Hemangioblastoma of the Kidney—A Comprehensive Clinical, Pathological, and Genetic Analysis of Four Cases

ABSTRACT

Hemangioblastoma (HB) is a benign central nervous system (CNS) tumor associated with mutations in the von Hippel–Lindau (VHL) gene. Although rare outside the CNS, the pathological and genetic features remain poorly understood. We analyzed four renal hemangioblastomas (RHB). Demographics, clinical presentation, and follow‐up data were collected. After assessing hematoxylin and eosin‐stained slides, immunophenotyping was conducted using CA9, α‐inhibin, AE1/AE3, CD10, CD56, PAX8, S100, MelanA, HMB45, CD117, FH, SDHB, and brachyury antibodies, alongside mismatch repair (MMR) deficiency examination. Additionally, whole‐exome sequencing (WES) was performed in 3 tumors. Our cohort comprised 3 male and 1 female patients, with a median age of 49 years. No data on VHL disease were available. Well‐circumscribed tumors (median size: 25.5 mm) displayed clear vacuolated cytoplasm with a vascular component. Immunostaining revealed expression of PAX8, α‐inhibin, AE1/AE3, S100, and cytoplasmic brachyury. WES analysis detected no pathogenic mutations. No cancer‐related deaths or progressions were observed. Histologically, RHB resembles low‐grade ccRCC and shares expression of PAX8, pancytokeratin, and CA9. However, RHB is uniquely positive for α‐inhibin, S100, and lacks VHL alterations. Its favorable prognosis underscores the importance of distinguishing it from ccRCC to prevent unnecessary treatments. Further research is warranted to elucidate the underlying genetic mechanisms.

1. Introduction

Hemangioblastoma (HB) is a rare benign mesenchymal neoplasm that predominantly arises within the central nervous system (CNS) [1]. Even in sporadic cases, a strong association exists with mutations in the von Hippel–Lindau (VHL) gene, with nearly half of VHL disease patients developing CNS hemangioblastoma (CNS‐HB) [1, 2]. While HBs can occasionally occur outside the CNS, they are most commonly found in spinal nerve roots and peripheral nerves, with rare occurrences in soft tissues, bones, and visceral organs [3]. Among visceral HBs, the kidney is the most frequently affected site. Since 2007, only about 30 cases of renal hemangioblastoma (RHB) have been documented in the English literature, highlighting its extreme rarity [4].

Macroscopically, RHBs exhibit diverse morphologic features. Liu et al. described them as encapsulated, homogenous brownish‐white masses, while Nonaka reported reddish lesions, sometimes cystic, with fibrotic areas [5, 6]. Microscopically, RHBs present a complex architecture comprising hypercellular and hypocellular, edematous regions. The hypercellular areas contain stromal cells with eosinophilic or clear cytoplasm, arranged in sheet‐like or solid patterns within a dense vascular network. Nuclear morphology varies, with round or oval nuclei of differing sizes, occasionally featuring intranuclear inclusions [4, 6, 7]. Additionally, hallmark features include cytoplasmic vacuolization resembling lipoblasts and hyaline/eosinophilic globules, as noted by Ip et al. [8]. A unique case reported by Yin et al. [9] further expanded the histological spectrum by displaying rhabdoid‐cell‐like features.

Immunohistochemical analysis has consistently demonstrated diffuse expression of S100, vimentin, and neuron‐specific enolase (NSE) across all documented cases. Notably, α‐inhibin positivity serves as a crucial diagnostic marker for both RHB and CNS‐HB being positive in the majority of the published cases [3]. However, Wu et al. [10] reported two cases of α‐inhibin‐negative RHB, which also lacked PAX8 and cytokeratin (CK7 and AE1/AE3) expression while maintaining positivity for S100 and vascular markers CD31 and CD34. These cases illustrate the diagnostic complexities associated with atypical marker profiles [10]. Additionally, focal cytokeratin (CK7 and AE1/AE3) positivity has been observed in three cases [11, 12, 13], while eight cases exhibited either focal or diffuse CD10 positivity [9, 10, 13, 14, 15, 16, 17]. With the exception of six cases, PAX8 positivity has been consistently reported, emphasizing the importance of organ‐specific marker expression in extraneuraxial HBs [11].

