2.1 Choice of surgery for tumour: Staging and surgical margins

Bone and soft tissue tumours are rare and constitute less than 1% of all malignancies. Such tumours should, when possible, be treated at specialized tertiary centres where there is a dedicated multidisciplinary team of medical, nursing, and counselling services to manage these often difficult problems.

A poorly performed biopsy at a non-specialist centre is reported to be the major cause of error and failure of limb salvage in two review articles. Although the procedure itself is not complicated, the indications for using this technique, the choice of needle, the appropriate setting, the anatomic site, the distribution of tissue, and the potential need for adjuvant treatment all must be carefully considered prior to performing the biopsy. Needle biopsies should be viewed as surgical procedures and are associated with risks and hazards.

Important issues that must be considered when performing biopsies for suspicion of tumour include:

Biopsies are best performed in specialist centres under the supervision of an orthopaedic surgical oncologist who will ultimately be responsible for the definitive care of the patient and who is working in close collaboration with an expert musculoskeletal radiologist and pathologist.

Accurate preoperative surgical staging of musculoskeletal tumours is currently possible because of advanced imaging techniques, providing diagnostic information and helping clinicians in choosing the most appropriate treatment option for the patient. The aims of surgical staging are to determine the surgical margins of resection and to facilitate interinstitutional and interdisciplinary communication regarding treatment data and results.

There are two aspects to staging musculoskeletal tumours, namely clinical staging and pathologic grading. Clinical staging defines the local and systemic extent of the disease which is required in the planning of treatment to achieve both local and distal control. Clinical staging relies on a variety of imaging studies which not only pinpoint the anatomical location of the tumour but also serve as baseline indices for post-treatment comparisons. Pathologic grading categorizes tumours according to their predicted biological behaviour which may range from indolence to widespread metastasis (determined by a combination of features including cellularity, degree of pleomorphism, mitotic activity, necrosis, and infiltrative nature). Pathologic grading relies on biopsy of the tumour and therefore should follow clinical staging.

Plain radiography is the initial imaging modality in the evaluation of bone tumours. Some benign lesions have characteristic radiographic features that make biopsy unnecessary. Examples include fibrous cortical defects, bone islands, simple bone cysts, and bone infarcts. Radiographic features can also help in distinguishing malignant from benign bone lesions in many patients. The important radiographic signs for grading bone tumours are listed as follows, in order of priority:

In the staging of bone tumours, computed tomography (CT) scanning has a role in the detailed evaluation of local disease and in assessing the lungs for pulmonary metastases. It can be used to assess disease in areas that are not easily visualized with plain radiographs, such as the spine and pelvis.

Magnetic resonance imaging (MRI) (with or without contrast) is the modality of choice for imaging and staging of musculoskeletal tumours. Accurate depiction of the soft tissues allows sensitive detection of soft tissue extension and medullary involvement by a tumour.

Certain lesions have appearances that are usually characteristic enough for a diagnosis to be made based on MRI findings. Examples of such lesions include lipomas, superficial and skeletal muscle haemangiomas, benign neural tumours, periarticular cysts, haematomas, and pigmented villonodular synovitis.

MRI is increasingly used to assess tumour response to preoperative chemotherapy. This assessment is achieved by evaluating changes in a tumour’s size, margins, signal intensity, and enhancement patterns. Moreover, MRI can be useful in differentiating tumour recurrence from chronic post-therapeutic changes.

Radionuclide bone scanning has a role in detecting metastases, skip lesions, lesion multiplicity, and postoperative tumour recurrence. Bone-forming, metastatic lesions in the lungs (e.g. osteosarcoma) are occasionally detected with bone scintigraphy.

Soft tissue masses and the soft tissue components of bony tumours may be visualized by using ultrasonography (US). The aim of ultrasonography in the evaluation of musculoskeletal lesions is to confirm the presence of a lesion, to determine if the lesion is cystic or solid, to assess the relationship of the mass to the surrounding structures (e.g. neurovascular bundle), to evaluate the vascularity of the mass, and to guide interventional procedures if indicated.

The role of angiography has largely been replaced by cross-sectional imaging modalities. Angiography can still have a preoperative role in decreasing tumour size by allowing embolization of feeding vessels and a postoperative role in decreasing haemorrhage, again through the embolization of tumour-supplying vessels.

Positron emission tomography (PET) scanning has been shown to have sensitivity similar to that of serial CT scanning and MRI for detecting lesions and for distinguishing postsurgical scarring from recurrent tumours. However, the specificity of PET is higher than that of serial CT scanning or MRI making it better in depicting residual/recurrent disease after treatment. The main disadvantage of PET scanning is the high cost of the equipment, therefore limiting availability.

Percutaneous needle biopsy is now standard practice. Given the increase in the use of CT and ultrasound for accurate needle placement, biopsy is now primarily a radiologist-led procedure.

