 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1998 The Journal of Bone and Joint Surgery, Inc.
Current Concepts Review - Evaluation and Staging of Musculoskeletal Neoplasms* , CHICAGO, ILLINOISInvestigation performed at the Section of Orthopaedic Surgery and Rehabilitation Medicine, Department of Surgery, University of Chicago Medical Center, Chicago
Staging is the process of classifying a tumor, especially a malignant tumor, with respect to its degree of differentiation as well as its local and distant extent, in order to estimate the prognosis for the patient. Evaluation of the histological and imaging studies of a tumor allows the physician to formulate a rational strategy for treatment with operative intervention, medical therapy, or radiation therapy. Staging a tumor is a valuable process in the management of individual patients. In contrast, staging systems are based on clinical, radiographic, and histological features that are believed to be of prognostic importance and may be useful when outcomes are compared between groups of patients. Staging systems may be used to compile data on patient survival and recurrence for tumors that have similar features and to evaluate the effectiveness of therapy for such tumors. The first staging system to be described, as far as we know, was for cervical carcinoma. The system, which was developed by the League of Nations nearly fifty years ago5, was based on the TNM system, in which T referred to the extent of the primary tumor; N, to the presence or absence of nodal metastases; and M, to the presence or absence of distant metastases. Staging systems such as this one have been valuable in the care of patients who have homogeneous and relatively common neoplasms. Treatment protocols based on the TNM system are used to determine which patients should be managed with an operation, radiation, or chemotherapy. There is no universally accepted staging system for musculoskeletal neoplasms because of the low incidence of such tumors, their heterogeneous nature and unpredictable behavior, and disagreement as to the relative importance of various prognostic factors. Also, an important prognostic variable in a staging system for malignant musculoskeletal tumors, unlike a system for carcinomas, is the grade of the tumor, as the grade represents an estimation of the likelihood of metastasis based on histological measures of cell differentiation and growth. Thus, the grade must be included in any staging system for musculoskeletal neoplasms. A high grade and evidence of metastasis are associated with a poor prognosis for all neoplasms of bone or soft tissue regardless of the staging system that is used. However, the staging systems differ in the way that the extent of the tumor is measured. The size of the tumor, the anatomical compartment where it is located, and the depth of the tumor may be considered important variables. In many ways, the differences among staging systems reflect the various perspectives of those who use them. Some staging systems are valuable in the determination of a prognosis, whereas others may be more useful in the determination of an operative strategy. The systems that are most commonly used for the staging of malignant soft-tissue tumors are that of the American Joint Committee on Cancer4, which is based primarily on the Memorial Sloan-Kettering staging system that was first described by Hajdu20, and the system of the Musculoskeletal Tumor Society, which was described by Enneking et al.16. There is a better consensus with regard to the important prognostic factors and the staging of bone tumors. The Musculoskeletal Tumor Society adopted staging systems, which were described by Enneking et al.15,16, for both benign and malignant bone tumors, and the American Joint Committee on Cancer developed, with minor adaptations, a similar staging system for malignant bone tumors3. As far as we know, none of the staging systems have been evaluated statistically in large multicenter studies. In fact, there is disagreement, even among experienced physicians, about the value of important variables, such as the grade, in the evaluation of particular tumors. Moreover, because some systems are based on studies that were done before the availability of sophisticated imaging methods and the use of adjuvant therapy, their usefulness is limited.
The true value of staging is not determined by the particular system that is used but rather it depends on the systematic evaluation of a patient by the physician with use of imaging and biopsy studies. This process allows the surgeon to determine the type and the extent of the operation that is necessary for a specific type of tumor in a particular anatomical location. By providing an indication of the behavior and the anatomical extent of the tumor, the operative margins that are necessary for the eradication of the tumor, and the value of adjuvant therapy, the evaluation allows the surgeon to plan and carry out a specific operative procedure (Figs. 1 and 2).
Staging systems, in contrast, are valuable in the identification of groups of patients who are at extremely low risk for recurrence or for metastatic disease as well as those who are at high risk. Such systems are less useful in predicting the outcome for patients who have an intermediate-stage tumor. Patients who are considered to be at high risk would be candidates for participation in investigational studies of novel treatments. Staging systems are used primarily in cooperative studies and in the collection of objective data for the evaluation of these treatment protocols. Staging systems are not particularly helpful in the estimation of the prognosis for individual patients.
