Common Diagnostic Challenges and Pitfalls in Genitourinary Organs, With Emphasis on Immunohistochemical and Molecular Updates

The urinary tract extends from the renal collecting tubules proximally to the urethral meatus distally, all of which are lined by urothelium. Any segment of the urothelium can be affected by malignant transformation, with the similar morphologic spectrum. Urinary bladder is the most common site, and the incidence of bladder cancer has continued to rise for decades.¹  Greater than 90% of urinary tract cancers diagnosed in the United States are urothelial carcinomas (UCs), followed by squamous cell carcinoma (SCC; 2%–5%), adenocarcinoma (2%), neuroendocrine (1%), and other rare tumors (<1%).²  Reactive, proliferative, and metaplastic changes in the urothelium can mimic urothelial carcinoma both clinically and pathologically. This section is intended to provide an overview of the pathologic features of common and unique variants of urinary tract cancers in urinary bladder and to offer a basic approach to distinguish them from benign mimickers.

Based on the growth patterns of the tumor cells, urothelial carcinomas are subdivided into papillary (exophytic papillae with well-defined fibrovascular cores lined by urothelium) and nonpapillary (flat lesions). Papillary lesions—according to the lining urothelium varying from normal to marked atypia—include papilloma, papillary urothelial neoplasm of low malignant potential, and papillary urothelial carcinoma (low grade or high grade; noninvasive or invasive), while nonpapillary/flat lesions include urothelial carcinoma in situ and invasive urothelial carcinoma. Invasive urothelial carcinoma may present as a mass lesion with polypoid, sessile, or fungating ulcerated gross appearance, in which tumor cells breach through basement membrane into the bladder wall as irregular nests, cords, trabeculae, small clusters, or single cells that usually exhibit retraction artifact and paradoxical differentiation, and are separated by a desmoplastic stroma. The tumor cells in these conventional invasive urothelial carcinomas are typically of moderate size to large and have modest amounts of pale to eosinophilic cytoplasm. Nuclear atypia is always obvious, and in some cases, bizarre hyperchromatic nuclei with marked anaplasia may be evident. One of the unique characteristics of urothelial carcinoma is its high propensity for divergent differentiation including glandular, squamous, micropapillary, signet ring cell/plasmacytoid, lymphoepithelioma-like, sarcomatoid, small cell/neuroendocrine, and others. Approximately up to 60% of the tumors exhibit squamous differentiation and 10% of the cases contain foci of glandular features.³  It is important to include this in the pathology report because these features are not only prognostically significant or management indicators of some morphology findings, such as micropapillary and signet ring cell/plasmacytoid, but also helpful knowledge to establish the association of metastasis with the primary tumor in the metastatic sites or a secondary involvement, such as glandular or squamous differentiation.

The morphologic spectrum of the urothelial carcinomas has been expanded to include unusual histologic variants. Some variants, like micropapillary, ring cell/plasmacytoid, or small cell neuroendocrine, are associated with significantly different clinical outcomes, thus the recognition of these variants is important for clinical management. On the other hand, awareness of these unusual patterns is critical to avoid diagnostic misinterpretation. Here we focus on the variants that may pose diagnostic confusion.

Nested variant urothelial carcinoma is a rare variant of urothelial carcinoma, with a reported prevalence of 0.3%.⁴  Grossly, these tumors can be papillary or solid or may manifest as a focus of hemorrhagic area or slight mucosal irregularity or thickening. They most commonly involve the bladder, commonly in the ureteral orifice and trigone. Sole involvement of ureter and renal pelvis has been reported in several case reports.⁵  Nested variant urothelial carcinoma behaves as a high-grade tumor, at least in the bladder, which is associated with a high rate of locally advanced disease at radical cystectomy.^5,6  However, stage for stage, patients with the nested variant have a similar rate of recurrence or adverse survival as those with conventional urothelial carcinoma.^6,7  The tumor is morphologically characterized by irregular nests of urothelial cells in variable sizes and shapes and haphazard distribution, with infiltrating base. In cystectomy specimens, tumor cells infiltrate the muscularis propria with retained nested pattern. The overlying urothelium usually is normal in appearance. The cells comprising this neoplasm show no significant cytologic atypia: mildly pleomorphic and showing slightly increased nuclear to cytoplasmic ratio and occasional prominent nucleoli. Although tumor cells appear histologically bland, some authors^4,6  have observed significant nuclear pleomorphism, particularly within regions of muscle invasion. Mitotic figures are not readily seen.

Owing to their deceptively bland cytologic features, the tumors morphologically resemble florid von Brunn nests or proliferative cystitis, especially in the biopsy materials.⁸  The morphologic evidence helpful to make a distinction is variability of size, shape, and distribution pattern of the nests and the absence or presence of infiltrative border and muscularis propria invasion, if possible. In contrast to those in nested urothelial carcinoma aforementioned, the nests seen in florid von Brunn nests or proliferative cystitis show regular and smooth contour and relatively uniform shape, are evenly spaced, and well confined to the lamina propria, creating a sharp, linear border at the base. On the other hand, tumor cells in nested urothelial carcinomas usually exhibit at least a greater degree of cytologic atypia, usually more obvious in the deeper portion. The presence of muscularis propria invasion is diagnostic for carcinoma, which is regarded as the most definitive distinguishing finding. If equivocal, it is important to correlate the microscopic impression with the clinical findings of the urologist, and additional sampling may be warranted to ascertain the presence of a more aggressive lesion.

Microcystic variant of urothelial carcinoma is another deceptively bland entity, which is characterized by the formation of numerous microcysts. It may lead to the misdiagnosis of cystitis cystica. The cysts range from round to oval to slitlike and may contain secretions. Like nested variant of urothelial carcinoma, the most reliable morphologic feature in distinguishing this variant from benign mimickers is the variation, often dramatic, in size and shape of the epithelial formation and infiltrative haphazard distribution into the wall of the bladder. As the major differential consideration, cystitis cystica tends to have a very organized, confined appearance and lacks the overt variation in size and shape. No apparent biological significance associated with this variant has been observed, except that it may represent a potential diagnostic pitfall, particularly in small biopsy samples.

Micropapillary variant of urothelial carcinoma is a rare but aggressive histologic subtype. The unique clinicopathologic features associated with it are its invariably high stage at initial presentation, and it frequently tends to have lymphovascular invasion and lymph node metastasis. Tumor cells are arranged in small tumor nests or balls aggregated in lacunae or stromal retraction spaces. Tumor nests often show peripherally oriented nuclei with reversed polarity. Among 15 individual morphologic features observed in a study involving 14 genitourinary pathologists, multiple tumor nests in the same lacunar space showed the strongest association with a diagnosis of classic micropapillary variant of urothelial carcinoma.⁹  Although a noninvasive micropapillary growth pattern has been described,¹⁰  it should not be interpreted as micropapillary variant of urothelial carcinoma. Micropapillary growth pattern may mimic lymphovascular invasion. It is important not to overdiagnose the presence of micropapillary component, which is regarded as an ominous finding in invasive urothelial carcinoma regardless of the stage of the disease, thus leading to subsequent aggressive clinical management. If equivocal morphology is present, immunostaining for vascular markers, including CD31, D2-40, and ERG, may provide helpful evidence. Metastasis from carcinoma with micropapillary histology of other sites, including lung, breast, pancreas, or serous carcinoma, has also been reported. Clinical and radiologic correlation is usually required. Identification of admixed classic urothelial carcinoma elements would be necessary for accurate diagnosis. Of note, GATA3 is reportedly the most sensitive marker of micropapillary variant of urothelial carcinomas, which often show a loss of expression of p63.¹¹  In such cases, the best immunohistochemistry panel is combining cytokeratin (CK) 7, CK20, uroplakin III, and GATA3 with other primary specific markers, such as TTF-1 (lung), estrogen receptor and mammaglobin (breast), and WT-1 and PAX8 (ovary), which can aid in differentiating primary urinary tract from metastasis. The largest cohort of micropapillary variant of urothelial carcinomas in 2018 demonstrated that HER2 overexpression and amplification are common genetic alterations.¹² 

Plasmacytoid variant of urothelial carcinoma (PUC) is another histologic variant with an aggressive clinical course, which usually presents at a higher pathologic stage, a higher rate of positive surgical margin, poorer overall survival, chemotherapy resistance, frequent local recurrence, and a peritoneal pattern of spread.¹³  It is combined with signet ring cell and diffuse variants in the 2016 World Health Organization (WHO) classification of tumors of the urinary system and male genital organs.¹⁴  Tumor cells classically exhibit eccentrically located nuclei growing in either diffuse singly scattered and discohesive pattern or loose aggregates forming linear cords with minimal stromal response. These morphologic features raise the differential diagnoses including metastasis or spread from the breast, gastrointestinal (GI) tract, lymphoma, plasmacytoma, or melanoma. In several studies, breast markers, including estrogen receptor and mammaglobin, usually show negativity, but progesterone receptor (4%–13.3%) and gross cystic disease fluid protein 15 (GCDFP15, 24.4%) may show positivity in some cases. Immunostaining for CDX2 is reportedly positive in a small subset of PUCs (15%–17.7%), while nuclear staining of β-catenin is not detected in all 26 reported cases. CD138 immunostain reportedly shows positivity in 83% of cases, which could be another potential pitfall.¹⁵  Therefore, caution should be exercised when these immunohistochemical markers are used to distinguish PUC from metastatic lobular breast carcinoma or GI signet ring cell carcinoma. Loss of E-cadherin expression by immunohistochemistry has been reported in 57% to 70% of PUCs, in contrast to 11% in conventional UCs. CDH1 loss-of-function mutations or promoter hypermethylation is described in greater than 80% of cases and specific to this histologic variant.¹⁶ 

