Misdiagnosis and RT

[beamseyeview]

---

Members don't see this ad.

Behind paywall, but an interesting read:

https://www.washingtonpost.com/lifestyle/2021/03/26/my-life-was-upended-35-years-by-cancer-diagnosis-doctor-just-told-me-i-was-misdiagnosed/

Essentially, patient treated for anaplastic astrocytoma as a teenager. Sounds like had resection w/ adjuvant RT and PCV. Original path shows "pilocytic astrocytoma" but then 2nd opinion path review at academic center upgrades to "anaplastic astrocytoma." Many decades later (he's now in his 40's), he's still alive and a random neuropathologist tells him he was originally misdiagnosed because he is still alive (does not review slides, only prior reports)

In my mind, it's an interesting read, but hindsight is 20/20.

What piqued my interest was this line from the article:

"I looked into whether I had grounds for legal action against the hospitals where I was treated, including the one that pummeled my brain with radiation without conducting its own assessment of whether my tumor was cancerous."

Any thoughts?

 

[medgator]

Behind paywall, but an interesting read:

https://www.washingtonpost.com/lifestyle/2021/03/26/my-life-was-upended-35-years-by-cancer-diagnosis-doctor-just-told-me-i-was-misdiagnosed/

Essentially, patient treated for anaplastic astrocytoma. Sounds like had resection w/ adjuvant RT and PCV. Original path shows "pilocytic astrocytoma" but then 2nd opinion path review at academic center upgrades to "anaplastic astrocytoma." Many decades later, he's still alive and comes into contact with a neuropathologist who tells him that he was originally misdiagnosed

In my mind, it's an interesting read, but hindsight is 20/20.

What piqued my interest was this line from the article:

"I looked into whether I had grounds for legal action against the hospitals where I was treated, including the one that pummeled my brain with radiation without conducting its own assessment of whether my tumor was cancerous."

Any thoughts?

Probably outside the statue of limitations i would think... Terrible nonetheless. In most states, i think you need to file within 3-7 years of when tx was given?

 

[Ray D. Ayshun]

A random neuropathologist tells him he was misdiagnosed , or one originally involved in the case? Doesn't seem impossible for an AA or gbm to be cured once or twice by surgery, chemo, and radiation.

 

[beamseyeview]

A random neuropathologist tells him he was misdiagnosed , or one originally involved in the case? Doesn't seem impossible for an AA or gbm to be cured once or twice by surgery, chemo, and radiation.

A totally random neuropathologist, who contacts him out of nowhere, who doesn't even review the slides (only the reports themselves), tells him that it is impossible that he had an AA because otherwise he would already be dead

I know.

 

[Lamount]

My take away from reading the article is that the pathologist who contacted him is a narcissist. It’s unknowable... the slides are gone and all that remains are conflicting reports. Maybe the pathologist at the original hospital spoke with the second opinion and was convinced by what he/she was told. Additionally the prognosis of IDH mutant AA isn’t terrible and it’s plausible that someone can survive. Given that it is impossible to be sure, why put this guy through the pain of irreconcilable doubt?

 

[Palex80]

I have a friend diagnosed with an anaplastic astrocytoma 15 years ago. Got surgery, chemo-rads. He is still in remission since and doing well.

 

[elementaryschooleconomics]

The "you were misdiagnosed because you beat the odds" attitude annoys me. I had a similar situation a few months ago where I saw a patient who was 10+ years out from treatment for a GBM and everyone instantly said "he must have not had GBM then".

Really? Out of all the people ever treated for GBM there aren't gonna be one or two that slip through the cracks and survive? I'm usually pessimistic but come on, not THAT pessimistic.

 

[medgator]

In the era of Optune/TTFs, I'm seeing more 1+ yr GBM survivors. Must be a lot of misdiagnoses going on

 

[mavrik13]

I don't get it... I've seen 20 year GBM follow-up (singular) post chemo RT where the path has been reviewed 4 times. While the cure rate for aggressive brain tumours is unfortunately low, it is not 0. The "pathologist" who reviews reports but didn't even look at the slides... ugh...

