Metastatic fractures

WAKENDA K. TYLER

Introduction

Diagnosis and clinical presentation

INTRODUCTION

As we age, the incidence of cancer increases exponentially, making the elderly population most susceptible to the development of cancer and unfortunately many of the sequelae associated with it. Two-thirds of all cancer deaths occur above the age of 65 in the United States.¹ The most commonly seen forms of cancer in the elderly, which are lung, breast, prostate and the haematopoietic cancers, all have a high propensity to involve bone. Overall, 70–80% of patients with metastatic prostate or breast cancer and 40% of patients with metastatic lung cancer will have bone involvement from their cancers.^2,3 Unfortunately, in the case of older patients with metastatic disease, the bone involvement occurs in the setting of age related osteopenia or osteoporosis. To further complicate the picture, many patients with metastatic cancer are receiving systemic therapies that can further compromise the quality of the bone. These therapies include the hormone blocking agents and frequently steroids.

Skeletal related events, such as compression fractures or long bone fractures, occur in more than half of patients with metastatic breast, prostate or lung cancer.⁴ A pathological fracture is often a devastating event for a patient with cancer as it not only dramatically impacts their functional status but can also greatly interfere with treatment. Patients who do not meet certain criteria for functional status often cannot participate in many of the available chemotherapy protocols. Without systemic treatment, cancer progression in vital organs and in bone is often greatly accelerated. Once a pathological fracture occurs, length of survival greatly diminishes.⁵ This phenomenon of decreased survival after a metastatic cancer related fracture is partly due to disruption of treatment and decreased mobility, but also often indicates a disease process that has reached a more terminal state and may no longer be responding to systemic and available local therapies.

Certain solid organ cancers have a higher predilection for bone than others. This is due to a complex interaction between the tumour cells and their host environment. There is some evidence to suggest that the tumour cells, either at the primary site or circulating in the blood stream, can induce the bone environment to be more prepared for the tumour cells’ adherence and growth in that environment.⁶ The tumour cells are capable of secreting a myriad of proteins that enhance their ability to adhere and invade into the bone microenvironment. Once in that environment, they can then replicate and further induce the local cells to increase or decrease bone formation and activate a series of events that leads to an appropriate condition for the tumour to propagate. The solid organ cancers most likely to induce this series of events are lung, breast, prostate, renal and thyroid. Although these are often reported as the most common cancers to spread to bone, it should be noted that other solid organ cancers, such as liver, colon, rectal, pancreatic and uterine, can all spread to bone and lead to pathological fracture (Figure 16.1). The haematopoietic cancers frequently involve bone and in some cases, like multiple myeloma, bone involvement is an essential component of the cancer’s pathogenesis. Any patient with a haematopoietic cancer should have close monitoring of their bone disease to detect early risk of pathological fracture.

Once cancer cells have entered into the bone environment, they frequently will induce osteoclast and/or osteoblast activation depending on the gene expression within the cell. The vast majority of lung, thyroid and renal cancers in bone are lytic (Figure 16.2), meaning that they have induced bone resorption, while 98% of prostate cancers in bone are blastic (Figure 16.3), with the cells inducing abnormal bone formation. Breast cancer can lie on a spectrum anywhere from purely blastic to purely lytic, but most breast cancers in bone are a mix of lytic and blastic. The significance of the type of lesion created by the tumour in bone is relevant to the risk of fracture in that a greater degree of bone lysis results in an increased risk of fracture. Blastic lesions also have an increased risk of fracture because the bone that is induced to form in this setting is abnormal both in structure and strength.

Figure 16.1 (a) Patient with metastatic colon cancer who presented with right leg pain. X-rays were obtained, but films were not closely scrutinized for a pathological process because of a perceived low incidence of metastatic bone disease in this patient population. Black arrows show area of concern. (b) Three months later the patient presented to the emergency room with this fracture.

Figure 16.2 A lytic bone lesion (circle) in the distal diaphysis of the humerus in a patient with multiple myeloma. Note the thinning of the cortex and loss of bone density around the lesion in comparison to the rest of the uninvolved bone.

Figure 16.3 Blastic bone lesions throughout the femoral shaft and pelvis (white arrows) in a patient with metastatic prostate cancer. Note the increased density on the plain film illustrating the abnormal bone formation that is induced by the prostate cancer cells.

