Elsevier

Gynecologic Oncology

Volume 92, Issue 1, January 2004, Pages 71-79
Gynecologic Oncology

Immunohistochemical markers in differential diagnosis of endometrial stromal sarcoma and cellular leiomyoma

https://doi.org/10.1016/j.ygyno.2003.08.038Get rights and content

Abstract

Objective. Distinction of endometrial stromal sarcoma (ESS) from benign smooth muscle proliferations like cellular leiomyoma (CL) is sometimes problematic. The purpose of this study was to evaluate the potential utility of a panel of antibodies in the differential diagnosis of ESS and CL.

Methods. Using a standard streptavidin–biotin method, the expression of desmin, alpha smooth muscle actin (SMA), calponin h1, h-caldesmon, estrogen receptor (ER), progesterone receptor (PR), CD10, CD44v3, proliferating cell nuclear antigen (PCNA), and mast cells (MCs) were evaluated in 26 cases of ESS (21 low grade, 5 high grade), 25 CL (17 common CL, 8 highly CL), 25 myometria, and 25 endometria.

Results. Among ESS, 20 of 26, 17 of 26, 9 of 26, 12 of 26, 14 of 26, and 22 of 26 were positive for expression of desmin, SMA, calponin h1, ER, PR, and CD10, respectively, while only 2 of 26 were positive for CD44v3 and all were entirely negative for h-caldesmon. Of CL, all were positive for SMA, calponin h1, PR, and CD44v3; 24 of 25, 24 of 25, and 19 of 25 were positive for desmin, h-caldesmon, and ER, respectively, whereas 1 of 25 focally marked with antibodies to CD10. There was no significant difference of PCNA expression between ESS and CL, although the ESS cases tended to have higher values. The MC counts were significantly higher in the CL group than in the ESS group (P < 0.01). When using the cut-off value of seven MCs per HPF to distinguish ESSs from CLs, the sensitivity and specificity of this cut-off value were 92.9% and 100%, respectively.

Conclusions A panel of h-caldesmon, CD10, and CD44v3 should be used and will distinguish ESS from CL in most cases. In addition, counting the number of MCs might be useful as part of a multivariate approach to the differential diagnosis of them. But the biological function of MC and CD44v3 in these tumors is worthy of further investigation.

Introduction

Endometrial stromal sarcomas (ESSs) are usually composed of uniform small oval cells with scanty cytoplasm and regularly distributed spiral arterioles, closely resembling proliferative phase endometrial stroma, while smooth muscle neoplasms more often have a fascicular arrangement of spindled cells with cigar-shaped nuclei and scattered larger thick-walled vessels. But distinction of ESS from cellular leiomyoma (CL) of the uterus can be sometimes problematic for even the experienced pathologist, because both can show a remarkable degree of architectural and cytologic similarity, particularly because ESS can have smooth muscle-like differentiation and CL, especially highly cellular leiomyoma (HCL) can exhibit dense cellularity, prominent vascularity, and irregular margins [1], [2], [3], [4], [5]. The difficulty in differentiating the two is more pronounced in curetting or myomectomy specimens, as the relation to the surrounding endometrium or myometrium, or both, is lost. But correct classification is of considerable clinical importance due to CL, a benign neoplasm, and ESS, a low-grade malignant neoplasm, are different in clinical behavior and treatment.

Immunohistochemical markers have recently been used in an attempt to provide aid in resolving the problems that arise because of this overlap [3], [4], [5], [6], [7], [8], [9], [10], [11], particularly when careful evaluation of routine morphology, obviously the first step, is inconclusive. Perhaps the most noteworthy new development in this area is the suggestion that desmin, and more recently calponin h1 and h-caldesmon, are useful in the identification of smooth muscle differentiation [3], [4], [5], [6], [7], [8]. Additionally, CD10 has been shown by some authors to be a highly specific marker in recognizing endometrial stromal cells [3], [5], [9], [10], [11].

