In soft tissue sarcoma, the primary tumor is categorized according to size (<5 cm = T1, >5 cm =

T2) and depth relative to the fascia (entirely above the fascia, or “superficial” = a; invading or entirely

below the fascia, or “deep” = b). Because depth of tumor seemed to add significant prognostic

information, this distinction was added to the 1998 edition of the AJCC staging system. One series of

215 patients with superficial extremity sarcomas documented a 10-year survival rate of 85% despite

that 53% of tumors were high grade and 25% of tumors were 5 cm or larger.49 Though the cutpoint

between small and large tumors is 5 cm, tumors larger than 10 cm seem to have an even worse

prognosis.50 When 316 patients with soft tissue sarcoma were grouped into four subgroups (<5 cm, 5

to 10 cm, 10 to 15 cm, and >15 cm), each subgroup had a distinct prognosis (84%, 70%, 50%, and

33%, respectively).51

2 For nonmetastatic sarcomas, location is a major determinant of survival (Fig. 108-7). Nodal

metastasis of soft tissue sarcomas is rare (far less than 5% of cases), with the exception of a few

histologic subtypes for which the incidence of nodal involvement may be between 10% and 20%. These

subtypes include angiosarcoma, embryonal rhabdomyosarcoma, epithelioid sarcoma, synovial sarcoma,

and clear cell sarcoma.52 In the past, node involvement in sarcoma received classification as stage IV

disease because of the associated poor prognosis. However, evidence suggested that lymph node

metastases were associated with significantly better outcomes than systemic disease in the seventh

edition of the AJCC staging system, and that survival was equivalent to stage III disease after directed

lymphadenectomy.53,54 Therefore, node involvement (N1) was reclassified as stage III disease in the

absence of distant metastasis (M0).

3184

Figure 108-7. Soft tissue sarcoma disease-specific survival by site. Hash marks indicate censored patients. (IA = intra-abdominal;

Retro = retroperitoneal). (From Brennan MF, Antonescu CR, Moraco N, et al. Lessons learned from the study of 10,000 patients

with soft tissue sarcoma. Ann Surg 2014;260:416–422.)

By far the most common site of metastasis is the lungs though metastases to other sites do occur.

Metastases to the liver should be considered with GIST or leiomyosarcoma.

TREATMENT

Principles of Surgical Resection

Complete surgical resection remains the cornerstone of treatment for soft tissue sarcomas. Performing a

wide excision of the tumor with negative margins is the goal of resection with curative intent. Simple

enucleation, if even possible, usually results in inadequate resection and should be avoided. Technical

aspects of resection must take into consideration the anatomic location and extent of disease. Often, soft

tissue sarcomas are surrounded by a zone of compressed reactive tissue that forms a pseudocapsule.

Care should be taken to avoid entry into the tumor and pseudocapsule during course of dissection.

Enucleation commonly results in microscopically positive margins and reresection can be considered

though anatomic location may preclude effective clearance of margins even with a second operation.

Metallic clips placed in the tumor bed following resection can help define the limits of resection and

aid in surveillance as well as the planning of future treatment. Suction drainage catheters are routinely

used to obviate postoperative seroma formation following resection of extremity or truncal sarcomas.

Drains should be placed close to the incision so that the site can be included in a postoperative radiation

field and to minimize the extent of proximal involvement if amputation ever becomes necessary. A

unique characteristic of sarcoma is the lack of metastasis to regional lymph nodes. If regional

lymphadenopathy is discovered in conjunction with a diagnosis of sarcoma, therapeutic

lymphadenectomy can be considered since clearance of disease may be associated with improved

outcomes.52,53 For selected histopathologic subtypes, in patients with clinically negative node

examination, there may be a role for sentinel lymph node biopsy to identify occult micrometastatic

disease.55

Taken together, sarcomas of the upper and lower extremities and trunk make up the majority of soft

tissue sarcomas. Large truncal tumors may require reconstruction with a myocutaneous flap or

prosthetic materials if resultant defects cannot be closed primarily (Fig. 108-8). Largely of historic

interest now, radical amputations were once the mainstay of treatment for extremity sarcoma, but

modern surgical approaches involve limb-sparing procedures (Fig. 108-9), which maximize functional

outcomes. As recently as the 1970s, more than 50% of all soft tissue sarcomas of the extremity were

treated with radical amputations. The accepted standard of care changed after the group at the National

Cancer Institute, led by Rosenberg,56 published the results of a randomized trial of limb-sparing

resection plus adjuvant radiation therapy compared to amputation, finding no differences in disease-free

survival rates (71% vs. 78%, respectively) and the overall survival rates (83% vs. 88%, respectively) at

