Section 7 -

Spindle Cell Proliferations of the Bladder Mahul B. Amin, MD Jesse K. McKenney, MD

Case 8 History: 36-year-old man with microscopic hematuria. Imaging studies revealed a 4 cm intramural urinary bladder mass. Diagnosis: Inflammatory Myofibroblastic Tumor/ Pseudosarcomatous Myofibroblastic Proliferation Case 9 History: 82-year-old woman who presented with gross hematuria. Cystoscopy revealed a 5 cm fungating mass in the trigone/posterior wall. Diagnosis: Sarcomatoid Carcinoma Spindle Cell Proliferations of the Bladder Spindle cell proliferations of the urinary bladder are uncommon, but may cause significant diagnostic difficulty due to the degree of morphologic overlap between clinically benign and malignant lesions. These difficulties may be amplified in small biopsies because some of the more specific diagnostic features may not be present for evaluation. In addition, the number of different diagnostic terms applied to the same entity has added confusion to this diagnostic area. We discuss the nomenclature, morphologic criteria, and immunohistochemical features used to classify spindle cell proliferations occurring in the urinary bladder including those with myofibroblastic, smooth muscle, skeletal muscle, epithelial (sarcomatoid urothelial carcinoma), fibroblastic, and neural differentiation. A separate discussion of five challenging differential diagnostic scenarios is also presented. Myofibroblastic proliferations Smooth muscle neoplasms Skeletal muscle neoplasms Sarcomatoid carcinoma Other mesenchymal tumors of the urinary bladder I. Myofibroblastic Proliferations Introduction Myofibroblastic proliferations of the urinary bladder have been described under a variety of names including postoperative spindle cell nodule, inflammatory pseudotumor, pseudosarcomatous fibromyxoid tumor, pseudosarcomatous myofibroblastic proliferation, pseudosarcomatous spindle cell proliferation, and inflammatory myofibroblastic tumor. [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 13] These myofibroblastic proliferations have historically been divided into three diagnostic categories: those associated with a prior history of bladder instrumentation (i.e., postoperative spindle cell nodule), those associated with overlying urothelial carcinoma, and de novo lesions with no known clinical association. It should be emphasized that, for practical purposes, these myofibroblastic lesions are morphologically indistinguishable regardless of the clinical setting, a finding that has led some authors to combine them under a single diagnostic term "pseudosarcomatous myofibroblastic proliferation (PMP)" [5] or "inflammatory myofibroblastic tumor". [49] There is currently no consensus on nomenclature for these myofibroblastic lesions, but some evidence does suggest that they are similar to proliferative lesions outside the bladder classified as "inflammatory myofibroblastic tumor" (IMT), a term that many genitourinary pathologists are adopting. [1, 6] Future consensus statements may combine all bladder myofibroblastic lesions into a single diagnostic entity, but for descriptive purposes, we include a discussion of the three different clinical settings in which they are encountered. Most importantly, no matter what diagnostic nomenclature is employed, these myofibroblastic proliferations can be treated with conservative excision as discussed below. For this discussion, I will use the combined term IMT/PMP when referring to the bladder lesions. Morphology The morphology of myofibroblastic proliferations can be described based on the cytology of the individual cells and the low power architecture. The following morphologic description is based on studies of IMT from varying sites, [46] but is applicable to bladder lesions. Cytologically, the individual cells may appear spindled or stellate. The ends of the cells are usually tapered with elongated cytoplasmic processes identifiable at high power examination. The cytoplasm varies from eosinophilic to more amphophilic. The nuclei are also tapered, have bland, non-condensed chromatin, and may show prominent nucleoli. Admixed ganglion-like cells, identical to those seen in proliferative fasciitis, are also rarely reported. The low power appearance/architectural arrangement of the cells in IMT has been described as three separate patterns that may be intermixed in a given case: myxoid, compact spindle cells, and sclerotic/hyalinized. [1] The myxoid pattern, which is by far the most common pattern in the bladder, has a very loose, disorganized appearance similar to that seen in nodular fasciitis. There is often an associated mixed chronic inflammatory infiltrate and an irregular network of poorly organized small blood vessels. The individual cells have features of myofibroblasts as described above, but prominent eosinophilic cytoplasm is common. The compact spindle cell proliferation is more cellular with a better-developed fascicular architecture. This pattern is the prototypical histology originally described in the lung. The fascicles are not as tightly formed and do not show the precise, acute intersecting fascicles typical of a sarcoma (fibrosarcoma, synovial sarcoma, etc.). This pattern also characteristically shows aggregates of plasma cells scattered among the spindle cells as well as nodules of small lymphocytes sometimes with germinal centers. The sclerotic pattern, as the name implies, is less cellular with extensive areas of dense collagen deposition. Tumors with any of these patterns typically show increased mitotic activity and necrosis of the lesional cells. 