Carbon-ion Radiotherapy for Oligometastatic Colorectal Cancer in the Liver or Lung

Background/Aim: We aimed to evaluate the clinical outcomes of oligometastatic colorectal cancer in the liver and lung treated with carbon-ion radiotherapy (C-ion RT). Patients and Methods: Nineteen consecutive patients with oligometastatic colorectal cancer in the liver or lung who received C-ion RT were analyzed. The doses of C-ion RT were 60.0 Gy [relative biological effectiveness (RBE)] in 4 fractions, 60.0 Gy (RBE) in 12 fractions, or 64.8 Gy (BRE) in 12 fractions. Results: The median follow-up duration was 19 months. There were 23 tumors in 19 patients. The 2-year overall survival and local control rates for the whole patient cohort were 100% and 67%, respectively. None of the patients developed grade 2 or higher acute or late toxicities. Conclusion: C-ion RT for oligometastatic colorectal cancer in liver and lung provides favorable clinical outcomes. These outcomes suggest C-ion RT is a treatment option for oligometastatic colorectal cancer in liver and lung.

Oligometastatic disease is an intermediate state between localization and widespread dissemination (1). Therefore, controlling oligometastatic disease by local treatment may improve survival. Colorectal cancer is one of the cancer types that are characterized by oligometastases of the liver and lung. Local treatment of oligometastatic colorectal cancer, with or without chemotherapy, is performed to improve survival (2)(3)(4)(5)(6)(7). In contrast, radiotherapy (RT) is performed as a local treatment approach for patients who are not indicated for surgery due to comorbidity or refusal of surgery (8)(9)(10)(11)(12)(13)(14)(15).
Carbon-ion (C-ion) RT is performed to treat various types of cancer, including hepatocellular carcinoma, lung cancer, and oligometastatic disease (16)(17)(18)(19)(20)(21). C-ion RT has biological and physical advantages over photon therapy. Owing to its biological properties, C-ion RT has a higher relative biological effectiveness (RBE) due to the high linear energy transfer in the Bragg peak. Furthermore, its physical properties allow administration of high doses while sparing normal tissues because of its higher dose localization ability with distal tail-off enabled by the Bragg peak and sharp lateral penumbra (22). Previous studies have demonstrated a dose distribution advantage, showing that C-ion RT delivered a reduced dose to the normal liver and lung compared with stereotactic body RT (SBRT) and intensity-modulated RT (23)(24)(25). The biological and physical advantages may contribute to favorable clinical outcomes. However, research on the clinical outcomes of oligometastatic colorectal cancer treated with C-ion RT is limited. Hence, we aimed to evaluate the clinical outcomes of oligometastatic colorectal cancer in the liver and lung treated with C-ion RT.