The more recent trends have suggested that patients with advanced cancer are living longer and as a result there is a trend towards an increased prevalence of older patients with metastatic disease of bone.¹ Although cancer metastasizing to bone is generally considered a terminal diagnosis for patients with cancer, patients with metastatic bone cancer can and often do live for many years after diagnosis of bone involvement. Not every patient with metastatic bone disease is at risk for a pathological fracture. Therefore it is important to determine who is at risk, so that we do not over- or undertreat this patient population. The goals of any treatment for this cohort are (1) to improve or maintain function, (2) to reduce pain and (3) to prolong survival if possible. Determining the best treatment for each individual patient can be a daunting task. Preventing pathological fracture from occurring in a patient who will regain function after treatment is the ideal approach. The development of a pathological fracture is not an imminent terminal event as once believed and these patients need equal consideration with a treatment plan that will best improve their quality of life.

Several recent analyses have looked at the cost of skeletal events in patients with metastatic cancer and have found that the burden to healthcare systems is very high.^{7,8 and 9} Spinal cord compression and the need for surgical intervention rank highest among cost, with US$20,000 per patient being spent on spinal cord compression treatment and approximately US$18,000 on surgical intervention for pathological fracture.¹⁰ Not surprisingly, spinal cord compression is also associated with the shortest life expectancy.¹⁰ Reducing the likelihood of these two catastrophic events would greatly reduce healthcare costs, but more importantly improve quality of life.

Metastatic fractures are a serious threat to the well-being of the elderly population. Identifying patients at risk for fracture and treating them before the event occurs is ideal. However, despite our best efforts, fractures can still occur. Patients with metastatic fractures need prompt identification of the fracture and appropriate treatment that takes into account both quality and quantity of life remaining.

DIAGNOSIS AND CLINICAL PRESENTATION

Patients at risk for metastatic fracture will often present as one of three scenarios: (1) known metastatic disease and usually known bone involvement, (2) a remote history of cancer and a new finding of a bone lesion or (3) no known history of cancer and a new bone lesion. Each of these scenarios requires a different protocol for workup and management of the bone lesion. The first priority in all of these scenarios is determining what the bone lesion is, followed by a determination of the risk of fracture. If the risk of fracture is determined to be high, the next important step is to identify if fracture in a particular part of the skeleton will lead to major loss of function or not, as this will ultimately influence treatment.

For the patient with a known history of metastatic disease, it is not uncommon to have multiple sites of bone involvement. In this situation it is appropriate to assume that a bone lesion is due to metastatic cancer and therefore a major workup to identify the aetiology of the lesion is not necessary. If a particular site is causing pain, this should raise the concern of possible fracture risk. Bone pain often manifests as pain associated with weight bearing and can also present as deep aching pain at rest, usually at the site of concern within the skeleton.

Bone pain is the most common presenting symptom of a patient with an impending metastatic fracture. In some instances, patients will report no pain prior to fracture, but an extremely low energy event, such as opening a door or standing up from a sitting position, results in fracture. Pain is not always an indication of impending fracture as back pain with nerve root compression may present with leg pain but without risk of femoral or tibial fracture. Radicular pain such as this will often originate in the buttock region and radiate down the entire length of the leg, lacking localization to a discrete area.

Two classification systems are frequently used to help determine risk of fracture in a patient with known metastatic disease to bone. These systems are good tools to help guide decision making, but should not be used as absolute rules. Tables 16.1 and 16.2 contain the Mire and Harrington classification systems for determining impending fracture risk. In the case of the Mirel classification, a score equal or greater than 9 is high risk of fracture and likely warrants fixation. In the case of the Harrington classification, any one of these criteria should warrant strong consideration for prophylactic fixation. Both use clinical presentation as well as findings on plain X-ray. Plain X-rays are considered the gold standard for assessing the structural integrity of bone and therefore the true risk of fracture. CAT scans can also be used to assess the structural integrity of bone, but are not necessary for determining fracture. Magnetic resonance imaging (MRI) is useful for looking at soft tissue involvement and extent of tumour involvement within the marrow space, but lacks the ability to assess the cortical structure of bone and is therefore not a good tool for determining fracture risk (Figure 16.4). Technitium-99 bone and positron emission tomography (PET) scans are useful in locating the site of bone involvement, but give no information about the structural integrity of bone and therefore should not be used as the sole method for assessing risk of fracture. PET scans and bone scans help determine the metabolic activity and cellular activity at a particular site and this may be a surrogate for the virulence of the tumour cells, but does not provide details on what the tumour cells are actually doing to the bone at that site. When obtaining plain films, full length anterior–posterior and lateral films of the entire limb should be ordered. It is common for patients with metastatic disease with bone involvement at the upper end of the bone to also have a significant lesion at the lower end of the same bone. Missing this could be quite serious with regard to treatment planning (Figure 16.5).

