Abstract
Introduction:
Soft tissue sarcomas (STS) of the head and neck (H&N) are rare malignancies that are challenging to manage. We sought to describe the outcomes of patients treated with curative intent using surgery and radiation therapy (RT) for H&N STS.
Methods:
We performed a retrospective review of patients with non-metastatic STS of the H&N. The Kaplan-Meier method was used to estimate disease-specific survival (DSS) and local control (LC). Multivariable analyses (MVA) were conducted using Cox proportional hazards model.
Results:
192 pts had a median follow-up of 82 months. Tumors arose in the neck (n=50, 26%), paranasal sinuses (n=36, 19%), or face (n=23, 12%). Most patients were treated with post-operative RT (n=134, 70%). Post-op RT doses were higher (median 60Gy, pre-op 50Gy, p<0.001). Treatment sequence was not associated with LC (pre-op RT 78% (63–88), post-op RT 75% (66–82), p=0.48). On MVA, positive/uncertain margin was the only variable associated with LC (HR 2.54 (1.34–4.82), p=0.004). LC was significant on MVA (HR 4.48 (2.62–7.67), p<0.001) for DSS. Patients who received post-op RT were less likely to experience a major wound complication (MWC) (7.5% vs 22.4%, HR 0.28 (0.11–0.68), p=0.005). There was no difference in the rate of late toxicities between patients who received pre-op or post-op RT.
Conclusions:
H&N STS continues to have relatively poorer LC than STS of the trunk or extremities. We found LC to be associated with DSS. Timing of RT did not impact oncologic or long-term toxicity outcomes, however pre-op RT did increase the chance of developing a MWC.
Keywords: sarcoma, head and neck, radiation, surgery
Precis:
Local control and disease outcomes are poorer for patients with H&N sarcoma treated definitively with surgery and radiation than for patients with extremity sarcomas. Local control was found to be significantly associated with disease-specific survival.
INTRODUCTION:
Soft tissue sarcomas (STS) are rare, with an annual incidence of approximately 5 in 100,000 people (1). STS of the head and neck (H&N) are even less common, and comprise only 5–10% of all sarcomas (2), or 1% of all H&N tumors (3). They are heterogeneous in nature and have differing local and distant recurrence risks depending on histology (4, 5). The mainstay of curative treatment is surgical resection, with or without the addition of chemotherapy and/or radiation therapy (RT) (6).
Historically, outcomes for H&N STS are poorer compared to STS of other sites such as the trunk and extremity (7, 8). This is likely due to the difficulty in achieving sufficient margins with surgery due to complex local anatomy with multiple adjacent essential structures (5, 8). In addition, treatment-related morbidity can be significant in these locations. Depending on the tumor location, this can include disfigurement, fibrosis, ocular side effects, dysphagia, and osteoradionecrosis, among others. Extrapolating from the extremity STS literature, we know that pre-operative RT increases the risk for acute post-operative wound complications. However, it offers a decreased risk for long-term and irreversible RT-related complications compared to post-operative RT (9), because of lower doses and smaller fields used in the pre-operative setting. Due to their rarity, there is heterogeneity in the management approaches to H&N STS with limited data to guide treatment decisions. We sought to retrospectively examine outcomes and toxicities for patients with H&N STS treated at a high-volume sarcoma center with curative-intent RT and surgical resection.
MATERIALS/METHODS:
After obtaining IRB approval we performed a retrospective review of all patients treated at our tertiary cancer referral center for non-metastatic STS of the head and neck treated definitively with combined modality local therapy consisting of both surgery and RT from 1968–2020. We excluded patients with cutaneous angiosarcoma, sarcoma of the bone, or rhabdomyosarcoma as multidisciplinary management for these histologies differs from that of classical adult STS. Dermatofibrosarcoma protuberans and desmoid tumors were also excluded. We abstracted from the medical record their demographics, clinical characteristics, treatment course, and outcomes. At diagnosis, all patients underwent a complete history and physical examination, standard blood tests, and appropriate imaging, typically consisting of a CT Head and Neck with contrast to delineate the extent of primary disease, and a CT Chest or chest radiograph (more common in the older treatment era) to rule out the presence of lung metastases. In a subset of patients, an MRI scan with and without contrast was also obtained at diagnosis. A histopathologic diagnosis of sarcoma was confirmed in each case through review by an expert sarcoma pathologist at our institution.