5 years. This trial demonstrated that amputation was not mandatory and in 1985, a National Institutes

of Health consensus statement recommended limb-sparing procedures for most patients with high-grade

extremity sarcomas.57 Radical amputations, such as hemipelvectomy, hip disarticulation, or forequarter

amputation, are now reserved for patients who are not suitable candidates for limb-sparing approaches,

usually because of extent of disease, bony or joint invasion, or for otherwise unsalvageable recurrence

after previous limb-sparing surgery (Fig. 108-10). There is no improvement in survival with this

3185

approach under these circumstances and serious morbidity must be taken into account.58

Acceptance of multimodality therapy has led to the ability to perform wide excisions while preserving

function when treating extremity or truncal soft tissue sarcomas.5,59,60 Radiation can be employed in the

neoadjuvant or adjuvant setting, and evidence from randomized trials as well as retrospective studies

support similar local control and progression-free survival rates with both strategies. A phase III trial in

which patients were randomized to preoperative or postoperative radiation showed similar primary

outcomes, but found that preoperative radiation was associated with much higher wound complications

(35% vs. 17%, respectively) and this was especially true for lower extremity tumors. The interesting

tradeoff was a higher rate of late treatment effects in the postoperative radiation group, attributed to

relatively higher treatment doses and volumes.61,62

Judicious use of radiation has allowed for the sparing of major neurovascular structures, though if

necessary, arteries and/or veins can be resected en bloc with the tumor and subsequent reconstruction

can be done with autologous or prosthetic graft materials. In cases of neurologic compromise,

customized adjuncts from experienced physical medicine departments can help restore reasonable

function. For instance, resultant foot drop from peroneal neuropathy can be managed with an ankle–

foot orthotic and gait training. In some cases, resection should be planned in conjunction with a plastic

surgeon to provide complex soft tissue reconstruction with microvascular free flaps if the size of defect

following resection prohibits primary closure.

Adjuvant and Neoadjuvant Therapies

3 The role of external beam radiotherapy is best defined by a trial comparing limb-sparing surgery

alone with limb-sparing surgery with adjuvant radiation, showing improved 10-year local recurrence

rates with radiation. In high-grade sarcomas, the recurrence rate was 0% versus 22%, respectively,

though no significant effect on overall survival was appreciated.63 Certainly, in the setting of involved

surgical margins which could not otherwise be resected, adjuvant radiation therapy is associated with

improved local control.3,64

There is controversy regarding the optimal use of radiation in the preoperative setting compared to

the postoperative setting for extremity and truncal sarcomas. Radiation in the adjuvant setting with

either external beam radiation or brachytherapy has long been the standard approach.3 Typically, the

entire surgical bed and drain sites are included in the field, along with wide margins around the

sarcoma. Hemaclips marking the site of resection can be placed at the time of resection to help with

radiation planning. It is important that an individual approach be taken with each patient since radiation

can result in severe complications when used inappropriately (Table 108-4). For example, the entire

circumference of the extremity should not be irradiated because massive lymphedema may result.

Severe fibrosis, necrosis, fractures, and contractures can also occur with focused treatment and result in

impaired function. Wound healing may be compromised in patients undergoing adjuvant radiation and

in selected cases, use of rotational flaps for coverage can prevent or minimize chronic wound healing

problems.65 Similarly, radiation must be used judiciously in sensitive areas since nearby visceral

structures are exquisitively sensitive to higher doses of radiation. Dose-limiting toxicity must be

considered when recommending such therapies.

86

Figure 108-8. Chest wall sarcoma. A 20-year old man with a 7.5-cm right posterior chest wall epithelioid sarcoma involves the

11th and 12th ribs as seen on CT scan (A) and on physical examination (B). Chest wall resection was performed, including a

portion of diaphragm. Resulting chest wall defect (C) was repaired using prosthetic mesh (D).

An alternate approach to adjuvant radiation is brachytherapy, which involves the placement of

multiple catheters or seeds in the tumor resection bed to administer iridium.192 Unlike the several week

course needed to complete external beam radiation, a course of brachytherapy can be completed in a

few days’ time. In theory, brachytherapy produces less radiation scatter in critical areas, decreasing

toxicity while obtaining equivalent functional outcome.66 However, brachytherapy requires special

equipment and can involve technically complex treatment planning by an experienced radiation

oncologist. From a therapeutic standpoint, brachytherapy and external beam radiation appear to be

equivalent when properly administered. There is increasing interest in the use of intraoperative

radiation therapy (IORT), though its effectiveness is unproven. If IORT is employed, it is currently done

in conjunction with a planned external beam boost dose.67

Patients with high-risk (large [>10 cm], high-grade) extremity sarcomas should be considered for

preoperative treatment with chemotherapy or with chemoradiation since overall local control rates are

disappointing with postoperative radiation alone.59 Local control rates were considerably higher when

large tumors were treated before surgery and in some cases, tumors initially considered unresectable

without amputation shrank sufficiently to permit limb-sparing resection. There are several benefits of

preoperative radiation. With the tumor in situ, there tends to be a smaller overall treatment volume and

the theoretical advantages of more effective radiation because it is being delivered to an undisturbed

tumor bed. There is also the potential benefit of decreased seeding during the course of resection and

improved margin status because the radiation may shrink the tumor’s pseudocapsule and render it

relatively acellular. However, there is one major disadvantage to preoperative radiation and that is its

detrimental effect on wound healing. Some groups have used an intraoperative or postoperative boost

dose of radiation to the resection bed if there are concerns that margins are involved.