Evaluation of Patients Radiographs provide important information with regard to the appearance, intraosseous extent, and internal characteristics (such as ossification or calcification) of bone tumors as well as information with regard to the clarity of the margin between the bone tumor and the host bone (Fig. 1). Cortical destruction and periosteal new-bone formation are associated with invasive neoplasms. However, conventional radiographs often are not useful in the evaluation of soft-tissue masses (Fig. 2). Computed tomography allows the surgeon to assess the character of bone tumors. Although not as accurate as magnetic resonance imaging in the determination of the extent of soft-tissue and bone-marrow involvement, computed tomography is superior in the evaluation of cortical erosion, fracture, and internal characteristics such as ossification or calcification. Computed tomography also reveals associated soft-tissue masses and important adjacent anatomical structures, especially in the pelvis. Computed tomography of the lungs is an essential part of staging for a patient who has a suspected malignant musculoskeletal tumor. The lung is the most common site of metastatic disease in these patients, and computed tomography is a more sensitive method for the detection of metastatic disease than is conventional radiography of the chest. In addition, computed tomography of the pelvis and abdomen may detect multifocal disease in patients who have liposarcoma. Magnetic resonance imaging, however, is usually the best imaging system for the evaluation of a soft-tissue mass or the extent of soft-tissue or bone-marrow involvement by a bone tumor. Magnetic resonance imaging demonstrates the extraosseous and intraosseous extent of a bone tumor as well as the involvement of a joint. It also detects skip metastases, which are defined as neoplastic foci that are located at a distance from the principal tumor mass but are usually in the same anatomical compartment. Magnetic resonance imaging demonstrates the depth, size, and local extent of soft-tissue masses as well as their relationship to the surrounding neurovascular structures. In addition, magnetic resonance imaging may be diagnostic as it is for most deep or superficial lipomas. Scintigraphy is a sensitive but non-specific tool for the assessment of bone tumors. The value of technetium bone-scanning in the evaluation of patients who have a bone tumor is in the detection of polyostotic bone disease or bone metastases. Bone-scanning cannot distinguish benign from malignant lesions, and it cannot accurately determine the extent of a tumor. Similarly, technetium bone-scanning of soft-tissue tumors is of little value except for the detection of metastases to bone. Bone metastases are the second most common type of distant disease associated with soft-tissue sarcomas. Periosteal or osseous involvement is visualized better with magnetic resonance imaging or computed tomography than with bone scintigraphy. Gallium scanning is of limited value and is not used routinely in the evaluation of bone and soft-tissue tumors. Exceptions are childhood rhabdomyosarcoma, epithelioid sarcoma, and synovial sarcoma, in which non-pulmonary lymphatic metastases may be detected. Positron-emission tomography and thallium scanning are considered investigational tools at this time. They may be of value in the determination of local disease, non-pulmonary metastases, response to therapy, or recurrence of a tumor. The way in which these tests complement plain radiography, computed tomography, and magnetic resonance imaging has yet to be determined. In summary, imaging studies are an essential part of the process of staging a bone or soft-tissue tumor. These tests allow the clinician to arrive at a differential diagnosis before biopsy. If these studies, in addition to clinical information, suggest a primary malignant bone tumor, the patient should be referred without additional tests or biopsy to the surgeon who ultimately will be responsible for the treatment. It is often impossible to distinguish benign from malignant soft-tissue tumors on the basis of imaging tests. Large, deep tumors are more likely to be malignant, and the patient should be referred to a musculoskeletal oncologist before biopsy. In any case, when performing a biopsy or an excision, or both, the surgeon should adhere to the principles that follow.
Biopsy The biopsy is a technically simple procedure, but it should be done by a thoughtful, knowledgeable, and experienced surgeon. A poorly placed biopsy incision, a poorly performed biopsy, or the complications of a biopsy make it difficult to salvage an extremity and, in some instances, may affect the survival of the patient31. Histological evaluation of the biopsy specimen determines whether the tumor is benign or malignant. If it is malignant, the grade is estimated on the basis of the tumor cellularity, nuclear atypia, mitotic activity, and necrosis.
Systems have been developed for the staging of benign and malignant bone tumors and malignant soft-tissue tumors. These systems represent an attempt to predict the prognosis and to evaluate the effect of therapeutic intervention by stratifying similar tumors according to various prognostic factors. Commonly used factors are the grade, size, and compartmentalization of the tumor and the presence or absence of metastases. The true value of staging systems is in the evaluation of the results of treatment in a large number of patients. Staging systems are of little value in the formulation of an operative strategy for an individual patient.
