemic neoplasms by neuropathologist Arvid Lindau, led to the discovery of von Hippel–Lindau (VHL) disease 
. Officially coined in the 1930s, VHL disease represents a hereditary, autosomal-dominant condition that may manifest with a variety of tumor types including central nervous system (CNS) and retinal hemangioblastomas, pancreatic neuroendocrine tumors (pNETs), pheochromocytomas, inner ear endolymphatic tumors, epididymal or broad ligament cystadenomas, clear cell renal cell carcinomas (RCC), and/or pancreatic or renal cystic lesions 
. Occurring in approximately one in 35,000 live births, VHL disease is caused by a germline mutation and/or deletion in the VHL
tumor suppressor gene, located on the short arm of chromosome 3 
. VHL disease can be diagnosed based on clinical criteria incorporating the presence of syndromic tumors, with or without a familial VHL disease history 
, and phenotypic subtypes have been assigned depending on the absence or presence of pheochromocytoma (Type 1 or Type 2 disease, respectively) 
Historically, the management of VHL disease-associated neoplasms has entailed careful imaging and clinical surveillance, with the deployment of serial local interventions including surgical, ablative, or ophthalmologic procedures in the event of significant or symptomatic growth. In particular, RCCs and CNS hemangioblastomas often require serial high-risk, invasive interventions, thereby contributing to significant physical and psychologic morbidity for patients. Patients with VHL disease have a lifetime risk of RCC malignancies of up to 70% and face the potential for metastatic dissemination of RCC if tumors grow beyond 3 cm in diameter 
. Although surgical resection can be curative therapy for VHL-associated RCC, the risk of local recurrences and/or incident tumors in the remaining or contralateral kidney is inherently elevated, thereby limiting the potential durability of local interventions 
. Thus, the management of VHL disease-associated RCC has entailed the use of active surveillance for RCC tumors <3 cm in size, with selective nephron-sparing interventions such as partial nephrectomy or percutaneous ablative procedures, reserved for the treatment of tumors beyond the 3 cm threshold or for rapidly growing lesions 
. Nevertheless, the recurrent nature of RCC in VHL disease can lead to significant cumulative patient morbidity including potential metastatic dissemination or the need for hemodialysis or renal transplantation 
. Similarly, both CNS (brain and spinal cord) and retinal hemangioblastomas demonstrate recurrence and progression, resulting in serial interventions and significant cumulative morbidity 
Given the requirement for serial invasive procedures to manage disease-associated neoplasms throughout a VHL patient’s lifetime, effective systemic therapy for VHL disease could impart significant clinical benefit by reducing the need for invasive interventions and the resulting iatrogenic morbidity. In addition, the unifying genetic predisposition and biology of VHL disease-associated neoplasms lends well to potential therapeutic targeting 
. Indeed, an improved understanding of the VHL
gene and its downstream gene products has informed the successful development of numerous anti-cancer agents for sporadic clear cell RCC and ultimately culminated in regulatory approval by the United States Food and Drug Administration (US FDA) of the first systemic agent for patients with VHL disease-associated RCC, pNETs, or CNS hemangioblastomas 
2. VHL Biology and Inappropriate Angiogenesis
gene was identified in 1993, and the structure and function of the VHL
gene product (pVHL) were elucidated over the subsequent decade 
. pVHL forms a ternary complex with the transcription elongation factors elongin C and B, forming the VCB complex 
. Through cross-stabilization between pVHL and elongins B and C, the VCB complex is resistant to proteasomal degradation 
. pVHL plays a key role in the cellular signaling resulting from changes in oxygen tension by functioning as the substrate recognition subunit of an E3 ubiquitin ligase complex that targets the α subunit of the heterodimeric hypoxia-inducible factor (HIF) transcription factor for proteasome degradation 
. Under normoxic conditions, pVHL recognizes the oxygen-dependent prolyl-hydroxylation of HIFα and targets it for ubiquitylation 
. Polyubiquitylated HIFs are subsequently sequestered and degraded by the cellular proteosome. However, under hypoxic conditions, or in the absence of functional pVHL due to a VHL-disease associated mutation, HIFα constitutively accumulates and forms heterodimers with HIF1β 
. These heterodimers subsequently translocate to the nucleus and bind to hypoxia-response elements (HREs), inducing downstream gene expression and promoting cellular adaptation to acute or chronic hypoxia 
The HIF transcriptional complex promotes the expression of over 100 proteins, and key factors regulated by the pVHL–HIF pathway include HRE-related pro-angiogenic proteins as well as proteins involved in cellular growth and metabolism. In particular, HIF-mediated transcription induces the gene expression of the vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), cyclin D1, and glucose transporter 1 (GLUT1) 
. Due to the downstream inappropriate pro-angiogenic upregulation occurring in the setting of dysfunctional pVHL, the highly vascular neoplasms present in VHL disease are known to overproduce such angiogenic peptides 
. Moreover, consistent with the tumorigenesis implicated in hereditary VHL disease, somatic gene aberrations are identified in the vast majority (up to 90%) of sporadic clear cell RCCs 
. Indeed, this understanding of VHL biology and the inappropriate pro-angiogenic signaling resulting from VHL inactivation in clear cell RCC has revolutionized the systemic therapy of advanced RCC over the past two decades with the clinical development and eventual regulatory approvals of numerous VEGF receptor tyrosine kinase inhibitors (VEGFR TKIs) including sunitinib, pazopanib, and axitinib.
