Is there any benefit to early spinal radiation of metastases?

Is there any benefit to early treatment of asymptomatic men with many bone metastases? We saw that, at least in a retrospective study, there was no oncological benefit to treatment of oligo (1-5) bone metastases. But even if there is no benefit in delaying cancer progression, perhaps spinal radiation can prevent pain and crippling spinal compression?

Dearnaly et al. reported the results of a large multicenter randomized trial in the UK among 420 men with castration-resistant prostate cancer with asymptomatic spinal metastases on a bone scan. The goal was to determine whether early detection with MRI and treatment (SBRT radiation) could prevent clinical spinal cord compression (cSCC).

• The MRI detected early signs of spinal cord compression in 31% (61 patients), and they were treated with SBRT radiation using 20 Gy in 5 treatments.

The 1-yr and 2-yr results for all 210 men given MRIs (the "intervention group"), whether SCC was detected radiographically or not:

• At 12 months after randomization, the incidence of cSCC was 4.3% for the intervention group and 6.7% for the control group (no statistically significant difference)
• At 24 months after randomization, the incidence of cSCC was 9.2% for the intervention group and 12.6% for the control group (no statistically significant difference)
• Pain scores and severity of SCC were similar in both groups.
• Chemotherapy was more often used in the control group.
• Prostate cancer-specific mortality was similar in both groups.
• Overall survival was the same (22 months) in both groups. Note: this was a very progressed population of patients

The 1-yr and 2-yr results for only those 61 men who showed early signs of MRI detected (MRI+) SCC vs the remaining 139 patients who were MRI-:

• At 12 months after randomization, the incidence of cSCC was 11.5% for the MRI+ group and 1.3% for the MRI- group (note: statistically significant difference even after treatment)
• At 24 months after randomization, the incidence of cSCC was 13.2% for the MRI+ group and 7.6% for the MRI- group (note: statistically significant difference even after treatment)
• MRI+ patients had more clinical SCC, regardless of early detection and therapy.

So, there was no benefit to early detection and treatment. Perhaps there is a sub-group that could benefit?

Gillespie et al. reported the results of a small multicenter randomized trial among 78 men and women (22% had prostate cancer) who had more than 5 metastatic lesions identified on a bone scan, where none were yet painful ("asymptomatic"), but at least one of the bone metastases was "high risk." High risk was defined as any of:

• 2 cm or more in diameter
• at a junction in the spine between the cervix, thorax, lumbar, and sacral vertebrae
• in the hip or sacroiliac joint
• in a long bone (arms or legs) (note: this is rare for prostate cancer)

Patients were randomly treated with non-ablative radiation. After at least 1 year of follow-up, they observed the rate of skeletal-related events (SREs) and mortality. SREs could be pain, spinal compression, or fractures:

• SREs occurred in 1.4% of those receiving radiation vs 29% of those receiving standard care
• After 3 months, but not afterward, there was less pain reported by those who received radiation.
• There were no quality of life differences at any point in time.

After a median of 2.4 years of follow-up:

• Survival was 1.5 years for those with no SRE vs 1.1 year for those with an SRE

The discrepancy with the larger Dearnaly trial may be due to the identification of high-risk spinal mets; or the lower incidence of prostate cancer, the small sample size, and limited follow-up.

Dosimetry

A single ablative dose for painful non-spinal metastases has been found to be superior in this study. Fewer higher doses (24 Gy in 2 treatments) were more effective than 20 Gy in 5 treatments for relieving pain from spinal metastases in this study.  Sahgal et al. in a Canadian/Australian multi-institutional trial, found there was a better pain response with SBRT (24 Gy in 2 fractions) compared to IMRT (20 Gy in 5 fractions). After 3 months, significant pain improvement was 35% for SBRT vs 14% for IMRT. 

There is similar data from Heidelberg. Sprave et al.  reported better pain response at 6 months (but not at 3 months) with SBRT (24 Gy in 1 fraction) vs 3DCRT (30 Gy in 10 fractions). 

Conflictingly, Ryu et al. reported the results of the NRG Oncology/RTOG 0631 randomized clinical trial. They tried to obtain proof that SBRT (16-18 Gy in one dose) was superior to IMRT in one dose (8 Gy) in terms of pain response. After 3 months, 61% of those treated with IMRT had a significant pain improvement vs 41% of those treated with SBRT. After 1 year, there was no difference in pain scores. After 2 years, there was no difference in spinal fractures or compression. 

Spinal dosimetry depends on extent and soft-tissue involvement, and may require a neurological consult.