Testicular tumors--some new and a few perennial problems.

Testicular tumors--some new and a few perennial problems.

Abstract:	* The histopathology of testicular tumors is presented, emphasizing new, unusual, or underemphasized aspects. Within the category of seminoma of the usual type, the recent literature has drawn attention to the presence in occasional tumors of solid or hollow tubules or spaces of varying sizes and shape that may result in cribriform or microcystic patterns, causing potential confusion with other neoplasms, most notably Sertoli cell tumor or yolk sac tumor. Although regions of typical neoplasia and awareness of this phenomenon usually will be diagnostic, immunohistochemistry may play a role in excluding Sertoli cell tumor or yolk sac tumor. Although immunohistochemistry can play an undoubted helpful role in this and selected other areas of testicular tumor evaluation, careful evaluation of the gross and routine microscopic features will solve the vast majority of diagnostic problems. An excellent review of immunohistochemistry in this area by R. E. Emerson, MD, and T. M. Ulbright, MD, is cited herein. Spermatocytic seminoma remains a crucial pitfall in diagnosis, and the pathologist must always be alert to the possible diagnosis when looking at a seminomatous neoplasm, particularly in an older patient, although about one third of these tumors occur in the usual seminoma age range. The embryonal carcinoma has a great diversity of patterns, which are briefly noted. The enigmatic and picturesque tumor, polyembryoma, which virtually never occurs in pure form but may be a confusing component of a variety of mixed germ cell tumors, is discussed and illustrated. The phenomenon of burnt-out germ cell neoplasia is also briefly noted and an excellent recent contribution is referred to. Within the sex cord-stromal family of neoplasms, recent contributions and elaborations of unusual morphologic features of Leydig cell tumors and Sertoli cell tumors are presented. Within the Leydig cell family, cyst formation, adipose metaplasia, calcification or ossification, and spindle cell patterns may be particularly confusing, and in the Sertoli cell family, a great array of patterns caused by differing admixtures of tubular, solid, and stromal components occur. The peculiar lesion, intratubular large cell hyalinizing Sertoli cell tumor, of young boys with Peutz-Jeghers syndrome, is briefly discussed. Some of the problems in the family of hematopoietic neoplasms are reviewed, these processes posing diverse problems in differential diagnosis and their correct recognition having crucial therapeutic implications. Although secondary tumors to the testis have not received the same attention in the literature as the similar phenomenon in the female gonad, remarkable examples of testicular spread of diverse neoplasms, usually carcinoma but rarely melanoma, are seen, and the pathologist should be alert to this possibility, particularly when examining an unusual morphology in an older patient. Finally, a few comments are made on the common paratesticular neoplasm, the adenomatoid tumor, highlighting its varied patterns and recent description of some of the issues that may arise when they undergo total or subtotal infarction. (Arch Pathol Lab Med. 2008;132:548-564)

Subject:	Tumors (Diagnosis) Melanoma (Diagnosis) Cancer (Care and treatment) Immunohistochemistry
Author:	Young, Robert H.
Pub Date:	04/01/2008
Publication:	Name: Archives of Pathology & Laboratory Medicine Publisher: College of American Pathologists Audience: Academic; Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2008 College of American Pathologists ISSN: 1543-2165
Issue:	Date: April, 2008 Source Volume: 132 Source Issue: 4
Topic:	Canadian Subject Form: Tumours
Product:	Product Code: 8000432 Cancer Therapy NAICS Code: 621 Ambulatory Health Care Services
Accession Number:	230151949
Full Text:	In this review, I discuss the pathologic diagnosis of tumors of the testis, emphasizing newly appreciated issues as well as some that are long known but still occasionally a problem. One common tumor of the adnexa, the adenomatoid tumor, also is briefly discussed. References will be selective, a standard text having a comprehensive listing. (1) Illustrations must be similarly selective because of space constraints and will tend toward the unusual, as most readers will be very familiar with common patterns of the typical neoplasms, which additionally have been comprehensively depicted by Thomas M. Ulbright, MD, and colleagues in the testis fascicle. (1) SEMINOMA (USUAL TYPE) The features of the usual seminoma are well known, but a few points merit being emphasized. First of all, from the gross perspective, although they are typically creamy white to yellow neoplasms, a minority can have a beefy red sectioned surface, and foci of necrosis are not rare. On microscopic examination, the almost invariable framework of delicate fibrous septa with associated blood vessels and a sprinkling of lymphocytes should ideally be present to diagnose seminoma with confidence, as other tumors that may mimic seminoma usually lack this feature. This architecture imparts the classic alveolar pattern of seminoma (Figure 1). However, as we shall see later when Sertoli cell tumors are discussed, the typical low power of seminoma may be seen rarely with other tumors, so high-power scrutiny to make sure the typical cytologic features of seminoma cells are present is crucial. In an occasional seminoma, the septa are inconspicuous; the diagnosis of seminoma should be made cautiously in such cases and consideration given to other entities, if only to exclude them by careful conventional morphologic study and, if indicated, immunohistochemical evaluation. In some seminomas, the fibrous septa are exaggerated such that broad fibrous bands are present. In these and some other cases, discrete nests and cords are conspicuous. Although seminomas usually obliterate the underlying parenchyma, they may have a conspicuous interstitial growth pattern with tubular preservation, particularly peripherally, and rarely the cells are so widely dispersed in the interstitium that a tumor is grossly inapparent. (2) A granulomatous infiltrate is much more common in seminoma than any tumor ever in the differential with it, but embryonal carcinoma also rarely has such an infiltrate. About 5% of seminomas contain varying numbers of syncytiotrophoblast cells, which may secrete chorionic gonadotropin. There is no evidence that these tumors have a prognosis worse than the usual seminoma. They may be associated with striking foci of hemorrhage and by definition lack an association with cytotrophoblast; otherwise, the focus in question would be choriocarcinoma. It is more common to see syncytiotrophoblast in a germ cell tumor unassociated, as opposed to associated, with cytotrophoblast, in part because syncytiotrophoblast is so common in cases of embryonal carcinoma. [FIGURE 1 OMITTED] [FIGURE 2 OMITTED] [FIGURE 3 OMITTED] [FIGURE 4 OMITTED] [FIGURE 5 OMITTED] One aspect of the morphology of seminomas of otherwise typical type that may be confusing is the presence in some cases of solid or hollow tubules or other spaces of various types that may result in patterns that range from cribriform to microcystic (Figure 2, A). (3) The differential ranges potentially from Sertoli cell tumor to yolk sac tumor to even spermatocytic seminoma or embryonal carcinoma. The lesional cells in these patterns, however, have the typical features of usual seminoma cells, which is crucial in the differential diagnosis. The typical immunohistochemical staining of seminoma cells for placental alkaline phosphatase and Sertoli cells for inhibin and other markers (4) will be helpful in the Sertoli cell tumor differential if problems persist after routine evaluation. Mistaking a malignant Sertoli cell tumor for a seminoma also occurs; this is discussed in the section on Sertoli cell tumors. Staining for OCT-4 (positive in seminoma but not in yolk sac tumor) may help in the differential with a yolk sac tumor. This differential was the most problematic one in a recently reported series of 28 seminomas with spaces of various types. (3) The spaces in that series ranged from small, closely packed, and relatively regular to dilated, more dispersed, and somewhat irregular. Thirteen tumors either lacked or had scant lymphocytes, suggesting that perhaps this may have had something to do with the morphology, although that is speculative. It is, however, worth remembering that in this unusual variant of seminoma the typical lymphocytic infiltrate may not be as conspicuous as usual. These tumors may have edema, and that finding and the paucity of lymphocytes may suggest the erroneous diagnosis of spermatocytic seminoma, but the varied cell types of that neoplasm are absent. A microcystic pattern of seminoma is unassociated with other features, such as hyaline bodies, that would make yolk sac tumor a more realistic consideration, and seminomas with tubules (that may be confused with glands of embryonal carcinoma) do not have as primitive nuclei, as seen in embryonal carcinoma. In occasional seminomas with syncytiotrophoblast giant cells, prominent intracytoplasmic lacunae may form small cysts, which occasionally may be striking on low power, (5) another explanation for spaces in seminoma. An even rarer aspect of otherwise typical seminomas than that