3. Anti-Angiogenic Systemic Therapies in VHL Disease
Given the shared pVHL–HIF pathway tumor biology implicated in both VHL-associated disease and sporadic RCC, its functional consequence of inappropriate angiogenesis, and the proven anti-tumor efficacy and regulatory approvals of VEGFR TKI therapies in sporadic advanced clear cell RCC, these antiangiogenic agents were among the first tested for the systemic treatment of patients with VHL disease-associated RCC and other neoplasms (Table 1
). In a prospective, open-label, single-arm phase 2 clinical trial, 15 patients with genetically-confirmed VHL disease and at least one measurable VHL disease-associated lesion (CNS hemangioblastoma, RCC, pancreatic cysts or NETs) received sunitinib 50 mg daily on a 28-day on/14-day off schedule for up to four treatment cycles (total duration of 6 months) 
. The primary objective was to evaluate the safety of sunitinib, and the secondary objectives included efficacy evaluation assessed by radiographic response. Patients may also have had retinal hemangioblastomas, which were followed by regular direct ophthalmoscopy. While all patients received at least two treatment cycles (50% of planned therapy), six patients (40%) discontinued therapy prior to the completion of four treatment cycles. Grade 3 toxicity occurred in five patients (33%) and included fatigue (33%), hand-foot syndrome (13%), and nausea (13%). Sunitinib dose reduction was required in 10 patients (66%) (37.5 mg daily in six patients; 25 mg daily in four patients). The observed anti-tumor response appeared to be lesion-dependent, with six out of 18 (33%) evaluable RCCs demonstrating a partial response (PR), and no responses observed in 21 evaluable CNS hemangioblastomas. A best response of stable disease (SD) occurred in five evaluable pNETs, and none of the seven retinal lesions demonstrated response per ophthalmoscopy 
. Similarly, in a single arm prospective study of five patients with genetically-confirmed VHL disease treated with the same dose and schedule of sunitinib, the best observed response was SD and treatment was limited by unacceptable toxicity (occurring in three out of five patients within 6 months of treatment initiation) 
. Overall, these findings indicate that while sunitinib anti-angiogenic therapy offered some anti-tumor response in VHL disease-associated RCC, the minimal efficacy against extra-renal disease manifestations and the significant treatment-related toxicity limited its effective long-term use in this setting.
Table 1. Prospective studies of systemic therapy for VHL disease-associated neoplasms.
The analysis of sporadic RCC and VHL-associated hemangioblastoma archival tissues was subsequently conducted in an effort to evaluate the differential response observed between RCC and CNS hemangioblastoma lesions exposed to sunitinib in VHL disease 
. Notably, while RCC tissues demonstrated higher levels of phosphorylated VEGFR-2 when compared to hemangioblastomas, the protein levels of fibroblast growth factor receptor 3 (FGFR3) and FGFR ligand (FGFR substrate 2) were higher in hemangioblastomas relative to RCC 
. This intriguing finding provided a mechanistic rationale for the clinical testing of dovitinib, a multi-kinase inhibitor targeting FGFR, VEGFR, and PDGFR, in patients with genetically-confirmed VHL and at least one measurable hemangioblastoma 
. Unfortunately, the study was terminated following the treatment of only six patients due to intolerable toxicity. Three of six patients (50%) discontinued therapy due to adverse effects including maculopapular rash, vomiting, and dyspnea, despite dose reduction and supportive measures. The best response observed was SD in hemangioblastomas, which occurred in five out of six patients 
The largest prospective study to date evaluating a VEGFR-directed approach for patients with VHL disease was a non-randomized phase 2 single-center open label clinical trial of pazopanib 
. Thirty-one eligible patients with genetically-confirmed or clinically-defined VHL disease received pazopanib of 800 mg daily for a planned treatment duration of at least 24 weeks. Co-primary endpoints of the study were safety and objective response rate. Common adverse events included fatigue, hypertension, and diarrhea, and four patients (13%) required treatment discontinuation due to grade ≥3 transaminitis. An additional three patients (10%) discontinued due to the intolerance of lower-grade cumulative toxicities. Ultimately, 21 patients (67%) required pazopanib dose reduction due to toxicity. However, pazopanib therapy demonstrated clear therapeutic activity in VHL disease. Across the 59 total RCC tumors, a PR was observed in 29 (49%) and a complete response (CR) occurred in two tumors (3%). While pazopanib achieved a PR in nine out of 17 (53%) pancreatic tumors, only two out of 49 (4%) CNS hemangioblastomas demonstrated PR. Only seven patients (23%) elected to remain on pazopanib therapy beyond 24 weeks. Therefore, while pazopanib demonstrated clinical activity in VHL disease, particularly in RCC tumors, the clinical benefit for patients was limited by treatment-related toxicity and suboptimal efficacy toward VHL-disease associated hemangioblastoma.