Sarcomatoid urothelial carcinoma is a relatively rare variant with overt epithelial histology admixed with sarcomatous components with or without heterologous differentiations.^3,17  It almost always presents at a high stage, frequently with nodal and/or distant metastasis, and has an ominous prognosis. The sarcomatoid areas most commonly resemble high-grade sarcomas, which may merge with foci of urothelial carcinoma, SCC, adenocarcinoma, or small cell carcinoma. Prior history or coexistence of urothelial carcinoma, invasive or in situ component, is helpful to make a diagnosis of sarcomatoid carcinoma over a primary sarcoma. Major differential diagnosis considerations are benign, or locally aggressive lesions, such as inflammatory myofibroblastic tumor (IMT). A zonal appearance, namely more myxoid and hypocellular regions toward the surface, and greater cellularity with a fibrous background toward the base, the absence of nuclear atypia, and epithelial elements are key features in distinction from a malignant process. A judicious immunohistochemistry panel (pan-cytokeratin, smooth muscle actin, desmin, ALK-1, p63, CK5/6, and high-molecular-weight cytokeratin), in the context of thorough morphologic evaluation, may be helpful in problematic circumstances. Sarcomatoid carcinomas are positive for pan-cytokeratin, p63, CK5/6 and high-molecular-weight cytokeratin, smooth muscle actin, and rarely desmin. IMT can be positive for pan-cytokeratin, smooth muscle actin, desmin, and ALK-1, but is usually negative for p63, CK5/6, and high-molecular-weight cytokeratin. Leiomyosarcomas show positivity for smooth muscle actin, desmin (usually diffuse and strong), and occasionally for pan-cytokeratin (focal to moderate staining extent),¹⁸  but negativity for p63, CK5/6, high-molecular-weight cytokeratin, and ALK-1.¹⁹ 

With the advance in molecular pathology, urothelial carcinoma is characterized as a molecularly diverse disease with heterogeneous clinical outcomes. Variable nomenclature further creates confusion, particularly the terms noninvasive urothelial carcinoma, non–muscle-invasive urothelial carcinoma, and muscle-invasive urothelial carcinoma. From a treatment standpoint, the dichotomy of muscle-invasive versus non–muscle-invasive disease is emphasized by treating physicians. Noninvasive papillary urothelial carcinoma, urothelial carcinoma in situ, and invasive cancers limited to lamina propria are often lumped together as non–muscle-invasive tumors when designing molecular studies. Most notably, TERT promoter mutations have been observed in up to 80% of urothelial carcinomas, regardless of histologic type, grade, and stage.^20–22  Furthermore, TERT genotypes have been shown to be conserved across spatially, temporally, and morphologically distinct components within a tumor, supporting its application as a relatively stable and reliable molecular biomarker for bladder cancers. Therefore, TERT promoter mutations may provide a useful indicator to distinguish urothelial carcinoma from benign mimics or primary versus metastatic tumors with predominant glandular features. Of note, TERT promoter mutations are not detected in enteric-type primary bladder adenocarcinoma.²³  Another notable molecular alteration is FGFR3 mutations. Up to 80% of stage Ta tumors have activating point mutations in FGFR3, which is associated with favorable outcome. In contrast, FGFR3 mutations are less common in tumors with stage T2 or above, which is reported in 10% to 20% of cases. In vitro studies suggest that FGFR3 mutations could contribute to early clonal expansion within the urothelium in vivo. Multiple studies have shown mutations in TP53 and homozygous deletions of CDKN2A on 9p21 may play a role in stage progression of papillary urothelial carcinomas.^24,25  Unlike papillary bladder cancers, most urothelial carcinoma in situ lesions harbor TP53 mutations.²⁶  The Cancer Genome Atlas study of 412 muscle-invasive bladder cancers, using multiple analytic platforms, reveals high overall mutation burden, most of which appears to be passenger mutations without any functional significance.²⁷  The overall mutational load is associated with APOBEC signature mutagenesis. Significantly mutated genes include TP53, KMT2D, KDM6A, ARID1A, PIK3CA, KMT2C, and RB1, which are seen in greater than 17% of cases. FGFR3 mutations are identified in 14% of muscle invasive carcinomas.²⁷  Several molecular classifications have been proposed.²⁸  For example, 5 tumor-cell phenotypes of advanced urothelial carcinoma have been recommended, including urothelial-like, genomically unstable, basal/SCC-like, mesenchymal-like, and small-cell/neuroendocrine-like.²⁹  All of these molecular-based classifications have not yet translated to routine practice.

The spectrum of primary glandular lesions of the bladder is broad and includes benign (cystitis glandularis with or without intestinal metaplasia) and malignant (primary bladder adenocarcinoma) lesions. Although rare, a glandular lesion usually poses a diagnostic challenge to distinguish it from secondary involvement via either direct extension or distant metastasis, owing to their remarkable morphologic, immunophenotypic, and molecular similarities. To avoid misdiagnosis, it is paramount for practicing pathologists to be aware of the full spectrum of differential diagnosis.

Primary adenocarcinoma is a rare entity in urinary bladder, accounting for 0.5% to 2% of all primary bladder malignancies.^30,31  By definition, the tumor exclusively comprises glandular differentiation, without any concurrent or previous noninvasive or invasive conventional urothelial carcinoma. Thus, the diagnosis cannot be rendered in a biopsy specimen and the generous sampling of a resection specimen is required. It usually affects middle-aged to elderly individuals, with a male preponderance. The histologic spectrum of the tumor includes enteric, mucinous, signet ring cell, and mixed subtypes, which represent the morphologic heterogeneity in primary bladder adenocarcinoma.^32,33  As a clinically aggressive tumor, it usually presents with advanced-stage disease at initial presentation with nodal involvement in 30% to 40% of cases. The clinical outcome, therefore, is poor unless detected at a very early stage.^34,35  Owing to the morphologic similarity with primary adenocarcinoma of other sites, such as colorectum, confirming urinary bladder origin of an adenocarcinoma is a significant diagnostic dilemma. There are no site-specific immunohistochemical markers that can aid in the differential diagnosis. Primary urinary bladder adenocarcinoma generally expresses CK7, CK20, and CDX2. Furthermore, the nuclear staining of β-catenin can be seen in a small subset of primary bladder adenocarcinomas. As known, a similar portion of colorectal adenocarcinomas lack nuclear β-catenin accumulation.³⁶  A recent targeted next-generation–based molecular study of a cohort of 15 well-characterized primary bladder adenocarcinomas demonstrates the substantial genomic alteration overlap with colorectal adenocarcinoma.³⁶  In their study, 11 of 15 cases harbor at least 1 genomic alteration in TP53, KRAS, PIK3CA, CTNNB1, APC, TERT, FBXW7, IDH2, and RB1. CTNNB1 and APC mutations are restricted to enteric subtype only. A few studies have investigated molecular alterations, using single gene assay in primary bladder adenocarcinoma, to reveal low frequency of KRAS and TERT promoter mutations in 11.5% (3 of 26) and 28.5% (4 of 14) of primary bladder adenocarcinomas, respectively.^37,38  Therefore, in routine practice clinical exclusion of extravesical primary adenocarcinoma is necessary to establish a diagnosis of primary bladder adenocarcinoma.

Testicular neoplasms are rare solid tumors in men. Among them, testicular germ cell tumor (GCT), accounting for 95% of cases, is the most common solid malignancy affecting males between the ages of 15 and 35 years.³⁹  The spectrum of histologic subtypes, particularly among GCTs, is broad, including seminoma, embryonal carcinoma, yolk sac tumor (YST), choriocarcinoma, and teratoma, and their clinical behaviors are distinctive. More than half of testicular GCTs are composed of more than 1 histologic type.⁴⁰  Therefore, the accurate histologic evaluation and staging is critical to disease management—either surgical treatment (orchiectomy, and/or retroperitoneal lymph node dissection) or chemotherapy if indicated.

