Bone Metastases: Epidemiology and Societal Effect

Prostate

Lung

Breast

Study years

1999–2007

1999–2010

1999–2007

Patients

23,087

29,720

35,912

Mets at diagnosis

569 (3 %)

254 (0.9 %)

178 (0.5 %)

Developed mets

2578 (11.5 %)

1692 (5.8 %)

1272 (3.6 %)

Developed SRE

1329 (5.9 %)

905 (3 %)

590 (1.6 %)

1-year survival

− no bone mets

− bone mets no SRE

− bone mets + SRE

87 %

47 %

40 %

37.4 %

12.1 %

5.1 %

93.3 %

59 %

40.2 %

Reference

Nørgaard [12]

Cetin [13]

Jensen [11] and Yong [14]

In lung cancer (most studies being of NSCLC), a review by Kuchuk reports an incidence at diagnosis of skeletal metastases of 20–40 % [15]. Bone-only metastases were present in less than 7 %. The presence of bone-predominant metastases did not improve survival. However, an SRE was not further detrimental to survival.

Skeletal metastases will typically present to trauma surgeons, orthopedic oncologic surgeons, oncologists, and surgical oncologists—the latter two usually because they are managing the primary tumor. Primary management should incorporate early orthopedic opinion and appropriate surgical and oncologic management. The use of conventional internal fixation may be inappropriate and as such surgical treatment should be planned and undertaken in daylight hours with experienced anesthetists and in conjunction and following discussions with the managing oncologists. Heroic operations in the face of a short life expectancy are usually unjustified. Similarly, ill-thought-out internal fixation in a patient with a reasonable life expectancy can result in implant failure. Surgery in the absence of radiotherapy may result in disease progression and can result in complex periprosthetic fractures. Revision surgery is always more challenging than primary surgery for both the patient and the surgeon (and often the anesthesiologist).

Many patients with skeletal metastases will have concomitant visceral metastases. This is commonest in lung, renal, and breast cancer. Solitary bone metastases occur most frequently in renal cancer. Most patients have multiple skeletal metastases [16] rather than solitary ones.

The incidence of patients with bone metastases having an SRE is high. In a large study of 1819 patients with newly diagnosed skeletal metastases in breast, prostate, or lung cancer, 22 % of patients had an SRE concomitant with diagnosis of the metastasis. Of those not presenting with an SRE, 46.8 % of lung cancer patients experienced an SRE during follow-up. The figure was 46.4 % for prostate cancer and 51.9 % for breast cancer [17]. This figure is higher than from other series but suggests that the risk of developing an SRE in any patient with a skeletal metastasis approached 1 in 2.

Site of Bone Metastases

Swanson et al. followed 947 patients with renal cell cancer from first diagnosis. 252 (26.7 %) developed skeletal metastases. The most common sites were spine, pelvis, and proximal femur [8]. A similar distribution was seen by Lipton [18] as most common sites of metastasis.

Kakhi et al. utilized isotope bone scanning to review the most common site for bone metastases in prostate, breast, gastrointestinal, and lung cancers. The spine, ribs, and pelvis were the most common sites affected in all of the cancers with the addition of the sternum in breast cancer. The most common appendicular bone was the femur, most commonly the proximal femur [19].

Incidence of Skeletal Related Complications

Bone metastases are a common cause of morbidity and skeletal events are common in patients. They are detrimental to quality of life. They result in admission to hospital (Table 1.2) and once the patient has been admitted the rate of admission increases [20].

Table 1.2

3-year incidence rates of hospital admission due to MBD and admission following a previous SRE in 28,162 patients with breast, prostate , and lung cancer

	3-year incidence rate of admission per 1000 patients	Previously admitted following SRE—rate of admission per 1000 patients
Breast cancer	95	211
Prostate cancer	163	150
Lung cancer	156	260

Data adapted from Pockett et al. [20]

The placebo wings of multicenter randomized trials give evidence as to the incidence of different types of SREs in patients with skeletal metastases (Table 1.3).

Table 1.3

Incidence of SREs from placebo wing of multicenter trials in advanced malignancy

	Breast	Prostate	NSCLC and other solid tumors	Myeloma
Pathological fracture (%)	52	25	22	37
Radiotherapy (%)	43	33	34	34
Surgery (%)	11	4	5	5
Spinal cord compression (%)	3	8	4	3
Reference	Lipton [21]	Saad [22]	Rosen [23]	Berenson [24]

Cancer Survival

Survival varies dependent on primary tumor pathology and visceral tumor load. Longer mean survivals are seen in thyroid (26 months), breast (19 months), and prostate cancer (18 months). Poorer mean survivals are a feature of lung cancer (6 months) and cancer of unknown primary. The presence of visceral metastases results in poorer survival rates [25].