just considered is the presence in rare cases that we have seen of signet ring cells (54) (Figure 2, B). In one case, these were particularly conspicuous. Occasional seminomas undergo extensive hemorrhagic infarction, and indeed, this is occasionally massive, such that no nonnecrotic tissue is left. This occasionally results in the misdiagnosis of something innocuous from the prognostic viewpoint, such as a hemorrhagic infarct. In contrast to that nonneoplastic entity, in cases of necrotic seminoma, one can still almost invariably see the outlines of the seminoma cells, and the lack of glandular patterns of embryonal carcinoma and preservation of the septa of the seminoma, albeit in ghost outline, and scattered still recognizable lymphocytes, can result in the confident diagnosis of necrotic seminoma (Figure 3). The cytologic features of seminoma, both cytoplasmic and nuclear, are well known (1) and will not be repeated here. It is important, however, to emphasize that although clear cell cytoplasm is the norm, a subset of tumors have appreciable eosinophilic, sometimes plasmacytoid-appearing, cytoplasm (Figure 4). Whether the general category of the usual type of seminoma (meaning seminomas other than the spermatocytic subtype) have within them a morphologically recognizable set of cases that can be reliably observed to have features indicative of a poorer prognosis has long been controversial. The entity of anaplastic seminoma, based on brisk mitotic rate, was popularized some years ago but never gained widespread acceptance and was not advocated in the 1999 Armed Forces Institute of Pathology (AFIP) fascicle. (1) Those tumors appeared to present at a higher stage than typical seminomas, but stage-for-stage there was no evidence for a worse prognosis. A relatively recent article (6) has revisited this general issue. In that study, 105 cases of testicular seminoma treated at one institution were investigated to discern whether there is any relation between morphology, immunohistochemical features, and stage. Those authors identified a group of cases they diagnosed as seminomas with atypia based on moderate to marked nuclear pleomorphism, nuclear overlapping, lack or paucity of a lymphocytic infiltrate, and lack or paucity of cytoplasmic clarity. Tumors with more than 50% of the tissue characterized by such features were classified as seminomas with atypia. Cytologic atypia in areas with granulomatous inflammation or fibrosis was disregarded. The authors of that study found that their seminoma with atypia presented at a higher stage than tumors without atypia. From the immunohistochemical viewpoint, the authors found that 43% of the seminomas with atypia were C-kit protein negative compared with only 5% of conventional seminomas. This was a statistically significant finding. There was some difference, but not a statistically significant difference, between Ki-67 positivity in the two groups, with almost 30% of seminomas with atypia showing more than 50% nuclear positivity compared with only 10% of conventional seminomas. These results are of obvious interest, but I have concerns about the reproducibility of the conventional morphologic observations, given the vagaries of fixation and staining that can alter the appearance of seminomas and make some of them look a little more "atypical" than others. However, if others independently confirm the findings in the article just summarized, it may result in identification of a clinically significant subset of tumors, but it is too early at this time to define a specific new subset, and even the authors of the article referred to do not recommend that. In the differential diagnosis of seminoma with embryonal carcinoma, and for that matter certain other issues, immunohistochemistry may play a helpful role. An excellent discussion of this was recently contributed by R. E. Emerson, MD, and T. M. Ulbright, MD, (4) and is strongly recommended to the reader. Through the courtesy of Drs Emerson and Ulbright, tables from that essay are presented here (Table). With regard to the specific issue of seminoma and embryonal carcinoma, I will just briefly note that staining of seminoma for C-kit and negative staining for CD30, and the converse results in embryonal carcinoma are perhaps the most helpful pair of staining results. SPERMATOCYTIC SEMINOMA The spermatocytic seminoma has little or no relation to the typical seminoma, except that it is another type of germ cell tumor that almost always lacks the ability to differentiate into other tissue. (7,8) These tumors are composed mainly of cells resembling spermatogonia, with focal differentiation into larger cells that suggest primary spermatocytes in routine sections. DNA studies, however, have not revealed a haploid number of chromosomes in their nuclei, raising a question about their spermatocytic nature. This rare tumor has a proud lineage from the perspective of those who "put it on the map," described in French by Pierre Masson, MD, (9) and the first article in English by Robert E. Scully, MD, (10) and then further observations, including one on its ultrastructure by Juan Rosai, MD, and colleagues. (11,12) In a classic example of diligence, Dr Rosai was able to have some purported malignant cases that had been described by others reviewed and determined to be cases of malignant lymphoma. (11) Only one case unassociated with sarcomatous transformation (see below) has spread beyond the testis. Although one of the triad of tumors to be particularly considered in the older man (lymphoma and metastasis being the other two), about one third of spermatocytic seminomas occur in the usual seminoma age range, an important point to remember. On gross examination the neoplasms are often large and typically pale and edematous or gelatinous, sometimes with cyst formation. They are more often bilateral than the usual seminoma but are still infrequently bilateral. Rarely, the clinical scenario of a long-known testicular mass with sudden rapid enlargement is seen and correlates with the enigmatic phenomenon of sarcomatous transformation. (13,14) Microscopic examination shows features that differ from those of the typical seminoma in a number of regards. The tumor is never associated with another germ cell component, only rarely has lymphocytes or a granulomatous inflammation, is glycogen and placental alkaline phosphatase negative, and never has syncytiotrophoblast cells. Although there is often a diffuse growth, there is in many cases an edematous fluid in the background that can cause irregular aggregates of tumor cells (Figure 5). The septa of typical seminoma are usually absent. The more varied trilineage cell population is the crucial cellular distinction from usual seminoma. The most common cell type is about the size of the typical seminoma cell but has denser cytoplasm and more uniform, rounder nuclei. Smaller, degenerating cells with dense, homogeneous nuclei are usually present as well and represent the second cell type; it is important not to confuse these cells with lymphocytes. The third form of the cell is the largest; they suggest primary spermatocytes, usually occur singly, and contain very large nuclei with chromatin that may have a filamentous quality. The entity of "anaplastic" spermatocytic seminoma has recently been described. (15) We agree that the morphology in these cases differs from that of typical spermatocytic seminoma but think use of the word "anaplastic" not indicated, as there is no evidence that these tumors have a clinical behavior that one might expect for tumors that are "anaplastic." Additionally, those reported in this category have had foci of conventional neoplasia with the well-known three cell types noted above. Seven cases of sarcomatous transformation of spermatocytic seminoma have been reported. (13,14) The sarcomatous component is usually nonspecific but may have the features of rhabdomyosarcoma. This makes a tumor with a good prognosis change to one with a poor prognosis. EMBRYONAL CARCINOMA These tumors are usually smaller and have a more granular, tan-sectioned surface than the seminoma. The classic microscopic patterns are solid, glandular (Figure 6A), or papillary. The tumor cells are larger than those seen in both the seminoma and yolk sac tumor, with more hyperchromatic, irregular nuclei. Syncytiotrophoblast cells are often scattered among the embryonal carcinoma cells. The features of this tumor, both architectural and cytologic, are somewhat more varied than we and others have sometimes implied on the basis of descriptions and illustrations provided in the literature. For that reason, a panel of 4 illustrations is provided (Figure 6, A through D) to convey some of the spectrum. Among other things, the stroma is sometimes prominent and can range from relatively acellular and sometimes slightly hyaline in quality to more typically densely cellular. The presence of cellular mesenchyme on its own does not warrant a designation of teratoma, in my opinion. The glands, although usually lined by exceedingly primitive-appearing cells, sometimes appear more differentiated, and rarely on evaluation of an individual gland one may have some uncertainty as to whether the gland is a gland of embryonal carcinoma or of the frequent associated teratomatous component. This is, thankfully, a distinction that is almost never, if ever, of any consequence, but it is of some slight academic interest. Although