Distinguishing oncocytomas from their malignant mimics is very challenging. This review highlights our approach to classifying low-grade oncocytic tumors on both resections and biopsies. We also discuss how we use immunohistochemical stains in this challenging differential diagnosis.

Discussing the differential diagnosis of low-grade oncocytic renal neoplasms is difficult because the main differential is between low-grade oncocytic renal cell carcinomas, which generally behave indolently, and benign oncocytomas. In general, low-grade oncocytic renal cell carcinomas have a real but very low risk of metastases (typically years after presentation). There is also, unfortunately, no way to define these entities (for example, through molecular phenotype), so although individual pathologists may vary in their threshold for oncocytoma versus low-grade oncocytic renal cell carcinoma, there is no way to determine what threshold is actually correct.1  However, there are cases that the majority of genitourinary pathologists would agree are oncocytomas, cases that the majority agree are not oncocytomas, and cases in which we do not agree. This review discusses our general approach to low-grade oncocytic renal neoplasms on both resections and biopsies.

The differential diagnosis of oncocytic tumors on resections includes an oncocytoma; an eosinophilic variant of chromophobe renal cell carcinoma; a low-grade renal cell carcinoma with oncocytic features, unclassified type; a succinate dehydrogenase (SDH)–deficient renal cell carcinoma; and a hybrid oncocytic tumor.

Differential Diagnosis on Resection

Oncocytoma

A classic oncocytoma has a central hyalinized or edematous scar with rounded nests of tumor cells; toward the periphery of the tumor these nests can be back to back, imparting a more solid growth pattern. The cells in an oncocytoma have even, dense, granular cytoplasm with indistinct cell borders, and the nuclei are relatively evenly spaced within the tumor because of nuclei being centrally placed within the tumor cells. Most importantly, the nuclei in oncocytomas are uniform, monomorphic, and round with smooth nuclear contours. Nucleoli are usually present, but the chromatin otherwise looks fine and nonhyperchromatic (Figure 1, A). Variable features that can be seen in oncocytomas include cystic dilatation of the tumor nests; very focal pseudopapillae within tumor nests; focal trabecular architecture, particularly in the area of the hyalinized scar; and focal loss of cytoplasm in the tumor cells, imparting an “oncoblastic” appearance (Figure 1, B). Degenerative-type nuclear atypia, generally in the form of focally enlarged nuclei with hyperchromatic, smudgy chromatin, is allowable (Figure 1, C). Oncocytomas can extend into the perinephric or renal sinus adipose tissue without any stromal response and can grow into the renal vein (Figure 1, D). These features can cause consternation for the pathologist but do not impart any implication for clinical behavior or prognosis.24  Features that are not allowable in an oncocytoma are more than a very occasional mitosis; readily identified mitoses at scanning power essentially excludes an oncocytoma. Gross or conspicuous necrosis, diffuse or extensive papillary areas, and spindled/sarcomatoid areas exclude an oncocytoma. It is somewhat controversial whether very focal chromophobe-like areas are allowed in an oncocytoma; although some argue that focal chromophobe-like areas can be seen in oncocytomas, in our practice, we generally would not consider true chromophobe-like areas to be compatible with an oncocytoma.3,5 

Eosinophilic Variant of Chromophobe Renal Cell Carcinoma

The eosinophilic variant of chromophobe renal cell carcinoma can have a low-power architecture that can closely mimic an oncocytoma, however, often the eosinophilic variant of chromophobe renal cell carcinoma has a more solid and trabecular architecture than a classic oncocytoma. Often subtle areas of cells with more flocculent cytoplasm and accompanying prominent cell borders are present. The key to the correct diagnosis lies in the nuclei; although the nuclei are not as pleomorphic as those seen in a classic chromophobe renal cell carcinoma, subtle irregularities in nuclear contours, subtle pleomorphism, and focal perinuclear haloes are usually present upon careful examination (Figure 2, A).5  Nuclei in the eosinophilic variant of chromophobe renal cell carcinoma are also not as evenly spaced as those seen in an oncocytoma because of variations in overall cellularity and nuclear size and shape.

