D2-40 is a sensitive and specific marker in differentiating primary adrenal cortical tumours from both metastatic clear cell renal cell carcinoma and phaeochromocytoma
- 1Department of Cellular Pathology, Wycombe Hospital, High Wycombe, UK
- 2The John Radcliffe Hospital, Oxford, UK
- Dr D Bailey, Department of Cellular Pathology, Wycombe Hospital, High Wycombe HP11 2TT, UK;
- Accepted 11 July 2007
- Published Online First 27 July 2007
Background: The morphological similarities between the cells of clear cell renal cell carcinoma (CCRCC) and those of the adrenal cortex impose diagnostic difficulties, for example in the context of a solitary nodule in the adrenal gland in a patient with renal cell carcinoma (RCC). This problem is confounded by the variable and patchy staining seen with the established panel of antibodies utilised in this context, namely EMA, cytokeratins, vimentin, inhibin, melan-A, and RCC marker; particularly on biopsy material. D2-40, an antibody commonly used to highlight lymphatic endothelial cells, is consistently positive in the normal adrenal cortex.
Aims: To investigate the utility of D2-40 in distinguishing neoplastic and non-neoplastic adrenal cortical cells from those of CCRCC, and from phaeochromocytoma.
Methods: D2-40 antibody was applied to tissue sections from 10 normal adrenal glands, 15 renal carcinomas (13 clear cell, 2 papillary variants), 1 metastatic CCRCC in the adrenal gland, 6 adrenal cortical hyperplasias, 5 adrenal cortical adenomas, 3 adrenal cortical carcinomas, and 4 phaeochromocytomas.
Results: D2-40 was strongly and diffusely positive in the cells of the neoplastic and non-neoplastic adrenal cortex, but negative in the cells of the CCRCC, both primary and metastatic, in 100% of the cases. The cells of the adrenal medulla, and those of the phaeochromocytomas, were negative for D2-40.
Conclusions: D2-40 may be a useful marker for distinguishing primary adrenal cortical neoplasms from both metastatic CCRCC and phaeochromocytoma.
The histological similarities between clear cell renal cell carcinoma and adrenal cortical cells and their neoplastic counterparts, have long been recognised. Indeed, Grawitz introduced the term ‘hypernephroma’ in 1883 to describe clear cell renal cell carcinoma (CCRCC), which it was believed was derived from ‘hypernephroid’ (heterotopic adrenal rests) within the kidney. These similarities pose a problem in the setting of diagnostic surgical pathology; for example, in distinguishing metastatic clear cell renal cell carcinoma from a primary adrenal neoplasm in a biopsy from a solitary nodule in the adrenal gland, particularly in relation to a staging procedure for renal carcinoma. Until now, the establishment of such a diagnosis has been reliant on a panel of antibodies, including EMA, cytokeratins, vimentin, and more recently melan A, CD10, inhibin-α, and renal cell carcinoma marker.1 However, the inconsistency in staining patterns observed with these antibodies in this setting reduces the reliability of this panel.
D2-40 is a monoclonal antibody that was developed for use on paraffin-embedded, formalin fixed tissue. The antibody is directed to the M2A antigen; an O-linked sialoglycoprotein that is a membrane antigen present in the testis on fetal gonocytes, and which is re-expressed on testicular germ cell tumours and intratubular germ cell neoplasia.2 It has subsequently been recognised that the M2A antigen is identical to podoplanin,3 thus D2-40 recognises podoplanin. It has been shown that podoplanin is expressed by several normal human tissues, such as peritoneal mesothelial cells, osteocytes, glandular myoepithelial cells, ependymal cells, and by stromal reticular cells, and follicular dendritic cells of lymphoid organs,3 although the role of this membrane protein remains to be firmly elucidated.4
The diagnostic utility of D2-40 in testicular and ovarian germ cell neoplasms is well-documented,4 and it has also been shown that D2-40 is a sensitive marker for seminoma metastases and primary extragonadal seminomas.5
More recently it has been recognised that D2-40 highlights lymphatic endothelial cells,6 a characteristic that is being utilised particularly in the diagnosis of lymphatic invasion. D2-40 has also been of use in the differential diagnosis of vascular tumours, and several studies have reported D2-40 positivity in the neoplastic cells of other solid tumours.4
We have investigated the utility of D2-40 in differentiating non-neoplastic adrenal cortical cells and those of primary adrenal cortical neoplasms, from metastatic clear cell renal cell carcinoma, and from phaeochromocytoma.
