Clinicopathologic characteristics, metastasis-free survival, and skeletal-related events in 628 patients with skeletal metastases in a tertiary orthopedic and trauma center

Background Skeletal-related events (SREs) due to bone metastases (BM) significantly impact the morbidity and mortality of cancer patients. The present study sought to investigate clinicopathological characteristics, metastasis-free survival (MFS), and SREs in patients referred to a tertiary orthopedic and trauma center. Methods Data were retrieved from electronic health records (n=628). Survival curves were estimated utilizing the Kaplan–Meier method. The Cox regression model was used to determine factors influencing MFS based on estimated hazard ratios (HRs). Results Breast (55.8%) and lung (18.2%), and lung (32.9%) and prostate (16.8%) cancer were the most common cancer types in our cohort in women and men, respectively. Fifteen percent of patients presented with BM as the first manifestation of tumor disease, 23% had metastasis diagnosis on the same day of primary tumor diagnosis or within 3 months, and 62% developed BM at least 3 months after primary tumor diagnosis. Osteolytic BM were predominant (72.3%) and most commonly affecting the spine (23%). Overall median MFS was 45 months (32 (men) vs. 53 (women) months). MFS was shortest in the lung (median 15 months, 95% CI 8.05–19) and longest in breast cancer (median 82 months, 95% CI 65.29–94). Age (≥ 60 vs. < 60 years) and primary cancer grading of ≥2 vs. 1 revealed prognostic relevance. Conclusion Women with breast or lung cancer, men with lung or prostate cancer, age ≥60 years, male sex, and primary cancer grading ≥2 are associated with increased risk for MBD. Intensified follow-up programs may reduce the risk of SREs and associated morbidity and mortality.

Due to significant improvements in the diagnosis and therapy of many malignancies, the survival time of tumor patients has steadily increased in the last years and will much likely advance [7,8]. Thus, MBD can be a chronic condition, as most of the patients with bone metastases are incurable [7,9,10]. Bone metastases (BM) often remain asymptomatic for a long time or present with non-specific symptoms until complications such as pathological fractures, e.g., occur. Consequently, early detection, as well as an interdisciplinary stage-adjusted therapy of bone metastases, has recently gained attention to avoid complications like SREs. SREs are associated with increased morbidity, decreased function in activities of daily living (ADLs), overall decreased quality of life, and, finally, mortality [3,11].
Profound knowledge of specific tumor characteristics and prognostic factors for predictive risk stratification models may help prevent severe SREs and reduce subsequent morbidity and mortality. Therefore, the present study sought to determine clinicopathologic characteristics, metastasis-free survival (MFS), and skeletal-related events of patients who were referred to our orthopedic oncology outpatient clinic due to osseous metastases.

Ethical considerations
The study was approved by the local institutional review board (protocol number 224/16) at the University Medical Center Freiburg, Freiburg, Germany. All studies were performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Study population
We conducted a single-center retrospective cohort study of 628 patients at our university medical tertiary orthopedic and trauma center and comprehensive cancer center, respectively. The main inclusion criteria were skeletal metastases as the cause of presentation. Patients with lymphoma and/or multiple myeloma were excluded. Before attending the oncologic, orthopedic outpatient clinic, patients underwent a standardized diagnostic algorithm according to current guidelines for the diagnosis and treatment of the respective tumor entity. In patients who had metastases as the first site of diagnosis, these diagnostic algorithms were performed after BM diagnosis.

