Abstract
Background:
The optimal margin of resection for high-grade extremity sarcomas and its impact on survival has long been questioned in the setting of adjuvant radiotherapy. The objective of this study was to investigate the impact of resection status on recurrence and survival.
Methods:
All patients with primary, nonmetastatic, high-grade extremity sarcomas that underwent surgical resection from January 2000 to April 2016 in the U.S. Sarcoma Collaborative (USSC) were retrospectively reviewed. Recurrence patterns, recurrence-free survival (RFS), and overall survival (OS) were examined in multivariate analyses (MVA).
Results:
A cohort of 959 patients was identified with a median follow-up of 34.7 months from diagnosis. R0 resection was achieved in 86.7% (831) while R1 resection in 13.3% (128). Locoregional recurrence for R0 and R1 groups occurred in 9.1% (76) versus 14.8% (19; p = .05) while distant recurrence occurred in 24.7% (205) versus 26.6% (34; p = .65), respectively. Median RFS was 171.2 versus 48.5 (p = .01) while median OS was 149.8 versus 71.5 months (p = .02) for the R0 versus R1 group, respectively. On MVA, female gender (hazard ratio [HR] = 0.69, p = .007) and adjuvant radiotherapy (0.7, p = .04) were associated with improved OS, whereas older age (HR = 1.03, p < .001) and tumor size (HR = 1.01, p < .001) were associated with worse OS. R0 resection status was associated with improved locoregional RFS (HR = 0.56, p = .03) but not with distant RFS (HR = 0.84, p = .4) or OS (HR = 0.7, p = .052).
Conclusions:
In high-grade extremity sarcomas, tumor size and gender are predictive of OS while R0 resection status is associated with improved locoregional recurrence rate without a significant impact on distant RFS or OS.
Keywords: high-grade extremity sarcomas, negative margins, predictors of recurrence, R resection status
1 |. INTRODUCTION
Soft tissue sarcomas are rare mesenchymal tumors arising within the extremities in 43% of cases.1 A paradigm shift in the treatment of extremity sarcomas occurred in the 1980s moving from amputation to limb salvage as the preferred surgical management.2–4 A landmark randomized control trial at the National Cancer Institute demonstrated that limb salvage with adjuvant radiotherapy had similar disease-free and overall survival (OS) to amputation.5 The role of adjuvant radiotherapy was further studied in two randomized trials demonstrating improved local control rates, specifically for high-grade histologies, but in neither of these studies did an improvement in local control rates translate into an OS difference.6,7
Even though the significance of an adequate oncologic resection cannot be overemphasized, a balance between negative margin resection and acceptable functional outcomes remains challenging. In this setting, radiation treatment is a key component of the multidisciplinary care of those patients. Preoperative radiotherapy is favored in the National Comprehensive Cancer Network (NCCN) guidelines and has been associated with decreased local recurrence rates, albeit at the cost of increased rates of wound complications.8 However, the question of optimal margins has not been definitively answered, and in a recent paper by Bonvalot et al.,9 neither margin status nor local recurrence were found to correlate with survival. This is in contrast to a study by Zagars et al.10 that correlated negative margin status to improved disease-specific survival.
Given the paucity of studies looking at high-grade histologies alone, in this cohort, we limited the inclusion criteria to high-grade soft tissue extremity sarcomas, in an effort to reduce potential biases. The aim of the current study was to identify the impact of resection status margins (R0 vs. R1) on recurrences and OS in patients with primary, nonmetastatic, high-grade extremity sarcomas using a large, multi-institutional database.
2 |. MATERIALS AND METHODS
All patients with primary high-grade extremity soft tissue sarcomas that underwent surgical resection from January 2000 to April 2016 in the United States Sarcoma Collaborative (USSC) database were retrospectively reviewed. This collaborative is compiled of seven high-volume academic centers (Wake Forest University, Winston-Salem, North Carolina; Winship Cancer Institute, Emory University, Atlanta, Georgia; Stanford University, Palo Alto, California; The Ohio State University, Columbus, Ohio; University of Chicago Medicine, Chicago, Illinois; Washington University, St. Louis, Missouri; Medical College of Wisconsin, Milwaukee, Wisconsin). This study was approved by the respective Institutional Review Boards at the study sites. Patients with desmoid tumors, multifocal or metastatic disease at presentation, palliative intent resection, or incomplete survival data were excluded. OS was calculated from the date of surgery to the date of the last follow-up or date of death. Recurrence-free survival (RFS) was calculated based on imaging/clinical findings suggestive of recurrence, from the date of surgery. Patients with no evidence of disease at the time of the last follow-up were censored. The surveillance modality included computed tomography or magnetic resonance imaging based on institutional preferences at 3- or 6-month intervals for the first few years and annually thereafter.
