Radiotherapy (RT) has been increasingly utilized as the local modality for hepatocellular carcinoma (HCC) [1]. Various RT techniques have been developed, that enable the delivery of high-dose radiation to the target volume accurately and safely [2]. Proton beam therapy (PBT), one of the advances in RT techniques, has dosimetric advantages over X-ray therapy due to its physical properties such as the use of the Bragg peak and the absence of an exit dose [1,3]. Therefore, PBT allows for dose escalation in the RT of HCC without increasing the probability of radiation-induced liver disease or other gastrointestinal toxicities. On the contrary, by dose escalation, the chest wall proximal to the tumor is also irradiated with high dose. Some studies have reported high rates of chest wall toxicity, such as rib fracture (16%) [4] and chest wall pain of grade 2 or more (30%) [5] after PBT for HCC. Herein, we report a case of radiation-induced myositis after PBT of HCC.

A 58-year-old male patient was referred to our hospital with incidentally detected huge liver mass on abdominal ultrasonography. He had a history of chronic hepatitis B virus (HBV) infection, but was not under regular HCC surveillance. The patient had no symptoms except dyspepsia. There were no remarkable findings including jaundice, palpable mass or ascites on physical examination. He had been consuming about 15 standard drinks per week for 30 years. He was a never smoker, was self-employed, and had a family history of hypertension and thyroid cancer (younger brother). The initial laboratory findings at our hospital were as follows: normal complete blood cell count with 181,000 platelets per microliter of blood; total bilirubin 0.7 mg/dL; albumin 4.3 mg/dL; aspartate aminotransferase (AST) 95 IU/L; alanine aminotransferase (ALT) 62 IU/L; alkaline phosphatase (ALP) 69 IU/L; gammaglutamyl transpeptidase (GGT) 164 IU/L; international normalized ratio (INR) 1.04; serum creatinine 1.05 mg/dL (estimated glomerular filtration rate of 78.4 mL/min/1. 73 m²). Alpha-fetoprotein (AFP) level was 17.3 mg/dL and prothrombin induced by vitamin K absence-II (PIVKA-II) level was 13,000 mAU/mL. Serum HBV DNA level was 495,573 IU/mL. The patient had good liver function with Child-Pugh class A (score 5) and the indocyanine green retention rate at 15 minutes (ICG R15) of 11.6%.

Initial contrast-enhanced magnetic resonance images (MRI) using Primovist showed a 20 cm sized HCC with several daughter nodules in the right liver, and borderline sized lymph nodes around the hepatic artery and the portocaval space (Fig. 1). There was no distant metastasis or major vascular invasion. The clinical staging was stage III (T3N0M0) according to the modified Union for International Cancer Control (UICC) staging and stage B according to Barcelona Clinic Liver Cancer (BCLC) staging. The patient was started with antiviral treatment for HBV using entecavir, and HCC treatment using transarterial chemoembolization (TACE) followed by PBT.

Conventional TACE was performed with an intra-arterial injection of a mixture of doxorubicin hydrochloride (Adriamycin; Dong-A Pharm, Seoul, Korea) 40 mg and iodized oil (Lipiodol; Laboratoire Andre Guerbet, Aulnay-sous-Bois, France) 10 mL after a femoral approach, celiac angiogram and superselection of tumor feeder at the level of the segmental or subsegmental artery with a micro-guidewire and a 2.0-Fr microcatheter. The feeder were embolized with gelatin sponge pledgets (Cutanplast, Mascia Brunelli S.P.A, Milano, Italy) until hemostasis was achieved.

The day after the initial TACE, serum total bilirubin, AST, and ALT increased up to 1.7 mg/dL, 2,270 IU/L, and 484 IU/L, respectively. Serum C-reactive protein (CRP) level was elevated over 6 mg/dL and fever developed 2 days after procedure. There was no evidence of definitive infection on microbiological cultures or significant complication on immediate follow-up computed tomography (CT) scan. The patient fully recovered and was discharged after 7 days of conservative treatment under the diagnosis of post-TACE syndrome without an increase of Child-Pugh score.