Data management
To assess clinical parameters, oncologic characteristics (histopathology, tumor grading, the time point of diagnosis of the primary tumor and associated bone metastases (BM), and localization of osseous lesions) and SREs were extracted from our local hospital information system. Patients were assigned into three groups depending on the occurrence of BM to examine differences in the frequency of clinicopathologic characteristics and SREs: (1) patients with detection of metastases before primary cancer diagnosis ("primary bone metastases," PBM group), (2) patients with the diagnosis of metastases at the same time or within 3 months after the primary cancer diagnosis ("primary cancer with bone metastases," PSBM group), and (3) patients with bone metastases > 3 months after the primary cancer diagnosis ("bone metastases during follow-up," FBM group). Furthermore, BM were stratified by localization as well as osteolytic, osteoblastic, or mixed osteolytic-osteoblastic characteristics, respectively. The localization of osseous lesions was separated into six groups for statistical analysis: spine, spine and pelvis and others, spine and others, pelvis, pelvis and others, others (upper extremity, lower extremity, skull, rib, sternum, clavicle, scapula, multiple localizations, or combinations of the aforementioned localizations). SREs were divided into the following four groups for statistical analysis: (1) fracture with neurological complications, (2) fracture without neurological complications, (3) neurological complications without fracture, (4) no complications. The majority of malignant primary tumors were graded according to the Union International Contre le Cancer (UICC) classification [12]. For histologic grading of prostate cancer, Gleason Score was used [13]. A Gleason score ≤ 6 was defined as low-grade cancer, a Gleason score of 7 as medium-grade cancer, and a Gleason score ≥ 8 was defined as high-grade cancer [14]. Three groups were established to examine the prognostic relevance of primary cancer grading scores with regards to MFS: tumor grading score 1 (patients with a UICC grade 1 or a Gleason score of ≤ 6), tumor grading score 2 (patients with a UICC grade 2 or a Gleason score of 7), and tumor grading score 3 (patients with a UICC grade of 3 or a Gleason score of ≥ 8). Patients with melanoma were excluded from this examination as the grading data did not allow an assignment to either of these groups. Treatment data were extracted to analyze the distribution of the following therapies for descriptive statistics: conservative treatment (e.g., orthotic braces or prolonged bed rest), surgery, radiotherapy (RTx), chemotherapy (CTx), and the combination of the aforementioned. Subsequently, three groups were built: patients who underwent surgery, local radiotherapy, and chemotherapy, respectively; patients who underwent only surgery, or patients who did not undergo any surgery (nonsurgical group).

Statistical analysis
Descriptive statistics was performed to assess and compare distributions of clinicopathological factors. Median is shown with its 95% confidence interval (95% CI). To describe the boxplots, the median is shown with its interquartile range (IQR). Kaplan-Meier curves and log-rank tests were used for metastasis-free survival analysis in the FBM group. Multivariate analysis was performed to investigate the relationship between MFS and sex, age, the grade of the primary tumor, and the primary cancer types in the FBM population. A non-proportionality test based on Schoenfeld residuals was conducted to check for proportional hazards assumption. If the assumption was fulfilled, the Cox proportional hazard regression model was used to calculate the hazard ratios (HRs) and its 95% confidence interval. A two-sided p-value < 0.05 was considered to be statistically significant. Statistical analysis was performed using Stata Statistical Software Release 15 (StataCorp. 2011, College Station, TX, USA).

Basic demographics and clinicopathological characteristics
The present study population comprised 330 females (52.6%) and 298 males (47.5%) (   (Fig. 2a). The distribution of SREs dependent on the time point of BM manifestation was similar across the study groups (Fig.  2). However, complications resulting in neurological deficits were slightly more frequent in the PBM group (

Discussion
The diagnosis and treatment of bone metastases are challenging due to different time points of manifestation and possible skeletal complications. They have a significant impact on patients' morbidity and mortality.
Hence, thorough knowledge of specific tumor characteristics helps to prevent severe SREs. The present study sought to determine clinicopathologic characteristics, metastasis-free survival (MFS), and SREs of patients who were referred to our orthopedic oncology outpatient clinic at a tertiary center at a comprehensive cancer center. The outcome of the present study reveals that women having breast and lung cancer and men having lung and prostate cancer are at increased risk for MBD and associated SREs. Overall, more than half of the patients had SREs and fractures, respectively, due to BM. Furthermore, our findings suggest that male sex, patients over 60 years of age, and higher primary cancer grading ≥ 2 are predictive factors that are associated with a poorer bone MFS. Most commonly, patients received local radiotherapy and chemotherapy, often combined with adjuvant antiresorptive therapy. Surgery in combination with local radiotherapy and chemotherapy was conducted in nearly half of all patients (266/ 628, 42.4%), emphasizing the relevance of a multidisciplinary treatment approach.