Soft-Tissue Sarcomas in Adults
The grade of a tumor is a histological measure of its tendency to metastasize. Lesions that are characterized by cellular atypia, frequent mitoses8, extensive necrosis12, marked vascularity, and poor differentiation seem to be more likely to metastasize than tumors that exhibit fewer of these histological features. However, even among expert pathologists, there is a great deal of disagreement with regard to the histological grading of soft-tissue sarcomas1,39. Despite this disagreement about the appropriate number of grades and the manner of assigning them, it is clear that the grade should be weighted heavily in any staging system19,23,32,36,41-43. The staging system described by Enneking et al.16 defines only two grades, which are based on histological criteria. A high-grade tumor is considered to have a metastatic potential of more than 25 per cent; however, this is, at best, a crude estimate. The staging system of the American Joint Committee on Cancer4 defines four grades: well differentiated (G1), moderately well differentiated (G2), poorly differentiated (G3), and undifferentiated (G4). However, the system essentially contains two grades; well differentiated and moderately well differentiated (G1 and G2) are grouped as low grade, and poorly differentiated and undifferentiated (G3 and G4) are considered high grade. Size is an important prognostic factor11,28,36,42,47. It reflects not only the biological behavior of the tumor itself but also the ability of the host to inhibit the growth of the tumor. Size is likely to be a continuous variable, with an increased risk for both local recurrence and metastases associated with larger tumors. Musculoskeletal neoplasms, like most tumors, tend to grow in a centrifugal manner, but the size and shape of the tumor are influenced by the local anatomy, the rate of growth, the response of the host, and the invasiveness of the tumor. A perilesional reactive zone or pseudocapsule of reactive fibrous tissue, inflammatory cells, vascular structures, and microscopic foci surround these tumors. An excision of a tumor performed in the plane of the pseudocapsule is defined as a marginal resection and is associated with a high risk of local recurrence. The preferred operative treatment of any malignant musculoskeletal neoplasm is a wide resection, which is described as an excision of the tumor in continuity with surrounding circumferential normal tissue16. The clinical quantity of normal tissue as a measurable dimension and the necessary quality of the margin (fascia compared with fat) have not been defined. Musculoskeletal neoplasms may be intracompartmental or extracompartmental16. Intracompartmental tumors are bounded in all dimensions by natural barriers (for example, deep fascia) to extension of the tumor. Extracompartmental tumors are found in an extracompartmental location (for example, the popliteal space) or have extended beyond the natural barriers either by growth or by contamination from fracture, hemorrhage, or an operative procedure. Extracompartmental extension may be an indication of invasiveness. Compartmentalization was an important concept in the selection of operative treatment for the control of a local tumor before the introduction of modern imaging techniques, adjuvant chemotherapy, and radiation therapy. Before those techniques and treatments became available, a radical resection—that is, an excision of the entire compartment—was recommended in order to minimize the potential for local recurrence due to microscopic residual tumor. Small intracompartmental tumors can be resected with wide margins of normal tissue. This offers a high probability of local control, and adjuvant radiation therapy may not be needed. Extracompartmental tumors, however, are commonly adjacent to bone or neurovascular structures and only a marginal resection can be performed. Adjuvant chemotherapy or radiation therapy may be necessary to minimize the prevalence of local recurrence. The importance of the depth of the tumor as an independent prognostic factor is controversial. In one study44, 129 (32 per cent) of 404 soft-tissue sarcomas in adults were subcutaneous and these tumors generally were associated with a better prognosis than were deep sarcomas. Other studies11,20,36 have shown a similar percentage of sarcomas in a subcutaneous location. Some studies have shown depth to be an independent prognostic factor, whereas others have suggested that subcutaneous sarcomas are smaller than deep sarcomas at the time of diagnosis and that this smaller size is more important for predicting the outcome. The presence of a regional or distant metastasis is the most important prognostic variable, and it needs to be included in any staging system. Sarcomas most commonly metastasize to the lungs through a hematogenous route. Metastasis of musculoskeletal neoplasms in adults, with the exception of synovial sarcomas and epithelioid sarcomas, infrequently is clinically evident in regional lymph nodes34. According to the systems of Enneking et al.16 and the American Joint Committee on Cancer4, regional and distant metastases confer an equally poor prognosis.