1) Postoperative Spindle Cell Nodule [7] By definition, these proliferations occur after trauma, typically following bladder instrumentation. The interval of time between the procedure and the development of a proliferation is typically a few weeks to months; however, some authors have suggested that rare cases may occur a year or more after the initial insult. These post-instrumentation proliferations are usually small and rarely exceed 1 cm. At least one well-documented post-instrumentation myofibroblastic proliferation of the bladder has been shown to have an ALK gene alteration by FISH, a finding that may add support to the theory that all bladder myofibroblastic lesions represent a single entity. [49] 2) Inflammatory Myofibroblastic Tumor/Pseudosarcomatous Myofibroblastic proliferation (IMT/PMP) [2, 3, 4, 5, 6, 47, 49, 8, 9] Although individual cases of IMT/PMP not related to instrumentation are indistinguishable from those with known trauma (postoperative spindle cell nodules), they are typically more myxoid, reach a larger size (usually 2-8 cm, but over 30 cm reported), and may have somewhat more variation in the size of the cells. Recent studies of vesical IMT/PMP have documented both ALK 1 (anaplastic lymphoma kinase) expression by immunohistochemistry [5, 6, 47, 49] and ALK gene rearrangements by FISH analysis [6, 49], findings typical of IMT in other anatomic sites. The reported incidence of ALK 1 expression by immunohistochemistry in bladder IMT/PMP has been as high as 89%. [6] Freeman et al. confirmed ALK gene rearrangement by FISH in 4 of 6 cases demonstrating ALK immunoreactivity. [6 ] More recently, Montgomery et al. documented ALK gene rearrangements in 13/18 tested cases. [49] Unfortunately, ALK expression is not entirely specific for IMT and is reported in other mesenchymal neoplasms as well (e.g., up to 20% of rhabdomyosarcomas). [11] As in all diagnostic immunohistochemistry, close morphologic correlation is essential when interpreting ALK immunostains. A low rate of recurrence for IMT/PMP (up to 25%) has been reported in some series that included all anatomic sites. [1] The risk of metastases from IMT (all anatomic sites) is controversial because it has been difficult to distinguish metastases from multifocal, non-contiguous disease; however, the risk seems to be very small (reportedly less than 5%, but probably lower). The recurrence rate for lesions specifically in the urinary bladder has ranged from 0-25% [5, 6, 47, 49, 9, 13] and no metastases have been reported. One patient death has been reported from urinary obstruction due to local mass effect of an unresected vesical IMT. [9] As noted in the introduction, there is controversy regarding diagnostic terminology for these lesions, which can be confirmed by a survey of the nomenclature employed in current textbooks. Some authors suggest that vesical myofibroblastic proliferations are distinct from IMT at other sites. Whether the presence or absence of ALK reactivity has any clinical significance or defines subsets of myofibroblastic lesions in the bladder remains to be determined. [48] Under current clinical management standards, however, this is mostly a semantic argument because histologically benign appearing myofibroblastic proliferations are followed clinically after a simple excision procedure. The clinically relevant decision is their distinction from malignant spindle cell lesions (e.g., leiomyosarcoma or sarcomatoid carcinoma) in which a more radical excision procedure is indicated. This discussion excludes IMT on the serosal surface of the bladder, a proliferation that should be classified as intraperitoneal, not a bladder primary lesion. Omental/mesenteric/peritoneal IMTs [1] are more typical in children, are often multifocal, and are associated with systemic symptoms including fever, weight loss, anemia, thrombocytosis, increased erythrocyte sedimentation rate, and increased gamma globulins. The frequency of these generalized symptoms in patients with bladder IMT/PMP is not well documented. 3) Pseudosarcomatous Stromal Reaction Associated with Neoplasia [49, 10, 14] Myofibroblastic proliferations have been rarely described in association with an overlying carcinoma. I have personally seen an increasing number of these cases in consultation, so this may an underreported finding. The main differential diagnosis in this scenario is a sarcomatoid carcinoma and distinguishing features are described below in the differential section. Sometimes, the distinction from muscularis propria may also be questioned. Rarely, myofibroblastic proliferations are associated with sarcomatoid carcinoma. [49] Immunohistochemistry The immunophenotype of myofibroblastic proliferations in the urinary bladder is somewhat different than that reported at other anatomic sites (summarized below). Most notably, cytokeratin reactivity is frequently encountered in the bladder. Although h-caldesmon was initially touted as a specific marker of smooth muscle, [12] h-caldesmon immunoreactivity has been reported in 66 and 57% of vesical IMT/PMP in two separate studies. [6, 49] Immunohistochemistry in Inflammatory Myofibroblastic Tumor ALK1 +/-* Actin +/- Desmin -/+ ■ Cytokeratin +/-^ EMA - Myogenin - h-Caldesmon +/-» *Immunoreactivity for ALK1 is variable with reports of 8-89% ■ Marked variation in desmin reactivity between studies ^Reports of cytokeratin reactivity have ranged from 36-85% » Recent studies have reported immunoreactivity in 57 and 66% II. Smooth Muscle Neoplasms True smooth muscle neoplasms of the bladder are rare, but are divided into two categories: clinically/histologically benign (leiomyoma) and malignant (leiomyosarcoma). 1) Leiomyoma [15, 16, 17] Leiomyomas of the bladder have the same morphologic features as in other anatomic locations and are characterized by intersecting fascicles of spindle cells with abundant eosinophilic cytoplasm. The cytoplasm is generally tapered and the nuclei are oval or elongated with blunted ends. By definition, no significant nuclear atypia (hyperchromasia, pleomorphism, irregular nuclear membranes) is allowable for a diagnosis of leiomyoma. In addition, the lesions are defined as well circumscribed with no irregular infiltration into the surrounding muscularis propria. Coagulative tumor cell necrosis is not found when these strict rules are followed. To our knowledge, myxoid change has not been described in vesical leiomyomas. 