Incidence, type, and localization of bone metastases
The reported overall incidence proportion of BM varies among different study cohorts due to the different follow-up times and methodological heterogeneity. It is reported to be 65-75% for breast, 65-75% for prostate, 20-40% for lung, 20-25% for renal, and 2.9-23% for gastrointestinal cancer [15][16][17][18][19][20][21][22]. In a study of 382,733 cancer patients, Hernandez et al. showed that mean time to BM was 400 days (1.1 years), and cumulative  Table 2. Harries and Jensen et al. calculated incidence rates instead of cumulative incidences and reported incidence rates of 22 per 1000 person-years, and 172.5 per 1000 person-years, respectively, for BM in breast cancer patients [28,29]. The incidence rate for breast cancer in our FBM population was 136.96 per 1000 person-years, which is in accordance with the aforementioned. Incidence rates have the advantage of using person-time for each subject and therefore taking into account whenever an event occurs. However, most studies report cumulative incidences that do not account for lost to follow-ups and are therefore less accurate compared to incidence rates, especially for long follow-up time [30]. Nevertheless, our retrospective observational analysis indicated that breast and lung cancer in women, as well as lung and prostate cancer in men, were the most common tumors leading to BM. The distribution patterns of primary tumors among the included patients is in accordance with recently published studies [16,23,31,32]. Osteolytic lesions were predominant in our study cohort, correlating with the results provided by Singh et al. [31]. In a study conducted by Yücel et al., osteoblastic BM lesions were reported to be predominant in patients with lung cancer, prostate, gastrointestinal, and head and neck cancer, whereas osteolytic lesions were predominant in patients with breast and genitourinary cancer [32]. While osteolytic BM lesions are considered to be the most frequent type of bone lesions among different tumor entities, osteoblastic metastases are predominantly found in prostate cancer, which is in accordance with our data [32,33].
Usually, the spine, ribs, pelvis, and proximal femur are the most common sites for skeletal metastases [34,35]. In our study, most patients had multiple skeletal lesions rather than solitary ones. As described in other sources, up to 20% of all cancer patients will develop symptomatic spinal metastases [36,37]. Equivalent to previous reports, the spine was the most frequently involved site of BM in our study [38,39].

Bone metastasis-free survival
To avoid under-or overtreatment of patients with BM and help set patient, family, and physician expectations, caregivers need accurate survival estimates. The present study indicated that MFS was different for primary cancer types. MFS was shortest in patients with lung and longest in patients with breast cancer. Numerous recent studies found low bone MFS intervals for lung cancer between 6 and 23 months, which is also in accordance with our data (median 14.5 months, 95% CI 8. [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]. In our study, one breast cancer patient developed BM approximately 31 years after the initial diagnosis of her primary tumor. This extraordinary case highlights the importance of sufficient follow-up visits to cancer patients. About 30% of women with early-stage breast cancer will experience recurrence of disease and develop metastases [40]. Besides regular follow-up visits, new onset of pain must be evaluated, and referral to an orthopedic oncologist should be considered in these cases. Moreover, sex (male vs. female), age (≥ 60-year-old vs. < 60-year-old), and higher cancer grade were associated Fig. 6 Box plot of metastasis-free survival (MFS) stratified by sex, primary cancer type, and metastases characteristics. The median MFS and its IQR are shown in the box plots grouped by primary cancer type, metastasis characteristics (mixed osteoblastic-osteolytic, osteoblastic, and osteolytic metastases), and sex (women (w) and men (m)). The dots represent outliers with poorer MFS in our study cohort. In a recent study, female sex was proposed to be associated with a lower risk of BM [41]. This prognostic factor was also found to be relevant by Zhang et al., who included these factors in a predictive scoring system [42]. Sex, primary tumor grading, and age were also found to be relevant risk predictors by other recent cohort studies [43,44].