Statistical analysis and follow-up
Follow-up time for survival analyses was computed from the last local therapy date: last RT date for those treated with post-operative RT or the surgical date for those patients treated with pre-operative RT. The Kaplan-Meier method was used to estimate local control (LC), distant metastasis-free survival (DMFS), disease-free survival (DFS), disease-specific survival (DSS), overall survival (OS) and development of treatment-related complications (10). The log-rank statistic was used to test for differences between curves. Covariates examined included: age, gender, tumor grade, tumor size, tumor location (neck versus other) surgical margins, pre-operative or post-operative RT, treatment era (pre-2005 and 2005 or after, chosen as this is when IMRT became more standard at our institution), RT modality, and receipt of chemotherapy. RT-related toxicities were retrospectively categorized into mild (requiring no treatment, noted by provider or patient during follow-up), moderate (requiring medical management), or severe (requiring surgical management or hospitalization). Major wound complications (MWC) after surgery were defined as occurring within 120 days of surgery as per the NCIC SR2 criteria (9). Differences between demographic and treatment variables were analyzed using chi-square test or Fisher’s exact test as appropriate (11). Multivariable analyses were conducted using Cox proportional hazards model. The analyses were conducted using IBM SPSS Statistics (Version 26) and Python (version 3.7, Python Software Foundation).
RESULTS:
We identified 192 patients eligible for analyses. Median follow-up for those alive at last follow-up was 82 months [IQR 69–95], and approximately half of the patients were treated prior to 2005 (n=101, 53%) (Table 1). The most common location was the neck or supraclavicular (SCV) area (n=61, 31.8%), followed by the paranasal sinuses (n=36, 18.8%), and face (n=23, 12.0%). The most common histology was sarcoma, not otherwise specified (n=44, 22.9%), followed by undifferentiated pleomorphic sarcoma (n=32, 16.7%), and neurogenic sarcoma (n=15, 7.8%) (Table 1). The majority of patients received post-operative RT (n=134, 79.8%), were male (n=111, 57.8%), and White (n=158, 82.3%) (Table 1). Median age at diagnosis was 49.5 years [range 18–84], which was significantly higher in the pre-operative RT cohort (56.5 years [18–83]) compared to the post-operative RT cohort (44 [18–84], p=0.009). Median tumor size was similar for pre-operative RT (4.8cm [0.8–26]) and post-operative RT patients (4.3cm [1–13], p=0.205). Median pre-operative RT dose was 50Gy [46–58], and 60Gy [44–70] for post-operative RT.
Table 1:
Demographics
| Entire Cohort N (%) Median [range] |
Preop N (%) Median [range] |
Postop N (%) Median [range] |
p-value | |
|---|---|---|---|---|
| n= | 192 | 58 (30%) | 134 (70%) | |
| Age | 49.5 [18–84] | 56.5 [18–83] | 44.0 [18–84] | 0.009* |
| Sex | 0.866 | |||
| Female | 81 (42%) | 25 (31%) | 56 (69%) | |
| Male | 111 (58%) | 33 (30%) | 78 (70%) | |
| Race | 0.497 | |||
| Black | 7 (4%) | 3 (43%) | 4 (57%) | |
| Hispanic | 27 (14%) | 6 (22%) | 21 (78%) | |
| White | 158 (82%) | 49 (31%) | 109 (69%) | |
| Max Size | 4.3 [0.8–26] | 4.75 [0.8–26] | 4.3 [1–13] | 0.205* |
| Histology | 0.385 | |||
| Sarcoma, NOS | 44 (23%) | 15 (34%) | 29 (66%) | |
| Undifferentiated Pleomorphic Sarcoma | 32 (17%) | 8 (25%) | 24 (75%) | |
| Neurogenic sarcoma | 15 (8%) | 2 (13%) | 13 (87%) | |
| Other | 101 (53%) | 33 (33%) | 68 (67%) | |
| Grade | 0.404 | |||
| 1 | 28 (23%) | 8 (29%) | 20 (71%) | |
| 2 | 20 (16%) | 5 (25%) | 15 (75%) | |
| 3 | 75 (61%) | 29 (39%) | 46 (61%) | |
| Margins | 0.017 | |||
| Negative | 104 (54%) | 39 (38%) | 65 (62%) | |
| Positive/uncertain | 88 (46%) | 19 (22%) | 69 (78%) | |
| Location | <0.001 | |||
| Other | 131 (68%) | 26 (20%) | 105 (80%) | |
| Neck/SCV | 61 (32%) | 32 (52%) | 29 (48%) | |
| Neoadj chemo | 48 (25%) | 20 (42%) | 28 (58%) | 0.046 |
| Adj Chemo | 33 (17%) | 8 (24%) | 25 (76%) | 0.412 |
| Simultaneous Chemo | 2 (1%) | 1 (50%) | 1 (50%) | 0.54 |
| Any Chemo | 79 (41%) | 28 (35%) | 51 (65%) | 0.187 |
| XRT Dose | 60 [44–70] | 50 [46–58] | 60 [44–70] | <0.001* |
| XRT fx | 30 [22–50] | 25 [23–31] | 30 [22–50] | <0.001* |
| XRT modality | 0.741 | |||
| 3D | 32 (25%) | 11 (34%) | 21 (66%) | |
| IMRT | 57 (44%) | 21 (37%) | 36 (63%) | |
| Electrons | 36 (28%) | 14 (39%) | 22 (61%) | |
| Protons | 5 (4%) | 3 (60%) | 2 (40%) | |
| Advanced Treatment Modality | 0.076 | |||
| 3D/electrons | 130 (68%) | 34 (26%) | 96 (74%) | |
| IMRT/protons | 62 (32%) | 24 (39%) | 38 (61%) | |
| Treatment Era | <0.001 | |||
| Pre-2005 | 101 (53%) | 46 (46%) | 55 (54%) | |
| On or after 2005 | 91 (47%) | 12 (13%) | 79 (87%) |
Mann-Whitney U Test
Size available for 177
Grade available for 123
XRT modality available for 130 pts
Table 1 shows patient, tumor, and treatment characteristics. A higher proportion of patients with tumor located in the head (versus neck/SCV) were treated with post-operative RT (n=105, 80%) than pre-operative RT (n=26, 20%). Also, final surgical margins were more commonly negative in patients who received pre-operative RT (75% vs. 48.5% of the post-op RT cohort; p=0.017).