 

[Upgrayedd]

In the era of Optune/TTFs, I'm seeing more 1+ yr GBM survivors. Must be a lot of misdiagnoses going on

I mean the standard of care arm in the TTF trial was quite good

 

[medgator]

I mean the standard of care arm in the TTF trial was quite good

Must have been some pretty subpar central path review...

 

[TheWallnerus]

The twist here is that the narcissistic, pessimistic pathologist cold calls the guy and says "look at your old path reports, something's rotten in Denmark" and uh-oh, plot twist, there's a major discrepancy in the decades-old path reports. I can believe there are long-term survivors of AA. However, in the case of a *15 year old* with an initial diagnosis, in fact diagnoses, of a pilocytic, that is later modified to anaplastic by an outside academic pathologist, and the family and patient are completely oblivious to the conflicting path reports... if forced to bet... I bet misdiagnosis.

So in some ways narcissistic, pessimistic, cold-calls-patient pathologist has proved his point. Neuropath has the highest rate of misdiagnosis in all of pathology. It could scare you, if you really wanted to perseverate on it, the amount of times we have mis-irradiated patients in a long career WITHOUT EVER KNOWING. This is not an if, it is a DEFINITE, mathematically.

Study: Pathology Errors Can Have Serious Effect on Cancer... : Oncology Times

An abstract is unavailable.

journals.lww.com

Neuropathology on Trial

To the Editor:

academic.oup.com

 

[communitydoc13]

It’s unknowable...

Agree it's unknowable and also agree that misdiagnosis is most likely explanation. I don't worry about this in terms of a malpractice scenario as I think the definition of malpractice is protective of downstream physicians in cases like this. Does bring up important questions.

1. What is the risk of repeated path review when a given pathologist usually (as in really usually) gets the right answer and there is confirmation bias (either by the patient or practitioner) at play?

I've seen this happen a couple times, including a friend who had path reviewed by the best places and went with the most favorable diagnosis when it was clearly wrong in retrospect. In the case here, the converse happened. It would seem to be that in cases where a given pathologist is likely to get it right, repeated review (without some sort of consensus generating meeting) could actually be a bad thing?

2. What do we do with exceptional events?

These happen all the time, just at low relative frequency to unexceptional events. I'm not sure we have statistical tools to study them well and am not even sure that they are very susceptible to our usual tools of science. I have seen exceptional events on occasion over the years (spontaneous regression of tumor, cessation of progression with woo stuff). If you look closely at the outcomes of the placebo arm in almost any trial, you will see a small relative number of exceptional events (e.g. 90% PSA response to placebo). There was a database a while back (is it still there?) for reporting abscopal events as exceptional events for study. Not sure how this has gone.

As a clinician, I don't dismiss exceptional events or take credit for them. Hopefully, when they are good, the patient can revel in them.

 

[Lamount]

Agree it's unknowable and also agree that misdiagnosis is most likely explanation. I don't worry about this in terms of a malpractice scenario as I think the definition of malpractice is protective of downstream physicians in cases like this. Does bring up important questions.

1. What is the risk of repeated path review when a given pathologist usually (as in really usually) gets the right answer and there is confirmation bias (either by the patient or practitioner) at play?

I've seen this happen a couple times, including a friend who had path reviewed by the best places and went with the most favorable diagnosis when it was clearly wrong in retrospect. In the case here, the converse happened. It would seem to be that in cases where a given pathologist is likely to get it right, repeated review (without some sort of consensus generating meeting) could actually be a bad thing?

2. What do we do with exceptional events?

These happen all the time, just at low relative frequency to unexceptional events. I'm not sure we have statistical tools to study them well and am not even sure that they are very susceptible to our usual tools of science. I have seen exceptional events on occasion over the years (spontaneous regression of tumor, cessation of progression with woo stuff). If you look closely at the outcomes of the placebo arm in almost any trial, you will see a small relative number of exceptional events (e.g. 90% PSA response to placebo). There was a database a while back (is it still there?) for reporting abscopal events as exceptional events for study. Not sure how this has gone.