A key diagnostic challenge arises from the microscopic similarities between RHB and clear cell renal cell carcinoma (ccRCC), often referred to as the most critical differential diagnosis. The difficulty has been further amplified since Montironi et al. [18] described a potential new morphological subtype of ccRCC in 2014, where classic ccRCC and HB‐like areas coexist, complicating the distinction between these entities. The significance of focal cytokeratin and CD10 positivity in reported cases remains controversial, while inconsistent PAX8 expression further complicates diagnosis.

This article explores the evolving differential diagnostic challenges of RHB, emphasizing the persistent uncertainties surrounding its classification. These challenges prompted us to conduct a thorough reevaluation of our own RHB cases to gain deeper insights into this rare neoplasm.

2. Methods

2.1. Statement of Ethics

This retrospective study was conducted with the permission of the Regional and Institutional Human Medical Biological Research Ethics Committee, University of Szeged (No. 188/2019‐SZTE), and the Scientific and Research Ethical Committee of Hungarian Scientific Council (ETT TUKEB, 49585/2‐2019/EKU). Here, the data reviewed were collected from patients as part of the routine standard of care; no diagnostic or therapeutic interventions were performed, and no patient contact was involved. Therefore, patient consent was not required in accordance with local or national guidelines.

2.2. Study Design and Review

In this study, we present four cases of RHB, sourced from participating institutes. All cases comprised tumor resections and nephrectomy specimens exclusively, with biopsy samples excluded from analysis. None of these cases had been previously published or subjected to detailed analysis. Each case underwent review by two pathologists (B.P. and L.K.), who meticulously evaluated morphology, immunohistochemical features, and molecular genetic data. Clinical characteristics such as age, sex, symptoms, and any underlying renal disorders, as well as follow‐up data, were retrieved from the electronic medical records of the patients. Details regarding tumor size, laterality, and surgical technique were extracted from the original histopathological reports.

2.3. Immunohistochemistry

All immunohistochemical stains were performed at the same laboratory (Department of Pathology, Albert Szent‐Györgyi Medical School, University of Szeged), applying Leica Bond‐Max Automated IHC Staining System (Leica Biosystems). We used the following primary antibodies: CA9 (polyclonal, 1:2000, Novus Biologicals), CD117 (EP10, monoclonal, 1:100, BioSB), CK AE1/AE3 (monoclonal, 1:600, Cell Marque), CD10 (56C6, monoclonal, 1:50, Cell Marque), α‐inhibin (R1, monoclonal, 1:100, BioSB), MelanA (A103, monoclonal, 1:200, Labvision), HMB45 (hmb‐45, monoclonal, 1:200, Cell Marque), SDHB (BSB‐131, monoclonal, 1:200, BioSB), FH (J‐13, monoclonal, 1:2000, Santa Cruz), CD56 (123C3.D5, monoclonal, 1:200, Cell Marque), PAX8 (MRQ50, monoclonal, 1:400, Cell Marque), S100 (4C4.9, monoclonal, 1:800, Cell Marque), and brachyury (SC374‐321, monoclonal, 1:100, Santa Cruz). We studied the presence of the mismatch repair proteins (pMMR) with MLH1 (ES05, monoclonal, 1:100, Novocastra), MSH2 (79H11, monoclonal, 1:200, Novocastra), MSH6 (PU29, monoclonal, 1:100, Novocastra), and PMS2 (EP51, monoclonal, 1:100, BioSB) IHC. Primary antibodies were visualized using the Bond Polymer Refine Detection kit (Leica Biosystems). In parallel, we stained appropriate positive and negative controls. The reactions were appreciated in a semiquantitative fashion (< 1% positivity of tumor cells: − (negative); 1%–25% positivity of tumor cells: +; > 25%–50% positivity of tumor cells: ++; > 50%–75% positivity of tumor cells: +++; and > 75% positivity of tumor cells: ++++) except for FH, SDHB, and pMMRs because these were evaluated as retained or lost.

