Abstract
Background/Aim: Radiotherapy for angiosarcoma of the scalp has not been standardised yet. Hence, we aimed to retrospectively analyse the outcomes of patients treated with electron beam therapy or intensity-modulated radiation therapy (IMRT) for unresectable angiosarcoma of the scalp. Patients and Methods: Data from patients treated with chemoradiotherapy or radiotherapy alone for unresectable angiosarcoma of the scalp between March 2009 and March 2021 were evaluated. Survival and local control rates were analysed using the Kaplan-Meier method, and the log-rank test was used to compare groups. Adverse events were analysed using the Common Terminology Criteria for Adverse Events ver. 5.0. Results: Sixteen patients were eligible for the study. Eight patients were treated with electron beam therapy and eight patients with IMRT. The median follow-up period was 18.0 months. The median radiation dose was 57 Gy in 19 fractions in the electron beam therapy group and 70 Gy in 35 fractions in the IMRT group. In the IMRT group, acute non-haematologic toxicity was observed in two patients with grade 3 dermatitis. The one-year overall survival rate, progression-free survival rate, and local control rate in the electron beam therapy group were 80.8%, 56.3%, and 77.4%, respectively, and the corresponding values in the IMRT group were 100%, 75%, and 100%, respectively. One-year local control was significantly better in the IMRT group compared to that in the electron beam therapy group (p=0.016). Conclusion: IMRT for angiosarcoma of the scalp may improve local control rates compared to electron beam therapy, but long-term follow-up studies are required to validate this finding.
Angiosarcoma, a soft tissue sarcoma derived from vascular endothelial cells, occurs predominantly in the scalp of older adults, and has a poor prognosis (1, 2). It is a high-grade tumour that may appear as a benign haemangioma; however, it may also ulcerate, show a tendency to invade surrounding tissues, proliferate, bleed, and then disseminate. Local lesions often extend beyond what can be recognised by visual examination and palpation (1-5), and radical resection with a wide margin is required. However, securing sufficient margins is often challenging; in such cases, concurrent chemoradiation is recommended (6).
Electron beam therapy is usually performed for patients with cutaneous angiosarcoma of the scalp that are ineligible for surgical resection. However, depending on the size and 3-dimensional extent of the tumour, an electron beam may fail to cover the tumour area adequately or provide sufficient dose distribution. Intensity-modulated radiation therapy (IMRT) is commonly used for head and neck tumours, and its therapeutic efficacy and safety have been reported (7, 8). Concerning angiosarcoma of the scalp, a planning study of IMRT with X-rays has shown IMRT to be useful for large and complicated lesions (9, 10). However, only a handful of studies have compared the efficacy of IMRT compared to that of electron beam therapy with respect to adverse events in clinical cases.
At our hospital, angiosarcoma had been treated with electron beam therapy until 2015, when IMRT was introduced to patients with extended disease to improve the dose coverage of the target. In addition, the dose fractionation of IMRT was changed from 57 Gy in 19 fractions to 70 Gy in 35 fractions.
This study aimed to retrospectively analyse the therapeutic efficacy and adverse events of radiotherapy for angiosarcoma of the scalp and examine the benefits and disadvantages of IMRT.
Patients and Methods
Patients. This single-centre retrospective study was approved by the institutional review board of Gunma University Hospital (approval number: HS2021-079). Patients with pathologically confirmed diagnoses of unresectable angiosarcoma and treated with radiotherapy at our hospital between March 2009 and March 2021 were eligible for this study. Since 2015, treatment with electron beam therapy was replaced by that with IMRT for cases presenting with a large angular variation or irregular target shape that precluded effective therapy with a single irradiation field. Consent to participate was obtained in an opt-out format that allows patients to freely express their refusal.
Chemotherapy. The combination of radiotherapy with chemotherapy was the standard approach; docetaxel or paclitaxel was used. The chemotherapy regimen mainly included sequential paclitaxel until 2013, concurrent tri-weekly docetaxel (60 mg/m2) from 2014 to 2018, and concurrent weekly paclitaxel (70 mg/m2) from 2019. Chemotherapy indication was determined by the patient’s overall condition.
Electron beam radiotherapy. A tissue-equivalent 5-mm-thick bolus was applied to the surface in all cases. The gross tumour volume (GTV) was defined as the total tumour area based on imaging findings and mapping by a dermatologist. The clinical target volume (CTV) was adjusted to have a margin of 2 cm from the GTV. The radiation field was defined as CTV plus a 5-10 mm margin. The beam energy was selected to cover the CTV at >90% of the dose referenced with a computed tomography (CT) image. The planning dose was 57 Gy in 19 fractions (three fractions per week), the equivalent dose in 2 Gy was 61.75 Gy (α/β=10 Gy) for CTV in 7 weeks, and the total dose was adjusted based on safety and systemic patient profiles.