Chemotherapy was administered in a little under half of patients (n=83, 43%), and more commonly delivered neoadjuvantly (n=48, 58%) compared to adjuvantly (n=33, 40%) (Table 1). There was no difference in the receipt of chemotherapy overall between patients receiving pre- or post-op RT (pre-op n=28, 53.8%, post-op n=51, 38.1%, p=0.187). A higher proportion of patients treated with chemotherapy had grade 2–3 tumors (93% vs 65%, p<0.001), larger tumors (61% ≥5cm vs 28%, p<0.001), and they also were younger at diagnosis (chemotherapy median age 43 yrs [IQR 18–68], no chemo 53 yrs [25–81], p=0.001).
Patients treated after 2005 were more likely to receive chemotherapy (50% vs 32%, p=0.013) and to be treated with an advanced radiotherapy treatment modality (IMRT or protons) (92% vs 34%, p<0.001). Also, patients treated in 2005 or after were more likely to receive post-operative RT (n=79, 87%) compared to those treated prior to 2005 (n=55, 54%, p<0.001).
Disease Outcomes
Overall, 81 (42%) patients developed disease recurrence, either local, distant, or a combination of the two. Forty-two patients (22%) experienced local recurrence, 10 (17%) in the pre-operative RT cohort and 32 (24%) in the post-operative RT group (Table 2). 5-year LC for the entire group was 76% (95%CI 69%−82%). There was no significant difference in 5-year LC between the pre-operative RT and post-operative RT cohorts [78% (63%−88%) and 75% (66%−82%), respectively, p=0.483] (See Table 2). Neither size, grade, nor treatment era was found to be associated with LC on univariate analysis (all p>0.05). Patients with negative margins had improved LC at 5 years [86% (77%−92%)] compared to those with positive margins [65% (53%−74%), p=0.003]. Radiotherapy modality had no effect on LC [3D/electrons 76% (67%−82%), IMRT/Protons 77% (62%−86%), p=0.711]. On multivariable analysis, positive margin status was the only variable associated with LC (HR 2.54 [95% CI 1.34–4.82), p=0.004] (Table 3).
Table 2:
5-year Local Control, Disease Free Survival, Distant Metastasis-Free Survival and Disease Specific Survival
| 5-yr LC | p-value | 5-yr DFS | p-value | 5-yr DMFS | p-value | 5-yr DSS | p-value | |
|---|---|---|---|---|---|---|---|---|
| Entire Cohort | 76% (69%−82%) | 58% (51%65%) | 74% (67%−79%) | 74% (67%−80%) | ||||
| Age | 0.329 | 0.152 | 0.331 | 0.423 | ||||
| <65 | 78 (70–84) | 61 (52–68) | 76 (68–82) | 75 (66–81) | ||||
| ≥ 65 | 70 (51–82) | 49 (33–63) | 68 (52–80) | 72 (55–83) | ||||
| Sex | 0.700 | 0.107 | 0.027 | 0.440 | ||||
| Female | 78 (67–86) | 67 (55–76) | 85 (74–91) | 76 (65–84) | ||||
| Male | 74 (64–82) | 52 (42–61) | 66 (56–75) | 72 (62–80) | ||||
| Max Size | 0.553 | 0.012 | 0.008 | 0.117 | ||||
| <5 | 77 (68–84) | 67 (57–75) | 82 (73–88) | 76 (67–83) | ||||
| ≥ 5 | 74 (62–83) | 46 (35–57) | 62 (50–72) | 71 (59–80) | ||||
| Grade | 0.801 | 0.112 | 0.019 | 0.088 | ||||
| 1 | 80 (65–89) | 77 (57–89) | 93 (74–98) | 88 (67–96) | ||||
| 2 and 3 | 78 (65–86) | 57 (46–67) | 72 (61–80) | 78 (68–85) | ||||
| Location | 0.692 | 0.032 | 0.022 | 0.207 | ||||
| Other | 77 (68–84) | 64 (54–71) | 79 (71–86) | 76 (67–82) | ||||
| Neck/SCV | 73 (58–84) | 47 (33–59) | 63 (49–74) | 71 (57–81) | ||||
| Neoadjuvant Chemo | 0.415 | 0.186 | 0.102 | 0.627 | ||||
| No | 78 (70–84) | 62 (53–69) | 76 (68–83) | 74 (66–81) | ||||
| Yes | 68 (50–81) | 48 (32–62) | 67 (51–79) | 72 (55–83) | ||||
