|Home | About | Journals | Submit | Contact Us | Français|
Bone-modifying agents (BMAs) reduce the incidence of skeletal-related events (SREs) and are thus recommended for breast cancer patients with bone metastases. However, the risk factors for SREs during BMA treatment are not well-understood. This study evaluated the number and timing of SREs from case studies to identify these factors.
The medical records of 534 women with breast cancer who developed bone metastases between 1999 and 2011 were reviewed. SREs were defined as a pathologic fracture, spinal cord compression, or the need for bone irradiation or surgery. Multiple variables were assessed and were analyzed by using the Cox proportional hazard analyses and the Andersen and Gill method.
Multivariate analyses for both the time to the first SRE and the primary and subsequent SRE frequency demonstrated that significant baseline risk factors included luminal B type disease, a history of palliative radiation therapy, BMA treatment within 2 years, and elevated serum calcium levels at the time of the initial BMA dose. Additionally, for the time to the first SRE and for the primary and subsequent SRE frequency, the presence of extraskeletal metastases and BMA administration initiation ≥6 months after the detection of bone metastases were also significant risk factors, respectively.
In breast cancer patients with bone metastases, more vigilant observation should be considered for patients with the identified risk factors. To reduce the risk for SRE, BMAs should be administered within 6 months of bone metastases diagnosis and before palliative radiation therapy.
Retrospectively, risk factors were identified for skeletal-related events (SREs) in breast cancer patients with bone metastasis who were treated with bone-modifying agents (BMAs). For the time to the first SRE and for the SRE frequency, presence of extraskeletal metastases and BMA initiation ≥6 months after the detection of bone metastases were risk factors, respectively. Luminal B type disease, a history of palliative radiation therapy, BMA treatment within 2 years, and elevated serum calcium levels at initial BMA dose were risk factors for both first SRE and SRE frequency. More vigilant observation should be considered for patients with these risk factors.
背景. 由于骨调节剂 （BMA） 可降低骨骼相关事件 （SRE）, 因此建议用于乳腺癌骨转移患者。但是 BMA 治疗期间发生 SRE 的危险 因素并未彻底明确。本研究评价了病例研究中 SRE 的发生次数和时机, 从而对这些因素进行识别。
方法. 我们对 1999-2011 年间 534 例发生骨转移的乳腺癌女性患者病史进行了回顾。SRE 定义为病理性骨折、脊髓压迫症, 或者需要骨照射或手术。使用 Cox 比例风险分析和 Andersen-Gill 法对多重变量进行评估与分析。
结果. 对发生首次 SRE 的时间和主要及后续 SRE 频率进行多变量分析, 证实了存在显著的基线危险因素, 包括 luminal B 型、姑息性放疗史、过去 2 年内曾接受 BMA 治疗, 以及给予首剂 BMA 治疗时血钙水平升高。此外, 就发生首次 SRE 的时间而言, 存在骨外转移也是显著危险因素, 而首次 BMA 治疗距发现骨转移≥6 个月是主要和后续 SRE 发生频率的显著危险因素。
结论. 在乳腺癌骨转移患者中, 对存在前述已确认危险因素的患者应进行更为审慎的观察。为降低SRE风险, 应在骨转移确诊后6个月内以及姑息性放疗之前开始BMA治疗。The Oncologist 2016;21:508–513
对临床实践的提示: 本研究回顾性地鉴别出了接受骨调节剂 （BMA） 治疗的乳腺癌骨转移患者发生骨骼相关事件 （SRE） 的危险因素。存在骨外转移是发生首次 SRE 时间的危险因素, 首次 BMA 治疗距发现骨转移≥6 个月是 SRE 发生频率的危险因素。Luminal B 型、姑息性放疗史、过去 2 年内接受过 BMA 治疗以及给予首剂 BMA 治疗时血钙水平升高均为首次 SRE 和 SRE 频率这两者的危险因素。对于存在这些危险因素的患者应该进行更为审慎的观察。
Breast cancer is the most prevalent malignancy among women worldwide, and its incidence continues to increase [1, 2]. Although mortality has been falling recently as a result of earlier diagnosis and treatment and improved therapies, breast cancer remains the leading cause of female cancer-related death worldwide [3, 4]. The median survival for a stage IV breast cancer patient is approximately 2 years . A reported 50%–60% of metastatic breast cancer patients had bone metastases at diagnosis or developed them subsequently [6, 7], and in more than 70% of patients postmortem examination showed bone metastases . These findings demonstrate that bone is the most common site for metastases from this malignancy. In patients with bone metastases, skeletal-related events (SREs), such as pathologic fractures, spinal cord compression, and the need for bone irradiation or surgery, remain a major cause of mortality and morbidity and reduce the patient’s quality of life [8–10].