Figure 16.4 (a) T2 weighted MRI of the knee in a patient with metastatic breast cancer. White arrows indicate significant oedema seen on MRI within the bone, but without plain film (b), one cannot accurately determine the degree of loss of bone architecture. In this case bone architecture is intact.

Figure 16.5 Anterior–posterior (AP) view of the femur of a patient with metastatic renal cancer. (a) A lesion was first identified at the proximal end of the femur. (b) Full length femur films revealed a lesion at the distal end of the bone (white arrow). An intramedullary implant should not end at the location of this distal lesion or a stress riser will be created.

Harrington first described a fairly simplistic method for assessing risk of fracture in the early 1980s. His system has stood the test of time and is one that is easy to remember for those who may not treat or see patients with impending fractures on a daily basis (Table 16.1 and Figure 16.6). The other classification system that is commonly used is the Mirels classification system (Table 16.2). Unlike the Harrington system, it takes into account type of metastatic lesion (lytic or blastic) and location. For the Harrington system, if any one of the criteria is met, the patient is considered at high risk of fracture.¹¹ For the Mirels system (Table 16.2), the patient is assigned a number from 1 to 3 for each of the four criteria assessed: site, pain with activity (functional pain), type of lesion and size of lesion in relation to the diameter of the bone region being assessed for fracture risk. The numbers are then added together. A score of <7 has a very low risk of fracture (<4%), while a score of >9 is considered to have a fairly high risk of fracture (33% or more likely to fracture in the next 6 months).¹²

Using the images in Figure 16.6 as an example, the patient has a lesion in the peritrochanteric region of the femur. It is lytic and greater than 2.5 cm in size as well as about two-thirds the diameter of the entire bone in that location of the femur. If you are also told that the patient has significant pain with both taking a step and straight leg raise when supine on a bed, then you also know that the patient has significant functional pain. Based on this clinical scenario and image presentation, the Mirels score would be 12, the highest possible score indicating a very high risk of fracture over the next 6 months. By Harrington’s criteria (>2.5 cm and persistent functional pain), the patient also would be considered at high risk for fracture. Treatment to prevent fracture would likely be recommended assuming no other mitigating factors.

Table 16.1 Harrington system for determining risk of fracture

Cortical bone destruction of 50% or more

Lesion of 2.5 cm or more in the proximal femur

Pathological avulsion fracture of lesser trochanter

Persistent stress pain (functional pain)

Source: Harrington KD. Instr Course Lect 1986;35:357–381.

Figure 16.6 Lateral hip (a) and anterior–posterior (AP) view of the pelvis (b) of a patient with metastatic lung cancer in the left proximal femur. Note the cortical erosion on both the lateral and AP views (white arrows). Compare the contralateral hip on the AP pelvis to see the more medially located erosion. This patient has a very high risk of metastatic fracture using both Harrington’s and Mirels’ criteria.

Table 16.2 Mirels algorithm for determining risk of pathological fracture

	1	2	3
Site	Upper extremity	Lower extremity	Peritrochanteric
Pain	Mild	Moderate	Functional
Lesion	Blastic	Mixed	Lytic
Size	<1/3	1/3–2/3	>2/3

Source: Mirels H. Clin Orthop Relat Res 1989;(249):256–264.

Note: <1/3, 1/3-2/3 and >2/3 is in reference to diameter of bone on the AP or lateral plain radiographs.