| Adjuvant Chemo | 0.003 | 0.036 | 0.11 | 0.003 | ||||
| No | 80 (72–86) | 62 (54–69) | 77 (69–83) | 77 (69–83) | ||||
| Yes | 56 (36–72) | 41 (24–57) | 60 (40–75) | 58 (39–73) | ||||
| Any Chemo | 0.002 | 0.003 | 0.005 | 0.004 | ||||
| No | 85 (76–90) | 68 (58–76) | 82 (73–88) | 80 (71–86) | ||||
| Yes | 62 (49–73) | 43 (31–54) | 63 (50–73) | 65 (52–75) | ||||
| Treatment sequence | 0.483 | 0.834 | 0.928 | 0.823 | ||||
| Pre-op RT | 78 (63–88) | 56 (41–69) | 74 (60–84) | 76 (62–85) | ||||
| Post-op RT | 75 (66–82) | 59 (50–67) | 74 (66–81) | 73 (64–80) | ||||
| Margins | 0.003 | 0.145 | 0.273 | 0.38 | ||||
| Negative | 86 (77–92) | 66 (56–75) | 73 (63–81) | 76 (66–83) | ||||
| Positive/Uncertain | 65 (53–74) | 49 (38–60) | 75 (63–83) | 71 (60–80) | ||||
| Treatment Era | 0.26 | 0.30 | 0.63 | 0.04 | ||||
| Pre 2005 | 72 (61–81) | 55 (44–64) | 72 (62–81) | 68 (57–76) | ||||
| On/After 2005 | 79 (68–86) | 62 (51–71) | 76 (66–83) | 80 (70–87) | ||||
| Advanced RT | 0.711 | 0.884 | 0.522 | 0.18 | ||||
| 3D/electrons | 76 (67–82) | 59 (49–67) | 76 (67–82) | 72 (63–79) | ||||
| IMRT/Protons | 77 (62–86) | 57 (43–69) | 71 (57–81) | 78 (63–87) | ||||
| Local Recurrence | 0.485 | <0.001 | ||||||
| No | -- | -- | 75 (67–81) | 82 (75–88) | ||||
| Yes | -- | -- | 70 (52–83) | 45 (29–60) |
Table 3:
Multivariable analyses of disease control and survival outcomes
| LC | DMFS | DFS | DSS | |||||
|---|---|---|---|---|---|---|---|---|
| OR (95% CI) | p-value | OR (95% CI) | p-value | OR (95% CI) | p-value | OR (95% CI) | p-value | |
| Margins | ||||||||
| Negative | Ref | -- | -- | -- | -- | -- | -- | |
| Positive/Uncertain | 2.54 (1.34–4.82) | 0.004 | -- | -- | -- | -- | -- | -- |
| Grade | ||||||||
| 1 | -- | -- | Ref | -- | -- | -- | -- | |
| 2 and 3 | -- | -- | 4.38 (1.04–18.51) | 0.044 | -- | -- | -- | -- |
| Location | ||||||||
| Other | -- | -- | Ref | -- | -- | -- | -- | |
| Neck/SCV | -- | -- | 2.16 (1.04 – 4.49) | 0.040 | -- | -- | -- | -- |
| Local Recurrence | ||||||||
| No | -- | -- | -- | -- | -- | -- | Ref | |
| Yes | -- | -- | -- | -- | -- | -- | 4.48 (2.62–7.67) | <0.001 |
| Size | ||||||||
| < 5cm | -- | -- | -- | -- | Ref | -- | -- | |
| ≥ 5cm | -- | -- | -- | -- | 1.73 (1.12–2.68) | 0.014 | -- | -- |
Fifty-one patients (26.6%) developed distant metastases (Table 2). The most common site was the lungs (n=39, 20.3%), followed by bone (n=10, 5.2%). Treatment sequence did not influence the development of distant metastasis, with 5-yr DMFS in the pre-op RT cohort 74% (60–84%) and 74% in the post-op RT cohort (66%−81%, p=0.928). DMFS was better at 5 years in those with tumors < 5cm [82% (73%−88%) vs ≥5cm 62% (50%−72%), p=0.008], lower grade tumors [grade 1 93% (74–98%) vs grade 2–3 72% (61%−80%), p=0.019], and those with tumors of the head [79% (71%−86%) vs SCV/neck 63% (49%−74%), p=0.002]. Male patients also had worse 5-yr DMFS [66% (56%−75%) vs female 85% (74%−91%), p=0.027]. A higher portion of patients who received chemotherapy developed distant metastases at 5 years [63% (50%−73%) vs 82% (73%−88%), p=0.005], likely reflecting that chemotherapy was more commonly recommended for larger and high-grade tumors, which have at baseline an increased risk for distant metastases. In adjusted analyses, only tumor location and grade remained significant for DMFS [Neck/SCV HR 2.16 (1.04–4.49), p=0.040; Grade 2–3 HR 4.38 (1.04–18.51), p=0.044] (Table 3).
