Pharmacological intervention compared with no intervention for prevention of radiation‐related insufficiency fractures and avascular necrosis

Patient or population: Adults undergoing pelvic radiotherapy

Settings: Intervention prior or during radiotherapy; hospital

Intervention: Zoledronic acid

Comparison: No intervention

The 2 included studies did not fully describe the radiotherapy techniques used.

The dose, delivery of treatment (conformal, intensity modulated, brachytherapy) and, therefore, volume of pelvis irradiated could influence the pattern of bone fractures, changes in BMD and risk of avascular necrosis.

Studies were heterogeneous in other aspects so meta‐analyses were not performed and results were reported as a narrative.

Insufficiency fractures and avascular necrosis

• Kachnic 2013 reported 1 fracture occurring in each intervention group.

• Denham 2014 reported 72 non‐spinal fractures (38 in intervention group). Only 3 fractures were near the radiation field and 1 case of avascular necrosis but details of the intervention group for these was not provided.

• Data were sparse, and the imaging and investigations used did not meet the review criteria.

BMD and fracture risk

• Improvements in BMD in both studies.

• 1 RCT reported significant percentage increases in BMD at 36 months in the lumbar spine (6% with zoledronic acid versus ‐5% with control; P < 0.0001), left hip (1% with zoledronic acid versus ‐8% with control; P = 0.0002) and left femoral neck (3 with zoledronic acid versus ‐8% with control; P = 0.0002) (Kachnic 2013).

• In comparison, the other RCT had a sub study assessing the change in mean total hip BMD at 24 and 48 months (Denham 2014). For the 6‐month androgen deprivation group, there was no change in BMD at 24 months (0.6%; P = 0.18) but a significant increase at 48 months (1.8%; P = 0.003). In the 12‐month androgen deprivation group, there was no significant change in BMD at 24 months (0.5%; P = 0.41) and a tendency towards significance at 48 months (1.2%; P = 0.09).

• As different time points and sites were used to measure changes in BMD, a further analysis was difficult to perform.

• Only 1 RCT reported fracture risk, but the risk was assessed using a non‐validated method, so further analysis was not conducted (Denham 2014).

Bone turnover markers

• The 2 included RCTs did not report changes in bone turnover markers between control and intervention groups, preventing further reporting on these outcomes.

QoL

• Only 1 RCT reported QoL and suggested there were no differences in QoL between groups (Kachnic 2013). However, this study only accrued small patient numbers and only 58% QoL data were obtained at 36 months so there was likely to be a lack of statistical power to detect any true differences.

• Neither trial reported bone symptoms or unplanned hospital stays.

Mortality

• Neither trial reported cancer‐specific mortality of overall mortality.

Adverse events

• Both included RCTs did not fully describe the radiotherapy techniques used. The dose, delivery of treatment (conformal, intensity modulated, brachytherapy) and, therefore, volume of pelvis irradiated could influence the pattern of bone fractures, changes in BMD and risk of avascular necrosis.

• 1 RCT had an imbalance in the sub study groups reporting on changes in BMD, where 1 group may have had higher risk factors for osteoporosis that could bias the results (Denham 2014).

• The results reported by Kachnic 2013 need to be interpreted with caution as the study had inadequate accrual, therefore, lacked statistical power to detect differences.

• Adverse events and compliance to the intervention reported descriptively.

BMD: bone mineral density; FACT‐G: Functional Assessment of Cancer Therapy ‐ General; QoL: quality of life; RCT: randomised controlled trial.

GRADE Working Group grades of evidence
High quality: further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: we are very uncertain about the estimate.