Staging System of the American Joint Committee on Cancer Critics of the system of the American Joint Committee on Cancer point out that it is based largely on single-institution studies that have demonstrated that size, grade, and depth are independent prognostic factors. In addition, the system has not been subjected to multi-institutional tests of validity—that is, its ability to predict survival and the system's intraobserver and interobserver error. In fact, it is possible that the prognostic variables are time-dependent. High-grade tumors have been noted to be associated with early metastatic disease, whereas the negative effect of a large size on survival may not be seen for several years38. There is no means in the current system for differentially weighing the variables on the basis of time. In addition, other studies36,44 have demonstrated that the depth of the tumor is not as important a prognostic factor as are the grade and the presence or absence of metastasis. It is clear that not all prognostic variables should be weighted equally. The use of five centimeters as an important dimension in the determination of the prognosis with this system must be considered arbitrary. Certainly, patients who have a very small tumor, which often is subcutaneous, have an excellent prognosis. However, to our knowledge, no study has shown that five centimeters is a critical measure for distinguishing low, intermediate, and high-risk populations of patients. Size is probably a continuous variable, and it may be that there is no single important dimension or that the volume or weight of the tumor may be more predictive of outcome. An additional criticism of the system is that it was based on studies that included a number of lesions that were not located in an extremity but, instead, were in the retroperitoneum, thorax, head, or neck. Tumors in these locations are more difficult to eradicate and are more likely to recur than are those in the extremities. It is clear that sarcomas of the extremities should be considered separately. This system may be effective, however, in the identification of patients who have a very low risk for metastases (that is, patients who have a small, low-grade, subcutaneous lesion) as well as patients who have a high risk for metastatic disease (that is, those who have a large, high-grade tumor). Clearly, any investigation of the value of any adjuvant therapy should be directed at patients who have a high risk for metastatic disease.
Staging System of the Musculoskeletal Tumor Society The anatomical site (T) may be either intracompartmental (A) or extracompartmental (B). At present, this information is obtained preoperatively by means of computed tomography and magnetic resonance imaging studies and is confirmed by the operative findings. A tumor is classified as intracompartmental if it is bounded by natural barriers to extension, such as bone, fascia, synovial tissue, periosteum, or cartilage. An extracompartmental tumor may be primary (occurring in an extracompartmental location) or secondary (an intracompartmental tumor that has penetrated into an additional compartment by means of natural extension, fracture, hemorrhage, or contamination at the time of the operation). A tumor is assigned to stage III if a metastasis is present in the regional lymph nodes or at a distant site (M1). Thus, according to the system of Enneking et al.16, a tumor may be either stage I (a metastatic potential of 25 per cent or less) or stage II (a metastatic potential of more than 25 per cent) depending on the grade, with additional classification of the stage as A or B depending on the site, or it may be stage III if a metastasis is present. The system of the Musculoskeletal Tumor Society has been criticized for being overly simplistic and outdated, and most pathologists have difficulty using a two-grade system. In addition, a large number of musculoskeletal tumors do not fit neatly into the definition of intracompartmental or extracompartmental. For example, subcutaneous sarcomas, which account for 30 per cent of all sarcomas, are poorly defined by this system. Many tumors are either bicompartmental or extracompartmental. More importantly, however, the system of the Musculoskeletal Tumor Society was defined before the availability of advanced imaging (that is, magnetic resonance imaging) and before the widespread use of adjuvant therapy. At that time, radical or whole-compartment resection or amputation was recommended so that skip lesions would not be missed and the surgeon would not inadvertently cut across a tumor. At present, magnetic resonance imaging allows the surgeon to define precisely the extent and location of a tumor, including any skip metastases, so that a wide resection can be performed in most patients. Adjuvant chemotherapy or radiation therapy may also decrease the prevalence of local recurrence. Similar to the system of the American Joint Committee on Cancer, the system of the Musculoskeletal Tumor Society has never been subjected to statistical validation tests, to our knowledge. The system is valuable, to some extent, in the formulation of an operative strategy, but it is less useful in the evaluation of other therapeutic modalities. Both of these staging systems for soft-tissue sarcomas directly or indirectly incorporate the important prognostic variables of grade, compartmentalization, size, depth, and metastasis to regional lymph nodes or a distant site. They differ with respect to the number of histological grades and the relative assignment of other prognostic factors. It is difficult to devise an adequate grading system that is acceptable to all pathologists and that can be used by all in a reproducible manner. The size of the tumor is considered an important prognostic factor in the system of the American Joint Committee on Cancer. Although size may influence the prognosis, the proximity of a tumor to neurovascular structures and the size of the tumor relative to the compartment involved may be of more practical importance from an operative point of view. The system of Enneking et al.16, which uses the concept of anatomical compartments, addresses the anatomical site from more of an operative perspective than does the system of the American Joint Committee on Cancer4. The staging system of the American Joint Committee on Cancer4 now includes the depth of the tumor as a separate prognostic variable. Metastasis to the regional lymph nodes is infrequent in adults who have a soft-tissue sarcoma, except in those who have an epithelioid sarcoma or a synovial sarcoma, and the prognosis associated with such a metastasis is similar to that associated with a distant metastasis34. For that reason, no practical distinction is made between lymphatic or distant metastasis in either the staging system of the American Joint Committee on Cancer4 or that of Enneking et al.16. Thus, in the two staging systems for soft-tissue sarcomas in adults, the histological grade and the presence or absence of metastasis are the most important prognostic variables. The size, depth, and anatomical site of each tumor are deemed important, but the hierarchy of their importance is unknown and how they should be weighted in a staging system remains unresolved.
