Abstract
Background/Aim: Carbon-ion radiotherapy (CIRT) has been reported to obtain favorable results in the treatment of bone and soft tissue malignancies; however, studies on CIRT for soft tissue sarcomas (STS) of the extremities are limited. Here, we have retrospectively evaluated the therapeutic efficacy and adverse events associated with scanning CIRT (sCIRT) for STS of the extremities at our institution. Patients and Methods: Thirteen consecutive patients with STS who underwent sCIRT between January 2017 and January 2020 were included in the study. The total dose of sCIRT was set at 67.2-70.4 Gy (RBE), which was provided in 16 fractions. Overall survival (OS), progression-free survival (PFS), and local control (LC) were estimated using the Kaplan–Meier method. Toxicity was evaluated using Common Terminology Criteria for Adverse Events v5.0. Results: The cohort consisted of 10 males and 3 females with a median age of 69 years (range=38-95 years). Median duration of observation was 31.8 months (range=7.4-56.4 months). Tumors were localized to the upper extremity in 2 cases and to the lower extremity in 11 cases. Median maximum tumor diameter was 11.7 cm (range=3.0-36.6 cm), while 3-year OS, PFS, and LC were 61.5%, 44.9%, and 79.1%, respectively. Acute toxicity of grade 3 or higher was not observed. Late toxicity included grade 3 peripheral nerve palsy and decreased range of motion in 1 and 1 patient each. Late toxicity of Grade 4 or higher was not observed. Conclusion: sCIRT for STS of the extremities demonstrates favorable therapeutic results with acceptable toxicity.
Soft tissue sarcoma (STS) is a rare and heterogeneous tumor that accounts for approximately 1% of all malignant tumors in adults (1). It can occur anywhere in the body; however, it is most commonly seen on the extremities, which account for approximately 60% of all STSs (2, 3). Surgery is the primary treatment modality for STS of the extremities (4), and while amputation may be useful for local control (LC), it does not increase overall survival (OS) or progression-free survival (PFS) (5) and is also highly invasive. Therefore, in recent years, the main treatment strategy for STS of the extremities has shifted to a combination of extensive local excision and radiation therapy (RT) as it results in favorable LC (6, 7). Nevertheless, treatment options are limited if patients refuse surgery or if the tumor is inoperable. Additionally, STS is radioresistant, and RT for unresectable STS is not efficacious (8).
Carbon-ion radiation therapy (CIRT), first performed in 1994 at the National Institute of Radiological Sciences (Chiba, Japan) (9), has physical and biological advantages over conventional X-ray therapy; specifically, higher dose concentration due to its Bragg peak and sharp penumbra (10), and approximately three times greater biological effectiveness than that of conventional X-rays (11, 12). These properties confer a favorable local effect while suppressing normal tissue toxicity. In fact, promising results have been reported with CIRT for malignant bone and soft tissue tumors (13-17). CIRT was initiated at our institution in 2015 (18) and is delivered by the raster scanning method (sCIRT) as it can provide better dose distribution based on tumor geometry (19).
Phase I/II trials of CIRT for malignant bone and soft tissue tumors of the extremities have reported 5-year OS and LC rates of 56% and 76%, respectively (20), but clinical outcomes after sCIRT alone for STS of the extremities have not yet been reported. Therefore, we retrospectively analyzed the therapeutic efficacy and toxicity of sCIRT for STS of the extremities.
Patients and Methods
Patients. The institutional review board approved this study (approval number: 2022-14). Written informed consent was obtained from all patients. Study subjects were patients with STS who underwent sCIRT at Kanagawa Cancer Center between January 2017 and January 2020. Eligibility criteria for inclusion were (i) histopathologically diagnosed STS, (ii) no lymph node or distant metastasis, (iii) gross tumor measurement possible on computed tomography (CT) or magnetic resonance imaging (MRI), and (iv) performance status of 0-2. Clinical data were obtained in March 2022.
sCIRT. All patients underwent sCIRT. Gross tumor volume (GTV) was delineated using CT, MRI, and positron emission tomography (PET)-CT. Clinical target volume (CTV) was defined as GTV plus 1 cm margin circumferentially and 2-4 cm margin longitudinally, and was excluded for areas with low potential for invasion. Planning target volume (PTV) was CTV plus 0.5-1 cm margin circumferentially. Total dose was set at 70.4 Gy (relative biological effect, RBE), typically provided in 16 fractions over 4 weeks, but was set at 67.2 Gy (RBE) in 16 fractions for cases where tumor was close to risk organs. Further, 95% of the PTV volume was scheduled to be delivered by 95% of the prescribed dose. If dose constraints of risk organs could not be met, either the PTV margin or the dose was reduced to accommodate the constraints. Dose limitations for risk organs were as follows; maximum dose (Dmax) <60 Gy (RBE) for skin, maximum dose covered 1 cc (D1cc) <40 Gy (RBE) for small and large intestine, Dmax <60 Gy (RBE) for bladder, Dmax <30 Gy (RBE) for spinal cord, D 10cc <67.2 Gy (RBE) for the sciatic nerve.
