What is the evidence for the use of bisphosphonates in the treatment of bone pain?

Guideline contents > What is the evidence for the use of bisphosphonates in the treatment of bone pain?

Author(s):

Professor Villis Marshall — Author
Professor Christopher Sweeney — Co-author
Cancer Council Australia Advanced Prostate Cancer Guidelines Working Party — Co-author

Funding received from

Cite this page

Marshall, V, Sweeney, C, Cancer Council Australia Advanced Prostate Cancer Guidelines Working Party. Clinical question:What is the evidence for the use of bisphosphonates in the treatment of bone pain? . In: Clinical practice guidelines for the management of locally advanced and metastatic prostate cancer. Sydney: Cancer Council Australia. [Version URL: https://wiki.cancer.org.au/australiawiki/index.php?oldid=90259, cited 2023 Dec 11]. Available from: https://wiki.cancer.org.au/australia/Guidelines:Prostate_cancer/Management/Locally_advanced_and_metastatic.

Last modified:
29 October 2014 04:08:27

What is the evidence for the use of bisphosphonates in the treatment of bone pain?

Castrate-resistant prostate cancer with existing bone pain

Seven low-quality randomised trials (five were double blind) focus on the treatment of existing pain of patients with painful bone metastases. Measurements of pain and pain outcomes varied. In three studies men in both arms received anti-neoplastic therapy. Due to many confounding factors and study limitations, no firm conclusions can be made regarding the ability of bisphosphonates to manage existing bone pain related to prostate cancer.

The two larger and more informative studies are presented below:

Small et al 2003^[1] examining two multi-centred randomised placebo controlled trials reported that pamidronate disodium 90mg (iv) administered every three weeks for 27 weeks did not provide any improved palliation of worst (p=0.89) or average (p=0.71) bone pain compared with placebo in menwith bone pain at study entry. At nine weeks the mean decrease in worst pain score on a 0–10 pain scale was 0.86 in the pamidronate arm and 0.69 in the placebo group (p=0.58). No significant difference was seen in the use of radiation for bone pain relief (p=0.88). However a pre-planned subgroup analysis of men with stable or falling analgesia showed at nine weeks (n=121) a 2.13 unit decrease in mean worst pain score on a 0–10 scale in the pamidronate group, which was significantly (p=0.008) larger than the 0.79 unit decrease in the placebo group. Retrospective subgroup analysis for men with moderate rather than mild baseline pain was reported to show a significant reduction in pain (p=0.004) at nine weeks.

Ernst et al 2003^[2] using a randomised double blind design (n=227) compared clodronate (1500 mg iv every three weeks) combined with mitoxantrone and prednisone with placebo and mitroxantrone and prednisone. Similar levels of pain control were noted in both arms; 43% experiencing pain improvement in the clodronate arm and 38% experiencing pain improvement in the placebo group (p=0.52). In this study analgesia was considered in determining the pain response. Similar proportions of men required local radiotherapy with over 12 months follow-up; 16% of men in the clodronate arm and 14% of the men in the placebo arm (p=0.85). However, in a subgroup analysis of men with moderate rather than mild baseline pain (n=49), 58% of men receiving clodronate treatment experienced pain palliation whereas only 26% of men receiving placebo experienced pain palliation (odds ratio=4.6, 95% CI=1.3 to 15.5, p=0.04).

For completeness and to appreciate the limitations of the remaining studies, a brief outline follows.

Smith 1989^[3] examined the effect of sodium etidronate on the control of bone pain in a small (n=28) double-blind RCT. Men received either sodium etidronate (iv) 7.5 mg/kg/day for three days then orally 2x200mg/day for one month, or placebo. No significant differences were observed between the two groups (p=1.00).

Adami and Mian 1989^[4] randomised 13 men with radiographic evidence of bone metastases to treatment with either 300mg of sodium clodronate (iv) daily or placebo for 14 days. It was reported that pain scores measured using a visual analogue scale and analgesic consumption fell in the treated group, but the statistical significance of the difference compared with the placebo group was not reported. In a multi-centre double-blind RCT without any documented chemotherapy, Strang et al 1997^[5] treated 55 men with either sodium clodronate 300mg (iv) for three days followed by oral sodium clodronate at a dose of 3200mg daily or placebo for four weeks. They did not find a significant improvement in pain scores between groups at 32 days follow up, however their data suggested that patients with higher initial pain scores (n=20) may have a better response than those with lower scores. This trial was terminated prematurely because of recruitment difficulties and as a result numbers may not have been sufficient to show any statistically significant differences in men with high pain scores.

Elomaa et al 1992^[6] and Kylmala et al 1993^[7] reported a randomised trial of sodium clodronate 3.2g/day for one month followed by 1.6g/day for a further five months. At one month there was a reduction in pain from baseline in both the control and the clodronate arms. Although the reduction was greater in the treated group, with 28% of men no longer experiencing pain compared with 15% of men in the control arm, the difference was not statistically significant (p=0.26). The probable explanation for this finding (ie pain relief without clodronate in the control group) was that both groups were started on estramustine phosphate 2x280mg per day at the same time as the clodronate. The symptomatic improvement in the control group was almost certainly due to the effect of the extramustine, thus rendering it difficult to determine the true effect of clodronate.

Kylmala et al 1997^[8] reported the results of a similar double-blind placebo controlled trial. Fifty-seven men, most of whom had painful bone metastases, began estramustine therapy and were randomised to either clodronate (iv) 300mg a day for five days and then oral clodronate 1.6g/day for 12 months, or placebo. Again, no significant difference was found between the groups, with mean pain scores in the treatment group not significantly improved from those in the control group over the 12 months.

