Data from the phase 3 NRG-RTOG 0631 may inform future research assessing spinal radiosurgery in the oligometastatic setting, according to an expert from Stony Brook University Medical Center.
Stereotactic radiosurgery did not demonstrate superiority vs conventional external-beam radiotherapy (EBRT) in terms of 3-month pain response in patients with cancer who had spine metastases, although data may inform further investigation of spine radiosurgery in the oligometastatic setting, according to findings from the phase 2/3 NRG Oncology Radiation Therapy Oncology Group (RTOG) 0631 study (NCT00922974)1.
The incidence of pain responses at 3 months was 41.3% for patients receiving EBRT vs 60.5% for those receiving radiosurgery (1-sided P = .99; 2-sided P = .01). Additionally, the mean change from baseline in pain score at 3 months was –3.83 with EBRT vs –2.98 with radiosurgery (P =.07).
At 12 months, pain response from radiosurgery was 57.6% vs 55.3% among those receiving EBRT (P = .49). In the radiosurgery and EBRT groups, respectively, sensitivity analysis results revealed no difference in the appearance of new metastases (43.5% vs 43.9%; P = .96) or progression of known metastases (34.0% vs 42.3%; P = .12). Additionally, 12-month and 24-month survival rates were 44.3% and 31.5% in the radiosurgery group vs 53.1% and 31.5% in the EBRT group (Hazard ratio, 0.91; 95% CI, 0.69-1.20; P = .51).
“It is important to note that NRG RTOG 0631 is the first phase 2/3 multi-center trial assessing the safety and efficacy of [stereotactic radiosurgery] for the treatment of vertebral metastases,” lead study author Samuel Ryu, MD, radiation oncologist in the Department of Radiation Oncology at Stony Brook University Medical Center, said in a press release on the study.2 “The data collected from this trial could also inform future trials, especially those assessing [stereotactic radiosurgery] for epidural tumor decompression, spinal cord compression, and treating oligometastases where durable local tumor control improves survival.”
Investigators of the phase 2/3 NRG RTOG 0631 study randomly assigned patients 2:1 to receive treatment with stereotactic radiosurgery or conventional EBRT. In the radiosurgery group, patients received a single dose of 16 or 18 Gy to the margin target volume encompassing the involved vertebral bone, with prescription isodose line selections based on institutional preference. In the EBRT group, patients received a single dose of 8 Gy to the isocenter and delivered to the involved vertebral levels.
The primary end point was 3-month patient-reported pain response at the index vertebral level using the Numerical Rating Pain Scale (NRPS). Secondary end points included the rapidity and duration of pain response, adverse effects (AEs), long-term effects on the spinal cord, vertebral compression fractures on follow-up MRIs, and quality of life (QOL).
Patients 18 years and older with a Zubrod score of 0 to 2 and 1 to 3 treatment-naïve vertebral metastases were eligible for inclusion in the trial. Additional eligibility criteria included having solitary vertebral metastasis, 2 contagious vertebral levels involved, or a maximum of 3 separate sites.
Of 339 patients, 209 received stereotactic radiosurgery, and 130 received conventional EBRT. Between both arms, 76.7% of patients only had 1 vertebral metastasis, and 87.3% received pain medication. Overall, 22.0% of those in the radiosurgery group and 10.0% of those in the EBRT group had a Zubrod score of 2 (P = .02). The mean pain score at the index vertebra at baseline was 6.06 in the radiosurgery group and 5.88 in the EBRT group.
Modeling of complete case data showed that the most predictable factor for pain control was a Zubrod score of 0, although Zubrod performance status did not have a significant impact on multiple imputation (Odds ratio, 0.62; 95% CI, 0.37-1.06; P = .08). Investigators also reported no significant differences in the rapidity or duration of pain response between treatment arms.
Investigators identified no significant differences in changes in QOL scores between treatment arms. A Functional Assessment of Cancer Therapy-General total score longitudinal model did not reflect any differences between treatment groups, which investigators confirmed on multiple imputation.
Overall, 7.7% of patients in both the radiosurgery and EBRT groups had any-grade acute AEs. Additionally, 2 patients in the radiosurgery group and 1 other in the EBRT group had grade 4 late AEs attributed to sepsis or lymphopenia.
The 24-month proportion of vertebral compression fractures was 19.5% in the radiosurgery group and 21.6% in the EBRT group (P = .59). At 24 months, the spinal cord signal change was 3.6% and 1.7% in each respective group (P =.38).