2.4. Whole Exome Sequencing and Data Analysis

Three RHB cases underwent WES analysis according to the methodology detailed in a previous publication of our working group [19]. Ten serial sections of 10‐μm thickness per formalin‐fixed, paraffin‐embedded sample were taken, and deoxyribonucleic acid (DNA) was extracted. DNA concentration was measured by Quant‐iT 1× dsDNA HS Assay kit (Thermo Fisher Scientific) with Fluostar Omega (BMG Labtech) plate reader. For whole exome sequencing (WES) library construction, Twist Library Preparation EF Kit 2.0 with Universal Adaptor System and Exome 2.0 Panel (Twist Bioscience) was applied. The fragment size distribution of the precapture and postcapture libraries was determined by capillary electrophoresis on LabChip GX Touch HT Nucleic Acid Analyzer by using X‐Mark HT Chip and DNA NGS 3K Assay kit (PerkinElmer). The libraries were quantified by Quant‐iT 1× dsDNA HS Assay kit (Thermo Fisher Scientific) with Fluostar Omega (BMG Labtech). In average, more than 24 Gbp raw data was generated per sample; demultiplexing, adapter trimming, Q30‐filtering, and somatic variant calling of the sequenced data was performed on Dragen Bio‐IT platform (Illumina). Genomic variants of vcf files were annotated by using the Varseq Golden Helix software package. Tumor mutation burden (TMB) was calculated by the number of non‐synonymous somatic mutations (single nucleotide variants and small insertions/deletions) per megabase in coding regions [20].

3. Results

3.1. Clinicopathological Findings and Follow‐Up Data

We reviewed all ccRCC cases in our archive (n = 2139) and identified 4 RHBs. The clinicopathological features are summarized in Table 1. Our cohort comprised three male and one female patients, with a median age of 49 years (ranging from 35 to 65 years). All tumors were sporadic, and none of the patients exhibited any features of VHL syndrome. None of the patients presented with tumor‐related symptoms or underlying renal disorders. Radical nephrectomy was performed in two cases, while the remaining two patients underwent tumor resection. All four examined cases were unilateral, right‐sided, and unifocal. Follow‐up information was available for every case, with all patients being alive and without showing any signs of disease progression. The follow‐up duration ranged from 2 to 11 years, with a 6.53 median follow‐up time of years and a 6.68 mean of years.

TABLE 1.

The clinicopathological features of the cases investigated.

Case ID	Age (years)	Sex	Surgery	Side	Size (mm)	Original diagnosis	Follow‐up (mo)	Status
1	36	M	RN	R	27	ccRCC	30	Alive
2	62	M	RN	R	42	ccRCC	89	Alive
3	65	F	TR	R	24	ccRCC	68	Alive
4	35	M	TR	R	24	ccRCC	134	Alive

Abbreviations: ccRCC, clear cell renal cell carcinoma; mo, months; R, right; RN, radical nephrectomy; TR, tumor resection.

3.2. Macroscopic and Microscopic Characteristics

Macroscopically, the tumors exhibited well‐circumscribed characteristics, with some instances presenting a thick fibrotic capsule. The median tumor diameter measured 25.5 mm (mean size: 29.25 mm), ranging from 24 mm to 42 mm in its largest dimension. Furthermore, the cut surface displayed a heterogeneous appearance, comprising gray‐brown areas interspersed with white‐grayish fibrotic regions (Figure 1A). The surrounding renal parenchyma maintained its typical appearance.

FIGURE 1.

(A) Grossly, a grayish‐brown tumor is present on the cut surface of the kidney resection specimen. (B) Renal hemangioblastoma is composed of stromal cells and an extensive capillary network (×400). (C) The stromal cells have clear and vacuolated cytoplasm. In some instances, the stroma contains fibrotic tissue and smooth muscle bundles (×400). (D) In Case 1, noncaseating granulomas were dispersed among the tumor cells (×400). (E) All tumors displayed some degree of regression, with stromal edema, fibrosis, calcification, and hemosiderin deposition (×200). (F) The tumors were well‐defined and demarcated from the surrounding renal parenchyma and fat tissue, sometimes with a pseudocapsule (×200).

