Prevention of locoregional recurrence and distant metastasis in Japanese breast cancer patients using Japanese standard postoperative radiation fields: Experience at a single institution
Abstract
Background
Radiotherapy is an effective local control therapy for breast cancer. Locoregional control is associated with distant metastasis risk and survival after surgery.
Aim
We aimed to evaluate whether Japanese standard postoperative radiotherapy after surgery correlates with disease-free survival (DFS) and overall survival and clarify the characteristics of patients who benefit from it.
Method and Results
This retrospective study included 626 operable breast cancer patients. Tumor characteristics and survival outcomes were compared between patients who received radiotherapy and those who did not. Cox proportional hazard analysis was used to analyze prognostic factors for DFS and perform subgroup analysis. Propensity score matching was used to evaluate the efficacy of radiotherapy using a logistic regression model in patients who received radiotherapy or did not. The median follow-up duration after diagnosis of breast cancer was 63 months. DFS and overall survival were better in the irradiated group (P= .002 and P = .001, respectively). Radiotherapy was more effective for estrogen receptor (ER)–positive disease and for early breast cancer without lymph node metastasis. Multivariate analysis revealed that radiotherapy was a dependent risk factor for recurrence or metastasis.
Conclusion
Radiotherapy prevents distant metastasis and recurrence in early breast cancer patients. In particular, ER-positive, node-negative patients benefit from Japanese standard tangent field radiation.
1 INTRODUCTION
Whole-breast radiation therapy (WBRT) after breast-conserving surgery and postmastectomy radiation therapy (PMRT) for node-positive patients reduces locoregional recurrence rates and overall mortality.1-4 Local control after breast cancer surgery is associated with distant metastasis and overall survival (OS). Thus, radiation therapy is performed appropriately as primary therapy for early-stage breast cancer patients. However, aggressive irradiation including regional lymph nodes increases the risks of pneumonitis, lymphedema, cardiac disease, and late secondary neoplasms.5-7 The WBRT area recommended by Japanese Society of Therapeutic Radiology (JASTRO) guidelines8 usually includes a wider region, especially in the axillary area, than that recommended by the American Society of Radiation Oncology (ASTRO)9 or European Society of Radiotherapy and Oncology (ESTRO)10 guidelines; this method might be efficient when performed for Japanese women with less subcutaneous tissues. We named this Japanese standard postoperative radiation fields.
The aim of this study is to evaluate whether the Japanese standard fields after surgery correlates with disease-free survival (DFS) and OS and the characteristics of patients who benefit from it.
1.1 Patients and methods
1.1.1 Patients
From 2007 to 2017, 626 patients with early breast cancer underwent breast surgery at Shiga General Hospital (Moriyama, Shiga, Japan) in this single-center, retrospective cohort study. Following surgery, radiotherapy was recommended for the patients who were treated with breast-conserving surgery or PMRT for node-positive patients. Patients were required consent and ethical approval for this study.
Eligible patients were recruited at our institution referring patients after breast cancer surgery. Eligibility criteria of adjuvant radiotherapy were derived from international guidelines for suitable patients for breast radiotherapy.8-10 Eligible patients are with a confirmed histological diagnosis of invasive breast cancer and/or ductal carcinoma in situ, after breast-conserving surgery with axillary node dissection or sentinel lymph node biopsy. Patients were divided into two groups, the group with patients who received radiotherapy (RT+) and who did not (RT−). RT− group includes both the patients who were in low risk without radiotherapy and did not need to receive it and who could not receive it due to contraindications like preexisting collagen vascular diseases.