Follow-up. After completion of sCIRT, a radiation oncologist and a musculoskeletal tumor surgeon followed-up all patients every 1-3 months. Imaging studies, namely, CT and MRI were performed every 1-3 months and PET-CT was acquired to diagnose recurrence or distant metastasis, if necessary. Recurrence or distant metastasis were diagnosed by both the radiation oncologist and the musculoskeletal tumor surgeon. Local failure was defined as relapse within the PTV (20). Observation period and time-to events were calculated from the date of the initiation of sCIRT. Toxicity was assessed using the Common Terminology Criteria for Adverse Events ver 5.0, with those occurring less than 3 months after sCIRT initiation defined as acute, while those occurring after 3 months considered to be late toxicities. The worst toxicity grade observed was used for toxicity evaluation.
Statistical analysis. OS, PFS, and LC were estimated using the Kaplan-Meier method. All statistical analyses were performed using STATA software (version 17.0, College Station, TX, USA).
Results
Patient characteristics. Thirteen patients were included in this study. Table I summarizes patient characteristics. The cohort comprised 10 men and 3 women with a median age of 69 years (range=38-95 years). Median follow-up period was 31.8 months (range=7.4-56.5 months). Tumor location was upper extremity in 2 patients and lower extremity in 11 patients, and the median value for maximum tumor diameter was 11.7 cm (range=3.0-36.6 cm). Histological types were liposarcoma (n=5), undifferentiated pleomorphic sarcoma / undifferentiated sarcoma (n=4), spindle cell sarcoma (n=2), myxofibrosarcoma (n=1), and solitary fibrous tumor (n=1). Of 5 liposarcoma patients, the subtypes were dedifferentiated (n=2), myxoid (n=2), and well-differentiated (n=1). Reasons for undergoing sCIRT were refusal of surgery (n=8), non-resectable tumor (n=3), and inoperability due to complications (n=2). Primary tumors were seen in eight patients while postoperative recurrence was seen in five. Of these 5 patients, one patient received chemotherapy as adjuvant therapy for surgical resection, and 2 patients received chemotherapy as salvage treatment for recurrence diseases after surgical resection. A total dose of 70.4 Gy (RBE) and 67.2 Gy (RBE) was delivered in 11 and 2 patients, respectively.
Patient characteristics (n=13).
OS, PFS, and LC. OS at 3 years was 61.5% [95% confidence interval (CI)=30.8-81.8%] (Figure 1) while median duration of OS was 31.8 months. There were 6 deaths during the observation period; of these, 5 were due to primary disease. PFS at 3 years was 44.9% (95%CI=17.7%-69.0%; Figure 2) and median duration of PFS was 8.3 months. There were 7 recurrences during the observation period with the first recurrence modes being distant metastasis and local failure in 5 and 2 cases, respectively. The first distant metastases were observed in the lung (n=3), lymph nodes (n=1), and at different sites in the irradiated limb (n=1). In one patient with metastasis in the upper limb, chemotherapy was administered after the recurrence, but as the disease later progressed, a shoulder disarticulation was performed 25 months after sCIRT. Importantly, extremities were preserved in all cases except this. LC at 3 years was 79.1% (95%CI=36.7-94.7%; Figure 3). Two cases of local failure were recorded during the observation period – time-to event was 4.6 and 14.1 months, respectively, with tumor diameters of 14.7 cm for spindle cell sarcoma and 10.8 cm for dedifferentiated liposarcoma, respectively. Both patients had received 70.4 Gy (RBE) of sCIRT.
Overall survival (OS) of patients with soft tissue sarcoma who underwent scanning carbon-ion radiotherapy. OS rate at 3 years was 61.5% (95% confidence interval=30.8-81.8%). Median OS duration was 31.8 months.
Progression-free survival (PFS) of patients with soft tissue sarcoma who underwent scanning carbon-ion radiotherapy. PFS rate at 3 years was 44.9% (95% confidence interval=17.7-69.0%). Median PFS duration was 8.3 months.
Local control (LC) of patients with soft tissue sarcoma who underwent scanning carbon-ion radiotherapy. LC rate at 3 years was 79.1% (95% confidence interval=36.7-94.7%). There were 2 cases of local recurrence during the observation period.
Toxicity. Table II summarizes toxicity data. Acute toxicity included radiation dermatitis of grades 1 and 2 in 7 (54%) and 5 (38%) patients, respectively, with no occurrence of acute dermatitis that was grade 3 or higher. No other acute toxicity was observed. Late dermatitis, grade 1, was observed in 2 (15%) patients, and other late toxicities included grade 2 neuralgia in 4 (31%) patients and grade 3 peripheral neuropathy in 1 (8%) patient. The grade 3 neuropathy presented as paralysis of the ankle joint due to sciatic neuropathy and appeared 23 months after sCIRT. Limitations in range of motion (ROM) of the joints of grades 2 and 3 were observed in 1 (8%) patient each. The grade 3 ROM limitation appeared 6 months after the sCIRT, gradually worsened, and resulted in complete loss of joint function. Grade 1 edema was observed in 1 (8%) patient.
