Treatment of resectable intrathoracic sarcomas: a single institution experience over twenty years

Background There are very few published data on the management of primary intrathoracic sarcoma, defined as sarcomas arising from the lung, pleura, and mediastinum, excluding the chest wall. Therefore, the aim of this study was to evaluate the outcome of patients with intrathoracic sarcoma treated at an academic referral center over a 21-year period. Methods A retrospective search was performed to identify patients with intrathoracic sarcomas treated with surgical resection from January 1990 to November 2011 at the University of Washington Medical Center. Local control and overall survival were analyzed in relation to the treatment received. Results Thirty-five patients were identified. Hazard ratios for local control, adjusted for tumor margin status, at 5 years were 0.74 (95 % CI [0.21, 2.58]) for the addition of chemotherapy (CT) to surgery, 0.57 (95 % CI [0.15, 2.23]) with the addition of (radiation therapy) RT, and 0.50 (95 % CI [0.06, 4.03]) with the addition of both CT and RT. At 7 years, the ratios for local control were 0.69 (95 % CI [0.20, 2.36]) for CT added to surgery, 0.58 (95 % CI [0.15, 2.27]) for RT, and 0.41 (95 % CI [0.05, 3.33]) with the addition of both CT and RT. Hazard ratios for overall survival, adjusted for sarcoma stage, at 5 years were 0.61 (95 % CI [0.16, 2.39]) for the addition of CT to surgery, 1.03 (95 % CI [0.26, 4.08]) for the addition of RT, and 0.54 (95 % CI [0.11, 2.69]) for the addition of both CT and RT. The 7-year hazard ratios for overall survival were 0.77 (95 % CI [0.23, 2.60]) for CT added to surgery, 0.99 (95 % CI [0.25, 3.84]) for the addition of RT, and 0.42 (95 % CI [0.09, 2.05]) for both CT and RT with surgery. At 10 years, hazard ratios for overall survival were 0.71 (95 % CI [0.21, 2.38]) for added CT, 0.81 (95 % CI [0.21, 3.08]) for added RT, and 0.33 (95 % CI [0.07, 1.65]) for the addition of both CT and RT to surgery. Conclusions Our series is the largest published study of intrathoracic sarcoma which focuses on the survival benefit of adding RT, chemotherapy or both to surgery in resectable intrathoracic sarcoma. Our data suggest a potential benefit in local control and survival from adjuvant therapy, with the greatest benefit likely to come from combined CT and RT, though none of the results achieved statistical significance. As intrathoracic sarcomas are rare and histologically heterogeneous, larger collaborative studies are necessary to determine treatment efficacy and elucidate which histologic subtypes are likely to benefit most from adjuvant therapy.


Introduction
Sarcomas are malignancies that arise from cells of mesenchymal origin, and consequently can be found in bone, muscle, fat, cartilage and other connective tissue (http://cancer.gov/ cancertopics/cancerlibrary/what-is-cancer). They are rare, accounting for less than 1 % of cancers in adults [1], and for approximately 15-20 % of tumors in children [2]. In addition these tumors are very heterogeneous with over 50 distinct histologic subtypes with a wide anatomical distribution [2]. Consequently due to this heterogeneity, studies of systemic therapy are very challenging. Historically they have been divided into soft tissue sarcomas, representing the majority of cases, and malignant bone sarcomas, which make up a little over 10 % [2].
Soft tissue sarcomas occur most commonly in the extremities, followed by the abdomen/pelvis and head and neck region [3]. In the USA, approximately 11,000 new cases are reported each year, and close to 4000 attributed deaths occur [4,5]. They arise in all age groups, but are more common with age (51.7 % >60 years, 27.6 % 40-60 years, and 20.7 % 20-40 years), with a slight male predominance (1.1 male to 1.0 female) [6]. The etiology is usually unknown, though exposures to radiation, chemicals, and longstanding lymphedema have been suggested as risk factors [6]. Prognosis is most strongly correlated with histological grade [7,8], and to a lesser extent with tumor size, positive surgical margins and histological subtype [9,10].
Primary intrathoracic sarcomas are rare soft tissue sarcomas of the lung, pleura, and mediastinum. The most common  [11]. The rarity of these tumors is reflected in the paucity of literature specifically examining the efficacy of various treatment modalities to guide clinical therapeutic approaches. Frequently they are studied in combination with other soft tissue sarcomas, and often are not assessed as a distinct subset or not included at all [12][13][14][15][16]. They are commonly treated in a similar manner to soft tissue sarcomas of the extremities, for which more established guidelines exist [17].
Surgery is the primary treatment for localized intrathoracic sarcomas. The role of radiation and chemotherapy, however, is less established. Radiation therapy (RT) has been effective as an adjuvant to surgery for soft tissue sarcomas of the extremity [16,18] and has been shown to reduce local recurrences with either pre-or post-surgery treatments [19]. The utility of chemotherapy (CT) is less clear, with no trial demonstrating a clear survival benefit in resectable soft tissue sarcoma [20][21][22][23]. Novel chemotherapeutic agents and combinations are currently being investigated in the metastatic setting, and may prove more efficacious [24]. Given the paucity of published data on the management of primary intrathoracic sarcoma, the purpose of this study was to assess the treatment and outcome of patients with intrathoracic sarcoma treated at the University of Washington, with particular attention to the effect of radiation and chemotherapy on local control and overall survival.