Radiation-induced myositis is a rarely reported adverse effect of RT [6]. Histologically, radiation-induced myositis is late radiation injury to muscle, causing focal muscle degeneration, loss of capillaries and increase in collagen [7]. Radiation-induced myositis was reported as muscle injury after stereotactic body RT (SBRT) using high doses in 1 to 3 fractions. Lockney et al. [8] identified 11 (1.6%) patients showing radiographic evidence of myositis following high dose spinal SBRT among 667 patients. The median times to the development of radiographic evidence and clinical symptoms from myositis were 4.7 and 1.4 months after SBRT, respectively. On the other hand, various literatures also showed cases of myositis as radiation recall phenomenon, which is defined as acute inflammation in previously irradiated tissues, typically occurring weeks to months following RT and subsequent systemic therapy [9]. Not only cytotoxic agents such as gemcitabine, carboplatin, and paclitaxel [10,11], but also targeted agents [12,13] are known to be associated with radiation recall myositis. Mostly, radiation recall myositis was reported as case report due to its extremely low incidence [11]. In case of PBT which enables high dose irradiation on tumors without increasing the normal liver dose, the external tissue proximal to the tumor can also be irradiated with the high dose. Some literatures have reported chest wall toxicity such as rib fracture (16%) [4] and chest wall pain of grade 2 or more (30%) [5] after PBT for HCC.

In our case, both the symptom and the radiographic finding consistent with myositis in chest wall appeared at six months after the completion of PBT. The area of myositis was the area proximal to the gross tumor volume targeted to receive the prescribed dose of 72.6 CGE in 22 fractions. Therefore, high doses of up to 71.1 CGE in 22 fractions were delivered in areas of chest wall where myositis occurred. There is no consensus regarding dose constraints for chest wall toxicity resulting from radiation myositis. However, efforts to reduce the dose delivered to the chest wall should be made. Especially, during the planning of PBT, which is more beneficial in normal liver sparing than X-ray RT, dose escalation is frequently tried. Therefore, cautions to save adjacent normal tissues reported to be not vulnerable to RT, are necessary because these normal tissues can be irradiated with high doses according to dose escalation.

In addition to high dose irradiation to the chest wall, sorafenib or regorafenib might be trigger factors for radiation myositis due to radiation recall phenomenon. The mechanisms of radiation recall phenomenon by sorafenib are poorly understood. Phillips and Fu [14] suggested that drug sensitivity has changed by radiation, through alteration in cell age distribution. Kim et al. [15] suggested that RT induces the release of certain cytokines, including interleukin-1, interleukin-6, platelet-derived growth factor-beta, tumor necrosis factor, and transforming growth factor, which cause inflammatory response. However, the mechanisms for radiation recall are not clearly established. In addition, reports of radiation recall phenomenon by sorafenib are mainly limited to dermatitis and there is no relevant report about myositis. However, the course of myositis is similar to that in previous reports showing radiation recall dermatitis after the administration of sorafenib [15,16]. In our case, sorafenib was administered at two months after the completion of PBT because of pulmonary metastasis. Myositis occurred at four months after the start of sorafenib treatment. Previous literatures reported the onset of radiation recall dermatitis after the initiation of sorafenib as 7 to 14 days which is shorter than the present case. The differences in the onsets of the phenomena might result from various factors such as difference in sensitivity according to the tissue, RT regimen, or the interval between the RT and sorafenib treatment. Although there is no report relevant to radiation recall myositis induced by sorafenib or regorafenib specifically, our case implies the probability of radiation recall myositis associated with the targeted agents and further studies elucidating this mechanism would be required.

While the pain induced by radiation myositis was very severe, it was well controlled by steroid in our case. Previous literatures also reported that the pain by myositis was well controlled by steroid and/or non-steroidal anti-inflammatory drugs (NSAID) [8]. The discontinuation of the causative drugs also can be helpful for radiation recall myositis [11]. Because radiographic change usually precedes the presentation of symptoms, close observation would be required. When radiologic findings representing muscle edema such as increased T2 signal with bandlike demarcation from nonirradiated tissue in magnetic resonance image after high dose RT is shown [17], one should consider the possibility of radiation myositis and early intervention may be warranted.

In conclusion, dose escalation by PBT can induce myositis adjacent to the treatment area. Sorafenib after PBT may be one of the causes for radiation myositis as a radiation recall phenomenon. Therefore, effort to reduce the dose in muscles adjacent to target volumes and careful observation regarding radiation recall myositis after use of sorafenib are required. Steroid and/or NSAID are found to be effective in the control of radiation myositis symptoms.