In recent years, numerous investigations have concentrated on CD44 isoform expression in various malignant tumors, since CD44 has been linked to tumor progression and metastasis [12], [13]. CD44 is encoded by at least 20 exons and alternative splicing of the CD44 gene creates multiple CD44 variant isoforms (CD44v) which are present in many cells throughout the body [14]. It has been reported recently that no CD44v3 expression was detected in leiomyosarcomas, whereas myometrium and leiomyomas expressed CD44v3. So Poncelet et al. [15] suggested that the loss of CD44v3 expression could be used as a putative diagnostic tool for uterine leiomyosarcomas. As far as we are aware, no data could be available about the expression of CD44v3 in ESS.

Mast cells are important effector cells, providing granule and membrane mediators as well as cytokines in allergic and inflammatory diseases. Also, they are commonly seen in a variety of malignant and benign human neoplasms and many have clinical significance in such cases [16], [17], [18], [19]. It has been proven that the intratumoral mast cell (MC) counts in uterine leiomyosarcomas were significantly lower than in atypical and cellular leiomyomas and it could be used as a criterion in their differential diagnosis [20], [35]. But MCs in ESS have not been investigated previously.

The purpose of this study was to evaluate the potential utility of these and selected other antibodies applied as a panel in the differential diagnosis of ESS and CL.

Section snippets

Case selection

The files of the Department of Pathology, Women's Hospital of Medical College of Zhejiang University were searched from 1995 to 2002 for the diagnosis of uterine ESS and CL. Twenty-six cases of ESS (21 low-grade ESS and 5 high-grade ESS) and 25 CL (17 common CL, 8 HCL) of uterus were studied. In addition, 25 examples of normal myometrium and 25 examples of non-neoplastic endometrial stroma were examined. The tissues were used after obtaining written consent from the patients.

Hematoxylin and

Results

The immunohistochemical results are summarized in Table 2 and selected illustrations are provided Fig. 1, Fig. 2, Fig. 3, Fig. 4.

Conclusion

Differentiating ESS from CL generally is not difficult given the distinctive histologic features of each tumor [1], [2]. However, ESSs have been shown previously to have overlapping features with smooth muscle tumors by histology, immunohistochemistry, and electron microscopy [3], [4], [5], [8], [9]. Because it is thought that both endometrial stroma and myometrium derive from the Müllerian duct embryonically, it is not surprising, therefore, that there is considerable histologic and

References (35)

  • M.R. Nucci et al.

    h-Caldesmon expression effectively distinguishes endometrial stromal tumors from uterine smooth muscle tumors

    Am. J. Surg. Pathol.

    (2001)
  • E. Oliva et al.

    An immunohistochemical analysis of endometrial stromal and smooth muscle tumors of the uterus—A study of 54 cases emphasizing the importance of using a panel because of overlap in immunoreactivity for individual antibodies

    Am. J. Surg. Pathol.

    (2002)
  • M.M. Miettinen et al.

    Calponin and h-caldesmon in soft tissue tumors: consistent h-caldesmon immunoreactivity in gastrointestinal stromal tumors indicates traits of smooth muscle differentiation

    Mod. Pathol.

    (1999)
  • D.S. Rush et al.

    h-caldesmon, a novel smooth muscle-specific antibody, distinguishes between cellular leiomyoma and endometrial stromal sarcoma

    Am. J. Surg. Pathol.

    (2001)
  • W.G. McCluggage et al.

    CD10 is a sensitive and diagnostically useful immunohistochemical marker of normal endometrial stroma and of endometrial stromal neoplasms

    Histopathology

    (2001)
  • T. Toki et al.

    CD10 is a marker for normal and neoplastic endometrial stromal cells

    Int. J. Gynecol. Pathol.

    (2002)
  • M.A. Noordzij et al.

    The prognostic value of CD44 isoforms in prostate cancer patients treated by radical prostatectomy

    Clin. Cancer Res.

    (1997)
  • Cited by (0)

    View full text