Prevention/delay of bone pain in patients with hormone refractory disease which is either asymptomatic or with minimal symptoms from metastatic bone disease

In a quality double-blind multi-centre RCT, Saad et al 2002 and 2004 ^[9]^[10] (N = 422) were unable to show consistent differences at 15 months follow up in reducing pain progression between men treated with 4mg zoledronic acid (iv) and placebo (p = 0.13). However, with a follow up of 24 months they were able to show significantly smaller increases in brief pain inventory scores at 21 (p = 0.01) and 24 months (p = 0.02) for the treated group when compared with placebo. As the authors point out, the study was not primarily designed to assess pain improvement.

Weinfurt et al 2006^[11] published a re-analysis of the pain data from this trial. They found that a typical zoledronic acid patient had a 33% chance of a better pain response than a placebo patient, whereas a placebo patient had a 25% chance of a better pain response than a patient receiving zoledronic acid. This difference was statistically significant (p = 0.04). Nearly one third of patients in this trial had no baseline pain and for those experiencing baseline pain, changes in pain levels were not reported. Furthermore, many of these may have had osteoarthritis non-cancer-related pain at entry.

Comments

Back to top

Side effects of bisphosphonate In most trials, the reported incidence of moderate to severe side effects for sodium clodronate, pamidronate and zoledronic acid is low. In randomised trials these agents were reported to cause hypocalcaemia in 2–4% of patients^[12]^[9] and in the zoledronic acid trial^[9] all men under study were given calcium supplement 500mg/day and vitamin D 400–500 IU/day. In the case of zoledronic acid^[9], the incidence of fatigue, anaemia, myalgia, dizziness and lower limb oedema were at least 5% greater in the treated group than the control group. Also in their study, the dose of 8mg per treatment was discontinued because of the increased incidence of renal impairment at that dose level. No men were accepted onto the trial if they had a serum creatinine greater than 265umol/L. A case series reported by Oh et al 2007^[13] indicates that renal impairment may be more frequent than previously appreciated in patients treated with zoledronic acid.

Recently, there have been reports of osteonecrosis of the jaw in patients undergoing treatment with bisphosphonates. In a case series of patients undergoing treatment with bisphosphonates for a variety of cancers, Bamias et al 2005^[14] reported an incidence of 6.7% for osteonecrosis of the jaw, and an incidence of this condition in 6.5% of men with prostate cancer. They identified length of exposure as the most important risk factor and that zoledronic acid may be more prone to produce this complication than pamidronate. In a recent publication of a cohort study with matched controls, Wilkinson et al 2007^[15] have confirmed the earlier case series reports. They showed that cancer patients treated with intravenously administered nitrogen containing bisphosphonates had an absolute risk at six years for any jaw toxicity of 5.48% compared with matched controls, who had an absolute risk of 0.30%. When adjustments were made for a number of potential confounders, the hazard ratio was 11.48 (95% CI = 6.49 to 20.33). They also confirmed that length of exposure was a risk factor for jaw toxicity with those receiving 14–21 infusions of bisphosphonates significantly more likely to experience jaw toxicity than those who received less than three infusions (hazard ratio = 3.02 with 95% CI = 1.28 to 7.10). They were unable to find a clear association between types of cancer, comorbidities, or with known risk factors for jaw and facial bone disease such as diabetes, alcoholism, smoking, obesity, hyperlipidaemia or parenteral corticosteroids. Atrial fibrillation may also be a concern. In a RCT of once-yearly infusions of zoledronic acid for the treatment of post-menopausal osteoporosis, there was a significant increase in ‘serious’ atrial fibrillation.^[16]

Comments

Evidence summary and recommendations

Evidence summary	Level	References
For men with initially asymptomatic or mildly symptomatic bone metastases, zoledronic acid has been shown to be associated with less bone pain at 21 and 24 months. As the trials reported did not have pain control as their primary objective, this has diminished the power of these observations.	II	^[9], ^[10], ^[11]
Sodium etidronate has not been shown to be effective in controlling bone pain in metastatic prostate cancer.	II	^[3]
Three underpowered randomised controlled trials have failed to show a significant improvement in the control of metastatic bone pain over placebo either with clodronate alone or in combination with estramustine sulphate. A fourth study found a significant improvement with the addition of clodronate to mitoxantrone/prednisone only in a subgroup of men with moderate rather than mild baseline pain.	II	^[6], ^[8], ^[2], ^[10], ^[7]
Pamidronate disodium has not been shown to be effective in controlling bone pain in metastatic prostate cancer, except in a subgroup of men with stable or falling analgesia in which a significant but modest benefit was observed in worst pain.	II	^[1]

Evidence-based recommendation	Grade
On the basis of the available evidence, bisphosphonates are not recommended for routine palliation of symptomatic bone disease in men, with hormone-resistant prostate cancer with a possible exception of zoledronic acid, where there is some evidence of a benefit in castrate-resistant men.	C

In Australia, zoledronic acid (4mg in 5ml) is the only bisphosphonate listed on the PBS for ‘bone metastases from hormone-resistant prostate cancer with demonstration of biochemical progression of disease despite maximal therapy with hormonal treatments’. It is also listed along with clodronate for the ‘treatment of hypercalcemia of malignancy refractory to anti-neoplastic therapy’. (<www.pbs.gov.au> accessed 3/06/2009).

Comments

Back to top