Alexander Marks

Monoclonal Antibody D2-40, a New Marker of Lymphatic Endothelium, Reacts with Kaposi's Sarcoma and a Subset of Angiosarcomas

There is controversy over the histogenesis of Kaposis sarcoma (KS) from lymphatic or blood vessel endothelium. D2-40 is a novel monoclonal antibody to an Mr 40,000 O-linked sialoglycoprotein that reacts with a fixation-resistant epitope on lymphatic endothelium. We sought to establish the selectivity of D2-40 for lymphatic endothelium in normal tissues and compare its reactivity with the expression of the widely used vascular endothelial marker CD31 in a series of 62 formalin-fixed and paraffin-embedded vascular lesions including KS. In normal tissues, D2-40 stained the endothelium of lymphatic channels but not of blood vessels, including arteries and capillaries defined by reactivity with the blood vessel endothelial marker PAL-E. In our series of vascular lesions, D2-40 stained lymphangiomas (10/10), benign tumors of undisputed lymphatic origin, but not benign neoplasms or tumorlike lesions of blood vessel origin, including hemangiomas (0/10), glomus tumors (0/3), angiolipomas (0/2), pyogenic granulomas (0/2), vascular malformations (0/2), hemangiopericytoma (0/1), or hemangioendothelioma (0/1). D2-40 stained all cases of cutaneous KS (24/24) at all stages of progression, including patch, plaque, and nodular stages, supporting the concept that this disease originates from a cell type capable of undergoing lymphatic differentiation. D2-40 also stained three of seven angiosarcomas, indicating that a subset of these tumors can undergo at least partial differentiation along the lymphatic endothelial lineage and could be classified as lymphangiosarcomas. In comparison, CD31 was expressed in all benign and malignant vascular lesions, except for glomus tumors (0/3) and 5/10 lymphangiomas, in which staining was absent. We conclude that D2-40 is a new selective marker of lymphatic endothelium in normal tissues and vascular lesions and is valuable for studying benign and malignant vascular disorders in routinely processed tissue specimens.

A new monoclonal antibody, D2-40, for detection of lymphatic invasion in primary tumors.