Intensity-modulated radiotherapy. In all cases of IMRT, a thermoplastic shell (ORFIT Industries n.v., Wijnegem, Belgium) was used for head fixation, and a tissue-equivalent 5-mm-thick bolus was attached to the surface of the target (Figure 1A and B). CT images with 2.5-mm thick slices were obtained. GTV was defined as the area where the tumour was determined to be present based on imaging findings and mapping by a dermatologist. To reduce the dose to the brain, CTV was adjusted to have a margin of 2 cm from the GTV and to not exceed the skull depth. Two types of planning target volumes (PTVs) were made: PTV1 was made with a 5-mm margin from the GTV, and PTV2 was made with a 3-mm margin from the CTV. The planning dose was 70 Gy and 63 Gy in 35 fractions for PTV1 and PTV2, respectively, and the total dose was adjusted based on safety and systemic patient profiles. The treatment dose was administered to 95% of the target (Figure 1C). For treatment planning, we used Eclipse software (version 13.0; a collaboration product of Elekta AB, Stockholm, Sweden, and Mitsubishi Electric, Tokyo, Japan).
Images of a patient treated with intensity-modulated radiotherapy (IMRT). (A) Tumour localisation and marking of tumour extent (inner line). (B) Installation of the fixing shell and bolus. (C) The prescribed dose distribution of the IMRT was 70 Gy in 35 fractions.
Follow-up and evaluation. After treatment, the patients were followed up every 3 months. In the case of patients who refused regular follow-ups at our hospital, their clinical course was evaluated via telephone, letters, and patient referral documents. Adverse events were evaluated using the Common Terminology Criteria for Adverse Events version 5.0 (11). Overall survival (OS), progression-free survival (PFS), and local control (LC) rates were defined as the periods from the first day of irradiation to death from any cause, disease progression at any site or death from any cause, and tumour regrowth or recurrence in the PTV, respectively, or the last follow-up.
Statistical analysis. We calculated OS, PFS, and LC rates from the first day of irradiation using the Kaplan-Meier method. The log-rank test was used to compare outcomes between groups. p-Values of <0.05 were considered statistically significant. All statistical analyses were performed using Statistical Package for the Social Sciences software, version 25.0 (IBM Inc., Armonk, NY, USA).
Results
Patient characteristics. A total of 16 patients received radiotherapy: eight were treated with electron beam therapy, and eight underwent IMRT. All tumours were larger than 5 cm. Primary tumour characteristics, such as invasion of the face, bleeding from the tumour, and presence of skip lesions, are presented in Table I. Six and nine patients received combination chemotherapy with docetaxel and paclitaxel, respectively. The prescribed doses in electron beam therapy were 54-60 Gy in 18-20 fractions (equivalent dose in 2 Gy of α/β=10 Gy: 58.5-65 Gy), and 70 Gy in 35 fractions except for one patient in the IMRT group. The median follow-up period was 18.0 months.
Patient characteristics.
Efficacy. The 1-year OS rate, 1-year PFS rate, and 1-year LC rate were 80%, 56.3%, and 77.4% in total, 62.5%, 37.5%, and 50% in the electron beam group, and 100%, 75%, and 100% in the IMRT group, respectively. IMRT was preferable; there was no statistical superiority in OS and PFS, but it was significantly better in LC than in the electron beam group, and the p-value was 0.119, 0.084, and 0.016, respectively (Figure 2).
Survival rates and control rates of the total (red) and intensity-modulated radiotherapy (IMRT) groups (blue). (A) Overall survival rate, (B) progression-free survival rate, and (C) local control rate.
The details of the first failure site are presented in Table II. No local recurrence was observed in patients treated with IMRT. Cases of regional recurrence included a case of scalp skin recurrence outside of the irradiated field and a neck lymph node recurrence. The latter case was treated with docetaxel, which was terminated due to drug-induced pneumonia; multiple lung metastases were observed at the end of the radiotherapy course; the cervical lymph node recurrence was subsequently observed. In the electron beam therapy group, one patient treated with radiotherapy alone developed splenic metastasis at the end of treatment.
First failure site (including duplicates).
Adverse events. Adverse events are presented in Table III. Grade 3 dermatitis was observed in two (12.5%) cases. Grade 3 dermatitis was observed in a patient treated with IMRT, which was terminated after the dose of 66 Gy was delivered in 33 fractions. In addition, grade 3 mucositis was observed in one patient treated with IMRT. Moreover, three and two cases of drug-induced pneumonia and treatment discontinuation were observed in the IMRT and electron beam therapy groups, respectively. No other acute non-haematologic adverse events of grade ≥2 were observed. All patients that presented with grade 3 dermatitis had experienced tumour bone invasion before treatment initiation. No severe adverse events were observed. Acute haematological adverse events of grade ≥4 were observed in one case in the IMRT group. One patient required treatment interruption because of adverse events and an overall poor condition.
