When to do a Brain MRI - 1259840

Sat, 10/12/2013 - 17:37

My wife was diagnosed in August of 2012 with NSCLC and had a CT/PET scan and brain MRI done initially. The MRI was clear as were lymph nodes and the rest of her body. The cancer was declared T4N0M0 and stage IIIA. A pneumonectomy for the tumor was performed and we crossed our fingers. Unfortunately in March, 2013 a scan showed that the cancer metastasized in her right hemithorax and 6 months of induction chemo followed.

So here we are 13 months from the initial diagnosis and the brain MRI and 7 months since metastasis. Although my wife has no symptoms seeming to indicate the need for another brain MRI, shouldn't we be getting one? Our onc says without symptoms, they are usually not prescribed. Does this make sense?


Revision log message
Created by FeedsNodeProcessor


Hi Tom,

What your doctor is doing is pretty typical. As Dr. Weiss has said "I typically acquire an MRI at diagnosis, then again if symptoms suggest. Some oncologist obtain more regular MRIs. I have no objection to this, but am not convinced that it improves actual patient outcomes compared to having a low threshold to get MRI once symptoms develop." - http://cancergrace.org/lung/topic/should-brain-mri-be-done-even-in-abse…

And in the next reply, Dr. Aggarwal agreed: "I agree with Dr. Weiss. I obtain MRI brain if symptoms (headache, blurry vision, uncontrolled nausea, seizures etc.) develop, after getting a baseline scan."

Forum moderator

Dr West

I follow the same approach. It is not the standard of care at this time to do surveillance brain MRI scans after diagnosis in people with advanced NSCLC who don't have evidence of brain involvement at their initial work-up.

-Dr. West


I'm going to do a big time disagree with the doctors.

I had read too many instances where people had unmonitored brains (especially on tarceva which I was on), only to have symptoms, get checked and have multiple mets, some very large. People who had had brain mets over the years, strongly suggested getting a brain mri once every six months. I discussed with my doctor's nurse practitioner and she also felt that every six months made sense and I started to have them regularly every six months. In May, without any symptoms whatsoever, 5 brain mets were discovered on MRI, the largest 8mm on the cerebellum. I had them zapped a little over a week later with gamma knife.

BTW, when they were found, the doctor was surprised that I was getting them checked every six months and wanted to know who recommended. I told him his NP. He told me that surprised him because he only checks with symptoms.

Now, because I am being followed by my GK neurosurgeon and have already had brain mets, my brain is scanned every 3 months. I just had one of those scans and it was fine.

I don't know that you can sway your doctor, but especially in light of your progression elsewhere (I progressed too around the time of the mets although I did not know when I had them taken care of because it was prior to my pet/ct) I would push for the scan. I wont say it is the most above board thing to do, but I've heard of many patients who lie about having headaches to get those scans if that's what it takes.


Dr West,

I appreciate your response to my post, but I have to say that this issue is puzzling to me. Granted I am a layman and you are the expert, so perhaps I shouldn't question further. Nevertheless, I'd appreciate it if you could share a bit more of your experience on this issue. Specifically,

1 - Are there any studies as to what percentage of Stage IV NSCLC metastasis include brain mets?
2 - Are there any studies which evaluate the success of gamma knife or other brain met treatments as a function of how large the brain met was? (as a surrogate for how long it was growing?)
3 - I see anecdotal evidence from forum members about having brain mets with no symptoms. Is that usual?

It just seems that with no questions as to the need for PET/CT scans every 2 months during chemo, that the cost of brain MRIs can't be that prohibitive - even at once a year. It's just puzzling to me, especially when a brain MRI is not deemed appropriate without symptoms even for someone who never had a brain MRI after metastasis was first found.

As always, thank you for your contributions to this site. It is greatly appreciated.


Dr West

About 10-15% of people with advanced lung cancer will have brain metastases at diagnosis.

It is worth noting that that there is NO evidence that people with metastatic lung cancer live longer or do significantly better when getting repeated PET/CT scans instead of CT scans, and the example you provide is a great one of colossal waste. Look through the website and you'll find countless examples of me and many other experts saying that they are done gratuitously, and most experts don't favor them. They are often done in part because doctors may profit by sending for them.