In 1995 Bauer reported that after surgical treatment of skeletal metastases the 1-year survival was 30 % and the 3-year survival was 8 % [26]. Pathologic fracture, visceral or brain metastases, and lung cancer were negative prognostic variables for survival whereas solitary bone metastases, breast and kidney cancer, myeloma, and lymphoma were positive. In 2004, Hansen, on behalf of the Scandinavian Sarcoma Group (SSG) , reported 1-year survival of 40 % and a 3-year survival of 20 % [27]. In 2013, the SSG reported 1195 surgically treated non-spinal metastases. The 1-year survival was 41 % and the 5-year survival was 2 %. The longest median survival was in myeloma patients (26.3 months), thyroid cancer (22.7 months), breast cancer (12 months), and kidney cancer (10 months). Melanoma had the worst prognosis (2.3 months) [16].

Implications of Increasing Survival

Increasing survival of patients with bone metastases has a number of effects for the orthopedic surgeon treating the metastases:

Tumor that is not adequately treated (en bloc excision or surgery plus radiotherapy) will continue to grow resulting in some cases in extreme bone destruction or stresses being put on implants (Fig. 1.1).

Fig. 1.1

Seventy-six male with known diffuse large B-cell lymphoma sustained a pathological femoral neck fracture (a) treated by hemiarthroplasty (b). Adjuvant radiotherapy was not given resulting in bone loss around the implant (c). The hemiarthroplasty was converted to a proximal femoral replacement
- Fixation that is reliant on bone healing is likely to fail because of implant failure (Fig. 1.2) leading to more complex and more costly operations, prolonged inpatient stays, and increasing mortality.
  
  Fig. 1.2
  
  Male with multiple myeloma . Pathological fracture proximal femur (a) treated by long Affixus nail (Biomet) (b). The nail failed (c) and was revised to a proximal femoral replacement

Incidence of Pathological Fractures

The majority of the workload for metastatic bone disease for non-spinal metastases is for pathological fracture. The incidence of pathological fracture varies between different primary tumors. Tumors that tend to produce lytic metastases have a higher fracture rate than those that produce sclerotic metastases. Table 1.4 highlights some of the evidence for pathological fracture rate. The majority of evidence comes from the placebo wing of randomized controlled trials of the efficacy of bisphosphonate therapy.

Table 1.4

Pathological fracture rate based on longitudinal studies and placebo wing of bisphosphonate studies (solid tumor study was of non-breast and prostate metastatic malignancy—tumors included NSCLC (54 %), renal (10 %), small-cell lung cancer (8 %), thyroid (2 %), head and neck (2 %), cancer of unknown primary (7 %), and others (23 %))

Tumor type	Reference	Criteria	Pathological fracture rate
Breast cancer	Coleman [28]	Breast cancer with bone metastases	78/498 (16 %)
Prostate cancer	Saad [29]	Prostate cancer with bone metastases	46/208 (22.1 %)
Lung cancer	Joshi [30]	Lung cancer with bone metastases	21.6 %
Renal cancer	Lipton [18] Swanson [8] Forbes [31]	Renal cancer with bone metastases Newly diagnosed renal cell cancer	42 % 15 % 12 %
Other solid tumors (see description)	Rosen [23]	Bone metastases from non-breast/prostate cancers	55/250 (22 %)

Predicting Pathological Fracture

While this is covered elsewhere in the text, a pragmatic approach is recommended by the authors. If the patient has functional pain and a large lytic metastasis then prophylactic surgical stabilization should be considered.

Life expectancy is an important consideration in planning any surgical intervention in skeletal metastases. The Scandinavian Sarcoma Group proposed the following scoring system [16] (Table 1.5). A score of 0–1, the majority survive 12 months; a score of 2–3 six months; and a score of 4 is associated with a survival that may not reach 3 months.

Table 1.5

SSG life expectancy after bone metastases

Score	0	1
Number of metastases	Single	Multiple
Visceral metastases	None	Yes
Breast/thyroid/renal/myeloma	Yes	Other
Karnofsky score 70	Above (self-care)	Below (needs help)

Data from Ratasvuori [16]

In addition to the published literature issues such as patient weight, comorbidities, compliance, ability to bear weight, local and systemic pain, use of pain medication, use of bisphosphonates, concurrent chemotherapy, function both current and previous, specific concurrent bone sites of tumor involvement, overall disease load including non-bone lesions, response of other sites to nonsurgical oncologic treatment, activity level, patient and functional expectations, among others may be important [32].

Impact on Survival of Pathological Fractures

A pathological fracture is associated with reduced survival. In a study of 3049 patients with bone metastases a pathological fracture had up to a 32 % increased risk of death compared to the absence of a pathological fracture [33] (Table 1.6).

Table 1.6

Incidence of pathological fracture and implications on survival: data based on Saad et al. [33]. Hazard ratios are adjusted for previous skeletal related events and ECOG performance status of more than 2