Multidisciplinary approach for a high-risk, localized soft tissue sarcoma of the trunk after unplanned nononcological resection.

A man aged 45 years with hypertension and hyperlipidemia presented to his primary care physician with a 3-month history of a golf ball-sized right upper back mass. Ultrasound was performed, revealing a well defined lesion with mixed echogenicity, thought to be consistent with a lipoma. Clinically, however, the mass rapidly enlarged and became painful. The patient was referred to dermatology and subsequently to a surgeon. He underwent surgery at his local institution 2 months after initial presentation. The mass was excised in two fragments measuring 8 × 7 cm and 6 × 6 cm. Pathology revealed a high-grade pleomorphic liposarcoma (PLPS). Necrosis was observed, and the mitotic rate was 26 per 10 high-power fields. The tumor involved the margins of the piecemeal excision specimens. Computed tomography (CT) imaging of the chest, abdomen, and pelvis was notable for tiny, indeterminate pulmonary nodules and a 4.3 × 2.4 × 4.7 cm, well circumscribed area with central low density in the right posterolateral chest wall soft tissues, possibly reflecting a postoperative seroma (Figure 1A). The patient noted progressive enlargement of the operative site; therefore, repeat CT imaging of the chest was performed 3 weeks later and showed an enlarging soft tissue mass concerning for residual disease (Figure 1B). Magnetic resonance imaging (MRI) of the chest showed a 4.2 × 5.0 × 6.0 cm solid mass with enhancement and diffusion restriction contiguous with the latissimus dorsi muscle (Figure 1C,D). The patient was referred to our sarcoma specialty center and had a multidisciplinary evaluation with medical oncology, radiation oncology, and surgical oncology. Preoperative radiation therapy (RT) and surgery followed by adjuvant chemotherapy with doxorubicin and ifosfamide were recommended. Radiographs before oncological resection. (A) Axial contrast-enhanced CT of the chest after nononcological resection shows postoperative collection with eccentric focus of nodular enhancement in the deep subcutaneous tissues of the right lateral chest wall, suspicious for residual/recurrent tumor. (B) Axial unenhanced CT of the chest 23 days after the image in A showing an increase in the lobulated soft tissue attenuation lesion. Axial (C) T2 fat-suppressed and (D) T1 postcontrast MR images show a T2 intermediate enhancing mass consistent with recurrent tumor. CT indicates computed tomography; MR, magnetic resonance. Ultrasonography (US) is often the first imaging examination performed to evaluate palpable, superficial soft tissue masses because of its wide availability, low cost, and ease of performance. US is most useful in distinguishing solid from cystic lesions without additional intravenous contrast, and it can identify specific benign lesions. However, the US findings of benign and malignant solid soft tissue tumors overlap, and accurate tissue diagnosis is not feasible in most instances.1, 2 In this case, the US findings were reported as likely lipoma; therefore, the mass was resected because of rapid growth and pain. Findings concerning for malignancy are rapid growth, vascularity, and pain, which should prompt further workup with MRI or CT.2 MRI is the imaging modality of choice for soft tissue tumors or tumor-like masses because of its superior contrast, allowing for tissue characterization and multiplanar capability without radiation exposure. In cases of suspected myxofibrosarcoma, MRI can be helpful diagnostically because these tumors are often infiltrative and extend along the fascial planes, referred to on MRI as the tail sign.3, 4 CT is used primarily to guide the biopsy and can help in the identification of mineralization and osseous involvement, although it is not routinely used except in patients who have contraindications to MRI. Postoperative seromas occur in up to 10% of patients and appear as a well marginated, fluid density structure with or without thin rim enhancement on contrast-enhanced CT or MRI. Local recurrence after excision of soft tissue sarcoma (STS) is seen in 5%–35% of cases, usually occurring within the first 2 years. Risk factors associated with recurrence are >5 cm, intermediate-grade or high-grade tumors, and positive or close surgical margins. It is often difficult to distinguish recurrent tumor from postoperative seroma on CT, especially for tumors with myxoid elements, which can mimic low-density fluid on CT. Contrast-enhanced MRI is the modality of choice to distinguish between posttreatment changes and tumor recurrence.5, 6 Recurrence is often seen as a discrete, T2, hyperintense, enhancing nodule or mass, with imaging characteristics often similar to the primary tumor.7 In this case, the MRI features were consistent with recurrence. Histologically, the tumor showed extensive myxoid stroma with curvilinear blood vessels and scattered