Histopathologic type, TNM grade according to the 7th edition American Joint Committee on Cancer Staging system and French Federation of Cancer Centers Sarcoma Group grade were documented.11,12 A combined grade variable was utilized and classified all tumors coded as G2 or G3 as “high-grade.” Only high-grade tumors based on the combined grade variable were included in the analysis. The completeness of resection was classified as microscopically complete (R0), microscopically incomplete (R1), and macroscopically incomplete (R2).13 The type of resection was classified as “excisional” for isolated tumor removal, “radical” for en-bloc removal of surrounding muscle or at-risk bone, and “compartmental” when resection of the entire anatomic compartment containing the tumor was performed. Re-resections were performed either after an unplanned resection at a facility outside of the collaborative or to achieve a clear margin after R1 resection.
2.1 |. Statistical analysis
Baseline characteristics were compared based on resection status. R2 resections were excluded from this analysis. Student’s t test was used to analyze continuous variables and χ2 statistic was used to analyze nominal variables.
Proportional hazards univariate regression models were utilized based on all clinically significant variables, according to prior studies. Multivariate models were then created including all variables significant at p < .20. Age and gender were included in the multivariate analysis irrespective of the univariate p value. We examined whether the proportional hazards assumption was met in our final multivariate model and this was confirmed through an examination of time by measure interactions across all components of that model (p = .19 for the overall test). Primary outcomes among subgroups were compared using Kaplan–Meier survival methods using log-rank χ2 tests. Statistical significance was assumed for all variables with p < .05. SAS (version 9.4) was used for statistical processing.
2.2 |. Histologic grouping
Due to the rarity of certain histologic subtypes, a simple grouping algorithm was used to create a “Sarcoma—not otherwise specified (NOS)” group. Histologic subtypes with 10 cases or less were clustered together and analyzed as part of the Sarcoma NOS category. The breakdown of the different histologies can be seen in Table S1.
2.3 |. Outcomes
The primary outcomes were RFS and OS. Secondary outcomes included locoregional RFS and distant RFS.
3 |. RESULTS
3.1 |. Clinicopathological patient characteristics
The study cohort included 959 patients with a high-grade extremity sarcoma. Median follow-up was 34.7 months from initial diagnosis (interquartile range [IQR]: 17.7, 64.9 months). There were 831 patients that underwent R0 resection (86.7%) and 128 patients that underwent R1 resection (13.3%). The median age for the R0 group was 60.4 years versus 65.3 years for the R1 group (p = .06), with 54.3% versus 56.4% being male in the R0 versus R1 group (p = .28). As shown in Table 1, re-resection was undertaken in 15.4% (128/831) to achieve an R0 margin while 19.5% (25/128) had an R1 margin after re-resection (p = .22). Amputations were performed in 8.5% patients in the R0 group and 3.9% in the R1 group (p = .07). There was also an increased rate of bone resection with R0 compared to R1 9.9% versus 3.1% (p = .01). The most common pathologic subtypes in the R0 versus R1 group were: undifferentiated pleomorphic sarcoma 45.5% versus 47.7%, myxofibrosarcoma 10.1% versus 13.3%, synovial sarcoma 11.4% versus 6.3%, leiomyosarcoma 8.2% versus 5.5%, and sarcoma NOS 8.2% versus 7%, (p = .17). A larger proportion of patients in the R0 group received neoadjuvant chemotherapy 22.8% versus 11% in the R1 group (p = .003). Adjuvant chemotherapy was administered to 21.4% patients in the R0 group versus 17.3% in the R1 group (p = .3), while more patients in the R1 group received adjuvant radiotherapy 37.5% versus 23.6%, (p = .001; Table 2).
TABLE 1.
