The Japanese nationwide cohort data of proton beam therapy for liver oligometastasis in breast cancer patients

Abstract

A nationwide multicenter cohort study on particle therapy was launched by the Japanese Society for Radiation Oncology in Japan in May 2016. We analyzed the outcome of proton beam therapy (PBT) for liver oligometastasis in breast cancers. Cases in which PBT was performed at all Japanese proton therapy facilities between May 2016 and February 2019 were enrolled. The patients were selected based on the following criteria: the primary cancer was controlled, liver recurrence without extrahepatic tumors and no more than three liver lesions. Fourteen females, with a median age of 57 years (range, 44–73) and 22 lesions, were included. The median lesion size, fraction (fr) size and biological effective dose were 44 (20–130) mm, 6.6 (2–8) gray (Gy) (relative biological effectiveness)/fr and 109.6 (52.7–115.2) Gy, respectively. The median follow-up period was 22.8 (4–54) months. The 1-, 2- and 3-year local control (LC) rates of liver metastasis from breast cancer were 100% for all. The 1-, 2- and 3-year overall survival rates were 85.7, 62.5 and 62.5%, respectively. The 1-, 2- and 3-year progression-free survival (PFS) rates were 50.0%, 33.3%, and 16.7%, respectively. The median PFS time was 16 months. Only one patient did not complete PBT due to current disease progression. One patient had Grade 3 radiation-induced dermatitis. None of the patients experienced radiation-induced liver failure during the acute or late phase. Owing to the low incidence of adverse events and the high LC rate, PBT appears to be a feasible option for liver oligometastasis in breast cancers.

INTRODUCTION

Approximately, 50% of all patients with metastatic breast cancer have liver metastasis, and 5–12% of breast cancer recurrence patients develop liver metastasis as the main recurrence site [1–5]. Breast cancer patients with liver metastasis have a poor prognosis of 4–8 months if they have no treatment and a prognosis of 18–24 months even with systemic therapies [1, 2, 6]. Most metastatic patients may have systemic disease; therefore, only a limited number of patients are candidates for local treatment. To improve the treatment outcome, various local treatments, such as surgery, radio frequency ablation (RFA), transarterial chemoembolization (TACE) and radio therapy, including stereotactic body radiotherapy (SBRT) and particle therapy (PT), have been applied, in combination with chemotherapy, or performed alternative to chemotherapy. Many cases are inoperable, or the patients do not wish to undergo surgery due to the underlying disease, age or other conditions. If surgery is not possible, radiotherapy is one alternative treatment.

PT, such as proton beam therapy (PBT) and carbon-ion beam therapy (CIBT), has been reported to have excellent therapeutic effects on liver tumors due to the physical properties of the Bragg peak [7–9]. However, the clinical benefit of oligometastatic liver tumors is still controversial. There is also little evidence on the use of PBT for liver metastatic tumors [10–13]. The number of cases of liver metastasis at each institute is not large, and the number of institutions offering PBT is also small. Therefore, there are few studies that can provide sufficient evidence for the treatment of liver metastasis. There are no reports of CIBT, and there are only two reports of PBT on liver metastasis from breast cancer [14, 15].

In Japan, a nationwide multicenter cohort study on PT was started in May 2016 for all facilities that performed PBT and/or CIBT. Using these data, we analyzed the data on PBT for liver metastasis of breast cancer. The present study was performed and managed by the Oligometastatic Cancer Working Group in the Particle Beam Therapy Committee and the Subcommittee at the Japanese Society for Radiation Oncology (JASTRO).

MATERIALS AND METHODS

Study design and patient enrollment

This was a nationwide multi-institutional cohort study. The study population included patients for whom PBT was performed at all Japanese PBT facilities between May 2016 and February 2019. These criteria almost all corresponded to the requirements to provide PT for oligometastatic diseases in the JASTRO system as follows:

(i) Analysis on the Japanese multi-institutional cohort study dataset of PT. Among patients registered in the Japanese multi-institutional cohort study, the dataset of patients who received PBT between May 2016 and Feb 2019 was reviewed to evaluate;

(ii) Number of metastatic liver tumors ≤3, with breast cancer.

(iii) Absence of recurrence in primary disease site after primary curative treatment.

(iv) Absence or control of the other cancers and clinically detectable recurrent or metastatic diseases other than the metastatic regions.