In breast cancer, SREs occur in about half to two thirds of patients with bone metastases [11–13]. Therefore, treatments that reduce the incidence of SREs are potentially of great clinical benefit. One such approach is the use of bone-modifying agents (BMAs) . To delay the time to first and subsequent SREs, it is recommended that breast cancer patients with bone metastases be administered BMAs even if they are asymptomatic [15–17]. In addition, because a longer treatment duration with BMAs reduces the risk for SREs , recent guidelines from the American Society of Clinical Oncology (ASCO) recommend that BMA administration be continued until the patient’s general performance status declines substantially.
However, early clinical trials have evaluated BMA treatment only for up to 2 years, and thus the optimal duration and scheduling of BMAs remain undefined . Moreover, the risk factors for multiple SREs during a long period are not well-understood. Identifying these risk factors is important not only to determine the optimal duration of BMA administration and its scheduling but also to predict SREs, which would allow the early initiation of treatment and thus help to maintain a patient’s quality of life. Hence, the aim of the present study was to evaluate the number and timing of SREs to identify risk factors for these events.
We retrospectively reviewed the medical records of 677 women with bone metastases from breast cancer who had been administered BMAs between 1999 and 2011 at the National Cancer Center Hospital, Tokyo, Japan. SREs were defined as a pathologic fracture, spinal cord compression, or the need for bone irradiation or surgery. Findings from discharge summaries, physician progress notes, radionuclide and radiographic bone scans, surgical procedure notes, radiation treatment summaries or records, and pathology reports were followed to track the development of bone metastases or SREs.
Multiple variables were assessed, including patient demographic characteristics, tumor subtypes with respect to HER2 and hormone receptor status, current and previous chemotherapy and hormone treatments, hematologic measures at BMA initiation, Eastern Cooperative Oncology Group performance status when the bone metastases was diagnosed, and the appearance and location of skeletal and extraskeletal metastases.
Of 677 patients screened, 147 were excluded because they had hypercalcemia when BMA administration was started (n = 64), incomplete follow-up (n = 61), or insufficient records or data (n = 18). Consequently, the number of patients analyzed was 534 (Fig. 1). The institutional review board of the National Cancer Center Hospital approved the study protocol. All the participants gave informed consent before treatment initiation.
The frequencies and descriptive statistics of the demographics and clinical variables for 534 patients were obtained. There were two primary outcomes. The first was the time to the first SRE, defined as the time from the first bone metastases diagnosis to the first SRE that developed after the initiation of BMA treatment. The second was the frequency of subsequent SREs. The time to the first SRE was analyzed by using the Cox model, and the SRE frequency was analyzed by using the Andersen and Gill method . All the analyses were exploratory in nature. A two-sided p value <.05 was considered to represent a statistically significant difference. The analyses were performed by using SAS software, version 9.3 (SAS Institute Inc., Cary, NC, https://www.sas.com/).
The baseline patient characteristics are summarized in Table 1. The median duration of patient follow-up (date of primary bone metastases diagnosis to date of the last visit or death) was 35 months (range, 1–329 months), and the median time to the development of metastatic bone disease from the diagnoses of primary breast cancer was 39.5 months (range, 0–367 months). At the time of reporting, 375 of 534 (70%) patients had died, with an average duration between breast cancer diagnosis and death of 72 months (range, 4–435 months).
For the initial BMA dose, pamidronate was administered to 54% of patients and zoledronic acid to 46% of patients. Subsequently, 78 patients (15%) switched from pamidronate to zoledronic acid, but none from zoledronic acid to pamidronate. The median time from the first bone metastases diagnosis to the start of BMA administration was 3 months (range, 0–301 months). Three hundred twenty-eight patients (61%) began receiving a BMA within 6 months of the first bone metastases diagnosis, and 206 (39%) began treatment more than 6 months after the first diagnosis. The median duration of BMA administration was 15 months (range, 1–105 months; interquartile range, Q1 5.9 months, Q2 17.8 months, and Q3 31.4 months). At the time of reporting, 490 patients (92%) had discontinued BMAs as a result of deterioration or death (71%), a recommendation from their doctor (20%), side effects (4%), the end of follow-up (4%), or a decision made by the patient (1%).
Osteonecrosis of the jaw and renal failure, both of which are well-established side effects of BMA administration, developed in 10 patients (1.8%) and 1 patient (0.1%), respectively. The other reported side effects were a single case each of arthralgia, dental cavities, hot flushes, thrombocytopenia, anemia, and neuralgia and 2 cases each of fever and liver dysfunction.
The median time from bone metastasis diagnosis to the first SRE was 26 months (range, 1–323 months). During the entire study period, 196 patients (37%) experienced at least one new SRE. The median time from the first BMA dose to the first SRE following treatment initiation was 13.5 months (range, 1–130 months).