glands may be seen in yolk sac tumor, primitive glands are more typical of embryonal carcinoma, and my bias is to that diagnosis in problematic causes, but this is another academic distinction. Although the cytoplasm of embryonal carcinoma cells is usually basophilic, occasionally it is clear, which rarely causes some confusion with seminoma which, of course, more typically has cells with clear cytoplasm. [FIGURE 6 OMITTED] Although poorly fixed seminomas may present a problem in differential diagnosis, the arrangement of cells one to another in embryonal carcinoma is much more irregular than in the seminoma. The so-called applique pattern of embryonal carcinoma1 should not lead to confusion with choriocarcinoma. That pattern results when cells that are degenerative at the periphery of better-preserved cells superficially suggest the biphasic pattern of choriocarcinoma. Helpful immunohistochemical results relevant to embryonal carcinoma are present in the Table and the CD30-positive, C-kit-negative profile noted earlier. YOLK SAC TUMOR The pure yolk sac tumor is almost always encountered in infants and young children but is rare in the first 6 months of life, something important in the differential with the juvenile granulosa cell tumor (see below). Rarely, the yolk sac tumor occurs as a pure tumor in the adult, but it is more often a component of a mixed germ cell tumor, being seen in about 40% of such cases. Gross examination typically reveals a mass that is predominantly solid and is soft, white, gray, or pale yellow; cystic degeneration as well as necrosis and hemorrhage are often present. Tumors without hemorrhagic or necrosis may mimic seminoma, and a seminoma-like gross appearance in someone younger than 15 years (when seminoma is rare) should suggest yolk sac tumor, although there are, of course, other options on the gross alone. The tumor has many microscopic patterns, most elucidated by Teilum in his seminal observations. The most common pattern is reticular, characterized by a loose meshwork of irregular spaces lined by flat, cuboidal, or columnar cells with a primitive appearance. This pattern merges imperceptibly with one that is microcystic and even macrocystic. Often, solitary papillae with a connective tissue core containing a central vessel project into some of the spaces; these structures, known as Schiller-Duval bodies, are highly characteristic of yolk sac tumor but are often absent. Other patterns and features of yolk sac tumor are discussed in detail in the testis fascicle. (1) The tumor cells in yolk sac tumors can almost always be stained at least focally for alpha fetoprotein. They are also stainable for cytokeratin and placental-like alkaline phosphatase; negative staining for OCT-4, as noted earlier, may be of diagnostic utility and is a helpful new feature from the immunohistochemical viewpoint. The yolk sac tumor has a limited differential diagnosis, but a few comments are merited. It should be remembered that one can see papillae with blood vessels in embryonal carcinoma, but the papillae are covered by the typical cells of embryonal carcinoma. A variety of tumors may have a somewhat loose morphology, which can be misconstrued as the reticular pattern of yolk sac tumor. In children, the differential with a juvenile granulosa cell tumor can arise, something considered under that neoplasm later. A pattern of yolk sac tumor that may be confusing is the solid pattern, because it may suggest the diagnosis of a seminoma. However, the typical septa with scattered lymphocytes of seminoma are absent in these cases. Additionally, most of the cells in the solid foci of yolk sac tumor are smaller than seminoma cells and do not have their "squared off" nuclear membranes and typically have less prominent nucleoli. Occasionally, the solid foci in yolk sac tumor have cells with scant cytoplasm, imparting a blastema-like look. TROPHOBLASTIC-TYPE TUMORS Choriocarcinoma occurs very rarely as a pure neoplasm; much more often it is a component of a mixed germ cell tumor (Figure 7). Microscopic examination shows patterns similar to those encountered in uterine choriocarcinomas, but rarely is there a monophasic pattern. (16) The tumors stain immunohistochemically for human chorionic gonadotropin, placental lactogen, and various placental proteins. One testicular tumor indistinguishable from the placentalsite trophoblastic tumor of the uterus has been described in the testis of a child. (16) It is important that choriocarcinoma be distinguished from seminoma or embryonal carcinoma with syncytiotrophoblast giant cells. In those instances, there is not the admixture of cytotrophoblast with syncytrophoblast that is usually seen in choriocarcinoma, even allowing for the fact that some examples of the latter have a less classically formed biphasic pattern than some others. In seminomas and embryonal carcinomas, the cells adjacent to the giant cells are typical seminoma or embryonal carcinoma cells. TERATOMAS Teratomas of the testis are rare in pure form, and there is limited new information on these tumors, which have one particularly unusual clinical feature; namely, the fact that those occurring in prepubertal boys are invariably benign, regardless of their histology, whereas those occurring after puberty have a malignant potential, even when histologically mature.17 The one exception to the latter comment is the rare mature cystic teratoma (dermoid cyst) of the testis. Although sporadically reported previously, there is only one sizable series of dermoid cysts of the testis reported recently. (18) The diagnosis should be reserved for lesions that are grossly typical of a dermoid cyst and are unassociated with adjacent intratubular germ cell neoplasia unclassified. The more common mature teratoma that has a malignant potential has a solid and cystic gross-sectioned surface contrasting with the predominantly cystic nature of the dermoid cyst. [FIGURE 7 OMITTED] In cases of highly immature teratomas, there may be prominent immature skeletal muscle, and judging whether such is still compatible with either immature teratoma or rhabdomyosarcoma arising out of a teratoma may be difficult. There should be unequivocal expansile overgrowth of the skeletal muscle to make the rare diagnosis of rhabdomyosarcoma. Monodermal Teratomas The most common monodermal teratoma is the carcinoid tumor of the testis, which has not been the subject of significant new information in the literature. (19) Their features and those that distinguish them from metastatic carcinoid are well known.1 In contrast, it has been appreciated recently that primitive neuroectodermal components in testicular germ cell tumors may in some instances be conspicuous and even dominate the microscopic picture, rarely representing most of or the entire neoplasm. (20) These foci of malignant small cell neoplasia with varyingly prominent differentiation of neuroectodermal type are usually recognized when there is an associated teratomatous component from which they arise, but when the latter are inconspicuous, problems in differential diagnosis may result. Foci resembling various forms of central nervous system tumors, such as medulloepithelioma or ependymal neoplasms or neuroblastoma, may be encountered. Overgrowth of a neuroectodermal tumor component of whatever form is only diagnosed when the spatial distribution of tissues, primitive or otherwise, typical of teratoma is no longer present, and architectural features indicate overgrowth of the neuroectodermal tissues. [FIGURE 8 OMITTED] POLYEMBRYOMA This exceptionally rare gonadal neoplasm has its own place in the classification of testicular tumors and, despite its exotic nature, it is academically interesting, morphologically intriguing (Figure 8, A through D), and as a component is not rare in a mixed germ cell tumor, meriting these comments. The one polyembryoma we have personally seen grossly in the testis was a beefy mass. These neoplasms are composed entirely or predominantly of embryoid bodies, which when fully formed are remarkably pretty structures, constituted by a central germ disc, underlain by yolk sac epithelium, and with recapitulation of the amniotic cavity anteriorly and the yolk sac cavity posteriorly, as oriented in the illustration provided (Figure 8, D). From the embryologic viewpoint, though, the amniotic cavity-like space is, of course, dorsal and the yolk sac-like space ventral. The yolk sac tissue is often associated with extraembryonic mesenchyme, and trophoblast is occasionally present. Whether a neoplasm composed entirely of embryoid bodies has ever occurred is arguable; in our experience, even the examples that get closest still contain some typical teratomatous elements (Figure 8, C). For this reason, Robert E. Scully, MD, has remarked that conceptually, the polyembryoma can be viewed as the most immature of all possible forms of immature teratoma, recapitulating as it does early embryonic life (R. E. Scully, personal communication). An alternative view, which some take, is to consider the neoplasm a special form of mixed germ cell tumor, containing as it does embryonal carcinoma-type epithelium and yolk sac epithelium. Regardless of the interesting but practically inconsequential debate on histogenesis, it is satisfying to recognize a polyembryoma, whether in close to pure form or, as is more frequent, when a component of varying prominence within a conventional