It's not surprising that people with brain mets say scans are good, but that's a biased group, and not the same as what the evidence says. Yes, some people may develop asymptomatic brain mets. But there isn't actual evidence that doing scans based on symptoms leads to worse outcomes than doing scans every 6 months. It's not that doing scans every 6 months is a terrible idea, but most people who don't have brain mets at their initial diagnosis won't have them later. A few will, but:

1) Gamma knife is promoted over WBR in part because it's very profitable to do compared with WBR. It also doesn't treat the rest of the brain and has a high risk of future brain mets that can then be treated with more (highly profitable) gamma knife (bonus!). Practice patterns would be different if WBR were more profitable.

2) Gamma knife is not associated with living longer than WBR.

3) It may be possible to do 200,000 extra head MRIs/year, at $2500/MRI, to find ~5% with new brain mets. It's not clear if even in that 5% benefit by finding the brain mets a month earlier than they'd have been found based on a new headache or vision changes, but a unilateral policy change would cost the US health care system $500,000,000/year that isn't available.

If people want to pay for them, by all means, but how can we demand that society cover the cost with unclear/dubious benefit?

-Dr. Wes


With all due respect, the percentage of patients presenting with it are lower than the percentage of patients who either present with brain mets or go on to develop it. I cannot find the info, but isn't that more in the 40% range. Also, isn't it possible that with some of the better outcomes from targeted therapy for mutations, as more people live longer, they may be more susceptible. Since I've been on tarceva, I focused in on brain mets in people on tarceva, many who did not have regular mris, first sign of a met was a seizure and brain was riddled with mets at that point. I always figured as I do not get headaches, that I would notice mets even with a slight headache. Nope. I did have a very small fuzzy feeling develop between two toes - my only possible symptom and one that most doctors were attributing to neuropathy.

GK is extremely expensive. However, for a young person whose cognitive ability may be impaired by wbr, it often makes sense. Plus, I have seen many times where people have wbr and the pesky mets come back but they seem more limited in their further treatment by stereotactic surgery. Highlight that with patients who have several gk's and then keep progressing and then use wbr as a last stand. They seem to go a little longer. When you are young (relatively) like I am (early 50's with two teens), the name of the game is going as long as possible. Not because I fear death, but because I need to help ease my kids into adulthood.

As to cost of us healthcare. I think it is as much to blame on the layers that insurance companies have added to the process to deny claims. The cost of MedSolutions and companies like it. The costs of extra time spent by doctors in peer reviews and written appeals and the need for additional staff. The amount of extra staff needed for it. There are plenty of ways that insurance could reign in costs just by putting more control back into the hands of doctors.


Interesting discussion. Please allow me to chip in with my humble opinion.
I do understand from an evidence based perspective, there is no proven beneficial treatment intervention from early detection of progression vs detection upon appearance of symptoms on advanced lung cancer. However, medical research is pushing new boundary of either (1) New drug or treatment for old problem, or (2) New or expanded application of old drug/treatment. For example, the following link quoted what was possible benefit of early SBRT treatment to oligometastases as "Dr. Okunieff suggests that perhaps metastases can generate more metastases and so just leaving oligometastatic disease alone when it appears stable may mean that we miss a window of opportunity".
While medical professionals have the perspective of the whole spectrum of patients, we as patients (and caregivers) may be allowed to focus on a narrow interest of the well being of one individual case, and can rightly choose to maximise the benefit not based on evidence based knowhow alone, but also on judgemental calls based on individual unique circumstances. I, for one, have elected to do regular scans on body and brain for my wife. Multiple brain mets and possible lepto was detected and treated 9 months ago (with WBR) before any related symptom appeared, and she is now again undergoing SBRT to deal with an isolated and asymptomatic lung nodule. She is having a relatively symptom free quality of life which may or may not be due to the early intervention. Even if this is not helping longitudinally, I am satisfied and happy if it has helped with quality of life.

dr loiselle

Hi -

Dr. West pointed out this thread to me.

I am happy to comment from the perspective of a radiation oncologist. I like this topic because it is one that I often ponder.