Staging of Childhood Rhabdomyosarcoma In 1972, the first Intergroup Rhabdomyosarcoma Study, a prospective, randomized investigation, was begun to conduct a systematic clinical trial for children with rhabdomyosarcoma who had not had previous treatment21,22. The investigators evaluated a comprehensive treatment program for these children and assessed the effectiveness of various chemotherapy and radiation-therapy regimens. The patients were managed with the various regimens according to specific clinical groups (Table IV), which, in the absence of metastatic disease, were determined by the type of operative procedure that had been performed initially22. Twenty-one per cent (117) of 561 patients had a primary tumor in an extremity, and the results of the study suggested a poorer prognosis for those children than for those who had a tumor in an orbital or genitourinary location. In a subsequent study, lesions of the extremity were assigned a higher stage and were treated with a more intensive chemotherapy regimen24. In addition, the alveolar histological subtype appeared to be associated with a higher prevalence of relapse and with a low rate of survival in the Intergroup Rhabdomyosarcoma Study22. However, it is likely that these two factors were acting in combination because 38 per cent (forty-five) of the 117 lesions that involved an extremity were of the alveolar subtype compared with only 20 per cent (115) of 561 lesions in the group as a whole. In the second Intergroup Rhabdomyosarcoma Study24, although the relative contributions of site and histological subtype were unknown, lesions in the extremity were treated with an intensive chemotherapy regimen, placing these patients in clinical group III or IV.
There is controversy with regard to the validity of the staging system used in the Intergroup Rhabdomyosarcoma Study. Ideally, a staging system characterizes the natural history and the extent of the disease, not the initial therapeutic intervention14,25,27,37. Critics of the staging system used in the Intergroup Rhabdomyosarcoma Study have also noted that, as adjuvant chemotherapy and radiation therapy have become more effective, the role of operative treatment has become more limited27,37. When managed with adjuvant radiation and chemotherapy, children who have a small, non-invasive tumor often survive despite incomplete resection. These problems with the classification system of the Intergroup Rhabdomyosarcoma Study led to the development of other staging systems for childhood rhabdomyosarcoma. The International Union against Cancer proposed a TNM classification and staging system for tumors, including rhabdomyosarcoma, in children27,37,48. The system differs from that of the American Joint Committee on Cancer for sarcomas in adults4 in that T indicates confinement or lack of confinement of the tumor to the organ or the muscle of origin, rather than the size of the tumor. T1 refers to a tumor that is confined to the muscle group of origin, and T2 refers to a tumor with local extension through the fascial plane, extending beyond the muscle of origin or invading neural, vascular, or osseous structures. The system of the International Union against Cancer places more emphasis on the invasiveness of the tumor than on the size of the tumor. However, it is important to note that, in two studies, 91 per cent (twenty) of twenty-two tumors that were larger than five centimeters exhibited local extension25,27. The American Joint Committee on Cancer modified the staging system for soft-tissue sarcoma in children to include the size of the tumor, with use of the designation a to indicate a tumor that is five centimeters or less in its greatest dimension and b to indicate a tumor that is more than five centimeters (Table V)2. The most recent edition of the staging manual of the American Joint Committee on Cancer eliminated the separate section on soft-tissue tumors in children.
The two staging systems for rhabdomyosarcoma are similar in that a distant metastasis places patients in the highest category of risk. If one assumes that large, invasive tumors are not amenable to complete resection, then the systems are even more alike. The fundamental difference between them is that the staging system of the Intergroup Rhabdomyosarcoma Study is based primarily on the level of operative resection, with the local extent of disease as a secondary consideration; the TNM system of the International Union against Cancer, conversely, stratifies patients into different prognostic groups on the basis of the invasiveness of the tumor. Recently, the Pediatric Oncology Group and the Children's Cancer Group for the Study of Rhabdomyosarcoma developed a new staging system for patients who have rhabdomyosarcoma (Table VI) 49. This system uses the TNM groupings, and anatomical site is considered to be important in determining the prognosis for patients who have rhabdomyosarcoma.