mixed germ cell tumor of the adult testis. [FIGURE 9 OMITTED] [FIGURE 10 OMITTED] [FIGURE 11 OMITTED] Foci of polyembryoma often have a prominent loose myxomatous mesenchyme in the background (Figure 8, A) upon which the embryoid bodies are set. Large, well-circumscribed lobules of embryoid bodies surrounded by the prominent mesenchyme produce a rather typical low-power appearance in many cases (Figure 8, A and B). Cysts may be prominent (Figure 8, B). The embryoid bodies themselves often are not perfectly formed and appear fragmented, as if they are breaking down. Cysts may be conspicuous. In some cases, one sees the associated yolk sac or embryonal carcinoma epithelium forming a microscopic aggregate, emanating from its constituent parts in the embryoid body, and the point at which it is enough to separately merit the diagnosis of a microscopic component of embryonal carcinoma or yolk sac tumor is debatable, but we have proposed a size of 3 mm as an arbitrary cutoff point. Because of their component of trophoblast, these tumors may be associated with elevation of the serum human chorionic gonadotropin level. MIXED GERM CELL TUMORS I will make little comment on these tumors other than to note that a bewildering array of patterns may result because of the diversity of tumor components and their varied prominence from case to case. Essentially, any admixture of the germ cell tumors as seen in pure form may be seen, one of the most common admixtures being embryonal carcinoma and teratoma (the old teratocarcinoma). Minor foci of yolk sac tumor are common (Figure 9), although it is usually overshadowed by other components, such as embryonal carcinoma. As is typical of embryonal carcinoma when seen in pure form, epithelium is often associated with syncytiotrophoblast giant cells when seen as part of a mixed germ cell tumor (Figure 9). Although seminoma may be seen as part of a mixed germ cell tumor, it has struck me that in some cases one sees seminoma separate from a dominant mass of nonseminomatous mixed germ cell neoplasia, and in such cases it is probably truly multicentric neoplasia, although for signout purposes it is probably sufficient to consider the seminoma together with the other neoplastic components under the one designation of mixed germ cell tumor with the traditional rough quantitation of the various components in descending order of frequency. BURNT-OUT (REGRESSED) GERM CELL TUMORS A most remarkable phenomenon in testicular pathology, which interestingly has no companion in the female gonad, is that seen when germ cell tumors undergo apparent spontaneous regression, resulting in complete or almost complete hyaline scarring. This phenomenon, which is variously referred to as spontaneous regression or "burntout" germ cell tumor, has been the subject of a recent study by the Indiana University group, (21) that is recommended to the reader. In these cases, the testis typically is firm and white and on microscopic examination hyaline scarring is seen, sometimes with a minor component of associated viable neoplasia, such as seminoma, teratoma, or intratubular germ cell neoplasia. The latter may be a subtle clue to the diagnosis that what one is observing is something more specific than just an old hyalinized infarct on nonspecific atrophy. Sometimes, the scar shows rather prominent vessels (Figure 10), hemosiderin-laden macrophages, chronic inflammatory cells and, particularly in cases that are presumptively "dead" embryonal carcinoma, intratubular calcification. Although these regressed tumors may be sometimes clinically apparent as testicular abnormalities on palpation, they often are only discerned on ultrasound, the classic situation being when a patient is investigated after presenting with a retroperitoneal germ cell tumor, which is ultimately shown to be a metastasis from the regressed testicular primary. [FIGURE 12 OMITTED] [FIGURE 13 OMITTED] [FIGURE 14 OMITTED] SEX CORD-STROMAL TUMORS Leydig Cell Tumors These tumors account for 1% to 3% of testicular neoplasms. (22,23) They occur at all ages but are most common between 20 and 50 years of age. Gynecomastia is the initial symptom in some cases. Children with Leydig cell tumors almost always have isosexual pseudoprecocity. Leydig cell tumors are typically sharply circumscribed, and the neoplastic tissue is almost always uniformly solid and usually yellow or yellow-tan, but occasionally gray, white, brown, or green-brown. Necrosis and hemorrhage may be seen and cause concern for malignancy. The most common microscopic pattern is diffuse (Figure 11), but growth as large nodules is not rare. Insular, trabecular, pseudotubular, and ribbonlike arrangements of the tumor cells are occasionally encountered. Necrosis may impart a pseudopapillary appearance (Figure 12). The stroma may be hyalinized and occasionally edematous or myxoid. The neoplastic cells resemble normal Leydig cells in most cases; they typically have abundant eosinophilic, slightly granular cytoplasm, but the cytoplasm may be spongy as a result of lipid accumulation. Crystals of Reinke are found in approximately one third of the cases. In two recent papers, some unusual morphologic manifestations of Leydig cell tumors have been described for the first time or highlighted. (24,25) The first of these contributions recorded cystic spaces, usually small, but sometimes large (Figure 13), resulting in confusion with various neoplasms, such as yolk sac tumor. (24) It is hard to think of a better example of the importance of thorough sampling in evaluating gonadal tumors, because when this results in other slides showing typical Leydig cell tumor, mere awareness of the phenomenon will enable the correct interpretation to be made. Particularly in cases with unusual morphology, staining for inhibin may be helpful, as it may in cases of sex cord tumors of other types, such as Sertoli cell tumors, granulosa cell tumors, and those in the unclassified group. The second series contained 12 tumors with adipose differentiation, 8 with spindle cells, and 3 with calcification and ossification. (25) These features had only rarely been described previously. The adipose differentiation is important in that it also is a feature of some cases of the testicular tumor of the adrenal genital syndrome and, accordingly, does not reliably facilitate the differential diagnosis of these two entities. A spindle cell pattern, if prominent, may obscure the underlying diagnosis of Leydig cell tumor and bring other entities, such as sex cord-stromal tumor, unclassified, or even testicular sarcoma, into the differential diagnosis, but thorough sectioning and awareness of the phenomenon usually resolves the problem. Calcification can potentially erroneously suggest the diagnosis of a large cell calcifying Sertoli cell tumor, but the latter entity shows unequivocal tubular differentiation. Ossification is only a pathologic curio. None of these unusual features, including spindle cell, appears to have prognostic importance. No single pathologic criterion reliably distinguishes between benign and malignant Leydig cell tumors. (22) However, the latter are typically larger, have infiltrative margins, invade lymphatics or blood vessels, and contain foci of necrosis. They also have a mitotic rate of more than 3 per 10 high-power fields and significant nuclear atypicality much more often than benign tumors. Rarely, however, a tumor with little or no suggestion of malignancy on gross or microscopic examination metastasizes, usually late in the course of the disease. Additional issues in differential diagnosis not related to tumors with unusual features already discussed are briefly noted here. Perhaps the most important remains the occasional confusion of this neoplasm with the remarkable nonneoplastic entity, the so-called testicular tumor of the adrenal genital syndrome. (1) Individual fields may be indistinguishable. A triad of microscopic findings should, certainly, if all present, lead to serious consideration of the nonneoplastic lesion, namely, particularly broad fibrous bands (Figure 14, A), spotty nuclear atypia without mitotic activity (Figure 14, B), and cytoplasmic lipochrome pigment. The latter, when conspicuous, may result in the nonneoplastic lesion having a particularly dark brown color, but it must be acknowledged that occasional Leydig cell tumors may have a similar color, albeit less often. Should a mass lesion be bilateral and have any extra testicular nodularity, the nonneoplastic lesion should be strongly considered, particularly if any of the microscopic findings already noted are seen. Occasionally, true Leydig cell hyperplasia or at least relative prominence of Leydig cells in an atrophic testis may suggest a small Leydig cell tumor, but these are typically multifocal and separated by seminiferous tubules, a setting that should always favor a nonneoplastic proliferation. Rarely, malakoplakia may be an issue, but associated inflammatory cells, sometimes even abscess formation, as well as, of course, the distinctive Michaelis-Gutmann bodies, should be diagnostic. Sertoli Cell Tumors Sertoli cell tumors account for less than 1% of testicular neoplasms. They occur at all ages. On gross examination they are typically well circumscribed, sometimes lobulated, yellow, tan, or white masses. Cysts may be present. Hemorrhage and necrosis may be seen, particularly in malignant tumors. Microscopic examination typically