Multivariate Cox regression analysis indicated that significant baseline risk factors for the time to first SRE included luminal B type disease (as opposed to luminal A type disease; hazard ratio [HR], 1.7; p = .009), the presence of extraskeletal metastases (HR, 1.6; p = .04), a history of palliative radiation therapy (HR, 1.7; p = .0002), and elevated serum calcium (HR, 1.3; p = .001) at the time of the initial BMA dose. The time to the first SRE was significantly longer in patients who received BMA treatment for more than 2 years than in those who were treated for less than 2 years (HR, 0.62; p = .0024) (Table 2).
One hundred ninety-six patients experienced at least 1 SRE. Of these, 78 developed multiple SREs, of whom 46 developed 2 SREs, 20 developed 3 SREs, and 12 developed 4 SREs. In total, there were 449 SREs, the most frequent of which was need for radiotherapy (59%), followed by fractures (18%), spinal cord compression (15%), and surgery (8%) (Table 3).
The Andersen and Gill analysis indicated that significant baseline risk factors for an SRE included luminal B type disease rather than luminal A type disease (HR, 1.5; p = .029), a history of palliative radiation therapy (HR, 1.4; p = .02), elevated serum calcium at the time of the initial BMA dose (HR, 1.3; p < .0001), and BMA treatment ≥6 months after the detection of a bone metastases (HR, 1.4; p = .03). Furthermore, patients who underwent BMA treatment lasting more than 2 years had a significantly lower risk for SREs than those treated for less than 2 years (HR, 0.67; p = .0075) (Table 2).
To the best of our knowledge, the current study includes the most breast cancer patients with bone metastases treated with BMAs in a clinical setting. In the entire period studied, 37% of patients experienced at least one new SRE during a median follow-up period of 35 months. This was relatively low compared with that in previous studies, in which the SRE incidence in patients who did not receive BMA treatment was 68% at 24 months ; this finding suggests that BMA treatment reduces the incidence of SREs.
This rate of SRE was also similar to that reported in previous retrospective studies. In a retrospective chart review of 100 women with metastatic breast cancer, Liauw et al. found that only 30% of patients experienced one or more SREs within 12 months of starting therapy , and Trinkaus et al. reported that 38% of patients with bone metastases from breast cancer developed an SRE during a median follow-up period of 267 days . However, randomized controlled trials have reported higher rates of SREs, ranging between 50% and 60%, after initial BMA treatment [22, 23]. These discrepancies between clinical trials and retrospective studies of BMA treatment have two main causes. One is the population bias inherent in these clinical trials, which have generally included patients who seem to be more predisposed to experience an SRE. The other is the use of regular radiographic skeletal surveys in the trial setting, which, because they are not routinely used in clinical practice, may have resulted in the detection of clinically silent fractures and thus may have overestimated the clinical utility of BMA treatment for SREs. Thus, only clinically relevant skeletal events were likely to be identified, such as cases where patients were symptomatic from an SRE or were experiencing pain. Consequently, clinical trials may overestimate the frequency of SREs by including those that are clinically silent and thus not fully reflect the positive effect of BMAs in clinical practice.
Several clinical trials have demonstrated that BMA treatment can delay the onset and reduce the incidence of SREs, regardless of whether the patient has sustained a previous SRE. This indicates that early and continuous BMA treatment is beneficial [12, 13, 18]. On the basis of the findings of these clinical trials, an expert panel of the ASCO issued guidelines recommending that BMAs be used until there is evidence of a substantial decline in the patients’ general performance status, despite the occurrence of SREs in women with breast cancer bone metastases . However, to date, clinical trials have not yet determined the optimal start time, frequency, or duration of BMA administration.
In this study, we found that patients who initiated BMA treatment ≥6 months after diagnosis of the first bone metastases had a significantly greater risk for multiple SREs than those who were administered BMA within 6 months of diagnosis. This finding suggests that BMA initiation within 6 months of bone metastases diagnosis is optimal. In addition, patients with a history of palliative radiation therapy developed SREs significantly sooner than those who did not undergo this therapy, suggesting that starting BMA treatment before palliative radiation therapy becomes necessary would delay the onset of SREs. This finding is consistent with that of a previous study that analyzed data from clinical trials comparing BMAs with a placebo for patients with breast cancer bone metastases. The analysis revealed that patients with prior radiotherapy had an increased risk for SREs, which suggests that BMA treatment initiation before development of the first SRE may be beneficial . This concept is further supported by the observation that patients with extraskeletal metastases develop an SRE significantly sooner than those with bone-only metastases (HR, 1.6; p = .045). Moreover, the findings of the current study support the ASCO guidelines that recommended BMA initiation upon the first radiographic evidence of bone metastases, even for asymptomatic patients .
