Biopsy Techniques, Diagnostic Methods, and Reporting
INTRODUCTION
Soft tissue tumors comprise a group of entities showing mesenchymal differentiation which can be located in skin, subcutis, or deep soft tissue. The latter includes subfascial limb and limb girdle tumors and those located in head and neck, abdomen, retroperitoneum (including paratesticular region), pelvis, and thoracic and intracranial cavities. Similar lesions can also involve viscera. Typically, soft tissue tumors are classified by the type of differentiation which they display,1 and the principal objectives of the surgical pathologist are to identify the lineage of the lesional cells and assess their malignant potential. Nonmesenchymal lesions such as carcinoma or melanoma can also present as soft tissue neoplasms.
In many cases, the tumor is seen initially as a spindle cell, epithelioid cell, small round cell, or pleomorphic lesion that needs to be characterized further with the aid of ancillary techniques including immunohistochemistry and molecular pathologic analyses. Lesions with adipose, osteochondroid, or vascular space formation can be identified morphologically but often present difficulties in precise subcategorization and in assessment of malignancy. This book is, therefore, organized into chapters representing the main morphologic categories as they present to the pathologist. Each entry includes relevant clinical data, morphologic features, and information from ancillary techniques wherever appropriate. The differential diagnosis within each category is presented in detailed tabular form. It is hoped that this approach will reflect the pathologist’s diagnostic process and will facilitate diagnosis and provision of relevant information to the clinician in this rare and difficult group of neoplasms. Necessarily, some entities appear in more than one category, with the main entry corresponding to the most common pattern or location in each case.
BIOPSY
Most superficial soft tissue tumors can be directly excised in their entirety, but deep tumors often grow to a large size and require diagnosis by biopsy. Interpretation can be difficult because a biopsy represents only a very
small proportion of the tumor and the appearances in sarcomas can be heterogeneous and modified by myxoid, fibrous, or inflammatory stromal changes. Accurate diagnosis involves two major decision making steps: Is the lesion malignant? If so, what sort of a sarcoma is it? Tumor grade should also be assessed when possible. Because of the rarity of soft tissue tumors, a patient with a soft tissue mass should ideally be referred to a specialized center for evaluation, biopsy (or review of pathology if carried out elsewhere), and, if necessary, treatment by a multidisciplinary team that includes an experienced pathologist.
small proportion of the tumor and the appearances in sarcomas can be heterogeneous and modified by myxoid, fibrous, or inflammatory stromal changes. Accurate diagnosis involves two major decision making steps: Is the lesion malignant? If so, what sort of a sarcoma is it? Tumor grade should also be assessed when possible. Because of the rarity of soft tissue tumors, a patient with a soft tissue mass should ideally be referred to a specialized center for evaluation, biopsy (or review of pathology if carried out elsewhere), and, if necessary, treatment by a multidisciplinary team that includes an experienced pathologist.
Biopsy Techniques
Biopsy of a soft tissue tumor can be excisional, open incisional, or closed.
1. Excisional biopsy aims to remove the whole lesion for both diagnostic and therapeutic purposes and is usually indicated for superficial lesions and those <5 cm in diameter. The excision should include a margin of normal tissue, although microscopic examination might later reveal tumor at the margin, indicating further excision. This applies especially to lesions that are “shelled” out as they are generally removed through the false capsule within which tumor cells remain.
2. Open incisional biopsy involves surgical opening of skin and subcutaneous tissues and direct sampling of the tumor from which a wedge can be removed. This approach results in a larger amount of tissue for diagnosis but has the potential complications of a formal surgical procedure, including wound dehiscence and infection, and compromise of subsequent definitive surgery or radiation therapy. It can also facilitate spread of tumor from the main lesion into adjacent or overlying tissues. In addition, an open procedure also requires more resources and is therefore less cost-effective than closed methods.
3. Closed biopsies carry minimal risk of infection or seeding of the wound with tumor and provide adequate material for all investigative modalities. They are simple and relatively inexpensive office procedures that are readily repeated if more material is required. Techniques include core needle biopsy (CNB) and fine needle aspiration cytology (FNAC).
