Cognitive Outcomes Improved With Proton Therapy in Pediatric Brain Tumors

Jeffrey Gross, MD

Favorable neurocognitive outcomes were observed in pediatric patients with brain tumors with the use of proton radiation therapy (PRT) compared with X-ray radiation therapy (XRT), according to findings from a study published in the International Journal of Radiation Oncology, Biology, and Physics.

Investigators assessed for IQ, processing speed, visual motor integration (VMI), parent-reported function, executive function, and memory in 125 pediatric patients treated for brain tumors. When compared with XRT, PRT was associated with a higher full-scale IQ, processing speed, and parent-reported practical function.

“Essentially, it has been well described that within years after completing radiation therapy for brain tumors, children develop and experience a number of problems related to attention and processing speed,” said lead author Jeffrey Paul Gross, MD, MS, radiation oncology resident, Northwestern University, in an interview with OncLive. “Often, they had a lower IQ than their peers who had not undergone any such treatments. A lot of that is likely due to the cancer itself in the brain, the surgery, chemotherapy, and the radiation. But, we know that radiation plays a big part in that.”

This led to the interest of using proton therapy, which delivers radiation to less of the healthy brain tissue than the conventional approach with XRT, Gross explained. During PRT, protons deposit their dose directly to the tumor, while conventional radiation by XRT continues beyond the tumor. PRT is hypothesized to be less harmful to children’s brains, which are more sensitive to radiation than adults as they are in rapid development, said Gross.

In the multivariate analysis, investigators also tested interaction with follow-up time. Findings showed higher full-scale IQ of β = 10.6 points per year, (P = .048); processing speed of β = 12.6, P = .02; and parent-reported practical function of β = 13.8, (P = .049); following PRT relative to XRT. Following craniospinal irradiation (CSI), there was lower processing speed of β = -15.9, (P = .04); and VMI of β = -14.0, (P = .006). Additionally, there was a higher verbal IQ of β = 0.84, (P = .02); and full-scale IQ of β = 1.03, (P = .01) observed in older patients.

According to the univariate analyses, the mean verbal IQ for those who received PRT was 99.6% compared with 92.9% for those who received XRT (P = .03). In the PRT and XRT populations, the assessment showed a full-scale IQ of 99.6% versus 88.6%, processing speed of 86.9% versus 80.0%, VMI of 87.1% versus 80.8%, general adaptive composite of 91.4% versus 80.7%, conceptual function of 94.6% versus 84.1%, social function of 94.8% versus 86.2%, and practical function was 91.1% versus 78.9%. There were no significant differences in long-term memory or digit span.

"With proton therapy, less of the child's healthy brain tissue is receiving a significant dose of radiation. We did find that this seemed to be supported by the data, in that children who were treated with protons had less of an impact with regard to their full-scale IQ, and also in terms of their parent-reported adaptive skills," Gross said.

The subgroup analysis conducted showed PRT was associated with a higher full-scale IQ compared with XRT following craniospinal irradiation (CSI) or whole ventricular radiotherapy (n = 76), β = 17.2, (P = .02); and higher processing speed, β = 22.6 (P = .01); following partial brain radiotherapy (n = 49).

In patients receiving CSI or whole-ventricular radiotherapy, there were differences in full-scale IQ, investigators concluded. There were also differences in processing speed in those who received partial brain RT.

The median age was 7.0 years at diagnosis. Of the 125 enrolled, 53.6% had tumors located in the posterior fossa, 28.8% were in the midline compartment, and 17.6% were in the supratentorial compartment. Of all subgroups, parent-reported outcomes were lowest for those with posterior fossa tumors, with β = -10.8, (P = .048).

"We assumed that there was a cognitive benefit, but we have never been able to demonstrate that this study was really the first comparison of conventional radiation with proton therapy, to provide that demonstration of benefit from a cognitive perspective,” noted senior author Vinai Gondi, MD, director of Research & Education at the Northwestern Medicine Chicago Proton Center in an interview with OncLive. “What we are observing in this analysis is somewhat of a proof of principle that when we are limiting the exposure of healthy brain tissue to radiation dose with proton therapy, we are limiting the impact the radiation has on the development of that child's brain.”

Investigators noted that patients who received PRT had a higher socioeconomic status than that of patients who underwent XRT. There were also different distributions of race and tumor location between the groups, and a shorter median time from treatment conclusion to last neuropsychological assessment for those who received PRT. Although, Gondi noted that proton therapy is becoming more accessible.

“The use of proton therapy has become increasingly accessible to patients because of the development of new proton therapy centers across the country and the world. What we have come to understand is that children with brain tumors can live a very long time and can often be cured of their tumors. We have come to appreciate that there are some long-term side effects of radiation therapy,” he concluded.

Gross J, Powell S, Zelko F, et al. Neuropsychological outcomes of pediatric brain tumor patients treated with proton (PRT) or X-ray (XRT) radiation therapy. Int J Radiat Oncol Biol Phys. 2018;102(3):S52. doi: 10.1016/j.ijrobp.2018.06.154.