Descriptive, operative, and pathologic characteristics of patients stratified by resection status
| R0 n = 831 | R1 n = 128 | p | |
|---|---|---|---|
| Age, median (y, IQR) | 60.4 (48.3, 72.7) | 65.3 (50.7, 74.5) | .06 |
| Tumor size, median (cm, IQR) | 8.4 (5, 13.6) | 7.2 (4.3, 13) | .16 |
| Gender, male | 451 (54.3%) | 76 (56.4%) | .28 |
| Re-resection | 128 (15.4%) | 25 (19.5%) | .22 |
| Tumor location | |||
| Upper extremity | 191 (23%) | 33 (25.8%) | .48 |
| Lower extremity | 640 (77%) | 95 (74.2%) | |
| Tumor depth | |||
| Superficial | 82 (9.9%) | 12 (9.4%) | .86 |
| Deep | 748 (90.1%) | 116 (90.6%) | |
| Histologic type | |||
| Liposarcoma | 92 (11.1%) | 17 (13.3%) | .17 |
| Leiomyosarcoma | 68 (8.2%) | 7 (5.5%) | |
| Myxofibrosarcoma | 84 (10.1%) | 17 (13.3%) | |
| Osteosarcoma, extraskeletal | 35 (4.2%) | 4 (3.1%) | |
| Rhabdomyosarcoma | 11 (1.3%) | 5 (3.9%) | |
| Sarcoma, NOS | 68 (8.2%) | 9 (7%) | |
| Synovial sarcoma | 95 (11.4%) | 8 (6.3%) | |
| Undifferentiated pleomorphic sarcoma | 378 (45.5%) | 61 (47.7%) | |
| Extent of resection | |||
| Excisional | 96 (11.7%) | 15 (11.7%) | .27 |
| Compartmental | 23 (2.8%) | 7 (5.5%) | |
| Radical | 705 (85.6%) | 82.8%) | |
| Amputation | 71 (8.5%) | 5 (3.9%) | .07 |
| Arterial resection | 73 (8.8%) | 6 (4.7%) | .12 |
| Venous resection | 65 (7.9%) | 5 (4%) | .12 |
| Nerve resection | 91 (11%) | 10 (7.9%) | .29 |
| Bone resection | 82 (9.9%) | 4 (3.1%) | .01 |
| Closure type | |||
| Primary | 445 (53.6%) | 80 (62.5%) | .37 |
| Primary + mesh | 6 (0.7%) | 0 | |
| Skin graft | 54 (6.5%) | 9 (7%) | |
| Tissue flap | 251 (30.2%) | 30 (23.4%) | |
| Skin graft + tissue flap | 74 (8.9%) | 9 (7%) | |
| Closure required plastic surgeon | 112 (13.5%) | 12 (9.4%) | .2 |
Abbreviations: IQR, interquartile range; NOS, not otherwise specified.
TABLE 2.
Multimodality treatments, morbidity, recurrence patterns for the R0 and R1 groups
| R0 n = 831 | R1 n = 128 | p | |
|---|---|---|---|
| Neoadjuvant chemotherapy | 188 (22.8%) | 14 (11%) | .003 |
| Adjuvant chemotherapy | 175 (21.4%) | 22 (17.3%) | .3 |
| Neoadjuvant radiotherapy | 302 (36.3%) | 28 (21.9%) | .001 |
| Adjuvant radiotherapy | |||
| Boost | 6 (0. 7%) | 3 (2.3%) | .001 |
| Full dose | 193 (23.6%) | 48 (37.5%) | |
| Intraoperative radiation | 18 (2.4%) | 0 | .15 |
| Brachytherapy | 8 (1.1%) | 0 | .6 |
| Postoperative complication | 219 (26.4%) | 38 (29.7%) | .43 |
| Postoperative drainage procedure | 39 (4.8%) | 10 (7.9%) | .14 |
| Bleeding—clinically significant (requiring intervention: embolization or reoperation) | 13 (1.6%) | 3 (2.3%) | .47 |
| Reoperation | 104 (12.5%) | 22 (17.2%) | .15 |
| For bleeding | 7 (6.7%) | 2 (9.1%) | .85 |
| For infection | 73 (70.9%) | 16 (72.7%) | |
| For other reason | 23 (22.3%) | 4 (18.2%) | |
| Readmission | 103 (12.5%) | 22 (17.2%) | .14 |
| Recurrence pattern | |||
| Locoregional recurrence | 76 (9.1%) | 19 (14.8%) | .05 |
| Distant recurrence | 205 (24.7%) | 34 (26.6%) | .65 |
3.2 |. Survival analysis
Median RFS for R0 resections was 171.2 months (IQR: 106.2—not reached) compared to 48.5 months (IQR: 20.2–81.5) for R1 resections (p = .01; Figure 1). For R0 resections, 3- and 5-year RFS were 62.5% and 58.8%, while for R1 resections, were 52% and 45.1%, respectively.
FIGURE 1.