Patients and Methods
Patients. We reviewed the medical records of patients with oligometastatic colorectal cancer in the liver or lung treated with Cion RT at Gunma University Heavy Ion Medical Center between October 2013 and March 2020. We enrolled 19 consecutive patients who met the following criteria: i) Liver or lung metastases from colorectal cancer as confirmed by histology or radiography; ii) curative resection for primary disease and regional lymph nodes, without gross or microscopic residual disease; iii) absence of local primary colorectal lesion and lymph node recurrence; iv) absence or control of extrahepatic or extra-thoracic disease; v) ≤3 synchronous liver or lung metastases at the time of treatment; vi) not indicated or refused surgery for metastatic disease of liver or lung ; vii) radiographically measurable tumor; and viii) performance status ≤3 by the Eastern Cooperative Oncology Group classification. Cases were excluded if they had received prior RT to the target area, had intractable infections in the target area, or had received chemotherapy/molecular targeted therapy within 4 weeks before the initiation of C-ion RT. The treatment protocol was reviewed and approved by the Gunma University Institutional Review Board (approval number: HS2019-130), and all patients signed an informed consent form before the initiation of therapy.
Carbon-ion radiotherapy. A heavy ion accelerator at Gunma University Heavy Ion Medical Center generated C-ion beams, and the beam energy was either 290 MeV/u, 380 MeV/u, or 400 MeV/u according to the tumor depth. The XiO-N system (version 4.47; collaborated product of Elekta AB, Stockholm, Sweden, and Mitsubishi Electric, Tokyo, Japan) was used for treatment planning. This system incorporates a dosing engine for ion beam RT (K2dose) (25). We calculated the clinical radiation dose based on the physical dose multiplied by the RBE of the C-ions. Before C-ion RT, patients were immobilized using tailor-made fixation cushions and thermoplastic shells to allow computed tomography (CT); respiratorygated and 4-dimensional CT images were acquired. In actual treatment, the gating level for respiratory-gated irradiation was within 30% of the wave height around the peak exhalation. Patients received C-ion RT once daily, 4 days a week (Tuesday to Friday).
The gross tumor volume (GTV) was delineated by the treatment planning CT images, which were merged with the contrast-enhanced CT images, contrast-enhanced magnetic resonance imaging (MRI), with/without 2-deoxy-2-[ 18 F]fluoro-D-glucose (FDG)-positronemission tomography (PET)/CT images if necessary. The clinical target volume had 5-10 mm margin around the GTV to include microscopic disease. The internal margin was assessed using 4dimensional CT images for tumor movement. The planning target volume was defined as the summation of the clinical target volume, internal margin, and setup margin. The prescribed doses were 60 Gy (RBE) in four fractions for cases with peripheral metastatic tumor, 60 Gy (RBE) in 12 fractions for cases with metastatic tumor close to the gastrointestinal tract, and 64.8 Gy (RBE) in 12 fractions for cases with large metastatic tumor (>5 cm). The treatment aim was to cover 95% of the PTV with at least 95% of the prescribed dose. The dose constraints were as follows: Dose to 1 cm 3 (D 1cc ) <40 Gy (RBE) administered to the gastrointestinal tract in standard cases; D 1cc <45 Gy (RBE) administered to the gastrointestinal tract in the cases treated with 12 fractions; organ volume that received at least 10% of the administered dose (V 10 ) <55% and V 20 <40% administered to the liver; V 20 <20% administered to the lung; maximum dose (D max ) <30 Gy (RBE) administered to the spinal cord; D max <52.8 Gy (RBE) administered outside the PTV at the porta hepatis (including the first branch of the portal vein and hepatic duct); D max <45 Gy (RBE) administered to the skin in standard cases; and D max <50 Gy (RBE) administered to the skin in the cases treated with 12 fractions. Figures 1 and 2 show radiographical images before C-ion RT and typical clinical dose distribution with C-ion RT of patients with oligometastatic colorectal cancer in the liver and lung.
Evaluation during follow-up. Patients were followed-up for 1 month after the completion of C-ion RT and every 3 months thereafter. Follow-up examinations comprised routine testing of blood cell counts and chemistry and diagnostic imaging using CT, MRI, or FDG-PET. Acute and late toxicities were graded according to the Common Terminology Criteria for Adverse Events (version 4.0) of the National Cancer Institute (27). Acute and late toxicities were evaluated as the highest grade of toxicity that occurred within 3 months and after 3 months of initiating C-ion RT, respectively.
Dose-volume histogram analysis. Dose-volume histogram (DVH) analysis was used to evaluate the dose of C-ion RT to normal liver and normal lung (i.e. total liver or lung volume minus GTV). We assessed the mean liver or lung dose, and the percentage of the normal liver or lung that received at least 5, 10, 15, 20, 25, 30, 40, and 50 Gy (RBE) (V 5 , V 10 , V 15 , V 20 , V 25 , V 30 , V 40 , and V 50 ) based on the DVH.
Statistical analysis. All statistical analyses were performed using the Statistical Package for the Social Sciences software (version 25.0; IBM Inc., Armonk, NY, USA). Survival was measured from the date of C-ion RT initiation to death or the most recent followup. Local control (LC) was defined as no evidence of local recurrence without an increase in tumor size on CT or MRI and no increase in FDG uptake on PET. Progression-free survival (PFS) was defined as the absence of progression of both local and distant metastases. PFS was measured from the date of initiation of C-ion RT to the date of tumor progression or death from any cause. The probabilities of overall survival (OS), LC, and PFS rates were calculated using the Kaplan-Meier method. Next, we evaluated the potential prognostic effect of sex (male or female), age (<65 or ≥65 years), performance score (0 or 1), primary tumor site (rectum or colon), tumor location (liver or lung), tumor size (<3 or ≥3 cm), GTV volume (<10 cm 3 or ≥10 cm 3 ), clinical target volume (<40 or ≥40 cm 3 ), serum carcinoembryonic antigen level (<5.0 or ≥5.0 ng/ml), duration between surgery and C-ion RT (<27 or ≥27 months), and surgical indication (not indicated or refused) in OS and LC using the log-rank test.