pleomorphic cells, features characteristic of myxofibrosarcoma (Figure 2A). However, in addition, there were focal areas with pleomorphic lipoblasts, diagnostic of pleomorphic liposarcoma (Figure 2B). Pleomorphic liposarcoma may show a range of histologic appearances. Most examples are morphologically similar to undifferentiated pleomorphic sarcoma, other than the presence of lipoblasts (which range from few to extensive); some cases closely mimic myxofibrosarcoma, whereas others show strikingly epithelioid morphology, mimicking carcinomas (especially adrenal cortical carcinoma).8, 9 The distinction between pleomorphic liposarcoma and myxofibrosarcoma has implications for both prognosis and surgical management.10 The risk of distant metastasis for pleomorphic liposarcoma is around 50%, compared with 25%–30% for high-grade myxofibrosarcoma. Pleomorphic liposarcoma is generally well circumscribed, similar to most other adult sarcoma types; therefore, local control for extremity and trunk wall tumors is usually straightforward when managed in sarcoma centers with surgical and radiation oncology sarcoma expertise. In contrast, myxofibrosarcoma shows markedly infiltrative margins along fascial planes, often extending far beyond grossly and clinically apparent margins; without wide excision, the risk of recurrence is very high: myxofibrosarcoma is one of the few sarcoma types in the modern era that often necessitates amputation after multiple local recurrences.11 Histologic features of pleomorphic liposarcoma. (A) This tumor showed large areas closely mimicking myxofibrosarcoma, including abundant myxoid stroma, curvilinear blood vessels, and pleomorphic cells with marked nuclear atypia. (B) In focal areas, pleomorphic lipoblasts (large cells with hyperchromatic nuclei indented by lipid vacuoles (arrowheads) could be identified, confirming the diagnosis of pleomorphic liposarcoma. Optimal treatment for localized, high-grade STS of the extremity and trunk is combined oncologic resection and RT. Three landmark randomized trials established this role for conservative surgery and RT. The first trial randomized patients with extremity STS to amputation (the standard of care at that time) versus conservative surgery and postoperative RT; all patients received adjuvant chemotherapy.12 There was no statistically significant difference in overall survival (OS) between arms, and the patients who received surgery and RT experienced a local recurrence rate of only 15%. The next two trials assessed the need for RT and compared conservative surgery alone with conservative surgery and RT. The first trial randomized patients with low-grade and high-grade STS of the extremity to conservative surgery alone versus conservative surgery plus postoperative RT (63 grays [Gy]); all patients with high-grade disease also received postoperative chemotherapy.13 Local recurrence rates were lower for patients who received RT. For patients with high-grade disease, the local recurrence rate in the RT arm was 0% compared with 20% for those who underwent surgery alone (p = .003). For patients with low-grade disease, the local recurrence rates were 4% and 33%, respectively (p = .016). There were no significant differences in OS between arms. The second trial randomized patients with STS of the extremity and trunk to conservative surgery alone versus conservative surgery and adjuvant RT (42–45 Gy) in the form of brachytherapy. The brachytherapy technique consisted of low-dose-rate iridum-192, which was after-loaded into catheters sewn into the tumor bed at the time of surgery.14 Among patients with high-grade disease, the local recurrence rate in the RT arm was 9% compared with 30% for those who underwent surgery alone (p = .0025). No difference in local recurrence was observed in patients with low-grade STS, and, for this reason, brachytherapy is not typically recommended as a radiation modality for low-grade STS. Among all patients, there was no survival difference between arms. Because the role for RT in addition to oncologic surgery had been established, the next question pertained to the optimal sequencing of RT and surgery. O’Sullivan et al. performed a seminal randomized trial in which they compared preoperative RT (50 Gy with or without a 16–20 Gy postoperative boost) with postoperative RT (66–70 Gy) for patients who had extremity STS.15 Local control rates were not different between groups and were 93% and 92% for the preoperative and postoperative groups, respectively (p = .72). A 7-year update of the trial also showed no statistically significant difference in survival between treatment arms.16 The side-effect profile was different, however. For patients who received preoperative RT, the rate of acute major wound complications after surgery was higher for patients who received preoperative RT compared with those who received postoperative RT (35% vs. 17%; p = 1). These complications are