Low-Grade Oncocytic Renal Cell Carcinoma, Unclassified Variant

There are occasional tumors that have the architecture of an oncocytoma but also have nondegenerative-type nuclear atypia that is too marked or diffuse for a diagnosis of oncocytoma. These cases typically have easily recognizable nuclear atypia at scanning power, but do not have the irregular nuclear contours or perinuclear haloes of a chromophobe renal cell carcinoma and do not have the classic features of other oncocytic renal cell carcinomas. In these cases, in which the tumor resembles an oncocytoma but has nuclear atypia beyond what is allowable for oncocytoma, the diagnosis of a low-grade oncocytic renal cell carcinoma, unclassified variant, is warranted (Figure 2, B). These renal cell carcinomas are not low grade by strict Fuhrman nuclear grading criteria but are generally indolent, with only exceptionally rare cases of metastases/recurrence reported. Therefore, I generally do not grade these tumors but instead sign these out with a comment that these are generally indolent tumors.

SDH-Deficient Renal Cell Carcinomas

There is a complex of SDH-deficient syndromes in which there is a loss of function of the SDH enzyme, which is involved in the Krebs cycle and in the electron transport chain. Succinate dehydrogenase complex deficiency syndromes are associated with multiple tumors, including extra-adrenal paragangliomas, pheochromocytomas, gastrointestinal stromal tumors, and renal cell carcinomas.6  Succinate dehydrogenase–deficient renal cell carcinomas generally have a more solid architecture at scanning power without the typical rounded nests seen in oncocytomas. A central scar (albeit generally associated with more irregular as opposed to rounded nests of tumor cells) can be present. In general, these tumors have low-grade–appearing nuclei, but there is readily identifiable nuclear irregularity, and the nuclei are generally not as round as those seen in an oncocytoma (Figure 2, C). One characteristic feature is the presence of pale eosinophilic to clear cytoplasmic inclusions, although this is not present in all cases. SDH-deficient renal cell carcinomas have a metastatic rate of 11% at long-term follow-up.7  A confirmatory immunostain for SDHB that shows loss of staining is available.8 

Hybrid Oncocytic Tumors

There has been some disagreement and confusion as to the proper context in which to use this diagnosis.1  We do not use this as a catchall term for any oncocytic tumor that is difficult to classify, but rather use it to refer to a specific tumor that is classically associated with Birt-Hogg-Dube syndrome. These tumors often have a more solid and alveolar architecture than is typically seen in oncocytoma and are typically of low nuclear grade. A distinctive morphologic feature is clusters of cells with clear cytoplasm; this abrupt change in cytoplasm can be appreciated at scanning power (Figure 2, D). In patients who are known or suspected to have Birt-Hogg-Dube syndrome, tumors with a mixture of oncocytoma and chromophobe renal cell carcinoma–like areas may also be diagnosed as hybrid oncocytic tumors.9  Of note, patients with Birt-Hogg-Dube syndrome usually have multiple, bilateral tumors on imaging.10  Birt-Hogg-Dube is also one of the few syndromes for which patients may reach adulthood without detection; it is one of the few syndromes for which a pathologist raising the possibility of Birt-Hogg-Dube not uncommonly leads to initial detection.

Oncocytic Tumors on Resection With Borderline Atypia

There will be some cases in which individual pathologists may disagree as to whether or not the amount of atypia (typically nuclear atypia) is still compatible with an oncocytoma. Unfortunately, there is no consensus on how to handle these “gray-zone” cases. One abstract has attempted to tackle classification of these cases, with the ultimate finding that even those cases in which 2 genitourinary experts did not agree (1 favoring oncocytoma and 1 favoring nononcocytoma) did not have documented metastases or disease-related deaths on follow-up.11  As in any threshold cases, there will always be some disagreement on how much atypia is allowable in an oncocytoma, even among genitourinary pathologists at a single institution. There have been exceptional gray-zone cases in which I have used the diagnosis of low-grade oncocytic tumor. In a comment, I state that the differential includes both an oncocytoma and a low-grade indolent renal cell carcinoma and that the atypia in the tumor is somewhat beyond what is typically seen in an oncocytoma; therefore, the patient may benefit from some clinical follow-up. This approach is supported in a recent study by Williamson et al1  in which 82% of genitourinary pathologists support such an approach in oncocytic tumors with features inconclusive for definitive benign or malignant classification.