(v) Delivery of PBT to all metastatic regions with curative intent.

Procedure of PBT

The treatment indication and strategy were discussed at the cancer boards in each institution. The dose and fraction (fr) of PBT were determined by referring to the unified treatment policy established by JASTRO. We used the following protocol: 64 gray (Gy) (relative biological effectiveness [RBE])/8 fr, for the hepatic periphery area away from the GI tract, and 72.6 Gy (RBE)/22 fr for the adjacent hilar region type. Regarding safety, the HCC irradiation protocols of 66 Gy (RBE)/10 fr, 72.6–76 Gy (RBE)/20–22 fr and 74–76 Gy (RBE)/37–38 fr could be used.

Outcomes

The local control (LC), overall survival (OS), progression-free survival (PFS) and adverse events (AEs) were examined. LC was defined as the date of the event when the response evaluation criteria in solid tumor evaluation resulted in progressive disease (PD) from PBT initiation. OS was defined as the duration from the date of PBT initiation to death from any cause. PFS was defined as the duration from the date of PBT initiation to the recurrence or death from any cause death. We evaluated AEs using the Common Terminology Criteria for Adverse Events version 5.0.

Statistical analysis

The LC, OS and PFS rates were calculated using the Kaplan–Meier method. Factors possibly related to OS, such as tumor size (for multiple lesions, the largest size), number of liver tumors, intervention of chemotherapy and/or hormone therapy, were investigated. The cut-off values were estimated using the receiver operating characteristic curve and area under the curve. The final cut-off value was selected as the point where the sum of sensitivity and specificity was maximized. Univariate analysis was performed using the log-rank test. P-values of <0.05 were considered to be statistically significant. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R d4 (R Foundation for Statistical Computing, Vienna, Austria). EZR is a modified version of the R commander designed to add statistical functions commonly used in biostatistics [16].

RESULTS

Characteristics of patients and metastatic liver tumors

Fourteen females, with a median age of 57 years (range, 44–73) and 22 lesions, were included. Nine patients had one lesion, two patients had two lesions and three patients had three lesions. All metastatic tumors were diagnosed as liver oligometastatic recurrence after the primary cancer was controlled. PBT was selected because 13 patients had unresectable tumors, and 1 patient’s data were missing. Tumors in 9 of the 13 patients were unresectable because of the underlying disease, and those in the remaining 4 patients were unresectable because of patients’ general conditions (age, etc.). The median lesion size, fr size and biological effective dose (BED) using the linear quadratic model with α/β = 10 Gy ((BED)₁₀) were 44 (20–130) mm, 6.6 (2–8) Gy (RBE)/fr and 109.6 (52.7–115.2) Gy, respectively. The characteristics of patients and metastatic liver tumors are shown in Table 1. Precedence chemotherapy and/or hormone therapy does not mean adjuvant standard systemic therapy. It means systemic therapy for liver metastatic recurrence before PBT. Of the 11 patients who received precedent chemotherapy, 2 had stable disease, 4 had PD, 3 had inevaluable disease and 2 were unknown. Three patients did not receive any chemotherapy after the recurrence of liver metastases.

Table 1.

Characteristics of patients and metastatic liver tumors in each patient

Case	Sex	Age	PSa	No. of lesions	Largest tumor size (mm)	Hepatectomy	Chemotherapy			Total dose (Gy[RBE]/fr)	AEs worst grade	Survival time (months)	Event
							Precedent	Concurrent	Post
1	F	44	0	2	34	NA	P	A	P	66/10	0	17.1	Deceased
2	F	51	0	1	52	Unresectable	P	A	P	66/10	0	53.9	Alive
3	F	73	0	1	35	Unresectable	P	A	P	66/10	0	50.5	Alive
4	F	54	0	1	32	Unresectable	P	P	A	64/8	3	21.5	Deceased
5	F	70	0	1	56	Unresectable	P	A	P	64/8	0	44.1	Alive
6	F	52	2	3	130	Unresectable	P	A	A	68/20	0	5.9	Deceased
7	F	51	1	2	90	Unresectable	A	A	A	72.6/22	0	4.3	Deceased
8	F	54	0	1	26	Unresectable	A	A	A	64/8	0	27.2	Alive
9	F	66	0	1	64	Unresectable	P	A	A	66/10	0	31.4	Alive
10	F	70	1	3	30	Unresectable	P	P	A	72.6/22	0	17.0	Deceased
11	F	60	1	1	71	Unresectable	P	A	P	72.6/22	0	21.0	Alive
12	F	48	1	1	110	Unresectable	P	P	A	39.6/12	0	24.0	Alive
13	F	63	0	3	30	Unresectable	A	A	A	74/37	0	17.8	Alive
14	F	61	1	1	20	Unresectable	P	A	A	66/10	0	24.8	Alive

Abbreviations: P = presence, A = absence, NA = not available.