Seminoma is the most common subtype of testicular GCT and accounts for about 50% of all GCTs, including 30% to 45% of pure GCTs and 15% to 18% of mixed GCTs.⁴¹  It commonly occurs in patients aged 20 to 49 years. Grossly, seminoma is a solid homogeneous mass, often with lobulated cut surface bulge above the surrounding parenchyma. Typical growth pattern of a seminoma is diffuse arrangement of tumor cells with pale to clear cytoplasm intersected by fibrovascular septa containing small lymphocytes. Other variant morphologies include corded, microcystic, and tubular-like patterns and signet ring cell changes, although these patterns are usually focal findings in otherwise typical tumor morphology. The tumor cells are characterized by crisp cell membranes, polygonal nuclei frequently with flattened edges, fine chromatin, and 1 or more large, centrally located nucleoli. The overall prognosis is excellent: 95% to 98% 5-year survival rate for patients with clinical stage I disease and 83% for patients with clinical stage IIC-IIIC receiving cisplatin-based chemotherapy.^42,43 

Embryonal carcinoma (EC) is the second most common GCT and the most common subtype of nonseminomatous GCTs, occurring in 40% of all GCTs and 87% of nonseminomatous GCTs.⁴⁴  The mean patient age at diagnosis is a decade younger than for patients with seminoma, which is approximately 25 to 30 years. Unlike seminoma, EC usually shows variegated gross appearance with areas of granular, solid, white to tan tumor intermixed with hemorrhagic, necrotic, and cystic foci. The most common architectural patterns are solid, glandular, and papillary. Less commonly, nested, micropapillary, anastomosing, and blastocyst-like patterns are seen. The tumor cells are classically large, with dense amphophilic cytoplasm, indistinct cell borders, and pleomorphic vesicular nuclei with prominent nucleoli. Nuclear overcrowding and overlapping are frequent findings, which are helpful morphologic features, in addition to marked nuclear pleomorphism, to distinguish EC from other GCT subtypes, in particular YST. Clinically, pure ECs behave more aggressively than mixed GCTs and lymphovascular invasion is a frequent finding, with subsequent metastasis to retroperitoneal lymph nodes, mediastinal lymph nodes, lung, and other sites. A study of 479 patients with EC of the testis demonstrated 74% of the patients had metastatic disease at the time of diagnosis, and 50% of these had distant metastases, attesting to the aggressiveness of EC and its tendency to early hematogenous spread.⁴⁵ 

Yolk sac tumor is a malignant GCT that differentiates to resemble extraembryonic structures, including the yolk sac, allantois, and extraembryonic mesenchyme. It almost always presents as one component of mixed GCTs. Pure YST represents only 0.6% to 1.8% of the testicular GCTs.^45,46  One well-known clinical feature is that in 98% of cases the presence of YST elements strongly correlates with elevated serum levels of α-fetoprotein, usually more than 100 ng/mL.⁴⁷  Gross examination of YST shows solid to partially cystic mass, usually with a grayish white to tan cut surface. A wide variety of morphologic patterns has been described in YST, including microcystic/reticular, myxomatous, macrocystic, solid, glandular/alveolar, endodermal sinus/perivascular, hepatoid, papillary, sarcomatoid/spindle cell, parietal, and polyvesicular vitelline. Two morphologic features shared among different architectural patterns are recognized as intracytoplasmic and extracellular hyaline globules and irregular often bandlike intercellular deposits of basement membrane materials. Based on the presence or absence of germ cell neoplasia in situ (GCNIS), it is categorized as postpubertal and prepubertal type, although prepubertal type is rare and usually occurs in young children.⁴⁶  Although it still remains unproven, YST tends to be relatively chemoresistant given that patients with GCTs containing YST elements had poorer outcomes.⁴⁸  Certain morphologic patterns appear to be adverse findings, including glandular, hepatoid, parietal, and sarcomatoid, based on the observation of these patterns disproportionately found in chemoresistant patients and late recurrence.⁴⁸ 

Choriocarcinoma is a less common GCT subtype that differentiates to resemble the trophoblastic cells of the extraembryonic chorion, including cytotrophoblastic, intermediate trophoblastic, and syncytiotrophoblastic cells. Like YST, choriocarcinoma is commonly seen in mixed GCTs (6.4%–17.8%) and pure form is uncommon (0.3% of GCTs).⁴⁹  Serum human chorionic gonadotropin is invariably significantly elevated, often above 50,000 IU/L.⁴⁹  Most choriocarcinomas are hemorrhagic with foci of solid, greyish tan cut surface. Correspondingly, microscopic evaluation classically demonstrates a hemorrhagic background with dual cell populations consisting of mononucleated trophoblasts (small cytotrophoblasts and medium-sized intermediate trophoblasts) and multinucleated syncytiotrophoblasts. Choriocarcinoma is regarded as the most aggressive subtype of GCT, with high propensity for early hematogenous spread, high stage at initial presentation, and hemorrhagic complications. The 3-year survival with choriocarcinoma is only 21% even when treated with combination chemotherapy.⁴⁹  Therefore, it requires early aggressive treatment to improve patients' chance of survival. The most common distant metastatic site is the lung, which prompts radiologic assessment of the lung.⁴⁰ 

Teratoma is composed of several types of tissues representing 1 or more of the germinal layers (endoderm, mesoderm, and ectoderm). Like YST, teratoma is subcategorized as postpubertal and prepubertal type from the presence or absence of GCNIS. Although prepubertal-type teratoma commonly occurs before puberty, it can be seen in postpubertal patients. Age is not considered as a discriminatory feature. Prepubertal-type teratomas do not recur or metastasize, while postpubertal type is malignant, with metastasis seen in 22% to 37% of cases,⁵⁰  and chemoresistant, being the most common residual component in treated GCTs.⁵⁰ 

On macroscopic evaluation, the postpubertal-type teratomas are nodular and firm, and exhibit heterogeneous cut surfaces with solid and cystic areas corresponding to the tissue types (Figure 1, A). Microscopically any type of epithelial or mesenchymal tissue, either well differentiated, mature or immature, embryonic appearing, with varying degree of atypia, can be seen (Figure 1, B). GCNIS is present in the background testicular parenchyma (Figure 1, C and D). The prepubertal-type teratomas could show similar macroscopic findings to postpubertal type (Figure 1, E). However, they are microscopically distinct from postpubertal type: ciliated epithelium, squamous cysts, and smooth muscles particularly prominent, lacking cytologic atypia, and presence of normal surrounding seminiferous tubules (no GCNIS, tubular atrophy, parenchymal scars, tubular microlithiasis, necrosis, or impaired spermatogenesis) (Figure 1, F). The differences between prepubertal and postpubertal teratomas are tabulated in Table 1.

Spontaneous regression of GCTs is a well-recognized phenomenon. In the setting of a diagnosis of retroperitoneal GCT in the absence of a known testicular primary tumor, the recognition of spontaneous regression is critical to render an accurate diagnosis. GCNIS in a scarred testis is the single most specific feature for GCT regression. The presence of GCNIS in combination with testicular scarring and atrophy is accepted as diagnostic criteria of GCT regression, even in the absence of known metastatic GCT. An additional relatively specific but even less sensitive finding for GCT regression is presence of coarse irregular intratubular calcifications. It is important not to misinterpret the smaller, round, often psammoma body–like calcifications within atrophic tubules that characterize microliths. In cases that lack GCNIS, a constellation of morphologic findings after thorough sampling of a testis, including coarse intratubular calcifications, scarring, and atrophy, may permit a greater degree of confidence to diagnosis of a regressed GCT.⁵¹  If only scarring is present, it is paramount to rule out nonneoplastic causes of scarring before rending a diagnosis of regression of GCT. Such nonneoplastic scars usually occur in testes lacking diffuse atrophy, more often are multifocal, may be associated with vascular lesions, such as thrombi, or vasculitis, and consistently lack the association with GCNIS or coarse intratubular calcifications. If the features remain equivocal, the patient deserves thorough clinical evaluation by radiologic methods and serum markers studies. If these prove negative, ongoing follow-up is a more appropriate approach.

Despite the distinctive morphologic subtypes of testicular GCTs, a common genetic thread is the presence of isochromosome 12p, in which there is simultaneous loss of 12q and gain of 12p. The remaining cases contain extra copies of a portion of 12p.^52–54  It is nearly universal in testicular GCTs and considered as an early event. However, the exact mechanisms of 12p gain in testicular GCTs are still unclear. Interestingly, GCNIS rarely exhibits 12p gain, although GCNIS cells are typically aneuploid, with hypertriploid to subtetraploid karyotypes. Some genes on 12p may play a significant role in the development, pluripotency maintenance, and/or progression of testicular GCTs, such as CCND2, KRAS, TNFRSF1A, GLUT3, REA, NANOG, DPPA3, and GDF3.^55–58  Testicular GCTs have ploidy exceeding 2, but no nonseminomatous GCTs demonstrate significantly lower ploidy than seminomas.⁵⁹  Multiple studies have demonstrated a number of molecular alterations. A recent integrated molecular characterization of testicular GCTs using comprehensive assays of genomic, epigenomic, transcriptomic, and proteomic features reveals high aneuploidy and a paucity of somatic mutations in 137 testicular GCTs. Previously unappreciated diversity within seminoma has been identified. Furthermore, somatic mutations of only 3 genes achieve significance—KIT, KRAS, and NRAS—and are exclusively identified in seminoma or seminoma components in mixed GCTs. In the same study, the authors postulate that only seminomas without KIT mutations may be capable of acquiring nonseminomatous histology, because all nonseminomatous GCTs, including mixed GCTs with seminoma components, lacked KIT mutations. More studies are certainly required to draw a definite conclusion.

MicroRNA (miRNA) is an emerging regulator of testicular GCTs. The miRNA expression profiles of testicular GCTs and normal testis tissue, using small RNA sequencing, reveal numerous dysregulated miRNAs.^60,61  Among them, 2 specific clusters of miRNAs, miRNA-371-373 and miRNA-302-367, are stable and can be detected in patients, and have recently emerged as potential serum tumor markers for GCT monitoring.⁶² 

An accurate distinction of the histologic subtypes of testicular GCTs is of great importance for subsequent patient management. For example, seminomas are highly sensitive to both radiation and chemotherapy, whereas nonseminomatous GCTs are resistant to radiation. Additionally, a correct recognition of various histologic elements in a mixed GCT may indicate prognostic value and serve as an important consideration in deciding whether patients who have clinical stage I disease are good candidates for surveillance management protocols. The most common and most clinically significant challenges in testicular pathology are differential diagnoses between seminoma and nonseminomatous GCTs.