4. HIF2α Inhibition in VHL Disease
Since an understanding of VHL disease biology and pathogenesis led to the preclinical development of drugs targeting HIF2α transcription activity, VHL disease represents a model disease system for the clinical testing of belzutifan. The clinical activity of belzutifan in VHL disease was specifically tested in a phase 2, non-randomized, open-label clinical trial of 61 patients with germline VHL alteration and at least one measurable, non-metastatic primary RCC 
. Eligible patients did not have a RCC lesion larger than 3 cm that necessitated immediate surgical resection to mitigate metastatic risk. Belzutifan was administered at a starting dose of 120 mg daily, and the primary study endpoint was the objective response rate of RCC lesions. Secondary endpoints included safety and the objective response rate in non-RCC VHL disease-associated neoplasms (including CNS hemangioblastomas and pancreatic lesions). Consistent with the serial invasive interventions typically required for the management of VHL disease, 40 patients (66%) had undergone prior partial or radical nephrectomy, and 47 patients (77%) had undergone prior neurosurgical intervention. After a median follow-up period of 21.8 months (range 20.2–30.1), a total of 56 patients (92%) had an observed reduction in the sum of RCC target lesions including 30 patients (49%) with a confirmed PR. An additional 30 patients (49%) had SD. At the time of the initial data cut-off, the median duration of response had not been not reached (2.8+–22.3+ months), and 54 patients (89%) remained on belzutifan therapy. Similar anti-tumor efficacy was observed in non-RCC VHL disease-associated neoplasms, with 47 of 61 patients (77%) having a confirmed response in pancreatic lesions (including a 91% confirmed response rate in 20 patients with pancreatic NETs). Among the 50 patients with CNS hemangioblastomas, the objective response rate was 30% including three patients (6%) with complete radiographic responses. Treatment-related toxicity was mostly low grade (grade 1 or 2), and belzutifan dose-reduction was required in only nine patients (15%). Treatment-related grade 3 events occurred in nine patients (15%) and included anemia (8%), fatigue (5%), and hypertension (8%). Although four patients (7%) received blood transfusion and 12 patients (20%) received erythropoietin-stimulating agents, treatment-related anemia typically stabilized without intervention in most patients. No treatment-related grade 4 or 5 events occurred.
Overall, these clinical findings compared favorably to the observed experience with VEGFR TKI therapies for the treatment of VHL disease. For example, while pazopanib TKI therapy demonstrated comparable objective response rates (42%) in VHL disease-associated neoplasms (RCC, pancreatic, and CNS neoplasms), the treatment exposure was limited by more significant toxicity (23% toxicity-related discontinuation rate and 68% dose-reduction rate) 
. The patient acceptability and ongoing clinical efficacy of belzutifan was reinforced with longer-term follow-up data of the phase 2 trial in VHL disease 
. After a minimum follow-up period of 24 months (range 27.6–37.5), 50 out of 61 patients (82%) remained on belzutifan therapy. Among patients discontinuing study therapy (N = 11), treatment discontinuation was due to the progression of RCC neoplasms (N = 4), patient decision (N = 4), adverse events (N = 2), or unrelated patient death (N = 1). The RCC objective response rate improved to 59% including 3% CR (Table 1
). The median duration of response was still not reached, with some patients having an objective response lasting greater than two years. In non-RCC VHL disease-associated lesions, the objective response in pancreatic NETs (N = 20) and CNS hemangioblastomas (N = 50) remained 90% and increased to 38%, respectively. Similarly, among the 12 patients (16 total eyes) with baseline retinal hemangioblastomas, all demonstrated ophthalmologic improvement. No new safety signals emerged with this longer-term follow-up. Ultimately, these overall findings led to the regulatory approval of belzutifan in August 2021 as the first systemic treatment for adult patients with VHL disease who require therapy for RCC, CNS hemangioblastomas, or pancreatic NETs, not requiring immediate surgery