^aAccording to the Eastern Cooperative Oncology Group.

Survival outcomes

The median follow-up period of breast cancers was 22.8 (4–54) months. The 1-, 2- and 3-year LC rates of liver metastasis from breast cancers were 100% for all, as shown in Fig. 1a. The 1-, 2- and 3-year OS rates were 85.7, 62.5 and 62.5%, respectively (Fig. 1b). The 1-, 2- and 3-year PFS rates were 50.0, 33.3 and 16.7%, respectively, and the median PFS time was 16 months (Fig. 1c).

Fig. 1.

Kaplan–Meier plot of estimated (a) LC, (b) OS and (c) PFS rates.

OS-related factors

Age, PS, tumor size, the number of liver tumors and the intervention of chemotherapy and/or hormone therapy at any time were not significantly related to the OS. The results are summarized in Table 2.

Table 2.

Results of univariate analysis on OS

Factor	Median	Range	Cut-off value	AUC	95% CI	Group	n	Median OS (months)	P-value
Age	57	44–70	54	0.7	0.372–1	<54	5	17.1	0.121
						≥54	9	NA
PS	0	0–2	1	0.667	0.364–0.97	<1	8	17.1	0.202
						≥1	6	4.3
Tumor size	44 mm	20–130	90	0.57	0.201–0.932	<90	11	NA	0.0768
						≥90	3	5.9
Number of tumors	1	1–3	2	0.82	0.581–1	<3	11	NA	0.085
						≥3	3	17.1
Precedent						Presence	9	NA	0.488
Chemotherapy and/or hormone therapy						Absence	5	NA
Concurrent						Presence	3	21.53	0.349
Chemotherapy and/or hormone therapy						Absence	11	NA
Post						Presence	5	17.1	0.365
Chemotherapy and/or hormone therapy						Absence	9	4.27

PS = performance status, AUC = area under the curve, CI = confidence interval.

Safety outcomes

Only one patient did not complete PBT due to current disease progression. The other patients completed PBT without interruption. One patient had Grade 3 radiation-induced dermatitis. The other patients had no Grade 3 or higher AEs. None of the patients experienced radiation-induced liver failure during the acute or late phase.

DISCUSSION

Analysis of 439 patients of postoperative recurrence found that median survival times (MSTs) of patients with and without liver metastases were 12 and 26 months, respectively. Patients with liver metastases have a poor prognosis (P < 0.001) [2], and liver metastasis of breast cancer is a fatal and incurable disease.

Reports on the local treatment of patients with liver metastasis of breast cancer are not sufficient. In surgical outcomes, the 3- and 5-year OS rates were 49–75 and 41–61%, respectively, with a MST of 42–64 months [4, 17–20]. In TACE outcomes, the 1- and 3-year OS rates were 63–76 and 13–48%, respectively, with a MST of 18.5–31 months [21–23]. In RFA outcomes, the 1- and 3-year OS rates were 68–87 and 25.3–49.3%, respectively, with a MST of 26–33.5 months [24–27] There are few papers reporting treatment of SBRT for breast cancer liver metastases [28–30] However, one of the studies was also mixed on cases of lung metastases, and the results for patients with liver metastases simply are not clear [28] Regarding SBRT outcomes, the 1- and 2-year OS rates were 85 and 57%, respectively [29] Regarding PBT outcomes, the 1- and 3-year OS rates were 88–94.1 and 71.7–73%, respectively, with a MST of 39.3 months [14, 15]. The summary of clinical results of local treatment for liver metastasis in breast cancer patients is shown in Table 3. A study of prognostic factors predicting survival from first recurrence in patients with metastatic breast cancer reported that the (i) size of primary tumor, (ii) axillary lymph node status, (iii) hormonal receptor status, (iv) adjuvant radiotherapy, (v) disease-free interval and (vi) number of recurrence site were prognostic factors [2].

