Screening for Brain Metastases

An Interview with Andrew Seidman, MD, Memorial Sloan-Kettering Cancer Center

Part 1: Screening for Brain Metastases 

Musa Mayer: I’d like to begin by asking you about the advisability of screening for brain metastases, both for HER2-positive and HER2-negative metastatic breast cancer patients.  

Andrew Seidman:  First, I would like to distinguish between screening patients who have earlier stages of breast cancer—Stages I, II and III—and patients who already have known metastatic breast cancer (Stage IV).  I don’t want patients with early breast cancer to think that when they’re diagnosed with early breast cancer, with or without lymph node involvement, that the standard approach would be to do a brain MRI or CT scan in the absence of any neurological symptoms.  That’s an important message to get across, that this is not standard of care.  For women with early-stage breast cancer, the likelihood of having brain metastases at the time of diagnosis in the absence of any neurological symptoms is so small that it wouldn’t be needed as a screening test.

 MM:  Would that be true for even for locally advanced (with positive lymph nodes) HER2-positive disease? 

AS:  Well, that’s a very slippery slope. That’s a scenario that is driven more by opinion than by any data, I’ll be the first to admit. I think in the scenario of the patient with locally advanced HER2-positive breast cancer, it is not unreasonable to screen for brain metastases, since we know that patients with HER2-positive breast cancer have a higher chance of having brain metastases.  Even in the absence of finding other sites of metastases, such as bone, lung, liver or others, it’s not unreasonable in someone who has locally advanced disease and is HER2-positive to do screening, an imaging study of the brain.  Certainly for patients who present with, or relapse with, metastatic breast cancer that is HER2-positive, doing a baseline or screening brain scan—typically we do MRI, some centers prefer to do CT scans—is a reasonable scenario. 


You asked me a question, though: does any woman with metastatic breast cancer who has no neurological symptom need to have brain imaging done as part of her initial extent-of-disease evaluation, or on a regular basis?  I would probably say no.  I think the standard of care is that most of us are not in the habit of doing periodic brain imaging when there are no neurological symptoms whatsoever.  The yield of finding what you’re looking for, which is metastatic disease in the brain or leptomeninges, is exceedingly small.  The chances are that possibly you might find other abnormalities that might have nothing to do with breast cancer.  We know, for example, that women with breast cancer do have a higher incidence of meningiomas, which are often benign, non-threatening growths in the brain.  So, absent neurological symptoms in patients with HER2-negative metastatic breast cancer, it’s not my habit to do screening brain imaging.

 MM: Is it your belief, or is there any data suggesting that there is an advantage to catching a brain metastasis early, before it is symptomatic? 

AS: This is my feeling and my bias: patients who present with neurological symptoms for their brain metastases do so for one of two reasons.  One is related to the volume of disease.  Simply having a large enough lesion, or a large enough number of lesions in the brain can lead to the development of headaches, discoordination, problems with gait or walking, or with speech, etc. 


The other scenario is that it may not be the size or the number of lesions, but simply the critical location.  So a patient can have a small volume of cancer to the brain in a critical place in this very complex organ that we call the brain, with its very complicated neural interconnections. Even a relatively small lesion, in a bad place, can result in significant neurological disability.  So, it’s my bias, even though any survival advantage for screening and earlier detection hasn’t been clearly demonstrated, one might be able to reduce neurological morbidity, even if it doesn’t affect mortality.  Earlier interventions by other modalities such as craniotomy for resection of a solitary lesion, or stereotactic radiation or Gamma Knife approaches, could potentially offer the patient a better quality of life because of the absence of neurological symptoms and deficits. 


This really remains to be better studied and proven in the context of prospective clinical trials.  It’s such a difficult question that I’m not certain that those trials could really be done, that patients would be willing to be randomized, although I believe that there are such trials that have been proposed and may even be getting underway, looking at the role of screening and therefore early detection of brain metastases compared to when patients present symptomatically. 

You’ll recall that George Sledge (research oncologist, also a member of the COE) and I agreed to disagree about the proper role of screening.  George cited the study that his colleague Kathy Miller, at Indiana University, published in the Annals of Oncology, where they compared a group of women who were screened because they were going on a clinical trial that involved Avastin, which excluded patients who would be at a higher risk of central nervous system disease bleeding due to the Avastin.  In screening, they identified patients who had occult (undiagnosed), previously unsuspected brain metastases, and they compared their outcomes—specifically, just their survival—with a group of patients whose brain metastases were not diagnosed with this approach, by screening, but who were diagnosed as part of their routine care, when they developed symptoms.  In this analysis, they didn’t show any difference in overall survival, these women whose cancers were detected by screening, as compared with the women whose brain metastases were detected by routine clinical care.  I don’t think this is the end of the story, and this study didn’t address neurological outcomes.  It only looked at survival.


Interview continues in Part 2...