Health resource utilization associated with skeletal-related events in patients with advanced breast cancer: results from a prospective, multinational observational study

Breast cancer is the most common cancer in women, with nearly half a million new cases diagnosed in Europe in 2008 (Steliarova-Foucher et al.2012). Up to 75% of patients with advanced breast cancer develop bone metastases (Coleman1997), which are often associated with skeletal or bone complications, otherwise referred to as skeletal-related events (SREs) and commonly comprising pathologic fracture, radiation to bone, surgery to bone, and spinal cord compression (Coleman2001). The high frequency of bone metastases and associated complications in patients with breast cancer means SREs contribute significantly to the clinical and economic burden of the disease.

SREs can be debilitating and patients frequently experience a reduction in quality of life, with significant decreases in physical and functional well-being (Weinfurt et al.2004). SREs are associated with considerable morbidity, the worst being spinal cord compression, which can lead to paralysis, and pathologic fractures, which may result in disability or the need for surgery, with the potential for additional perioperative morbidity (Coleman2006; Katzer et al.2002). As might be expected, SREs often result in considerable pain, which further impairs patient mobility and reduces quality of life (Katzer et al.2002; DePuy et al.2007; Costa et al.2008). SREs are also associated with an increased risk of death (Norgaard et al.2010; Saad et al.2007).

In addition to the patient impact, SREs impose a substantial burden on healthcare resources. Planning future resource requirements and estimating the value of new treatment options requires data on the resource burden of SREs. However, there is a lack of prospective data on the impact of SREs on health resource utilization (HRU).

Several retrospective US studies have suggested that SREs increase HRU (Delea et al.2004,2006; Lage et al.2008; Schulman and Kohles2007), and studies in Spain, France, and Portugal have demonstrated that SREs increase HRU and costs (Pockett et al.2010; Decroisette et al.2011; Felix et al.2011; Svendsen et al.2013). However, these studies were restricted to individual countries and to our knowledge there have been no large, international, prospective studies investigating the differential contribution of the various types of SREs to HRU.

We therefore designed a prospective, observational, multinational study to evaluate the HRU associated with each type of SRE in patients with bone metastases or lesions secondary to breast, prostate or lung cancer, or multiple myeloma in Canada, Germany, Italy, Spain, the United Kingdom (UK) and the United States of America (USA). Here, we report the data for patients with advanced breast cancer from the four European countries (Germany, Italy, Spain, and the UK).

This is the first large, prospective study investigating HRU associated with different types of SRE in patients with breast cancer metastatic to bone in Europe. We found that all SREs were associated with substantial HRU, with lengthy hospitalizations frequently required. This is also the first study in which HRU has been assigned to specific SREs independently by the investigators. This ensures that HRU was recorded only when it was deemed by investigators’ expert opinion to be a direct consequence of the SRE, rather than owing to the underlying disease.

Patient characteristics at enrollment were broadly similar across the different countries. However, patients in the UK had a higher rate of SREs per patient-year than those from the three other European countries studied, which may reflect the longer median time from the detection of bone metastases to enrollment in the study in the UK (approximately double that seen for the other countries). Despite this, the ECOG performance status of patients in the UK showed a similar distribution to those from Germany and Italy, whereas Spain had a higher proportion of patients with a poorer ECOG performance status of 2.

In general, patterns of HRU were similar across the four European countries studied. The overall patterns of HRU observed in our study are similar to those reported in studies by Gunther et al. in Austria, the Czech Republic, Poland, Sweden, and Switzerland (Gunther et al.2011), by Felix et al. in Portugal (Felix et al.2011), and by Svendsen et al. in Denmark (Svendsen et al.2013). All of these studies found SREs were associated with lengthy inpatient stays, which were major contributors to the overall cost of care.