Table 3.

Summary of clinical results of local treatment for liver metastasis of breast cancer patients

Author	Country	Year	Treatment	Number of patients	Number of tumors	Median tumor size (cm)	OS rate (%)			MST (months)	AE grade ≥ 3
							1-year	2-year	3-year
Porcard [18]	France	2000	Surgery	49	≤3	3.8	86	79	65	42	NA
Li [19]	China	2005	TACE	28	S/M	2.8	63	30	13	28	0%
Vogl [20]	Germany	2009	TACE	208	≤5	NA	69	40	33	18.5	NA
Duan [21]	China	2011	TACE+C	44	S/M	4.6	76.2	66.7	47.6	31	NA
Meloni [22]	Italy	2009	RFA	52	≤5	2.5	68	NA	43	29.9	0%
Kümler [23]	Denmark	2015	RFA	32	≤6	2.0	87	68	48	33.5	3.1%
Bai [24]	China	2019	RFA	69	S/M	2.9	81.8	50.1	25.3	26	1.1%
Schullian [25]	Austria	2021	RFA	42	S/M	3.0	84.1	NA	49.3	32.3	12%
Onal [27]	Turkey	2018	SBRT	22	≤ 5	2.1	85	57	NA	NA	4.5%
Mahadevan [28]	USA	2018	SBRT	42	S/M	40 cm3	66.4	NA	NA	21	0%
Fukumitsu [10]	Japan	2017	PBT	8	S/M	4	88	NA	73	NA	NA
Kim [11]	Korea	2021	PBT	17	≤2	2.4	94.1	88.2	71.7	39.3	0%
Current study	Japan	2023	PBT	14	≤3	4.4	85.7	62.5	62.5	NA	7%

Abbreviations: S = solitary, M = multifocal, TACE = transarterial chemoembolization.

Many studies have reported that the indications for local treatment (TACE, RFA and SBRT) are metastatic liver tumors less than 5 or 6 cm from the viewpoint of safety of treatment [22, 24–26, 28, 29]. One strength of PBT is its ability to treat larger tumors than other local treatments. The present study included seven cases of tumors larger than 5 cm, and the largest tumor was 13 cm. Only one patient suffered from G3 dermatitis, where a 3.2 cm tumor was located near the skin. PBT can be used to treat significantly larger tumors without severe adverse effects. In a past study of PBT for patients with large hepatocellular carcinoma, 22 patients whose tumors were larger than 10 cm were treated with PBT without the late treatment-related toxicity of Grade 3 or higher [31].

In a past report, liver metastases occurred in 47 out of 912 breast cancer patients, an incidence of 5.2%. Ten of 47 patients (22.7%) presented with liver metastases only; 11 patients (25%) had synchronously locoregional recurrence and/or extrahepatic metastases and 23 patients (52%) developed locoregional recurrence and/or extrahepatic metastases following liver metastases [1] Liver metastases of breast cancer patients without extrahepatic tumors were only 1.1% among all patients having postoperative recurrence [1]. In breast cancer, tumor cells spread to the liver via systemic circulation from the primary breast local site; thus, isolated liver metastasis without extrahepatic lesion is rare.

Systemic chemotherapy is the most often used treatment for breast cancer metastases; however, local treatment may be significant if the case is appropriately selected. Systemic chemotherapy regimens with new molecular-targeted agents have been developed, but the survival of liver metastasis of breast cancer is a median of 3–15 months [32–34] Although these prognostic factors, as shown in a previous report [2], have not been examined and cannot be definitively determined, a comparative study of the effect of systemic chemotherapy with or without TACE as local treatment for liver metastases of breast cancer after mastectomy showed that the group that received TACE had a longer survival time (P = 0.027) [23]. Therefore, the addition of local treatment to chemotherapy is considered to be significant.