2) Leiomyosarcoma [17, 18, 19, 20] Leiomyosarcomas (LMS) are diagnosed by the criteria typically applied to true smooth muscle neoplasms of deep soft tissue, not uterine smooth muscle tumors. The typical morphologic appearance is a cellular proliferation of spindle cells with prominent eosinophilic cytoplasm arranged into well-formed, intersecting fascicles. Stromal myxoid change is frequently present, but epithelioid variants are exceedingly rare in the bladder. Under the published criteria, any one of the following findings is sufficient for a diagnosis of malignancy in a true smooth muscle neoplasm of the bladder: significant nuclear pleomorphism with hyperchromasia and irregular nuclear membranes (usually readily identifiable at low power), coagulative tumor cell necrosis, and infiltration of the muscularis propria. As discussed above (see leiomyoma), the significance of low mitotic activity in the absence of other features of malignancy is uncertain. In the urinary bladder, there is no documented counterpart to the atypical (symplastic) leiomyoma of the uterus; therefore, marked nuclear atypia alone (hyperchromasia, pleomorphism, irregular nuclear membranes) is sufficient for a diagnosis of LMS under current published criteria. Studies have used slightly different criteria for separating tumors into low and high grade groups, [17, 19] but it is clear that tumors with significant nuclear atypia, mitotic activity (>5 mits/10 HPF), and coagulative necrosis have a high risk for recurrence, metastasis, and death due to disease whereas tumors with less severe degrees of nuclear atypia, <5 mits/10 HPF, and no necrosis have a much smaller risk of aggressive behavior. Many, but not all, patients with low-grade tumors have had no evidence of disease after relatively long follow-up intervals. Patients with higher grade tumors also follow a more rapidly fatal course (mean time to death = 7 months for high grade tumors vs. 65 months in low grade tumors in one study). [17] The distinction of LMS from leiomyoma and high grade LMS from low grade LMS are summarized in the tables below. Summary of distinguishing features between Leiomyoma and Leiomyosarcoma Leiomyoma Leiomyosarcoma Cytology No pleomorphism or hyperchromasia Obvious pleomorphism and hyperchromasia Circumscription Well-circumscribed Infiltrative border Stroma Stromal changes uncommon Myxoid change common Mitoses None or minimal (<1/10 HPF) Usually > 1/10 HPF Coagulative Necrosis Absent Present Immunohistochemistry in Smooth Muscle Tumors of the Urinary Bladder LMS Leiomyoma Actin +/- (70%) + Desmin +/- (50%) + h-Caldesmon + + Cytokeratin -/+* - EMA -/rare focal + - ALK1 -/rare focal + - Myogenin - - Summary of features that stratify leiomyosarcoma into low and high risk groups Low-grade High-grade Nuclear pleomorphism Mild+ Moderate-severe+ Mitotic activity <1/10HPF or <5/10 HPF* >5/10 HPF Coagulative necrosis None or Focal (<15%)* Extensive (>15%) + No objective criteria for nuclear "atypia" have been studied * Two different thresholds have been reported [17, 19] III. Skeletal Muscle Neoplasms Children and adolescents may occasionally undergo biopsies for bladder masses. The most common urinary bladder tumor in this age group is the botryoid type of embryonal rhabdomyosarcoma (RMS). Rarely, vesical rhabdomyosarcoma may occur in adults, but the histologic subtype is not as prognostically significant and does not guide clinical management in that setting. The existence of a true benign skeletal muscle tumor (rhabdomyoma) of the bladder is questionable with no convincing evidence in the literature. RMS is subclassified primarily into embryonal and alveolar types. Embryonal RMS may be further classified as botryoid, spindle cell, anaplastic, or not otherwise specified. Rare alveolar RMSs may be classified as anaplastic. This review focuses mainly on the diagnostic criteria for RMS and the distinction of embryonal and alveolar subtypes. 1) Embryonal rhabdomyosarcoma (RMS), botryoid type Botryoid RMS is by far the most frequently encountered neoplasm in pediatric bladder biopsies and responds well to current treatment protocols. [21] Grossly, these tumors present as polypoid, predominantly intraluminal bladder masses, and the clinical diagnosis is usually straightforward. Morphologically, the main diagnostic feature is the "cambium layer": a proliferation of spindle cells condensed underneath and abutting the surface urothelial or squamous lining, typically forming an exophytic, polypoid growth pattern somewhat reminiscent of phyllodes tumor of the breast at the superficial aspect. The spindle cells show varying degrees of differentiation ranging from undifferentiated mesenchymal cells to fully developed rhabdomyoblasts with prominent eosinophilic cytoplasm and cytoplasmic cross-striations. The botryoid type typically has areas with an abundant, loose myxoid stroma and relatively bland spindle cells. If these bland foci are the only areas represented in the biopsy, one must have a high index of suspicion based on the clinical setting and seek immunohistochemical confirmation as described below. Some studies of RMS have suggested that the growth pattern within the wall of the bladder might substratify patients with diffuse infiltration being an adverse prognostic factor, [24] but this feature is not currently used for defining risk groups by the COG Soft Tissue Sarcoma Study Group (formerly the Intergroup Rhabdomyosarcoma