Results
Patient characteristics. The clinical characteristics of the 19 patients are summarized in Table I. The median follow-up duration after the initiation of C-ion RT was 19 months (range=4-55 months). The median age at the time of registration for C-ion RT was 65 years (range=47-86 years). Four patients had two metastatic tumors that received C-ion RT. The number of oligometastatic tumors in the liver and lung was 14 in 11 patients and nine in eight patients, respectively. The median tumor size of 23 lesions was 2.6 cm (range=1.1-6.5 cm) in maximum diameter, in two cases larger than 5 cm. At the time of C-ion RT, six patients had chemoresistant disease or were unable to continue chemotherapy due to toxicities; 11 patients were unsuitable due to comorbidity and age, or refused chemotherapy. In patients who received chemotherapy before C-ion RT, the median duration from the initiation of chemotherapy for oligometastatic disease to the initiation of C-ion RT was 10 months (range=5-28 months). All patients with liver metastases had Child-Pugh class A and had no liver cirrhosis, and no patients with lung metastases had chronic lung disease or interstitial pneumonia. The dose-fractionation schedules for liver metastases were as follows: Nine lesions received 60 Gy (RBE) in four fractions, one lesion received 60 Gy (RBE) in 12 fractions, and four lesions received 64.8 Gy (RBE) in 12 fractions. For lung metastases, there were nine lesions, which received 60 Gy (RBE) in four fractions. Three patients did not achieve dose coverage of the 95% PTV with at least 95% of the prescribed dose as priority was given to the dose constraint of normal organs such as the gastrointestinal tract or skin.

. A 58-year-old male with oligometastic colorectal cancer in lung treated with carbon-ion radiotherapy. A: Plain computed tomography (CT) before treatment. Yellow arrow shows the tumor with contrast enhancement. B: 2-Deoxy-2-[ 18 F]fluoro-D-glucose (FDG) positron-emission tomography before treatment. Yellow arrow shows the tumor with abnormal FDG uptake. C: Dose distribution on axial CT images. The area within the red outline is the gross tumor volume. The 95% (red), 90% (orange), 80% (yellow), 65% (green), 50% (blue), and 20% (purple) isodose curves are highlighted [100% was 60 Gy (relative biological effectiveness)]. D: Plain CT 12 months after treatment. No recurrence or residual tumor evident (green arrow). E: FDG positron-emission tomography 12 months after treatment. FDG uptake was reduced compared to that before treatment (green arrow).
with local recurrence also developed distant or lymph node metastases. Twelve patients developed distant or lymph node metastases. Three patients died of colorectal cancer.
The observed cases of acute and late toxicities are shown in Table II. None of the patients developed grade 2 or higher acute and late toxicities. No patients developed radiationinduced liver disease (RILD) or Child-Pugh class decline. Table III lists DVH parameters. Analysis did not reveal any significant prognostic factors for OS and LC (Table IV).