L ymphatic invasion is a commonly reported histopathological finding in the assessment of primary tumors. In some studies, it has been found to be a predictor of shorter disease-free or overall survival in melanoma and cancers of breast, colon, and cervix (Birner et al, 2001; Borgstein et al, 1999; Clemente et al, 1992; Compton, 1999; Lauria et al, 1995). In the absence of effective immunohistochemical markers of lymphatic endothelium in paraffin sections, lymphatic invasion is currently identified on the basis of conventional hematoxylin and eosin (H&E) staining as the presence of tumor emboli within vascular channels distinctly lined by a single layer of endothelial cells. Pitfalls in the technique arise mainly from the difficulty in visualizing the lymphatic vessel wall following H&E staining. They include the inability to discern tumor emboli that obliterate the lumen of lymphatics and to distinguish retraction artifacts that isolate tumor aggregates due to tissue shrinkage during fixation from true tumor emboli in lymphovascular spaces. We presently report that monoclonal antibody (mAb) D2– 40, a new selective marker of lymphatic endothelium (Kahn et al, 2002), clearly demarcates tumor emboli in lymphatics in paraffin sections of primary tumors, including cancers of breast, colon, prostate, cervix, endometrium, and skin (melanoma and squamous cell carcinoma). In a pilot study of 50 breast cancer cases, D2–40 identified lymphatic invasion in 44% lymph node-negative and 88% lymph node-positive cases. In comparison, we found that 18% of the specimens examined by H&E staining were false-negative and 4% were false-positive for detection of lymphatic invasion. This new mAb has the potential for increasing the accuracy of detection of lymphatic invasion in primary tumors and, therefore, could be widely applied for this purpose in surgical pathology. MAb D2–40 (IgG1) to a Mr 40,000 0-linked sialoglycoprotein was purified from ascitic fluid, as previously described (Marks et al, 1999), and stored at a concentration of 0.7 mg/ml under sterile conditions at 4° C. Paraffin blocks of primary cancers of breast, colon, prostate, cervix, endometrium, and skin (melanoma and squamous cell carcinoma) were obtained from the Department of Pathology, Sunnybrook and Women’s College Health Sciences Centre, University of Toronto. Sections of paraffin-embedded tissues (5 m) were first incubated in methanol containing 3% H 2O 2 to inactivate endogenous peroxidase. For immunostaining, the sections were incubated with mAb D2–40 (0.1 ug/ml), followed sequentially by biotinylated goat antimouse immunoglobulin antibody (Zymed, San Francisco, California) at a 1:200 dilution and a horseradish peroxidase-avidin conjugate (Dako, Carpinteria, California) at a 1:500 dilution. For color development, the sections were incubated with 3:3 diaminobenzidine. Examples of the application of D2–40 for the accurate detection of lymphatic invasion are shown in Figure 1. In Figure 1A, a tumor embolus that obliterates the lumen of a lymphatic in breast cancer is clearly outlined by positive immunostaining of the vessel wall with D2–40 (arrowhead). In Figure 1B, an adjacent H&E section shows the difficulty in discerning a tumor embolus. In Figure 1C, a retraction artifact that was initially spuriously diagnosed as lymphatic invasion on an H&E section in breast cancer was subsequently correctly identified as a retraction artifact on the basis of negative immunostaining with D2–40 (open arrowhead). Immunostaining was also negative for CD-31, CD-34 and factor VIII-related antigen (data not shown) ruling out a blood vessel tumor embolus. Neighboring lymphatics stain positively with D2–40 (arrows). In Figure 1D, positive immunostaining with D2–40 outlines multiple lymphatics containing tumor emboli that obliterate the lumen (arrowheads) and uninvolved lymphatics containing lymphocytes (arrows) in colon cancer. In Figure 1E, positive immunostaining with DOI: 10.1097/01.LAB.0000028824.03032.AB

Ascitic fluid from human ovarian cancer patients contains growth factors necessary for intraperitoneal growth of human ovarian adenocarcinoma cells.

Human ovarian cancer, the leading cause of death from gynecologic malignancy, tends to remain localized to the peritoneal cavity until late in the disease. In established disease, ascitic fluid accumulates in the peritoneal cavity. We have previously demonstrated that this ascitic fluid is a potent source of in vitro mitogenic activity including at least one unique growth factor. We now report that the human ovarian adenocarcinoma line, HEY, can be induced to grow intraperitoneally in immunodeficient nude mice in the presence (23/28 mice), but not absence (0/21 mice) of ascitic fluid from ovarian cancer patients. Ascitic fluid from patients with benign disease did not have similar effects on intraperitoneal growth of HEY cells (1/15 mice). Once tumors were established by injections of exogenous ascitic fluid, they could progress in the absence of additional injections of ascitic fluid. The mice eventually developed ascitic fluid which contained potent growth factor activity, suggesting that the tumors eventually produced autologous growth factors. This nude mouse model provides a system to study the action of ovarian cancer growth factors on tumor growth in vivo and to evaluate preclinically, therapeutic approaches designed to counteract the activity of these growth factors.

Fluorescent Proteins Expressed in Mouse Transgenic Lines Mark Subsets of Glia, Neurons, Macrophages, and Dendritic Cells for Vital Examination

To enable vital observation of glia at the neuromuscular junction, transgenic mice were generated that express proteins of the green fluorescent protein family under control of transcriptional regulatory sequences of the human S100B gene. Terminal Schwann cells were imaged repetitively in living animals of one of the transgenic lines to show that, except for extension and retraction of short processes, the glial coverings of the adult neuromuscular synapse are stable. In other lines, subsets of Schwann cells were labeled. The distribution of label suggests that Schwann cells at individual synapses are clonally related, a finding with implications for how these cells might be sorted during postnatal development. Other labeling patterns, some present in unique lines, included astrocytes, microglia, and subsets of cerebellar Bergmann glia, spinal motor neurons, macrophages, and dendritic cells. We show that lines with labeled macrophages can be used to follow the accumulation of these cells at sites of injury.