reveals tubules that are sometimes hollow but are usually solid, as well as cords, nests, and masses of cells consistent with Sertoli cells. In our study of 60 Sertoli cell tumors, not otherwise specified, (26) several morphologic findings found were noteworthy in that they had not been emphasized in the prior literature: the presence in many cases of a prominent, diffuse growth with rather limited evidence of the tubular component that enables the diagnosis to be rendered with confidence, the presence in some instances of cells with copious eosinophilic cytoplasm, bringing the differential diagnosis of a Leydig cell tumor into consideration, abundant hyalinization, and prominent, sometimes dilated, vessels in the stroma. A subsequent article on malignant Sertoli cell tumors (27) further emphasized the problem the above noted diffuse pattern may cause. First of all, not surprisingly, a diffuse growth is particularly likely to be seen in malignant tumors because of the limited differentiation seen in such cases. The 13 cases in this category were usually initially misdiagnosed as seminoma, usually of the classic type but occasionally of the spermatocytic type. In many of them, the correct diagnosis was only made retrospectively after failure of the patient to respond to therapy, usually successful in cases of seminoma. The problems on microscopic examination were related primarily to the diffuse growth but were compounded by other features common in seminoma: a nested pattern, fibrous septa (Figure 15, A), cells with clear cytoplasm that was positive for glycogen in 3 of 4 tumors tested, prominent nucleoli (5 cases), and lymphoid infiltrate (10 cases). Focal tubular differentiation was crucial diagnostically but was often limited in amount (Figure 15, B). Other helpful findings were nuclei that were smaller and less pleomorphic than those of seminomas and immunohistochemical staining for inhibin and negative staining for placental-like alkaline phosphatase. As expected, the testis adjacent to the tumor did not show any intratubular germ cell neoplasm. Small nodules that are composed of immature Sertoli cells and, rarely, germ cells that occur most often in cryptorchid but occasionally in descended testes are generally considered to be proliferating clusters of immature tubules rather than true neoplasms and should be referred to as "Sertoli cell nodules." Other issues in differential diagnosis are limited and are primarily as follows. There may be a pattern superficially reminiscent of the cystic pattern of yolk sac tumor in rare neoplasms, but the primitive cytology of yolk sac tumor is lacking. Rete cystadenomas may have a Sertoliform pattern (28) but have a distinctive rete location. Finally, adenomatoid tumors may have a solid tubular appearance but the typical vacuoles, even when few in number, will be diagnostic, as will immunohistochemistry if indicated. (29) An unusual phenomenon associated with an occasional Sertoli cell tumor and, for that matter, sex cord stromal tumors of the testis of other types, such as unclassified forms, arises when these often slowly growing tumors incorporate nonneoplastic germ cells within them, resulting in a mimicry of true mixed germ cell-sex cord stromal tumors. We recently reported 10 cases of this type. (30) The admixed germ cells were usually at the periphery and in clusters, but occasionally were in the center of the neoplasm or more diffusely distributed (Figure 16). None of the entrapped germ cells stained with markers for neoplastic germ cells. In our experience, this peculiar phenomenon of entrapment of germ cells within a sex cord tumor is more common than the exceptionally rare mixed germ cell sex cord-stromal tumor unclassified. Even vacuolated lutein cells in sex cord-stromal tumors may be erroneously considered germ cells, resulting in a similar misdiagnosis. [FIGURE 15 OMITTED] [FIGURE 16 OMITTED] [FIGURE 17 OMITTED] From the behavior viewpoint, our first study of Sertoli cell tumors disclosed that of 16 patients with follow-up during 5 years, 4 were alive with disease and 3 died of disease. (26) The pathologic features that correlated best with a clinically malignant course were a tumor diameter of 5 cm or greater, necrosis, moderate to severe nuclear atypia, vascular invasion, and a mitotic rate of more than 5 per 10 high-power fields. Only 1 of 9 benign tumors with follow-up data of 5 years or longer had more than one of these features, whereas 5 of the 7 malignant tumors had at least 3. However, the unpredictability of any one feature in reliably predicting prognosis was borne out by the subsequent series of Henley et al (27) reporting malignant Sertoli cell tumors that have a propensity to be misdiagnosed as seminoma. In that study, 5 of the tumors were under 5 cm in diameter, and enigmatically the tumors typically had low mitotic rates, less than 5 mitotic figures per 10 high-power fields. The architectural lack of differentiation seen in that series in and of itself seems to be somewhat predictive of a potentially malignant course. Two subtypes of Sertoli cell tumors that are distinctive, the large cell calcifying Sertoli cell tumor and the sclerosing Sertoli cell tumor, are now considered separately. LARGE CELL CALCIFYING SERTOLI CELL TUMOR This subtype of Sertoli cell tumor is associated in almost half of the reported cases with other endocrine lesions, such as pituitary adenomas, bilateral primary adrenocortical hyperplasia, and testicular Leydig cell tumors. (31) Cardiac myxomas and spotty mucotaneous pigmentation have also been reported. The clinical associations have included acromegaly, gigantism, hypercortisolemia, sexual precocity, sudden death, and the Peutz-Jeghers syndrome, although these lesions occurring in the latter syndrome appear to have unique features (see below). The large cell calcifying Sertoli cell tumor is usually 4 cm or less in diameter and is multifocal or bilateral in almost half of the cases, although this figure may be spuriously high due to the tendency for syndrome-associated cases to be reported. Sectioning discloses firm yellow to tan tissue in which granular calcific foci may be detectable. The tumors appear grossly well circumscribed in most cases. Microscopic examination reveals that the neoplastic cells are arranged in sheets, nests, trabeculae, cords, small clusters, or solid tubules. The characteristic feature is the presence of calcification, which is usually conspicuous and sometimes massive, with the formation of large, wavy, laminated nodules. The neoplastic cells are typically large and rounded, but occasionally are cuboidal or columnar, and rarely are spindle shaped. In most of the cases, the cytoplasm is abundant, eosinophilic, and finely granular. The nuclei are round or oval without conspicuous nucleoli Mitotic figures are generally rare. One study (32) focused on the differing features of malignant and benign large cell calcifying Sertoli cell tumors. It showed that the malignant tumors were unilateral and solitary and occurred at a mean age of 39 years, whereas the benign neoplasms were more often bilateral and multifocal in 28% of the cases and occurred at a mean age of 17 years. Only one malignant tumor occurred in a patient with evidence of a genetic syndrome (Carney syndrome), whereas 36% of the benign tumors had various genetic syndromes and/or endocrine abnormalities. Most of the tumors in the latter cases were bilateral or multifocal. There were strong associations of a malignant behavior with size larger than 4 cm, extratesticular growth, gross or microscopic necrosis, high-grade cytologic atypia, vascular space invasion, and a mitotic rate of more than 3 mitoses per 10 high-power fields. All of the malignant cases exhibited at least 2 of these features, whereas all of the benign cases lacked any of them. The presence of any 1 of these features in a solitary large cell calcifying Sertoli cell tumor, especially in a patient older than 25 years, should be considered suspicious for a malignant behavior, whereas 2 or more of these indicate a strong probability of a malignant course. INTRATUBULAR LARGE CELL HYALINIZING SERTOLI CELL TUMOR This lesion, which occurs in boys with Peutz-Jeghers syndrome, is hybrid in appearance, and resembles in some cases, at least to an extent, both the large cell calcifying Sertoli cell tumor and the sex cord tumor with annular tubules. (33,34) It represents a distinctive variant with likely no metastatic potential. It typically occurs in young boys, may be estrogenic, and is bilateral and multifocal. Typically, no gross abnormality is seen, although in one of our cases, multiple small, white foci were noted in a biopsy specimen, and in the one patient treated by orchiectomy there was ill-defined nodularity. Microscopic examination is distinctive, showing a patchy distribution of expanded tubules, sometimes solitary tubules, several times the diameter of adjacent uninvolved tubules. The abnormal tubules are filled and enlarged by Sertoli cells and globular eosinophilic basement membrane deposits that connect up with thickened peritubular basement membrane (Figure 17). The Sertoli cells typically have abundant pale to eosinophilic, frequently vacuolated cytoplasm and have a fibrillar quality. Nuclear characteristics are typically bland, and mitotic figures are exceptional. Two of our 8 cases had focal psammoma body-like calcification within the basement