The distinction of seminoma from nonseminomatous GCTs is critical owing to the significantly different clinical management. The major differential diagnoses in this category include seminoma, EC, and YST. Although most cases are straightforward and highly rely on the classic architectural and cytologic features of tumor types, some degree of overlapping morphology or some variants may pose diagnostic problems. The classic morphology of seminoma is a tumor with a diffuse arrangement of tumor cells with fibrovascular septa containing small lymphocytes. The tumor cells typically have clear cytoplasm, polygonal nuclei that often show a “squared-off” edge, centrally located large nucleoli, and distinct cell border. In some cases, the nuclei may have diffusely “smudged” appearance, denser, and amphophilic cytoplasm, more irregular and closely packed, which may cause concern for solid pattern EC. The presence of overall architectural pattern of seminoma is helpful to aid in the distinction between seminoma and EC. If doubt persists, immunohistochemical workups can provide objective evidence for making a distinction. Positive staining for both CD30 (membranous staining pattern) and AE1/AE3 cytokeratin and negativity for CD117 strongly support a diagnosis of EC.

Another nonseminomatous GCT that may mimic seminoma is solid variant of YST frequently with clear cytoplasm. A number of morphologic features are helpful to facilitate a diagnosis of YST: the more variable nuclear appearance of solid YST, lack of fibrous septa with lymphocytes inside, presence of intracytoplasmic hyaline globules and bandlike deposits of extracellular basement membrane, and usually coexistence of other architectural patterns. The converse situation is corded nuclei, solid or microcystic pattern of seminoma, which may cause confusion with YST. Close attention should be paid to the cytologic features: uniform, polygonal cells with large nuclei, prominent nucleoli, and squared-off nuclear edges. An immunohistochemical workup panel, including OCT3/4 and glypican 3, in combination with AE1/AE3, can be used to distinguish immunoreactivity for OCT3/4 and negativity for glypican 3 in seminoma, with a reverse staining profile in YST. The different staining extent and pattern of AE1/AE3 is helpful as well: a small subset of seminomas (reportedly 20%–36%) shows focal immunoreactivity for AE1/AE3, which is paranuclear with a dotlike staining pattern, although most seminomas are nonreactive, while YST exhibits diffuse reactivity of AE1/AE3.^63,64 

Lymphovascular invasion is considered an adverse prognostic factor for testicular GCTs. As an indicator of increased risk of clinically occult metastases or relapse, the presence of lymphovascular invasion is a relative contraindication for surveillance management in patients with clinical stage I nonseminomatous GCTs. A helpful clue is to closely search the periphery of the tumor and the tunica albuginea adjacent to the tumor, which are the most common locations of lymphovascular invasion. Considering the retraction artifact of the tumor cells, one should be highly cautious when diagnosing lymphovascular invasion within a tumor. The morphologic features, including a smooth outline, attachment to the endothelium, and admixture with fibrin, are strongly supportive morphologic evidence of a real tumor thrombus. Additionally, positive staining for vascular marker, preferably ERG, can help establish lymphovascular invasion in cases with light microscopic ambiguity.

Adrenal gland is an endocrine organ composed of an outer cortex and an inner medulla, each of which is related to distinct spectrum of entities. The widespread use of imaging has increased detection of incidental adrenal lesions, thus potentially increasing intervention for adrenal masses. Characterization of an adrenal mass as benign or malignant is critical for the clinical management of patients.

Compared to normal adrenal gland, adrenal hyperplasia is characterized as a smooth, diffuse, bilateral enlargement of the adrenal glands, wherein the glands retain the normal adrenal shape. It is not an adrenal neoplasm but rather a physiologic overgrowth of adrenocortical tissue secondary to long-term hormonal stimulation. Adrenal hyperplasia can be further subclassified as macronodular, micronodular, and diffuse hyperplasia on the basis of pathologic assessment. Grossly evident nodularity is usually seen in macronodular hyperplasia at gross evaluation, generally more than 1 cm in size. A significant increase in adrenal gland weight is common, although generally not exceeding 100 g. Micronodular hyperplasia, in contrast, shows multiple nodules with an average diameter of less than 1 cm. Diffuse hyperplasia is characterized by a generalized expansion of the adrenal gland with retention of adrenal normal architecture and absence of macroscopic or microscopic nodularity.

Subtype of hyperplasia may be idiopathic or associated with a heritable syndrome. Primary macronodular adrenal hyperplasia is mostly associated with Cushing syndrome. Other associated syndromes include familial adenomatous polyposis, hereditary leiomyomatosis, and renal cell carcinoma and McCune-Albright syndrome. Up to 90% of micronodular hyperplasia cases present in the setting of Carney complex.

The most common primary adrenal lesion is adrenal cortical adenoma, a benign neoplasm deriving from cortical cells, which may be functioning or nonfunctioning. Functioning adrenal adenomas are often associated with various cortical endocrine syndromes, Conn syndrome (aldosterone secretion), Cushing syndrome (cortisol secretion), and virilization and feminization (sex steroid secretion). Nonfunctional adenomas usually present as incidental findings. Grossly, a conventional adrenal adenoma shows a well-circumscribed solid mass, in most cases surrounded by a grossly evident capsule, with yellow to tan gross appearance secondary to high lipid content. Correspondingly, it is microscopically composed of pale to eosinophilic, lipid-rich, vacuolated cells arranged in nests and cords invested by a thin capillary vasculature. Other less common architectural features include tubular and sinusoidal growth patterns. Lipid-poor cells can be predominant or admixed with lipid-rich cells.

Several rare morphologic variants have been reorganized. Pigmented adenoma is characterized by a brown to black gross appearance and the presence of abundant lipofuscin granules microscopically.^65,66  It is most commonly associated with Cushing syndrome, when functional. As the name implies, myxoid variant shows abundant extracellular mucin and pseudoacinar architecture.⁶⁷  Oncocytic adenoma derives its name from oncocytic microscopic appearance resulting from abundant mitochondrial content.⁶⁸  Despite the distinct gross and microscopic appearance of each variant, no difference in genomic associations or risk of malignant transformation has been observed or reported in these patients. Adrenal adenomas are molecularly characterized by frequent activation of protein kinase A (PKA) signaling and cyclic adenosine monophosphate (cAMP) production, which is similar to adrenal gland hyperplasia.

Adrenal cortical carcinoma is a rare heterogeneous malignancy with a poor prognosis, with an annual incidence of less than 1 per million. Sixty percent to 80% of cases present with a clinical syndrome associated with hormone overproduction. A number of genetic syndromes associated with adrenocortical carcinoma have been recognized, including Li-Fraumeni syndrome, Beckwith-Wiedemann syndrome, familial adenomatous polyposis, Lynch syndrome, and multiple endocrine neoplasia type 1 syndrome. On the other hand, genetic profile and signaling pathways associated with sporadic adrenal cortical carcinoma have been elucidated.

On gross examination, adrenal cortical carcinoma is typically a large mass, with yellow to tan, irregular, and lobulated cut surface, obvious fibrous septations, frequent hemorrhage and necrosis. A background uninvolved adrenal gland is not commonly evident. A spectrum of architectural and cytologic features is recognized in adrenal cortical carcinoma. A low-magnification view usually reveals nests or sheets of tumor cells with pale to eosinophilic cytoplasm separated by fibrous bands. Tumor necrosis and hemorrhage are common findings. Nuclear features ranging from bland to marked pleomorphism can be seen within tumors. Immunohistochemical markers commonly used to establish adrenal cortical origin include inhibin, steroidogenic factor 1 (SF-1), calretinin, and Melan-A. Adrenal cortical carcinoma also shows weak immunoreactivity for cytokeratin and synaptophysin.

Strictly speaking, adrenal cortical lesion must be evaluated for malignant potential. However, assessing the malignant potential of a localized adrenal cortical neoplasm remains challenging. Several multiparametric risk stratification schemes have been generated, including the Hough scoring system, the Weiss scoring system,^69,70  the Van Slooten scoring system, the Weiss modified index, and recently diagnostic algorithms such as the stepwise discriminate diagnostic system and the simplified diagnostic algorithm.^71,72  Among these, the Weiss modified scoring system is the most popular owing to its reliability and relative simplicity. Using this paradigm, adrenal cortical lesion is scored on criteria including cytoplasm (clear cells comprising 25% or less of the tumor cells), presence of more than 5 mitotic figures per 50 high-power fields, atypical mitosis, necrosis, and capsular invasion. Each criterion is scored 0 when absent and 1 when present in the tumor. A total score is calculated by 2× mitotic rate criterion + 2× clear cytoplasm criterion + abnormal mitoses + necrosis + capsular invasion. A score of 3 or more suggests malignant potential.⁷⁰  Despite numerous studies and schemas established to distinguish between adrenal cortical adenoma and carcinoma, practical limitations still remain. Some of the limitations related to Weiss system include its applicability to pediatric adrenocortical neoplasms and relevance of “borderline” lesions with a Weiss score of 2 to 3.