CNBs are usually taken with a Tru-Cut needle (CareFusion Corp., San Diego, CA), which can be inserted through a tiny skin incision and angled to sample different parts of the tumor. Several studies of cutting needle biopsies for soft tissue tumors have demonstrated that the larger the needle, the more optimal the specimen. Ultrasound- or computerized axial tomography-guided biopsy can be performed for less accessible lesions such as those deep in body cavities, although they usually yield thinner cores because of the use of a smaller gauge needle. The use of CNB has become more widespread in recent years, and practiced operators can achieve a very high proportion of adequate biopsies. An experienced pathologist
can differentiate benign from malignant tumors with a sensitivity of 99.4%, a specificity of 98.7%, a positive predictive value of 99.4%, and a negative predictive value of 98.7%.2 Ninety-five percent accuracy can be obtained in subtyping and 85% in grading of soft tissue neoplasms on needle biopsies. Grading is less accurate than histologic typing, usually because a higher grade area is identified in the excised specimen. However, an “at least” grade, or an indication of low- or high-grade malignancy, can be given. For some pediatric tumors, grading, classification, and prognostic categorization have specific issues that may limit the use of CNBs.
FNAC is used for initial diagnosis of soft tissue tumors in some large centers and gives good results in expert hands.3 However, tumor architecture and mitotic index cannot be assessed, and it is less effective in determining histologic subtype and grade. Compared with CNB, FNAC has limitations in the areas of adipose, myxoid, and fibrous neoplasms.4 FNAC can have a role, however, in identifying recurrent or metastatic sarcomas.
Following prior biopsy diagnosis, frozen section is rarely indicated for immediate management, as margins are better assessed on the excised specimen. Intraoperative pathologic examination is, however, sometimes required at open biopsy, either for diagnosis or to confirm that adequate viable tissue is available.
SPECIMEN COLLECTION AND HANDLING
Core Biopsies
A minimum of three needle biopsy cores is optimal for morphology, immunohistochemistry, touch imprints and molecular genetic analysis (which may be done on paraffin sections), cytogenetics, and storage of frozen tissue for future molecular studies. Specimen adequacy can be evaluated with touch imprints or frozen section at the time of the biopsy procedure. Touch imprints are useful for a variety of possible genetic and cell marker studies, and some experts have advocated performing touch imprints until the pathologist is comfortable with the adequacy of the entire specimen. This is not, however, usually necessary in routine use. If required, one core can be used for frozen section examination and a portion can be placed in glutaraldehyde for possible ultrastructural studies. The remainder of the cores should be submitted in formalin.
CNBs should be counted and measured. If the cores are not fragmented, and resources allow, it is preferable to separate and embed them in more than one cassette to allow maximal use of the available material.
Open Biopsies
These should also be submitted in formalin, but if the specimen is large enough, a small portion can be frozen in liquid nitrogen and stored at −80°C.
Resection Specimens
Surgical excision specimens should be sent to the laboratory directly after removal from the patient, without formalin, to allow orientation and sampling of fresh tissue for genetic studies. The latter might include freezing a suitable portion in liquid nitrogen and storage at −80°C in an appropriate facility. All likely investigative options should be considered at the time that fresh tissue arrives in the laboratory to maximize potential diagnostic and prognostic information.
After the specimen is weighed, orientated, and measured, its surface should be painted completely with specimen-marking ink and the ink allowed to dry. For large specimens and to prevent spillage of ink onto the cut surface, ink can be dabbed at the surface where the block is to be taken (e.g., the closest resection margin). One color is generally sufficient unless there is a particular clinical need to identify different aspects of the specimen. Intra-abdominal and retroperitoneal sarcomas do not usually require surface marking unless requested by the surgeon. The specimen should then be sliced transversely at 1-cm intervals. After small portions of tumor are removed for freezing, the specimen is placed in formalin in an adequately sized container and allowed to fix for 24 to 48 hours. After fixation, a suitable portion of the specimen should be photographed (this can also be done with the fresh specimen).
