Differential diagnosis of malignant tumours in the abdominal cavity of rats after intraperitoneal injection of crocidolite or benzo[a]pyrene.

In our investigation (i.p. test), crocidolite and benzo[a]pyrene, both caused a progression from initially reactive, then autonomously transformed proliferation of myofibroblasts and undifferentiated mesenchymal cells to malignant, multidirectionally differentiated (desmin and ED-1 positive) fibro-histiocytic tumours. Immunohistochemically these tumours showed no morphological characteristics (for example co-expression of vimentin and keratin in spindle-shaped tumour cells) of human asbestos-associated malignant mesotheliomas. On the other hand many tumour cells induced by crocidolite and benzo[a]pyrene had an ultrastructural appearance resembling fibroblasts and myofibroblasts. These have been demonstrated in only a few desmoplastic and sarcomatous mesotheliomas in human beings. None of the tumours revealed the typical ultrastructural features of epitheloid or transitional mesotheliomas. Apparently, both carcinogenic substances induce the transformation of undifferentiated pluripotent mesenchymal cells in rat peritoneum, regardless of their localization in the submesothelial compartment or perivascular connective tissue (preferentially after crocidolite application) or in the connective tissue pseudocapsule of major benzo[a]pyrene containing beeswax/tricaprylin depots in the mesometrium and mesenterial fatty tissue. In this way asbestos fibres in this animal experiment do not seem to induce an arrest in differentiation of intermediate or immature mesothelial cells as supposed formerly, but rather affect undifferentiated mesenchyme cells and myofibroblasts. This is an explanation for the immunohistochemical expression of markers of muscular differentiation in these tumour cells, which is known to occur in human malignant fibro-histiocytic tumours. If supplementary immunohistochemical investigations with different keratin antibodies also fail to confirm the mesothelial differentiation of the tumours induced in our i.p. test, the decision to call them "mesotheliomas" should be reconsidered. Further immuno-transmission-electron microscopical investigations with intermediate filament or macrophage antibodies are needed to clarify whether the term malignant "fibrohistiocytic sarcoma", "mesenchymoma" or "mesothelioblastoma" would be more correct from the morphological point of view.