Recurrence-free survival Kaplan–Meier curves for the R0 versus R1 groups (log-rank, p = .01)
In the R0 group, 9.1% developed a locoregional recurrence compared to 14.8% in the R1 group (p = .05). Three- and five-year locoregional RFS was 88.9% and 86% for the R0 group versus 78.4% and 71.4% for the R1 group. In the R0 group, 24.7% developed a distant recurrence with a median time to recurrence of 171.2 months. In the R1 group, 26.6% developed a distant recurrence and the median time to recurrence was 81.5 months (p = .23). Kaplan–Meier curves depicting locoregional and distant RFS are demonstrated in Figures 2 and 3. Median OS for R0 resections was 149.8 months (IQR: 114.3—not reached) compared to 71.5 months (IQR: 51.5—not reached) for R1 resections (p = .02; Figure 4). Three-year and five-year OS were 78.3% and 70.3% for R0 resections, 72.1% and 54% for R1 resections, respectively.
FIGURE 2.
Locoregional recurrence-free survival Kaplan–Meier curves for the R0 versus R1 groups (log-rank, p = .005)
FIGURE 3.
Distant recurrence-free survival Kaplan–Meier curves for the R0 versus R1 groups (log-rank, p = .23)
FIGURE 4.
Overall survival Kaplan–Meier curves for the R0 versus R1 groups (Log-rank, p = .02)
R0 resection status was associated with improved locoregional recurrence (HR = 0.56, p = .03) but was not associated with distant recurrence (HR = 0.84, p = .4) or recurrence-free survival (HR = 0.76, p = .09). Age (HR = 1.02, p = .003), tumor size (HR = 1.01, p = .002), re-resection (HR = 1.77, p = .04) were also associated with locoregional recurrence on multivariate analysis (Table 3). Similarly, tumor size (HR = 1.007, p = .01), female gender (HR = 0.68, p = .007), subfascial (deep) tumor location (HR = 2.84, p = .007), re-resection (HR = 0.48, p = .02), and neoadjuvant radiotherapy (HR = 1.36, p = .03) were predictive of distant recurrence-free survival (Table 4). The univariate and multivariate analysis for recurrence-free survival can be seen in Table 5. Tumor size (HR = 1.008, p < .001), female gender (HR = 0.75, p = .02), subfascial tumor location (HR = 1.68, p = .03), and neoadjuvant radiotherapy (HR = 1.34, p = .02) were associated with recurrence-free survival. On multivariate analysis for OS, age (HR = 1.03, p < .001) and tumor size (HR = 1.01, p < .001) were associated with a worse OS while female gender (HR = 0.69, p = .007) and adjuvant radiotherapy (HR = 0.71, p = .04) were associated with an improvement in OS (Table 6).
TABLE 3.
Cox proportional hazards univariate and multivariate regression analysis for locoregional recurrence-free survival
| Locoregional recurrence-free survival | Hazard ratio univariate | 95% CI | p | Hazard ratio multivariate | 95% CI | p |
|---|---|---|---|---|---|---|
| Age | 1.01 | 0.99–1.03 | .09 | 1.02 | 1.007–1.04 | .003 |
| Tumor size (cm) | 1.01 | 1.003–1.02 | .004 | 1.01 | 1.004–1.02 | .002 |
| Gender, female | 0.96 | 0.62–1.48 | .84 | 1.007 | 0.66–1.55 | .97 |
| Lower extremity | 1.24 | 0.72–2.14 | .44 | |||
| Deep tumor | 0.95 | 0.5–1.79 | .87 | |||
| Reresection | 1.68 | 0.95–2.98 | .08 | 1.77 | 1.03–3.05 | .04 |
| Histologic type | ||||||
| Liposarcoma | Ref | |||||
| Undifferentiated pleomorphic sarcoma | 1.49 | 0.69–3.21 | .3 | |||
| Leiomyosarcoma | 1.63 | 0.58–4.57 | .35 | |||
| Myxofibrosarcoma | 1.71 | 0.69–4.19 | .24 | |||
| Extraskeletal osteosarcoma | 1.14 | 0.24–5.52 | .87 | |||
| Sarcoma NOS | 1.43 | 0.51–3.99 | .5 | |||
| Synovial sarcoma | 0.59 | 0.15–2.29 | .45 | |||
| Neoadjuvant chemotherapy | 0.73 | 0.38–1.42 | .35 | |||
| Adjuvant chemotherapy | 0.95 | 0.5–1.81 | .88 | |||
| Neoadjuvant radiotherapy | 0.91 | 0.53–1.58 | .75 | |||
| Adjuvant radiotherapy | 0.82 | 0.48–1.39 | .47 | |||
| R0 resection status | 0.57 | 0.33–0.98 | .04 | 0.56 | 0.33–0.95 | .03 |
Abbreviations: CI, confidence interval; NOS, not otherwise specified; Ref, reference.