Looking through the above, while surveillance brain scans for asymptomatic patients with lung cancer do not have an evidence basis toward an outcome of prolonging life, treating small asymptomatic brain metastasis with radiosurgery is indeed the lowest impact to a patient's quality of life of any effective brain targeted treatment (surgery, whole brain, radiosurgery, etc.). Of course, some of these patients may have never develop symptoms, and may have lived out their life with asymptomatic brain metastases never requiring treatment. However, others would go on to develop symptoms, perhaps require surgery and radiation that dramatically lowers their quality of life (and is very expensive).

Recently, a patient of mine with Stage IV NSCLC was planning a long trip outside of the country to remote geographic areas. Systemic disease was controlled - we discussed various risks for progression and decided to rescan the brain. It turns out there were multiple small brain metastases. We treated them with radiosurgery before the trip....

Overall, I agree that restaging scans can be done recklessly and can rack up dramatic costs to patients and the healthcare system. However, in select cases, especially where disease is otherwise controlled, I think the relative risk for a brain failure heightens, and I think that early intervention for at least some patients is ideal.

With regard to costs, reimbursement for radiosurgery has generally gone down. It may be less expensive than hippocampal sparing whole brain radiation (IMRT). With regard to the cost of the MRI, it is possible to do a limited screening brain MRI... it is quick, and should be less expensive.

Dr Loiselle


So, we have an interesting discussion now! I appreciate Dr West passing this on to a radiation oncologist for a different perspective.

Dr. Loiselle, I find your views an important contribution to this discussion. It interesting, however, that you suggest a criterion of "when the disease is otherwise controlled." In my layman mind, I would suggest that a criterion for brain checks should be when the disease is NOT under control and the potential for metastasis to the brain is higher.

Nevertheless, I go back to my original scenario with a patient (in this case, my wife), having gone through induction chemo only to see progression, yet never having a brain MRI AFTER discovery of metastasis. Her only MRI was pre-metastasis and was clear. Doesn't it make sense that a new "baseline" scan be done post metastasis, notwithstanding absence of symptoms?


Dr West

It's clearly an arguable, controversial question, but I would say that it's wrong to argue about the merits of a policy working back retrospectively from one person's experience. The question is whether your wife's case is 1 in 4 or 1 in 25, what the cost is for surveillance scans, and whether there is a major, demonstrable difference in real outcomes whether a patient has metastases identified by surveillance scans or symptoms.

People's stories are so different that it's not instructive to work backwards from one person's story to draw conclusions about what everyone should do. That's a common marketing strategy -- "my local doctor said I had 4 months to live and here I am, 4 years later, doing well because I decided to go to ____ for my treatment" -- but most doctors have patients who are doing unfathomably well and others who are doing unfathomably poorly.

I don't think this question is going to be or should be settled by using one person's case as a representative sample for a national policy.

-Dr. West


I think if you go on a forum site answered more by patients, many patients will have a different opinion on the value of scans. Thankfully I did and had listened to them. I think your gut is strongly telling you something. I know that my gut did when something was wrong even though I did not have symptoms. You know your wife best and subconsciously you may be perceiving something that does not feel right and makes you want that scan. Sometimes you need to go with your gut feeling. Doctors can give you rational explanations, but until they are the patient and know the sort of anxiety that can be alleviated by a scan, they will not understand its true value. It is more than a tool to find mets. It is a tool to not find them and to put worries over them at ease.

Dr West

But again, you're saying that half a billion dollars of funds that aren't available and would have to paid in additional taxes or health care premiums that lots of people don't have should be paid for no reason other than the opinion that patients simply want them.

My point is that national policies that require hundreds of millions of societal dollars to follow are not and should not be made by the opinion or gut feeling of patients or doctors or anyone. That doesn't mean that patients can't decide to pay for interventions based on their opinion or gut feelings, but we don't have societal resources to have the best evidence supporting a policy be a gut feeling.

-Dr. West


I believe that the consequences on macro level health care costs for spending high sums on low probability events is a reasonable discussion, and the differences between how a patient, a doctor and a health care policy maker thinks will likely be different. I, for one, have learned quite a bit and respect the views presented.