Although the present study showed a high LC rate, the PFS rate deviates considerably from and is inferior to the LC rate. This is due to the appearance of new distant metastatic lesions, some of which may appear in the liver after initial PBT. This situation can occur with any local treatment. In surgery reports, the additional metastasis rates (recurrence of remaining liver and other metastatic site) after surgery were 26.6% at 1 year, 54.8% at 2 years and 63.8% at 3 years [20] Liver function is considerably impaired after hepatectomy, sometimes resulting in adhesions, making repeat hepatectomy difficult. On the other hand, PBT can be used for repeated liver treatment, as evidenced by some past reports [11, 12, 35].

The present study has several limitations. First, this was not a strictly prospective study, and treatment methods were not necessarily consistent across facilities. In order to eliminate these biases, we standardized the treatment protocols, such as dose fractionation, at all PT facilities in Japan. Furthermore, we have tried to minimize inter-institutional bias by enrolling patients who had been discussed by the cancer board at each facility; each board included physicians and surgeons. Second, the sample size was small. As previous reports have shown, liver metastasis only without extrahepatic lesions is extremely rare, and it is difficult for a single institution to accumulate a fixed number of cases for these diseases. We have been collaborating with nationwide facilities to register and analyze the data. In the present study, we tried to increase the number of cases. Third, since this cohort consisted of patients from all PBT facilities in Japan, it was necessary to limit the information to be collected to the minimum. This situation allowed us to enlarge the number of cases enrolled and had the advantage of being able to analyze many cases at once. Molecular subtype is very important information for the prognosis of breast cancer patients. However, there were insufficient data regarding subtype in the current study. Therefore, it was difficult to discuss whether the PBT contributed to the prolonged life expectancy compared to standard chemotherapy. Furthermore, because of the short observation period, there is room for reconsideration of OS and LC data. The above Japanese nationwide cohort survey has been conducted since 2016, and we will expand the observation period. Our goal is public medical insurance coverage on liver metastatic tumors. We plan to accumulate the long-term follow-up data on more cases in the future. As the number of cases and observation period increase, we plan to add multifaceted analyses, including comparisons of the size of the primary tumor, axillary lymph node status, hormonal receptor status, disease-free interval and treatment prior to and after PBT.

We analyzed the outcome of PBT for hepatic oligometastases of breast cancer using our nationwide multi-institutional cohort study. This cohort study showed a satisfactory LC rate for liver oligometastasis of breast cancer patients. There are few reports of PBT for liver metastases in breast cancer patients and we think the current data are useful.

Although chemotherapy is the standard therapy, most patients who undergo PBT are refractory, physically intolerant or those who refuse to undergo chemotherapy. Breast cancer patients who undergo PBT have a better prognosis than those who do not undergo any treatment. However, it remains unclear as to whether PBT is a better option than chemotherapy for the treatment of breast cancer in patients with oligometastasis because the results of the present study do not clearly indicate an improved prognosis. Therefore, further investigation into the effectiveness of PBT compared to chemotherapy alone is necessary.

In conclusion, based on the low incidence of adverse events (AEs) and the high LC rate, PBT appears to be a feasible option when local treatment is necessary for liver oligometastasis in breast cancer patients.

CONFLICT OF INTEREST

None.

FUNDING STATEMENT

This work was supported by AMED under Grant Number JP16lm0103004.

AUTHOR CONTRIBUTIONS

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by all authors. The first draft of the manuscript was written by H.Y., and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

PRESENTATION AT A CONFERENCE

This manuscript has not been published or presented elsewhere in part or in entirety and is not under consideration by another journal.

CLINICAL TRIAL REGISTRATION NUMBER

The study was registered and managed by the University Hospital Medical Information (UMIN). In addition to the UMIN clinical trial registry system, the UMIN case data repository system was implemented (approved number UMIN000022917). All study participants provided informed consent or did not withdraw, and the study design was approved by the appropriate ethics review board (approved number 016-0106).

DATA AVAILABILITY

The datasets generated during and/or analyzed during the current study are not publicly available due to including patients’ personal information but are available from the corresponding author on reasonable request.

ETHICS APPROVAL

All procedures involving human participants performed in this study followed the ethical standards of the institutional and national research committees as well as the 1964 Declaration of Helsinki and its subsequent amendments. The study was approved by the ethics committee of our institution (approval number 016-0106) and was registered and managed by the UMIN. In addition to the UMIN clinical trial registry system, a UMIN case data repository system was implemented (approval number UMIN000022917).

CONSENT TO PARTICIPATE

Informed consent was obtained from all individual participants included in the study.