Adverse events.
Discussion
In this study, the overall 1-year OS, PFS, and LC rates were 80.8%, 56.3%, and 77.4%. The corresponding values in the IMRT group were 100%, 75.0%, and 100%, respectively. LC was better in the IMRT group than that in the electron beam therapy group. Adverse events did not interrupt radiotherapy. Grade 3 dermatitis was observed in cases with deep tumour invasion. Chemotherapy was discontinued due to drug-induced lung injury in five cases.
The improved LC rate in the IMRT group, compared to that in the electron beam therapy group, may be due to several factors. First, the higher dose in the IMRT group compared to the latter treatment may have achieved the improvement in LC. Kashihara et al. reported the total biological effective dose of >80 Gy to be associated with improved LC (12). The median dose in the electron beam group was 57 Gy in 19 fractions (EQD2=61.75 Gy, α/β=10 Gy), compared to 70 Gy in 35 fractions in the IMRT group, which corresponds to an 11.3% relative increase. Second, IMRT may have enabled sufficient doses to be delivered to the entire target, even in cases of extensive lesions or where dose reduction to organs at risk is difficult. Other radiotherapy outcomes include surgery and concomitant immunotherapy. Sasaki et al. examined the outcomes of 30 angiosarcoma patients treated with electron beam therapy (median dose: 68 Gy) and surgery or immunotherapy, reporting a 1-year LC rate of 57% (13). Suzuki et al. reported outcomes of 14 patients with angiosarcoma of the face and scalp treated using electron beam radiotherapy (median dose: 70 Gy) with surgery or immunotherapy, estimating a 1-year LC at 71% (14). Kashihara et al. reported outcomes of 27 patients with angiosarcoma of the scalp treated with volumetric-modulated arc therapy and IMRT (median dose: 70 Gy), estimating a 2-year local progression-free rate of 48.4% (12). Overall, this evidence suggests that IMRT may improve local outcomes.
In our study, cases of grade 3 dermatitis were observed in the IMRT group; in contrast, no case of grade 2 dermatitis was observed in the electron beam therapy group; the increased treatment dose may have caused the dermatitis. Additionally, all cases of grade 3 dermatitis occurred in patients with deep tumour invasion into the skull; achieving tumour control was challenging in these cases. The observed incidence and severity of dermatological and haematological adverse events are consistent with those previously reported (13-15). IMRT increases the dose to the brain; thus, long-term brain-related adverse events should be examined.
Roy et al. showed that concurrent paclitaxel chemoradiotherapy (CRT) for cutaneous angiosarcoma improved 2-year OS rates, but not LC or PFS rates, compared to a non-CRT approach (15). Herein, all patients received chemotherapy except one. There was no case in which chemotherapy was suspended due to radiotherapy-related adverse events. Notably, chemotherapy was discontinued in five (31.3%) patients due to drug-induced lung injury. These findings suggest a need for future studies that help identify optimal regimens as concomitant systemic therapy in this context.
This study had some limitations. First, the follow-up period was short. Long-term follow-up in this context is challenging because of the poor prognosis associated with the disease. Nevertheless, late adverse events associated with IMRT, such as brain necrosis, may have been inadequately evaluated. Second, the sample size was small. Angiosarcoma is a rare and aggressive disease and poses recruitment challenges; however, future studies should focus on achieving larger sample sizes.
Conclusion
Definitive radiotherapy for unresectable angiosarcoma of the scalp was associated with 1-year OS, PFS, and LC rates of 80.8%, 56.3%, and 77.4%, respectively; the LC rate was better in the IMRT group than in the electron beam therapy group. High-dose IMRT with chemotherapy showed improved LC with acceptable adverse events. Long-term large follow-up studies are required to confirm the safety and efficacy of IMRT with systemic chemotherapy.
Acknowledgements
The Authors are grateful to Editage for reviewing the English language.
Footnotes
Authors’ Contributions
Conceptualisation, N.O., H.Y.; Formal analysis, N.O., H.Y.; Data collection and curation, N.O., H.Y., S.S.; Writing manuscript, N.O., H.Y., N.K.; Review and editing, S.S., N.K., K.O., H.K., M.Y.; Supervision, H.K., M.Y., T.O. The Authors read and approved the final manuscript.
Conflicts of Interest
N.O. and T.O. received an honorarium from Hitachi Industries. The other Authors have no conflicts of interest directly relevant to the content of this article.
- Received September 27, 2022.
- Revision received October 6, 2022.
- Accepted October 7, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