Study). [21, 22, 23] 2) Alveolar rhabdomyosarcoma Although extremely rare, alveolar RMS may occur in the urinary bladder. This is an important distinction, because the prognosis for pediatric patients with alveolar RMS is much worse than embryonal, and a more aggressive chemotherapeutic regimen is required. [21] Morphologically, alveolar RMS can present with three different growth patterns: classic alveolar, solid, and mixed alveolar-embryonal. All histologic patterns of alveolar RMS share the same cytology: the neoplastic cells resemble lymphoma with high nuclear-to-cytoplasmic ratios, round nuclear contours, and vesicular nuclear chromatin. Focal areas may suggest myogenic differentiation, but alveolar RMSs frequently appear completely undifferentiated and mimic other small round cell malignancies. In the classic pattern, the low power architecture is characteristic with fibrovascular septa dividing the neoplasm into separate nests. The neoplastic cells located centrally in the nests are typically discohesive, but residual cells line the septa in a hobnail or "picket fence" pattern. The discohesive cells may show considerable crush-like artifact and appear homogeneously "hyperchromatic". The solid-type of alveolar RMS loses the low power nested architecture, and grows in confluent sheets; however, the cytology of the neoplastic cells is identical to that described above. Diagnostic difficulties may arise in tumors showing mixed alveolar and embryonal patterns. Under the current COG Soft Tissue Sarcoma Study Group (Intergroup Rhabdomyosarcoma Study) definitions, the presence of any alveolar component warrants designation as alveolar type for treatment purposes regardless of extent. [21] This definition of alveolar is based on outcome data that demonstrates a worse prognosis associated with any focal area of alveolar histology. The pattern of alveolar component in these mixed cases is most frequently the classic nested morphology, but solid patterns may be seen as well. Additional features described in alveolar RMS include giant, multinucleated cells with rhabdomyoblastic differentiation and clear cell change. The distinguishing features of alveolar and embryonal RMS are summarized in the table below. Difficulties in Subclassification Most cases of RMS show the classic morphologies as described above and are readily sub-classified as either embryonal or alveolar, but occasional morphologic patterns are very difficult to classify under the current published criteria. Focal cellular, monomorphic areas without fibrovascular septa in an otherwise typical embryonal rhabdomyosarcoma may suggest either a focal solid alveolar pattern or simply a cellular focus of embryonal. In addition, RMS with marked nuclear pleomorphism may also cause difficulty in separating anaplastic embryonal from anaplastic alveolar RMS. In these morphologically ambiguous cases, there is some evidence suggesting the utility of using diffuse myogenin reactivity [50, 51, 52] or PAX translocations [51, 53] as surrogate markers of the alveolar subtype. Immunohistochemistry Because of the consistent immunophenotype and the therapeutic implications, the diagnosis of RMS should be confirmed by immunohistochemistry in most cases. RMS usually shows strong cytoplasmic immunoreactivity for desmin and nuclear reactivity for either/both MyoD1 and myogenin. MyoD1 and myogenin are very specific markers of skeletal muscle differentiation, and can help to avoid confusion with other neoplasms expressing desmin such as leiomyosarcoma and desmoplastic small round cell tumor. [25, 26] Because of the improved specificity of MyoD1 and myogenin, myoglobin and other previously useful markers of skeletal muscle differentiation have been largely replaced. After heat-induced epitope retrieval, MyoD1 may show non-specific cytoplasmic staining, but only nuclear staining should be regarded as positive. Summary of Features Distinguishing Embryonal and Alveolar Rhabdomyosarcoma* Alveolar Embryonal Cytology Round cells with vesicular chromatin ("lymphoma-like") Variably sized spindled cells Architecture Nests of cells divided by fibrovascular septa No nested arrangement or fibrous septa; Variably cellular with variable fascicular architecture Maturation Predominantly undifferentiated by routine morphology Varying degrees of maturation to rhabdomyoblasts Immunohistochemistry Desmin + Diffuse nuclear myogenin + Desmin + Scattered nuclear myogenin + *Any focal morphologic alveolar pattern is sufficient for classification as alveolar rhabdomyosarcoma regardless of extent IV. Sarcomatoid Carcinoma As in any anatomic site, carcinomas of the urinary bladder can "de-differentiate" into a spindle cell neoplasm that histologically resembles a sarcoma. [27, 28, 29, 30, 31, 54, 55] These carcinomas with a sarcomatous appearance have been reported under a variety of names including sarcomatoid carcinoma, metaplastic carcinoma, carcinosarcoma, malignant mixed mesodermal tumor, and spindle and giant cell carcinoma. Sarcomatoid urothelial carcinoma may be biphasic with both a spindle cell and epithelial component or have a completely sarcomatoid morphology. The epithelial component may be a papillary urothelial carcinoma, invasive urothelial carcinoma, urothelial carcinoma in-situ, squamous cell carcinoma, adenocarcinoma, or small cell carcinoma. [28] Extensive tissue sampling may be required to identify minor foci of obvious epithelial differentiation. The spindle cell or "sarcomatoid" component typically shows marked nuclear pleomorphism and hyperchromasia, and usually has the appearance of an undifferentiated malignant spindle cell tumor or may acquire a specific line of heterologous differentiation (so-called carcinosarcoma) histologically identical to subtypes of known sarcomas. The presence of heterologous