TABLE 4.
Cox proportional hazards univariate and multivariate regression analysis for distant recurrence-free survival
| Distant recurrence-free survival | Hazard ratio univariate | 95% CI | p | Hazard ratio multivariate | 95% CI | p |
|---|---|---|---|---|---|---|
| Age | 0.99 | 0.99–1.01 | .47 | 0.99 | 0.98–1.005 | .44 |
| Tumor size (cm) | 1.006 | 1.001–1.01 | .02 | 1.007 | 1.002–1.01 | .01 |
| Gender, female | 0.67 | 0.51–0.9 | .007 | 0.68 | 0.51–0.9 | .007 |
| Lower extremity | 1.15 | 0.81–1.62 | .44 | |||
| Deep tumor | 2.87 | 1.33–6.18 | .007 | 2.84 | 1.33–6.08 | .007 |
| Reresection | 0.48 | 0.26–0.89 | .02 | 0.48 | 0.26–0.89 | .02 |
| Histologic type | ||||||
| Liposarcoma | Ref | |||||
| Undifferentiated pleomorphic sarcoma | 1.32 | 0.83–2.11 | .24 | |||
| Leiomyosarcoma | 1.67 | 0.89–3.14 | .11 | |||
| Myxofibrosarcoma | 0.88 | 0.45–1.73 | .72 | |||
| Extraskeletal osteosarcoma | 2.08 | 1.03–4.18 | .04 | |||
| Sarcoma NOS | 1.53 | 0.85–2.85 | .16 | |||
| Synovial sarcoma | 1.4 | 0.78–2.49 | .26 | |||
| Neoadjuvant chemotherapy | 0.77 | 0.54–1.09 | .14 | |||
| Adjuvant chemotherapy | 1.1 | 0.78–1.55 | .58 | |||
| Neoadjuvant radiotherapy | 1.42 | 1.02–1.97 | .04 | 1.36 | 1.03–1.8 | .03 |
| Adjuvant radiotherapy | 0.93 | 0.65–1.33 | .69 | |||
| R0 resection status | 0.81 | 0.54–1.22 | .31 | 0.84 | 0.56–1.26 | .4 |
Abbreviations: CI, confidence interval; NOS, not otherwise specified; Ref, reference.
TABLE 5.
Cox proportional hazards univariate and multivariate regression analysis for recurrence-free survival
| Recurrence-free survival | Hazard ratio univariate | 95% CI | p | Hazard ratio multivariate | 95% CI | p |
|---|---|---|---|---|---|---|
| Age | 1.001 | 0.99–1.01 | .76 | 1.002 | 0.99–1.01 | 0.62 |
| Tumor size (cm) | 1.007 | 1.003–1.01 | <.001 | 1.008 | 1.004–1.01 | <.001 |
| Gender, female | 0.75 | 0.59–0.95 | .02 | 0.75 | 0.59–0.95 | .02 |
| Lower extremity | 1.16 | 0.87–1.55 | .32 | |||
| Deep tumor | 1.69 | 1.05–2.73 | .03 | 1.68 | 1.05–2.69 | .03 |
| Reresection | 0.83 | 0.55–1.25 | .38 | |||
| Histologic type | ||||||
| Liposarcoma | Ref | |||||
| Undifferentiated pleomorphic sarcoma | 1.37 | 0.92–2.05 | .12 | |||
| Leiomyosarcoma | 1.65 | 0.97–2.83 | .07 | |||
| Myxofibrosarcoma | 1.13 | 0.67–1.92 | .63 | |||
| Extraskeletal osteosarcoma | 1.88 | 0.99–3.54 | .05 | |||
| Sarcoma NOS | 1.5 | 0.9–2.5 | .12 | |||
| Synovial sarcoma | 1.23 | 0.73–2.07 | .44 | |||
| Neoadjuvant chemotherapy | 0.78 | 0.57–1.06 | .11 | 0.8 | 0.59–1.08 | .15 |
| Adjuvant chemotherapy | 1.05 | 0.78–1.42 | .74 | |||
| Neoadjuvant radiotherapy | 1.26 | 0.95–1.66 | .11 | 1.34 | 1.04–1.72 | .02 |
| Adjuvant radiotherapy | 0.9 | 0.67–1.22 | .5 | |||
| R0 resection status | 0.71 | 0.52–0.99 | .04 | 0.76 | 0.55–1.04 | .09 |
Abbreviations: CI, confidence interval; NOS, not otherwise specified; Ref, reference.