Adrenal cortical lesions composed of predominantly (>90%) eosinophilic cells belong to oncocytic neoplasms, which range from adrenocortical oncocytoma, borderline oncocytic neoplasm of uncertain malignant potential, to oncocytic adrenocortical carcinoma. Owing to the nature of the lesion (predominant oncocytic cell composition and diffuse growth pattern), it would inadvertently predict aggressive behavior in this variant if using Weiss scoring system. Major and minor criteria have been proposed to predict malignant behavior (Lin-Weiss-Bisceglia system), including major criteria such as mitotic rate greater than 5 per 10 high-power fields, atypical mitosis, and venous invasion, and minor criteria such as size greater than 10 cm and/or weight more than 200 g, microscopic necrosis, capsular invasion, and sinusoidal invasion.⁷³  Benign classification using this system is predicted by an absence of all criteria, borderline classification by the presence of at least 1 of the minor criteria, and malignant classification by at least 1 of the major criteria. Another unique morphologic variant of adrenal cortical lesions exhibits the diffuse background Alcian-blue–positive, periodic acid-Schiff–negative matrix, which is defined as myxoid adrenocortical neoplasm. It is often subclassified according to the Weiss criteria. Both oncocytic and myxoid variant of adrenocortical neoplasms mirror the immunoprofile of conventional adrenocortical carcinoma.

A dominant nodule may arise in a background of hyperplasia and mimic an adenoma. Therefore, it may be difficult to distinguish a dominant nodule arising in the background of adrenal cortical hyperplasia from an adrenal adenoma. Making it more challenging, no discriminatory immunohistochemical or molecular markers are available to distinguish these 2 lesions. Usually a clinical and radiologic correlation and careful evaluation of the background adrenal gland can inform the diagnosis, because the identification of microscopic nodularity in the background of adrenal gland favors a dominant nodule in the setting of hyperplasia.

The distinction between adrenal cortical adenoma and carcinoma carries significant clinical implications. For nononcocytic adrenal cortical mass lesions, modified Weiss system has been adopted by pathologists worldwide and arguably has become the standard for the assessment and categorization of these types of adrenal cortical lesions.^70,74  Oncocytic adrenal cortical lesions are categorized as benign, borderline with uncertain malignant potential, and malignant with the Lin-Weiss-Bisceglia system.

Adrenal gland is the fourth most common site of metastasis owing to its rich vascular network, following lung, liver, and bone. Metastasis accounted for approximately 7% of lesions on adrenalectomy.⁷⁵  In a study of 1000 autopsied cases of carcinoma the incidence of adrenal involvement is seen in 27% of cases.⁷⁶  The most common metastatic lesions include those from lung, breast, and GI tract as well as lymphoma and melanoma. Other malignancies, including sarcoma, may be primary or reflect secondary spread from the retroperitoneal soft tissue or vasculature. Renal cell carcinoma, often clear cell renal cell carcinoma, can involve the adrenal gland by metastatic spread or direct extension. Most metastases occur in older populations frequently with prior history of a primary cancer diagnosis at another anatomic site.⁷⁵  Incidental adrenal metastasis as initial manifestation without known malignancy may sometimes cause diagnostic challenges. A high degree of suspicion coupled with careful clinicoradiologic correlation and immunohistochemical workup could prove invaluable in the detection of rare metastatic lesions. Adrenocortical carcinoma is the main differential diagnosis consideration of most metastases. Occasionally metastatic neuroendocrine tumors bring pheochromocytoma into differential diagnoses. Therefore, familiarity with immunomarkers of adrenal origin is helpful and the distinction between a primary adrenal lesion and a metastasis can be made through straightforward immunohistochemical stains for cytokeratin, α-inhibin, calretinin, and SF-1 as well as stains related to sites of origin. Adrenal cortical carcinoma usually has underexpression of epithelial markers, including AE1/AE3, CAM 5.2, and epithelial membrane antigen (EMA), while diffusely and strongly expressing Melan-A, inhibin, SF-1, calretinin, and synaptophysin (Table 2).

Pheochromocytoma (PHEO) is a chromaffin-derived tumor within the adrenal medulla and closely related to the less common extra-adrenal paraganglioma arising from extra-adrenal sympathetic or parasympathetic autonomic paraganglia. An estimated annual incidence is reported as 2 to 8 cases per million.⁷⁷  The tumor occurs most frequently in the fourth and fifth decade of life and most tumors are sporadic, with more than one-third of cases associated with heritable syndromes including MEN2 (RET mutation), von Hippel–Lindau disease (VHL mutation), neurofibromatosis type 1 (NF1 mutation), hereditary paraganglioma syndromes (SDH family mutation), and Sturge-Weber disease (GNAQ mutation). Grossly, PHEO is a solitary well-circumscribed noncapsulated mass, with cut surface ranging from mottled and hemorrhagic to firm, and grey-white in color. Microscopic evaluation of classic PHEO usually shows nests of basophilic cells with immunoreactivity for chromogranin and synaptophysin and surrounding sustentacular cells highlighted by S100 immunostain. According to the update on adrenal tumors in the 2017 WHO classification of endocrine tumors, all PHEOs could have metastatic potential. Presence of distant metastasis at the time of initial surgery is so far the most reliable evidence of malignancy.⁷⁸  Clinical signs and symptoms of patients with malignant PHEO are similar to those of patients with benign disease. Most PHEOs are nonmetastatic, which are not life-threatening and can be successfully cured by surgery; however, approximately 15% to 20% would finally metastasize.^79,80 

Although many relevant indicators such as genetics, histology, and molecular markers have been reported to be related to the metastasis of PHEO, none of them is 100% predictive.^79,80  Various types of prediction systems have been created, including the pheochromocytoma of the adrenal gland scaled score (PASS) system, the grading system for adrenal pheochromocytoma and paraganglioma (GAPP), the composite pheochromocytoma/paraganglioma prognostic score (COPPS) system, and the age, size, extra-adrenal location, secretory type score (ASES). Among these systems, PASS was developed in 2002 and consists only of histopathologic indicators: vascular invasion, capsular invasion, periadrenal fat invasion, diffuse growth/large nests, necrosis, increased cellularity, spindling, cellular monotomy, greater than 3 mitotic figures per 10 high-power fields, marked nuclear pleomorphism, and hyperchromasia. A score of 4 or higher suggests increased malignant behavior. Compared to the PASS system, GAPP was created in 2014 by combining 4 histologic parameters on the basis of PASS (histologic pattern, cellularity, comedonecrosis, and capsular/vascular invasion) and both Ki-67 and catecholamine type, which is the first time the clinical characteristics of the tumor are considered (catecholamine type). Modified GAPP is modified by combining the significant parameters in the GAPP system and the loss of SDHB expression, the first gene mutation incorporated into the scoring system. The COPPS system is based on 3 clinical-histopathologic features (tumor size, necrosis, vascular invasion) and the loss of S100 and SDHB immunostaining to predict the risk of metastasis, with a COPPS score of 3 or more being significantly associated with the occurrence of metastases.⁸¹  ASES score is completely based on 4 clinical characteristics of patients: age, tumor size, extra-adrenal location, and secretory type. There is no unified clinical standard to differentiate metastatic from nonmetastatic disease and a highly effective prediction system is of urgent need.

Although the inherited basis of PHEO has been well characterized, limited somatic profiling has identified mutations at various frequencies in several genes, including EPAS1 (HIF2alpha), RET, VHL, RAS, NF1, and ATRX.⁸²  Some recurrent copy number alterations have been reported.⁸³  A recent multiplatform comprehensive molecular study reveals CSDE1 as a somatically mutated driver gene, complementing 4 known drivers: HRAS, RET, EPAS1, and NF1. Fusion genes in PHEO involving MAML3, BRAF, NGFR, and NF1 have been discovered. Integrated analysis classifies them into 4 molecularly defined groups: a kinase signaling subtype, a pseudohypoxia subtype, a Wnt-altered subtype driven by MAML3 and CSDE1, and a cortical admixture subtype.

All PHEOs could have metastatic potential, which is endorsed in the 2017 WHO classification of endocrine tumors. It is difficult to predict metastasis based on histopathologic features alone, and none of the aforementioned histologic scoring systems reach the level of accurate metastasis prediction. Recent molecular studies on PHEO suggest that the presence of SDH mutations is associated with increased risk of developing aggressive disease by altering intracellular metabolism, especially the tricarboxylic acid cycle.^84–86  In practice, SDHB immunohistochemistry is a reliable tool to identify patients with SDHx mutations.⁸⁷ 

Renal cell carcinoma (RCC) is the most common type of kidney cancer in adults, accounting for approximately 85% of neoplasms arising from the kidney, followed by urothelial carcinoma arising from renal pelvis (5%–10% adult kidney cancers).⁸⁸  Other types of malignancy are rare in the kidney. In this section, we mainly focus on RCCs and discuss other rare entities in the differential diagnosis.

The recent advances on the morphology, immunohistochemistry, genomics, and epidemiology of RCCs contribute to histologic classification. Renal cell carcinomas are recognized as a group of histopathologically and molecularly heterogeneous tumors, with different sets of genetic and epigenetic abnormalities. In accordance with this increased understanding, the histologic subtypes of RCCs have been significantly expanded to 14 types in the most recent WHO classification of renal cell tumors. New emerging and provisional renal neoplasms have been advocated by WHO and The Genitourinary Pathology Society (GUPS), including eosinophilic solid and cystic RCC, RCC with fibromyomatous stroma, and ALK rearrangement–associated RCC, which are validated by multiple independent studies and considered as novel entities.⁸⁹  Pathologists play critical roles in guiding treatment by providing proper tumor classification of kidney lesions. In most cases the diagnosis is generally straightforward by close assessment of the cytologic growth pattern characteristics. However, immunohistochemistry and rarely genetic evaluation may be needed in some challenging cases. Here we provide diagnostic algorithms of RCCs and divide them into different groups with distinct cytologic features and architectural patterns, although overlapping features are frequently seen among different renal tumors.