Overall, 5-yr DFS for the entire cohort was 58% (51%−65%) (Table 2). Size ≥5cm was significantly associated with worse 5-yr DFS [46% (35%−57%), vs <5cm 67% (57%−75%), p=0.012]. Tumor location was also associated with DFS with primaries of the neck/SCV having worse DFS [47% (33%−59%)] vs those of the head [64% (54%−71%), p=0.032]. The use of chemotherapy was associated with worse DFS at 5 years [43% (31%−54%) vs 68% (58%−76%), p=0.003]. Treatment sequence, use of an advanced RT modality, treatment era, and margin status were not associated with DFS. On multivariable analysis the only variable that remained associated with DFS was size, with patients having tumors ≥ 5cm demonstrating worse DFS [HR 1.73 (1.12–2.68), p=0.014] (Table 3).
Five-year DSS for the entire cohort was 74% (67%−80%), and 5-year OS was slightly lower at 71% (64%−77%) (Table 3). DSS was worse in the pre-2005 treatment era [68% (57%−76%)] than on or after 2005 [80% (70%−87%), p=0.04]. Patients experiencing a local recurrence had decreased 5-yr DSS [45% (29%−60%)] compared to those that did not have local recurrence [82% (75%−88%), p<0.001]. There was no significant difference in 5-yr DSS based on treatment sequence, margin status, size, or location (all p>0.05). On multivariable analysis only local recurrence remained associated with DSS [HR 4.48 (2.62–7.67), p<0.001] (Table 3).
Toxicities
MWCs occurred in 23 patients (12%) overall, 13 (22.4%) in the pre-op RT cohort and 10 (7.5%) in the post-op RT cohort (p=0.003). On univariate analysis, treatment sequence was the only variable associated with the development of a MWC (post-op HR 0.28 (0.11–0.68), p=0.005; (Table 4)). There was no significant association with plastic surgery closure at time of surgery (HR 1.74 (0.19–15.5), p=0.62), treatment era (pre-2005 HR 0.68 (0.28–1.66), p=0.399), or the use of an advanced RT modality [HR 1.41 (0.57–3.46), p=0.456] with the risk of developing a MWC.
Table 4:
Long-term radiation toxicities and major wound complications from treatment
| Any RT complication HR (95% CI) | p-value | MWC HR (95% CI) |
p-value | |
|---|---|---|---|---|
| Treatment Sequence | ||||
| Pre-op RT | ref | ref | ||
| Post-op RT | 1.83 (0.58–5.74) | 0.299 | 0.28 (0.11–0.68) | 0.005 |
| Any Chemo | ||||
| No | ref | ref | ||
| Yes | 1.49 (0.59–3.78) | 0.398 | 2.03 (0.84 – 4.89) | 0.115 |
| Age | 0.98 (0.95–1.01) | 0.158 | 1.02 (0.99 −1.04 ) | 0.239 |
| Size | 0.85 (0.68–1.07) | 0.162 | 0.93 (0.77 – 1.12) | 0.45 |
| Advanced RT | ||||
| 3D/electrons | ref | ref | ||
| IMRT/Protons | 1.84 (0.72–4.70) | 0.204 | 1.41 (0.57 – 3.46) | 0.456 |
| Treatment Era | ||||
| On/After 2005 | ref | ref | ||
| Pre-2005 | 0.90 (0.35–2.28) | 0.821 | 0.68 (0.28 −1.66) | 0.399 |
| Reconstruction with Plastic Surgery | ||||
| No | Ref | ref | ||
| Yes | 0.24 (0.05–1.2) | 0.085 | 1.74 (0.19 −15.5) | 0.62 |
| Radiation Dose | 1.04 (0.96–1.12) | 0.393 | - | - |
There were 20 patients (10%) with long-term toxicities associated with RT. The most common toxicity noted was osteoradionecrosis (n=4, 2.1%) and fibrosis (n=3, 1.6%). Other patients developed fractures, brain necrosis, complete nasolacrimal duct obstruction, persistent dry eye, and thyroid dysfunction. Eleven patients required surgical intervention to treat these toxicities. Numerically there were more patients in the post-operative RT group with late RT-related toxicities (n=16, 12%; pre-op n=4, 7%). On univariate analysis treatment sequence was not associated with the development of any late RT-related complication (post-op HR 1.83 (0.58–5.74), p=0.299) (Table 4). Treatment era, size, and the use of chemotherapy were also not associated with the development of a late toxicity (pre-2005 HR 0.90 (0.35–2.28), p=0.821; size HR 0.85 (0.68–1.07), p=0.162; received chemotherapy HR 1.49 (0.59–3.78), p=0.398). Neither the use of an advanced radiation modality (IMRT or protons) (HR 1.84 (0.72–4.70), p=0.204) nor RT dose (HR 1.04 (0.96–1.12), p=0.393) were associated with an increased likelihood of developing a long-term RT-related complication. When looking at patients who required a medical or surgical intervention for their toxicity, there were 12 patients in the post-operative group (9%), and 3 in the pre-operative group (5%), but this was not significantly different (p=0.376). Likewise, there was no significant association found between moderate to severe long-term RT toxicities and radiation dose (p=0.246).