TABLE 6.
Cox proportional hazards univariate and multivariate regression analysis for overall survival
| Overall survival | Hazard ratio univariate | 95% CI | p | Hazard ratio multivariate | 95% CI | p |
|---|---|---|---|---|---|---|
| Age | 1.03 | 1.02–1.04 | <.001 | 1.03 | 1.02–1.04 | <.001 |
| Tumor size (cm) | 1.01 | 1.003–1.01 | <.001 | 1.01 | 1.004–1.01 | <.001 |
| Gender, female | 0.68 | 0.51–0.89 | .006 | 0.69 | 0.52–0.9 | .007 |
| Lower extremity | 1.23 | 0.87–1.74 | .24 | |||
| Deep tumor | 1.59 | 0.93–2.74 | .09 | 1.6 | 0.94–2.72 | .09 |
| Reresection | 0.71 | 0.43–1.16 | .17 | 0.69 | 0.42–1.12 | .13 |
| Histologic type | ||||||
| Liposarcoma | Ref | |||||
| Undifferentiated pleomorphic sarcoma | 0.98 | 0.64–1.5 | .93 | |||
| Leiomyosarcoma | 0.69 | 0.36–1.36 | .29 | |||
| Myxofibrosarcoma | 0.88 | 0.5–1.56 | .66 | |||
| Extraskeletal osteosarcoma | 1.24 | 0.6–2.58 | .57 | |||
| Sarcoma NOS | 1.35 | 0.79–2.3 | .28 | |||
| Synovial sarcoma | 1.03 | 0.55–1.92 | .93 | |||
| Neoadjuvant chemotherapy | 0.93 | 0.65–1.33 | .69 | |||
| Adjuvant chemotherapy | 1.07 | 0.75–1.53 | .71 | |||
| Neoadjuvant radiotherapy | 0.82 | 0.59–1.13 | .22 | 0.81 | 0.59–1.1 | .18 |
| Adjuvant radiotherapy | 0.7 | 0.5–0.98 | .04 | 0.71 | 0.51–0.98 | .04 |
| R0 resection status | 0.69 | 0.48–1.005 | .053 | 0.7 | 0.48–1.004 | .052 |
Abbreviations: CI, confidence interval; NOS, not otherwise specified; Ref, reference.
4 |. DISCUSSION
In this multicenter study of patients with high-grade extremity soft tissue sarcomas, R0 resection was associated with an improvement in locoregional disease control however it was not associated with an improvement in distant metastasis-free survival or overall survival. Independent of resection margin, older age, increased tumor size, and male gender were independently associated with worse OS while adjuvant radiotherapy correlated with improved OS.
Multiple studies have investigated the impact of adequate surgical resection to survival outcomes with variable findings. To the best of our knowledge, this is the largest study looking solely at high-grade extremity sarcomas. R1 margins have been shown to be predictive of local recurrence, and disease-specific survival in single-institution studies.10,14,15 However, a recent study by Bonvalot et al.9 did not show a difference in OS after R1 resections and surgical margins were not a predictor of survival on multivariate analysis. Similar to our findings, the local recurrence rate was lower after R0 resection but there was no difference in distant recurrence-free survival.9 It is important to note that none of the aforementioned studies have shown any correlation between margin status and distant recurrence, which represents the most common site of failure for high-grade extremity sarcomas. The R0 resection rate of 86.7% in our study compares favorably to published reports that range between 66% and 87.7%. Similarly, the local recurrence rate of 10% is on the lower end of prior reports that vary from 9% to 20%. Despite the relative adequacy of the surgical resections and the low rate of locoregional recurrences, distant recurrences were observed in 24.9%, unchanged from historic reports ranging from 22% to 30%. More importantly though, survival remains relatively unchanged compared to large series composed predominantly, although not exclusively, of high-grade tumors.9,10,14,14,15
Radiation has an integral role in the multidisciplinary management of extremity sarcomas with more than half of the patients studied receiving radiation either in the neoadjuvant or adjuvant setting (Table 2). Adjuvant radiotherapy was associated with improvement in OS; however, neoadjuvant radiation was associated with worse recurrence-free survival due to its correlation with worse distant recurrence-free survival. Those results are potentially explained by the bias which led to a referral for preoperative therapy in higher risk lesions, as well as the tumor biology of this heterogeneous group of patients that are at high risk for distant failure as the main mode of recurrence. Even though the retrospective nature of this study does not allow for definitive conclusions to be made regarding the decision to defer radiation in 40% of the patients analyzed, there are several factors that likely contributed. It is possible that a portion of patients were not candidates for radiation, particularly in the adjuvant setting, given the complication rate of 26% and the relatively advanced age of the group analyzed. Another factor that has to be considered and has been investigated in a recent analysis of the National Cancer Database is patient access to specialized care.16 Out of a sample of 21,763 patients with soft tissue sarcoma of the extremity, 47%–49% did not receive radiation despite advanced clinical stage and moderately/poorly differentiated tumors. Interestingly, “patients who traveled long distance were less likely to receive radiation if they had negative margins.” This raises concerns about potential inequalities in access and reflects upon the difficult decisions that patients and providers are faced with day to day.