Staging Systems for Sarcomas of Bone
System of the Musculoskeletal Tumor Society
System of the American Joint Committee on Cancer
A careful comparison of the systems for the staging of bone sarcomas developed by the American Joint Committee on Cancer3 and by Enneking et al.16 (Tables II and III) shows only minor differences. The system of the American Joint Committee on Cancer defines four stages, but they are grouped in such a way that, in practice, there are only three. It also seems that only two histological grades are needed for bone sarcomas and that clinical and radiographic criteria should be considered in the determination of grade. In essence, the staging system for bone sarcomas developed by the American Joint Committee on Cancer3 is an adaptation of the system of Enneking et al.16. The anatomical site (T) is essentially identical in both systems. The only real difference is that the system of the American Joint Committee on Cancer has a designated but undefined stage III. It appears that the American Joint Committee on Cancer used the four-stage system as a standard in the staging of any type of tumor, perhaps because it is more acceptable worldwide than a three-stage system. Similar to the staging systems for other musculoskeletal neoplasms, neither staging system for bone sarcomas has been subjected to tests of validity, to our knowledge. Although it is clear that intraosseous low-grade lesions are associated with a more favorable prognosis and a metastasis is associated with a poor prognosis, the predictive value of staging systems with regard to estimating the survival of patients who have a bone tumor is not known.
Staging of Benign Tumors of Bone
Latent benign bone tumors are classified as stage 1. Such tumors are usually asymptomatic and are commonly discovered as an incidental radiographic finding. Thick, dense reactive bone is present on both plain radiographs and computed tomography scans. These lesions remain dormant and heal spontaneously in many instances. Examples include fibrous cortical defects, non-ossifying fibromas, and osteoid osteomas. Operative treatment usually is not necessary. Stage 2 indicates an active benign bone tumor. These tumors are actively growing and enlarging and therefore may be associated with physical signs and symptoms. In some instances, destruction of the cortex leads to a pathological fracture. A stage-2 tumor has a thin rim of reactive bone but remains intracompartmental. Examples include slowly expanding aneurysmal bone cysts and chondroblastomas. Operative curettage and bone-grafting are commonly used procedures for such tumors. A stage-3 benign bone tumor has little associated reactive bone and often breaks through the bone cortex. Stage-3 lesions are likely to be growing faster than are stage-1 or 2 lesions, and they are found both clinically and radiographically to be more locally invasive. Pain and tenderness are common, and a mass may be detected on physical examination. Radiography, computed tomography, and magnetic resonance imaging demonstrate extension into surrounding tissues. Malignancy may be suspected in such instances. Some examples are giant-cell tumors and rapidly expanding aneurysmal bone cysts. Curettage and bone-grafting or insertion of methylmethacrylate may be used, depending on the diagnosis. A more extensive operation involving arthrodesis or prosthetic reconstruction may be necessary in patients who have severe loss of bone. Campanacci et al.9, in 1987, described a grading system for giant-cell tumors of bone that is based solely on radiographic appearance. A grade-I tumor is well demarcated by a thin rim of reactive bone, with an intact or slightly thickened cortex without deformation. Grade-II giant-cell tumors have relatively well defined margins but no radiopaque rim. The cortex is thin and expanded but intact. Grade-II lesions associated with a fracture are considered separately. Grade-III lesions demonstrate permeative growth in bone, with destruction of bone cortex and an associated soft-tissue mass. These staging systems are valuable in defining the characteristics that are associated with locally invasive behavior of benign bone tumors. However, most often the stage is merely a reflection of the histological diagnosis. Similarly, treatment options and oncological issues such as local recurrence are related more to the diagnosis than to the stage. More troubling, however, is a high level of interobserver disagreement that limits the clinical application and usefulness of these systems.