DISCUSSION:
The advantages and disadvantages of pre-operative versus post-operative RT have been well studied for extremity and trunk STS, which comprised the vast majority of patients enrolled historically on prospective STS trials testing the different approaches (9, 12). This is not the case for H&N STS, however. Patients with STS in H&N sites have not been included in the modern, relatively large, prospective cohort trials that inform the decision-making regarding treatment of STS with RT (9, 12, 13). As such, optimal management of H&N STS remains under-investigated due to the rarity of these tumors and lack of data available regarding acute and long-term toxicities.
The literature has firmly established a local control benefit using a combined modality approach with surgery and RT compared to surgery alone for extremity and superficial trunk STS deemed to be at high risk for local recurrence after surgery (4, 5, 14–16). Unfortunately, reports also consistently have shown H&N STS to have poorer outcomes than STS of the extremity or trunk, with local control rates of anywhere from 50–80% (5, 7, 8, 16–18). In this cohort, 5-year LC was 76%, which is lower than the 85–95% LC observed for extremity/trunk STS. The only significant factor which predicted for LC in our cohort was margin status, which is consistent with previous studies (5, 8, 14, 18). Interestingly, use of pre-operative radiation was associated with an increased likelihood of achieving negative surgical margins. Selection bias may account for this trend in our retrospective cohort study, and precludes definitive conclusions about the association between margin status, LC, and timing of RT.
Importantly, LC was found to be associated with improved DSS in this cohort. This has been seen in prior studies of extremity soft tissue sarcomas (19), though other studies have shown equivalent survival in patients who had local recurrence to those who do not (20). Notably, extremity sarcomas have good salvage options including repeat surgery or amputation, which may decrease the survival detriment of a local recurrence. The negative impact on DSS with local recurrence in our cohort may be due to difficulty salvaging local recurrences with surgery in the head and neck region as compared to extremity or truncal locations. Thus, our data suggests that optimal treatment upfront which minimizes local recurrence is important in H&N STS. The feasibility of a salvage surgery should be part of the decision making when seeing patients in the de novo setting and when making decisions about the use of pre- or post-op RT.
Distant metastases were more common in patients with higher grade and larger tumors in this study, consistent with prior reports (5, 8, 14, 18, 21, 22). They were also more common in patients with tumors of the neck than in those with tumors of the head. We believe that this increased risk of distant spread, could partially be attributed to the rich lymphatic drainage and blood supply in the neck (23). Treatment with chemotherapy should be considered for these patients with high risks of distant metastatic spread (4, 5, 15).
In this study, we observed a slight increase in MWCs for patients receiving pre-operative RT compared to post-operative RT, which is consistent with the STS literature regarding RT sequencing for extremity/truncal STS cases. Our observed MWC rate of 12% was lower than historically reported for extremity and superficial trunk STS. The rate of MWCs in the NCIC SR2 trial was 35% for the pre-operative RT arm and 17% in the post-operative RT arm (9), while we observed 22% in pre-operative patients and 7% for post-operative patients in our H&N specific cohort. There were 31% of patients with MWCs in the HYPORT-STS study, and 36.6% in RTOG 0630 (12, 13); both of which are more contemporary single-arm prospective studies investigating pre-operative RT in STS. Previous studies examining major post-operative complications after H&N surgery for squamous cell malignancies report rates between 7–18% with and without pre-operative RT (24–26). Taken together, these findings are consistent with our data and likely are attributable to an overall lower risk of MWC in the H&N region vs the extremity.
ASTRO guidelines currently recommend pre-operative RT for STS of the extremity and trunk to decrease the risk of long-term, function-limiting complications, even with the known risk of higher MWCs, which are usually temporary in nature (27). Long-term side effects from H&N RT including xerostomia, osteoradionecrosis, and dysphagia have all been shown to be dose dependent and dependent on size of the treatment field (28, 29), both of which are generally larger with post-operative RT for sarcomas. We did not see a statistically significant association between RT dose and long-term side effects in this cohort, however this may be due to the low event rate of reported toxicities overall. This could also be partially due to the heterogeneity of treatment locations and modalities used. When there are nearby structures at risk with relatively low dose thresholds, it is important to discuss the difference in dose and field size required for pre-operative versus post-operative RT with the multi-disciplinary team when determining the treatment sequence.