3187

Figure 108-9. Wide excision involves removal of the tumor with a margin of normal tissue. If necessary, major vascular structures

or nerves may be resected.

Figure 108-10. A: Radical amputations for extremity sarcomas include one joint above the most proximal extent of tumor. Now

largely of historic interest, tumors of the proximal thigh or buttock were previously routinely treated with hemipelvectomy. B:

While limb-sparing procedures are now done whenever possible, amputation may be necessary for certain cases. This 60-year-old

woman had high-grade angiosarcoma of the right leg with extensive satellite metastases extending above her knee. She required

above-knee amputation.

Table 108-4 Complications of Treatment for Extremity Sarcoma

3188

Increasing data support the use of resection alone for selected patients; those who have small (<5

cm), low-grade tumors seem to have acceptable local control and excellent long-term survival without

radiation if wide margins are obtained at the time of resection.68,69 Most patients undergoing surgical

excision may be candidates for additional radiation therapy because of high risk for recurrence, though

its role in lower- or intermediate-risk patients is somewhat undefined, and the so-called therapeutic

ratio of risk and benefits to radiation treatment must be carefully considered. Validated prognostic tools

may be helpful for individual patients weighing the benefits of available treatment options. Kattan et

al.70 developed a nomogram to help better predict individual risk for 12-year disease-specific mortality

in sarcoma patients using multiple prognostic factors simultaneously.

For extremity sarcomas, the importance of pathologically negative margins must be emphasized.

While differing widths of margins have been called “optimal,” it is generally agreed that 1 to 2 cm are

adequate, but that microscopically clear margins may be sufficient. Involved margins, along with age,

recurrent presentation, and fibrosarcoma or MPNST subtypes, significantly increases the risk of local

recurrence.3,71 Patients presenting with involved margins following the index operation should be

offered reresection in order to obtain surgical clearance of margins. Several studies have shown no

compromise in survival if complete resection is achieved with a second operation.72 Interestingly, it is

unclear if local recurrence predisposes to subsequent distant metastasis or if margin status is merely a

proxy for aggressive disease.71,73

Since the main cause of death in patients with soft tissue sarcoma is distant metastatic disease,

continued efforts have been made to develop effective systemic therapies.5,74 Postoperative adjuvant

chemotherapy has been controversial because risk of adverse toxic effects may not be outweighed by

the relatively low response rates and lack of durable results. The histologic subtypes of sarcoma vary in

their responsiveness to chemotherapy. For example, osteogenic sarcoma, rhabdomyosarcoma, and

Ewing sarcoma (the “pediatric” sarcomas) have had high rates of success with multimodality treatments

including adjuvant chemotherapy.59 Unfortunately, most randomized trials are underpowered to detect

modest differences in survival and promising response rates reported in smaller nonrandomized clinical

trials are not duplicated in subsequent larger randomized trials. In general, adjuvant chemotherapy is

not ever indicated for patients with low-grade sarcomas and patients with small sarcomas of higher

grades. Initial interest in adjuvant chemotherapy for extremity soft tissue sarcoma was piqued by a

randomized trial reported by Rosenberg et al.75 which demonstrated an improvement in disease-free

and overall survival with adjuvant doxorubicin, cyclophosphamide, and high-dose methotrexate. After a

longer follow-up period (median follow-up 7.1 years), however, both disease-free and overall survival

were not statistically significant.76

Over the years, single agents as well as combination chemotherapy regimens have been used in the

adjuvant treatment of soft tissue sarcomas. Two of the more active agents include doxorubicin and

ifosfamide. However, multiple randomized trials of postoperative chemotherapy have not demonstrated

an improvement in disease-free or overall survival with intermediate or long-term follow-up though

there was a trend toward chemotherapy in many studies.59,77–79 The Sarcoma Meta-analysis

Collaboration evaluated the effect of adjuvant doxorubicin-based chemotherapy in 1,568 patients from

14 trials.80 Although the time to local and distant recurrences, as well as recurrence-free survival, were

significantly better in the treatment group, an overall survival advantage was not seen (HR 0.89, p =