Many different tumors in the kidney can exhibit clear cytoplasm, including clear cell RCC, papillary RCC, and chromophobe RCC, which account for 90% to 95% of all malignant renal tumors in adults. Among them, clear cell RCC is the most common renal epithelial malignancy and, with the exception of some rare tumors, the most fatal subtype of RCCs. As compared to other subtypes of RCCs, management paradigm is well established for the patients with clear cell RCC. Immune checkpoint inhibitor combinations show significant benefits with respect to progression-free survival and overall survival in several phase 3 randomized controlled trials,^90,91  and are recommended as first-line treatments for patients with untreated advanced clear cell RCC. They have been recently approved by the US Food and Drug Administration. Therefore, accurate recognition of clear cell RCC is paramount not only for prognostication but also for patient management. What is important for pathologists to keep in mind is 2-fold: familiarity with the classic gross and morphologic features of clear cell RCC and its mimickers, and generation of differential diagnosis based on histologic evaluation of the tumors.

Gross appearance of clear cell RCCs is classically described as solid golden yellow masses. Significant variations are common, including areas of necrosis, hemorrhage, and prominent cyst formation. Fleshy and tan areas reflect microscopic sarcomatoid differentiation. Sampling all grossly distinct areas often aids greatly in reaching the correct diagnosis. Microscopic characteristics of clear cell RCC are acinar/alveolar architecture of optically clear tumor cells, completely invested by small-caliber blood vessels. Although tumor cells often show eosinophilic cytoplasm in high-grade areas, this distinctive vasculature usually remains, albeit sometimes focal, regardless of tumor grade. Identification of focal areas of classic morphology of clear cell RCC in high-grade tumors warrants a diagnosis of clear cell RCC, despite coexisting with broad morphologic variety within a tumor. Additional sections from golden yellow areas of the tumor during gross examination are often useful, especially in a large tumor. The diagnosis of clear cell RCC in most cases is straightforward on hematoxylin-eosin–stained slides. Ancillary workup may be used in some difficult cases. However, there is no single marker that is specific for clear cell RCC. Usually a panel of immunohistochemical markers is performed on the basis of the differential diagnosis generated from the microscopic features. As a general rule, it is important to evaluate not only the presence and/or absence of immunoreactivity, but also the pattern and extent of staining (membranous, cytoplasmic or nuclear; diffuse or focal), which is critical to interpret the immunostaining results and eventually aids in distinguishing from the differential diagnosis and arriving at the correct diagnosis.

Papillary RCC is the second most common malignancy of renal epithelial origin, which historically is subdivided into type 1 and type 2 on the basis of cytologic features (basophilic or eosinophilic cytoplasm, nuclear grade, absence or presence of nuclear pseudostratification). The GUPS update on existing WHO entities states that the previous division into type 1 and type 2 papillary RCC is not recommended owing to poor interobserver reproducibility and lack of proven clinical significance,⁹²  although these 2 subtypes are currently listed in the WHO GU tumor classification 2016 edition. A number of molecularly distinctive entities, such as fumarate hydratase–deficient RCC and MiTF family translocation–associated RCC, have been separated from the group of tumors previously classified as “type 2 papillary RCC.” Grossly, papillary RCCs are typically well circumscribed with a well-defined pseudocapsule. The gross color often varies from a dull yellow to red brown. Hemorrhage and cyst formation are often seen to a variable extent. Extensive necrosis is common, but it is not considered as an adverse prognostic factor, in contrast to clear cell RCC and chromophobe RCC. As the name implies, most papillary RCCs comprise papillae, characterized by a central fibrovascular core lined by cuboidal tumor cells with round to oval nuclei, small inconspicuous nuclei, and scant pale basophilic cytoplasm. Although most of the time it is focal, clear cytoplasm can be seen in typical papillary RCC. It is important to note the quality of this clear cytoplasm is reticular or granular,⁹³  in contrast to that seen in clear cell RCC, which is typically described as optically clear.

Chromophobe RCC is the third most common subtype of RCC, with a better prognosis than clear cell RCC. Typical gross features include light brown, well-circumscribed, unencapsulated solid mass. Classic chromophobe RCC consists of 3 cell populations in varying proportions: polygonal cells with pale reticulated cytoplasm and distinct cell border; smaller cells with eosinophilic cytoplasm; and large cells with abundant foamy cytoplasm. When tumor cells with pale cytoplasm become prominent within the tumor, it may cause diagnostic dilemma. Useful morphologic clues to distinguish it from other entities in the differential considerations include cytologic (distinct or “accentuated” cell border, perinuclear halos, and hyperchromatic wrinkled or “raisinoid” nuclei) and architectural (sheets of cells separated by incomplete vascular septae) features.

Clear cell papillary RCC is uncommon, accounting for about 1% of RCCs,⁹⁴  and less aggressive than clear cell RCC, papillary RCC, and chromophobe RCC.⁹⁵  A cohort of 89 clear cell papillary RCCs from Memorial Sloan Kettering Cancer Center (New York, New York) demonstrates several unique epidemiologic characteristics as compared to clear cell RCC and papillary RCC: a higher proportion of females (45%) and African American individuals (19%), increased odds of presenting with additional ipsilateral masses (odds ratio, 4.4) and bilateral diseases (odds ratio, 4.8).⁹⁶  Grossly, clear cell papillary RCCs have capsules and exhibit cystic changes to varying extent. By definition, these tumors morphologically have clear tumor cells with low nuclear grade and the nuclei show apical alignment: usually located toward the apex and away from the basement membrane. Frequently, fibrous capsule and hyalinized or sclerotic stroma are present focally or diffusely. The typical immunoprofile of clear cell papillary RCC is “cuplike” staining distribution of CA-IX and diffuse CK7 positivity. Molecularly, clear cell papillary RCC harbors fewer somatic aberrations and a greater degree of mitochondrial DNA depletion,⁹⁶  in addition to lacking a chromosomal 3p deletion and VHL gene mutations, and no copy number abnormality of chromosomes 7, 17, and Y.^97,98 

MiTF family translocation–associated RCC is a relatively newly recognized entity initially described in pediatric patients in scattered case reports.^99,100  Nevertheless, numerous studies have demonstrated that these neoplasms can occur in adults as well.^101,102  This entity has been included into the 2016 WHO classification under the umbrella term MiTF/TFE family translocation–associated carcinoma, which covers the tumors that have translocations involving TFE3 or TFEB gene. TFE3 gene is located at the Xp11.2 locus and can fuse by translocation to one of numerous partner genes, which results in the overexpression of TFE3. Although the list of fusion genes is growing, the 3 most common Xp11 translocation RCCs are those bearing t(X;1)(p11.2;q21), which fuses the PRCC and TFE3 genes; t(X;17)(p11.2;q25), which fuses the ASPL and TFE3 genes; and t(X;1)(p11.2;p34), which fuses the SFPQ (PSF) and TFE3 genes.¹⁰³ TFEB gene is located at 6p21 locus and tumors with t(6;11)(p12;q12) will have a resulting fusion of TFEB and the alpha gene, ultimately resulting in nuclear expression of TFEB.

Histologically, Xp11 translocation RCCs can mimic almost all subtypes of RCCs.¹⁰⁴  Among the helpful morphologic clues for including it in the differential diagnosis are heterogeneous architectural and cytologic features within a tumor, and tumor cells with voluminous clear to eosinophilic cytoplasm and high nuclear grade. Psammoma bodies are often present. t(6;11) RCCs are classically described as tumors with a biphasic morphology: large clear to eosinophilic cells and small tumor cells with scant cytoplasm with condensed chromatin, which often form nests of rosettes clustered around the basement membrane. As stated in the GUPS update on existing WHO entities, t(6;11) RCC can overlap morphologically with Xp11 translocation and vice versa. When the smaller cell population is limited or absent, the differential diagnosis of t(6;11) RCC is broader and also includes clear cell RCC and various oncocytic renal neoplasms.⁹² 

More recently, RCCs with TFEB amplification have been identified and appear to be associated with a more aggressive clinical course than TFEB translocation–associated RCCs. To date, 39 cases of high-level amplification of TFEB in RCCs have been reported.^105–108  Morphologically, TFEB-amplified RCC frequently shows nests of high-grade epithelioid cells with eosinophilic cytoplasm associated with pseudopapillary formation and necrosis, or true papillary formations.