One issue that may have been lost in the more global discussion is the question of whether or not a Brain MRI should be considered AFTER METASTASIS IS DISCOVERED. I recognize that many people initially diagnosed as Stage IV would have had their MRI done then and according to some, shouldn't have more without symptoms. However, how about for those that were initially diagnosed as Stage I, II or III, had their brain MRI done then and then progressed to Stage IV afterwards. Isn't that a significant difference?



tvoltagg - I think you will still get a different answer from patients and doctors. I think that if there is progression, it is possible, and it has happened to enough of us (asymptomatic brain mets that present with progression) that an mri is a reasonable request.

As to Dr. West's last comment, for some brief historical context - my husband worked for a major health insurance company in the 1990's around when contracting of radiation decision services to outside companies (i.e. Med Solutions) was started. He has told me that they were contracted specifically to make denials easier while buffering litigation against the health insurance company - a profit motive. Also, my mom worked for the largest pharmaceutical lobby in the 1980's. There is more than enough blame to go around for the high cost of healthcare including the cost of maintaining companies like Med Solutions, the costs to the doctors for their extra staff necessary to fight with insurance companies and the money that such time costs. The health lobbies have Washington in its pocket and you would not believe some of the stories my mom can tell about major recognizable Washington figures hitting her up for funds when she was a minor player at the lobby. So, while I recognize there are costs involved to healthcare, I don't fall hook, line and sinker that it is somehow the patients who need to put aside their best interest to bolster all of the extra healthcare costs. I paid thousands of dollars a month in healthcare costs (when I was healthy) in the individual market. If I need an mri now, so be it. Insurance made money off of me for years. Now that I require healthcare, they can pay it back (believe me I would rather them still make money off of me and not have cancer). If you really want to see healthcare costs come down, move to single payer, put more control in the hands of the doctors to make decisions and vote in politicians who are not supported by corporate interests.


Dear Dr. West, Here is a link to a recent study that was done on brain mets that you may or may not have seen. While it does not make or alter current practice on when to do an mri, it does mention a particular population that is at higher risk of brain mets and makes a point to say it is better to catch them earlier. I hope that with such research out there you will work with your oncologist colleagues to gauge whether the "wait for symptoms" is really better for this subgroup or if more aggressive monitoring should be standard practice so that catching brain mets early enough is an actual possibility. This should not be a decision made on cost benefit only unless one holds that the value of an advanced cancer patient does not hold enough value to be worthy of the cost of possibly life saving mris.

Please note - I fall into the subgroup.



I just read through this, thanks to all for such an intelligent and interesting debate. I agree that there is no 'right' answer here. In general, an MR scan should be prompted by symptoms, or situations where the results may have a significant impact on management. As Dr Loiselle says, currently we don't have solid evidence to show that frequent MR scanning improves survival. Even in lung cancer clinical trials, we don't routinely do MRIs to assess for progression.
I agree with Dr West that PET/CT is an inferior and more expensive tool than plain CT scan for assessing response to chemotherapy. It seems the use of PET/CT is motivated in part due to practice revenue. So comparing MR cost to PET/CT does not seem like a tenable argument.

The recent study mentioned above is rather simple and nothing can really be said from it. It did not adequately address immortal-time bias, this type of study is not published in renowned journals. For example, the population the author asserts had a higher risk of brain mets, is simply the subgroup that has longest survival. This group more often lives long enough to eventually suffer this complication (female, adenocarcinoma, young). Living longer with metastatic lung cancer will always put a population at higher risk of subsequent brain mets, so it is not a novel finding.


I'm trying to understand what was written in this thread...

1. So, if the patient is asymptomatic, the standard of care in US is not to do any imaging for brain mets, even if, let's say, the last time you (the oncologist), checked for brain mets was 2 years ago? This might make more sense in a patient with an activating mutation and a longer response.

2. Suppose that you can order a brain scan easily (CT), but, if it finds brain mets, access to treatment might be problematic, for sure taking months. Would this change the approach from the previous point? To rephrase it, if you can't treat immediately brain mets if they appear, would you be more proactive about finding them?


Dr West

The standard of care in the US is to not repeat a brain MRI in the absence of neurologic symptoms after an initial negative scan.