differentiation (e.g., chondrosarcoma, osteosarcoma, rhabdomyosarcoma) does not seem to have independent prognostic significance, but the tumors are usually high-grade. One series has described sarcomatoid urothelial carcinomas with "myxoid and sclerosing" features that may closely mimic an inflammatory myofibroblastic tumor. [31] The diagnostic criteria for distinguishing a sarcomatoid urothelial carcinoma from a primary vesical sarcoma and IMT are discussed in the differential diagnosis section. The immunoprofile of sarcomatoid urothelial carcinoma is variable. Although cytokeratin and EMA immunoreactivity is usually present in 67-100% of cases, the spindle cell component may have only focal reactivity or may be completely negative. [6, 47, 9, 29, 30, 56] The spindle cell component may also variably express actin (15-80% in reported series) and/or desmin (0-40 %), which may be diffuse in cases with myogenous differentiation. Cases showing a specific line of heterologous differentiation typically have the expected immunoprofile for that tissue type. Diagnostic Criteria for Sarcomatoid Carcinoma Spindle cell neoplasm with marked nuclear pleomorphism and hyperchromasia (histologically malignant), and one of the following: Associated papillary or invasive carcinoma of any type Associated urothelial carcinoma in-situ Prior history of urothelial carcinoma Strong, diffuse cytoplasmic cytokeratin and/or EMA immunoreactivity V. Other Mesenchymal Tumors with Monotonous Spindle Cells Solitary fibrous tumor/ hemangiopericytoma [32, 33, 34] Solitary fibrous tumor (SFT)/ hemangiopericytoma (HPC) may occur as a primary bladder neoplasm. [32, 33, 34] These two tumors are closely related and are thought to represent ends of a common morphologic spectrum of mesenchymal neoplasia. [35] More cellular cases are generally termed HPC, while tumors with more collagenized stroma are classified as SFT. The main histologic features shared by both include a typically well-delineated mass, a proliferation of bland spindle cells with even chromatin arranged in a disorganized, "patternless" architecture, and a distinct hemangiopericytic vasculature. There is usually varying cellularity with more cellular areas separated by thick bands of hyalinized collagen. Hemangiopericytic vasculature is defined as varying sized, slit-like to ectatic vessels showing branching or angulated ends, typically surrounded by some degree of perivascular hyalinization. SFT/ HPC have also been reported with admixed adipose tissue (lipomatous HPC)38 and myxoid stromal changes (myxoid HPC). [39] To date, the reported cases of vesical SFT have followed a benign clinical course, but there are few cases and follow-up data is limited. We use the general prognostic factors that have been described for soft tissue SFT/HPC. [35, 36, 37] It is often stated that outcome is difficult to predict for these tumors. While there may be some outliers, in general, most tumors that follow an aggressive clinical course have one or more of the following features: mitoses > 4/10 HPF, marked nuclear atypia (hyperchromasia and pleomorphism), and coagulative tumor cell necrosis. Long-term follow-up care is warranted for all patients with SFT/HPC. Neurofibroma [40, 41] Neurofibromas of the bladder are identical to those in other anatomic sites. They classically present as a hypocellular proliferation of spindle cells arranged into loose and disorganized, hypocellular fascicles with scattered, variably shaped bundles of collagen (so-called "shredded carrot" collagen). The individual cells are characterized by small bland nuclei with curved or wavy shapes. One unusual feature that may be helpful is that, unlike other tumors with associated collagen bundles (spindle cell lipoma, fibromatosis, etc.), the neoplastic cells in neurofibroma are frequently present within the collagen bundles. Both cytoplasmic and nuclear immunoreactivity for S-100 protein should confirm the diagnosis. These tumors are benign and generally require no additional therapy. Fibroepithelial Polyp [42, 43] Although it is not truly a mesenchymal lesion, for differential diagnostic reasons, fibroepithelial polyp will be discussed. These rare bladder lesions occur in all ages, and are more common in the ureter and renal pelvis. In children, they are the most commonly identified benign bladder lesion. Morphologically, they are characterized by large polypoid excrescences covered by urothelium. The central cores are comprised of loose, edematous stroma with scattered blood vessels. Rare cases have been reported with atypical stromal cells similar to those seen in fibroepithelial polyps of the lower female genital tract and myxoid stroma with associated cystitis cystica. Other lines of differentiation Rarely, mesenchymal neoplasms common to other anatomic sites are seen in the bladder as primary tumors. A discussion of these neoplasms is beyond the scope of this course, but the list of case reports and small series includes: malignant fibrous histiocytoma/pleomorphic undifferentiated sarcoma, angiosarcoma, extraosseous osteosarcoma, fibrosarcoma, liposarcoma, granular cell tumor, clear cell sarcoma, Ewing's/PNET, schwannoma, and angiomyolipoma/PEComa. Because of their rarity as primary bladder neoplasms, one should exercise caution in rendering a sarcoma diagnosis in the urinary bladder (or in any parenchymal organ); the possibility of secondary involvement should be excluded clinically before the diagnosis of an unusual primary sarcoma is accepted. In addition, the possibility of heterologous differentiation within a sarcomatoid carcinoma should be considered and ruled out by extensive tissue sampling. Based on probability, specific lines of mesenchymal differentiation in patients with a past history of urothelial carcinoma (especially high grade carcinomas) most