Toxicity data (n=13).
Discussion
We retrospectively evaluated clinical outcomes of sCIRT for STS of the extremities in a cohort of 13 patients and show that OS, PFS, and LC at 3 years were 61.5%, 44.9%, and 79.1%, respectively. Additionally, no severe acute toxicity was observed and grade 3 peripheral nerve palsy and joint dyskinesia were observed in 1 and 1 patient each in late phase. To best of our knowledge, this is the first report to describe clinical outcomes after sCIRT for STS of the extremities.
Even though surgery is the main treatment modality for STS of the extremities (4), RT plays an important role as adjunct therapy for local control and functional preservation (7). A randomized trial (n=43) that compared amputation (n=16) versus conservative surgery and postoperative RT (n=27) for STS of the extremities reported 5-year LC rates of 100% and 85% in the amputation and conservative groups, respectively, with no significant difference in LC, OS, or PFS between the two groups (21). Yang et al. randomized 91 patients to conservative surgery alone (n=44) or conservative surgery and postoperative RT (n=47) groups and reported a significantly higher LC rate of 99% in the postoperative RT group compared to 70% in the surgery alone group, with no significant difference in OS between the two groups (22). Thus, while favorable outcomes have been reported for the combination of conservative surgery and RT, the local effects of RT alone without surgery are inadequate. Specifically, Kepka et al. have analyzed outcomes with RT in 112 patients with unresectable STS involving the extremities and found that 5-year LC rate for all patients was 45%. Furthermore, the 5-year LC rate was only 9% when the tumor was >10 cm (8). In contrast, we demonstrate that sCIRT for STS of the extremities shows good local efficacy even though our cohort included cases deemed inoperable, patients who rejected surgery, and median tumor size exceeding 10 cm.
A phase I/II dose escalation study of CIRT for malignant bone and soft tissue tumors has reported 3-year OS and LC rates of 46% and 73%, respectively (13) and another report states that 5-year OS and LC rates after CIRT for unresectable retroperitoneal sarcoma were 50% and 69%, respectively (16). Similarly, CIRT for unresectable head and neck sarcomas resulted in 3-year OS and LC rates of 74.8% and 91.8%, respectively (23), while 5-year OS and LC rates after CIRT for unresectable axial STS have been reported to be 46% and 65%, respectively (24). In another phase I/II dose escalation study on CIRT for STS of the extremities, the 3-year OS and LC rates were 68% and 76%, respectively (20), and we show 3-year OS and LC rates after sCIRT for STS of the extremities to be 61.5% and 79.1%, respectively. Thus, CIRT for malignant bone and soft tissue tumors shows good therapeutic efficacy, regardless of tumor location, even for unresectable lesions.
A prospective phase II trial that compared conventional RT to intensity-modulated RT (IMRT) during postoperative irradiation of STS of the extremities showed reduced wound complications in the IMRT group (25). Another prospective phase II trial for STS of the extremities reported significantly less late toxicity such as fibrosis, edema, and joint contractures in the IMRT group compared to the conventional RT group (26). Notably, CIRT has been shown to exhibit better dose distribution compared to IMRT in several diseases (27-30), and the above results on IMRT suggest that dose reduction for normal tissues by IMRT may be useful in lowering RT-related toxicity. Therefore, CIRT for STS of the extremities is expected to reduce toxicity, and a previous report has revealed that, among 17 cases of CIRT for STS of the extremities, only one instance of Grade 3 or greater late toxicity, which manifested as femoral fracture, was observed (20). In our study, there were also few severe adverse events (late toxicity, Grade 3), namely, one case each of peripheral nerve palsy and joint contracture.
Our study has several limitations such as the single-center nature of the study, small number of cases, and a short observation period. Nevertheless, as there are few reports on CIRT for STS of the extremities, data on optimal margin setting and tolerated doses for normal organs such as the bones, nerves, and joints are sparse, and we believe that our report provides relevant and valuable data. Further studies are needed to increase the number of cases and to examine longer-term clinical outcomes after CIRT for STS of the extremities.
Conclusion
As sCIRT for STS of the extremities showed good therapeutic efficacy and acceptable toxicity, we believe that sCIRT may be a promising treatment option for unresectable STS or patients who refuse surgery.
Footnotes
Authors’ Contributions
YT collected and analyzed the data and drafted the manuscript. H Katoh, DY, TH and TK analyzed the data and contributed to the final draft of the manuscript. H Koge, KK and SS collected and analyzed the clinical data. IS, KT and NM aided in writing the manuscript and contributed to the final draft of the manuscript. All Authors read and approved the final manuscript.
Conflicts of Interest
Hiroyuki Katoh and Daisaku Yoshida received research funding from Toshiba Energy Systems and Solutions Corporation (Kanagawa, Japan).
- Received May 30, 2022.
- Revision received June 16, 2022.
- Accepted June 17, 2022.
- Copyright © 2022 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
