Fernandez-Flores, Angel MD, PhD

To the Editor:

We have read with great interest the report on cutaneous epithelioid angiomatous nodule (CEAN) by Sangüeza et al.¹ We want to thank them for having mentioned a previous report that we published on the subject.²

In their series, the authors studied the expression of some vascular markers by the CEANs, such as CD31 and CD34, and found expression of them in all their 10 cases.¹

In recent years, many vascular conditions have been studied for the expression of podoplanin (marked with the antibody D2-40).³ Because the latter has been considered as a good marker of lymphatic differentiation,^4,5 it is expected that tumors traditionally considered as derived from blood vessels fail to express the marker. That was the case with some blood vessel tumors, such as papillary hemangioma⁶ or retiform hemangioendothelioma.³ On the contrary, the expression of D2-40 by some tumors, such as angiosarcomas^4,7,8 or hobnail hemangiomas,⁹ has reinforced the belief of their lymphatic origin. Nevertheless, different types of immunostaining have been shown in other tumors of alleged blood vessel origin, such as tufted angioma.¹⁰

We thought that it would be relevant to check the expression of podoplanin by the case of CEAN that we reported in 2005.² For that, we used the anti-human D2-40 antibody of DAKO (monoclonal mouse antibody, code N1607).

To our surprise, we found a peculiar pattern of immunostaining. The vessels that were more centrally placed in the nodule failed to express podoplanin, whereas there was an intensive expression by the peripheral vessels (Fig. 1). This pattern did not seem to depend on the “epithelioid” aspect of the vessels because some of the positive peripheral vessels showed an epithelioid endothelium (Fig. 2). In these peripheral areas, some negative, nonepithelioid blood vessels could be found surrounded by the positive vessels (Fig. 3).

Figure 1

Figure 2

Figure 3

When we examined again the hematoxylin-eosin slide, we realized that the central part of the lesion was the one made of solid sheets of epithelioid cells with open channels dispersed within the sheets (Fig. 4, top). On the contrary, the D2-40-positive vessels of the periphery were not immersed in sheets of epithelioid cells, although they had an epithelioid endothelium, as already seen in the immunohistochemical slide (Fig. 4, bottom).

Figure 4

This makes us think if the peripheral component of the lesion could in fact be reactive and not really part of the tumor.

A similar pattern of staining (positive in the peripheral area) has also been previously described in kaposiform hemangioendothelioma.¹⁰

The pattern of immunostaining with D2-40 that we observed in CEAN contrasts with the one described in angiosarcomas, which is diffuse.¹¹ This latter fact can be of great help in the differential diagnosis between CEAN and angiosarcoma, which sometimes can really be difficult.¹

Angel Fernandez-Flores, MD, PhD

Service of Anatomic Pathology

Hospital El Bierzo

Ponferrada, Spain and the

Service of Cellular Pathology

Clinica Ponferrada

Ponferrada, Spain

REFERENCES