Microscopically, the tumors manifested as sheet‐like or solid growths of stromal cells interwoven with a capillary meshwork (Figure 1B). The stromal cells exhibited relatively large, round, or polygonal shapes, characterized by eosinophilic or clear cytoplasm (Figure 1C). Intracytoplasmic vacuolization was observed within the stromal cells. Nuclear pleomorphism was evident in three cases, yet mitotic figures were absent. Notably, Case 1 presented intratumoral noncaseating granulomas (Figure 1D). Additionally, regressive areas featuring stromal edema, fibrosis, and calcification were observed across all cases (Figure 1E). It is noteworthy that prominent nucleoli, microscopic tumor cell necrosis, and sarcomatoid changes were absent in our cases. Furthermore, no invasive growth or infiltration into extrarenal sites such as the sinus, fatty capsule, or renal pelvis was observed (Figure 1F). Cases 1 and 4 exhibited morphological features consistent with HBs, prompting the reporting pathologist to diagnose them accordingly. Conversely, Cases 2 and 3 initially received diagnoses of ccRCC and RCC with fibromyomatous stroma, respectively.

3.3. Immunomorphological and Genetic Characteristics

Immunohistochemical findings, summarized in Table 2, revealed consistent patterns. All RHBs displayed strong positive reactions for α‐inhibin, S100, and PAX8 (Figure 2A–C). Half of the cases exhibited diffuse cytoplasmic positivity for brachyury (Figure 2D), while the remainder showed only focal positivity for this immunoreaction. Furthermore, three out of four RHB cases demonstrated strong positivity for CA9 (Figure 2E). Cases 3 and 4 exhibited focal immunoreactivity for CD56. Two cases were strongly positive for AE1/AE3 (Figure 2F), with one of these cases also displaying focal positivity for CK7. One case showed diffuse, strong positivity for CD10. Notably, all cases tested negative for CD117, MelanA, and HMB45. Additionally, FH and SDHB expression were retained in all tumors.

TABLE 2.

Immunohistochemical features of the tumor investigated.

Case ID	Inhibin	S100	CA9	PAX8	AE1/AE3	Brachyury (cytop)	CD10	CK7	CD56
1	++++	++++	++++	++++	++	++++	Neg	Neg	Neg
2	++++	++++	Neg	++++	++	++++	Neg	Neg	Neg
3	++++	++++	++++	++++	++++	++	++++	+	++
4	++	++++	++++	++++	++++	++	+	Neg	+

Abbreviations: CA9, carbonic anhydrase 9; CD, cluster of differentiation; CK, cytokeratin; cytop, cytoplasmic; Neg, negative; PAX8, paired box protein 8.

FIGURE 2.

(A) α‐inhibin diffusely labeled the stromal cells (×400). (B) The tumors showed diffuse S100 expression (×400). (C) PAX8 was strongly expressed in the nuclei of the tumor cells (×400). (D) In two tumors, diffuse cytoplasmic brachyury positivity was observed (×200). (E) Carbonic anhydrase 9 labeled the membranes of the stromal cells in all but one case (×400). (F) Cytokeratin expression was observed in each tumor (×400).

No somatic pathogenic variant in VHL, TSC1/2, and MTOR, and in other genes was identified. Additionally, no germline variants were detected in our cohort. Hence, all cases were considered low TMB, having less than 10 mutations per megabase.

4. Discussion

At present, 31 RHB cases have been published in the literature, to which we add the four cases reported in this study. In the following, we will discuss our morphological, immunohistochemical, and molecular findings, as well as clinical and differential diagnostic factors, comparing them with experiences reported in the literature.