Inpatient stays were particularly common with spinal cord compression and surgery to bone. Although spinal cord compression was rare, it required the longest hospital stays. This is consistent with two separate economic analyses illustrating the high cost of each spinal cord compression event (Felix et al.2011; Barlev et al.2010). Notably, surgery to bone required shorter inpatient stays than pathologic fracture, probably because surgery facilitates faster stabilization of bone. Radiation to bone is one of the most common SREs in patients with advanced breast cancer (Lipton et al.2000). Despite requiring inpatient stays less frequently, this SRE was still associated with a considerable proportion of hospitalizations. This may have been owing to complications requiring an inpatient stay or patients being required to travel substantial distances for multifractionated radiotherapy.

Approximately three-quarters of SREs were associated with outpatient visits, with a mean of five visits per SRE. Therefore, outpatient visits may also impose a considerable burden on healthcare resources in Europe. In an analysis of the economic burden of bone metastases in the USA, outpatient visits were found to be the biggest contributor to incremental costs associated with bone metastases, accounting for 63–71% of the additional costs compared with patients without bone metastases (Schulman and Kohles2007).

Radiation to bone was associated with the greatest number of procedures per SRE, in addition to the highest number of outpatient visits. Our findings therefore suggest that radiation to bone is a major contributor to the increased burden that SREs place on healthcare systems, consistent with similar results in other studies (Lage et al.2008; Decroisette et al.2011; Felix et al.2011). The high numbers of outpatient visits and procedures observed in our study probably reflect the use of multiple fractions of radiotherapy. A study in patients with breast or prostate cancer with bone metastases in Portugal reported a mean of 8.4 sessions per treatment for radiation to bone (Felix et al.2011). In a retrospective study of palliative radiotherapy for bone metastases in Spain, only 25% of cases received single-fraction (800 cGy) treatment, with 59% receiving five- or ten-fraction regimens (Expósito et al.2012). The lower number of procedures and outpatient visits reported in the UK for radiation to bone is probably owing to a preference for single-fraction radiotherapy over multiple-fraction treatment. Surveys of radiotherapy for bone metastases in the UK reported a mean of about three fractions per treatment, with approximately 60% of treatments delivered as a single fraction (Williams et al.2007; Royal College of Radiologists2007). The length of inpatient stay for radiation to bone was also much shorter in the UK than elsewhere, which could again reflect the preference for single-fraction over multiple-fraction radiotherapy.

Our study had some limitations. As discussed in Hoefeler et al., the duration of follow-up for this study was shorter than planned (6.9–10.9 months) owing to slow recruitment (possibly because the trial was non-interventional) and early withdrawal from the study as a result of patient death (Hechmati et al.2011). The sample sizes for surgery to bone and spinal cord compression were limited and may not have been sufficient to provide a generalizable HRU estimation. Similarly, the applicability of estimations of the duration of inpatient stays may have been limited by the small sample sizes. Other limitations to the study included some data being inaccessible to investigators at all study sites. For example, information about home health visits or nursing home stays was not always relayed back to the main hospital, making it difficult to capture resource use associated with these events. It should also be noted that the incidence of SREs reported here is not representative of the real-world distribution of SRE types, because they are impacted by the index SRE recruitment cells and the inclusion criteria stating patients must have an ECOG performance status of 2 or less and a life expectancy of at least 6 months. Furthermore, pain was not defined as an SRE, although this common complication from poorly treated bone metastases could have led to substantial HRU and lengthy inpatient stays. These limitations would be expected to result in an underestimation of the overall HRU associated with SREs, and therefore the true burden of SREs may be even greater than suggested by these results.

These data indicate that preventing SREs in patients with bone metastases secondary to breast cancer would substantially reduce HRU. Bisphosphonates, such as zoledronic acid, have been widely used to reduce the occurrence of SREs. Denosumab, a monoclonal antibody to RANKL, has recently been approved for the prevention of SREs in patients with bone metastases from solid tumors (European Medicines Agency2012), and has been shown to be superior to zoledronic acid at reducing the incidence of SREs (Fizazi et al.2011; Henry et al.2011; Stopeck et al.2010).

In conclusion, all types of SRE arising from bone metastases in patients with breast cancer are associated with considerable HRU. Therefore, treatments that delay or prevent SREs may reduce the burden imposed on healthcare systems across Europe.