Role of Immunostains in Oncocytic Tumors on Resections

Immunostains can be helpful in problematic oncocytic tumors on resections. Chromophobe renal cell carcinomas are typically c-kit (CD117)–positive and variably positive for cytokeratin 7 (CK7), whereas oncocytomas are typically positive for c-kit (CD117) and show very focal scattered positivity for CK7.12  However, in contrast to biopsies (see below), we do not feel it is necessary to routinely perform immunostains in all oncocytic tumor resections.

Classifying oncocytic tumors on core biopsies is somewhat controversial; there are some pathologists who may argue that accurate classification of oncocytic tumors on a biopsy is impossible.1  At the University of Michigan (Ann Arbor), we have been classifying oncocytic tumors on core biopsies for many years. We are generally more conservative with oncocytic tumors on biopsies. Our terminology includes “oncocytic tumor, favor oncocytoma” and “oncocytic tumor, favor/consistent with renal cell carcinoma (with subtype if possible).” There are some cases in which the tumor has architectural or nuclear atypia beyond what we typically see in oncocytoma but not sufficient atypia to favor a renal cell carcinoma. We have used the terminology “oncocytic tumor, cannot exclude renal cell carcinoma” in these cases. In general, our clinicians have presumptively treated such cases as if they are low-grade renal cell carcinomas.13,14  It is important not to use this diagnosis too often; at our institution, such cases account for only 21% of our oncocytic neoplasm biopsies.13 

Immunostains in Oncocytic Core Biopsies

At least at our institution, we perform immunostains on nearly all of our oncocytic core biopsies. At a minimum, we perform a c-kit (CD117) and a CK7. The c-kit should be diffusely positive in both oncocytomas and chromophobe renal cell carcinomas, can be positive in unclassified oncocytic renal cell carcinomas, and is usually negative in all other types of renal cell carcinomas. Cytokeratin 7 is usually positive in single scattered tumor cells and focal clusters of cells in oncocytomas and is usually diffusely positive in classic chromophobe renal cell carcinomas. Cytokeratin 7 can show scattered tumor cell positivity in the eosinophilic variant of chromophobe renal cell carcinoma and in unclassified oncocytic renal cell carcinomas. Other immunostains that can be ordered but are of more questionable value are vimentin, which generally shows an inverse staining pattern compared with c-kit (negative in oncocytomas and chromophobe renal cell carcinomas, and can be negative in unclassified oncocytic renal cell carcinomas; positive in all other renal cell carcinomas) and usually does not add much more information than c-kit. S100A1 is an immunostain that is reportedly positive in oncocytomas and negative in chromophobe renal cell carcinomas. In our recent experience, S100A1 appears to not be as sensitive and specific as previously reported and is not widely available.13  The Table summarizes expected immunoprofiles in oncocytic renal neoplasms.

Approach to Oncocytic Tumors on Core Biopsy

In general, we are more cautious with oncocytic neoplasm biopsies; we would rather have a few patients with oncocytoma undergo excision than undertreat a low-grade renal cell carcinoma. Therefore, at our institution, in order to diagnose a biopsy as oncocytic tumor, favor oncocytoma, both the morphologic features and the immunohistochemical profile need to be consistent with an oncocytoma.

