For prognostic assessment in women who receive radiotherapy (RT) for bone metastases (BMs) from breast cancer (BC), prognostic factors specific for BMs from BC were investigated in the present study. The prognostic assessment was performed by retrospectively reviewing 143 women who received first-time RT for BMs from BC between January 2007 and June 2018. The median follow-up time and median overall survival (OS) time from the first-time RT for BMs were 22 and 18 months, respectively. In the multivariate analysis, nuclear grade 3 (NG 3) [hazard ratio, 2.18; 95% confidence interval (CI), 1.34-3.53], brain metastases (hazard ratio, 1.96; 95% CI, 1.01-3.81), liver metastases (hazard ratio, 1.75; 95% CI, 1.17-2.63), performance status (PS) (hazard ratio, 1.63; 95% CI, 1.10-2.41) and previous systemic therapy (hazard ratio, 1.58; 95% CI, 1.03-2.42) were significant factors for OS, whereas age, hormone-receptor/human epidermal growth factor receptor 2 status, number of BMs and synchronous lung metastases were not significant factors. When points according to risk levels [unfavorable points (UFPs)] were assigned to each risk factor (1.5 points for NG 3 and brain metastases; and 1 point for PS ≥2, previous systemic therapy and liver metastases), the median OS times of patients with a total number of UFPs ≤1 (n=45), 1.5–3 (n=55) and ≥3.5 (n=43) were 36, 17 and 6 months, respectively. Overall, in patients who received first-time RT for BMs from BC, NG 3, brain/liver metastases, poor PS and previous systemic therapy were unfavorable prognostic factors. Comprehensive prognostic assessment using these factors seemed to be useful for the prediction of prognoses in patients with BMs from BC.
The bones are the primary metastatic site in 46% of patients with breast cancer (BC), and bone metastases (BMs) have been reported to occur in 71% of metastatic BC cases (
Svensson
All procedures performed in the present study were in accordance with the ethical standards of the Institutional and/or National Research Committee and the 1964 Helsinki Declaration and its later amendments. Informed consent for the use of clinical data was obtained by opt-out methods. This retrospective study was approved by the Ethics Committee of the National Hospital Organization Shikoku Cancer Center (Matsuyama, Japan; approval no. RIN2021-71).
The cases of 167 consecutive female patients who received first-time RT for BMs from BC between January 2007 and June 2018 in the National Hospital Organization Shikoku Cancer Center were retrospectively reviewed. A total of 24 patients were excluded for the following reasons: i) No computed tomography (CT) of the chest and abdomen within 3 months of beginning RT (n=14); ii) discontinuation of RT (n=2); iii) synchronous and/or sequential double cancer (n=5); and iv) follow-up duration of <6 months despite survival (n=3). Thus, 143 patients were included for analysis. Among the 143 patients, 117 patients (82%) had died and 26 patients (18%) were alive at the last follow-up day of clinical examination.
BMs were detected with CT, bone scintigraphy and/or 18F fluorodeoxyglucose positron-emission tomography/CT (FDG-PET/CT). Visceral metastases were evaluated with CT and/or FDG-PET/CT within 3 months of the beginning of first-time RT to BMs. Patients who had not undergone contrast-enhanced magnetic resonance imaging (MRI) were classified in a no brain metastases cohort. Performance status (PS) was evaluated according to the Eastern Cooperative Oncology Group scale (
Patients received three-dimensional conformal RT. RT was delivered using 4- or 10-MV photons with a linear accelerator (Clinac 21-EX; Varian Medical Systems, Inc.).
The primary endpoint was overall survival (OS) time, which was defined as the time from the beginning of the first-time RT for BMs to death. OS was calculated using the Kaplan-Meier method, and statistical differences in OS were evaluated by the log-rank test. The Cox proportional hazard model was used for univariate and multivariate analyses. Factors, including age, PS, estrogen receptor status, progesterone receptor status, HER2 overexpression, triple-negative BC type, nuclear grade (NG) (
A total of 128 patients (90%) had multiple BMs (only vertebral, 7 patients; only non-vertebral, 5 patients; both vertebral and non-vertebral, 116 patients) and 15 patients (10%) had a single bone metastasis (vertebral, 11 patients; non-vertebral, 4 patients). Overall, 98 patients (69%) had undergone previous systemic therapy, including chemotherapy, hormone therapy and/or anti-HER2 agent therapy for >3 months before RT. A total of 58 patients (41%) underwent FDG-PET/CT before the RT. In addition, 39 patients (27%) underwent contrast-enhanced T1-weighted MRI (CE-MRI) within 3 months of the beginning of the RT or during RT, and brain metastases were detected in 14 patients (10%).
