Mitotic count in seminomas--an unreliable criterion for distinguishing between classical and anaplastic types.

Testicular seminomas occur in various forms of which the classical, the spermatocytic, and that with syncytiotrophoblastic giant cells are distinctly defined. Anaplastic seminomas, however, are less clearly distinguished. Forty-five seminomas, 39 pure and 6 combined tumors, were examined from the perspective of the current definition that 3 or more mitoses per high power field (m/hpf) distinguish the anaplastic from the classical form. Our resultant yield of over 80% of "anaplastic" seminomas is clearly incompatible with general clinical experience, indicating that the arbitrary criterion of 3 m/hpf does not segregate anaplastic forms from neoplasms that are mitotically active but relatively non-aggressive. Moreover, the evidence indicates that mitotic activity does not adequately define a tumor form that is near the undifferentiated end of the spectrum which extends from "embryonal carcinoma" to the spermatocytic type. If it should prove, however, that the mitotic rate must be used as an arbitrary watershed criterion until a more reliable one is found, it should then be set at more than 5 m/hpf, yielding a percentage of anaplastic seminomas (about 9%) that is compatible with clinical experience.