4.1. Immunohistochemical Features

Our immunohistochemical findings showed consistent expression of α‐inhibin, S100, and PAX8 in all cases. Cases 1, 3, and 4 showed strong membrane staining with CA9. Case 3 presented diffuse, strong positivity and Case 4 showed weak, focal positivity for CD10. Furthermore, focal or diffuse cytoplasmic expression of brachyury was detected in all cases.

Previous studies showed that RHB can express both PAX2 and PAX8 markers, which support that HBs generally can vary their immunoprofiles due to the primary development site, and it is well known that PAX8 and PAX2 are specific transcription factors involved in renal organogenesis [11, 21].

Consequently, PAX8 should be used in RHB cases with caution, especially when RHB is distinguished from RCC, which is also a PAX8 positive malignancy [22]. Similar considerations apply to CA9 expression, which can also be confusing in differentiating from the far more common ccRCC of the kidney.

According to Barresi et al., brachyury may be a useful diagnostic marker for separate RHB and ccRCC. They tested brachyury expression in 22 HB arising from the central nervous system, in 16 primary and in 8 metastatic RCC. Nuclear and cytoplasmic staining in stromal cells was considered positive, and no such staining was detected in any of the RCC cases. Brachyrury showed high specificity and sensitivity in HB cases [23]. Based on their experience, another review work by Barresi et al. suggested that the same brachyury positivity may be found in RHB cases [24]. There is no data in the literature on brachyury expression in RHB. In our cohort, brachyury showed strong, diffuse cytoplasmic positivity in stromal cells of 2 cases and showed only focal staining in the other 2 cases. Another study suggests that brachyury is not a reliable marker. This study analyzed 22 peripheral HBs, including 3 RHBs; however, it did not specify in which two cases weak focal positivity was observed [25].

Regarding the CD10 expression in RHBs, in the literature 8 RHB cases were described with focal or diffuse CD10 positivity. In these cases, histomorphological features, diffuse α‐inhibin, and S100 positivity confirmed the diagnosis of RHB [9, 10, 13, 14, 15, 16, 17].

In our cohort, we have not examined the expression of other endothelial markers, such as CD31 and ERG. However, some studies demonstrate that stromal cells in nonrenal visceral HBs were negative for CD31 [3], and ERG only highlighted the neoplastic vascularization in cases of CNS‐HB [26]. These results clearly contradict the endothelial origin of stromal cells. Nevertheless, there are some theories about the cell origin of stromal cells in HBs. Most of them assume that stromal cells originate from hemangioblasts, which were impeded in normal embryogenesis due to the loss of VHL. This is also supported by brachyury positivity in these cells, as brachyury is normally only expressed during the early stages of embryogenesis. In addition, they demonstrated that the neoplastic stromal cells of origin in HBs from VHL patients are derived from embryologically arrested hemangioblasts with hematopoietic and endothelial cell potential as well [27]. It is also important to note that the latter findings explain the cell origin only of the VHL‐associated HBs, but not the sporadic cases.

CNS‐HBs have conventionally been reported as PAX8‐negative entities [28], and except one, all peripheral, nonrenal HB cases in the literature are documented as PAX8‐negative tumors as well [25]. In consequence PAX8 positivity in RHB may support an organ‐specific antigen theory, meaning these tumor cells can be derived from organ‐specific pluripotent cells or express site‐specific antigens due to local microenvironmental influence [11]. In contrast, Eichberg et al. [29] reported 7 cerebellar HB with PAX8 expression. They also described that the granular layer of the cerebellum was found to be PAX8 positive, so it may refer to the similar theory from Zhao et al. [11].

4.2. Molecular Characterization

None of the cases in our experience were related to VHL syndrome; therefore, our series included only sporadic tumors. As noted above, we did not identify any pathogenic or likely pathogenic mutations in TSC1, TSC2, or mTOR genes. Actually, the WES analysis described no pathogenic mutations in the three analyzed cases. Based on these findings, we suggest that RHB has not only a distinct immunophenotype, but also a definite genotype, which differs from CNS‐HB.