likely represent heterologous sarcomatoid carcinoma even when an associated carcinomatous component cannot be demonstrated. One extremely rare caveat to this rule arises in patients with a history of radiation therapy and a possible radiation-induced sarcoma; there is typically a long symptom-free latent period that may approach 10 years. [44] VI. The Differential Diagnosis of Spindle Cell Lesions Myofibroblastic Proliferation (IMT/PMP) vs. Leiomyosarcoma/Sarcomatoid Carcinoma This is potentially the most difficult distinction in the category of spindle cell lesions. In general, the nuclear morphology is the most useful feature in this setting. Although there may be some variation in the size of nuclei in an IMT/PMP, the chromatin is evenly distributed giving the nucleus a bland cytologic appearance, often with one or more nucleoli. Malignant lesions, in contrast, have irregular chromatin with marked hyperchromasia and may focally have obvious nuclear pleomorphism. Photomicrographs contrasting these nuclear features can be found in reference 45. The associated stroma and the low power architecture of IMT/PMP may be indistinguishable from a malignant neoplasm. Myxoid stroma is common in IMT/PMP, sarcomatoid carcinoma, and leiomyosarcoma. Both types of malignant tumors may have areas with a loose, hypocellular arrangement or more cellular fascicles. All three lesions commonly infiltrate the muscularis propria, have increased mitotic activity, and show necrosis superficially making them unreliable distinguishing features. Other morphologic features that are reportedly more common in malignant neoplasms include necrosis at the tumor/muscularis propria interface, acute inflammation, and lack of prominent blood vessels. Deep necrosis (at the tumor/muscularis propria interface) has been reported as a specific histologic feature of malignancy, [9] but this has been questioned in a more recent series. [5] Because of the significant overlap in immunophenotypes, immunohistochemistry plays little role in this differential diagnostic setting and is potentially misleading. The utility of h-caldesmon to distinguish smooth muscle from myofibroblasts has been recently questioned. [6, 49] Comparison of Benign Myofibroblastic Proliferations and Malignant Spindle Cell Neoplasms (SCA and LMS) Benign Myofibroblastic Malignant Cytology* Evenly distributed (fine) Coarse (hyperchromatic) Nucleoli Typically present Variable Mitoses Easily identifiable Easily identifiable Invasion Invades muscularis propria Invades muscularis propria Inflammation Usually chronic Usually acute Vasculature Small vessels common Prominent vessels uncommon Stroma Often myxoid May be myxoid Cellularity Variable Variable SCA: Sarcomatoid urothelial carcinoma; LMS: Leiomyosarcoma * Cytology is the most specific finding and supersedes other features Leiomyosarcoma vs. Sarcomatoid Carcinoma Once a diagnosis of histologic malignancy is determined, the main distinction is generally between a leiomyosarcoma and a sarcomatoid urothelial carcinoma because these two tumors may be morphologically indistinguishable. An association with a papillary urothelial neoplasm, invasive urothelial carcinoma, or urothelial carcinoma in-situ is generally diagnostic of sarcomatoid urothelial carcinoma. Because the identification of a carcinomatous component is the most specific finding, extensive tissue sampling is recommended to identify small foci of an obvious epithelial component. When the tumor has the morphologic appearance of a sarcoma and no clues are present to aid in the distinction, immunohistochemistry and clinical history may be useful. The past history of a high-grade urothelial carcinoma is usually sufficient for classification as a sarcomatoid carcinoma. The immunohistochemical profiles have significant overlap in the spindle cell component. Although cytokeratin immunoreactivity has been reported in a large number of leiomyosarcomas, the pattern is usually focal or patchy. EMA may be expressed in sarcomatoid carcinoma, but not typically in leiomyosarcoma. Actin and desmin reactivity have also been reported in sarcomatoid urothelial carcinoma, and may be diffuse, particularly in cases with morphologic smooth or skeletal muscle differentiation. In the setting of a malignant tumor without an obvious carcinomatous component, strong, diffuse desmin and actin reactivity, with only focal or absent cytokeratin expression, would support a leiomyosarcoma. The opposite immunophenotype (strong, diffuse cytokeratin expression in the absence of actin and desmin) would support sarcomatoid urothelial carcinoma. Rhabdomyosarcoma vs. Fibroepithelial Polyp Although fibroepithelial polyps may have a polypoid growth similar to botryoid RMS, the stroma is usually less cellular than rhabdomyosarcoma and a cambium layer is not evident. Immunostains for myogenin and MyoD1 should resolve most cases because the stromal cells in fibroepithelial polyps are non-reactive for these markers. IMT may also closely mimic the deceptively bland pattern of embryonal RMS, but again, specific skeletal muscle markers resolve most cases. ALK-1 immunostains are not helpful in this scenario because up to 20% of RMS show immunoreactivity with this antibody. Sarcomatoid Urothelial Carcinoma vs. Urothelial Carcinoma with Associated Pseudosarcomatous Myofibroblastic Proliferation Spindle cell proliferations associated with urothelial carcinoma should be morphologically evaluated using the criteria discussed above for IMT vs. sarcomatoid urothelial carcinoma. Pseudosarcomatous myofibroblastic proliferations associated with carcinoma are histologically indistinguishable from de novo IMT/PMP (Figure 8), while sarcomatoid urothelial carcinomas contain a spindle cell component that typically shows both obvious nuclear pleomorphism and