significant but reversible, and function was equivalent between treatment arms 1 year after surgery. For patients treated with the larger treatment field sizes typical for postoperative RT, grade ≥2 late toxicities were higher compared with patients treated with smaller field sizes. These toxicities include subcutaneous fibrosis, joint stiffness, and edema and are typically permanent.17 Series have also demonstrated higher rates of bone fractures for patients who received postoperative RT compared with preoperative RT.18 To summarize, the advantages of preoperative RT include irradiation of smaller volumes to lower doses compared with postoperative RT, albeit at the expense of a higher risk of wound complications, which are reversible. With postoperative RT (which uses higher doses to larger volumes), there is a higher rate of permanent long-term complications, such as edema, joint stiffness, fibrosis, and fracture. Both approaches are acceptable, but expert panels recommend preoperative RT for most situations.19 Because STS is so rare, as was the case for the current patient, initial management is often inadvertent marginal surgery performed either for a presumed benign lesion or for a presumed malignant lesion without appreciating that wide margins are required for STS. As described above, the initial surgery for the patient in this case was piecemeal, and postoperative imaging showed gross residual disease. This commonplace situation is referred to as an unplanned excision. Expert guidelines in this scenario recommend preoperative RT followed by definitive oncologic re-resection.19 Essentially, patients who undergo an unplanned excision are typically approached as though they have de novo disease. Accordingly, our recommendation for this patient with gross residual disease is standard preoperative RT (intensity-modulated RT to 50 Gy) followed by oncologic resection 4–6 weeks after the completion of RT. With respect to defining the RT treatment fields, careful review of both the initial preoperative MRI (when available) and the postoperative/unplanned excision MRI with an experienced radiologist is crucial. Both studies should be fused with the radiation planning CT scan to define the sum of the initial de novo and postexcision gross tumor volumes. A wire should be placed on the incision at the time of simulation to help define the underlying tumor bed. Once the gross tumor volume has been established, typical clinical target volume expansions for a subcutaneous STS are 3–4 cm radially in every direction and 0.5–1.0 cm deep into underlying muscle/chest wall. Planning target volume expansions are institution-dependent and usually 0.5 cm if daily pretreatment volumetric imaging guidance is used.19 There are also some planning nuances with respect to histology. Pathology review for this case showed high-grade pleomorphic liposarcoma mimicking myxofibrosarcoma. When planning local therapy, one's antenna should always go up if the histology shows myxofibrosarcoma because these tumors are particularly infiltrative and are associated with higher local recurrence rates if the extent of tumor is underappreciated.11, 20, 21 Knowing the diagnosis in this case was critical to formulating the appropriate management plan. As discussed above, defining the sarcoma subtype often has implications for the comprehensive management plan, including surgery. Before any definitive surgical intervention for a presumed sarcoma, a core-needle biopsy should be performed by interventional radiology. A core-needle biopsy is preferred over fine-needle aspiration because it provides better tumor architectural information critical to identifying the type of sarcoma. Furthermore, the risk of needle-track seeding is very low at 0.0%–0.7%.22-26 A core-needle biopsy is also preferred over an excisional biopsy for establishing a diagnosis. This allows for one definitive operation when the time comes for surgery, rather than one operation for excision followed by a second more definitive operation, if indicated. When evaluating a patient with a soft tissue mass, it is important to consider that the differential is extensive, and surgery may not be indicated at all. Examples include various subcutaneous manifestations of hematologic malignancies (such as lymphoma) that only require systemic therapy, benign neoplasms (such as desmoid fibromatosis, for which surgery is no longer considered first-line therapy), or even benign conditions (such as nodular fasciitis) that are self-limited and eventually resolve entirely without any intervention. If sarcoma is suspected, confirmation of the diagnosis and histologic subtype by an experienced sarcoma pathologist is critical to ensure appropriate management. Different histologic subtypes under the broader sarcoma umbrella commonly require different management. For instance, in this case, distinguishing between myxofibrosarcoma and PLPS affects the extent of the