TFE3 and TFEB immunohistochemical staining are reported to be sensitive and specific ancillary studies as long as diffuse and strong staining is present. However, these 2 immunostains are not widely available. A panel of surrogate markers can be useful in distinguishing these 2 entities from others among the differential diagnosis, in particular clear cell RCC, which includes 1 or 2 epithelial markers, melanocytic markers (HMB-45 and Melan-A), and cathepsin K. Translocation-associated RCC frequently shows negativity or patchy positivity for cytokeratin and epithelial membrane antigen. TFE3 or TFEB gene rearrangement demonstrated by fluorescence in situ hybridization (FISH) or gene fusion products detected by sequencing techniques are the gold standard for establishing the diagnosis. A most recent study highlights that TRIM63 can serve as a sensitive and specific biomarker for MiT family aberration–associated RCCs, including TFEB-amplified RCCs. A combination of TRIM63 RNA in situ hybridization (ISH) and TFE3/TFEB FISH assays would improve the accuracy and efficiency of MiTF RCC diagnosis, especially because TRIM63 RNA ISH is strongly positive in TFE3 FISH false-negative cases with RBM10-TFE3 inversion.¹⁰⁹ 

In addition to RCCs, other renal lesions with clear tumor cells are also considered in the differential diagnosis, especially epithelioid angiomyolipoma. Although areas of classic triphasic angiomyolipoma (mature adipose tissue, dysmorphologic vessels, and spindle cells) often coexist with epithelioid component—which significantly aids in the distinction from other differential diagnoses—in cases lacking any typical morphologic features a panel that includes PAX8, cytokeratin, and EMA, in conjunction with melanocytic markers (HMB-45, Melan-A), cathepsin K, and smooth muscle actin is generally helpful.

All the entities listed in renal tumor with clear cytoplasm can include a varying proportion of tumor cells showing eosinophilic cytoplasm, in particular high-grade tumors. Thoroughly examining a tumor for the classic morphologic features aids in differentiating these tumors. In this section, oncocytoma, fumarate hydratase (FH)–deficient, and succinate dehydrogenase (SDH)–deficient RCC will be discussed in greater detail.

Oncocytoma is a benign renal epithelial neoplasm. Grossly, it is a well-defined mahogany brown to tan-yellow tumor. Microscopically, the tumor is composed of variably sized solid nests of tumor cells, with round to ovoid nuclei and dense chromatin (Figure 2, A). Amin et al¹¹⁰  described a variety of architectural patterns including archipelaginous, tubulocystic, trabecular, and with scarring and haloes, and cytologic features including prominent nucleoli, degenerative pleomorphism, and oncoblasts.¹¹⁰  Fat invasion without stromal reaction is allowable in oncocytomas.

FH-deficient RCC usually presents at an advanced stage at initial presentation and is clinically aggressive. It grossly exhibits a bubble-wrap appearance. Histologically, the tumor displays a variety of growth patterns, with papillary being the most frequent (Figure 2, B), followed by cribriform growth.¹¹¹  Cytologically, tumor cells show eosinophilic cytoplasm with inclusion-like nucleoli (Figure 2, C). FH protein loss can be demonstrated by immunohistochemistry. However, the sensitivity of immunostain can be reduced owing to missense mutations in FH gene. In this setting, S-(2-succino) cysteine (2SC) immunostain can be used as a surrogate marker (Figure 2, D).^112,113  Molecular analysis of FH gene alterations is the key to establish the final diagnosis and to determine germline or sporadic mutations.¹¹⁴ 

SDH-deficient RCCs have germline mutations in the SDH complex gene. Loss of SDHB protein expression via immunohistochemistry serves as a marker for dysfunction of SDH complex. Grossly, SDH-deficient RCCs are cystic, and bilateral lesions can be seen in 26% of cases.¹¹⁵  The histologic hallmark is flocculent cytoplasmic vacuoles in the background of pale eosinophilic cytoplasm (Figure 2, E and F), imparting a wispy or bubbly appearance.^116,117 

Prostate adenocarcinoma is the most common malignancy involving the prostate. Prostate cancer histology, especially Gleason pattern, plays a critical role in clinical management to identify patients for active surveillance or more aggressive therapy. Attempts have been made to refine histologic classification and reporting in prostate cancer to facilitate patient risk stratification.

Although the Gleason score system has maintained the original concept of using architectural features to stratify morphologic patterns associated with prognosis for more than 5 decades since its introduction,¹¹⁸  significant modifications have been incorporated to refine the risk stratification and widely adopted after consensus meetings organized by the International Society of Urologic Pathology (ISUP) in 2005 and 2014.^119–122  Notably, Gleason patterns 1 and 2 have been abolished in needle biopsies and are now rarely reported in resection specimens; Gleason pattern 4 includes poorly formed glands, glomeruloid and cribriform glands, and it becomes the most heterogeneous group associated with diverse clinical outcomes. From a study of more than 20,000 prostate cancer cases treated with prostatectomy and greater than 5000 cases treated with radiation therapy, Epstein and his colleagues¹²⁰  from Johns Hopkins Hospital propose a 5-tiered contemporary grading system, namely group 1 (Gleason score ≤6), group 2 (Gleason score 3 + 4), group 3 (Gleason score 4 + 3), group 4 (Gleason score 8), and group 5 (Gleason scores 9 and 10).¹²⁰  This new grading system shows more accurate grade stratification than the current Gleason system.^123,124  Subsequently, the new system was endorsed by ISUP¹²¹  and WHO in 2016 with the recommendation that pathology reports use the new system and the modified Gleason score in conjunction, till it becomes widely accepted and practiced. Later, some of the unresolved questions and emerging challenges were addressed in ISUP and GUPS proceedings papers on the grading of prostate cancer in 2019, including advanced imaging, molecular diagnostics, and artificial intelligence.^125–127 

Based on the 2014 ISUP conference and 2016 WHO classification, the percentage of Gleason pattern 4 should be reported in tumors with grade groups 2 and 3 in needle biopsy and resection specimens.¹²⁰  The major rationale for this recommendation in needle biopsy is the potential clinical decision for active surveillance. The National Comprehensive Cancer Network guideline now considers active surveillance for select patients with favorable intermediate-risk prostate cancer, especially if extent of Gleason pattern 4 is less than 10%.^128,129  Accumulating evidence further supports this recommendation.^130–132  In addition, reporting percentage of Gleason pattern 4 in needle biopsy with grade group 3 tumor improves prediction of the probability of higher-grade tumor in prostatectomy specimens, which may be beneficial for patient counseling and treatment decisions.^131,133  For example, a grade group 3 tumor with 60% of Gleason pattern 4 on a biopsy specimen has a relatively higher chance of reflecting a grade group 2 or 3 tumor in a resection specimen, whereas 90% of Gleason pattern 4 seen in grade group 3 tumor is more likely to represent a grade group 3 or higher tumor in a prostatectomy specimen.

Invasive adenocarcinoma glands with poorly formed or fused lumens, cribriform or glomeruloid architecture, and ductal features are regarded as Gleason pattern 4. The most common Gleason pattern 4 seen in needle biopsies are poorly formed glands (57%), followed by fused glands (53%) and cribriform glands (25%).¹³⁴  Among them, cribriform pattern is described as glands composed of sheets of tumor cells that form cohesive rounded or irregularly shaped trabeculae with perforations or multiple “punched out” lumina. The presence of cribriform pattern on needle biopsy is strongly associated with upgrading and upstaging,¹³³  with more advanced pathologic stage on prostatectomy, independent of grade.¹³⁵  Thus, it suggests the patients who have cribriform pattern in their tumor may not be good candidates for active surveillance. When cribriform pattern is seen on prostatectomy, it is an adverse independent predictor for distant metastases-free survival and disease-specific survival, compared to fused, poorly formed, and glomeruloid patterns.^136,137  Although it is not an official recommendation in 2014 ISUP consensus and 2016 WHO classification, it became increasingly common practice to include whether cribriform pattern is present in Gleason pattern 4 both on biopsy and prostatectomy.¹²⁶ 

Although intraductal carcinoma of the prostate (IDCP) is recognized as a new entity in the 2016 WHO classification of GU tumors, the debate on whether to incorporate it into grading still continues.¹³⁸  The current major disagreement between ISUP and GUPS regards the issue of IDCP: GUPS recommends not including IDCP in determining the final Gleason score,¹²⁶  whereas ISUP group reaches an agreement to incorporate IDCP into the Gleason score if IDCP is admixed with invasive carcinoma.¹²⁵  Similarly, the pathogenesis remains to be determined. One of the theories is that it is conceived as the intraductal spread of invasive aggressive carcinomas, rather than a preinvasive lesion.

By definition, intraductal carcinoma is a tumor with malignant epithelial cells filling large acini and prostatic ducts with preservation of basal cell layer, and either solid or dense cribriform pattern (>50% of lumen filled by epithelial cells) or loose cribriform or micropapillary patterns if associated with marked nuclear atypia (nuclear size 6× normal) or comedonecrosis.¹³⁹  If only loose cribriform and micropapillary patterns are seen, they must be distinguished from high-grade prostatic intraepithelial neoplasia (HGPIN) and atypical intraductal proliferation (AIP).¹²⁶  HGPIN glands typically show smooth rounded contours and are similar in size to the adjacent benign glands, with slightly enlarged nuclei (2 to 3 times the size of the benign glands), whereas AIP exhibits a greater degree of architectural complexity and/or cytologic atypia than HGPIN, yet falls short of the diagnostic criteria for IDCP. The most common diagnostic scenarios for AIP are listed as follows: loose cribriform architecture without significant nuclear pleomorphism or comedonecrosis, solid or dense cribriform structure partially spanning the glandular lumen, or any architecture with significant nuclear atypia or pleomorphism beyond HGPIN, but not fulfilling the current diagnostic criteria for IDCP.