membrane material. Occasionally, the tubules are almost totally replaced by the basement membrane deposits, resulting in the hyalinized appearance. There are only two series of this phenomenon and a small number of case reports, so there is still much to learn about it, but a rather unique profile is emerging. SCLEROSING SERTOLI CELL TUMOR This tumor (35) is typically small and is characterized by the presence of tubules, nests, and cords of generally well-differentiated Sertoli cells in an abundant hyaline stroma. Although 1 in our series of 10 cases was not sharply circumscribed but invaded the surrounding tissue, no case has been reported with a malignant outcome. GRANULOSA CELL TUMORS Granulosa cell tumors of the type encountered in the ovary of an adult woman are very rare in the testis. (36) They resemble their ovarian counterparts. They may be clinically malignant. Juvenile granulosa cell tumors are more interesting because they almost invariably occur in the first few months of life, when they are the most common form of testicular tumor, (37) and may pose major problems in differential diagnosis. These neoplasms may occur in infants with mixed gonadal dysgenesis. On gross examination, these tumors vary greatly in size, with many of them replacing most or all of the testis. They are usually well circumscribed and are composed of white to yellow, often lobulated tissue. Cysts may be present. Microscopic examination reveals a spectrum of patterns (Figure 18), ranging from predominantly epithelial to predominantly stromal, but a mixed pattern, usually with at least focally prominent follicles, is most common. The follicles are often variable in both size and shape (Figure 18, A). In some cases, lipid vacuoles are conspicuous within the tumor cells (Figure 18, B). One may see solid or hollow tubules, or cords composed of or lined by cells resembling Sertoli cells. Occasionally, a more or less diffuse pattern is seen, and when the cells are small with hyperchromatic nuclei and briskly mitotic, misdiagnoses of various forms may be made (Figure 18, C and D). Usually, the nuclei of the granulosa cells lack the features of those of ovarian adult granulosa cell tumors, such as nuclear grooves. There may be varying degrees of nuclear pleomorphism, and mitotic activity is often striking. Prior to the delineation of the juvenile granulosa cell tumor, this neoplasm was most likely occasionally misdiagnosed as yolk sac tumor, a better-known tumor of the testis of the young. The follicular pattern of juvenile granulosa cell tumor is perhaps the major clue to the diagnosis, but it is also crucial to merely be aware of the existence of juvenile granulosa cell tumor as an entity in the testis and its great tendency to occur in the first few months of life, whereas the yolk sac tumor occurs generally from 1 year on. I have seen the mitotic activity of this tumor and spindle cells lead to the misdiagnosis of rhabdomyosarcoma. Among other things, the gross is important, as such sarcomas are almost always paratesticular rather than testicular. It may not, however, always be possible to make this distinction when the sarcoma reaches a large size. [FIGURE 18 OMITTED] There is some significant degree of subjectivity in the diagnosis of the juvenile granulosa cell tumor and its distinction from other neoplasms. For example, when criteria similar to ours are used, these tumors greatly outnumber Sertoli cell tumors in children. (38) However, in a study from another group, (39) in which different criteria were used, Sertoli cell tumors accounted for a much higher percentage of neoplasms in children. Unfortunately, the overlap between these two neoplasms will result in considerable variation in interpretation from center to center for the foreseeable future, in all probability. MALIGNANT LYMPHOMA, PLASMACYTOMA, AND LEUKEMIA, INCLUDING MYELOID AND MONOCYTIC SARCOMA Malignant lymphomas account for 5% of testicular tumors and approximately half of those in men older than 60 years. (40,41) The tumor is bilateral in about 10% to 15% of the cases. In about 25% of the cases, testicular lymphoma is a manifestation of widespread disease that is evident at the time of diagnosis or within a short interval after treatment of the testicular tumor. Approximately an additional 25% of the patients have Ann Arbor stage II disease (involvement of lymph nodes below the diaphragm), and the remaining patients have disease confined to the testis or with only involvement of adjacent structures. In a recent series (41) focusing on the last group of stage I cases, the prognosis was relatively favorable, particularly compared with that previously associated with this disease. Gross examination of a testicular lymphoma reveals partial or complete replacement of the organ by a fleshy or firm homogeneous mass, which may be cream-colored, tan, pale yellow, or slightly pink. The tumor resembles a typical or spermatocytic seminoma but invades the epididymis or spermatic cord much more often than those tumors. On microscopic examination, the tumor is often conspicuously or sometimes entirely intertubular but may efface tubules or invade them, creating an erroneous impression that it arose within them in the last circumstance. Unlike the seminoma, however, lymphoma is rarely seen within tubules in the adjacent testis uninvolved by the main tumor mass. Nearly all testicular lymphomas are diffuse large B-cell lymphomas. (41) Rare primary testicular lymphomas are follicular lymphomas. (42) The majority of these have been described in children, but they occasionally occur in adults who are usually relatively young. In contrast to nodal follicular lymphomas, follicular lymphomas arising in the testis are usually localized (Ann Arbor stage I). In contrast to nodal follicular lymphomas, most testicular follicular lymphomas are grade III, rather than grade I or grade II. The immunophenotype is similar to that of nodal follicular lymphoma, with the important difference being that testicular follicular lymphoma lacks expression of BCL2 protein in those that have been reported. In addition, testicular follicular lymphomas have lacked a BCL2 gene rearrangement, again, in contrast to the vast majority of nodal follicular lymphomas. The patients have had an excellent prognosis. (42) Only one case of Hodgkin lymphoma presenting with testicular involvement has been reported. (43) The differential diagnosis of malignant lymphoma and germ cell tumors, with which they are most often confused, should not really be difficult because of the contrasting features of the cells in lymphomas and germ cell tumors. Nonetheless, errors in diagnosis are not rare. (41) In one case, an anaplastic large cell lymphoma was difficult to distinguish from an embryonal carcinoma because of the carcinoma-like appearance that is a feature of this particular lymphoma. (44) It should be noted that staining for CD30 is not discriminatory in this situation, but staining of the lymphoma for leukocyte common antigen and T-cell-associated antigens was helpful. In the largest series of cases of malignant lymphoma of the testis reported to date, the features that correlated best with a long disease-free survival were stage I disease and sclerosis. Among stage I patients, enigmatically, those with right-sided tumors had a better prognosis. (40) Six cases of testicular and 1 case of epididymal plasmacytoma were recently described by Ferry and colleagues. (45) All of the patients had plasma cell neoplasia distant from the testis identified before (3 cases), concurrent with (3 cases), or after (1 case) the testicular or epididymal plasmacytoma. Three of the 4 consultation cases in this series were submitted with diagnoses such as spermatocytic seminoma, "anaplastic" seminoma, or lymphoma. The neoplastic cells in lymphoma have less abundant cytoplasm than plasma cells and lack a paranuclear hof. In addition, sclerosis is common in testicular lymphomas but was not seen in any of the plasmacytomas. Most testicular lymphomas present with stage I disease, whereas in cases of testicular plasmacytoma there is usually widespread disease at diagnosis, usually in the form of plasma cell myeloma, as in 6 of the 7 cases in the study of Ferry et al. (46) Only occasionally do seminomas have the appreciable eosinophilic cytoplasm of plasmacytomas. The pleomorphism of some plasmacytomas may, on low power, suggest the heterogeneous cell population of spermatocytic seminoma, but on high-power examination the neoplastic cells have the appearance of mature or immature plasma cells, in contrast to the cytologic features of the 3 cell types in spermatocytic seminomas. One of the 2 granulocytic sarcomas of the testis mentioned below was initially misinterpreted as plasmacytoma because of a prominent component of myelocytes with eccentric nuclei and abundant cytoplasm. Unlike most testicular plasmacytomas, however, the granulocytic sarcoma was associated with extensive extratesticular (intrascrotal) spread, associated sclerosis, and cytoplasmic granularity. Immunohistochemical and enzyme histochemical stains will be definitive if this problem cannot be resolved by evaluating routine material. Microscopic involvement of the testis has been found at autopsy in about two thirds of patients with acute leukemia and about one fifth of those with chronic leukemia. The testis may serve as a sanctuary for the neoplastic cells in patients being