Soft tissue sarcoma (STS) is a rare tumor accounts for less than 1% of all newly diagnosed malignant diseases and extremity is the most common site of origin [1]. STS extensively invades surrounding soft tissue, so local recurrence rates are as high as 30% to 50% with local excision only [2]. Radical excision and amputation were used to improve local control and multimodality approach with conservative surgery plus postoperative radiotherapy is being used since 1970s for the purpose of preserving limb function and local control [2].

STS has a tendency to metastasize early and approximately one-half of the patients died of their disease [3]. The relationship between local control and survival has been under controversy, but Lewis et al. [4] suggested that there is a strong association of local recurrence with development of subsequent metastases and tumor mortality. Prognostic factors for local control and disease free survival are known to be histologic grade, tumor size and depth. Histologic grade is the most important prognostic factor for overall survival and predicting metastases [3].

In this article, we retrospectively analyzed the treatment outcomes and prognostic factors of localized extremity STS patients treated with conservative surgery plus radiotherapy.

The aim of the treatment in STS is to achieve local control and cure while ensuring organ preservation and limb function. Conservative surgery and postoperative radiotherapy are the preferred treatment in order to avoid amputation and this conservative approach offers high local control similar to those reported with amputation [3,10-13]. Whether improved local control translates to improved survival remains controversial [11,13]. In Lewis's analysis local recurrence has strong association with tumor-related mortality [4]. However, this may be because locally recurrent tumors are biologically more aggressive rather than a causal relationship [14].

Several prognostic factors for local recurrence have been reported, and these were included in our analyses. High grade, distal extremities and positive microscopic margin have been analyzed as prognostic factors influencing local recurrence [3]. Specific histologic types have been associated with high risk of local recurrence, especially in MFH [15]. In our study, we were not able to find any correlation between these factors and local recurrence and there was no correlation between treatment characteristics (radiation dose, field margin, and use of chemotherapy) and local recurrence. However, the patients who underwent RT were more likely to have had deep location, high grade tumors excised with a close or positive margin. In our analysis, histology of 3 recurrent patients was MFH, chondrosarcoma and synovial sarcoma. The lack of associated factor of local recurrence may be due to the small number of patients involved in this study, further studies are needed.

In our series the 5-year actuarial local control rate was 90.7% and it was comparable to other reported studies which achieve 5-year local control rates of 85% to 90% [14]. Three patients had local recurrence and two had recurred within the field, one in the margin of the field. Our result was comparable to other studies previously reported that the majority of the local recurrence was found within the field [14,16]. Cleator et al. [14] demonstrated 68% of local recurrence was within the field, 16% were out of the field, and 16% were on the margin of the field. In Dickie et al. [16]'s analysis of 768 patients 49 cases recurred within the field (6.4% overall), 9 were out of the field (1.1% overall), and 2 were on the margin of the field (0.3% overall).

Initial field margin of at least 5 cm around the tumor bed/scar are known to achieve superior local control in the postoperative radiotherapy [13]. But field margin needed to include potential microscopic disease is poorly defined [3]. A margin of only 2 cm around the tumor was added for brachytherapy and actuarial local control rate for high grade tumor are about a 90% [11] and this was comparable local control with wide field margin added for external beam radiotherapy. Afterloading catheters were placed intraoperatively in the tumor bed for brachytherapy and this made more accurate targeting thereby limited volume was irradiated and did not introduce long term functional morbidity [11]. Our initial filed margin was 0 to 5 cm (median, 2 cm) to the tumor bed including scar and this was applied individually according to surgical extent. However with this limited margin a 91% of local control was achieved. Possible explanation may be that the surgical technique may have a considerable impact on the degree of dissemination of tumor cells [3] or may be that only limited margin is needed to control disease when accurate targeting was done. And reduction of normal tissue irradiation with limited planning target volume margin in an evidence-based and quantitative manner should not increase the risk of local recurrence [16]. To reduce normal tissue irradiation, traditional 5 cm longitudinal margin from the end of tumor bed may be shortened. Further studies are needed to define optimal margin for radiotherapy.

In general, the postoperative radiotherapy (PORT) dose used in most centers is 50 Gy (range, 40 to 50 Gy) and boost of at least 10 Gy is given to high risk areas [3]. Although adequate dose in PORT has not been established, improved local control rate are associated with higher dose above 60 Gy [3,10,17]. In the literature, dose was escalated up to 70 Gy for patient with residual tumor or positive resection margin. In our analysis, almost local recurrence occur within the field, so it seems to be dose escalation to tumor bed is required for tumor control. Dose escalation might improve local control when target volume can be exactly defined and covered. But, severe complications are accompanied with higher dose above 63 Gy [13]. Thus selective dose escalation of radiotherapy might be considered for high risk patients and reappraisal of adequate dose to low risk patients might be considered. Patient with negative margins can be treated less than 60 Gy without compromising local control and fewer complications [13]. Tailored treatment might be needed according to risk group and whether dose escalation can be translated into improved outcome will be assessed after a longer follow up period.

In our series, the 5-year DRFS and OS are 73.4% and 69.2%, respectively and it was comparable to other reported studies [2,15,18]. The well-established prognostic factors for tumor related death from STS are histological grade, tumor size and age [3]. Our multivariate analysis also demonstrated higher OS and DSS rate for low grade tumor. And histologic subtype was not a significant determinant of outcome in our series. Univariate analysis revealed that histologic grade, AJCC stage and resection margin status were independently correlates with development of distant metastasis. But these were not significant determinants on multivariate analysis. STS have a tendency to metastasis early and the poor survival of such patients has led to a search for systemic treatments that might be applied early in the course of the disease [2]. Development of effective systemic treatment might be considered.

The 14% incidence of acute complications reported in our series is similar to that reported by other authors. As for acute side effects, skin desquamation was occurred in 4 patients, 1 patient suffered from pruritus and 1 patient had skin ulceration. Our data suggests patient with chemotherapy may be associated with an increased risk of complications than patient without chemotherapy. The RT treatment volume defined for STS multimodality management has been predictive of late RT effects such as fibrosis, edema, and joint stiffness and a reduction of normal tissue irradiation may translate into a decrease in such radiation-related toxicities [16]. Modern treatment techniques, such as intensity-modulated radiotherapy (IMRT), have facilitated highly conformal targeted approaches with a reduction in the overall treatment volume exposed to higher doses to reduce complication. However, this reduction may potentially expose the patient to an increased risk of local recurrence (LR).

In conclusion, conservative surgery with postoperative radiotherapy achieves a satisfactory rate of local control with acceptable complication rate in extremity STS. However, this disease has a high distant metastatic propensity that correlates with tumor grade and stage. Tumor grade was the most important factor for overall survival and there was no significant factor associated with local recurrence. The majority of the LRs developed in the irradiation field regardless of surgical margin status or total accumulative RT dose. And this may indicating that the incidence of LR is not predominately caused by aberrations in RT dose or target volume coverage but by biologic behavior of tumor. Development of effective new systemic treatment for this disease is an urgent priority for tumor with high grade or high stage to reduce distant metastasis. And the newly developing radiation technologies such as IMRT and image-guide radiotherapy have a beneficial effect in dose escalation and sparing normal tissue.