Spatial and temporal expression of S100B in cells of oligodendrocyte lineage.

The analysis of oligodendrocyte (OL) lineage development has been facilitated by the immunocytochemical characterization of OL‐specific antigens and definition of the phenotypes sequentially acquired by differentiating OLs. The purpose of the present study was to address an enduring discrepancy between several reported cases of S100B immunodetection in CNS myelin and myelinating OLs on the one hand, and the systematic use of the S100B protein as an alleged astrocytic marker in studies of the mammalian CNS on the other. To resolve this discrepancy, we have compared the developmental distribution of EGFP+ cells in the CNS of s100b‐enhanced green fluorescent protein (EGFP) (Vives et al., 2003) and cnp‐EGFP (Yuan et al., 2002) mice, and examined the degree of overlap between EGFP expression and that of stage‐specific markers of OL differentiation during the embryonic and postnatal phases of development. We demonstrate that the S100B protein is expressed in postnatal and adult populations of NG2+ progenitors of mouse brain, as well as in immature and mature myelinating OLs present in the brain and spinal cord of embryonic and adult mice, respectively. Comparison between EGFP and endogenous S100B expression in the s100b‐EGFP and cnp‐EGFP mice indicates that S100B protein expression is upregulated in immature and mature OLs. These results argue against the current view that S100B expression is restricted to the astrocytic lineage in the CNS, and indicate that the use of S100B in combination with other molecular markers will help discriminate oligodendrocytes from astrocytes.

Nuclear expression of S100B in oligodendrocyte progenitor cells correlates with differentiation toward the oligodendroglial lineage and modulates oligodendrocytes maturation

The S100B protein belongs to the S100 family of EF-hand calcium binding proteins implicated in cell growth and differentiation. Here, we show that in the developing and the adult mouse brain, S100B is expressed in oligodendroglial progenitor cells (OPC) committed to differentiate into the oligodendrocyte (OL) lineage. Nuclear S100B accumulation in OPC correlates with the transition from the fast dividing multipotent stage to the morphological differentiated, slow proliferating, pro-OL differentiation stage. In the adult, S100B expression is down-regulated in mature OLs that have established contacts with their axonal targets, suggesting a nuclear S100B function during oligodendroglial cells maturation. In vitro, the morphological transformation and maturation of pro-OL cells are delayed in the absence of S100B. Moreover, mice lacking S100B show an apparent delay in OPC maturation in response to demyelinating insult. We propose that nuclear S100B participates in the regulation of oligodendroglial cell maturation.

Placental-like alkaline phosphatase as a marker of carcinoma-in-situ of the testis. Comparison with monoclonal antibodies M2A and 43–9F

In an immunohistochemical study of 59 routinely processed tissue specimens from 48 adult testes with isolated carcinoma‐in‐situ (CIS) changes and of 66 specimens from adult testes without neoplasia, placental‐like alkaline phosphatase (PIAP) was shown to be a reliable marker of CIS cells preceding the development of a testicular tumour. Thus, a positive reaction was encountered in all 36 biopsies treated with formaldehyde, or Bouins or Stieves fluid. However, only 11 of 23 specimens fixed with Clelands fluid were immunoreactive for PIAP. None of the non‐malignant components of seminiferous tubules, including the large abnormal spermatogonia, reacted with the antibody against PIAP. Besides the antibody against PIAP, monoclonal antibodies M2A and 43–9F were tested on CIS specimens fixed with the above‐mentioned fixatives. In the 17 specimens fixed with Stieves or Bouins fluid, a positive reaction was obtained in all sections with all three antibodies tested. However, for each antibody at least two specimens gave a weak staining reaction. When all three immunostainings were performed, in each case at least one of them gave a moderate or strong reaction, thus making CIS cells easily detectable. In the samples fixed with Clelands fluid, a negative reaction was found in one to three specimens, depending on the antibody used. However, at least one of the three antibodies gave a positive reaction if all three immunostainings were applied. In only one of the formaldehyde‐fixed paraffin specimens did CIS cells react with the monoclonal antibody 43–9F, whereas M2A gave no positive reaction at all if this method of fixation was used. Thus, the sensitivity of the immunohistochemical staining procedure in the detection of CIS is dependent on the fixative used and increases when immunostainings with all three markers are performed simultaneously.