surgical margin. Myxofibrosarcomas typically have microscopic extensions well beyond the palpable or radiographic margins and thus warrant wider margins of resection (approximately 3 cm) when feasible for this histology compared with others. Local recurrence is the predominant pattern of failure for myxofibrosarcoma; in fact, the 5-year local recurrence rates are higher (12% all grades, 31% intermediate-grade/high-grade) than they are for other histologies arising in the trunk or extremities.27, 28 However, with a diagnosis of PLPS, the surgical goal was to achieve margins of 2 cm, which is less than the margins needed for myxofibrosarcomas. Piecemeal resections or incisional biopsies should be avoided because these can increase the risk of recurrence. Ideally, resections should be macroscopically complete with negative microscopic margins (R0 resections) to minimize local recurrence rates (6% at 5 years; 8% at 10 years).29 Preplanned positive margins against critical structures (macroscopically complete with positive microscopic margins, or R1 resection) do not necessarily have higher rates of local recurrence (10% at 5 years; 12% at 10 years).29 Importantly, unplanned excisions or whoops operations (margin-positive resections when sarcoma was not suspected) are associated with high rates of local recurrence even after re-excision (18% at 5 years; 24% at 10 years).29 After the initial multidisciplinary consultation, the patient was not able to obtain insurance authorization to return for treatment at our sarcoma specialty center. Despite appeals motivated by improved outcomes and OS for patients managed at sarcoma specialty centers,30 the patient received neoadjuvant treatment in the community without expertise in sarcoma. This led to delays in initiating RT, and the patient's tumor progressed rapidly in the interim. Initiation of neoadjuvant systemic therapy with combined doxorubicin, ifosfamide, and mesna was considered; however, the center had insufficient staff and availability to administer ifosfamide. Ultimately, the patient completed RT locally. Restaging imaging (MRI of the chest to evaluate the primary site and CT of the chest, abdomen, and pelvis) was obtained for surgical planning and to re-evaluate for metastatic disease. Eventually, the patient was able to undergo definitive surgery at our institution. Based on the histology, the goal of the operation was to obtain a margin-negative resection. The tumor was primarily located in the subcutaneous soft tissue, but its deep margin was inseparable from the underlying latissimus dorsi muscle. To get an adequate margin superficially, a paddle of skin extending beyond the palpable mass circumferentially was resected, continuing with wide soft tissue radial margins around the tumor in the subcutaneous soft tissue down to the fascia. The deep margin largely consisted of the latissimus dorsi muscle; however, anteriorly, the specimen was more adherent to the serratus anterior muscle than anticipated based on imaging. Therefore, for the deep margin, a portion of the serratus along the anterior aspect of the mass and a portion of the latissimus along the posterior aspect of the mass were removed. This provided a uniform muscle layer as a deep margin. The thoracodorsal neurovascular structures to the uninvolved portion of the latissimus were preserved. Final pathology demonstrated a 9.3-cm, high-grade PLPS. Margins were >2 cm circumferentially around all aspects of the tumor. Only 5% of the tumor was viable, indicating treatment response. The role of perioperative chemotherapy in localized STS arising in the extremity or trunk is controversial. Several studies over recent decades resulted in conflicting conclusions, leading to varied practice patterns across sarcoma specialty centers. For example, a large meta-analysis published in 1997 included 14 trials that randomized patients with localized, resectable STS to receive chemotherapy versus no chemotherapy and concluded that anthracycline-based chemotherapy is associated with an absolute reduction in the risk of local recurrence by 6% (95% CI, 1%–10%), a 10% reduction in the risk of distant metastasis at 10 years (95% CI, 5%–15%), and a trend toward improved OS (HR, 0·89; 95% CI, 0.76–1.03).31 These conclusions have been interpreted cautiously, however, given the heterogeneity of the STS subtypes, sizes, grades, and sites of disease included, the limited use of RT (47% of cases), and the varied chemotherapy regimens and doses used. A subsequent meta-analysis combining data from the 1997 analysis with four additional trials also demonstrated a modest risk reduction in local and distant recurrence (HR, 0.67; 95% CI, 0.56–0.82) and an improvement in OS (HR, 0.77; 95% CI, 0.64–0.93), but similar concerns about the heterogeneity of the included studies have limited widespread adoption of perioperative chemotherapy.32 To address the limitations of the prior meta-analyses, The