Re-resections were performed in 16% of patients, and positive margin resections were more common in the re-resection setting (16.3%) than after primary resection (12.7%). Though this was not statistically significant, re-resection was independently associated with worse locoregional control on multivariate analysis. It is notable though, that re-resection correlated with improved distant-disease control. The above findings cannot be explained by clearance of residual disease that would have otherwise led to distant recurrences, as that should have also led to improved rates of locoregional control. The significance of residual disease has been investigated in prior studies with isolated local recurrences lacking any impact on metastatic recurrence.17 The exact mechanism by which local recurrences or residual local disease affect distant recurrences remains unclear and therefore the role of re-resection needs to be further elucidated.
There are several limitations to our study stemming from its retrospective nature. The inclusion of eight academic centers without central pathologic adjudication includes inherent variability in the assessment of surgical margins, especially due to the sheer volume of sarcoma specimens with a median size of 8 cm. The issue of accurate margin reporting has been studied extensively and a recent report emphasized the importance of adequate documentation especially as it pertains to high risk margins.18 In an era of more frequent use of neoadjuvant treatment in sarcoma, traditional pathologic grading becomes even more problematic, potentially limiting the accuracy of staging. Recent advances in molecular diagnostics, immunohistochemistry techniques, and even the use of magnetic resonance imaging can provide more information for the characterization of tumor grade.19,20 Another critical component is the difficulty to account for the impact of histologic subtype on survival outcomes given the variability among different pathologic types. Despite the fact that histologic subtype was not a predictor of DFS or OS, it is likely that differences in biology are responsible for the differences observed in terms of distant recurrence rate and ultimately survival. Furthermore, institutional differences regarding the pattern and use of imaging for surveillance might account for some differences in the identification of local or distant recurrences.
This large multi-institutional cohort of high-grade extremity sarcomas demonstrated that despite the significance of negative margin resection in preventing locoregional recurrences, it is not predictive of distant failure or ultimately survival. Even though the significance of an adequate, well-planned surgical resection cannot be overstated, the predictors of locoregional, distant recurrences, and survival appear to differ, indicating different aspects of the variable biology of these tumors. Systemic therapy did not correlate with improved survival with a quarter of patients suffering distant recurrence. Development of new diagnostic tools, which will be able to identify and treat patient-specific tumor biology at a personalized level, is direly needed as currently offered modalities seem to have reached their limits, with sarcoma outcomes being almost unchanged over the last two decades.21
Supplementary Material
ACKNOWLEDGMENT
This study was supported by the Wake Forest University Biostatistics shared resource NCI CCSG P30CA012197.
Footnotes
CONFLICT OF INTERESTS
Nathan Patel has stock/interest in Genomenon Inc. The other authors declare that there are no conflict of interests.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the supporting information tab for this article.
DATA AVAILABILITY STATEMENT
Data are available upon request from the author.
