Carlson, Robert H.
ATLANTA—Whole brain radiotherapy (WBRT) has earned a bad reputation, and one reason is certainly the resulting post-treatment decline in neurocognitive function
Now, however, a Radiation Therapy Oncology Group study has shown that some of that effect might be prevented by avoiding the hippocampus via the use of intensity-modulated conformal radiotherapy.
The Phase II study reported here at the American Society for Radiation Oncology Annual Meeting (Abstract LBA1—accessible atonline.myiwf.com/astro2013/Abstract.aspx) showed that conformal avoidance of the hippocampus during WBRT for brain metastases was associated with memory preservation at four and six months follow-up. The details were presented by Vinai Gondi, MD, Director of CNS Radiation Oncology and Co-Director of the Cadence Brain Tumor Center in Warrenville, Ill.
VINAI GONDI, MD
The study, RTOG 0933, compared prospective data on patients treated with WBRT and hippocampal avoidance with data from the Phase III PCI-P-120-9801 trial (Li et al: JCO 2007; 25:1260-1266), which did not include hippocampal avoidance. The earlier trial showed a 30 percent mean relative decline in the Hopkins Verbal Learning Test Delayed Recall from baseline to four months.
Other tests—the Hopkins Verbal Learning Test Delayed Recall, Hopkins Verbal Learning Test Recall, and Hopkins Verbal Learning Test Immediate Recognition—were administered to RTOG 0933 patients at baseline and at two, four, and six months. Among 42 patients analyzable at four months, the mean relative decline in Hopkins Verbal Learning Test Delayed Recall from baseline was 7.0 percent, statistically significant in comparison with the historical control, Gondi reported.
The mean relative decline in Hopkins Verbal Learning Test Recall and Hopkins Verbal Learning Test Immediate Recognition from baseline to four months was 3.6 and 1.6 percent, respectively. And among 29 patients analyzable at six months, the mean relative decline in the three tests from baseline was 2.0, about 3.0 percent, and 0.7 percent, respectively.
There were two treatment-related grade 3 adverse events reported (fatigue and headache) but no treatment-related grade 4 or 5 events. Median survival was 6.8 months. And three patients (4.5%) had progression in the hippocampal avoidance region, which the researchers said was consistent with the expected event-rate.
Quality-of-life scores were also preserved at six months of follow-up. “These results are promising and place the hippocampus at the center of our understanding of radiotherapy-related memory lapse,” Gondi said, acknowledging, though, that the results are from a single-arm Phase II study, and that validation in a Phase III trial is needed.
One such Phase III trial now under development, he noted, is RTOG 1317, which will assess cognitive changes in patients with brain metastases from small-cell lung cancer treated with prophylactic cranial irradiation with or without hippocampal sparing.
RTOG 0933 was supported by grants from the RTOG, Community Clinical Oncology Program, and the National Cancer Institute.
The Discussant for the study, Laurie E. Gaspar, MD, Professor and Chair of the Department of Radiation Oncology at the University of Colorado School of Medicine, called the findings very timely: “We all know that whole-brain radiation therapy is currently under attack. At tumor boards you're almost embarrassed to recommend it. Your medical oncology and neurosurgery colleagues wonder why you would do that, given their perception of the high toxicity that's likely to result from such a treatment.
“And the problem is that they're partially right,” she said. “The RTOG has done a number of studies that have documented that neurotoxicity.”
She cited two trials, RTOG 0212 (Lancet Oncology 2009;10: 467-474) and RTOG 0214 (JCO 2011;20:272-278) that showed declines in the Hopkins Verbal Learning test after prophylactic cranial irradiation for both small- and non-small cell lung cancer at both six and 12 months.
“At 12 months, 50 percent of patients still have self-reported decreased neurocognitive function,” Gaspar said. Those studies also showed an increased incidence of cognitive decline for patients who had an increased dose of WBRT, as did older patients and those with baseline cognitive dysfunction, she said.
Hippocampal involvement with loss of neuronal stem cells and their differentiation is one of the primary problems with radiation-induced cognitive impairment, Gaspar said, as well as inflammation resulting in vascular changes.
“The strategy of Dr. Gondi and his RTOG colleagues to reduce the radiation dose to a specific target has been shown to be very effective.” But there are also medical strategies available, she said, including treatment with memantine, lithium, proliferator-activated receptor (PPAR) agonists, and ACE Inhibitors.
Memantine has been studied the most and has been shown to reduce the decline and severity of memory loss, and in some cases temporarily improve cognition in patients who have Alzheimer's or other forms of dementia, Gaspar noted.
She cited the Phase 3 RTOG 0614 trial presented at the 2012 ASTRO Annual Meeting, which showed significant overall delay in the onset of cognitive decline with memantine during the course of WBRT for metastatic brain tumors.
Gaspar said she will support the upcoming RTOG 1317 trial that Gondi mentioned, of prophylactic cranial irradiation with or without hippocampal sparing.
But she also mentioned another planned Phase III RTOG study that will test memantine with or without hippocampal avoidance in prophylactic cranial irradiation for brain metastases.
“They're going to stratify patients for baseline cognitive impairment, which I think will further define who this study is going to be useful for.” But, she added, “I wish they were using 30 Gy instead of 37.5 Gy, because PCI [prophylactic cranial irradiation] studies have shown us that the dose is important.”
What Memory Protection Could Cost
“Somebody's probably going to ask at some time what all this costs,” Gaspar said, leading into an estimate she made using Medicare charges for a regimen of WBI 30 Gy in 10 fractions, with six months of memantine plus IMRT, such as would be needed for hippocampal avoidance.
She said this hypothetical regimen would approximately double Medicare charges for WBI alone. “Hopefully there won't be any insurance issues, but I'm going to test that in Colorado,” she said.
LAURIE E. GASPAR, MD
Gaspar concluded her discussion by saying that hippocampal avoidance is likely effective and feasible based on the data Gondi presented, and that it would be reasonable for clinicians to discuss both memantine and hippocampal avoidance with patients now “because [the treatments] work in different ways.”
Meanwhile, she said, she hoped better selection criteria could be found—“because eventually we will have to justify who this treatment should be used for.”
Asked to comment for this article, Christopher A. Barker, MD, Assistant Attending Radiation Oncologist at Memorial Sloan-Kettering Cancer Center, said via email, “These findings are encouraging, and suggest that avoidance of the hippocampus may help prevent some of the neurocognitive sequelae of whole-brain radiotherapy. However, a prospective randomized trial will be necessary to confirm this.”
Future studies, he added, will be necessary to know which patients are the best candidates for this approach, and whether avoidance of other regions of the brain involved in learning and memory would afford additional preservation of neurocognitive function.