Accumulating evidence demonstrates intraductal carcinomas convey independent prognostic value. Most patients with intraductal carcinoma associated with invasive adenocarcinoma Gleason 6 on needle biopsy have unfavorable prognosis such as metastasis at diagnosis, progression during active surveillance, or high-grade and high-stage tumor at prostatectomy. In a series of 21 pure intraductal carcinomas seen on prostate biopsy, with subsequent radical prostatectomy information, only 2 cases (10%) had no invasive prostatic carcinoma and tumors with Gleason scores of 8 and above were present in most cases.¹⁴⁰  Thus, when IDCP is seen in the absence of high-grade invasive cancer in needle biopsy, a comment is recommended to be added in the pathology report on the invariable association with aggressive invasive carcinoma according to the 2014 ISUP consensus.¹²¹  Intraductal carcinoma with concomitant invasive adenocarcinoma in needle biopsies is also associated with decreased cancer-free survival, even in the subgroup of patients with Gleason scores of 8 and above.¹⁴¹  In patients with intermediate-risk disease, treated by radical prostatectomy or radiotherapy, the presence of intraductal carcinomas is the predictor for early biochemical recurrence and metastasis.¹⁴²  In addition, intraductal carcinoma present in radical prostatectomy specimens has also been reported to be an independent predictor of biochemical recurrence^143,144  and cancer-specific survival.¹⁴³  Given these studies the presence of intraductal carcinoma should be reported in any prostatic specimen: needle biopsy, transurethral resection, and prostatectomy, which is recommended by ISUP and GUPS groups.

When dense cribriform or solid growth patterns are present, the major differential diagnoses of invasive acinar adenocarcinoma and urothelial carcinoma spreading into prostatic duct should be included in the differential diagnosis, in addition to intraductal carcinoma of the prostate.¹⁴⁵  An immunohistochemistry panel can often aid in resolving the diagnostic ambiguity. Both invasive and intraductal carcinoma of the prostate are positive for prostate-specific markers, including prostate specific antigen (PSA), prostate-specific acid phosphatase, prostate-specific membrane antigen, and NKX3.1. With the aid of basal cell markers, including CK5/6, 34βE12, and p63, basal cells at the periphery are highlighted in intraductal carcinoma, while absent in invasive carcinoma. In contrast, urothelial carcinoma is negative for prostate-specific markers, but positive for GATA3, p63, and 34βE12.

Penile lesions are not common specimens received by pathology departments. The rarity of these conditions and the low incidence of penile cancer contribute to the unique challenge of diagnosing premalignant penile lesions from benign penile and invasive carcinomas. This section will provide an overview of common benign, premalignant lesions and invasive malignancies, with emphasis on the current classification and histologic diagnosis, and diagnostic problems and pitfalls.

Penile intraepithelial neoplasia (PeIN) is regarded as an intraepithelial precursor lesion of invasive SCC, which is further subclassified into differentiated and undifferentiated types. Differentiated PeIN shows thickened epithelium with hyperkeratosis, parakeratosis, and hypergranulosis, elongated and anastomosing rete ridges, with subtle abnormal maturation (enlarged keratinocytes with abundant eosinophilic cytoplasm) and keratin pearl formation. Undifferentiated PeIN includes basaloid (epithelium replaced by a monotonous population of small to intermediate-sized blue cells with a high nuclear to cytoplasmic ratio), warty (thickened epithelium with an undulating and spiking surface and striking cellular pleomorphism), and warty-basaloid (characterized by overlapping features of both warty and basaloid types).

There is a significant association of the different types of PeIN with specific invasive SCC variants. Differentiated PeIN is seen preferentially associated with usual, papillary, pseudohyperplastic, verrucous, and sarcomatoid carcinomas. Undifferentiated PeIN is distinctively associated with warty, basaloid, and mixed warty-basaloid carcinomas.

Most malignant tumors of the penis are SCCs originating in the inner mucosal lining of the glans, coronal sulcus, or foreskin. Multiple histologic subtypes have been described, with conventional type being the most common (accounting for 50% of penile carcinomas). Compared to previous exclusively morphology-based classification schemes, the 2016 WHO classification presents a new classification based on clinicopathologic properties and relation to human papillomavirus (HPV) infection (Table 3), each with distinctive clinicopathologic and outcome features.

HPV-related penile SCCs include basaloid, warty, warty-basaloid, papillary-basaloid, clear cell, and lymphoepithelioma-like subtypes. As the most common HPV-related penile carcinoma, basaloid subtype accounts for 5% to 10% of cases, manifests in patients 10 years younger than those with usual SCC, and is an aggressive tumor, with approximately half of patients presenting with regional nodal metastasis and with mortality of 20% to 30%. In contrast, warty (condylomatous) carcinoma shows low mortality, with local recurrence in 17% to 18% of cases.

Non–HPV-related penile SCCs cover usual type, verrucous carcinoma, papillary carcinoma, pseudoglandular carcinoma, sarcomatoid squamous cell carcinoma, adenosquamous carcinoma, and carcinoma cuniculatum. Among them, verrucous carcinoma, papillary carcinoma, and carcinoma cuniculatum are less aggressive than usual-type SCC. In particular, verrucous carcinoma is locally aggressive but biologically indolent. No metastases are reported with pure verrucous carcinoma. Standard recommended treatment is complete local excision with clear margins, or partial or total penectomy.

Pseudoglandular and sarcomatoid SCCs are considered high-grade carcinomas with prominent acantholysis and pseudoglandular features, accounting for 1% to 2% of penile SCCs. Most pseudoglandular carcinomas invade into the corpora cavernosa, with regional metastasis rate of 42% and mortality rate of 29%, higher than usual SCC.¹⁴⁶  In sarcomatoid carcinoma, lymphovascular and perineural invasion are common, with early lymph node metastasis and distant metastasis (eg, lung, skin, bone, pleura). Sarcomatoid SCC is the most aggressive carcinoma of all penile carcinomas, with high mortality (45%–75%).

Squamous hyperplasia is a common penile lesion that can mimic squamous neoplastic processes, especially well-differentiated SCC. Microscopically, squamous hyperplasia is a flat lesion, although verrucous, papillary, and pseudoepitheliomatous appearance can be rarely seen. Careful examination of squamous maturation and cytologic atypia are helpful to distinguish hyperplasia from PeIN and other well-differentiated SCCs. A constellation of morphologic findings, including orderly squamous maturation and no atypical cells in the background of hyperkeratosis and hypergranulosis, help to render a diagnosis of hyperplasia. In contrast, PeIN lesions often show parakeratosis, with alteration of the squamous maturation and more prominent cytologic atypia even sometimes limited to the basal layers.

Distinguishing verrucous hyperplasia from verrucous carcinoma usually poses a diagnostic challenge, especially in small biopsy specimens, owing to their identical morphology under the microscope, other than the focal localization, small size, and often subclinical presentation of verrucous hyperplasia. In the event of microscopic assessment insufficient to make a distinction, resection of the entire lesion with adequate surgical margins would be advised to the urologist, rather than repeating the biopsy.

Another challenging diagnostic problem in the surgical pathology of penile lesions is the distinction of pseudoepitheliomatous hyperplasia from pseudohyperplastic carcinoma, which often requires a wide resection or even circumcision or penectomy for definite diagnosis. Thorough examination of the morphologic features is the key: in pseudohyperplastic carcinomas the epithelial nests usually vary in size and shape, with inconspicuous peripheral palisading, often surrounded by reactive stroma. Intraepithelial squamous pearl formation is more typically seen in carcinoma and rarely found in hyperplasia. Although the final diagnosis is based on pathologic findings, clinical information offers helpful clues, such as older age (>70 years), presence of lichen sclerosis, multicentricity, and foreskin location favoring pseudohyperplastic carcinomas.

Owing to their ominous prognosis, it is critical to recognize basaloid, sarcomatoid, and pseudoglandular subtypes of SCC. The variegated morphologic spectrum of basaloid SCC is often underrecognized because it was originally described as tumor nests growth pattern with uniform, monotonous, small basophilic tumors cells, with or without central comedonecrosis.¹⁴⁷  However, subsequent studies showed that neoplastic cells of basaloid subtype of SCC can be larger, pleomorphic, or spindle shaped, with retaining basophilic/amphophilic cytoplasm. A papillary variant resembling urothelial carcinoma has been reported.¹⁴⁸ 

Sarcomatoid SCC is a rare, but highly aggressive cancer. Differential diagnoses include true sarcoma and melanoma. The presence of foci of an otherwise conventional SCC or adjacent PeIN is usually diagnostic for sarcomatoid SCC. Immunohistochemical panel, including epithelial, smooth muscle, and melanocytic markers, could be helpful in problematic cases.

Pseudoglandular carcinoma is an aggressive variant of penile SCCs. It must be distinguished from penile mixed squamous-glandular or pure glandular neoplasms, such as mucoepidermoid, urothelial carcinoma with glandular differentiation, and adenocarcinoma of the Littre glands, and angiosarcomatoid variant of sarcomatoid carcinoma. The morphologic hallmark of pseudoglandular carcinomas is the presence of pseudoglands, described as an open space lined by high-grade atypical flat, cuboidal, or cylindrical cells. This space usually contains isolated necrotic cells floating within the lumen. In contrast, in true glandular or mixed squamous glandular neoplasms of the penis, the glandular lumina are either clear empty or filled with mucinous material.