hyperchromasia. Immunohistochemistry can be very confusing in this setting, because both lesions may express cytokeratin. Although this distinction should be based primarily on morphology, the co-expression of both actin and cytokeratin is typical of myofibroblasts. Urothelial Carcinoma with Associated Pseudosarcomatous Myofibroblastic Proliferation vs. Urothelial Carcinoma with Muscularis Propria Invasion Staging invasive urothelial carcinomas with associated pseudosarcomatous myofibroblastic proliferations may be difficult in some cases given the morphologic overlap with smooth muscle. The pseudosarcomatous myofibroblastic proliferations are characterized by individual myofibroblasts, usually separated by some degree of collagenous to myxoid stroma (Figure 8). In contrast, the smooth muscle of the muscularis propria is characterized by cells with more prominent eosiniophilic cytoplasm closely arranged into tight, rounded bundles without intervening stroma (Figure 9). This distinction should be primarily based on morphology, but normal muscularis propria does not typically express cytokeratin. Summary of Expected Immunophenotypes for Spindle Cell Proliferations of the Urinary Bladder IMT SCA LMS RMS Actin 63-100% 15-80% 43-100% 97% Desmin 27-80% 0-40% 0-60% 97% h-caldesmon 0-66% N/A 100% 100% CKAE1/3 36-89% 67-100% 0-38% 7% EMA 0-50% 50-100% 0-12% 2% ALK-1 75-89% 0% 0% 20% Myogenin 0% 0% 0% 76-100% MyoD-1 0% 0% 0% 91-100% IMT: inflammatory myofibroblastic tumor; SCA: Sarcomatoid urothelial carcinoma; LMS: Leiomyosarcoma; RMS: Rhabdomyosarcoma Selected References for Spindle Cell Proliferations of the Urinary Bladder General McKenney JK. An approach to the classification of spindle cell proliferations in the urinary bladder. Adv Anat Pathol 2005;12:312-323. Myofibroblastic Proliferations Hojo H, Newton WA, Hamoudi AB, et al. Pseudosarcomatous myofibroblastic tumor of the urinary bladder in children: a study of 11 cases with review of the literature. An Intergroup Rhabdomyosarcoma Study. Am J Surg Pathol 1995;19:1224-1236. Albores-Saavedra J, Manivel JC, Essenfeld H, et al. Pseudosarcoamtous myofibroblastic proliferations in the urinary bladder in children. Cancer 1990;66:1234-1241. Jones EC, Clement PB, Young RH. Inflammatory pseudotumor of the urinary bladder: a clinicopathological, immunohistochemical, ultrastructural, and flow cytometric study of 13 cases. Am J Surg Pathol 1993;17:264-274. Harik LR, Merino C, Coindre JM, Amin MB, Weiss SW. Pseudosarcomatous myofibroblastic proliferation of the bladder: a clinico-pathologic study of 42 cases. Am J Surg Pathol 2006;20:787-794. Freeman A, Geddes N, Munson P et al. Anaplastic lymphoma kinase (ALK1) staining and molecular analysis in inflammatory myofibroblastic tumours of the bladder: a preliminary clinicopathological study of nine cases and review of the literature. Mod Pathol 2004;17:765-771. Proppe KH, Scully RE, Rosai J. Post-operative spindle cell nodules of the genitourinary tract resembling sarcomas: report of eight cases. Am J Surg Pathol 1984;8:101-108. Ro JY, El-Naggar A, Amin MB, Sahin AA, Ordonez NG, Ayala AG. Pseudosarcomatous fibromyxoid tumor of the urinary bladder and prostate: immunohistochemical, ultrastructural, and DNA flow cytometric analyses of nine cases. Hum Pathol 1993;24:1203-1210. Iczkowski KA, Shanks JH, Gadaleanu V, et al. Inflammatory pseudotumor and sarcoma of urinary bladder: differential diagnosis and outcome in thirty-eight spindle cell neoplasms. Mod Pathol 2001;14:1043-1051. Young RH, Wick MR. Transitional cell carcinoma of the urinary bladder with pseudosarcomatous stroma. Am J Clin Pathol 1988;90:216-219. Cessna MH, Zhou H, Sanger W, et al. Expression of ALK1 and p80 in inflammatory myofibroblastic tumor and its mesenchymal mimics: a study of 35 cases. Mod Pathol 2002;15:931-938. Watanabe K, Kusakabe, Hoshi N, Saito A, Suzuki T. h-Caldesmon in leiomyosarcoma and tumors with smooth muscle-like differentiation: its specific expression in the smooth muscle cell tumor. Hum Pathol 1999;30:392-396. Hirsch MS, Dal Cin P, Fletcher CDM. ALK expression in reactive pseudosarcomatous myofibroblastic proliferations of the genitourinary tract. Mod Pathol 2003;16:153A (697). Young RH, Eble JN. Unusual forms of carcinoma in the urinary bladder. Hum Pathol 1991;22:948-65. Coffin CM, Watterson J, Priest JR, Dehner LP. Extrapulmonary inflammatory myofibroblastic tumor (inflammatory pseudotumor). Am J Surg Pathol 1995;19:859-872. Tsuzuki T, Magi-Galuzzi, Epstein JI. ALK-1 expression in inflammatory myofibroblastic tumor of the urinary bladder. Am J Surg Pathol 2004;28:1609-1614. Dehner LP. Inflammatory myofibroblastic tumor. The continued definition of one type of so-called inflammatory pseudotumor. Am J Surg Pathol 2004;28:1652- 1654. Montgomery E, Shuster DD, Burkhart AA, et al. Inflammatory myofibroblastic tumor of bladder: a clinicopathologic study, including a subset associated with high-grade carcinomas. Am J Surg Pathol 2006; 30:1502-1512. Smooth Muscle Neoplasms Yusim IE, Neulander EZ, Eidelberg I, Lismer LJ, Kaneti J. Leiomyoma of the genitourinary tract. Scand J Urol Nephrol 2001;35:295-299. Lake MH, Kossow AS, Bokinsky G. Leiomyoma of the bladder and urethra. J Urol 1981;125:742-743. Martin SA, sears D, Sebo TJ et al. Smooth muscle neoplasms of the urinary bladder. A clinicopathologic comparison of leiomyoma and leiomyosarcoma. Am J Surg Pathol 2002;26:292-300. Iczkowski KA, Shanks JH, Gadaleanu V, et al. Inflammatory pseudotumor and sarcoma of urinary bladder: differential diagnosis and outcome in thirty-eight spindle cell neoplasms. Mod Pathol 2001;14:1043-1051. Mills SE, Bova S, Wick MR, Young RH. Leiomyosarcoma of the urinary bladder. A clinicopathologic and immunohistochemical study of 15 cases. Am J Surg Pathol 1989;13:480-489. Swartz DA, Johnson DE, Ayala