The majority of patients received 30 Gy in 10 fractions. Overall, 130 patients (91%) received at least 30 Gy in 10 fractions and 13 patients (9%) received hypofractionated low-dose RT. A total of 108 patients (76%) received RT to the vertebral bone and 29 patients (20%) received RT to the pelvic bone. Patient characteristics are listed in
The median follow-up duration of survival was 22 months (range, 7–117 months) from first-time RT for BMs. The median OS time was 17.5 months. The OS curve is shown in
Upon univariate analysis, PS ≥2 (P=0.021), NG 3 (P<0.001), previous systemic therapy (P=0.003), synchronous brain metastases (P<0.001), synchronous liver metastases (P<0.001) and synchronous lung metastases (P=0.006) were significantly associated with poor OS after first-time RT for BMs (
The 1-year OS rate was as follows: PS 0–1 vs. PS 2–4, 76.8 vs. 45.5%; NG 1–2 vs. NG 3, 74.9 vs. 43.5%; previous chemotherapy yes vs. no, 56.6 vs. 73.3%; synchronous brain metastases yes vs. no, 23.1 vs. 65.8%; synchronous liver metastases yes vs. no, 46.5 vs. 71.4%; and synchronous lung metastases yes vs. no, 48.4 vs. 70.1% (
Upon multivariate analysis, the significant prognostic factors after first-time RT for BMs were NG 3 [hazard ratio, 2.18; 95% confidence interval (CI), 1.34-3.53; P=0.002], synchronous brain metastases (hazard ratio, 1.96; 95% CI, 1.01-3.81; P=0.046), synchronous liver metastases (hazard ratio, 1.75; 95% CI, 1.17-2.63; P=0.006), PS (hazard ratio, 1.63; 95% CI, 1.10-2.41; P=0.016) and previous systemic therapy (hazard ratio, 1.58; 95% CI, 1.03-2.42; P=0.038). Age, the timing of the appearance of metastases, hormone-receptor status, HER2 status, number of bone metastases (single vs. multiple) and synchronous lung metastases were not significant factors.
A comprehensive prognostic assessment using regression coefficients of significant prognostic factors in multivariate analysis (
For individualization of RT to BMs from BC, prognostic factors for patients who received first-time RT to BMs from BC were investigated in the present study. Based on the multivariate analysis, NG 3, synchronous brain metastases, synchronous liver metastases, PS ≥2 and previous systemic therapy were the statistically significant unfavorable prognostic factors for OS in patients who received first-time RT for BMs from BC. By contrast, hormone receptor status, HER2 status, the timing of appearance of metastases (relapse vs.
In clinical practice, RT with variable dose-fractionation schedules (e.g. single fraction RT of 8 Gy, traditional RT with 30 Gy in 10 fractions and stereotactic body RT) is administered to BMs. According to a meta-analysis, single-fraction RT was not inferior to multiple-fraction RT regarding pain relief from BMs (
Histological grade or NG has been one of the most often used major prognostic factors for operable BC, along with lymph node metastases and tumor size (
Synchronous metastases in other organs also affect the prognoses of patients with BMs from BC. According to previous reports regarding patients with BMs from BC, patients who have synchronous metastases in other organs showed shorter median OS times (median OS time, 9–17 months) compared with patients with bone-only metastases (median OS time, 31–33 months) (
Based on the results from multivariate analysis, previous systemic therapy was also a significantly unfavorable prognostic factor in the present study. Previous systemic therapy has the potential to induce resistance to systemic therapy and is likely to lead to an unfavorable prognosis.