0.12). Because of the rarity of disease and heterogeneity of tumor characteristics, most studies are too

small to provide adequate power for testing responses in specific histologic subtypes. In general, many

3189

regimens have combined two of the most active agents, doxorubicin (or other anthracycline) and

ifosfamide, and results from a randomized trial did show improved disease-free survival and overall

survival; distant relapse rates were not different however.81 Outside of clinical trials, adriamycin and

ifosfamide combination therapy is commonly used for high-risk, primary extremity and truncal

sarcomas in the adjuvant setting. Newer regimens with promising results include the combination of

gemcitabine and docetaxel which was initially discovered to be active in heavily pretreated uterine

leiomyosarcomas and is currently being considered for use in that subtype as well as in high-grade

undifferentiated pleomorphic sarcomas and other high-risk subtypes.82–84 A phase II study comparing

adriamycin and ifosfamide to the combination of gemcitabine and docetaxel found that the latter

regimen had a slightly lower hospitalization rate and trends toward longer disease-free survival rates.85

Even with advances in systemic chemotherapy and targeted agents, data supporting greatly improved

overall survival rates are sparse, especially in the adjuvant setting. For advanced tumors, treatments

with single agents or combination regimens such as dacarbazine, doxorubicin, epirubicin, ifosfamide,

gemcitabine, docetaxel, temozolomide, trabectedin, and eribulin have been trialed extensively. Newer

targeted therapies have shown some promise in certain histologic subtypes. For example, tyrosine

kinase inhibitor (TKI) agents such as pazopanib, imatinib, sorafenib, and sunitinib have shown efficacy

in multiple subtypes after initial success with GISTs. The class of mTOR inhibitors (sirolimus,

temsirolimus, and everolimus) has demonstrated promise for perivascular epithelioid cell tumors

(PEComas), lymphangioleiomyomatosis, and angiomyolipomas.

One potential advantage of upfront systemic chemotherapy is the ability to assess tumor response in

situ. By seeing whether the tumor responds to the chemotherapy, both radiologically and

pathologically, it may be possible to spare patients prolonged therapy if they have not shown a

response, or to continue therapy postoperatively if there is response. However, data from a trial of

preoperative doxorubicin and ifosfamide-based chemotherapy conducted by the European Organisation

for Research and Treatment of Cancer (EORTC) and the National Cancer Institute of Canada did not

demonstrate any survival benefit compared to surgery alone.86 Some agents such as gemcitabine may

have chemosensitizing properties,87 making preoperative chemoradiation strategies attractive. Based on

current evidence, neoadjuvant chemoradiation, either concurrent or sequential, should only be offered

to selected high-risk patients on clinical protocol.88 It is important to follow patients on neoadjuvant

therapy closely since tumor progression is expected to be seen in approximately 30% of patients.89,90

Commonly used measures of radiologic response to treatment may not be good surrogate endpoints

for treatment response.38 The pathologic correlations of treatment-induced necrosis or histologic

response to therapy with clinical outcomes have also not been well defined in soft tissue sarcoma.91

Some patients who have a clinical or pathologic response to chemotherapy may have improved local

control and decreased distant disease-free survival, but no overall survival advantage.92 In other cases,

lack of measurable response did not seem to predict significant differences in event-free endpoints.89

These issues must be carefully considered in the design of future trials.

Alternative approaches to standard systemic therapy have been attempted with limited success. For

example, preoperative radiation therapy in combination with intra-arterial doxorubicin chemotherapy

initially showed good local control, but results could not be duplicated in subsequent randomized trials

and treatments were associated with high morbidity.93,94 Regional chemotherapy administered via

hyperthermic isolated limb perfusion has been attempted for advanced extremity sarcomas.

Extracorporeal circulation allows the delivery of drug concentrations 10 to 20 times higher than with

systemic delivery and a European study showed a limb-salvage rate of 71% for unresectable extremity

sarcomas using melphalan and tumor necrosis factor-α.95,96 However, this technique is uncommonly

used in the United States and usually reserved for highly selected patient populations.

RETROPERITONEAL AND INTRA-ABDOMINAL SARCOMAS

Management of intra-abdominal (visceral) and retroperitoneal sarcomas deserves special consideration

since specific anatomic location may dictate workup and subsequent management. The grouping of

visceral sarcomas generally includes tumors of the gastrointestinal tract and gynecologic organs, and

less commonly, genitourinary organs. Location often precludes early diagnosis, unless the mass is found

incidentally. Though some patients have noted increased abdominal girth, an abdominal mass can be

appreciated with focused physical examination, but it is uncommonly the presenting complaint.

Typically, vague symptoms of mass effect intra-abdominally trigger medical evaluation. Neurovascular

symptoms related to compression or invasion can occur and manifest as paresthesia, dysesthesia,

3190