A dose of 50.0 Gy in 25 fractions was regularly prescribed both in WBRT and PMRT. For patients who were under 50 years of age, 10.0 Gy was added as a boost for all patents.8 When the resection margin was close to the cancer lesion after breast-conserving surgery, 10.0 Gy was added as a boost for all patients.8 For some of elderly patients aged over 70 years, hypofractionated radiation therapy (42.4 Gy in 16 fractions) with or without boost addition (10.6 Gy in 4 fractions)10, 11 and radiation omission were options if she or he could receive endocrine therapy for luminal A-like, stage I disease.12, 13
The aim of irradiation was to treat the whole breast with a tangential opposite-beam interpolation. In this technique, the fields for the WBRT involve the lower axillary lymph nodes as standard borders: medial border, midline of the skin of anterior chest wall; lateral border, mid-to-posterior axillary line; superior border, sternal notch; and the inferior border, 1 cm below the inframammary fold. In this technique, irradiated fields were wider than those recommended in ASTRO/ESTRO.9, 10 And for the PMRT, supraclavicular lymph nodes area was added to the WBRT.8 Wedges were used to ensure a uniform dose throughout the target volume. The dose was prescribed at a point of mid-separation at two-thirds of the distance between the skin and the base of the tangents. Treatment beam images were obtained to confirm adequate coverage. Patients were treated with four to six MV X-ray in the supine position with the arm abducted.
Adjuvant chemotherapy according to institutional practice was delivered before radiation therapy. Endocrine therapy with tamoxifen or aromatase inhibitor was administered either concurrently with or after radiotherapy. Trastuzumab was recommended for patients with human epidermal growth factor receptor 2 (HER2)–positive disease.
The primary outcomes were DFS, OS, and breast-cancer–specific survival (BCSS). DFS was defined as the time from surgery to the time of a first recurrence in the ipsilateral breast or in nodal or distant sites, contralateral breast cancer, or death from breast cancer. OS was defined as the time from diagnosis to death from any cause, and BCSS was the time to death from breast cancer.
Regional nodal status is one of the risk factors for OS, and radiotherapy is one of the effective local control therapies for these with lymph node metastasis.14-17 Therefore, we focused on the tumor characteristics which benefit from irradiation and these patients who experienced regional lymph node metastasis as primary disease or as recurrence.
1.2 Statistics
An independent t test and χ2 test were used to compare continuous and categorical variables, as appropriate. Kaplan-Meier estimates were performed to calculate DFS, OS, and BCSS. The propensity score for allocation to either the irradiated group or nonirradiated group was estimated using a logistic regression model. Potential risk factors related to the outcome (DFS) and confounding variables associated with the outcome were included as propensity score covariates on the basis of a literature review and prior clinical knowledge, as follows: age, cancer stage, ER, progesterone receptor (PgR), HER2 status, and adjuvant systemic therapy. To match subjects between the irradiated and nonirradiated groups, a nearest neighbor algorithm on a 1: 1 basis without replacement within a caliper of 0.25 × standard deviation of a log-transformed propensity score was used. Covariate balance was checked by using the absolute standardized mean difference with a threshold of 0.2 to determine any substantial imbalance. DFS between the matched groups was compared by plotting Kaplan-Meier curves with a log-rank test. Multivariate Cox proportional hazard models for DFS were created to identify the independent risk factors. A two-sided P value of less than 0.05 was considered significant.
We conducted all statistical analyses by using the R software package (version 3.4.1; R Development Core Team).
2 RESULTS
The baseline characteristics of the 626 patients in the unmatched cohort are presented in Table 1. Among the 438 patients in the irradiated group, five received hypofractionated radiation therapy (1.1%) and two (0.5%) of them with a 10.6-Gy boost, and 137 (31.3%) received a 10.0-Gy boost addition. Although 55 (12.1%) of the patients after breast conserving surgery omitted radiotherapy, 27 (5.9%) of them were over 70 years of age with luminal A-like diseases, seven (1.5%) were with interstitial pneumonia or collagen vascular diseases. Three (0.7%) had another malignant disease in severe clinical stage such as lung cancer and ovarian cancer, the remaining 18 (4.0%) rejected against expert recommendations to avoid complications due to radiotherapy. Table S1 shows patient characteristics of these 28 patients who omitted but needed WBRT. In patients who underwent mastectomy, 132 (76.7%) of the patients were in RT- group. This is because 84 (48.8%) of the patients were with N0 disease, and 11 (6.3%) had another severe malignant diseases, or collagen vascular diseases. Patient characteristics of the rest 37 (21.5%) who omitted radiotherapy were indicated in Table S2.
| Variable | Total (n = 626) | RTx+ (n = 438) | RTx− (n = 188) | P |
|---|---|---|---|---|
| Follow-up time (mo) | .003 | |||
| Median | 63 | 57 | 77 | |
| Range | 1–559 | 1–248 | 1–559 | |
| Age at primary breast cancer (y) | ||||
| Median | 59 | 58 | 62 | |
| Range | 24-95 | 26-84 | 24-95 | |
| Male, No. (%) | 6 (1.0) | 2 (0.5) | 4 (2.1) | .053 |
| BMI | .562 | |||
| Median ± SD | 22.67 ± 3.71 | 22.73 ± 3.74 | 22.54 ± 3.67 | |
| Range | 15.41–38.81 | 15.41–38.81 | 16.45–36.94 | |
| Laterality | .328 | |||
| Right | 285 (45.5) | 205 (46.8) | 80 (42.6) | |
| Left | 341 (54.5) | 233 (53.2) | 108 (57.4) | |
| Disease stage at diagnosis, No. (%) | <.001 | |||
| 0 | 70 (11.1) | 53 (12.1) | 17 (9.0) | |
| I | 271 (43.3) | 219 (50.0) | 52 (27.7) | |
| IIA | 146 (23.3) | 92 (21.0) | 54 (28.7) | |
| IIB | 60 (9.6) | 33 (7.5) | 27 (14.4) | |
| IIIA | 31 (5.0) | 16 (3.7) | 15 (8.0) | |
| IIIB | 23 (3.7) | 13 (3.0) | 10 (5.3) | |
| IIIC | 14 (2.2) | 12 (2.7) | 2 (1.0) | |
| Unknown | 11 (1.8) | 0 (0.0) | 11 (5.9) | |
| Histology, No. (%) | .340 | |||
| Invasive ductal carcinoma | 488 (77.9) | 342 (78.1) | 146 (77.7) | |
| Invasive lobular carcinoma | 16 (2.6) | 12 (2.7) | 4 (2.1) | |
| Mucinous carcinoma | 24 (3.8) | 13 (3.0) | 11 (5.9) | |
| Ductal carcinoma in situ | 70 (11.2) | 53 (12.1) | 17 (9.0) | |
| Other | 20 (3.2) | 14 (3.2) | 6 (3.2) | |
| Mixed | 8 (1.3) | 4 (0.9) | 4 (2.1) | |
| Histological grade | .020 | |||
| Grade 1 | 198 (38.3) | 126 (37.4) | 72 (45.3) | |
| Grade 2 | 128 (24.8) | 94 (27.9) | 34 (21.4) | |
| Grade 3 | 104 (20.1) | 68 (20.2) | 36 (22.6) | |
| Other | 87 (16.8) | 70 (14.5) | 17 (10.7) | |
| Receptor status, No. (%) | .006 | |||
| ER+/HER2- | 418 (66.8) | 299 (68.3) | 119 (63.3) | |
| ER+/HER2+ | 72 (11.5) | 53 (12.1) | 19 (10.1) | |
| ER-/HER2+ | 38 (6.1) | 27 (6.2) | 11 (5.9) | |
| ER-/HER2- | 73 (11.7) | 50 (11.4) | 23 (12.2) | |
| Unknown | 25 (3.9) | 9 (2.0) | 16 (8.5) | |
| Breast surgery, No. (%) | <.001 | |||
| Breast conserving surgery | 453 (72.4) | 398 (90.9) | 55 (29.3) | |
| Mastectomy | 172 (27.5) | 40 (9.1) | 132 (70.2) | |
| Not performed | 1 (0.1) | 0 (0.0) | 1 (0.5) | |
| Axillary surgery, No. (%) | <.001 | |||
| SLNB | 324 (51.8) | 260 (59.4) | 64 (34.0) | |
| Ax | 253 (40.4) | 160 (36.5) | 93 (49.5) | |
| Not performed | 49 (7.8) | 18 (4.1) | 31 (16.5) |
- Abbreviations: Ax, axillary lymph node dissection; BMI, body mass index; ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; SLNB, sentinel lymph node biopsy.
- Bolded values indicate statistically significance.
Two patients in RT+ group could not complete radiation therapy as planned because they dropped out of the treatment or experienced severe dermatitis related to radiation therapy. Compared with those in the irradiated group, those in the nonradiation group were more likely to be older, have bilateral breast cancer or advanced stage disease, and have undergone mastectomy and axillary lymph node dissection. Table 2 shows difference of the characteristic between the patients who received the WBRT and the PMRT. The PMRT was tended to be more performed in the patients with cancer in the center of breast and multiple lesions. As for tumor characteristics, the PMRT group included high-stage, high-grade, and ER-negative disease.
| Variable | Total (n = 438) | WBRT (n = 398) | PMRT (n = 40) | P |
|---|---|---|---|---|
| Follow-up time (mo) | .910 | |||
| Median | 57 | 57 | 56 | |
| Range | 1-248 | 1-211 | 6-248 | |
| Age at primary breast cancer (y) | .654 | |||
| Median | 58 | 58 | 59 | |
| Range | 26-84 | 26-84 | 31-83 | |
| Male, No. (%) | 2 (0.5) | 1 (0.3) | 1 (2.5) | |
| BMI | 0.260 | |||
| Median ± SD | 22.73 ± 3.74 | 22.67 ± 3.72 | 23.41 ± 3.94 | |
| Range | 15.41-38.81 | 15.41-38.81 | 17.19-33.38 | |
| Laterality | .926 | |||
| Right | 205 (46.8) | 186 (46.7) | 19 (47.5) | |
| Left | 233 (53.2) | 212 (53.3) | 21 (52.5) | |
| Disease stage at diagnosis, No. (%) | <.001 | |||
| 0 | 53 (12.1) | 53 (13.3) | 0 (0.0) | |
| I | 219 (50.0) | 216 (54.2) | 3 (7.5) | |
| IIA | 92 (21.0) | 88 (22.1) | 4 (10.0) | |
| IIB | 33 (7.5) | 26 (6.5) | 7 (17.5) | |
| IIIA | 16 (3.7) | 7 (1.8) | 9 (22.5) | |
| IIIB | 13 (3.0) | 3 (0.8) | 10 (25.0) | |
| IIIC | 12 (2.7) | 5 (1.3) | 7 (17.5) | |
| Unknown | 0 (0.0) | 0 (0.0) | 0 (0.0) | |
| Histology, No. (%) | .072 | |||
| Invasive ductal carcinoma | 338 (77.2) | 303 (77.9) | 35 (87.5) | |
| Invasive lobular carcinoma | 12 (2.7) | 10 (3.0) | 2 (5.0) | |
| Mucinous carcinoma | 13 (3.0) | 13 (3.0) | 0 (0.0) | |
| Ductal carcinoma in situ | 57 (13.0) | 57 (12.1) | 0 (0.0) | |
| Other | 14 (3.2) | 12 (3.4) | 2 (0.5) | |
| Mixed | 4 (0.9) | 3 (0.6) | 1 (2.5) | |
| Histological grade | <.001 | |||
| Grade 1 | 126 (37.4) | 119 (39.3) | 7 (20.6) | |
| Grade 2 | 94 (27.9) | 82 (27.1) | 12 (35.3) | |
| Grade 3 | 68 (20.2) | 53 (17.5) | 15 (44.1) | |
| Other | 49 (14.5) | 49 (16.1) | 0 (0.0) | |
| Receptor status, No. (%) | <.001 | |||
| ER+/HER2- | 299 (68.3) | 279 (70.1) | 20 (50.0) | |
| ER+/HER2+ | 53 (12.1) | 46 (11.6) | 7 (17.5) | |
| ER-/HER2+ | 27 (6.2) | 18 (4.5) | 9 (22.5) | |
| ER-/HER2- | 50 (11.4) | 46 (11.6) | 4 (10.0) | |
| Unknown | 9 (2.0) | 9 (2.2) | 0 (0.0) | |
| Axillary surgery, No. (%) | <.001 | |||
| SLNB | 260 (59.4) | 259 (65.1) | 1 (2.5) | |
| Ax | 160 (36.5) | 122 (30.7) | 38 (95.0) | |
| Not performed | 18 (4.1) | 17 (4.2) | 1 (2.5) | |
| Unknown | 0 (0.0) | 0 (0.0) | 0 (0.0) |
- Abbreviations: Ax, axillary lymph node dissection; BMI, body mass index; ER, estrogen receptor; HER2, human epidermal growth factor receptor 2; SLNB, sentinel lymph node biopsy.
- Bolded values indicate statistically significance.
There was no significant between-group difference in BCSS; however, DFS and OS were better in the irradiated group (hazard ratio, HR for DFS and OS in the irradiated group compared with the nonirradiated group, 0.61; 95% confidence interval, CI, 0.34-0.78; P = .002, and 0.43; 95% CI, 0.22-0.70; P = .001, respectively; Figure 1A,B). In order to ensure an even distribution of confounders between the irradiated and nonirradiated group and increase between group comparability, propensity score adjustment was performed in this observational study. After matching based on the propensity score, a total of 260 patients remained in the matched cohort, and the baseline characteristics were well balanced without significant differences between the irradiated and nonirradiated groups. Estimated propensity scores were used to match 130 patients in each group. The number of recurrent, death, and breast cancer specific death events were 50, 18, and 13, respectively. After propensity score matching, the multivariate analyses showed that nonirradiation was a significant risk factors for DFS (odds ratio; 0.29, 95% CI, 0.13-0.63; P = .002), however, not in OS (odds ratio; 0.38, 95% CI, 0.14-1.08; P = .069) and BCSS (odds radio; 0.44, 95% CI, 0.14-1.44; P = .177). In prespecified subgroup analyses, the treatment effects were greater for patients with early-stage, node-negative, hormone receptor-positive tumors (Figure 2).
3 DISCUSSION
In this study, we found that patients who received Japanese standard high tangent field radiotherapy after surgery had better DFS and OS than those in the nonirradiated group. Radiation therapy is sometimes omitted owing to older age or comorbidities.12, 13 Since this study included patients with various characteristics and age ranges, propensity-score–matched analysis was performed. Estimated propensity scores were used to match 130 patients in each group. After propensity-score matching, the multivariate analyses showed significant risk factors for recurrence or metastasis after primary therapy in the nonirradiated group. As described in previous reports, irradiation to prevent breast cancer recurrence was useful for operable breast cancer patients.1-4, 18 Early breast cancer without lymph node metastasis benefitted from radiotherapy in particular; however, the significance for patients with ductal carcinoma in situ (DCIS) was unknown. In a recent study, radiation therapy was reported to be more effective for high grade DCIS than low grade DCIS. All grades of DCIS were included in this study; thus, we could not show the significance for DCIS. Patients with ER-positive disease appeared to benefit more from irradiation than those with ER-negative tumors. This supports previous data. Since ER-negative or node positive patients usually receive multiagent chemotherapy before radiation therapy, regimens, or treatment intensity should have been considered.