In the present study, hormone-receptor status and HER2 status were not significant factors. There was a possibility that the hormone-receptor status and HER2 status at the initial presentation had changed after previous systemic therapy. Features of tumor cells in metastatic lesions sometimes differ from those in primary lesions (
Comprehensive prognostic assessment using the designated UFPs seemed to be useful for the prediction of the prognoses of patients with BMs from BC. The median survival time of the favorable group (total UFPs ≤1) was >3 years. Well-fractionated higher-dose RTs are likely to be suitable for these patients. By contrast, single fraction RT of 8 Gy is suitable for the unfavorable group (total UFPs ≥3.5). Prognostic assessment using total UFPs was helpful to determine the appropriate dose-fractionation schedules for patients with BMs from BC.
The present study has several limitations, including its retrospective nature. First, due to the lack of laboratory data in numerous cases, the prognostic assessment using UFPs cannot be compared to Katagiri's prognostic scoring system (
In conclusion, PS ≥2, NG 3, previous systemic therapy, synchronous brain metastases and synchronous liver metastases were significantly associated with poor OS time for patients with BMs from BC. A comprehensive prognostic assessment using UFPs based on these factors seemed to be useful to select patients who needed comparatively well-fractionated high-dose RT for BMs from BC.
Not applicable.
The datasets used and/or analyzed during the present study are available from the corresponding author on reasonable request.
KM, HK and YH designed the study concepts. KM, HK, YH, KN, MK, SO and TK collected patient data, and analyzed and interpreted the data. KM, HK and YH confirm the authenticity of all the raw data. KM, HK, YH, KN, MK, SO and TK collaborated in the discussion. KM, HK and YH prepared the manuscript, and KN, MK, SH and TK revised it critically for important intellectual content. All authors read and approved the final manuscript.
The present study was approved by the Ethics Committee of the National Hospital Organization Shikoku Cancer Center (Matsuyama, Japan; approval no. RIN2021-71). The opt-out method was applied with regard to consent for this study.
Not applicable.
SO received honorarium from AstraZeneca plc. All other authors declare that they have no competing interests.
Kaplan-Meier curve of overall survival rate in patients who received radiotherapy for bone metastases from breast cancer.
Kaplan-Meier curves of overall survival rates according to total UFPs in patients who received radiotherapy for bone metastases from breast cancer. The favorable group (total UFPs ≤1), the intermediate group (total UFPs=1.5-3) and the unfavorable group (total UFPs ≥3.5) are shown by the solid, dashed, and dotted lines, respectively. UFPs, unfavorable points.
Patient characteristics.
Characteristic | Value |
---|---|
Median age (range), years | 61 (33–88) |
PS, n (%) | |
0 | 9 (6.3) |
1 | 66 (46.2) |
2 | 35 (24.5) |
3 | 19 (13.3) |
4 | 12 (8.4) |
Unknown | 2 (1.4) |
Timing of appearance of metastases, n (%) | |
Relapse | 97 (67.8) |
|
46 (32.2) |
ER, n (%) | |
Positive | 117 (81.8) |
Negative | 19 (13.3) |
Unknown | 7 (4.9) |
PgR, n (%) | |
Positive | 95 (66.4) |
Negative | 41 (28.7) |
Unknown | 7 (4.9) |
HER2 overexpression, n (%) | |
Yes | 20 (14.0) |
No | 110 (76.9) |
Unknown | 13 (9.1) |
TN type, n (%) | |
Yes | 9 (6.3) |
No | 124 (86.7) |
Unknown | 10 (7.0) |
NG, n (%) | |
1 | 20 (14.0) |
2 | 24 (16.8) |
3 | 56 (39.2) |
Unknown | 43 (30.1) |
Previous systemic therapy, n (%) | |
Yes | 98 (68.5) |
No | 45 (31.5) |
Number of bone metastases, n (%) | |
Single | 15 (10.5) |
Multiple | 128 (89.5) |
RT sites, n (%) | |
Vertebral | 108 (75.5) |
Pelvis | 29 (20.3) |
Other | 6 (4.2) |
Brain metastases | |
Yes | 14 (9.8) |
No | 25 (17.5) |
Not examined | 104 (72.8) |
Liver metastases | |
Yes | 55 (38.5) |
No | 88 (61.5) |
Lung metastases | |
Yes | 56 (39.2) |
No | 87 (60.8) |
PS, Eastern Cooperative Oncology Group performance status; ER, estrogen receptor; PgR, progesterone receptor; HER2, human epidermal growth factor receptor 2; TN, triple-negative; NG, nuclear grade; RT, radiotherapy.