Unlike our results, Wang et al. [30] investigated the genetic characteristics of 10 RHBs and identified recurrent mutations in the TSC/mTOR pathway. Similarly, in a recent conference abstract, Baranova et al. [31] reported the same observation in three RHBs; however, their cohort consisted exclusively of patients with TSC. Trpkov et al. [32], in their latest research, confirmed TSC/mTOR pathway alterations in 3 RHB cases.

Brachyury expression is a specific diagnostic marker in chordomas, which also refers to the T gene duplication underlying the genetic background of the disease [24, 33]. Brachyury is a member of the T‐box transcription factor family, located at the 6q27 chromosomal locus. T gene duplication is observed in nearly one‐third of sporadic chordomas, causing brachyury overexpression [33]. In our study, we found no genetic explanation for brachyury expression, as there were no pathogenic or likely pathogenic alterations in chromosome 6.

4.3. Differential Diagnosis and Morphological Overlap

The most important differential diagnostic entity is the ccRCC. This is well represented by the fact that in our cohort there was one case with ccRCC as the original diagnosis. RHB can be easily misdiagnosed as ccRCC; however, the at most mild pleomorphism, the prominent capillary network, the lipoblasts‐mimicking vacuolated stromal cells, and the absence of invasive growth should be warning signs of RHB.

It should be noted that some authors divide a rare HB‐like subtype of ccRCC [18, 34], which does not facilitate the clear recognition of the two entities. This subtype of ccRCC with HB‐like features exhibits heterogeneous histology, with areas characterized by small spindle cells intermixed with complex vascular proliferation, nests of large eosinophilic cells, intracytoplasmic vacuoles, hyaline globules, and intranuclear inclusions within the HB‐like component. Some cases also feature focal oedematous, myxoid stroma, alongside regressive changes such as ossification and hemorrhage [18, 34]. Kojima et al. noted transitions and unclear borders between the components, with tumors encased by a thick fibromuscular capsule. α‐inhibin and S100 expression aid in identifying the HB‐like areas, as conventional ccRCC sections typically lack positivity for these markers, while both components express PAX8 [35]. No reported VHL alterations have been identified in ccRCC with HB‐like features. However, Kojima et al. detected recurrent mTOR variants in their analysis. They concluded that ccRCC with HB‐like features does not represent a distinct morphological variant but rather exhibits a relatively indolent behavior compared to conventional ccRCC [35]. To date, no ccRCC with HB‐like features case with pathogenic VHL abnormalities has been documented [4].

It is important to highlight the extreme rarity of these tumors; by the conclusion of 2023, only nine cases of ccRCC with HB‐like features had been reported [36, 37, 38]. Even more complicated is the extremely rare RCC subtype, RCC with fibromyomatous stroma (RCC FMS) features HB‐like areas. Based on the common morphologic and genetic background, Baranova et al. and Trpkov et al. [31, 32] proposed that RHB lies on the morphologic spectrum of RCC FMS.

While CNS‐HB and RHB are often considered the same entity occurring in different locations, the role of VHL alterations in RHB pathogenesis remains uncertain, as several studies showed no association with the VHL pathway in the genetic background of RHB.

Consequently, distinguishing between four entities—CNS‐HB, RHB, and ccRCC with HB‐like features—despite their similar morphological appearance, becomes imperative due to their vastly different clinicopathological features.

4.4. Clinical Implications

RHB typically follows an indolent or even benign clinical course, requiring no further treatment after surgical resection. While nephrectomy is often the definitive treatment for both RHB and ccRCC, the emotional burden and stigma associated with a malignant histological diagnosis may lead to unnecessary anxiety, overtreatment, and long‐term psychological distress for patients. Unlike to CNS‐HB there is no evidence for targeted genetic alteration in sporadic RHB cases. Secondary polycythemia associated with RHB can cause further diagnostic challenges, as Lee et al. [39] suggest in their case report describing an RHB case with erythrocytosis. On the other hand, polycythemia is a well‐known paraneoplastic symptom of RCC and CNS‐HB [40, 41].

