What is the effectiveness of local external beam radiotherapy (EBRT) in the palliation of uncomplicated bone pain?

External beam radiotherapy (conventional)

Treatment of bone pain

No trials specific to prostate cancer were identified. Most trials have accrued patients with any commonly seen malignant diseases associated with bony metastases. In the majority of the trials, however, prostate cancer patients are well represented, comprising more than 20% of the total (Nielsen 1998, 33% prostate cancer^[1]; Steenland 1999, 23% prostate cancer^[2]; Kirkbride 2000, 23% prostate cancer^[3]; Roos 2005, 29% prostate cancer^[4]; Hartsell 2005, 50% prostate cancer^[5]; Bone pain Trial Working Party 1999, 34% prostate cancer^[6]). When results for prostate cancer subgroups were available they did not differ from those of the entire cohort.^[2]^[7] Trials were not blinded and thus were assessed as low quality.

The role of local radiotherapy in the management of uncomplicated bone pain is well established. It is considered a standard therapy for painful bone metastases with published accounts of its efficacy dating back to the 1920s. As such, RCTs comparing radiotherapy with no therapy would be considered unethical. An indirect way to consider the efficacy of radiotherapy for the treatment of metastatic bone pain is to examine the effects of lowering radiation doses. Poorer outcomes at lower doses would support the notion that radiotherapy is an effective treatment of metastatic bone pain. This review does not address the issue of the optimal fractionation schedule for multiple fractions.

Overall response rates of around 65–80 % and complete response rates of around 10–50% may be achieved as seen in the studies by Steenland et al.^[2]^[8] Rates vary with the definition of pain response, period after treatment assessed and the percentage of patients lost to follow-up or for whom data are missing.

Low dose comparisons

Two randomised controlled trials (Hoskin 1992, n=270, 13% prostate cancer patients^[9]; Jeremic 1998, n=219, 17% prostate cancer patients^[7]) showed that overall pain responses were significantly (P<0.01 and P=0.002) worse when patients were treated with a single dose of 4Gy rather than 8Gy.

Single versus multi-fraction regimens (differing doses)

The main issue at hand has been the relative efficacy of various fractionation schedules in effecting pain relief. Nine RCTs compared a single fraction of 8Gy with multiple fractions ranging from 20Gy in five fractions to 30Gy in ten fractions. One of these trials^[4] examined the effect of different fractionation schedules for the treatment of neuropathic bone pain in particular.

Pain endpoints and patient survival rates varied and the periods assessed ranged from four weeks postradiotherapy to twelve months post-radiotherapy. Complete response is generally defined as resolution of pain relief without need for analgesic consumption; partial response is defined as pain reduction of 2 or more at the treated site on a 0–10 scale without analgesic increase, or analgesic reduction of 25% or more from baseline without an increase in pain.^[10] However, other integrated painanalgesia response systems exist.^[3]^[4]^[5]^[8] In any study, however, integrated pain-analgesic response estimates may be diluted by pain from symptomatic metastases outside of the irradiated area.

Two of these trials assessed less than 100 patients^[11]^[12] and thus in these studies an absence of a significant difference in response rates may not reflect equivalence but rather a failure to detect a difference. Two trials were designed to detect a difference in response rate greater than 15% (Nielsen 1998, n=24152; Bone Pain Trial Working Party 1999, n=76157), one trial was designed to detect a difference in response rate greater than 18% (Roos 2005, n=27255), and one trial was designed to detect a difference in response rate greater than 10% (Steenland 1999, n=115753).

Despite these differences, all nine trials were unable to detect any significant difference in overall pain response, whether crude or actuarial response rates, and in the three studies that examined duration of response no significant differences were seen. Prognostically favourable patients with longer life expectancy did not derive greater benefit from multi-fraction schedules.^[8]^[13]

Whilst these trials found that a single fraction was not significantly worse than a higher-dose multifraction regimen in terms of initial pain response, the question as to whether they were equivalent was rarely addressed. In the trial reported by Roos^[4], treatments were to be considered equivalent if the 90% confidence interval for the difference in risk ratios was greater than 15%. In this trial, pain palliation with single fraction radiotherapy was not a statistically worse treatment group, however they were unable to show that it was equivalent to the higher-dose multi-fraction regimen.