treated for acute lymphoblastic leukemia, although this complication is now uncommon because of improvements in therapy. On microscopic examination, the pattern of leukemic infiltration is similar to that of lymphoma. Rarely, a granulocytic sarcoma is the first manifestation of acute myeloid leukemia. (46) It can be confused easily with a large cell lymphoma unless special staining is done. SECONDARY TUMORS Metastases to the testis are rare in contrast to the frequency of metastatic tumors in the ovary. The most frequent sources of testicular metastasis have been carcinoma of the prostate gland, colon, kidney (Figure 19), stomach, and pancreas, and malignant melanoma of the skin. (47) Occasionally, carcinoid tumors also have metastasized to the testis. The testicular mass has been the presenting manifestation in almost 10% of the reported cases. Most of these tumors have distinctive features that enable them to be distinguished from primary testicular cancers. Metastatic carcinoid tumors have to be differentiated from primary carcinoid tumors. This distinction is often difficult, because a very small carcinoid tumor of the intestine may be difficult to detect clinically or radiologically. If other teratomatous elements can be found, the diagnosis of primary carcinoid tumor can be made. As in other areas of the body, metastatic malignant melanoma can be a problem. We described 3 examples of this phenomenon. (48) The patients were 33, 55, and 80 years of age, and in 2 of them the initial examining pathologist did not know of the history of malignant melanoma at the time the testicular tumor was first examined. All the tumors were unilateral. One had a discrete multinodular growth suggestive of metastasis, but the others were discrete solitary masses with no distinctive features. On microscopic examination, diffuse, nested, follicular, and fascicular patters, usually in combination, were seen. Two of the 3 tumors had neoplastic cells with moderate to abundant eosinophilic cytoplasm. Melanin pigment was identified in 2 tumors but was inconspicuous in them. The potential misinterpretations that can occur in these cases are widespread, ranging from seminoma to sex cord-stromal tumors to sarcoma. As this is a well-known facet of metastatic melanoma in general, one just simply has to think of this possible diagnosis when an unusual morphology is encountered that falls within the spectrum melanoma is known to produce. Immunohistochemistry can obviously play a major role in resolving the diagnosis in cases that truly are melanoma. Even ultrastructural examination may occasionally be indicated. We have also reported 4 cases of metastases to the testis of another well-known mimicker of diverse neoplasms because of its varied patterns, namely, renal cell carcinoma. (49) In 3 of them, the testicular metastasis accounted for the initial clinical presentation. The youngest patient was 46 years of age. All of the tumors were unilateral and ranged from 1.8 to 5 cm. On microscopic examination, all of the tumors were of the clear cell variant of renal cell carcinoma (Figure 19) and had varied patterns, including solid sheets, acini, cysts, alveoli, and trabeculae. The primary testicular tumor most often mimicked was a Sertoli cell tumor because of a pseudotubular pattern and the clarity of the cell cytoplasm. Seminoma may also be another misdiagnosis made. This is an area in which immunohistochemistry for inhibin may be very helpful diagnostically, obviously being negative in the metastatic renal tumors. The typical vascularity of renal cell carcinoma may be a helpful clue in evaluating routinely stained slides. [FIGURE 19 OMITTED] [FIGURE 20 OMITTED] [FIGURE 21 OMITTED] [FIGURE 22 OMITTED] [FIGURE 23 OMITTED] ADENOMATOID TUMOR This is the only paratesticular neoplasm about which there is new information since publication of the AFIP fascicle. (1) This tumor is most often a discrete 2- to 3-cm mass in the epididymis, usually at its lower pole, but may also arise in or beneath the tunica albuginea of the testis or in the spermatic cord. Occasionally, the testicular parenchyma is involved by local extension of an extratesticular tumor. Very rarely, the gross may suggest a bona fide parenchymal testicular neoplasm; a striking example of this has recently been depicted in the literature by Delahunt and Kenwright. (50) Typically, the tumor forms a round to oval, well-demarcated mass, but it may be plaquelike. It is composed of tan-gray-white glistening tissue and may resemble a germ cell tumor, especially a seminoma, on gross examination. The latter aspect is, of course, particularly relevant in the case of the rare parenchymal examples. On microscopic examination, it may appear to have an almost solid pattern on low-power magnification (Figure 20), but high-power magnification scrutiny reveals tubules (Figure 21), glandlike structures, cords, and clusters of cells and individual cells containing large intracytoplasmic vacuoles, which may be filled with a pale mucoid material. Lymphoid aggregates may be conspicuous (Figure 20), and in some instances may be a low-power "tip off" to the diagnosis. The stroma of the tumor ranges from inconspicuous to somewhat prominent, is usually fibrous and sometimes hyalinized, and may contain smooth muscle. Some tumors may have an infiltrative border and varying degrees of cytologic atypicality. Some particularly challenging cases are those that have been infarcted (Figure 22), as recently emphasized in the literature based on a report of 5 cases of this phenomenon. (51) Several issues resulted. First of all, the extensive necrosis often made it difficult to recognize that the neoplasm was an adenomatoid tumor in the first place. Additionally, the necrotic areas were surrounded by a florid reactive process characterized by fibroblasts and myofibroblasts with plump nuclei, often with prominent nucleoli and occasional mitoses. Furthermore, this resulted in blurring of the usual relatively well-delineated periphery of an adenomatoid tumor, potentially raising concern for a malignant neoplasm infiltrating the adjacent tissue. Recognition of ghost outlines of the typical formations of an adenomatoid tumor in the necrotic regions and awareness of the phenomenon are crucial in recognizing these cases as benign. CONCLUDING REMARKS Our knowledge of testicular tumors is now at a relatively advanced stage due to the many authors from prior years who have contributed greatly to this area. (52) Historic landmarks, such as the seminal description of seminoma by Chevassu in 1906 and the classic paper of Friedman and Moore from 1946, should never be forgotten. (53) Tribute is paid to other luminaries who have contributed elsewhere. (52) It is still, however, worthy of some reflection that a number of morphologic observations of great importance have been made only in recent years, and doubtless others remain to be detected by the astute eye. Although careful gross and microscopic evaluation of routine type, buttressed indeed in some instances by cognizance of the clinical background, including the mere age of the patient, will likely remain the fundamental aspect of testicular tumor interpretation forever, we are fortunate that immunohistochemistry (Figure 23) has come along to aid significantly, as expertly reviewed in a paper noted earlier. (4) Even newer modalities, such as molecular pathology, may potentially play a role in the future, although their role at this time is minimal. As in surgical pathology in general, a balanced approach incorporating the new where indicated, but never forgetting the "old," will hold the diagnostic pathologist in good stead and, more importantly, result in optimal patient care. The vast majority of what I have learned in this area has been from my mentor, Robert E. Scully, MD, whose knowledge of gonadal pathology is unsurpassable in my opinion. Although retired, Dr Scully is fortunately still available to review particularly difficult cases. It is appropriate that this review appear in this journal, coming as it does almost 60 years after the first original contribution of Dr Scully's remarkable career appeared in this very same journal and on the same topic (Archives of Pathology 1948;45:581-607). I would also like to express my gratitude to my good friend Thomas M. Ulbright, MD, of Indiana University for not only the pearls he has passed on to me given his great knowledge of this area, but also the enjoyment I have had working on manuscripts in testicular pathology with him. The contributions from Indiana University, headed up by him represent, in my opinion, a remarkable contribution to our knowledge of testicular pathology. Finally, and closer to home, another good friend, Judith A. Ferry, MD, has yet again assisted me with the portion of this essay discussing hematopoietic neoplasms involving the testis, an area in which she is an acknowledged authority. References (1.) Ulbright TM, Amin MB, Young RH. Tumors of the Testis, Adnexa, Spermatic Cord and Scrotum. Washington, DC: Armed Forces Institute of Pathology; 1999. Atlas of Tumor Pathology; 3rd series, fascicle 25. (2.) Henley JD, Young RH, Wade CL, UlbrightTM. Seminomas with exclusive intertubular growth: a report of 12 clinically and grossly inconspicuous tumors. Am J Surg Pathol. 2000;28:1 163-1 168. (3.) Ulbright TM, Young RH. Seminoma with tubular, microcystic, and related patterns: a study of 28 cases of unusual morphologic variants that often cause confusion with yolk sac tumor. Am J Surg Pathol. 