Overall, the sequencing of combined modality local therapy consisting of both RT and surgery requires multidisciplinary discussion and should be performed at a high-volume sarcoma center. It is reasonable to prioritize pre-operative RT in the management of H&N STS due to the increased likelihood of negative surgical margins at surgical resection, as well as the smaller treatment fields and lower dose. However, post-operative RT is still a reasonable option in certain cases. Factors that may help inform the multidisciplinary team include the proximity of the tumor to critical structures that would impact either surgical resection or post-operative dose constraints, and the ability to achieve negative margins.
There are limitations to this retrospective study. We cannot control nor determine the reasons for which patients were chosen for pre-operative or post-operative RT that could confound our analyses. In addition, there was selection bias with respect to who was treated with chemotherapy, and those with higher-risk disease received systemic therapy. Lastly, long-term complications are likely underreported in retrospective studies, such as this one, where physicians are not comprehensively assessing for nor reporting side effects such as dysphagia. This cohort also spans multiple treatment eras with evolution of imaging, RT techniques, chemotherapy, etc. Also, as with most studies looking at STS, there is significant heterogeneity in the tumor histology and location within the H&N.
CONCLUSION:
This study contributes to management considerations for patients with H&N STS in that it is a large cohort of patients with long-term follow up after treatment with RT and surgery for this presentation of STS. We continue to see poorer local control and survival outcomes for H&N STS than patients with STS of the trunk or extremities. Positive margins after surgery was the most important factor associated with local recurrence for H&N STS. Additionally, patients with low-grade tumors, treatment in the modern era, and without local recurrences have improved disease-specific survival. Radiation treatment sequence had no effect on disease outcomes; however, it did have a significant impact on the risk of developing acute post-operative wound toxicities. These data can contribute to multidisciplinary care planning for patients with H&N STS.
Figure 1.
Local control, disease-specific survival, disease-free survival, and distant metastasis–free survival of patients treated with surgery and preoperative or postoperative radiation therapy.
Funding:
Supported in part by Cancer Center Support (Core) Grant P30 CA016672 from the National Cancer Institute, National Institutes of Health, to The University of Texas MD Anderson Cancer Center.
Footnotes
Conflicts of Interest: None of the authors have any conflicts of interest to report
REFERENCES:
- 1.Stiller CA, Trama A, Serraino D, Rossi S, Navarro C, Chirlaque MD, et al. Descriptive epidemiology of sarcomas in Europe: report from the RARECARE project. Eur J Cancer. 2013;49(3):684–95. [DOI] [PubMed] [Google Scholar]
- 2.DeVita VT, Lawrence TS, Rosenberg SA. Cancer of the Breast: From Cancer: Principles & Practice of Oncology: Lippincott Williams & Wilkins; 2016. [Google Scholar]
- 3.Gorsky M, Epstein JB. Head and neck and intra-oral soft tissue sarcomas. Oral Oncol 1998;34(4):292–6. [PubMed] [Google Scholar]
- 4.Mücke T, Mitchell DA, Tannapfel A, Hölzle F, Kesting MR, Wolff KD, et al. Outcome in adult patients with head and neck sarcomas - A 10-year analysis. Journal of Surgical Oncology. 2010;102(2):170–4. [DOI] [PubMed] [Google Scholar]
- 5.de Bree R, van der Waal I, de Bree E, René Leemans C. Management of adult soft tissue sarcomas of the head and neck. Oral Oncology. 2010;46(11):786–90. [DOI] [PubMed] [Google Scholar]
- 6.Mendehall W Adult head and neck soft tissues sarcomas. Head Neck. 2005;10:916–9. [DOI] [PubMed] [Google Scholar]
- 7.Shellenberger TD, Sturgis EM. Sarcomas of the head and neck region. Curr Oncol Rep 2009;11(2):135–42. [DOI] [PubMed] [Google Scholar]
- 8.Levay J, O’Sullivan B, Catton C, Bell R, Fornasier V, Cummings B, et al. Outcome and prognostic factors in soft tissue sarcoma in the adult. International Journal of Radiation Oncology, Biology, Physics. 1993;27(5):1091–9. [DOI] [PubMed] [Google Scholar]
- 9.O’Sullivan B, Davis AM, Turcotte R, Bell R, Catton C, Chabot P, et al. Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the limbs: a randomised trial. Lancet. 2002;359(9325):2235–41. [DOI] [PubMed] [Google Scholar]
- 10.Kaplan EL, Meier P. Nonparametric-Estimation from Incomplete Observations. J Am Stat Assoc 1958;53(282):457–81. [Google Scholar]