4.5. Limitations

Before drawing conclusions, it is important to acknowledge the limitations of our study. First, the number of cases analyzed was low, although our findings were cross‐referenced with an extensive literature review. During the planning phase, we reviewed all ccRCC cases excluding biopsy samples in our archive (n = 2139), reaffirming that RHB is an exceptionally rare entity. Our experience indicates that RHB may have a frequency less than 1% among nephrectomy and resection specimens. The low incidence observed in our cohort aligns with the limited number of cases reported in the literature.

Second, we utilized WES to characterize our RHB cases. WES was chosen for its reliability, cost‐effectiveness, and ability to comprehensively analyze coding regions of tumor DNA. However, WES has inherent limitations: it cannot detect large genomic alterations and genetic changes in noncoding regions such as intronic, regulatory, or intergenic regions. Additionally, WES did not provide reliable data on the copy number alterations of genes and chromosomes.

Lastly, as discussed earlier, RHB, CNS‐HB, ccRCC, and ccRCC with HB‐like features share numerous histological and immunomorphological characteristics, which are summarized in Table 3. A larger comparative study is necessary to thoroughly investigate these overlapping features. Conducting such a study is a key objective in our ongoing research efforts.

TABLE 3.

Comparision of pathological and genetic features of central nervous system hemangioblastoma, renal hemangioblastoma, and clear cell renal cell carcinoma.

	CNS‐HB	RHB	ccRCC
Macroscopic appearance	Yellow cut surface, mostly cystic, well‐circumscribed	Gray‐brown cut surface, solid, well‐circumscribed	Golden‐yellow cut surface, hemorrhage, frequently cystic and fibrotic
Morphological characteristics	Clear, vacuolated stromal cells, abundant vascular network	Sheet‐like or solid growths, clear or eosinophilic or vacuolated stromal cells, hyaline globules, abundant vascular network	Clear cytoplasm, diverse growth pattern, vascular network
IHC	Inhibin+, S100+, Brachyury (cytop)+, CA9+, PAX8−	PAX8+, Inhibin+, S100+, CA9+/−, CD10−, Brachyury (cytop) +/−	PAX8+, CA9+, Inhibin−, S100−, Brachyury−
Genetic background	VHL alteration	Not known	VHL alteration
Incidence	2% of intracranial neoplasms [37]	31 reported cases	60%–75% of all RCCs [38]

Abbreviations: CA9, carbonic anhydrase 9; ccRCC, clear cell renal cell carcinoma; CD, cluster of differentiation; CNS‐HB, central nervous system hemangioblastoma; cytop, cytoplasmic; IHC, immunohistochemistry; PAX8, paired box protein 8; RCC, renal cell carcinoma; RHB, renal hemangioblastoma; VHL, von Hippel–Lindau.

5. Conclusions

In this study, we investigated the pathological, immunohistochemical, and genetic characteristics of four RHB cases. PAX8 was consistently positive in each case, reinforcing the tissue‐specific origin of these tumors. Distinguishing RHB from ccRCC is crucial for oncological management, and the strong expression of α‐inhibin and S100 may aid in establishing an accurate histological diagnosis.

We investigated only sporadic RHBs. Unlike CNS‐HB, the RHB cases in our cohort did not harbor VHL mutations or TSC/mTOR pathway‐related genetic alterations, contrary to recent findings in the literature. Based on our experience, RHB appears to be an indolent and potentially benign soft tissue tumor of the renal parenchyma.

Funding

This research was funded by the University of Szeged, Faculty of Medicine Research Fund‐Hetényi Géza Grant (Grant No. 5S 340 A202); the New National Excellence Programme (Grant No. ÚNKP‐21‐4‐SZTE‐131, ÚNKP‐22‐4‐305); and the HUN‐REN‐ONKOL‐TTK‐HCEMM Oncogenomics Research Group.

Conflicts of Interest

The authors declare no conflicts of interest.

Pósfai B., Jenei A., Forika G., et al., “Hemangioblastoma of the Kidney—A Comprehensive Clinical, Pathological, and Genetic Analysis of Four Cases,” APMIS 134, no. 1 (2026): e70147, 10.1111/apm.70147.

Data Availability Statement

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.