MATERIALS AND METHODS
Tissue was obtained from our archives of routine surgical specimens received in our respective departments during 2005–06, with the exception of one specimen from 1997. Sections were obtained from 10 normal adrenal glands received as part of radical nephrectomies for renal cell carcinoma. Ten sections were received from the corresponding renal cell carcinoma (8 clear cell variants, 2 papillary variants), with background normal renal cortex, medulla, +/− pelvicalyceal urothelium; 5 sections of similar tissue were obtained from nephrectomy specimens containing a clear cell renal cell carcinoma, each received without adrenal glands. One section from an adrenal gland with metastatic clear cell renal cell carcinoma was obtained. In addition, we obtained tissue from 6 adrenal glands with cortical hyperplasia (from 3 patients; bilateral adrenalectomy had been performed in each case), 5 adrenal cortical adenomas, 3 adrenal cortical carcinomas, and 4 phaeochromocytomas.
These tissue samples had been fixed in 10% neutral-buffered formalin, and embedded in paraffin wax. Sections (3 µm) were cut and stained with H&E, and further sections were immunostained with monoclonal anti-human D2-40 antibody (DakoCytomation Inc., Carpinteria, California, USA) according to the manufacturer’s instructions. Spurious immunostaining was controlled for by using a corresponding section from each paraffin block as a negative control for each case. This would also act to control for endogenous biotin expression; each negative control was verified as negative during the analysis.
In each case, an immunostaining reaction was interpreted as positive when there was strong cytoplasmic staining with D2-40 antibody, in accordance with the manufacturer’s datasheet. We used the lymphatic endothelium in each section as a comparative positive internal control. The slides were interpreted independently by two pathologists, and a consensus achieved regarding positive and negative staining in each case.
There was diffuse, strong granular cytoplasmic staining for D2-40 in the cells of all 10 (10/10, 100%) normal adrenal cortices. The nuclei were negative and there was no membranous accentuation of the immunostaining. The positive staining was graduated within the layers of the adrenal cortex; it was strongest within the cells of the zona reticularis compared to the zona fasciculata and weakest in the zona glomerulosa. The cells of the adrenal medulla were consistently negative (fig 1A). The malignant cells of the 13 clear cell renal cell carcinomas cases (13/13, 100%) were negative for D2-40 (fig 1C,D ). This was regardless of Fuhrmann nuclear grade, although the areas of sarcomatoid change in one case showed a faint blush of positivity, which was comparatively much less then the strong positivity observed in the adrenal cortices (fig 2A). In addition, the cells of the metastatic clear cell renal cell carcinoma in the adrenal gland (1/1) were negative for D2-40, while the cells of the adrenal cortex around the metastatic tumour nodule were positive (fig 2C). The two examples of papillary renal cell carcinoma showed variable positivity for D2-40 (fig 2B).
D2-40 was strongly and diffusely positive in all cases of adrenal cortical hyperplasia (6/6 glands, 100%), and cortical adenoma (5/5, 100%) (fig 3A,B). There was strong positive staining for D2-40 in the adrenal cortical carcinomas (3/3, 100%), although this appeared to be patchy in one case in which the cellular pleomorphism was most marked (fig 3C,D). The positive staining intensity and the pattern of staining in the adrenal cortical tumours was similar to that seen in the normal adrenal cortices. The tumour cells of the adrenal phaeochromocytomas were, by comparison, negative (4/4), although in each of these cases the corresponding adrenal cortex was strongly and diffusely positive (fig 2D).
Interestingly, there was strong granular cytoplasmic positivity for D2-40 in the renal tubules, although this was limited to the proximal tubules, the distal tubules and collecting ducts being consistently negative (fig 1B). The urothelium of the pelvicalyceal system, when present in the selected sections from the kidney, was consistently negative for D2-40.
In all cases, positive D2-40 expression by lymphatic endothelial cells within the sections stained was verified as an internal positive control.
Since the development of D2-40 as a monoclonal antigen to podoplanin, the spectrum of its diagnostic utility has rapidly broadened.4 D2-40 is now commonly employed in the assessment of lymphatic invasion in tumour resection specimens, and increasingly D2-40 positivity has been noted in the neoplastic cells of several tumours, in addition to its established positivity in testicular and ovarian germ cell neoplasms. These include those of cerebellar haemangioblastomas,7 and other brain tumours,8 squamous cell carcinoma of the skin,3 and of the lung4; and in the distinction of enchondroma and low grade chondrosarcoma, which are D2-40 positive, from chordoma in the central nervous system.9 The utility of D2-40 in pleural biopsy material,10 and subsequently in serous effusions,11 12 has been described whereby it has been reported that D2-40 highlights both benign and malignant mesothelial cells with a high degree of sensitivity and specificity.