European Organization for Research and Treatment of Cancer (EORTC) 62931 trial (ClinicalTrials.gov identifier NCT00002641) randomized 351 patients with intermediate-grade or high-grade, localized STS to receive adjuvant chemotherapy with doxorubicin 75 mg/m2 plus 5 g/m2 ifosfamide and mesna in a 21-day cycle for five cycles versus observation alone.33 The study demonstrated that adjuvant chemotherapy did not improve relapse-free survival (HR, 0.91; 95% CI, 0.67–0.22) or OS (HR, 0.94; 95% CI, 0.68–1.31). The varied results of these studies often leave the oncologist conflicted about what to recommend for an individual patient. To address this, several nomograms have emerged to predict outcomes in patients with STS and can be used as tools to inform clinical decisions.34-39 One of the most widely accepted nomograms with robust external validation is the Sarculator.40 The Sarculator was developed for extremity STS but was retrospectively applied to the EORTC 62931 trial, which included nonextremity tumors, and demonstrated that patients with a low predicted OS (pOS) (<60%) benefitted from adjuvant chemotherapy, with a significant improvement in disease-free survival (HR, 0.49; 95% CI, 0.28–0.85) and OS (HR, 0.50; 95% CI, 0.30–0.90).41 In addition, the Italian Sarcoma Group (ISG)-STS 1001 trial (ClincalTrials.gov identifier NCT01710176) randomized patients with localized, high-risk, extremity or trunk STS to receive neoadjuvant anthracycline and ifosfamide (AI) chemotherapy versus a histology-tailored regimen and demonstrated improved outcomes in the AI arm, suggesting that AI chemotherapy may also be better than no chemotherapy in this population.42 The Sarculator nomogram was also applied to the ISG-STS 1001 results and indicated that patients with low pOS had better outcomes than predicted by the Sarculator, further supporting a role for AI-based perioperative chemotherapy in these patients.43 Taken together, these emerging data suggest that recommendations for perioperative chemotherapy should be individualized to the patient, and discussions on risks and benefits may be informed by predictive nomograms. For the patient described in our case, the pOS was 57% based on the Sarculator's prediction for an extremity tumor, so adjuvant doxorubicin plus ifosfamide chemotherapy was discussed with the patient. After discussion of the potential side effects, long-term complications, and benefits, the patient agreed with adding systemic therapy after local treatment. The clinical trial landscape for STS is skewed toward treatment of advanced, inoperable, and metastatic disease; however, several notable combined modality studies are underway to optimize the management of localized disease. Several trials are exploring perioperative RT combined with checkpoint inhibitors, oncolytic viruses, tyrosine kinase inhibitors, MDM2 inhibitors, DNA damage repair inhibitors, and combination chemotherapy and targeted therapies. It is important to refer patients to a sarcoma center with expertise and active clinical trials so patients can have access to these novel approaches that may improve upon the standard of care. The patient has now completed local therapy for his trunk sarcoma and has no evidence of disease. He is anticipating initiating adjuvant chemotherapy followed by surveillance according to National Comprehensive Cancer Network guidelines. For this high-risk tumor, we would typically recommend CT of the chest and MRI of the primary site every 3 months for the first 2 years after completing therapy, then every 6 months through year 5, then annually through year 10. STSs are rare malignancies with over 70 subtypes that have distinct behaviors and biological underpinnings. Increasingly, treatment is becoming more nuanced and tailored to the specific subtype of sarcoma. Our case highlights that it is essential to accurately diagnose the subtype of STS and have a multidisciplinary consultation at a sarcoma center to provide a comprehensive plan for optimal treatment; this is endorsed globally by international guidelines and the National Comprehensive Cancer Network STS guidelines. For these reasons, we endorse early referral to a sarcoma center for multidisciplinary planning and local therapy. When feasible and indicated, systemic therapy can be delivered at a sarcoma center or at a local oncology practice after consultation with a sarcoma medical oncologist. Telehealth options and shared care models now allow for partnerships between community oncologists and sarcoma experts to manage patients on surveillance and seamlessly address concerns during long-term follow-up. Elizabeth H. Baldini reports personal fees from UpToDate outside the submitted work. Jason L. Hornick reports personal fees from Aadi Bioscience, Adaptimmune LLC, and TRACON Pharmaceuticals outside the submitted work. Candace L. Haddox, Jyothi P. Jagannathan, and Chandrajit P. Raut disclosed no conflicts of interest.