Patient selection
IRB approval was obtained to review medical charts for patients diagnosed with thoracic tumors at the University of Washington Medical Center (UWMC) from January 1990 to November 2011. Patients identified with intrathoracic sarcomas via pathology and imaging, who received surgical resection with or without adjuvant CT and RT were included in the study. Patients who received initial incomplete resections elsewhere or who presented to our institution with locally recurrent disease were included, provided there was adequate information about their initial or primary tumor treatment, respectively. Over the time period analyzed, patients were treated by an experienced multi-disciplinary team of physicians. Informed consent for treatment was obtained from all patients included in this study. Neoadjuvant/adjuvant chemotherapy was administered based on individual patient presentation with particular attention to histological subtype, tumor grade, and tumor size.

Data collection
Resectable intrathoracic sarcoma was defined as a nonmetastatic (stage I, II or III) primary sarcoma arising in the lung, pleura, or mediastinum. Each tumor was staged using the AJCC 7th edition staging criteria [25]. Patient demographic data (sex, age at diagnosis); tumor specific data (date of diagnosis, primary site, histology, size, grade, stage); margins status (positive <1 cm, negative >1 cm); treatment modalities received (neo-adjuvant radiation and/or chemotherapy or adjuvant radiation and/or chemotherapy); and outcomes (date of last follow-up, date of death) were obtained by retrospective chart analysis. The size of the tumor was defined as the maximum dimension (in centimeters) determined by CT at the initial diagnosis. All tumor pathology was examined centrally at the UWMC. Grade was classified according to the FNCLCC system as either low, intermediate or high-grade based on the number of mitoses, the amount of necrosis, and cellular differentiation [26].

Statistical analysis
Patients were followed from date of diagnosis to death of the patient, or end of the study period. Survival curves were plotted using the Kaplan-Meier method [27]. Hazard ratios for 5-, 7-, and 10-year local control and overall survival were obtained using Cox regression models [28], adjusting for tumor status (positive or negative) as a potential confounder in local control, and sarcoma stage (low or high) in overall survival. For the purposes of our analysis, Blow stage^was defined as any stage I or II sarcoma, while Bhigh stage^was defined as stage III. All analyses were conducted using R v.3.1.1 [29]. Local control was defined as the period from the date of first treatment to the date of first recurrence of disease at the original treatment site. Overall survival was defined as the interval between the first treatment and most recent follow-up or death.

Patient population characteristics
Thirty-five adults with non-metastatic intrathoracic sarcomas treated at the University of Washington between January 1990 172 J Radiat Oncol (2016) 5:169-177

Local control and overall survival
Survival results comparing four treatment groups To compare the effects of different treatment patterns on overall and disease-free survival, the patient population was divided into four groups based on treatments received. Group 1 received surgery only, group 2 received surgery and CT, group 3 received surgery and RT, and group 4 received surgery and both CT and RT. The survival results are displayed as Kaplan-Meier curves in Fig. 1. Cox hazard ratio models comparing these treatment groups for 5-, 7-, and 10-year local control are detailed in Table 2. Hazard ratios for 5-, 7-, and 10-year overall survival are detailed in Table 3. Treatment with surgery alone is the reference group.

Survival results comparing radiation vs. no radiation
To examine the effect of RT on local control and overall survival, the patient population was divided into two treatment groups based on whether RT was administered (RT vs no RT).
The survival results are displayed as Kaplan-Meier curves in Fig. 2. Cox hazard ratio models comparing these treatment groups for 5-, 7-, and 10-year local control and overall survival are detailed in Tables 4 and 5

Survival results comparing chemotherapy vs. no chemotherapy
To examine the effects of CT on local control and overall survival, the patient population was likewise divided into two groups (CT received vs. no CT). The survival results are displayed as Kaplan-Meier curves in Fig. 3. Cox hazard ratio models comparing these treatment groups for 5-, 7-, and 10year local control and 5-, 7-, and 10-year overall survival are detailed in Tables 6 and 7

Discussion
Intrathoracic sarcomas are a rare subgroup of soft tissue sarcoma with poor prognosis and no established standard guidelines. Our study is the largest series of intrathoracic sarcoma treated at a single institution focused on treatment and outcomes of this rare set of neoplasms. This study highlights the variability in histology and individualization of patient treatment for this disease. Though surgery is the mainstay of therapy for localized sarcomas, recent studies indicate that the addition of (neo) adjuvant radiation and/or systemic therapy can potentially improve outcomes [3,17,30]. Similar studies examining thoracic wall (rather than primary pleural or pulmonary) sarcomas found that chemotherapy, radiation, and both in combination, each had a positive impact on local control and survival as adjunctive treatments [23,31]. Our results, which are the first to address prognostic variables and treatment influence on local control and overall survival in intrathoracic sarcoma, are in agreement with these earlier findings for sarcomas presenting in other locations and underscore the prognostic power of high-grade disease in survival. Notable limitations include lack of statistical power due to limited sample size and variability in treatment over the period of the study. However, our study is the first to compare these treatment modalities for intrathoracic sarcomas and attempt quantification of their effect on local control and survival. Despite the limitations, this study will serve as a benchmark for further larger studies evaluating novel approaches in the management of intrathoracic sarcoma.
Due to the numerous histological subtypes and manifestations of intrathoracic sarcomas, a multi-disciplinary approach to their management is essential, as coordination between surgery, radiation, medical oncology, pathology and radiology will facilitate clearer classification and management strategies for these rare malignancies [17]. In order to improve the outcome of patients with rare diseases, it is important that treatment is undertaken at specialist centers [32]. Furthermore, international collaboration is required to assess the utility of novel therapies in such malignancies as well as identifying the underlying molecular drivers of these heterogeneous diseases.

Compliance with ethical standards
Conflict of interest The authors declare that they have no competing interests.
Statement of ethical standards This article does not contain any studies with human or animal subjects performed by any of the authors.
Funding sources No funding support is associated with this study.
Financial disclosures None.