Resection specimens should be measured in three dimensions, and the closest distance of each margin from the edge of the tumor noted, including the deep aspect for subcutaneous or intramuscular sarcomas. The tissue plane(s) in which the tumor is located is recorded. Color, consistency, presence of cysts, hemorrhage, and necrosis are noted. The amount of necrosis is assessed as a percentage of the whole tumor. Blocks should be taken to include the nearest resection edge and the deep margin where appropriate. It is not necessary to take a resection margin which is more than 3 cm from the main tumor with the exceptions of some superficial sarcomas, such as low-grade myxofibrosarcoma5 and epithelioid sarcoma,6 which can infiltrate microscopically. Lesions that are smaller than 5 cm in diameter should be processed in their entirety. It is generally recommended that one block be taken per centimeter of the longest dimension of the tumor, up to a maximum of 12, although cases previously diagnosed on biopsy as of high-grade malignancy need fewer blocks. Areas that appear visibly different require appropriate extra sampling. In large liposarcomas, especially those from the retroperitoneum, any firmer or differently colored areas should be sampled to detect dedifferentiation.
DIAGNOSIS
Clinical Features
Before interpreting the biopsy, attention should be given to (1) the patient’s age and sex; (2) the size and duration of the tumor (e.g., a sarcoma is generally larger with a longer history, whereas nodular fasciitis is smaller
and of very recent onset); and (3) its location, including anatomical plane (i.e., cutaneous, subcutaneous, fascial, or deep). Most pseudosarcomas and benign tumors, and less frequently sarcomas (e.g., myxofibrosarcoma, leiomyosarcoma, myxoinflammatory fibroblastic sarcoma, and epithelioid sarcoma), occur in the superficial soft tissues. Conversely, a mass located deep to the deep fascia is more likely to be a sarcoma. The plane in which the tumor is situated can also be determined by imaging (computed tomography or magnetic resonance imaging), which can additionally suggest the composition of the lesion.
and of very recent onset); and (3) its location, including anatomical plane (i.e., cutaneous, subcutaneous, fascial, or deep). Most pseudosarcomas and benign tumors, and less frequently sarcomas (e.g., myxofibrosarcoma, leiomyosarcoma, myxoinflammatory fibroblastic sarcoma, and epithelioid sarcoma), occur in the superficial soft tissues. Conversely, a mass located deep to the deep fascia is more likely to be a sarcoma. The plane in which the tumor is situated can also be determined by imaging (computed tomography or magnetic resonance imaging), which can additionally suggest the composition of the lesion.
Diagnostic Techniques
Light microscopy is usually the only useful technique for distinguishing reactive lesions or benign tumors from malignant neoplasms. In the Royal Marsden series of 424 treated adult patients with soft tissue masses, 20% were benign,7 and in a separate series of 426 previously undiagnosed patients with soft tissue tumor who underwent CNB, 47% were benign.8 In children, the majority of soft tissue masses are benign and the spectrum of sarcoma subtypes is different. Criteria of malignancy differ with lineage but commonly include nuclear pleomorphism, excessive or abnormal mitotic activity, necrosis, vascular invasion, and the fact of metastasis. However, benign lesions which are cellular, pleomorphic, or mitotically active can be mistaken for a similar-appearing sarcoma. Superficial lesions which simulate malignant tumors include cellular fibrous histiocytoma, the fasciitides, cellular myxoma, spindle cell and pleomorphic lipoma, chondroid lipoma, extraskeletal chondroma, solitary fibrous tumor, and pleomorphic hyalinizing angiectatic tumor. Examples in the deep soft tissues include cellular schwannoma and inflammatory myofibroblastic tumor, and fibromatoses can sometimes be difficult to distinguish from low-grade sarcomas. Conversely, some malignancies can be underdiagnosed, either because they have well-differentiated areas resembling their benign counterparts, such as well-differentiated leiomyosarcoma, or because they are deceptively bland, for example, epithelioid sarcoma, low-grade fibromyxoid sarcoma, and myxofibrosarcoma. Such problems can be resolved by careful assessment of often subtle features of architecture and cytomorphology, tumor-stromal interface, and associated reaction; by immunohistochemical or genetic findings; and by awareness of the diagnostic possibilities and of the relevant criteria for malignancy.