AG, Watkins DL. Bladder leiomyosarcoma: a review of 10 cases with 5-year follow-up. J of Urol 1985;133:200-202. Skeletal Muscle Neoplasms Newton WA, Gehan EA, Webber BL, et al. Classification of rhabdomyosarcomas and related sarcomas. Pathologic aspects and proposal for a new classification--an Intergroup Rhabdomyosarcoma Study. Cancer 1995;76:1073-85. Qualman SJ, Coffin CM, Newton WA, Hojo H, Triche TJ, Parham DM, Crist WM. Intergroup Rhabdomyosarcoma Study: update for pathologists. Ped Develop Pathol 1998;1:550-561. Parham DP. Pathologic classification of rhabdomyosarcomas and correlations with molecular studies. Mod Pathol 2001;14:506-14. Leuschner I, Harms D, Mattke A, Koscielniak E, Treuner J. Rhabdomyosarcoma of the urinary bladder and vagina. A clinicopathologic study with emphasis on recurrent disease: a report from the Kiel Pediatric Tumor Registry and the German CWS study. Am J Surg Pathol 2001;25:856-864. Cessna MH, Zhou H, Perkins SL, Tripp SR, Layfield L, Daines C, Coffin CM. Are myogenin and myoD1 expression specific for rhabdomyosarcoma? A study of 150 cases, with emphasis on spindle cell mimics. Am J Surg Pathol 2001;25:1150-7. Folpe , AL . MyoD1 and myogenin expression in human neoplasia: a review and update.Adv Anat Pathol 2002;9:198-203. Heerema-McKenney A, Wijnaendts LCD, Pulliam JF, et al. The prognostic significance of myogenin immunoreactivy in rhabdomyosarcoma. Socoety for Pediatric Pathology Annual Meeting, Feb. 11-12, 2006, abstract 18. Hostein I, Andraud-Fregeville M, Guillou L, et al. Rhabdomyosarcoma: value of myogenin expression analysis and molecular testing in diagnosing the alveolar subtype. An analysis of 109 paraffin-embedded specimens. Cancer 2004; 101:2817-2824. Dias P, Chen B, Dilday B, et al. Strong immunostaining for myogenin in rhabdomyosarcoma is significantly associated woith tumors of the alveolar subclass. Am J Pathol 2000; 156: 399-408. Sorenson PH, lynch JC, Qualman SJ, et al. PAX3-FKHR and PAX7-FKHR gene fusions are prognostic indicators in alveolar rhabdomyosarcoma: a report from the children's oncology group. J Clin Oncol 2002; 20:2672-2679. Sarcomatoid Carcinoma Ro JY, Ayala AG, Wishnow K, et al. Sarcomatoid bladder carcinoma: cinicopathological and immunohistochemical study of 44 cases. Surg Pathol 1988;1:359. Young RH, Wick MR, Mills SE. Sarcomatoid carcinoma of the urinary bladder: a clinicopathological analysis of 12 cases and review of the literature. Am J Clin Pathol 1988;90:653-661. Ikegami H, Iwaski H, Ohjimi Y, et al. Sarcomatoid carcinoma of the urinary bladder: a clinicopathologic and immunohistochemical analysis of 14 patients. Hum Pathol 2000;31:332-340. Lopez-Beltran A, Pacelli A, Rothenberg HJ et al. Carcinosarcoma and sarcomatoid carcinoma of the bladder: a clinicopathological study of 41 cases. J Urol 1998;159:1497-1503. Jones EC, Young RH. Myxoid and sclerosing sarcomatoid carcinoma of the urinary bladder: a clinicopathologic and immunohistochemical study of 25 cases. Mod Pathol 1997;10:908-916. Serio G, Zampatti C, Ceppi M. Spindle and giant cell carcinoma of the urinary bladder: a clinicopathological light microscopic and immunohistochemical study. Br J Urol 1995;75:167-72. Komatsu H, Kinoshita K, Mikata N, Honma Y. Spindle and giant cell carcinoma of the urinary bladder. Report of 3 cases. Eur Urol 1985;11:141-4. Torenbeek R, Blomjous CE, de bruin PC, et al. sarcomatoid carcinoma of the urinary bladder. Clinicopathologic analysis of 18 cases with immunohistochemical and electron microscopic findings. Am J Surg Pathol 1994; 18:241-249. Solitary Fibrous Tumor/ Hemangiopericytoma Mentzel T, Bainbridge TC, Katenkamp D. Solitary fibrous tumor: clinicopathological, immunohistochemical, and ultrastructural analysis of 12 cases arising in soft tissues, nasal cavity and nasopharynx, urinary bladder and prostate. Virchows Arch 1997;430:445-453. Westra WH, Grenko RT, Epstein J. Solitary fibrous tumor of the lower genitourinary tract: a report of five cases involving the seminal vesicles, urinary bladder, and prostate. Hum Pathol 2000;31:63-68. Corti B, Carella R, Gabusi E, et al. Solitary fibrous tumor of the urinary bladder with expression of bcl-2, CD34, and insulin-like growth factor type II. Eur Urol 2001;39:484-488. Guillou L, Fletcher JA, Fletcher CDM, Mandahl N. Extrapleural solitary fibrous tumour and haemangiopericytoma. In: Fletcher CDM, Unni K, Mertens F (eds.) WHO Classification of Tumours. Pathology and genetics. Tumours of soft tissue and bone. IARC Press: Lyon, 2002. Vallat-Decouvelaere AV, Dry SM, Fletcher CD. Atypical and malignant solitary fibrous tumors in extrathoracic locations: evidence of their comparibility to intra- thoracic tumors. Am J Surg Pathol 1998;22:1501-1511. Enzinger FM, Smith BH. Hemangiopericytoma. An analysis of 106 cases. Hum Pathol 1976;7:61-82. Folpe AL, Devaney K, Weiss SW. Lipomatous hemangiopericytoma: a rare variant of hemangiopericytoma that may be confused with liposarcoma. Am J Surg Pathol 1999;23:1201-7. Weiss SW, Goldblum JR. Hemangiopericytoma and solitary fibrous tumor family. In: Enzinger and Weiss's Soft Tissue tumors. 4th ed. Chapter 27: Perivascular tumors. Mosby, 2001: St. Louis, MO. Other Mesenchymal Tumors Winfield HN, Catalona WJ. An isolated plexiform neurofibroma of the bladder. J Urol 1985;134:542-3. Weiss SW, Goldblum JR. Neurofibroma. In: Enzinger and Weiss's Soft Tissue tumors. 4th ed. Chapter 30: Benign tumors of peripheral nerve. Mosby, 2001: St. Louis, MO. Young RH. Fibroepithelial polyp of the bladder with atypical stromal cells. Arch Pathol Lab Med 1986;110:241-42. Williams TR, Wagner BJ, Corse WR, Vestevich JC. Fibroepithelial polyps of the urinary tract. Abdom Imaging 2002;27:217-221. Laskin WB, Siverman TA, Enzinger FM. Postradiation soft tissue sarcomas: an analysis of 53 cases. Cancer 1988;62:2330.