Discussion
Surgery for oligometastatic colorectal cancer in the liver and lung is well established, and previous studies have reported 5-year survival of 30-61% (2-7). These results suggest that local treatment for oligometastatic colorectal cancer improves survival. Another local treatment approach is SBRT, which is indicated for patients who are unsuitable for surgery or refuse surgery. Previous reports of SBRT for oligometastatic colorectal cancer showed that the 2-year OS and LC were 57-75% and 36-91%, respectively, for patients with liver metastases and 68-89% and 58-80%, respectively, for those with lung metastases, with grade 3-4 toxicities developing in 0-9% of patients (8)(9)(10)(11)(12)(13)(14)(15). These results suggest that SBRT is an effective local treatment approach and might improve OS in oligometastatic colorectal cancer. In our study, C-ion RT was performed in patients with oligometastatic colorectal cancer who were unsuitable for surgery or refused surgery. The OS and LC rates were 100% and 61%, respectively, for those with liver metastases, and 100% and 83%, respectively, for those with lung metastases, with no grade 2 or higher toxicities. These outcomes are comparable to those in previous reports of SBRT. Additionally, we included patients with inoperable disease and those difficult to treat with chemotherapy; hence, local treatment with C-ion RT might improve OS, underscoring the value of C-ion RT.
We performed DVH analysis for the normal liver. A previous report investigating the relationship between normal liver DVH parameters and liver function after photon therapy found that a mean liver dose >23 Gy, V 5 >86%, V 10 >68%, V 15 >59%, V 20 >49%, V 25 >35%, V 30 >28%, and V40 >20% were risk factors of RILD (28). Another study on SBRT showed that V 25 >32% was a risk factor for Child-Pugh class decline in patients with hepatocellular carcinoma (29). In the present study, V 25 -V 40 DVH parameters of liver were exceeded in a few patients with multiple or tumors of 3 cm or larger. However, none of the patients developed RILD and Child-Pugh class decline. In previous studies, a risk factor of RILD and Child-Pugh class decline was liver cirrhosis in patients with hepatocellular carcinoma. However, we did not include patients with liver cirrhosis in our cohort, which explains the absence of RILD and Child-Pugh class decline in our study. The DVH analysis for normal lung in previous studies showed that risk factors for grade 2 or higher radiation-induced pneumonitis were a mean lung dose of >5 Gy and V 20 >7% in those treated with SBRT and V 30 >15% in C-ion those treated with RT (30,31). In the present study, V 30 did not exceeded 15% in any of the patients, and in only a few patients with multiple or large tumors (≥3 cm) did the mean Iung dose exceed >5 Gy and V 20 exceed 7%. However, none of the patients developed grade 2 or higher radiation-induced pneumonitis. These clinical results and DVH parameters suggest that the use of C-ion RT for oligometastatic colorectal cancer in patients with single 3cm or smaller tumors in liver and lung is a safe treatment approach, and in those with multiple tumors or 3 cm or larger tumor might be safe considering the liver or lung background. The improvement of LC might trigger more interest in Cion RT for oligometastatic colorectal cancer. In a previous report of C-ion RT for oligometastatic colorectal cancer in the liver, 3-year LC rates were 82% in those receiving single-fraction doses of 53 Gy (RBE) or higher and 28% in those receiving 48 Gy (RBE) or lower (p=0.01) (21). Additionally, 3-year LC was 86-93% after the administration of 70 Gy (RBE) in 16 fractions or higher in patients with pelvic recurrence of colorectal cancer (32,33). In the present study, all cases with local recurrence had central recurrence after administration of 60 Gy (RBE) in four fractions or 64.8 Gy (RBE) in 12 fractions. These results suggest a dose deficiency with the prescribed dose in our study. Higher dose irradiation such as 53 Gy (RBE) in a single fraction or 70 Gy (RBE) in 16 fractions without exceeding the tolerable dose with DVH in normal liver or lung might improve LC. In the present study, we considered that increasing the prescribed dose might be safe for small and single tumors.
Research on particle therapy, including proton beam therapy and C-ion RT, for oligometastatic colorectal cancer in the liver or lung is limited (21,34,35). All reports showed favorable and similar results to those using SBRT. Our ANTICANCER RESEARCH 41: 1997-2005(2021 2002  findings are also favorable and comparable to the SBRT results and previous particle therapy. Our study had several limitations. Firstly, this was a single-institutional retrospective analysis with a small number of patients and a short follow-up duration. Secondly, the patient backgrounds were heterogeneous. Thirdly, analyses of DVH and toxicities for C-ion RT were few, and the threshold risk value in the incidence of toxicities is unknown. Despite these limitations, this study confirmed the safety of C-ion RT. Our study offers useful information on the treatment of oligometastatic colorectal cancer, especially in patients who are unsuitable for surgery. In conclusion, C-ion RT is a favorable treatment approach for oligometastatic colorectal cancer in the liver and lungs V 5 -V 50 : The percentage of the liver or lung volume that received at least 5-50 Gy (relative biological effectiveness). and has comparable clinical outcomes to SBRT. Additionally, C-ion RT might be a safe treatment option, as exemplified by the absence of RILD and radiation-induced pneumonitis. Therefore, C-ion RT can be used in patients with oligometastatic colorectal cancer in the liver and lungs who are unsuitable for surgery.

Conflicts of Interest
The Authors declare no conflicts of interest in regard to this study.