Children's Solid Tumors Rarely Metastasize to the Brain

Daniel M. Keller, PhD

July 24, 2013

BARCELONA, Spain — An analysis of a series of cases of children with solid primary tumors challenges the notion that these tumors often metastasize to the brain.

But regardless of the management strategy, the prognosis remains poor once brain metastases appear, probably because of advanced disease when they are diagnosed, a study from M.D. Anderson Cancer Center shows.

Dima Suki, PhD, professor in the Department of Neurosurgery at the Houston, Texas, cancer center, said that "in the absence of routine surveillance, changes in neurological status should alert oncologists to the possible presence of a brain metastasis." In contrast to brain metastases, lung metastases were quite common.

Dr. Suki presented these data at the 23rd Meeting of the European Neurological Society (ENS).

Improved Detection

Brain metastases have traditionally been recognized in 1% to 20% of pediatric cancer cases vs 20% to 40% of adult cases. However, recently there has been a presumed increase in pediatric cases, possibly resulting from improved detection using better neuroimaging techniques and from better treatments of the primary cancer, resulting in longer survival and time for brain metastases to occur.

This analysis depended on a retrospective review of data from 1990 to 2012 from M.D. Anderson's tumor registry. Included were patients with parenchymal, epidural, and subdural metastases; patients with calvarial and supragaleal metastases invading the epidural space and beyond; and those with leptomeningeal involvement. Patients with metastases to the calvarium only or with direct tumor extension into the brain from an extracranial primary tumor were excluded. Autopsy reports were not reviewed.

Of 7625 cases extracted from the registry, 3950 had a solid primary cancer originating outside of the central nervous system. Of those, only 54 were diagnosed with a metastasis to the brain, for a frequency of 1.4%. The median age at which the primary cancer was diagnosed was 11.37 years (range, birth to 17.96 years). The median age at diagnosis of a brain metastasis was 15.03 years (range, 0.24 to 20.74 years). About half the patients were male.

Sixty-five percent had symptoms leading to a diagnosis of a brain metastasis, most commonly headache (33%), nausea/vomiting (27%), or seizure (23%). A single site was seen at the time of diagnosis of brain metastasis in 60% of cases, 34% had multiple sites, and 6% had only leptomeningeal disease.

Location, Histologic Type, and Origin

Most of the brain metastases were parenchymal (72%; 89% of which were supratentorial, 9% infratentorial, and 2% both), with the rest dura-based (20%), leptomeningeal (6%), or calvarial and epidural (2%).

In 85% of cases, a primary tumor was diagnosed before the brain masses with a median of 15 months between diagnosis of the primary and the brain metastasis. About half (48%) of the primaries were localized when they were diagnosed, 14% had regional invasion, and 38% had distant spread. At the time of brain metastasis diagnosis, 72% of patients had no evidence of the primary cancer, 16% were stable or responding to treatment, and only 12% of primary disease was progressing.

In contrast, extracranial metastasis was common (92%) when metastasis was detected in the brain. The most common sites were lung (70%), lymph nodes (26%), extraspinal bone (15%), and, less commonly, spine, liver, soft tissue, and other sites.

Excluding hematologic cancers, the most common primary tumors were sarcomas (54%), which were predominantly osteosarcomas. Fifteen percent of primaries were malignant melanomas. The most common histologic types of brain metastases were choriocarcinoma (43%) and clear cell sarcoma of the kidney (30%). Only 8% were of neuroendocrine or of peripheral neuroectodermal (3%) tumor histology.

Poor Survival Regardless of Treatment

"The prognosis remains bad regardless of the management strategy, probably because of very advanced disease at the time of diagnosis of the brain metastases," Dr. Suki reported. The median overall survival from the time of primary diagnosis was 29 months (95% confidence interval [CI], 24 - 34 months).

From the time of diagnosis of a brain metastasis, survival with treatment was a median of 8 months (95% CI, 6 - 11 months) from initiation of therapy vs 0.9 month (95% CI, 0.3 - 1.5 months) without therapy (P = .03).

Treatment of brain lesions (91% of cases) consisted of surgical resection (25% of treated cases), whole-brain radiation therapy (10%), chemotherapy (8%), stereotactic radiosurgery (4%), or combinations of these modalities.

Dr. Suki commented that brain metastases generally follow extensive systemic metastasis and disease progression but were rare as the initial site of presentation of a solid primary tumor elsewhere in the body. Lung metastases were common when brain metastases were first seen.

She said extensive disease at the time of diagnosis of brain metastases probably argues against routine brain surveillance imaging. Systemic chemotherapy may prevent systemic metastases but because of the blood-brain barrier does not prevent brain metastases.

Dr. Suki said the findings of this study challenge the notion of an increased incidence of brain metastases among children with a solid primary tumor. She said regardless of management strategies, prognosis is poor with current treatments. Any promising new agents are still in the early stages of development, she noted.

When asked for a comment, session moderator Riccardo Soffietti, MD, professor of neurology and neuro-oncology at the University Hospital of Turin, Italy, who did not participate in the research, told Medscape Medical News it was "a great study from a great group."

He said what impressed him the most is the low frequency of brain metastases in the pediatric cancer population compared with adult patients. "This is intriguing. The tumor types are different and probably for most tumor types in the pediatric population, the propensity to colonize the brain is lower. This could be interesting to look at biologically," he proposed.

In the adult population in the last 10 years, he said, the frequency of brain metastases has increased at least 10-fold, and there are "huge numbers." Some of the higher numbers may relate to better diagnostic techniques over time. But with a low frequency of brain metastases in the pediatric population, it is hard to detect any trends in the prevalence of pediatric brain metastases over time, and Dr. Soffietti said he suspects the prevalence has been fairly stable.

In practical terms, "If we look at some clinical implication in daily management, for sure [there is] no need for surveillance policies with these [pediatric] patients because the risk is very low," he advised. He noted the pediatric situation is different from adults, citing, for exampIe, breast cancer, where there is an increased incidence of brain metastases and therefore a need to monitor patients when they are symptomatic.

There was no commercial funding for the study. Dr. Suki and Dr. Soffietti have disclosed no relevant financial relationships.

23rd Meeting of the European Neurological Society (ENS). Abstract O257. Presented June 9, 2013.


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.
Post as: