Abstract
Introduction
Bone-only metastasis (BOM) is a distinct clinical phenomenon in which cancer cells disseminate exclusively to the bones, without involvement of other distant organs. We investigated the factors associated with the BOM state versus other states of metastasis in breast cancer patients with bone metastasis (BM) at their first relapse. The results could help tailor the screening and preventive therapy strategies for BM in breast cancer.
Methods
The study included 231 women who underwent mastectomy for primary unilateral non-metastatic breast cancer in 1997 or later and were subsequently diagnosed with BM at first relapse in 2008–2018 at the Fourth Hospital of Hebei Medical University in China. Factors such as patient age at primary breast cancer diagnosis, tumor clinicopathological characteristics, chemotherapy, radiotherapy, endocrine therapy (ET), time to progression (TTP), and others were analyzed. ET compliance was categorized from medication adherence. Multivariate logistic regressions were used to estimate the odds ratio (OR) and p value.
Results
Only three (3.8%, 3/79) human epidermal growth factor receptor 2-positive (HER2+) breast cancer patients (n = 79) used anti-HER2-targeted agents in the adjuvant setting. After excluding them, the remaining 228 patients were analyzed. They had an average age of 47.3 years and median TTP 29.4 months at their first relapse. Overall, patients with BOM accounted for 26.8%. The BOM state was similarly presented in the hormone receptor-positive (HR+) patients (n = 182) and in the HR-negative (HR−) patients (n = 45) (28.6% vs. 17.8%, p = 0.142). However, it was significantly lower in the HER2+ patients (n = 76) than in the HER2-negative (HER2−) patients (n = 129) (13.2% vs. 31.8%, p = 0.003). Multivariate analyses showed that the BOM state was not associated with the HR+ (vs. HR−, OR 1.253, p = 0.723) and full ET compliance (vs. no/partial, OR 1.346, p = 0.545) status. Nonetheless, the BOM state was significantly associated with a lower chance in the HER2+ patients overall (OR 0.240, p = 0.008) and in the HR+ patients (OR 0.145, p = 0.005) but not in the HR− patients (OR 1.012, p = 0.991) than one in the HER2− patients. A lower chance of BOM state was also associated with TTP ≥24 months (p < 0.05). There were no other associated factors identified.
Conclusion
Differently from HR status and other clinicopathological factors, the HER2+ status is associated with a lower chance of the BOM state in breast cancer patients with first BM. Such association appears to be reflected in HR+ patients only.
Keywords: Breast cancer, Bone metastasis, Bone-only metastasis, HER2, Hormone receptor, Endocrine therapy
Introduction
Breast cancer is the most commonly diagnosed cancer in women worldwide [1, 2]. Despite advancements in the treatment of localized breast cancer, managing metastatic disease remains a significant challenge [1, 2]. The common metastatic sites for breast cancer include the bone, lung, liver and brain, with bone being the most frequent organ involved [1, 2]. Bone metastasis (BM) can significantly impact a patient’s quality of life, leading to skeletal-related events such as bone pain, hypercalcemia, pathological fractures, and spinal cord compression [2, 3]. These complications can be managed with a combination of treatments, including radiotherapy (RT), surgery, and systemic therapies to improve the patient’s daily functioning and overall well-being [2, 3].
Bone-only metastasis (BOM) is a distinct clinical phenomenon in which cancer cells disseminate exclusively to the bones, without involvement of other distant organs [2, 4]. Patients with BOM, whether diagnosed at primary admission or during post-therapy surveillance, generally experience better progression-free survival and overall survival than patients with metastases to the lung, liver, or brain, with 5-year overall survival rates exceeding 20% [5–7]. This favorable prognosis is often attributed to early diagnosis and timely application of effective therapies on BM [2]. However, the exact mechanisms of the development of BOM and its more favorable prognosis remain unclear [8]. To investigate how BOM develops differently from BM combined with other sites or organ metastasis is important for optimizing BM management in breast cancer [3–5].
Many factors could be involved in the development, diagnosis, continuation, and prognosis of BOM in breast cancer [2, 9–11]. These include tumor clinicopathological features, factors related to BM diagnosis, and therapeutics for breast cancer and BM. Understandably, patient’s symptom severity of BM, frequency of office visits, choice of appropriate laboratory tests (e.g., alkaline phosphates, calcium levels), and selection of imaging modalities (e.g., X-ray, bone scan, CT, MRI, PET-CT) all influence the diagnosis and evaluation of BM [12–14]. The preventive and therapeutic administration of bone-modifying agents like bisphosphonates and denosumab impacts this course as well [2, 12–14]. Currently, the optimal timing and frequency schedule of these applications above are still largely unknown [14]. In this regard, knowing the factors related to the BOM state can boost the understanding of BM development, help tailor the surveillance strategies of BM, and assist the cost-effective use of bone-modifying agents in breast cancer [15]. So far, there are few published studies investigating the factors relating to BOM development or state in any metastatic breast cancer patients [7, 16].
For decades, the molecular subtype of breast cancer has played a critical role in guiding treatment decisions and predicting patient prognosis [17–19]. Patients with estrogen receptor or progesterone receptor-positive (ER+, PR+) tumors are more likely to develop BM compared to other site or organ distant metastasis [5, 8, 20, 21]. In contrast, patients with human epidermal growth factor receptor 2-positive (HER2+) breast cancer are more likely to experience metastases in organs other than bone [20–23]. Endocrine therapies (ETs) for hormone receptor-positive (HR+, defined as ER+ or PR+) tumors and anti-HER2-targeted therapies for HER2+ tumors have long been proven to improve tumor control and patient survival [18–23]. However, the interactive impact of these biomarkers and biomarker-based therapies on the development of BM or BOM remains underexplored [2, 20].
Trastuzumab, the first anti-HER2-targeted therapy developed, was approved for use in HER2+ breast cancer patients in 2002, initially for metastatic disease and later for adjuvant therapy. However, due to its limited availability and high cost, many Chinese patients with breast cancer could not access trastuzumab prior to 2010. In fact, even in 2008, only 20% of eligible patients in the nation’s capital city of Beijing were being treated with this standard Western medicine [24, 25]. Considering this circumstance, we designed an epidemiological study to investigate how factors such as TNM staging, ER/PR/HER2 status, and other clinicopathological characteristics are independently associated with the BOM state versus other forms of bone involvement in a cohort of Chinese breast cancer patients experiencing their first relapse. The results could help refine screening and preventive therapy strategies for BM in breast cancer. In this study, we present the findings from our analysis.
Methods
Data Source
This retrospective study analyzed data from patients registered at the Fourth Hospital of Hebei Medical University in China. The study was approved by the hospital’s Institution Review Board (protocol # 2020-190). As the leading cancer center in Hebei province, the hospital has provided healthcare services to over 15 million people in the Shijiazhuang metropolitan region for more than 5 decades. All patients enrolled in the study provided written informed consent, and all sensitive health information was excluded from the dataset.
Patient Selection and Study Design
Between 2008 and 2018, a total of 703 patients with breast cancer who experienced their first tumor relapse were identified from the hospital database (Fig. 1). Inclusion criteria for further study were (1) female patients with unilateral breast cancer, age ≥18 years, and who received breast surgery in 1997 or later, when the hospital’s breast cancer management protocol was implemented; (2) mastectomy, with or without axillary lymph node dissection; (3) clinical diagnosis of BM or other distant metastasis confirmed by two independent licensed radiologists, supported by imaging (e.g., X-rays, bone scan, MRI, CT, PET) and confirmed by biopsy if necessary. Exclusion criteria were (1) inflammatory, bilateral, metastatic primary breast cancer (2) ipsilateral locoregional recurrence (LRR) only (3) LRR, BM or other distant metastasis diagnosed prior to the completion of surgery, adjuvant chemotherapy, or post-operative RT, whichever was the last treatment for the primary tumor; and (4) metastases from other malignant diseases. Additionally, 3 patients who had received trastuzumab in the adjuvant setting were excluded from the final analysis given that the majority of patients in our cohort did not receive this treatment.
Fig. 1.
Flowchart illustrates the selection process of study patients.
According to the hospital’s breast cancer follow-up protocol, office visits were typically scheduled every 3 to 6 months during the first 2 to 3 years, every 6 to 12 months from year 3 to 5, and annually thereafter. Each visit included routine history-taking and physical exam, followed by selected imaging studies when necessary.
Variables Analyzed and Definition of Key Terms
In this study, we established clear criteria to define various forms of recurrence and metastasis in breast cancer, including LRR, distant metastasis, and BM lesions and types to ensure consistency in categorizing tumor relapses or metastases. LRR was defined as tumor relapses at the ipsilateral chest wall or lymph node within lymphatic drainage regions of the primary breast tumor. Tumor recurrence at the contralateral chest wall or breast or any other regional lymph node was regarded as a distant metastasis. BM was considered a form of distant metastasis, resulting from tumor cells entering the blood stream or lymphatic system. Single bone lesion was defined as solitary metastatic lesion confined to within one bone, while BM type was assigned based on all BM lesions in any involved bone. Pathological diagnosis was required for tumor relapses at the chest wall, breast, and most axillary and supraclavicular nodes.
The earliest diagnosis date for LRR, BM, or other distant metastasis was defined as the theoretical “BM date” in analysis. Time to progression (TTP) was calculated as 30 days/1 month from this date. For modeling purposes, a categorical TTP variable was used in logistic regression models after identifying a nonlinear relationship of analysis endpoint with TTP as a continuous variable.
Both ER+ and PR+ status were defined as ≥1% staining of primary tumor cell nuclei with immunohistochemistry (IHC). At our institution, Fluorescent in situ hybridization (FISH) was then recommended but not required to determine the HER2 status of breast cancer until 2011, and therefore, these data were not available for all patients. HER2-negative (HER2−) status was defined as IHC scores of 0 or +1 or a FISH-negative result. HER2+ status was defined as an IHC score of +3 or a FISH-positive result. For patients with an IHC score of +2 but no FISH analysis, HER2 status was classified as “undetermined.”
Since some HR+ breast cancer patients did not fully adhere to prescribed ET in terms of dose, frequency, or duration, we created a variable to indicate ET compliance status – categorized as full compliance or non/partial compliance. This variable reflected the patient’s adherence to prescribed ET up to the BM date, or at least for 5 years. Five ET drugs (Tamoxifen, Toremifene, Letrozole, Anastrozole, and Exemestane) were available during the study period. We observed that none of the patients included in the analysis had undergone oophorectomy, ovarian radiation, or received ovarian function suppression drugs or bone-modifying agents prior to the BM date.
Statistical Analysis
Continuous and categorical variables were summarized using descriptive statistics. To examine relationships between variables, we used Analysis of Variance, t tests, chi-squared tests, Fisher’s exact tests, or log-rank tests as appropriate. Given that BM dates could be influenced by factors such as physician office visits or imaging schedules, we opted to use logistic regression to explore the cross-sectional relationships of study factors. Odds ratio (OR), 95% confidence interval (CI), and p value were estimated. A two-sided p value <0.05 was considered the statistically significant level. All analyses were performed using SAS 9.4 for Windows. Final multivariate models were developed based on the results of univariate analyses and clinical relevance of the variables.
Results
Study Population and Their Characteristics
Figure 1 illustrates the selection process of study patients. A total of 231 breast cancer patients with BM at their first relapse were enrolled in the study. After excluding three (3.8%, 3/79) HER2+ patients (n = 79) who received trastuzumab in the adjuvant setting, 228 patients remained in the analysis. Table 1 presents the baseline characteristics of these patients at the time of their primary breast cancer diagnosis.
Table 1.
Baseline characteristics of breast cancer patients with BM at their first relapse
| Variable | All, n (%) | BOM, n (%) | p valuea | |
|---|---|---|---|---|
| yes | no | |||
| Patients | 228 (100.0) | 61 (26.8) | 167 (73.2) | |
| Age at primary breast cancer diagnosis, years | ||||
| Mean±std | 47.3±9.6 | 46.4±10.4 | 47.6±9.4 | 0.403 |
| Median (Q1 – Q3) | 47 (40–54) | 45 (39–54) | 47 (41–55) | 0.405 |
| Breast cancer diagnosis yearb | 0.164 | |||
| 1997–2010 | 95 (41.7) | 30 (49.2) | 65 (38.9) | |
| 2011–2016 | 133 (58.3) | 31 (50.8) | 102 (61.1) | |
| Tumor laterality | 0.027 | |||
| Left | 122 (53.5) | 40 (65.6) | 82 (49.1) | |
| Right | 106 (46.5) | 21 (34.4) | 85 (50.9) | |
| Tumor location | 0.118 | |||
| Internal | 54(23.7) | 10(16.4) | 44(26.3) | |
| External | 174(76.3) | 51(83.6) | 123(73.7) | |
| cT stage | 0.232 | |||
| T1 | 54 (23.7) | 16 (26.2) | 38 (22.8) | |
| T2 | 115 (50.4) | 35 (57.4) | 80 (47.9) | |
| T3/4 | 41 (18.0) | 6 (9.8) | 35 (21.0) | |
| Unknown | 18 (7.9) | 4 (6.6) | 14 (8.4) | |
| cN stage | 0.228 | |||
| N0 | 79 (34.6) | 26 (42.6) | 53 (31.7) | |
| N1-3 | 140 (61.4) | 32 (52.5) | 108 (64.7) | |
| Unknown | 9 (3.9) | 3 (4.9) | 6 (3.6) | |
| BC pathology | 0.534 | |||
| IDC | 189 (82.9) | 49 (80.3) | 140 (83.8) | |
| Other | 39 (17.1) | 12 (19.7) | 27 (16.2) | |
| Tumor grade | 0.438 | |||
| I/II | 145 (63.6) | 39 (63.9) | 106 (63.5) | |
| III | 51 (22.4) | 11 (18.0) | 40 (24.0) | |
| Unknown | 32 (14.0) | 11 (18.0) | 21 (12.6) | |
| Lymphovascular invasion | 0.358 | |||
| Negative | 59 (25.9) | 18 (29.5) | 41 (24.6) | |
| Positive | 52 (22.8) | 10 (16.4) | 42 (25.1) | |
| Unknown | 117 (51.3) | 33 (54.1) | 84 (50.3) | |
| HR status | 0.142 | |||
| Positive | 182 (79.8) | 52 (85.3) | 130 (77.8) | |
| Negative | 45 (19.7) | 8 (13.1) | 37 (22.2) | |
| Unknown | 1 (0.4) | 1 (1.6) | ||
| HER2 status | 0.002 | |||
| Positive | 76 (33.3) | 10 (16.4) | 66 (39.5) | |
| Negative | 129 (56.6) | 41 (67.2) | 88 (52.7) | |
| Undetermined | 16 (7.0) | 6 (9.8) | 10 (6.0) | |
| Unknown | 7 (3.1) | 4 (6.7) | 3 (1.8) | |
| Chemotherapy used | 1.000 | |||
| Yes | 219 (96.1) | 59 (96.7) | 160 (95.8) | |
| No | 9 (3.9) | 2 (3.3) | 7 (4.2) | |
| RT used | 0.182 | |||
| Yes | 103 (45.2) | 32 (52.5) | 71 (42.5) | |
| No | 125 (54.8) | 29 (47.5) | 96 (57.5) | |
| HR status and ET compliance | 0.057 | |||
| HR+ and full compliance | 99 (43.4) | 25 (41.0) | 74 (44.3) | |
| HR+ and no/partial compliance | 80 (35.1) | 25 (41.0) | 55 (32.9) | |
| HR− and no ET | 45 (19.7) | 8 (13.1) | 37 (22.2) | |
| Unknown | 4 (1.8) | 3 (4.9) | 1 (0.6) | |
BM, bone metastasis; BOM, bone-only metastasis; std, standard deviation; Q1, first quartile; Q3, third quartile; cT, clinical T; cN, clinical N; IDC, invasive ductal carcinoma; HR, hormone receptor; HER2, human epidermal growth factor receptor 2; RT, radiotherapy; ET, endocrine therapy.
a p value from ANOVA or chi-squared test on patients without the “unknown” or “undetermined” value.
bFISH test was mandatory for HER2 IHC 2+ tumor after 2010.
In this retrospective study, we limited the study population to patients with BM at their first relapse of any time, excluding those with only LRR (Fig. 1). In this regard, we decided to assign BM patients with LRR only to not have the BOM state. Overall, the proportion of patients presenting with BOM was 26.8%. Tumor laterality (left-sided 65.6% vs. 49.1%, p = 0.027) and HER2 status (p = 0.002) were statistically related to the BOM state. Specifically, the proportion of patients with BOM was significantly lower in HER2+ patients (n = 76) compared to in the HER2– patients (n = 129, 13.2% vs. 31.8%, p = 0.003). There was no significant difference in the proportion of BOM between HR+ patients (n = 182) and HR-negative (HR−) patients (n = 45, 28.6% vs. 17.8%, p = 0.142).
Table 2 presents the characteristics of BM patients at their first relapse of any type. They had an average age of 50.1 years and a median TTP of 29.4 (95% CI: 25.1–32.8) months. The TTP median did not differ significantly between the BOM patients and others (the non-BOM/BM patients) (20.7 vs. 30.9 months, p = 0.250). However, the proportion of BOM was statistically higher within the first 12 months (27.9% vs. 9.6%, p = 0.001) and lower (16.4% vs. 24.6%, p < 0.001) after 48 months. Additionally, BOM patients were more likely to have lumbar vertebrae involvement (57.4% vs. 32.3%, p = 0.001) and less likely to have rib involvement (31.1% vs. 46.7%, p = 0.035) compared to non-BOM/BM patients.
Table 2.
Characteristics of breast cancer patients with BM at first relapse
| Variable | All, n (%) | BOM, n (%) | p valuea | |
|---|---|---|---|---|
| yes | no | |||
| Patients | 228 (100.0) | 61 (26.8) | 167 (73.2) | |
| Age at BM, years | ||||
| Mean±std | 50.1±9.9 | 48.9±10.7 | 50.6±9.5 | 0.262 |
| TTP, months | ||||
| Median (95% CI) | 29.4 (25.1–32.8) | 20.7 (17.7–31.3) | 30.9 (27.4–35.0) | 0.250 |
| 0.006 | ||||
| <12 | 33 (14.5) | 17 (27.9) | 16 (9.6) | |
| <24 | 61 (26.8) | 17 (27.9) | 44 (26.3) | |
| <36 | 44 (19.3) | 7 (11.5) | 37 (22.2) | |
| <48 | 39 (17.1) | 10 (16.4) | 29 (17.4) | |
| ≥48 | 51 (22.4) | 10 (16.4) | 41 (24.6) | |
| LRR sites | ||||
| Chest wall or LN regions | 87 (38.2) | 87 (52.1) | ||
| Chest wall | 44 (19.3) | 44 (26.3) | ||
| LN regions | 64 (28.1) | 64 (38.3) | ||
| Other metastatic organ/site | ||||
| Liver | 66 (28.9) | 66 (39.5) | ||
| Lung | 63 (27.6) | 63 (37.7) | ||
| Brain | 8 (3.5) | 8 (4.8) | ||
| Others | 37 (16.2) | 37 (22.2) | ||
| BM lesion type | 0.266 | |||
| Osteolysis | 87 (38.2) | 18 (29.5) | 69 (41.3) | |
| Osteogenesis | 29 (12.7) | 9 (14.8) | 20 (12.0) | |
| Mixed | 112 (49.1) | 34 (55.7) | 78 (46.7) | |
| BM lesion(s) | 0.741 | |||
| Single | 90 (39.5) | 23 (37.7) | 67 (40.1) | |
| Multiple | 138 (60.5) | 38 (62.3) | 100 (59.9) | |
| BM site(s)b | ||||
| Thoracic vertebra | 116 (50.9) | 35 (57.4) | 81 (48.5) | 0.235 |
| Rib | 97 (42.5) | 19 (31.1) | 78 (46.7) | 0.035 |
| Lumbar vertebra | 89 (39.0) | 35 (57.4) | 54 (32.3) | 0.001 |
| Pelvis | 69 (30.3) | 24 (39.3) | 45 (26.9) | 0.071 |
| Sternum | 59 (25.9) | 12 (19.7) | 47 (28.1) | 0.196 |
| Lower limb bone | 29 (12.7) | 12 (19.7) | 17 (10.2) | 0.057 |
BM, bone metastasis; BOM, bone-only metastasis; std, standard deviation; CI, confidence interval; TTP, time to progression; LN, lymph node.
a p value from ANOVA or chi-squared or log-rank test.
bAnatomical bones found in fewer than 10% of patients are not listed.
Stratified Analysis by HR Status
Stratified analysis by HR status was performed to explore the relationship between HER2 status and BOM in HR+ and HR− subsets of patients with BM. Table 3 shows that a significant relationship between HER2 status and the BOM versus non-BOM/BM state was observed only in HR+ BM patients (n = 182, p = 0.001). In contrast, the median TTP was significantly shorter in BOM patients compared to non-BOM/BM patients (9.2 vs. 24.4 months, p = 0.001) in HR– patients (n = 45). The distribution of LRR sites and metastatic organs involved was similar between HR+ (non-BOM/BM) patients and HR− (non-BOM/BM) patients (p = 0.065–0.522). There were no statistically significant differences in the distribution of BM type or BM lesions (single vs. multiple) between BOM versus non-BOM/BM patients in either the HR+ or HR− subsets (p = 0.340–0.886).
Table 3.
Stratified analysis of breast cancer patients with BM at first relapse by HR status
| Variable | HR+ (n = 182) | p valuea | HR– (n = 45) | p valuea | ||
|---|---|---|---|---|---|---|
| BOM, n (%) | BOM, n (%) | |||||
| yes | no | yes | no | |||
| Patientsb | 52 (28.6) | 130 (71.4) | 0.036 | 8 (17.8) | 37 (82.2) | 0.036 |
| Age, years, mean±std | ||||||
| Breast cancer diagnosis | 45.5±10.1 | 47.1±8.9 | 0.276 | 50.0±10.4 | 49.2±10.7 | 0.852 |
| BM diagnosis | 48.2±10.5 | 50.2±9.1 | 0.194 | 51.1±10.5 | 51.9±10.9 | 0.862 |
| TTP, months | ||||||
| Median (95% CI) | 23.1 (18.0–35.9) | 32.8 (29.2–39.8) | 0.597 | 9.2 (0.8–20.5) | 24.4 (19.7–33.0) | 0.001 |
| 0.052 | 0.007 | |||||
| <12 | 12 (23.1) | 13 (10.0) | 5 (62.5) | 3 (8.1) | ||
| <24 | 15 (28.8) | 30 (23.1) | 2 (25.0) | 14 (37.8) | ||
| <36 | 6 (11.5) | 29 (22.3) | 1 (12.5) | 8 (21.6) | ||
| <48 | 10 (19.2) | 21 (16.2) | 8 (21.6) | |||
| ≥48 | 9 (17.3) | 37 (28.5) | 4 (10.8) | |||
| HER2 status | 0.001 | 1.000 | ||||
| Positive | 5 (9.6) | 45 (34.6) | 5 (62.5) | 21 (56.8) | ||
| Negative | 38 (73.1) | 75 (57.7) | 3 (37.5) | 13 (35.1) | ||
| Undetermined | 6 (11.5) | 8 (6.2) | 0 (0.0) | 2 (5.4) | ||
| Unknown | 3 (5.8) | 2 (1.5) | 0 (0.0) | 1 (2.7) | ||
| LRR sites | ||||||
| Chest wall or LN regions | 66 (50.8) | 21 (56.8) | 0.520 | |||
| Chest wall | 36 (27.7) | 8 (21.6) | 0.460 | |||
| LN regions | 45 (34.6) | 19 (51.4) | 0.065 | |||
| Other metastatic organ/site | ||||||
| Liver | 52 (40.3) | 14 (37.8) | 0.237 | |||
| Lung | 44 (34.1) | 19 (51.4) | 0.206 | |||
| Brain | 4 (3.1) | 4 (10.8) | 0.098 | |||
| Others | 29 (22.5) | 8 (21.6) | 0.522 | |||
| BM lesion type | 0.473 | 0.496 | ||||
| Osteolysis | 16 (30.8) | 52 (40.0) | 2 (25.0) | 17 (45.9) | ||
| Osteogenesis | 8 (15.4) | 15 (11.5) | 1 (12.5) | 5 (13.5) | ||
| Mixed | 28 (53.8) | 63 (48.5) | 5 (62.5) | 15 (40.5) | ||
| BM lesion(s) | 0.886 | 0.340 | ||||
| Single | 21 (40.4) | 51 (39.2) | 2 (25.0) | 16 (43.2) | ||
| Multiple | 31 (59.6) | 79 (60.8) | 6 (75.0) | 21 (56.8) | ||
HR, hormone receptor; BM, bone metastasis; BOM, bone-only metastasis; std, standard deviation; TTP, time to progression; CI, confidence interval; HER2, human epidermal growth factor receptor 2; LN, lymph node.
a p value from ANOVA or chi-square or log-rank test on patients.
bOne patient with an unknown HR status was excluded.
Logistic Regression Analysis
The HR and HER2 status and other clinicopathological factors of primary breast cancer were explored through univariate and multivariate logistic regressions to determine their relationships with BOM. Table 4 presents the results of the HR status, HER2 status, and ET compliance in the analysis of 3 patient populations. After adjusting for covariates, HR status was not independently associated with the BOM state (HR+ vs. HR−, OR 1.253, p = 0.723). In contrast, compared to HER2− status, HER2+ status was significantly associated with a 76.0% lower likelihood of having the BOM state overall (OR 0.240, p = 0.008), and an 85.5% lower likelihood in the HR+ subset (OR 0.145, p = 0.005), but not in the HR− patients (OR 1.012, p = 0.991). Furthermore, ET compliance status was not found to be associated with the BOM state in either the overall population (OR 1.346, p = 0.545) or in the HR+ subset (OR 1.210, p = 0.716).
Table 4.
Logistic regression analysis of the BOM state in breast cancer patients with BM at first relapse
| Patientsa | Class | BOM ratio | Univariate | Multivariateb | ||||
|---|---|---|---|---|---|---|---|---|
| y/zc (%) | OR | 95% CI | p value | OR | 95% CI | p value | ||
| All (n = 185–228) | HR+ | 52/182 (28.6) | 1.850 | (0.807–4.239) | 0.146 | 1.253 | (0.361–4.345) | 0.723 |
| HR− | 8/45 (17.8) | 1.000 | ref. | 1.000 | ref. | |||
| HER2+ | 10/76 (3.2) | 0.325 | (0.152–0.696) | 0.004 | 0.240 | (0.083–0.689) | 0.008 | |
| HER2− | 41/129 (31.8) | 1.000 | ref. | 1.000 | ref. | |||
| HER2u | 6/16 (33.5) | 1.288 | (0.438–3.784) | 0.646 | 0.393 | (0.095–1.620) | 0.196 | |
| ETfull | 25/101 (24.8) | 0.832 | (0.459–1.506) | 0.543 | 1.346 | (0.515–3.519) | 0.545 | |
| ETno/partial | 36/127 (28.4) | 1.000 | ref. | 1.000 | ref. | |||
| HR+ (n = 148–182) | HER2+ | 5/50 (10.0) | 0.219 | (0.080–0.598) | 0.003 | 0.145 | (0.037–0.564) | 0.005 |
| HER2− | 38/113 (33.6) | 1.000 | ref. | 1.000 | ref. | |||
| HER2u | 6/14 (42.9) | 1.480 | (0.497–4.574) | 0.496 | 0.476 | (0.104–2.191) | 0.341 | |
| ETfull | 25/99 (25.3) | 0.701 | (0.368–1.336) | 0.280 | 1.210 | (0.434–3.372) | 0.716 | |
| ETno/partial | 27/83 (32.5) | 1.000 | ref. | 1.000 | ref. | |||
| HR− (n = 40–45) | HER2+ | 5/26 (19.2) | 1.032 | (0.210–5.058) | 0.969 | 1.012 | (0.132–7.758) | 0.991 |
| HER2− | 3/16 (18.8) | 1.000 | ref. | 1.000 | ref. | |||
| HER2u | 0/2 (0.0) | 0.001 | (0.000–1) | 0.979 | 0.001 | (0.000–I) | 0.979 | |
BOM, bone-only metastasis; BM, bone metastasis; OR, odd ratio; CI, confidence interval; ref., Reference; HR, hormone receptor (+ [positive], − [negative]); HER2, human epidermal growth factor receptor 2 (+ [positive], − [negative], u [undetermined]); ET, endocrine therapy (subscript: full – full compliance, no/partial – no/partial compliance).
aPatient number (n) participated in models were varied because some variables had missing values.
bFinal covariates included BC diagnosis age, tumor laterality, tumor quadrant, cT stage, cN stage, tumor pathology, tumor grade, lymphovascular invasion, HR status (overall patient model only), HER2 status, axillary lymph node dissection (yes vs. not), chemotherapy, RT, ET compliance (full vs. no/partial, overall and HR+ patient model) and categorical TTP.
cy/z = number of patients with the BOM state/number of subset patients analyzed.
Interestingly, a relationship between TTP and the BOM state was observed. Compared to TTP <12 months, the following time intervals were associated with the BOM state, with corresponding ORs and p values: TTP <24 months (OR 0.361, p = 0.118), TTP <36 months (OR 0.089, p = 0.001), TTP <48 months (OR 0.185, p = 0.020), and TTP ≥48 months (OR 0.047, p < 0.001). Although the classification of TTP was somewhat arbitrary, it appears that when TTP was ≥24 months, the likelihood of BOM decreased by 81.1%∼95.3% compared to the TTP was <24 months. Finally, no other variables, including age, T and N stage, tumor grade, lymphovascular invasion, chemotherapy, or RT, were found to be associated with the BOM state (all p > 0.050).
Discussion
BOM is often observed in breast cancer patients with distant metastasis [5–7]. Compared to patients with metastases at other sites or organs, those with BOM tend to experience longer tumor control and have better survival [7]. For the study purpose, BOM patients can be categorized into three clinical subtypes: de novo, incident, and prevalent BOM. De novo BOM is typically defined as BOM diagnosed within the first 4 months following primary breast cancer diagnosis [16]. Incident BOM refers to the formation of BOM in patients with breast cancer who have not previously experienced organ metastasis. Prevalent BOM refers to the BOM observed in patients with existing metastatic breast cancer over a period of time. For epidemiological studies, a cohort of incident BOM patients is ideal for investigating factors related to the development of BOM, particularly among BM patients experiencing their first relapse of any type [26]. Identifying the clinicopathological factors related to the BOM state among first relapsed BM patients can help refine our understanding of the distinct nature of BM and inform tailored strategies for screening, prevention, and treatment of BM in breast cancer patients.
In this retrospective study, we consecutively identified 231 breast cancer patients relapsing with BM over an 11-year period at our institution for their first relapses of any type. Due to the delayed accessibility and high costs of anti-HER2-targeted therapy in China, the lack of widespread access to anti-HER2 therapy created a natural historical scenario allowing us to investigate the factors contributing to the development of BM in the absence of HER2-targeted therapy. In our cohort, only three HER2+ breast cancer patients (3.9%, 3/76) had received anti-HER2-targeted therapy in the adjuvant setting for their primary tumor, which led us to eventually and intentionally exclude these patients and to avoid the potential confounding or interaction effects introduced from this type of treatment. Results generated from the remaining 228 patients indicated that compared to HER2− patients, HER2+ patients were significantly less likely to develop BOM versus other forms of BM particularly in the HR+ subset. BOM appeared to emerge with a lower likelihood after 24 months. Clinicopathologic factors reported in the literature, such as age at diagnosis, T stage, N stage, pathology, tumor grade, lymphovascular invasion, chemotherapy, and RT were not found to be significantly associated with BOM state in these incident BM patients. Based on these findings, we believe that the combination of inherent HER2 and HR status plays a key role in distinguishing the development of BOM from other BM types at the first relapse of breast cancer. Therefore, in our clinical view, to detect BM and other distant metastasis timely and efficiently, bone scan is the preferably initial imaging for HER2− patients while other imaging (CT, MRI, PET) may be more applicable for HER2+ patients metastasis evaluation given the sensitivity and specificity features of these imaging for BM.
Most published studies on BOM in breast cancer patients focused on characterizing patient demographics, tumor characteristics, and prognostic factors associated with BOM [16, 26, 27]. Ahn et al. [26] investigated the prognostic factors of 91 Korea breast cancer women with incident BOM finding that their median age was 43 years at breast cancer diagnosis, and 47 years at the time of BOM; among these patients, 69% had ER+ tumors, 71% had a TTP of ≥2 years, and 30% presented with single bone lesion. In a 2018 study, Parkes et al. [16] characterized 1,445 breast cancer patients followed at MD Anderson Cancer Center from 1997 to 2015 for at least 6 months who presented BOM as their first metastases; among the 1,048 BOM patients with available tumor subtype data, the median age at breast cancer diagnosis was 50 years, with 73.5% of patients being White. The study also reported that 19.2% of patients had single bone lesion, 89.1% had axial bone involvement, and 49.7% exhibited lytic bone lesions [16]. The most common subtype was HR+/HER2− (78.2%), followed by HR+/HER2+ (11.4%), HR−/HER2− (7.3%), and HR−/HER2+ (3.1%); in the study, HR+ was defined as ER or PR ≥10% [16]. Although this study benefited from a large sample size, it is worth noting that de novo BOM patients accounted for 36.5% of the cohort, which was quite higher than ones in other studies [16]. More recently, Siregar et al. [27] analyzed 65 Indonesian women with prevalent BOM finding a mean age of 44.3 years at breast cancer diagnosis, with 58.5% diagnosed with invasive ductal carcinoma, 52.3% ER+, 50.8% PR+, and 40% HER2+; the most common bone sites were ribs (69.2%), femur (58.5%), vertebrae (36.9%) and pelvis (23.1%) [27]. In our review, it is truly challenging to compare these results with ours due to the dramatic differences of BOM subtype, primary breast cancer stage, treatment, and imaging tools presented in these studies.
Despite the clinical significance of BOM in breast cancer, this area remains understudied with only a few published studies investigating the factors related to the BOM state in either incident or prevalent metastatic breast cancer patients [3, 5, 9]. Park et al. [5] analyzed the patterns of first distant metastasis sites by breast cancer subtypes in a cohort of 313 Korean women who underwent breast surgery (mastectomy 95.2%) between 1994 and 2000 with a median follow-up of 93 months. Overall, 22.4% (70/313) patients developed first distant metastasis, and 9.6% (30/313) patients had incident BOM. The BOM proportions by subtypes were 10.1% (19/175, HR+/HER2−), 17.6% (6/34, HR+/HER2+), 7.1% (3/42, HR−/HER2+), and 3.7% (2/62, HR−/HER2−), respectively [5]. Among these 70 breast cancer patients with the first distant metastasis, the BOM was the most common metastasis pattern (42.9%, 30/70), compared to other patterns – extra-bone (excluding brain), bone and extra-bone (excluding brain), brain, brain and another organ site metastases; a significant relationship was found between these five patterns and breast cancer subtype (p = 0.04) [5]. In a prospective, multicenter, population-based cohort study in Germany, Schroder et al. [3] analyzed the metastatic site patterns in 1,094 breast cancer patients with prevalent BM at the beginning of palliative first-line therapy. The BOM proportion (35.3%, overall) was significantly higher in HR+ patients than in HR− breast cancer patients (36% vs. 20%, p < 0.05) [3]. However, neither Parks et al. [5] nor Schroder et al. [3] conducted multivariate analysis on the relationship between BOM and subtype or HR status. Diessner et al. [9] investigated a subgroup of 226 BOM patients with the first metastasis identified from a large retrospective German multicenter study (n = 9,625) recruited between 1992 and 2008. The proportion of BOM patients was significantly higher in luminal A (specifically defined as HR+HER2− and low tumor grade) or B/HER2− (defined as HR+HER2− and high tumor grade) or B/HER2+ (HR+HER2+) subtype patients than one in triple negative breast cancer (HR− HER2−) or HER2 overexpressing (HR− HER2+) patients (29.9% [202/676] vs. 11.4% [24/210], p < 0.0001). Five different mathematical models confirmed this correlation [9]. Multivariate models showed that breast cancer subtypes had the strongest influence on BOM development, followed by patient age at primary breast cancer diagnosis (average age of BOM patients vs. other patients: 65 vs. 60 years, p = 0.03). Primary tumor stage, histological subtype, tumor size, number of affected lymph nodes, grading, and the Nottingham Prognostic Index seemed to only minorly influence on the development of BOM [9]. These findings are consistent with our study findings. Although it is hard to make definitive conclusions from these multiple studies including our own, there seems to be a relationship between breast cancer subtype or biomarker (HR or HER2 status) and formation of BOM.
The effects of systematic treatments (e.g., ET, chemotherapy, anti-HER2-targeted therapy) and RT on the development of BM or BOM in breast cancer have been understudied [1, 9]. Liede et al. [1] analyzed a prospective cohort of 2,097 Canadian women with early breast cancer diagnosed between 1987 and 2000 with a median follow-up of 12.5 years. The cumulative incident rate of BM as the first recurrence was 6.5%, 10.3%, and 11.3% respectively at 5, 10 and 15 years. Multivariate analysis showed that tamoxifen (used by 47.8% of patients) was associated with a trend toward a lower risk of first BM (HR 0.74, 95% CI: 0.53–1.03, p = 0.07), while chemotherapy (HR 0.84, p = 0.44) and RT (HR 1.03, p = 0.87) were not significantly associated with reduced BM risk [1]. However, no analysis was conducted on risk factors for the development of BOM [1]. The biological and therefore statistical interactions of ET drugs and anti-HER2-targeted agents with breast cancer biomarkers could complicate the assessment of their independent effects on BM or BOM development [11]. In our study, we did not find the ET compliance status to be related to the BOM state overall (p = 0.545) or in HR+ patients (p = 0.716). Given that all HER2+ patients in our analyzed cohorts lacked the anti-HER2-targeted treatment, however, our analysis did indicate that HER2+ status was significantly associated with a lower likelihood of BOM. Since only a small number of patients were excluded due to the use of anti-HER2-targeted therapies, we are inclined to conclude that it is the HER2+ status itself, rather than treatment exposure, that is associated with a higher risk of extraosseous metastasis in breast cancer patients with high risk of BM. While other clinicopathological factors may play a role in BM development, they appeared to have little influence on the formation of the BOM state.
This retrospective study has several strengths. First, we excluded breast cancer patients with de novo BOM, which is often considered a distinct clinical entity with poorer prognosis compared to patients with incident BOM [26, 28]. Second, the study cohort was identified from a single cancer center, where breast cancer treatment protocols were consistently executed. None of the patients in this study received bone-modifying agents, ovarian function suppression procedures or drugs, or novel targeted therapies such as mTOR inhibitors, CDK 4/6 inhibitors, or immunotherapeutic agents prior to the development of their first BM. These agents have been shown in some studies to delay BM progression [12, 13]. Third, only 3.8% (3/79) of HER2+ breast cancer patients in our cohort received anti-HER2-targeted therapy in the adjuvant setting. Any selection biases from excluding these patients are likely to be minimal. Fourth, the compliance status of ET in HR+ patients was evaluated in multivariate analyses to assess its relationship with the BOM state. Fifth, we chose to use logistic regression models rather than time-dependent Cox model due to the potential for inaccuracies in the timing of BM development in this small retrospective cohort. Lastly, stratified analyses by HR status were conducted to provide a clearer understanding of the relationship between HER2 status and the BOM state.
This study also has limitations. First, the retrospective and cross-sectional nature of the study and relatively small size of BM patients may limit the generation of the study conclusion about cause-effect relationships. Second, BM and other organ metastasis had mostly not been diagnosed by the standard set of imaging tools and have not been pathologically verified by surgery or biopsy either. Third, the FISH test was not performed for many breast cancer patients with HER2 IHC score of 2+ tumor, although the “undetermined” status of HER2 was assigned in the analysis. Fourth, the effect of intrinsic subtype (i.e., combination of HR & HER2 status) and Ki-67 (mostly unavailable for our patient cohort) was not analyzed given the concerns about their significant interactions with ET compliance status and absence of anti-HER2-targeted therapy in statistical models. Last, we excluded 5.4% (35/649) relapsed patients with breast-conserving surgery history at 1 step of the patient selection process (Fig. 1). The inclusion of these patients could bring uncertain selection biases, given that such surgery was then highly selective in China. In our view, these limitations should have little impact on the study conclusion.
Clinical Implications
The HR and HER2 status in breast cancer are among the most influential factors in the early surveillance of breast cancer, particularly in relation to the development of bone and other organ metastasis. This study found that HER2+ status is associated with a lower chance of the incidental BOM state, especially after 24 months. These findings highlight the need for further investigation into the biological links between potential biomarkers and the development of BOM. The lower likelihood of BOM after 24 months among BM patients may assist clinicians in tailoring the timing and selection of imaging tools to monitor the progression of BM and other DM.
Conclusion
Differently from HR status and other clinicopathological factors, the HER2+ status is associated with a lower chance of the BOM state in breast cancer patients with first BM. Such association appears to be reflected in HR+ patients only. The underlying biological mechanisms and potential clinical applications of this finding for screening and preventing BM warrant further investigation.
Statement of Ethics
This study was approved by the Institution Review Board of the Fourth Hospital of Hebei Medical University (protocol # 2020-190). All study patients provided the written informed consent. All study methods were performed in accordance with the relevant guidelines and regulations of the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Conflict of Interest Statement
The authors declared that they have no competing interests.
Funding Sources
This study was not supported by any funding.
Author Contributions
Design and supervision of study and manuscript writing as the principal investigators: Z.L. and Y.L. Data collection, data analysis, and analysis discussion: L.C., H.H., Y.S., and Y.L. Data quality control and data interpretation: L.C., Y.S., and Z.J. Analysis discussion and revision of the manuscript: Z.J. All authors have read and approved the final manuscript.
Funding Statement
This study was not supported by any funding.
Data Availability Statement
The datasets generated and analyzed during the study are not publicly available due to the confidentiality agreement and research data policy of the Fourth Hospital of Hebei Medical University. They are available from the corresponding authors on reasonable request.