Univariate analysis for overall survival.
Characteristic | 1-year OS rate, % | P-value | HR (95% CI) |
---|---|---|---|
Age, years (<60 vs. ≥61) | 62.4 vs. 61.4 | 0.943 | 1.01 (0.70-1.46) |
PS (0, 1 vs. ≥2) | 76.8 vs. 45.5 | 0.021 | 1.55 (1.07-2.24) |
Timing of appearance of metastases (relapse vs. |
59.2 vs. 67.4 | 0.182 | 0.77 (0.52-1.13) |
ER (positive vs. negative) | 63.8 vs. 53.1 | 0.141 | 1.49 (0.88-2.55) |
PgR (positive vs. negative) | 62.0 vs. 61.9 | 0.630 | 1.11 (0.73-1.68) |
HER2 overexpression (positive vs. negative) | 65.0 vs. 61.4 | 0.441 | 1.24 (0.72-2.15) |
TN type (yes vs. no) | 40.0 vs. 63.6 | 0.311 | 0.68 (0.33-1.40) |
NG (1, 2 vs. 3) | 74.9 vs. 43.5 | <0.001 | 2.30 (1.46-3.60) |
Previous systemic therapy (no vs. yes) | 73.3 vs. 56.6 | 0.003 | 1.84 (1.22-2.76) |
Number of bone metastases (single vs. multiple | 60.0 vs. 62.1 | 0.774 | 1.10 (0.59-2.05) |
RT sites (vertebral vs. non-vertebral) | 64.6 vs. 53.3 | 0.093 | 1.42 (0.94-2.14) |
Brain metastases (no vs. yes) | 65.8 vs. 23.1 | <0.001 | 3.06 (1.67-5.62) |
Liver metastases (no vs. yes) | 71.4 vs. 46.5 | <0.001 | 2.20 (1.51-3.20) |
Lung metastases (no vs. yes) | 70.1 vs. 48.4 | 0.006 | 1.73 (1.17-2.54) |
OS, overall survival; HR, hazard ratio; CI, confidence interval; PS, Eastern Cooperative Oncology Group performance status; ER, estrogen receptor; PgR, progesterone receptor; HER2, human epidermal growth factor receptor 2; TN, triple-negative; NG, nuclear grade; RT, radiotherapy.
Multivariate analysis for overall survival.
Characteristic | Regression coefficient | P-value | HR (95% CI) |
---|---|---|---|
PS (≥2 vs. 0–1) | 0.4860 | 0.016 | 1.63 (1.10–2.41) |
NG (3 vs. 1–2) | 0.7776 | 0.002 | 2.18 (1.34–3.53) |
Previous systemic therapy (yes vs. no) | 0.4543 | 0.038 | 1.58 (1.03–2.42) |
Brain metastases (yes vs. no) | 0.6747 | 0.046 | 1.96 (1.01–3.81) |
Liver metastases (yes vs. no) | 0.5621 | 0.006 | 1.75 (1.17–2.63) |
Lung metastases (yes vs. no) | 0.1131 | 0.627 | 1.12 (0.71–1.77) |
HR, hazard ratio; CI, confidence interval; PS, Eastern Cooperative Oncology Group performance status; NG, nuclear grade.
UFPs for significant prognostic factors.
Factor | UFPs |
---|---|
PS | |
≥2 | 1 |
0–1 | 0 |
NG | |
3 | 1.5 |
1–2 | 0 |
Previous systemic therapy | |
Yes | 1 |
No | 0 |
Brain metastases | |
Yes | 1.5 |
No | 0 |
Liver metastases | |
Yes | 1 |
No | 0 |
UFP, unfavorable point; PS, performance status; NG, nuclear grade.