Pooled Analysis Reveals Long-Term Effect on QoL and Cognitive Outcomes From Radiation for Brain Metastases

Joshua Palmer, MD

In the first pooled prospective study to show the long-term outcomes of cranial radiation in patients with limited brain metastases, it was observed that whole-brain radiotherapy (WBRT) was associated with worse cognitive outcomes and mixed quality of life (QoL) compared with stereotactic radiosurgery (SRS) for these patients over time, according to a presentation from Joshua Palmer, MD, at the 2020 Virtual Conference on Brain Metastases, hosted by the Society for Neuro-Oncology.

When looking at the effects of SRS versus WBRT, investigators saw that from the time of radiation, there were mixed and changing outcomes in QoL for both treatments; patients also experienced a worse QoL when they had impaired cognition in the pooled analysis of the NCCTG (North Central Cancer Treatment Group) N107C/CEC.3 (NCT01372774) and NCCTG N0574 (Alliance) clinical trials.

These studies revealed that SRS was favored compared with WBRT. There were 194 patients in the N107c trial and 213 patients in the N0574 trial who had 1 and 3/4 brain metastasis with and without resection, respectively. For this analysis, the authors looked into the extended impact of radiation on 88 patients (54 from N107c and 34 from N0574) in terms of their QoL and cognitive ability, since long-term survivors may express late effects of cranial radiation.

"The long-term survival population was designed as those patients who lived more than 1 year from randomization,” said Palmer, assistant professor in the Departments of Radiation Oncology and Neurosurgery at The James Cancer Hospital at The Ohio State University Wexner Medical Center. “So, this patient is likely to have enough cognitive testing and quality of life measurements to demonstrate the impact of radiation, and because of the limited brain metastases, the hope is that these patients aren't succumbing to cognitive failure due to distant failure within the brain."

In this pooled study, the investigators also looked at differences in trends for cognitive measures (CMs). “Our hypothesis was that a 2SD [standard deviation], or what we would consider a clinically relevant SD drop in CM, would impact quality of life. That was tested with a 2-sided test with a P value of .05,” Palmer said.

The investigators looked at 2SD and 3SD drops in 1, 2, and 3 or more CMs to determine the how many patients were severely affected. More patients had a 2SD decline in 3 or more CMs at each time point in the WBRT group versus the SRS group, overall.

They observed that in the group of patients who received SRS, 2 CMs out of 6 had a linear trend that was negative; in the WBRT group, 5 out of the 6 CMs had a decreasing linear trend, according to Palmer. Word association testing was the only CM to improve in both arms. “Mostly the executive functioning and the processing speed were the main differences,” Palmer explained.

The 1SD drop with 1 CM favored the SRS arm at all time points from baseline. Once the investigators added in more CMs in the 1SD drop, it still favored SRS. For the 2SD and 3SD drops, there were not as many patients who succumbed severely to cranial radiation, but the trend toward SRS remained. Palmer said that “at 3, 6, and 9 months, there was still a statistically significant impact that favored SRS."

In terms of QoL, there were mixed results seen in the analysis. Generally, patients who received WBRT had lower QoL at 3 months compared with SRS, but SRS was lower at 2 years for patients’ physical, emotional, and functional well-being (FWB), as well as the Functional Assessment of Cancer Therapy-Brain (FACT-BR) and FACT-General assessments.

“What's important to patients is, likely, how does cognitive testing manifest into a change that may be demonstrated and affect their day-to-day life,” Palmer said. The measures showed that clinically significant declines in cognition predicts worse FWB and QoL. FWB showed a 6.4-point drop (95% CI, –11 to –1.75; P=.007) and overall QoL had a 5.1-point drop (95% CI, –7.7 to –2.5; P<.001) in patients who had 2SD or more in cognitive decline.

The median follow-up for the analysis was 24 months and patients self-reported their QoL with the FACT-BR; these scores were converted to a 0-to-100 point scale with 100 being the most favorable. One SD from baseline in 1 out of 6 CMs was used to define cognitive deterioration. The QoL and CM were modeled with baseline adjusted linear mixed models with an uncorrelated random intercepts for subject as well as random slopes for time.

_Reference:

_{Palmer J, Klamer B, Ballman K, et al. Abstract #25. Effect of stereotactic radiosurgery compared to whole-brain radiotherapy for limited brain metastasis on long term cognition and quality of life: A pooled analysis of NCCTG N107C/CEC.3 and N0574 (Alliance) randomized clinical trials. Slides presented at: 2020 Virtual Brain Metastases Conference; August 14, 2020.}