Skeletal-related events
Bone metastases are associated with an increased risk of SREs [9,11,19,45]. SREs, including pathological fractures, pain, and spinal cord compression requiring palliative radiotherapy or surgery to bone, occur on average every 3 to 6 months, regardless of the primary tumor entity [15,46,47]. Permanent neurological injury Fig. 7 Cox proportional hazard regression model for MFS survival. The hazard ratios (HR) of the respective dependent variables against their references (dummy variables), the 95% confidence interval (95% CI), and the p-value adjusted for the included covariates (age, sex, grade, and primary tumor type) in the model are shown and disability can occur via direct tumor compression, bony retropulsion, or deformity due to pathological fracture, or impingement from osteoblastic bone response [6]. Osteolytic BM are more likely to cause fractures than osteoblastic BM lesions, which was also seen in our study cohort [1]. Moreover, 60% (377/628) of our study cohort had pathological fractures, from which 16.3% (62/380) resulted in neurological deficits. Another 5.1% (32/628) of patients suffered from neurological complications of any other cause due to MBD. In a prospective study of 319 patients diagnosed with cancer, the median time interval between back pain and the development of neurological symptoms was 66 days [48]. This study indicates that cancer patients with severe back or spinal nerve root pain need an urgent assessment based on their symptoms.
Predictive risk stratification and structured follow-up programs can, therefore, help to prevent devastating neurological complications resulting from MBD, as an overall of 94/628 (14.97%) patients in our study cohort experienced neurological deficits as a result of metastatic bone disease. Treatment options for BM and possible consecutive complications include surgery, chemotherapy, and radiotherapy [49]. Further, bisphosphonate and analgesic treatment are integral components of the overall treatment strategy [2]. A large part of the cohort examined in the present study received a combined treatment, including surgery, local radiotherapy, and chemotherapy, respectively (266/628, 42.4%), most commonly combined with adjuvant antiresorptive agents. Orthopedic surgery is not only essential for the treatment of pathological fractures or spinal cord compression. Increasing evidence suggests that prophylactic surgical treatment improves patient outcomes (protective of survival) as pathological fractures of the upper or lower extremity are associated with increased risk for mortality [50]. Hence, profound knowledge on the prognosis and SREs according to the primary cancer type is essential for a successful multidisciplinary therapeutic approach.

Limitations
The retrospective study design is associated with certain intrinsic limitations, such as selection bias due to the inclusion of a selected cohort (patients at a tertiary orthopedic and trauma center), loss of information, loss of follow-up, variation in imaging techniques, and limited sample size for some tumors. Notably, the present retrospective cohort study focused on complication and metastasis-free survival rates rather than investigating diagnostical or therapeutical approaches. We provided descriptive statistics for nonsurgical and surgical treatment groups in our cohort but did not focus on a more detailed examination of the treatments within these groups. Data regarding surgical management of bone metastases and bone-targeted agents for preventing SREs were sufficiently provided recently by two systematic reviews, involving 59 trials with 10,999 patients [51,52]. The summarized evidence concluded that patients benefited from the surgical management of bone metastases to the long bones and pelvis/acetabulum, whereas for the pharmacological therapy, denosumab and zoledronate were effective in preventing SREs as well as reducing the risk of pathological fractures and radiation compared with placebo. In contrast, epidemiological data are currently strongly warranted, as there are a limited number of studies focusing on the bone metastasis-free survival rates in primary tumor patients.

Conclusion
The present study indicates that the incidence and the time point of first bone metastases manifestation vary depending on the primary tumor. Women having breast and lung, as well as men having lung and prostate cancer, are at increased risk for MBD and associated SREs. The time interval of manifestation of BM was lowest in patients with lung cancer and highest in patients with breast cancer. Furthermore, male sex, age ≥ 60 years, and primary cancer grading score ≥ 2 are associated with a poorer bone metastasis-free survival. Intensified follow-up programs for selected cohorts may reduce the risk of SREs and maintain the best affordable quality of life. In conclusion, we call for more extensive prospective cohort studies to close the lack of knowledge regarding bone metastasis-free survival and its underlying influencing factors in cancer patients.