Differential Diagnosis on Core Biopsy

Oncocytoma

Oncocytomas on core biopsies have similar morphologic features to those seen on resections; typical features that would favor an oncocytoma would be rounded nests, an edematous/hyalinized central scar, and dense, even granular cytoplasm. Most importantly, an oncocytoma on a core biopsy typically has relatively monomorphic, bland, round nuclei (Figure 3, A). In our series of 144 oncocytic neoplasm biopsies, we found a single mitosis in only 2 of 96 favor oncocytoma biopsies; the remainder of our favor oncocytoma biopsies did not have identifiable mitoses.13 

Eosinophilic Variant of Chromophobe Renal Cell Carcinoma

The main differential diagnoses would include an eosinophilic variant of chromophobe renal cell carcinoma. Again, chromophobe renal cell carcinomas generally have a more solid architecture, consisting of sheets or trabeculae of tumor cells. They also generally have some subtle areas of flocculent cytoplasm with associated prominent cell borders. Most importantly, nuclear pleomorphism, subtle nuclear irregularity, and focal perinuclear haloes are frequently present (Figure 3, B).

Papillary Renal Cell Carcinoma With Oncocytic Cytoplasm

We have seen several papillary renal cell carcinomas on core biopsy that mimic an oncocytic neoplasm; for some reason, without the benefit of a resection where the papillary architecture is apparent, papillary renal cell carcinomas with oncocytic cytoplasm can be (at least initially) mistaken for an oncocytoma. Careful attention to focal or subtle areas of papillary formation, subtle variability in nuclear contours, and the presence of macrophages within fibrovascular cores can lead to the correct diagnosis. In addition, hemosiderin is often present in papillary renal cell carcinomas (Figure 3, C). In difficult cases, immunostains should help resolve the diagnosis; papillary renal cell carcinomas are negative for c-kit, positive for vimentin, and variably positive for CK7.

Borderline Cases on Core Biopsy

In biopsies in which there is nuclear irregularity/atypia or unusual architectural features beyond our threshold for an oncocytoma, but not sufficient atypia to favor a renal cell carcinoma, a diagnosis of “oncocytic neoplasm, cannot exclude renal cell carcinoma” should be rendered (Figure 3, D), regardless of the immunohistochemical profile. In addition, any case that morphologically resembles an oncocytoma but has a discordant immunohistochemical profile (either negative for c-kit or more than focally positive for CK7) is generally also diagnosed as oncocytic neoplasm, cannot exclude renal cell carcinoma.

Follow-up of Oncocytic Core Biopsies

Under our classification schema, of 144 oncocytic core biopsies, 67% (96) were diagnosed as favor oncocytoma, 12% (17) were diagnosed as favor renal cell carcinoma, and 21% (31) were diagnosed as cannot exclude renal cell carcinoma. Excisions were limited, but of the 96 favor oncocytoma biopsies, 6 were excised, 5 of which were confirmed as oncocytoma. The sixth case was a challenging gray-zone case on excision that was diagnosed as an oncocytic neoplasm, cannot exclude renal cell carcinoma; on follow-up no metastasis or recurrence has occurred to date. Of 31 cannot exclude renal cell carcinoma biopsies, 10 were excised, 9 of which were confirmed as renal cell carcinoma. One was a hybrid oncocytic tumor on excision. Of 17 favor renal cell carcinoma biopsies, 12 were excised, all of which were confirmed as renal cell carcinoma on excision.13 

In conclusion, both architectural and cytologic features can be used to distinguish between oncocytomas and low-grade renal cell carcinomas; in general, nuclear features are more reliable than architectural features. The majority of oncocytic tumors on core biopsy can be categorized as favor oncocytoma or favor renal cell carcinoma, with good concordance on excision. Immunostains for c-kit and CK7 are performed on all of our oncocytic tumor biopsies, and it is generally prudent to be more cautious on core biopsies, such that a tumor must be morphologically and immunohistochemically compatible with an oncocytoma for a diagnosis of oncocytic neoplasm, favor oncocytoma, to be rendered.