On the basis of the histological characteristics of the tumor, the local extent, and the presence or absence of metastasis, a therapeutic strategy that includes operative treatment, chemotherapy, and radiation therapy can be devised. The relationship between the stage of the tumor and the prognosis is clear, but at the present time the stage plays only a remote role in decisions regarding operative management, especially with the advent of effective preoperative and postoperative chemotherapy and radiation therapy. These treatment modalities, with the addition of more sophisticated and accurate imaging systems, make the stage of the tumor (excluding instances of widespread metastases) of limited usefulness in planning the operative management. Thus, with respect to decisions regarding the operative procedure, the anatomical extent of the tumor is more crucial than the stage of the tumor. In no other orthopaedic subspecialty has the operative procedure to be carried out in a particular patient become so individualized in relation to the location, size, anatomical extent, and characteristics of the lesion as in musculoskeletal tumor surgery. As already described, the type of resection can be defined as intralesional (curettage or debulking), marginal (through the pseudocapsule or perilesional zone surrounding the tumor), wide (in normal tissue outside the pseudocapsule), or radical (the entire compartment) (Table IX)16. From an oncological point of view, whether the margin is achieved by a local procedure or by amputation is irrelevant. When the anatomical location of the tumor is such that the same margin can be achieved with or without an amputation, the surgeon and the patient of course would select the limb-saving procedure. However, if the anatomical location of the tumor makes it necessary to accept a margin that is less than ideal in order to preserve the limb, then there is a conflict between the desire of the surgeon and the patient to preserve the limb and the knowledge that preservation of the limb may lead to a poor oncological outcome. With effective adjuvant radiation and chemotherapy, many sarcomas can be treated with marginal resection, with a low risk of local recurrence. However, such an option requires frank discussion between the surgeon and the patient about the implications of such an operative procedure. All oncologists, whether they specialize in surgical, medical, radiation, or pediatric oncology, realize that local control does not guarantee long-term survival. However, the lack of local control of some sarcomas, such as Ewing sarcoma, osteosarcoma, and axial chondrosarcoma, most likely leads to or is associated with the death of the patient. These judgments must be made on an individual basis.
General Recommendations for Operative Margins for Musculoskeletal Tumors
Excision through the perilesional reactive zone is indicated for benign non-infiltrating soft-tissue tumors, recurrent stage-2 or stage-3 benign bone tumors, and possibly for some primary stage-3 tumors. Also, such excisions are used with success in patients who have a high or low-grade bone or soft-tissue sarcoma that has shown histological evidence of extensive necrosis in response to preoperative chemotherapy and radiation therapy. However, a higher prevalence of local recurrence can be expected in patients who have an excision through the perilesional reactive zone when no effective adjuvant treatment was administered. Wide margins are desirable for a few recurrent stage-3 benign bone tumors and for most bone and soft-tissue sarcomas when the margins have been adequately defined by contemporary staging techniques, such as magnetic resonance imaging or computed tomography. Before the introduction of adequate imaging techniques and adjuvant treatments, wide margins often were associated with a high rate of recurrence because of poor preoperative definition of the location and histological extension of the tumor. Radical resections, formerly the procedure of choice for most high-grade sarcomas, are now reserved for recurrent sarcomas after a pathological fracture through the bone sarcoma or when the margins cannot be defined accurately. Radical resections therefore are infrequently used for musculoskeletal sarcomas at the present time.
Examples
Operative Margins for Bone Tumors Wide resections usually are done for the typical stage-IIB osteosarcoma of the distal part of the femur. However, in some instances, especially after a displaced pathological fracture, a wide transfemoral amputation must be performed to achieve a wide margin. Radical resections, which necessitate the removal of the entire bone of origin of the tumor, are rarely performed for the treatment of bone tumors. A radical (above-the-knee) amputation may be indicated for a patient who has a large stage-IIB osteosarcoma of the proximal part of the tibia or for a patient who has a displaced pathological fracture associated with a stage-IIB osteosarcoma.
Operative Margins for Soft-Tissue Tumors A wide resection is performed for the treatment of most soft-tissue sarcomas. It is commonly performed for the treatment of small subcutaneous sarcomas or small-to-moderately sized deep soft-tissue sarcomas. If the soft-tissue sarcoma is in the distal part of the leg, then a wide amputation may be preferable. A radical resection for the treatment of a soft-tissue sarcoma can be achieved by excision of the entire anatomical compartment from origin to insertion. This can be performed with either a resection or an amputation. Radical resections are uncommon and are usually reserved for very large, deep intracompartmental tumors. A radical amputation is indicated for some massive sarcomas that encase vessels and nerves or for some recurrent sarcomas.