In the US, it's really not remotely common to have it take months to get these treated. I'd venture to say that if known brain metastases were left untreated, most oncologists would be inclined to follow these very closely.

-Dr. West


Thank you Dr. Creelan for your reply.

Taking it one step further though, if you had a patient who was perhaps at greater risk of brain mets (i.e. perhaps in the group that tends to survive longest and perhaps a young patient), would you strictly follow the standard of care for that patient (no brain mri unless symptomatic) or would you advise that patient individually taking into account their particular case history and specifics?


funnymom, I see where you are going with your questions. I had a brain met and never had a brain MRI until the day the met was discovered. However, I think your last question was unfair in that I believe that the doctors take all of their patients individually into account, They do have to have a place to start (ie, standard of care), but they have said many times that each person is different.


Sherry - I am not on cancergrace often enough to surmise a doctors viewpoint. I am looking at this thread as a stand alone and as a standalone, I see no evidence to the treatment of each patient as an individual whose unique case must be considered. If anything, when I mentioned my own situation, 5 symptomless brain mets (0ne that was up to 8mm on my cerebellum) the point that each individual case must be considered by an oncologist was never given - instead there was a cost benefits analysis. I understand that standards of care speak to the benefits of an audience at large, but there are some things that happen in these studies too that skew results. For instance, the overall population of lung cancer patients skews older. I think that skews data as an elderly skewed population will tend to have poor outcomes anyway. That is where it is so important to have oncologists judge the individual patients performance and prior history when making any decision, even one as simple as brain MRI for mets.

I'm sorry to have become involved with this thread. Obviously when you do not march in lockstep with the doctors, it becomes an excuse for followers...to attack. I will not be posting to this thread anymore. I think my opinion is already clear and that it is evident that I disagree with not only the doctors here, but this particular standard of care. But, perhaps it was my great fortune to have a doctor who looks beyond standards of care in decision making and that is why I am still here.


No code can account for all contingencies. I would advise a patient individually, taking into account their specifics. The definition of 'standard of care' even in legal parlance can vary significantly from region to region. It is probably true that asymptomatic brain mets are usually smaller than symptomatic ones. Like most clinicians, I practice shared decision-making, based on input from the patient and available guidelines.


funnymom, I'm truly sorry if you took my comments as an attack as they were not meant that way. On this forum we help each other and sometimes discuss our opinions, which differ at times. It has been my experience that doctors, in general, have a passion for what they do and they put the patient first. I'm sorry if it has not been yours.

Moderators - if you feel my comments should be deleted, please do.


Sherry, I'll post just one more time just say that the doctors on here are being of great service to the cancer community - of that we are in total agreement and I am thankful for the doctors response. I am not trying to have any great perry mason moment. I just want others reading the post to realize that sometimes, even with standards of care, each case is individual.

Dr West


Please let me assure you that I truly feel that there is plenty of room for different opinions and that yours is very welcome. I didn't feel attacked by yours and I didn't mean for my views to attack you; I hope that nobody else jumping in meant to cut you down either.

I completely understand your view. The challenge we face is that, as doctors, we have sometimes competing responsibilities. First, we want to do the best we can for our patients -- no doubt. But we also have a responsibility to be reasonable custodians of finite societal resources. This isn't something that we've historically done very well in the US health care system, especially in cancer care, where there is a certain temptation that anything that might possibly help a patient is the right path to pursue, no matter how great the cost. Frankly, that's a large part of how cancer care has operated for the past decade or more, but it's untenable. Doctors have been re-educated to recognize that we can't spend an infinite amount of societal resources on undefined benefits.

Most patients and caregivers will understandably prioritize the potential benefit for them and their loved ones over concerns for limited societal resources. It's the misalignment of responsibilities and priorities that leads to the lack of complete resonance between what is best for the individual patient and what the doctor is tasked to do: provide the best care for the patient while also being judicious about finite societal resources and prioritizing interventions that have a proven value on a population basis and not just in anecdotal cases.

Again, I hope you don't feel chased away. We are looking for good, respectful discussion with different points of view and don't want or need you to just nod assent if you don't feel the same way.

-Dr. West