Does renal haemangioblastoma exist? Evidence for inclusion within the MTOR‐associated renal cell carcinoma with fibromyomatous stroma spectrum

Abbreviations

RCC

renal cell carcinoma

TSC

tuberous sclerosis complex

VHL

von Hippel–Lindau

To the Editor,

Recently several studies have re‐examined renal haemangioblastoma and haemangioblastoma‐like renal cell carcinoma (RCC), including several published in Histopathology. ¹ , ² , ³ , ⁴ , ⁵ Historically, a number of reports described putative primary renal haemangioblastomas, many of which demonstrated PAX8 positivity, a finding that contrasts with central nervous system haemangioblastomas, in which negative PAX8 is used to differentiate from metastatic RCC. ⁶ This paradoxical positivity for PAX8 had been speculated to be potentially related to the organ of origin, in which a tumour originating in an unusual location may have different immunohistochemical features from its prototypical site, possibly due to the cell of origin or other factors. ⁶

However, recent molecular and pathological studies suggest that renal haemangioblastoma is not a counterpart of nervous system or soft tissue haemangioblastoma, but rather a morphological variation within the spectrum of TSC/MTOR–driven RCC with fibromyomatous stroma (Figure 1). As such, we believe that the majority of, if not all, putative primary renal haemangioblastomas (and haemangioblastoma‐like RCC) instead reflect haemangioblastoma‐like morphology within TSC/MTOR–associated RCC with fibromyomatous stroma (Table 1).

Figure 1.

(A) This tumour shows a mixture of epithelial and haemangioblastoma‐like areas. (B) At higher magnification, epithelial areas are shown at left and haemangioblastoma‐like areas at right. (C) Keratin 7 labelling is present only in the epithelial areas. (D) Labelling for GPNMB is present in both haemangioblastoma‐like areas (stronger) and epithelial areas (weaker). (E) S100 labels primarily the haemangioblastoma‐like areas, with similar results for inhibin (F). (G) This tumour is a renal cell carcinoma with fibromyomatous stroma that did not contain areas resembling haemangioblastoma. (H). Despite a lack of haemangioblastoma‐like morphology, there is substantial positive staining for inhibin.

Table 1.

Comparison of renal haemangioblastoma/haemangioblastoma‐like RCC to CNS haemangioblastoma and RCC with fibromyomatous stroma

	Renal haemangioblastoma /haemangioblastoma‐like RCC	CNS haemangioblastoma	TSC/MTOR RCC with fibromyomatous stroma
Carcinoma component	Subset	No	Yes
PAX8	Frequently positive	Negative	Positive
Keratin 7	Yes, especially if mixed morphology is present	No	Yes
S100/inhibin	Positive in haemangioblastoma‐like areas	Positive	Limited data †
GPNMB	Positive	Majority negative	Positive
Gene mutations	TSC/MTOR	VHL	TSC/MTOR
Occurrence in patients with tuberous sclerosis complex	Yes	No	Yes
Occurrence in patients with von Hippel–Lindau disease	No*	Yes	No

Shaded boxes = features that link renal haemangioblastoma to RCC with fibromyomatous stroma rather than true haemangioblastoma.

^*Some reports have implied the existence of renal haemangioblastoma in VHL patients, but no definitive report of this occurrence is noted.

^† Anecdotally, we have encountered positivity in some examples (Figure 1).

The majority of the tumours studied recently now show gene alterations in the TSC/MTOR pathways, even in those with ‘pure’ haemangioblastoma morphology. ² This is counterintuitive, as CNS haemangioblastoma is classically associated with von Hippel–Lindau (VHL) disease rather than tuberous sclerosis complex (TSC). However, accumulating evidence indicates that renal haemangioblastoma is not related to VHL and can occur in both sporadic and syndromic settings, particularly in association with TSC. It should be noted that, although earlier reports emphasized the purportedly ‘sporadic’ nature of renal haemangioblastoma, there is a lack of definitive documentation of true renal haemangioblastoma arising in patients with VHL disease.

Taken together, emerging molecular, immunohistochemical and morphological data suggest that most, or perhaps all, renal haemangioblastomas and haemangioblastoma‐like RCCs actually represent TSC/MTOR‐associated RCC. In addition to the paradoxical PAX8 positivity described previously, analyses of tumours with mixed morphologies demonstrate that haemangioblastoma markers such as S100 and inhibin are preferentially expressed in haemangioblastoma‐like areas, whereas epithelial markers characteristic of RCC with fibromyomatous stroma, including keratin 7, highlight the carcinoma component. ² In tumours dominated by haemangioblastoma‐like architecture, the absence of overt fibromyomatous carcinoma morphology may obscure the diagnosis. However, the recognition of tumours spanning a continuum, from predominantly fibromyomatous RCC to predominantly haemangioblastoma‐like, supports the concept that haemangioblastoma‐like morphology represents a pattern within the spectrum of TSC/MTOR‐altered RCC with fibromyomatous stroma. Accordingly, it seems likely that most previously reported cases of ‘renal haemangioblastoma’ correspond to haemangioblastoma‐like RCC lacking sufficiently conspicuous fibromyomatous features to reveal their true nature.

Author contributions

SRW and RA developed the concept for this editorial. SW prepared the first draft. RA provided critical review and editing.

Funding information

No funding was received for this work.

Conflict of interest

The authors declare no related conflicts of interest.

Consent for publication

Not applicable.

Data availability statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.