For a patient with a remote history of cancer who presents with a new bone lesion and the patient who presents with no known history of cancer and a bone lesion, it is imperative that the aetiology of the lesion in question be determined before proceeding with treatment. Primary bone sarcomas tend to present in two patient populations. The first is in patients under the age of 21, who will present with tumours such as rhabdomyosarcoma, Ewing’s sarcoma and osteosarcoma. The second is in patients over the age of 65, who will present with chondrosarcoma, high-grade pleomorphic sarcoma and osteosarcoma. Wrongfully treating or missing a primary sarcoma of bone can result in loss of limb (amputation) or more significantly, loss of life. This occurs most commonly in patients over the age of 65 in part because of the misconception that primary sarcomas of bone are only a paediatric condition. Therefore any patient presenting with a new bone lesion and no known history of metastatic bone disease should be investigated appropriately to exclude primary sarcoma of bone. The workup should include a CT scan of the chest, abdomen and pelvis (to look for a primary site of cancer origin), a bone scan (to look for other bony sites of involvement), serum protein electrophoresis and urine protein electrophoresis (to look for multiple myeloma) and often a biopsy. If the CT scan, bone scan and laboratory results do not reveal obvious metastatic disease or an obvious primary lesion, then a biopsy is warranted. In most cases, biopsies can be done through an image guided technique that does not require general anaesthesia. These biopsies are usually coordinated through the interventional or musculoskeletal radiology team. Rarely, an open biopsy is indicated if a needle biopsy fails to determine the diagnosis or if the lesion has features that would suggest a higher quantity of tissue may be necessary to determine the diagnosis.

Once a metastatic lesion is identified and confirmed, a decision regarding appropriate treatment can be made. The most common sites of bone involvement from metastatic disease are the spine (lumbar>thoracic>cervical), ribs, pelvis, skull and proximal long bones. In any patient with metastatic cancer, these sites should be monitored.¹³ It is also worth noting that patients with metastatic cancer have a multitude of conditions that can arise as a result of their cancer. In the case of patients with bone involvement, the treating physician should be aware of the possibility of hypercalcemia which is estimated to occur in 5–10% of all patients with metastatic bone disease.¹⁴ Hypercalcemia is a potentially fatal complication of metastatic bone disease and one that can be treated if identified early. It is recommended that any patient being evaluated for metastatic fracture or metastatic bone disease also be evaluated for hypercalcemia either through basic laboratory work or monitoring of symptoms for hypercalcemia (confusion, lethargy, excessive thirst and urination, nausea, vomiting and diffuse bone pain). Unfortunately the symptoms of hypercalcemia are also symptoms one finds in conjunction with chemotherapy, so good clinical judgment is required. First line treatment of hypercalcemia is hydration, followed by use of intravenous (IV) bisphosphonates if oral or IV hydration is not enough to bring calcium levels close to normal.

TREATMENT

Treatment of metastatic and impending metastatic fractures is determined by a multitude of factors. Type of malignancy, location of lesion and life expectancy are important factors to take into account. The vast majority of metastatic bone lesions do not require any direct intervention and even among those at increased risk of fracture, intervention is not necessarily surgical in nature. The top four most common sites of metastatic bone involvement (spine, ribs, pelvis and skull) rarely require surgical intervention for treatment of metastatic fractures. The majority of patients with metastatic bone disease do not wish to spend most of their remaining life recovering from an extensive surgical procedure and it is unfair to both the patient and their family not to take this into consideration when determining a treatment plan. Both surgical and non-surgical interventions should be considered together when deciding on an appropriate treatment plan for patients with metastatic or impending metastatic fractures (Box 16.1).

Radiation and radionucleotide therapy

Radiation therapy has several important roles in the treatment of patients with metastatic bone disease. It can be a very effective modality for reducing cancer related bone pain. It can also reduce the likelihood of metastatic fracture or progression of disease after surgical treatment of a metastatic fracture. It is also a useful tool in the prevention and treatment of spinal cord compression. Approximately 50–80% of patients will experience some level of pain relief following radiation therapy for painful metastatic bone lesions and up to 35% will report complete pain relief at the site of treatment.¹⁵ It is the treatment of choice at sites of bone involvement that are not amenable to surgical intervention and it can be used to avoid surgery in areas where surgery might be considered but radiographic or clinical presentation do not yet warrant that degree of intervention. Radiation therapy should be reserved for patients at risk for fracture or with significant pain, since there is a limit to the amount of radiation a given area of the body can receive. If that limit is reached early in the disease process, then radiation can no longer be used as a treatment modality later on when the patient may have greater need for it.

BOX 16.1: Treatments for metastatic fractures