References
- 1. Liede A, Jerzak KJ, Hernandez RK, Wade SW, Sun P, Narod SA. The incidence of bone metastasis after early-stage breast cancer in Canada. Breast Cancer Res Treat. 2016;156(3):587–95. [DOI] [PubMed] [Google Scholar]
- 2. Coleman RE, Croucher PI, Padhani AR, Clézardin P, Chow E, Fallon M, et al. Bone metastases. Nat Rev Dis Primers. 2020;6(1):83. [DOI] [PubMed] [Google Scholar]
- 3. Schröder J, Fietz T, Köhler A, Petersen V, Tesch H, Spring L, et al. Treatment and pattern of bone metastases in 1094 patients with advanced breast cancer: results from the prospective German Tumour Registry Breast Cancer cohort study. Eur J Cancer. 2017;79:139–48. [DOI] [PubMed] [Google Scholar]
- 4. Li X, Zhang X, Liu J, Shen Y. Prognostic factors and survival according to tumour subtype in women presenting with breast cancer bone metastases at initial diagnosis: a SEER-based study. BMC Cancer. 2020;20(1):1102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Park HS, Kim S, Kim K, Yoo H, Chae BJ, Bae JS, et al. Pattern of distant recurrence according to the molecular subtypes in Korean women with breast cancer. World J Surg Oncol. 2012;10(1):4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kast K, Link T, Friedrich K, Petzold A, Niedostatek A, Schoffer O, et al. Impact of breast cancer subtypes and patterns of metastasis on outcome. Breast Cancer Res Treat. 2015;150(3):621–9. [DOI] [PubMed] [Google Scholar]
- 7. Parkes A, Warneke CL, Clifton K, Al-Awadhi A, Oke O, Pestana RC, et al. Prognostic factors in patients with metastatic breast cancer with bone-only metastases. Oncologist. 2018;23(11):1282–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Salvador F, Llorente A, Gomis RR. From latency to overt bone metastasis in breast cancer: potential for treatment and prevention. J Pathol. 2019;249(1):6–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Diessner J, Wischnewsky M, Stüber T, Stein R, Krockenberger M, Häusler S, et al. Evaluation of clinical parameters influencing the development of bone metastasis in breast cancer. BMC Cancer. 2016;16(1):307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Body J, Quinn G, Talbot S, Booth E, Demonty G, Taylor A, et al. Systematic review and meta-analysis on the proportion of patients with breast cancer who develop bone metastases. Crit Rev Oncol Hematol. 2017;115:67–80. [DOI] [PubMed] [Google Scholar]
- 11. Othman A, Winogradzki M, Lee L, Tandon M, Blank A, Pratap J. Bone metastatic breast cancer: advances in cell signaling and autophagy related mechanisms. Cancers. 2021;13(17):4310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Coleman R, Cameron D, Dodwell D, Bell R, Wilson C, Rathbone E, et al. Adjuvant zoledronic acid in patients with early breast cancer: final efficacy analysis of the AZURE (BIG 01/04) randomised open-label phase 3 trial. Lancet Oncol. 2014;15(9):997–1006. [DOI] [PubMed] [Google Scholar]
- 13. Early Breast Cancer Trialists’ Collaborative Group EBCTCG . Adjuvant bisphosphonate treatment in early breast cancer: meta-analyses of individual patient data from randomised trials. Lancet. 2015;386(10001):1353–61. [DOI] [PubMed] [Google Scholar]
- 14. Coleman R, Hadji P, Body J, Santini D, Chow E, Terpos E, et al. Bone health in cancer: ESMO clinical practice guidelines. Ann Oncol. 2020;31(12):1650–63. [DOI] [PubMed] [Google Scholar]
- 15. Pulido C, Vendrell I, Ferreira AR, Casimiro S, Mansinho A, Alho I, et al. Bone metastasis risk factors in breast cancer. Ecancermedicalscience. 2017;11:715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Parkes A, Clifton K, Al-Awadhi A, Oke O, Warneke CL, Litton JK, et al. Characterization of bone only metastasis patients with respect to tumor subtypes. npj Breast Cancer. 2018;4(1):2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Gradishar WJ, Moran MS, Abraham J, Aft R, Agnese D, Allison KH, et al. Breast cancer, version 3.2022, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2022;20(6):691–722. [DOI] [PubMed] [Google Scholar]
- 18. Gennari A, André F, Barrios C, Cortés J, de Azambuja E, DeMichele A, et al. ESMO Clinical Practice Guideline for the diagnosis, staging and treatment of patients with metastatic breast cancer. Ann Oncol. 2021;32(12):1475–95. [DOI] [PubMed] [Google Scholar]
- 19. Cardoso F, Kyriakides S, Ohno S, Penault-Llorca F, Poortmans P, Rubio I, et al. Early breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2019;30(8):1194–220. [DOI] [PubMed] [Google Scholar]
- 20. Soni A, Ren Z, Hameed O, Chanda D, Morgan CJ, Siegal GP, et al. Breast cancer subtypes predispose the site of distant metastases. Am J Clin Pathol. 2015;143(4):471–8. [DOI] [PubMed] [Google Scholar]
- 21. Jiang X, Chen G, Sun L, Liu C, Zhang Y, Liu M, et al. Characteristics and survival in bone metastatic breast cancer patients with different hormone receptor status: a population-based cohort study. Front Oncol. 2022;12:977226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Kennecke H, Yerushalmi R, Woods R, Cheang MCU, Voduc D, Speers CH, et al. Metastatic behavior of breast cancer subtypes. J Clin Orthod. 2010;28(20):3271–7. [DOI] [PubMed] [Google Scholar]
- 23. Hess KR, Esteva FJ. Effect of HER2 status on distant recurrence in early stage breast cancer. Breast Cancer Res Treat. 2013;137(2):449–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Fan L, Strasser-Weippl K, Li J, St Louis J, Finkelstein DM, Yu K, et al. Breast cancer in China. Lancet Oncol. 2014;15(7):e279–89. [DOI] [PubMed] [Google Scholar]
- 25. Liao N. HER2-positive breast cancer, how far away from the cure?-on the current situation of anti-HER2 therapy in breast cancer treatment and survival of patients. Chin Clin Oncol. 2016;5(3):41. [DOI] [PubMed] [Google Scholar]
- 26. Ahn SG, Lee HM, Cho S, Lee SA, Hwang SH, Jeong J, et al. Prognostic factors for patients with bone-only metastasis in breast cancer. Yonsei Med J. 2013;54(5):1168–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Siregar KB, Al Anas M. Unveiling bone metastasis: exploring histological subtypes of breast cancer in Indonesia's tertiary referral hospital. Cancer Treat Res Commun. 2023;37:100764. [DOI] [PubMed] [Google Scholar]
- 28. Marie L, Braik D, Abdel-Razeq N, Abu-Fares H, Al-Thunaibat A, Abdel-Razeq H. Clinical characteristics, prognostic factors and treatment outcomes of patients with bone-only metastatic breast cancer. Cancer Manag Res. 2022;14:2519–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and analyzed during the study are not publicly available due to the confidentiality agreement and research data policy of the Fourth Hospital of Hebei Medical University. They are available from the corresponding authors on reasonable request.