We have shown that D2-40 is a specific marker of adrenal cortical cells in the normal and hyperplastic adrenal gland, and its neoplastic counterparts, ie adrenal cortical adenoma and carcinoma. The cells of the adrenal medulla do not stain with D2-40. The intensity of the positive cytoplasmic staining for D2-40 in each case was strong and uniform; this was similar in the cells of both the normal adrenal cortices, and the neoplastic adrenal cortical cells.
By contrast, the malignant cells of the clear cell variant of renal cell carcinoma, both primary in the kidney and metastatic to the adrenal gland, and regardless of Fuhrmann grade, did not stain with D2-40. This is consistent with the findings of a previous study in which the cells of CCRCC, both primary in the kidney and metastatic to the brain, were negative for D2-40,7 and with a study that included 35 cases of renal cell carcinoma, all of which were negative for D2-40.10 It was shown that among the malignant cells in both examples of papillary renal cell carcinoma (type 1), there was variable positivity for D2-40. These tumours are known to have a different immunohistochemical profile to the clear cell variant of renal cell carcinoma, and since the morphological appearance of a papillary renal cell carcinoma is distinct from the clear cell variant, this should not cause diagnostic difficulties in differentiation from adrenal cortical cells.
Renal cell carcinoma is a common clear cell tumour to metastasise to the adrenal gland, and it has been shown that of the subtypes of renal cell carcinoma, the clear cell variant is the most likely to metastasise.13 The morphological similarities between the tumour cells in CCRCC and the neoplastic and non-neoplastic cells of the adrenal cortex are well described, and the distinction is often reliant on a panel of antibodies (see table 1). However, the expression of some of the antigens targeted by these antibodies can be focal, and may in some cases be negative. This problem is confounded by the small amount of tissue present in biopsy material.
Our findings are significant in that D2-40 may be utilised in distinguishing metastatic CCRCC from a primary adrenal cortical tumour, particularly in the setting of a solitary adrenal mass on biopsy. It may also aid in distinguishing adrenal cortical tumours from phaeochromocytoma. The positive staining pattern for D2-40 in the cells of the normal and the neoplastic adrenal glands in our series was strong and diffuse, and it was 100% specific to these cells in comparison with CCRCC, both within the kidney and metastatic to the adrenal gland. This makes positivity of the D2-40 antibody in this context easy to interpret, and more reliable in small biopsy specimens.
Due to the similarity in the staining pattern between non-neoplastic and neoplastic adrenal cortical cells, D2-40 would not be useful in distinguishing between the normal adrenal cortex, cortical hyperplasia, or an adrenal cortical tumour.
Although we have shown the utility of D2-40 in surgical material, the fact that D2-40 has been used successfully in cytological preparations is promising, as FNA assessment of the solitary adrenal lesion is well-established,14 15 and the cytopathologist is faced with the same diagnostic difficulties as are posed to the surgical pathologist. Thus D2-40 may also prove useful in the assessment of FNA material from the adrenal gland.
In conclusion, we have shown that D2-40 may be a useful immunohistochemical marker for distinguishing primary adrenal cortical neoplasms from both metastatic clear cell renal cell carcinoma, and phaeochromocytoma. These findings, to the best of the authors’ knowledge, have not been described previously.
The morphological similarities between clear cell renal cell carcinoma and neoplastic and non-neoplastic adrenal cortical cells are well recognised.
Strong and diffuse cytoplasmic positivity for D2-40 in the cells of the neoplastic and non-neoplastic adrenal cortex, with negative staining in clear cell renal cell carcinoma may be useful in their distinction, even in small biopsy specimens.
D2-40 is negative in the cells of the adrenal medulla and thus may be useful in the distinction of adrenal medullary tumours from the neoplastic and non-neoplastic cells of the adrenal cortex.
D2-40 may not be useful in the distinction of neoplastic from non-neoplastic adrenal cortex due to the strong and diffuse cytoplasmic positivity in both cell types.
Thanks are extended to the laboratory staff at both Wycombe Hospital and the John Radcliffe Hospital for their technical assistance, in particular Katarzyna Gierejko, Claire Krelle and Angie Vockins; and to Helene Beard at the John Radcliffe Hospital for her assistance with the photography.