The recent emphasis on limb salvage rather than amputation for the treatment of bone and soft-tissue sarcomas of the extremities necessitates a therapeutic approach with multiple modalities in which a limited operative procedure is combined with adjuvant or neoadjuvant radiation therapy and chemotherapy. An essential aspect of the proper application of these treatment modalities is a fundamental knowledge of the factors that affect the survival of the patient. In the staging systems that we described, only tumor-related factors—that is, the grade, site, depth, size, and extent—are addressed. Although these variables reflect, to some degree, aspects of the complex host-tumor interaction, our understanding remains crude and superficial. In the coming decades, other biological factors, both cellular and molecular, will be identified as being fundamentally responsible for the behavior of a tumor. These factors will be even more important for the determination of the prognosis and will probably be included in future staging systems. Advances in the staging and classification of musculoskeletal tumors are likely to occur at the molecular and molecular-cytogenetic levels18. These approaches have been used successfully for some time in the clinical evaluation of malignant hematological disease. However, they have not been applied to solid tumors until recently because of the heterogeneous nature of the tumors, the difficulty in establishing cell cultures, and the complex and sometimes inconsistent karyotypic alterations. It is likely that the identification of specific genetic abnormalities will have prognostic importance. Although it is not possible to present a complete summary of these genetic abnormalities and their impact on survival, several examples, some of which might be included in future staging systems, can be described. One example of an abnormality that has recently been found to be associated with a poor prognosis is the presence of two oncogenes in osteosarcomatous tissue. Barrios et al.7 noted that when amplification of the c-myc gene was found in patients who had osteosarcoma, overall and disease-free survival was shorter than when no amplification was seen. Onda et al.35 demonstrated that overexpression of ErbB-2 was associated with early metastasis of osteosarcoma and a poor rate of survival. ErbB-2 is an oncogene that has been previously implicated as an indicator of a poor prognosis for patients who have carcinoma of the breast50. Two tumor-supressor genes commonly found in sarcomas are the p53 and retinoblastoma susceptibility (Rb) genes. Lonardo et al.29 showed that p53, a cell-cycle regulator involved in apoptosis, was mutated in approximately 27 per cent (twenty-two) of eighty-three patients who had osteosarcoma, but its prognostic importance was unclear. Not all reports have been in agreement as to whether the gene is likely to indicate a good or a poor prognosis30,33,45. The Rb gene is the prototypical tumor-supressor gene and was noted to be altered in more than 68 per cent (twenty-five) of thirty-seven osteosarcomas that were examined45,52. Feugeas et al.17 demonstrated an association between loss of heterozygosity of the Rb locus and a poor rate of survival. Evaluation of the multidrug-resistance gene that codes for P-glycoprotein expression may allow a prediction of the response to chemotherapy and serve as a prognostic marker10. P-glycoprotein is a transmembrane energy-dependent efflux pump that may allow cells to become resistant to certain anticancer drugs by preventing the accumulation of these agents within the cell. The overexpression of the multidrug-resistance gene (MDR-1) has been associated with a poor prognosis for patients who have osteosarcoma6,51. Whether the poor prognosis is due to the drug resistance or to the increased ability of the tumor to metastasize has not been determined definitively. However, recent evidence has strongly suggested a model in which P-glycoprotein overexpression produces a tumor that is less likely to metastasize initially but that is capable of surviving chemotherapy to metastasize at a later date26. Cytogenetic research has revealed consistent chromosomal translations in some bone and soft-tissue sarcomas (Table XI). In Ewing sarcoma, a characteristic gene translocation that results in a fusion gene has been identified. The fusion gene EWS-FLI is unique to that tumor in that its sequence is determined by two different genes on two different chromosomes juxtaposed by the translocation event13. The EWS portion of the gene is derived from chromosome 22, and the FLI portion is derived from chromosome 11. The messenger RNA associated with these proteins can be used for diagnostic and therapeutic benefit. The molecular technique of reverse transcriptase-polymerase chain reaction (RT-PCR) transcribes the mRNA to DNA and amplifies it to measurable levels. Reverse transcriptase-polymerase chain reaction does not require viable cells for cell culture as are needed in cytogenetic analysis, and it is capable of detecting fusion-gene expression from a small number of malignant cells in a background of normal tissue. This procedure is now being used in some centers for the evaluation of the extent of residual disease after resection, in surveillance for metastasis, and in preparation for bone-marrow transplantation. In addition, some evidence suggests that variations in the fusion-gene subtypes of a particular tumor may have prognostic importance53.
In summary, staging is best defined as a process whereby, on the basis of clinical findings and imaging studies, a patient is evaluated with regard to the histological characteristics as well as the local and distant extent of a tumor. The determination of the specific operative procedure depends on whether the goal of treatment is curative or, more rarely, palliative; the operative margin that is necessary for obtaining that goal; the anatomical location of the tumor and its relationship to bone, joint, or neurovascular structures; and the perceived efficacy of adjuvant therapy. The classification and staging of soft-tissue and bone neoplasms are evolving processes. The staging systems that are currently available are based on histological and clinical observations, and they serve to guide therapeutic strategies. However, these systems are based on variables that are crude and of limited usefulness. New discoveries in the fields of molecular and cellular biology are being made and will be correlated with the clinical course and outcome. These findings will almost certainly change the manner in which these tumors are classified and staged and will predict more accurately the behavior of a given tumor in a particular patient. Operative treatment, chemotherapy, and radiation therapy can then be used in a rational and effective manner to improve the functional and oncological outcome.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
|
Top
Introduction
Evaluation of a Musculoskeletal...