2005;29:500-505. (4.) Emerson RE, Ulbright TM. The use of immunohistochemistry in the differential diagnosis of tumors of the testis and paratestis. Semin Diagn Pathol. 2005; 22:33-50. (5.) Flynn MJ, Childerhouse A, Mead GM, Theaker JM. Unusual cystic change in classic seminoma of the testis. Am J Surg Pathol. 2006;30:137-139. (6.) Tickoo SK, Hutchinson B, Bacik J, et al. Testicular seminoma: a clinicopathologic and immunohistochemical study of 105 cases with special reference to seminomas with atypical features. Int J Surg Pathol. 2002;10:23-32. (7.) Burke AP, Mostofi FK. Spermatocytic seminoma: a clinicopathologic study of 79 cases. J Urol Pathol. 1993;1:21-32. (8.) Eble JN. Spermatocytic seminoma. Hum Pathol. 1994;25:1035-1042. (9.) Masson P. Etude sur le seminome. Rev Canad Riol. 1946;5:361-387. (10.) Scully RE. Spermatocytic seminoma of the testis: a report of 3 cases and review of the literature. Cancer. 1961;14:788-794. (11.) Rosai J, Silber I, Khodadoust K. Spermatocytic seminoma, I: clinicopathologic study of six cases and review of the literature. Cancer. 1969;24:92-102. (12.) Rosai J, KhodadoustK, Silber I. Spermatocytic seminoma, II: ultrastructural study. Cancer. 1969;24:103-116. (13.) True LD, Otis CN, Delprado W, Scully RE, Rosai J. Spermatocytic seminoma of testis with sarcomatous transformation: a report of five cases. Am J Surg Pathol. 1988;12:75-82. (14.) Floyd C, Ayala AG, Logothetis CJ, Silva EG. Spermatocytic seminoma with associated sarcoma of the testis. Cancer. 1988;61:409-414. (15.) Albores-Saavedra J, Huffman H, Alvarado-Cabrero I, Ayala AG. Anaplastic variant of spermatocytic seminoma. Hum Pathol. 1996;27:650-655. (16.) Ulbright TM, Young RH, Scully RE. Trophoblastic tumors of the testis other than classic choriocarcinoma: "monophasic" choriocarcinoma and placental site trophoblastic tumor: a report of two cases. Am J Surg Pathol. 1997;21:282-288. (17.) Ulbright TM. Germ cell tumors of the gonads: a selective review emphasizing problems in differential diagnosis, newly appreciated, and controversial issues. Mod Pathol. 2005;18:S61-S79. (18.) Ulbright TM, Srigley JR. Dermoid cyst of the testis. Am J Surg Pathol. 2001; 25:788-793. (19.) Berdjis CC, Mostofi FK. Carcinoid tumors of the testis. J Urol. 1977;118: 777-782. (20.) Michael H, Hull MT, Ulbright TA, Foster RS, Miller KD. Primitive neuroectodermal tumors arising in testicular germ cell neoplasms. Am J Surg Pathol. 1997;21:896-904. (21.) Balzer BL, Ulbright TM. Spontaneous regression of testicular germ cell tumors: an analysis of 42 cases. Am J Surg Pathol. 2006;30:858-865. (22.) Kim I, Young RH, Scully RE. Leydig cell tumors of the testis: a clinicopathological analysis of 40 cases and review of the literature. Am J Surg Pathol. 1985;9:177-192. (23.) Cheville JC, Sebo TJ, Lager DJ, Bostwick DG, Farrow GM. Leydig cell tumor of the testis: a clinicopathologic, DNA content and MIB-1 comparison of non-metastasizing and metastasizing tumors. Am J Surg Pathol. 1998;22:1361-1367. (24.) Billings SD, Roth LM, Ulbright TM. Microcystic Leydig cell tumors mimicking yolk sac tumor: a report of four cases. Am J Surg Pathol. 1999;23:546-551. (25.) Ulbright TM, Srigley JR, Hatzianastassiou DK, Young RH. Leydig cell tumors with unusual features: adipose differentiation, calcification and ossification, spindle-shaped tumor cells. Am J Surg Pathol. 2002;2 6:1424-1433. (26.) Young RH, Koelliker DD, Scully RE. Sertoli cell tumors of the testis, not otherwise specified: a clinicopathologic analysis of 60 cases. Am J Surg Pathol. 1998;22:709-721. (27.) Henley JD, Young RH, Ulbright TM. Malignant Sertoli cell tumors of the testis: a study of 13 examples of a neoplasm frequently misinterpreted as seminoma. Am J Surg Pathol. 2002;26:541-550. (28.) Jones MA, Young RH. Sertoliform cystadenoma of the rete testis: a report of two cases. J Urol Pathol. 1997;7:47-53. (29.) Kommoss F, Oliva E, Bittinger F, et al. Inhibin-[alpha], CD99, HEA125, PLAP, and chromogranin immunoreactivity in testicular neoplasms and the androgen insensitivity syndrome. Hum Pathol. 2000;31:1055-1061. (30.) Ulbright TM, Srigley JR, Reuter VE, Wojno K, Roth LM, Young RH. Sex cord-stromal tumors of the testis with entrapped germ cells: a lesion mimicking unclassified mixed germ cell sex cord-stromal tumors. Am J Surg Pathol. 2000; 24:535-542. (31.) Proppe KH, Scully RE. Large-cell calcifying Sertoli cell tumor of the testis. Am J Clin Pathol. 1980;74:607-619. (32.) Kratzer SS, Ulbright TM, Talerman A, et al. Large cell calcifying Sertoli cell tumor of the testis: contrasting features of six malignant and six benign tumors and a review of the literature. Am J Surg Pathol. 1997;21:1271-1280. (33.) Venara M, Rey R, Bergada I, Mendilaharzu H, Campo S, Chemes H. Sertoli cell proliferations of the infantile testis: an intratubular form of Sertoli cell tumor? Am J Surg Pathol. 2001;25:1237-1224. (34.) Ulbright TM, Amin MB, Young RH. Intratubular large cell hyalinizing Sertoli cell neoplasia of the testis: a report of a distinctive lesion of the Peutz-Jeghers syndrome. Am J Surg Pathol. 2007;31:827-835. (35.) Zukerberg LR, Young RH, Scully RE. Sclerosing Sertoli cell tumor of the testis: a report of 10 cases. Am J Surg Pathol. 1991;15:829-834. (36.) Jimenez-Quintero LP, Ro JY, Zavala-Pompa A, et al. Granulosa cell tumor of the adult testis: a clinicopathologic study of seven cases and a review of the literature. Hum Pathol. 1993;24:1120-1126. (37.) Lawrence WD, Young RH, Scully RE. Juvenile granulosa cell tumor of the infantile testis: a report of fourteen cases. Am J Surg Pathol. 1985;9:87-94. (38.) Goswitz JJ, Pettinato G, Manivel JC. Testicular sex cord-stromal tumors in children: clinicopathological study of sixteen children with review of the literature. Pediatr Pathol Lab Med. 1996;16:451-470. (39.) Harms D, Kock LR. Testicular juvenile granulosa cell and Sertoli cell tumours: a clinicopathological study of 29 cases from the Kiel Paediatric Tumour Registry. Virchows Arch. 1997;430:301-309. (40.) Ferry JA, Harris NL, Young RH, Coen J, Zietman A, Scully RE. Malignant lymphoma ofthe testis, epididymis, and spermatic cord: a clinicopathologic study of 69 cases with immunophenotypic analysis. Am J Surg Pathol. 1994;18:376-390. (41.) Al-Abbadi AM, Hattab E, Tarawneh M, Orazi A, Ulbright TM. Primary testicular and paratesticular lymphoma: a retrospective clinicopathologic study of 34 cases with emphasis on differential diagnosis. Arch Pathol Lab Med. 2007; 131:1040-1046. (42.) Bacon CM, Ye H, Diss TC, et al. Primary follicular lymphoma of the testis and epididymis in adults. Am J Surg Pathol. 2007;31:1050-1058. (43.) Seliem RM, Chikwava K, Swerdlow SH, Young RH, Ferry JA. Classical Hodgkin's lymphoma presenting as a testicular mass: report of a case. Int J Surg Pathol. 2007;15:207-212. (44.) Ferry JA, Ulbright TM, Young RH. Anaplastic large-cell lymphoma presenting in the testis: a lesion that may be confused with embryonal carcinoma. J Urol Pathol. 1996;5:139-147. (45.) Ferry JA, Young RH, Scully RE. Testicular and epididymal plasmacytoma: a report of 7 cases, including 3 that were the initial manifestation of plasma cell myeloma. Am J Surg Pathol. 1997;21:590-598. (46.) Ferry JA, Srigley JR, Young RH. Granulocytic sarcoma of the testis: a report of two cases of a neoplasm prone to misinterpretation. Mod Pathol. 1997;10: 320-325. (47.) Haupt HM, Mann RB, Trump DL, Abeloff MD. Metastatic carcinoma involving the testis: clinical and pathologic distinction from primary testicular neoplasm. Cancer. 1984;54:709-714. (48.) Datta MW, Young RH. Malignant melanoma mestastatic to the testis: a report of three diagnostically challenging cases with clinically significant manifestions. Int J Surg Pathol. 2000;8:49-57. (49.) Datta MW, Ulbright TM, Young RH. Renal cell carcinoma metastatic to the testis and its adnexa: a report of five cases including three that accounted for the initial clinical presentation. Int J Surg Pathol. 2001;9:49-56. (50.) Delahunt B, Kenwright DN. Benign solid and acinar mesothelioma (adenomatoid tumor). Pathol Case Rev. 2005;10:206-211. (51.) Skinnider BF, Young RH. Infarcted adenomatoid tumor: a report of five cases of a facet of a benign neoplasm that may cause diagnostic difficulty. Am J Surg Pathol. 2004;28:77-83. (52.) Young RH. A brief history of the pathology of the gonads. Mod Pathol. 2005;18:S3-S17. (53.) Freidman NB, Moore RA. Tumors of the testis: a reporton 922 cases. Milit Surg. 1946;99:573-593. (54.) Ulbright TM, Young RH. Seminoma with conspicuous signet ring cells: a rare, previously uncharacterized morphologic variant. Am J Surg Pathol. In press. Robert H. Young, MD Accepted for publication December 17, 2007. From the Department of Pathology, Massachusetts General Hospital, Harvard Medical School, Boston, Mass. The author has no relevant financial interest in the products or companies described in this article. Presented in part at the 28th Annual Course, Current Concepts in Surgical Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, November 2006. Corresponding author: Robert H.Young, MD, Department of Pathology, Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114.
Gale Copyright:	Copyright 2008 Gale, Cengage Learning. All rights reserved.

Previous Article: The thyroid gland: recurring problems in histologic and cytologic evaluation.
Next Article: Corrections.