Seven trials were unable to detect any significant difference in complete response. The trial reported by Hartsell 2005 (n=573)^[5] was designed to detect greater than 21.7% change in complete pain relief. However it is unclear whether the other trials were sufficiently powered to detect a difference.

Seven of these trials reported re-treatment rates. Re-treatment rates at the physician’s discretion were higher (8–18%) in the single-dose arm in four of the seven trials examining re-treatment rates (Price 1986, n=288, p=0.006^[14]; Bone Pain Trial Working Party 1999, p<0.001^[6]; Steenland 1999, p<0.001^[2]; Hartsell 2005, p<0.001)^[5]. These studies were not blinded and re-treatment may be subject to bias in the same way that initial pain response may be subject to bias.

Single fraction treatment did not have an adverse impact on quality of life^[1]^[2] ^[5]^[11]^[15] or significantly increase the incidence of spinal cord compression at the index site.^[4]^[6] Five studies examined the incidence of pathological fractures at the index site. Two studies found no difference in the incidence of pathological fractures^[4]^[5] whereas the larger Steenland study^[2] found a significant (p<0.05) increase in the incidence of pathological fractures within the single fraction treatment group. No significant difference was found in the incidence of femur fractures^[14] or long bone fractures^[6]

Eight trials examined acute toxicity.^[1]^[2] ^[4]^[5]^[6], ^[11]^[12]^[14] There were no statistically significant increases in short-term adverse outcomes with single-dose radiotherapy other than for the flare (p=0.03, Roos 200555). In one study, more severe toxicity (grade 2–4) was significantly decreased in the single fraction arm (p = 0.02, Hartsell 2005).56

These results support the conclusion that there is no evidence to suggest any dose response for initial pain response rates when comparing a single fraction of 8Gy versus multiple fractions ranging from 20Gy/5f to 30Gy/10f. That is, single fraction of 8Gy is not worse than a course of multi-fraction treatment for the endpoint of initial pain response. This is in agreement with the meta-analysis by Sze et al^[16] and updated,^[17] which included additional trials that did not meet the inclusion criteria for these guidelines.

Greater patient convenience and lower cost may make single fractions an attractive option for treatment even at the expense of higher re-treatment and fracture rates. However, two studies demonstrated that a significant proportion of patients may prefer multiple fractions if that will result in lower re-treatment and fracture rates.^[18]^[19]

Comments

Back to top

Evidence summary and recommendations

Evidence summary	Level	References
Low-dose external beam radiotherapy There are no controlled trials comparing EBRT with no treatment. As EBRT is a recognised treatment of metastatic bone pain, RCTs comparing radiotherapy with no therapy would be considered unethical. Poorer outcomes at lower doses support the notion that EBRT is an effective treatment of metastatic bone pain.	II	^[7], ^[9]
Single versus multi-fraction higher-dose EBRT No dose response exists for pain response rates when comparing a single fraction of 8Gy versus multiple fractions ranging from 20Gy/5f to 30Gy/10f. That is, a single fraction of 8Gy is not worse than a course of multi-fraction treatment for the endpoint of pain response. Fracture rates following radiation are low (<5%). There is no consistent evidence that fracture rates or spinal cord compression rates are higher in single fractions. Single fractions are associated with a higher re-treatment rate.	II	^[1], ^[2], ^[3], ^[4], ^[5], ^[6], ^[8], ^[11], ^[12], ^[14], ^[15]

Evidence-based recommendation	Grade
Radiotherapy is an effective and well-tolerated treatment for metastatic bone pain. A single dose of 8Gy is as effective as higher fractionated doses (eg 20–30Gy) in reducing bone pain. The higher incidence of re-treatment with lower-dose single fraction regimens should be considered as part of the decision-making process.	C