In order to elucidate longer survival, the patients who underwent surgery before 2012 were collected because BCSS was not significantly improved after radiation therapy. In the subgroup analysis (Figure 2), breast cancer in early stage benefit from radiation therapy as in T1 or N0. The patients were divided to those who received WBRT, which is usually performed in early breast cancer patients and PMRT with locally advanced breast cancer. Figures S1 and S2 show Kaplan-Meier plot comparing DFS, OS, and BCSS between those who received WBRT, PMRT, and the nonirradiated group who needed irradiation. The patients who did not receive radiotherapy but needed it omitted because of skin disease such as SLE, or complications like duplicate cancers, and refused to receive radiotherapy. Although the irradiated group included more advanced stage, and ER-negative disease (Supplemental Table 1)19 compared with the non-irradiated group, significant improvement in BCSS associated with WBRT was not shown (HR, 0.91; 95% CI, 0.21-3.94; P = .897). Furthermore, all the prognosis was poorer in PMRT group than the nonirradiated group (Figure S2). The patient characteristics, who received PMRT and who omitted but needed, are shown in Table S2. Interestingly, patients who received PMRT included eight patients whose BMI was over 25. Five patients in PMRT group experienced recurrence and two died for breast cancer in overweight group even after radiotherapy. Figure S3 shows irradiated field in PMRT. Radiation distributes ineffectively in such patients who have more subcutaneous tissue (Figure S3A) compared with these who have less (Figure S3B) if the same dose was irradiated. However, only 3.3% of the patients were classified as overweight in total, significance as risk of high BMI was not indicated because of small number.
Normally, systemic therapy has significant effects on the prognosis of breast cancer patients. However, we evaluated radiation therapy only in this study. Thus, radiation therapy did not directly lead to improvement of BCSS especially for such high-stage or ER-negative patients. In addition, over half of the patients who did not receive radiotherapy died from other diseases such as other cancers or cardiac dysfunctions (63.0%). This is one the reasons of significant difference for OS between irradiated group and nonirradiated group (Figure 1).
In subgroup analyses, node positive breast cancer patients did not significantly benefit from radiation therapy (Figure 2). However, MA.20 clinical trials previously showed that addition of regional nodal irradiation to WBRT improves DFS of patients with node-positive or high-risk node-negative breast cancer.20 We believe the patients who underwent radiotherapy still included those who could be omitted from irradiation, as low-grade DCIS19, 21, 22 or part of luminal A-like super early stage disease. Radiotherapy can be improved more efficiently by selecting those who needs it.
The limitation of this study is absence of long-term follow-up. Disadvantages of radiotherapy was unclear because the incidence of late complications related with radiotherapy was not observed enough. Since the prognosis of breast cancer is also affected by subtype and systemic therapy, further investigation with more patients is needed. The other limitation was that it was a retrospective observational cohort study and relatively small population and event number. Thus, these factors have limited the statistical power and residual confounding cannot be ruled out.
4 CONCLUSION
Japanese standard postoperative radiation fields for operable breast cancer patients prevents distant metastasis and recurrence. In particular, ER-positive, node-negative patients benefit from Japanese standard high tangent field radiation therapy.
ACKNOWLEDGEMENTS
None.
CONFLICT OF INTEREST
No conflict of interest relevant to this article are reported.
AUTHORS' CONTRIBUTIONS
All authors had full access to the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Conceptualization, H.T., W.T., F.Y., C.Y.; Methodology, H.T., W.T., F.Y., C.Y.; Investigation, H.T.; Formal Analysis, H.T., C.Y.; Resources, H.T., W.T., F.Y., C.Y.; Writing - Original Draft, H.T.; Writing - Review & Editing, H.T.; Visualization, H.T., C.Y.; Supervision, W.T., F.Y., C.Y.; Funding Acquisition, none.
DATA ACCESSIBILITY STATEMENT
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.