The Enneking system for the surgical staging of bone and soft tissue tumours is based on grade (G), site (T), and metastasis (M) and uses histological, radiological, and clinical criteria. It is the most widely used staging system and has been adopted by the Musculoskeletal Tumour Society. This system should be used for staging mesenchymal lesions only, because their biological behaviour differs from that of non-mesenchymal tumours (e.g. Ewing sarcoma, lymphoma, and leukemia).

Bone tumours are graded as follows:

Surgical grade generally follows histological grade; however, a higher surgical grade may be applied if the radiographical features and clinical behaviour of a lesion indicate an aggressiveness that is incompatible with its benign histological features.

The site and local extent of bone tumours are classified as follows:

Metastatic classification is as follows:

Under the Enneking system, malignant tumours are classified into stages I–III, with further subdivisions into A and B. Grade 1 and grade 2 tumours are stage I and stage II, respectively. T1 and T2 tumours are stage A and stage B, respectively. Tumours with distant metastasis are stage III (Table 2.1.1).

The Enneking staging system divides benign tumours into latent, active, or aggressive tumours (Table 2.1.2). Latent tumours are asymptomatic and are usually discovered incidentally. They reach a stage of non-growth after a period of slow growth. Active tumours are mildly symptomatic and may be discovered if pathologic fracture occurs or if the tumour is associated with mechanical dysfunction. Active tumours usually grow steadily. Aggressive benign lesions grow rapidly and usually are symptomatic and tender on palpation.

Surgery is the mainstay of treatment for musculoskeletal tumours. These tumours are surrounded by a non-tumorous reactive zone containing inflammatory cells and neo-vascular tissue. This pseudocapsule contains microscopic extensions of tumour which may be continuous with, or satellites of, the main tumour mass (Figures 2.1.1 and 2.1.2).

Intralesional margins imply a procedure which crosses the pseudocapsule and enters into the substance of the tumour. Often tumour tissue is removed piecemeal. Curettage is an example of intralesional surgery. Recurrence can be high with intralesional margins in incomplete curettages as macroscopic evidence of tumour remains after surgery. Intralesional margins are only used in benign tumours such as bone cysts, aneurysmal bone cysts, and giant cell tumours.

Marginal margins are those in which a tumour is removed in one piece and the plane of dissection passes through the pseudocapsule or reactive zone. Excisional biopsies or tumours which are shelled out are examples of surgery with marginal margins, and microscopic residual tumour may exist. Recurrent rates of malignant tumours are high after surgery with marginal margins if adjuvant chemotherapy is not used.

Wide margins include the entire tumour and its pseudocapsule or reactive zone and a cuff of normal tissue en bloc. During this procedure, tumour is never seen. Consensus has not been reached on the definition of what constitutes a cuff of normal tissue. In principle, the proximal or distal cut edge of normal muscle represents a wide margin if it is 5cm or more from the tumour border. Removal of bone at a similar distance beyond the tumour also constitutes a wide margin. Surgical margins which include the normal fascia surrounding the muscle of an intramuscular tumour are regarded as wide even though the absolute thickness of the muscle and fascial cuff may only be 1–2mm. Synovium or periosteum which covers the periosteal extension of a distal femoral tumour is regarded as an anatomic barrier and if excised en bloc with the tumour constitutes a wide margin even though again the thickness of the tissue may be 1–2mm.

Radical margins are achieved when the entire tumour-bearing compartment is removed en bloc. For bone sarcomas, radical margins include the en bloc removal of the bone of origin and the entire soft tissue compartment into which the tumour has entered. For soft tissue sarcomas, this includes anatomical fascia surrounding that compartment and bony borders when present. Radical margins are associated with the lowest risk of recurrence.

With limb-sparing surgery as the preferred technique, the majority of uncomplicated bone or soft tissue resections aim at complete uncontaminated surgical margins. The use of chemotherapy has improved the rate of local control in osteosarcoma while allowing wide excisional limb-sparing surgery. Similarly, radiotherapy has had a major influence on decreasing the size of surgical margins in soft tissue sarcoma resections. Specifically, combination radiotherapy and marginal surgery has had the same results as surgery with wide margins alone, and the recurrence rates for surgery with wide margins combined with radiotherapy are similar for radical surgical margins alone. If vital neurovascular structures are adjacent to the tumour, marginal surgery may be selected and combined with adjuvant therapy. In selected tumours, such as those which are completely within a single muscle, or subcutaneous and not engaging the deep fascia, surgery with wide margins alone may be sufficient to achieve local control. Decisions regarding choice and combination of modalities need to be tailored to each case.

Now that staging has been completed, the decision to perform a biopsy is considered. Plain radiographs of bone lesions are often diagnostic or can significantly narrow the differential and aid in determining whether or not a biopsy should be performed (i.e. a classic osteochondroma or osteoid osteoma needs no biopsy prior to excision).

