Kate Johnson

June 01, 2015

CHICAGO — Whole-brain radiotherapy (WBRT) should be avoided in patients with a limited number of brain metastases because it does not prolong survival, it reduces quality of life, and it causes cognitive decline.

That was the conclusion from investigators of the phase 3 N0574 study, presented here during a plenary session at the American Society of Clinical Oncology (ASCO) 2015 Annual Meeting.

But arguments countered by the expert discussant for the study indicate that the long-standing debate will continue as to which is more harmful ― the brain metastases or the radiotherapy that treats them.

"Since whole-brain radiation does not prolong life and since it increases the risk of cognitive decline when combined with stereotactic radiation, we expect that it will be reserved for palliation of symptomatic metastases in patients for whom stereotactic radiosurgery [SRS] is not possible," investigator Jan C. Buckner, MD, professor of oncology at Mayo Clinic in Rochester, Minnesota, told Medscape Oncology.

"For patients with newly diagnosed brain metastases that are amenable to SRS, we recommend initial treatment with SRS alone and close monitoring to better preserve cognitive function and quality of life," added lead investigator Paul D. Brown, MD, from the University of Texas MD Anderson Cancer Center, in Houston, during his plenary presentation.

But challenging that notion, discussant Andrew Lassman, MD, chief of neuro-oncology at Columbia University Medical Center, in New York City, countered that WBRT does boost survival "in the appropriate context" and that deferring WBRT "leads to a rapid and more numerous occurrence of brain metastases, which also cause neurocognitive injury."

"Rather than a broad statement that observation is always favored over WBRT, I think WBRT can and should be used in certain contexts," he concluded.

Cognitive Decline Documented

N0574 included 208 patients with one to three metastases who were recruited from 34 institutions during a 12-year period (2002-2013).

Approximately 70% had a primary lung cancer, and 50% had a single brain metastasis.

After stratification by age, extracranial disease status, number of brain metastases, and institution, the patients were randomly assigned to receive SRS alone or in combination with WBRT.

The primary endpoint of the study was cognitive function at 3 months, with cognitive decline defined as a drop of 1 standard deviation from baseline in at least one of a cognitive battery of tests assessing memory, processing speed, executive function, verbal fluency, and motor speed dexterity.

Assessments were performed at baseline, at 6 weeks, and at 3, 6, 9, and 12 months; in addition, serial MRIs were taken, and the Functional Assessment of Cancer Therapy–Brain instrument was administered to assess quality of life (QOL).

After a median follow-up of 7.2 months, the study showed that patients who were treated with WBRT were significantly more likely to have stable disease at 3 months, but this did translate to better survival.

Specifically, WBRT significantly improved time to intracranial progression in terms of central nervous system (CNS) failure at both 3 and 6 months (6.3% vs 24.7% and 11.6% vs 35.4% respectively) and beyond (P < .0001).

"But in spite of the imaging evidence of disease control, there was no overall impact on survival in these patients," explained Dr Buckner. In fact, median overall survival was 7.4 months in the WBRT arm vs 10.4 months in control patients (hazard ratio [HR], 1.02; P = .92).

WBRT patients were also more likely to have cognitive decline than patients treated with radiosurgery alone (91.7% vs 63.5%; P = .0007).

This reached statistical significance for immediate recall (30.4% vs 8.2%; P = .0043), delayed recall (51.1% vs 19.7%; P = .0009), and verbal fluency (18.6% vs 1.9%; P = .0098), which Dr Brown described as "the central component of communication."

This higher rate of cognitive decline persisted in WBRT patients at 6 months (97.9% vs 77.8%: P = .032) and even in a small subset of patients who were still alive at 1 year, he added.

Additionally, patients treated with WBRT were more likely to have a decline in almost every QOL test compared with those treated with radiosurgery alone, reaching both clinical and statistical significance for functional well-being and overall QOL, he said.

Finally, WBRT patients had a higher rate of acute radiation side effects at 6 weeks, namely, alopecia and dermatitis, although there were no differences in the rate of late radiation side effects, such as CNS necrosis.