1
Williamson
SR,
Gadde
R,
Trpkov
K,
et al.
Diagnostic criteria for oncocytic renal neoplasms
:
a survey of urologic pathologists [published online ahead of print March 14
,
2017]
. Hum Pathol. doi:.
2
Wobker
SE,
Przybycin
CG,
Sircar
K,
Epstein
JI
.
Renal oncocytoma with vascular invasion: a series of 22 cases
.
Hum Pathol
.
2016
;
58
(
12
):
1
6
.
3
Amin
MB,
Crotty
TB,
Tickoo
SK,
Farrow
GM
.
Renal oncocytoma: a reappraisal of morphologic features with clinicopathologic findings in 80 cases
.
Am J Surg Pathol
.
1997
;
21
(
1
):
1
12
.
4
Trpkov
K,
Yilmaz
A,
Uzer
D,
et al.
Renal oncocytoma revisited: a clinicopathological study of 109 cases with emphasis on problematic diagnostic features
.
Histopathology
.
2010
;
57
(
6
):
893
906
.
5
Tickoo
SK,
Amin
MB
.
Discriminant nuclear features of renal oncocytoma and chromophobe renal cell carcinoma: analysis of their potential utility in the differential diagnosis
.
Am J Clin Pathol
.
1998
;
110
(
6
):
782
787
.
6
Gill
AJ
.
Succinate dehydrogenase (SDH) and mitochondrial driven neoplasia
.
Pathology
.
2012
;
44
(
4
):
285
292
.
7
Gill
A,
Amin
MB,
Smith
SC,
Trpkov
K
.
Succinate dehydrogenase-deficient renal cell carcinoma
.
In
:
Moch
H,
Humphrey
P,
Ulbright
T,
Reuter
VE,
eds
.
WHO Classification of Tumours of the Urinary System and Male Genital Organs. 4th ed
.
Lyon, France
:
International Agency for Research on Cancer;
2016
.
8
Gill
AJ,
Benn
DE,
Chou
A,
et al.
Immunohistochemistry for SDHB triages genetic testing of SDHB, SDHC, and SDHD in paraganglioma-pheochromocytoma syndromes
.
Hum Pathol
.
2010
;
41
(
6
):
805
814
.
9
Hes
O,
Petersson
F,
Kuroda
N,
Hora
M,
Michal
M
.
Renal hybrid oncocytic/chromophobe tumors—a review
.
Histol Histopathol
.
2013
;
28
(
10
):
1257
1264
.
10
Pavlovich
CP,
Walther
MM,
Eyler
RA,
et al.
Renal tumors in the Birt-Hogg-Dube syndrome
.
Am J Surg Pathol
.
2002
;
26
(
12
):
1542
1552
.
11
Huang
Y,
Jadallah
S,
Sharma
N,
et al.
A clinicopathological study of renal oncocytoma and related unclassified oncocytic neoplasms: should the morphologic spectrum of renal oncocytoma be expanded?
Mod Pathol
.
2015
;
28
(
S2
):
230A
.
12
Carvalho
JC,
Wasco
MJ,
Kunju
LP,
Thomas
DG,
Shah
RB
.
Cluster analysis of immunohistochemical profiles delineates CK7, vimentin, S100A1 and C-kit (CD117) as an optimal panel in the differential diagnosis of renal oncocytoma from its mimics
.
Histopathology
.
2011
;
58
(
2
):
169
179
.
13
Alderman
MA,
Daignault
S,
Wolf
JS
Jr,
et al.
Categorizing renal oncocytic neoplasms on core needle biopsy: a morphologic and immunophenotypic study of 144 cases with clinical follow-up
.
Hum Pathol
.
2016
;
55
(
9
):
1
10
.
14
Shah
RB,
Bakshi
N,
Hafez
KS,
Wood
DP
Jr,
Kunju
LP
.
Image-guided biopsy in the evaluation of renal mass lesions in contemporary urological practice: indications, adequacy, clinical impact, and limitations of the pathological diagnosis
.
Hum Pathol
.
2005
;
36
(
12
):
1309
1315
.

Author notes

From the Department of Pathology, Michigan Medicine, University of Michigan, Ann Arbor.

The author has no relevant financial interest in the products or companies described in this article.

Competing Interests

Presented in part at the New Frontiers in Pathology: An Update for Practicing Pathologists meeting; October 13–15, 2016; Ann Arbor, Michigan.