Understanding the risk factors for SRE development enables clinicians to better judge which patients would benefit most from BMA treatment. Moreover, the stratification of patients at high risk for SREs, and targeting those who will most benefit from BMA treatment, may optimize BMA use and reduce patient risk. The present study identified two patient groups at high risk for SREs. One consisted of patients with luminal B type breast cancer. Breast cancer is heterogeneous, and thus classification into subgroups may help in choosing the best treatment. However, this classification varies with clinical and pathological subset, and the classification of the tumor type from gene expression is often regarded as the gold standard. Previous studies have shown that luminal B type cancers have an aggressive clinical behavior and a distinctive pattern of dissemination, with a predilection for bone [25, 26]. In the current study, a subset of patients with luminal B type cancer exhibited a significantly shorter time to the first SRE and a greater risk for multiple SREs than those with luminal A type disease. In addition, patients with other subtypes, such as HER2-positive and triple-negative, did not show any significant difference compared with those with luminal A type disease. This suggests that luminal B type cancer patients may have a higher SRE risk; thus, it would be better for them to continue BMAs whenever possible to reduce the incidence of SREs and delay their development.
The other risk factor for SRE development is an elevated serum calcium level when BMA treatment is initiated. Elevated serum calcium in these patients might result from enhanced bone osteoclastic resorption stimulated by the release of tumor-derived mediators . Recent studies indicated that the risk for skeletal complications in breast cancer is strongly related to the rate of bone resorption and that the bone resorption marker N-telopeptide of type 1 collagen is a useful biomarker for identifying patients at high-risk of skeletal complications [8, 28]. However, owing to the retrospective nature of the current study, such biological markers could not be investigated. Thus, further studies are needed.
The findings of the present study suggest that the administration of BMAs for longer than the first 2 years after diagnosis may delay the development of the first SRE and reduce the risk for subsequent SREs (Table 2). These observations are consistent with the findings of randomized clinical trials indicating that continuous BMA treatment is beneficial despite the occurrence of a first SRE, even though few of these have reported the efficacy of BMA treatment beyond 2 years [12, 13, 18]. However, in interpreting our results, other factors that might also affect SRE development should be considered. For example, patients in good condition, with less aggressive disease, or who responded well to anticancer agents may have been treated with BMAs for longer. Despite this limitation, the large number of patients included and a BMA administration period of up to 105 months make our findings more meaningful than those of previous studies. Another possible limitation of the current study is its retrospective nature. However, despite this, it provides additional data regarding risk factors for long-term BMA use, derived from a large number of patient histories, and identifies factors that are important in predicting which patients are most likely to benefit from BMA treatment. Another study limitation is the exclusion of patients who underwent BMA treatment initially with denosumab. This is because denosumab was only approved in 2012. Thus, further studies are needed to clarify the effects of long-term denosumab treatment and evaluate the risk factors for patients with bone metastases from breast cancer.
This long-term, large-cohort, retrospective study showed that patients with bone metastases from breast cancer with characteristics such as luminal B type disease, a history of palliative radiation therapy, and elevated serum calcium levels when BMA administration was started are more likely to develop SREs sooner, and at a greater frequency. Therefore, more vigilant observation should be considered for these patients. In addition, patients with extraskeletal metastases were more likely to develop the first SRE sooner than those with bone only metastases. Moreover, BMA administration initiation ≥6 months after initial bone metastases detection may increase the frequency of SREs. Thus, SRE prevention by initiating BMAs before palliative radiation therapy, or within 6 months of initial bone metastases detection, appears to be important. However, patients receiving BMA treatment for more than 2 years are less susceptible to SREs. Therefore, continued BMA treatment after initial bone metastases detection is important to delay the onset of the first SRE and reduce the frequency of subsequent SREs.
This article is available for continuing medical education credit at CME.TheOncologist.com.
The authors are thankful to Editage for assistance in writing the manuscript.
Conception/Design: Kan Yonemori
Provision of study material or patients: Kan Yonemori
Collection and/or assembly of data: Ryota Tanaka, Kan Yonemori, Naoki Takahashi
Data analysis and interpretation: Kan Yonemori, Akihiro Hirakawa, Fumie Kinoshita
Manuscript writing: Ryota Tanaka, Kan Yonemori, Akihiro Hirakawa
Final approval of manuscript: Kan Yonemori, Akihiro Hirakawa, Jun Hashimoto, Makoto Kodaira, Harukaze Yamamoto, Mayu Yunokawa, Chikako Shimizu, Manabu Fujimoto, Yasuhiro Fujiwara, Kenji Tamura
Yasuhiro Fujiwara: AstraZeneca KK, Eisai, Daiichi Sankyo, Chugai, Eli Lilly Japan, Yakult Honsha (H), Taiho, Takeda, Chugai, Eli Lilly Japan, Nippon Kayaku (RF). The other authors indicated no financial relationships.
(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board