S‐100 protein in tumors of cartilage and bone: An immunohistochemical study

Using the peroxidase‐antiperoxidase (PAP) immunohistochemical method, S‐100 protein was found in well‐differentiated chondrocytes of chondroma, chondroblastoma, mesenchymal chondrosarcoma, and osteosarcoma. It was not detected in osteoma, osteoblastoma, giant cell tumor, and Ewings tumor. The presence of S‐100 protein in tumorous chondrocytes and chondroblasts suggests that this protein may be a marker of chondrocyte origin and should not be considered a specific marker for nerve tissue.

Significance of lymph vessel invasion identified by the endothelial lymphatic marker D2-40 in node negative breast cancer

Monoclonal antibody D2-40, a marker of lymphatic endothelium, identifies tumor emboli in lymph vessels. The aim of the study was to assess whether D2-40+ lymph vessel invasion (LVI) correlates with clinicopathologic factors including lymphovascular invasion (LVI) as assessed by haematoxylin and eosin-stained sections (H&E+ or H&E−) and to assess the prognostic significance in node-negative breast cancer. The study group consisted of 303 node-negative breast cancer patients that had a median follow-up of 7.6 years. Clinical and pathological data were retrieved from the Henrietta Banting database. Immunohistochemical staining was performed on formalin-fixed, paraffin-embedded tissue sections of the primary invasive carcinoma using D2-40. Immunostaining with CD31 was performed on the discordant cases that were H&E+/D2-40−. D2-40+ lymph vessel invasion was detected in 82/303 (27%) cases. The foci of lymphatic invasion occurred predominantly at the invasive front of the tumor. The absence of D2-40 and CD31 in 13/17 discordant cases was suggestive of retraction artefact. D2-40+ lymph vessel invasion correlated significantly with age (P=0.0003), tumor size (P=0.005), histological grade (P=0.0001), H&E+ (P=<0.0001) and estrogen receptor status (P=0.005) but not with histological type or progesterone receptor status. Multivariate analysis revealed that D2-40+ lymph vessel invasion was the only significant predictor of distant recurrence. There was no significant association between D2-40 status and local recurrence (P=0.752) or regional recurrence (P=0.13). Both D2-40+lymph vessel invasion (P=0.009) and H&E+LVI cases (P=0.02) were associated with overall shorter survival in univariate analysis. These data indicate that D2-40 identifies lymphatic invasion in breast tumors and is a significant predictor of outcome in breast cancer.

Origin of the desmoplasia in desmoplastic malignant melanoma

Four cases of desmoplastic malignant melanoma were examined light microscopically and immunohistochemically. Electron microscopy was performed in three cases. Light microscopy showed that all tumors were composed of neoplastic spindle cells that infiltrated between mature collagen bundles in the reticular dermis. Some of the spindle cells had bizarre nuclei, whereas other spindle cells resembled normal fibroblasts. Melanin could not be demonstrated in any of the tumors by histochemical techniques. Electron microscopic examination of the spindle cells showed prominence of rough endoplasmic reticulum, which was dilated and filled with flocculent material and occasional collagen fibrils. The same cells contained aggregates of non-membrane-bound melanin granules and pre-melanosomes. Some cells also showed features of myofibroblasts. Immunoperoxidase staining with anti-S100 protein antibody demonstrated positivity of the spindle cells as well as of melanocytes in the basal layer of the epidermis. Scar tissue and fibroblasts did not stain. These findings show that the desmoplastic component of these malignant melanomas derives from melanocytes that have undergone adaptive fibroplasia. Therefore, in assessing depth of invasion in a malignant melanoma, measurements should include the desmoplastic areas.