Open incisional biopsy is the conventional method for obtaining a representative specimen and is considered to be the ‘gold standard.’ In carefully selected patients, closed needle or trephine biopsies are appropriate. This is a well established practice, with a reported accuracy of 78–97%. The site and method of approach must provide samples that are adequate for diagnosis and deliver them with the minimum amount of morbidity. Two attempts at percutaneous biopsy are justified. Percutaneous biopsy yield compares well with open biopsy and is more cost-effective. In experienced hands, percutaneous biopsy can be performed quickly, with less patient morbidity, and at a fraction of the cost of open surgical biopsies. Closed biopsies are ideal for difficult-to-access lesions in the vertebrae and pelvis and can be performed under CT or fluoroscopic guidance. Additionally, needle biopsies, when compared to open biopsies, lessen the risk of potential biopsy complications such as tumour spillage or haematoma and contamination, wound dehiscence, and infection. Finally, a needle biopsy tract is easier to excise en bloc with the final resection specimen. Different systems exist allowing percutaneous biopsy of musculoskeletal lesions:

All biopsy tracts must be resected en bloc with the tumour at the definitive surgical procedure. Even needle biopsies can leave behind viable tumour cells that are a potential source for local recurrence, despite adjuvant treatment. Therefore, the skin puncture must be marked with India ink or a modest stitch so it can be identified later, even after weeks of neoadjuvant therapy.

CT guidance is used primarily for bone-based pathology and for deep soft tissue tumours that are beyond the focus of a standard ultrasound probe. CT also allows biopsy of smaller lesions with confidence and the final position of the needle can be reliably recorded. This is important for retrospective confirmation of needle position in non-diagnostic pathology reports. Ultrasound is suitable for superficial soft tissue lesions and for bone-based pathology such as osteosarcomas which commonly have a large extraosseous ‘soft tissue’ component at the time of presentation. The avoidance of critical anatomical structures within the biopsy field will affect the choice of guidance system and both CT and ultrasound may play a role. In certain centres, specialist biopsy-designed MRI is now available.

Several needle systems have been devised for biopsies of the musculoskeletal system. In our unit we routinely use two needle types:

This cutting needle with a tapered end (aiding tissue retention) and a disposable trocar is used for lesions that are predominantly osteoblastic and centrally sited within bone, or for predominantly lytic or mixed intraosseous lesions that have an intact surrounding rim or cortex (Figure 2.1.3). It is unsuitable for soft tissue biopsy work.

These are predominantly used for extraosseous soft tissue specimens, or when cortical bone destruction allows access to the medulla (Figure 2.1.4). Hemorrhagic lesions may fail to produce a satisfactory core, but direction under ultrasound or CT guidance improves the diagnostic yield. Multiple specimens can be provided for the histopathologist.

Full aseptic conditions are observed in all cases, and the appropriate method of anaesthesia, depending on the age of the patient, is selected.

Having chosen the radiological technique that complements the percutaneous biopsy, a small stab incision is made with a No. 15 blade at the site of entry of the needle and the selected needle type inserted. Care must be taken in cases of suspected malignant but contained lesions not to take the needle beyond the point of natural tissue hold-up. For example, with a subperiosteal osteosarcoma contamination of the medulla from the biopsy must be avoided. In tumours close to joints it is imperative that the biopsy is located in such a way that the joint cavity is not contaminated by needle placement.

With all specimens care is taken to reduce tissue crush. Specimens should ideally be delivered fresh and unfixed to the histopathology department allowing imprint preparation to be made to ensure that lesional tissue is present in a sample. In cases of potential infective diagnoses, additional samples must be sent for microbiology examination.

Any biopsy is a surgical procedure and is associated with potential risks and complications. A poorly executed or interpreted biopsy increases the risk of error in diagnosis, wound complication, and a potentially deleterious change in treatment course or outcome, including the need for a more complex resection or amputation, increased risk for local recurrence, or death. because of the small sample size, needle biopsies are more subject to: 1) errors in diagnosis, particularly in grade of heterogeneous tumours; 2) anatomical misrepresentation resulting in obtaining non-diagnostic or indeterminate tissue; and 3) inability to perform research studies or special diagnostic studies such as cytogenetics or flow cytometry which may be necessary for definitive diagnosis, as is the case with small blue cell malignancies.

Therefore, when the diagnosis, based on needle biopsy, is in reasonable doubt having been reviewed by a specialist musculoskeletal pathologist, the clinician must proceed as though no biopsy had been performed. Frequently, this means repeating the biopsy, usually by the conventional open technique.

Optimizing surgical margins, and thus local control of tumour, requires accurate preoperative imaging of the tumour and surrounding anatomy. Surgical planning includes the selection of an appropriate position for the biopsy site. Improperly placed biopsy incisions may prevent the possibility of limb salvage surgery.

Needle biopsy of bone and soft tissue tumours is both safe and accurate when performed in a specialist centre. Radiologists are best placed to perform most biopsies as they have a number of imaging modalities at their disposal, enabling them to take the most appropriate tissue sample. The main disadvantage to pathologists is the small amount of tissue available.