- 11.McDonald JH. Handbook of Biological statistics. 3 ed. Baltimore, Maryland: Sparky House Publishing; 2014. [Google Scholar]
- 12.Guadagnolo BA, Bassett RL, Mitra D, Farooqi A, Hempel C, Dorber C, et al. Hypofractionated, 3-week, preoperative radiotherapy for patients with soft tissue sarcomas (HYPORT-STS): a single-centre, open-label, single-arm, phase 2 trial. The Lancet Oncology. 2022;23(12):1547–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wang D, Zhang Q, Eisenberg BL, Kane JM, Li XA, Lucas D, et al. Significant Reduction of Late Toxicities in Patients With Extremity Sarcoma Treated With Image-Guided Radiation Therapy to a Reduced Target Volume: Results of Radiation Therapy Oncology Group RTOG-0630 Trial. J Clin Oncol 2015;33(20):2231–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tran LM, Mark R, Meier R, Calcaterra TC, Parker RG. Sarcomas of the head and neck. Prognostic factors and treatment strategies. Cancer. 1992;70(1):169–77. [DOI] [PubMed] [Google Scholar]
- 15.Kotecha S, Williams M, White HB, Graystone J, Gibbons M, Cosker T. Head and neck sarcoma: three-year data from a tertiary referral centre. The Annals of The Royal College of Surgeons of England. 2021;103(10):762–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Barker JL Jr, Paulino AC, Feeney S, McCulloch T, Hoffman H. Locoregional treatment for adult soft tissue sarcomas of the head and neck: An institutional review. Cancer Journal. 2003;9(1):49–57. [DOI] [PubMed] [Google Scholar]
- 17.Singh RP, Grimer RJ, Bhujel N, Carter SR, Tillman RM, Abudu A. Adult head and neck soft tissue sarcomas: Treatment and outcome. Sarcoma. 2008;2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kraus DH, Dubner S, Harrison LB, Strong EW, Hajdu SI, Kher U, et al. Prognostic factors for recurrence and survival in head and neck soft tissue sarcomas. Cancer. 1994;74(2):697–702. [DOI] [PubMed] [Google Scholar]
- 19.Lewis JJ, Leung D, Heslin M, Woodruff JM, Brennan MF. Association of local recurrence with subsequent survival in extremity soft tissue sarcoma. J Clin Oncol 1997;15(2):646–52. [DOI] [PubMed] [Google Scholar]
- 20.Yang JC, Rosenberg SA. Surgery for adult patients with soft tissue sarcomas. Semin Oncol 1989;16(4):289–96. [PubMed] [Google Scholar]
- 21.Roos JH, Mäkitie AA, Tarkkanen J, Ilmarinen TT. Pretreatment tumor sampling and prognostic factors in patients with soft-tissue sarcoma of the head and neck. European Archives of Oto-Rhino-Laryngology. 2022;279(6):3147–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Weber RS, Benjamin RS, Peters LJ, Ro JY, Achon O, Goepfert H. Soft tissue sarcomas of the head and neck in adolescents and adults. The American journal of surgery. 1986;152(4):386–92. [DOI] [PubMed] [Google Scholar]
- 23.Ji H, Hu C, Yang X, Liu Y, Ji G, Ge S, et al. Lymph node metastasis in cancer progression: molecular mechanisms, clinical significance and therapeutic interventions. Signal Transduction and Targeted Therapy. 2023;8(1):367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Agra IMG, Carvalho AL, Pontes E, Campos OD, Ulbrich FS, Magrin J, et al. Postoperative Complications After En Bloc Salvage Surgery for Head and Neck Cancer. Archives of Otolaryngology–Head & Neck Surgery. 2003;129(12):1317–21. [DOI] [PubMed] [Google Scholar]
- 25.Chaukar DA, Deshmukh AD, Majeed T, Chaturvedi P, Pai P, D’Cruz AK. Factors affecting wound complications in head and neck surgery: A prospective study. Indian J Med Paediatr Oncol 2013;34(4):247–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Pellini R, Mercante G, Marchese C, Terenzi V, Sperduti I, Manciocco V, et al. Predictive factors for postoperative wound complications after neck dissection. Acta Otorhinolaryngol Ital 2013;33(1):16–22. [PMC free article] [PubMed] [Google Scholar]
- 27.Salerno KE, Alektiar KM, Baldini EH, Bedi M, Bishop AJ, Bradfield L, et al. Radiation Therapy for Treatment of Soft Tissue Sarcoma in Adults: Executive Summary of an ASTRO Clinical Practice Guideline. Practical Radiation Oncology. 2021;11(5):339–51. [DOI] [PubMed] [Google Scholar]
- 28.Strojan P, Hutcheson KA, Eisbruch A, Beitler JJ, Langendijk JA, Lee AWM, et al. Treatment of late sequelae after radiotherapy for head and neck cancer. Cancer Treatment Reviews. 2017;59:79–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Davis AM, O’Sullivan B, Turcotte R, Bell R, Catton C, Chabot P, et al. Late radiation morbidity following randomization to preoperative versus postoperative radiotherapy in extremity soft tissue sarcoma. Radiother Oncol 2005;75(1):48–53. [DOI] [PubMed] [Google Scholar]