REFERENCES
- 1.von Mehren M, Randall RL, Benjamin RS, et al. Soft tissue sarcoma, version 2.2018, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2018;16(5):536–563. [DOI] [PubMed] [Google Scholar]
- 2.Shiu MH, Castro EB, Hajdu SI, Fortner JG. Surgical treatment of 297 soft tissue sarcomas of the lower extremity. Ann Surg. 1975;182(5):597–602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Abbas JS, Holyoke ED, Moore R, Karakousis CP. The surgical treatment and outcome of soft-tissue sarcoma. Arch Surg. 1981;116(6):765–769. [DOI] [PubMed] [Google Scholar]
- 4.Lindberg RD, Martin RG, Romsdahl MM, Barkley HT. Conservative surgery and postoperative radiotherapy in 300 adults with soft-tissue sarcomas. Cancer. 1981;47(10):2391–2397. [DOI] [PubMed] [Google Scholar]
- 5.Rosenberg SA, Tepper J, Glatstein E, et al. The treatment of soft-tissue sarcomas of the extremities: prospective randomized evaluations of (1) limb-sparing surgery plus radiation therapy compared with amputation and (2) the role of adjuvant chemotherapy. Ann Surg. 1982;196(3):305–315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Yang JC, Chang AE, Baker AR, et al. Randomized prospective study of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas of the extremity. J Clin Oncol. 1998;16:197–203. [DOI] [PubMed] [Google Scholar]
- 7.Pisters PW, Harrison LB, Leung DH, Woodruff JM, Casper ES, Brennan MF. Long-term results of a prospective randomized trial of adjuvant brachytherapy in soft tissue sarcoma. J Clin Oncol. 1996;14(3):859–868. [DOI] [PubMed] [Google Scholar]
- 8.Albertsmeier M, Rauch A, Roeder F, et al. External beam radiation therapy for resectable soft tissue sarcoma: a systematic review and meta-analysis. Ann Surg Oncol. 2018;25(3):754–767. [DOI] [PubMed] [Google Scholar]
- 9.Bonvalot S, Levy A, Terrier P, et al. Primary extremity soft tissue sarcomas: does local control impact survival? Ann Surg Oncol. 2017;24(1):194–201. [DOI] [PubMed] [Google Scholar]
- 10.Zagars GK, Ballo MT, Pisters PWT, et al. Prognostic factors for patients with localized soft-tissue sarcoma treated with conservation surgery and radiation therapy: an analysis of 1225 patients. Cancer. 2003;97(10):2530–43. [DOI] [PubMed] [Google Scholar]
- 11.Edge SB, Byrd DR, Compton CC, Fritz AG, Greene FL, Trotti A, eds. AJCC Cancer Staging Manual. 7th ed. Springer; 2010. [Google Scholar]
- 12.Coindre JM. Grading of soft tissue sarcomas: review and update. Arch Pathol Lab Med. 2006;130(10):1448–1453. [DOI] [PubMed] [Google Scholar]
- 13.AJCCManual for Staging of Cancer. 3rd ed. JB Lippincott; 1988. [Google Scholar]
- 14.Pisters PW, Leung DH, Woodruff J, Shi W, Brennan MF. Analysis of prognostic factors in 1,041 patients with localized soft tissue sarcomas of the extremities. J Clin Oncol. 1996;14:1679–1689. [DOI] [PubMed] [Google Scholar]
- 15.Gronchi A, Lo Vullo S, Colombo C, et al. Extremity soft tissue sarcoma in a series of patients treated at a single institution: local control directly impacts survival. Ann Surg. 2010;251(3):506–511. 10.1097/SLA.0b013e3181cf87fa [DOI] [PubMed] [Google Scholar]
- 16.Moten AS, von Mehren M, Reddy S, Howell K, Handorf E, Farma JM. Treatment patterns and distance to treatment facility for soft tissue sarcoma of the extremity. J Surg Res. 2020;256:492–501. 10.1016/j.jss.2020.07.019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zagars GK, Ballo MT, Pisters PW, Pollock RE, Patel SR, Benjamin RS. Surgical margins and reresection in the management of patients with soft tissue sarcoma using conservative surgery and radiation therapy. Cancer. 2003;97(10):2544–2553. [DOI] [PubMed] [Google Scholar]
- 18.Gundle KR, Kafchinski L, Gupta S, et al. Analysis of margin classification systems for assessing the risk of local recurrence after soft tissue sarcoma resection. J Clin Oncol. 2018;36(7):704–709. 10.1200/JCO.2017.74.6941 [DOI] [PubMed] [Google Scholar]
- 19.Gibbs J, Henderson-Jackson E, Bui MM. Bone and soft tissue pathology: diagnostic and prognostic implications. Surg Clin North Am. 2016;96(5):915–962. 10.1016/j.suc.2016.06.003 [DOI] [PubMed] [Google Scholar]
- 20.Crombé A, Marcellin PJ, Buy X, et al. Soft-tissue sarcomas: assessment of MRI features correlating with histologic grade and patient outcome. Radiology. 2019;291(3):710–721. 10.1148/radiol.2019181659 [DOI] [PubMed] [Google Scholar]
- 21.Helman LJ, Meltzer P. Mechanisms of sarcoma development. Nat Rev Cancer. 2003;3(9):685–694. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available upon request from the author.