"To the best of our knowledge, these results apply to all patients with all types of cancer," said Dr Buckner.

"If at all possible, use either no radiation, low-dose radiation, hippocampal-sparing radiation, or a combination of radiation and memantine as a way to reduce the risk of cognitive decline because, essentially, the brain does not like to be irradiated," he concluded at a press conference.

Also speaking at the press conference, Brian Alexander, MD, ASCO expert on CNS cancer from the Center for Neuro-Oncology at the Dana-Farber Cancer Institute in Boston, said "this was a very large, well-done study to quantify both the risks and benefits of this treatment, so I think the burden of proof has probably now switched to say can we prove that WBRT is beneficial in a subset of patients."

And pediatric oncologist Gregory Armstrong, MD, from St. Jude Children's Research Hospital, Chicago, pointed to the special importance of preserving cognitive function in children. "If you consider that a 5-year-old with leukemia or a 10-year-old with medulloblastoma has 60 or 70 years ahead of them, preserving their cognitive ability by reducing radiation doses is essential, it's the hallmark of what we've done in our field and what we continue to do."

But the study discussant, Dr Lassman, noted, "I think there are other interpretations when placing this study in the context of other trials for brain metastases."

Dr Lassman said the N0574 study did not have survival as a primary endpoint and therefore was not designed to assess whether WBRT improves this outcome. "It is unrealistic to expect a difference in survival for more aggressive treatment of brain metastases unless the brain metastases are the life-limiting site of disease."

Additionally, he said survival results can look very different when assessed according to patients' Graded Prognostic Assessment (GPA) score, which takes into account several factors of known prognostic significance, including presence or absence of extracranial metastasis as well as age, performance status, and number of brain metastases.

For example a Japanese study first published in 2006 ( JAMA. 2006;295:2483-91) found no survival difference in patients treated with radiosurgery alone or with WBRT (7.5 months vs 8 months), but further analysis just published this month demonstrates a clear prolongation of survival among WBRT patients (16.7 months vs 10.6 months) with lung cancer and favorable GPA scores (from 2.5 to 4) but no difference among other patients (6.5 vs 4.75 months).

"The data from N0574 have not been analyzed yet according to GPA score, and we eagerly await that further analysis," he noted.

Dr Lassman also questioned the neurocognitive outcomes in the N0574 trial.

"This is the largest trial to date studying the neurocognitive effects of WBRT with such a detailed battery of neurocognitive evaluations...[but] the results are discordant with other studies that did not demonstrate such persistent deterioration and even showed some improvement," he said.

Specifically, a Japanese study as well as others have shown an initial neurocognitive decline after WBRT followed by improvement. "Patients who responded to WBRT had improvement in several neurocognitive domains, whereas those who did not respond had clear worsening," he explained. "The major cause of neurocognitive decline was recurrence of metastases and not WBRT."

Asked by Medscape Medical News to comment further on these findings, Gerald Fogerty, MD, from the Melanoma Institute of Australia, in Sydney, Australia, and lead investigator of an ongoing study of WBRT for intracranial melanoma metastases, said, "We need more data," explaining that different cancers predispose to different problems with WBRT and neurocognitive function.

"For example, smokers and those who have had much chemo will have worse reactions to WBRT," he said.

He also noted that a recent study showed that hippocampal-sparing WBRT can improve neurocognitive outcomes ( J Clin Oncol. 2014;32:3810-6).

"This will now be unable to be shown in a randomized controlled trial if WBRT is buried," he said.

This study received funding from the National Institutes of Health. Dr Buckner participated in a consulting or advisory role with Merck Serono and has received expenses for travel and accommodations from Genentech/Roche. Dr Brown, Dr Lassman, and Dr Fogerty have disclosed no relevant financial relationships.

American Society of Clinical Oncology (ASCO) 2015 Annual Meeting. Abstract LBA4. Presented May 31, 2015.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.