Sarcomas with adipose or osteochondroid differentiation or vascular space formation can readily be diagnosed morphologically. Many others, however, are seen as spindle cell, epithelioid cell, clear cell, small round cell, or pleomorphic tumors, which need to be characterized further with the aid of immunohistochemistry, ultrastructural examination, or genetic techniques.
Immunohistochemistry
Immunohistochemistry is an essential adjunct in soft tissue tumor diagnosis. In a few cases, specific antigens are expressed, but for most, a panel of antibodies is required, with a second panel according to the broad lineage
indicated thereby, in conjunction with the morphology. Immunohistochemical markers useful in paraffin section are listed in detail for each tumor type, but some general comments are appropriate here. Antibodies are almost never 100% specific or sensitive, and overreliance on a single marker should be avoided. Familiarity with the pattern of reactivity for each antibody employed helps to avoid diagnostic error. Also, artifactually positive immunostaining can occur at the edges of small needle biopsy cores. A panel of antibodies should always be used; both positive and negative results are relevant and no finding should be taken out of the context of the morphologic findings and the clinical picture.
indicated thereby, in conjunction with the morphology. Immunohistochemical markers useful in paraffin section are listed in detail for each tumor type, but some general comments are appropriate here. Antibodies are almost never 100% specific or sensitive, and overreliance on a single marker should be avoided. Familiarity with the pattern of reactivity for each antibody employed helps to avoid diagnostic error. Also, artifactually positive immunostaining can occur at the edges of small needle biopsy cores. A panel of antibodies should always be used; both positive and negative results are relevant and no finding should be taken out of the context of the morphologic findings and the clinical picture.
The initial panel will depend on the morphologic group into which the lesion falls and the clinical circumstances. However, for spindle cell and pleomorphic sarcomas, a first-line panel generally includes desmin, smooth muscle actin, S100 protein, a broad-spectrum cytokeratin, and CD34, perhaps with the addition of MDM2, CDK4, and p16 for intraabdominal pleomorphic sarcomas. For epithelioid and clear cell soft tissue tumors, an initial panel might include cytokeratin, EMA, CD34, CD31, CD30, desmin, and S100 protein, with the addition of INI1 in selected cases. For small round cell tumors, a useful panel is CK, desmin, CD99, S100 protein, CD56, CD45, and TdT, with the addition of FLI-1, WT1, myogenin, TLE1, NB84a, synaptophysin, chromogranin, neurofilament, neuron-specific enolase, and other markers as indicated.
The results of these (e.g., Tables 1.1,1.2 and 1.3) will determine further investigations. Second-line panels apply more specific markers of epithelial, mesothelial, myoid, endothelial, neural, melanocytic, or other types of differentiation, or those useful for individual tumor types. Examples might include cytokeratin subtypes, calretinin, thrombomodulin, CEA, BerEp4, TLE1, GFAP, CD56, chromogranin, synaptophysin; myogenin, SMM, h-caldesmon, calponin, beta-catenin, TFE3, bcl-2, STAT6; HMB45, melan A, inhibin, CD31, FLI-1, HHV8, CD117; CD45, CD30, CD21, CD23, and CD35. A suspected diagnosis of gastrointestinal stromal tumor can be confirmed with CD117 and/or DOG1.
Electron Microscopy
Ultrastructural examination now has a limited role,11 but its use should be borne in mind where specifically indicated so that a small portion of tissue can be appropriately fixed and saved for referral to a specialist unit if this is subsequently needed. This technique remains of value where immunohistochemistry is inconclusive and genetic data are not available: examples include low-grade myofibrosarcoma, adult fibrosarcoma, S100 protein-negative malignant peripheral nerve sheath tumor, and some synovial sarcomas. It can also be contributory to diagnosis of sarcomas with distinctive ultrastructural features such as the crystalline structures of alveolar soft part sarcoma.
TABLE 1.1 Immunohistochemistry of Some CD34-Positive Soft Tissue Tumors | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
TABLE 1.2 Immunohistochemistry of Selected Tumors with Desmin Positivity | |||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