As a clinician, I don't dismiss exceptional events or take credit for them. Hopefully, when they are good, the patient can revel in them.

Would agree a long term survivor is far more likely to have had a LGG... but again would point out that IDH mut AA has a mOS of 54 months.

I don’t think that it is wrong to question his pathology and to investigate with an internal review. I DO think it is wrong to burden the patient with one’s unprovable theories.

If you discovered that the 2nd pathologist had had diagnosed 100 people with AA, most were dead within 18 months but some lived to 7 years... and THIS one patient was the only one to be alive at 15+ years, would you think the diagnosis was likely wrong?? I wouldn’t. Uncommon things happen uncommonly, but they do happen.
On the other hand, if 50% of the pathologist’s “AA” diagnoses are long term survivors, I would be worried that many of them are false diagnoses.

if this new pathologist is concerned, perhaps he should have investigated -this would have been far more productive than what he did.

 

[communitydoc13]

If you discovered that the 2nd pathologist had had diagnosed 100 people with AA, most were dead within 18 months but some lived to 7 years... and THIS one patient was the only one to be alive at 15+ years, would you think the diagnosis was likely wrong?? I wouldn’t.

I agree.

The point I was making is that the likelihood of some pathologist making an unlikely mistake goes up the more pathologists review a case. And, if the clinician or patient has a bias towards the mistaken outcome, sequential review can actually be bad.

For example, say Mayo, MDA and JHH all have a 5% chance of inaccurately upgrading a low grade glioma. And, lets say you were already biased to treat per the highest grade given (as is typical). You would have a roughly 14% chance of pursuing a false upgrade if you picked the highest grade among the three reviewers, whereas this likelihood is only 5% with a single review.

I cannot speak to what happens when there is discordance between diagnoses by different pathologists at different institutions. If there is some sort of mutual review, then I'm sure this risk is mitigated, but I am not aware of any "mutual review" process.

At my community hospital, difficult diagnoses are sent to Mayo and they are final authority.

Now picking the answer that 2/3 pathologists got would be the way to go.

 

[TheWallnerus]

The point I was making is that the likelihood of some pathologist making an unlikely mistake goes up the more pathologists review a case.

In a roundabout way this is why it is bad to do multiple post hoc, or subset or what have you, analyses in a randomized trial. Do enough statistical tests on a group of data you're bound to get a positive result eventually. Every pathologic diagnosis (or lab result... or IGRT match...) is a statistical test. Chronologically, the patient got screwed. If you get an AA diagnosis, and then another pathologist reads out as pilocytic, everyone would be like "Whoa wait a sec." It would be sent off for one or maybe two more second reads. This is a war on cancer and someone cancelled the war we can't have that. But you could have a pathologist read low grade glioma, and then later in time another respected pathologist says AA, and everyone is like "Time to treat." Come to think... it's kind of weird that not every path report gets a second blinded read by a different pathologist. Nah, that'd make life too difficult, we'd get too many conflicting reads.

 

[evilbooyaa]

Misdiagnosis happens. The patient should have been informed that the first two pathologists who read his tumor locally called it a JPA (Grade 1), and that a third party neuropathologist called it a higher grade tumor (Grade 3). Who knows if the first two pathologists were neuropathologists.

Somebody, a second opinion, an academic neuropathologist called for his/her 'expert opinion', called this a AA. In a 15-year old, in the pre-molecular era. Maybe he had a IDH1 mutant, 1p/19q co-deleted tumor that was actually more oligodendroglioma.

According to this website, 5-year survival in AA ages 20-44 is 58%: Survival Rates for Selected Adult Brain and Spinal Cord Tumors
Even if the cure rate is 1/10th of that somehow, it'd still be ~6% of patients.

IMO this pathologist is talking completely out of his ass and it's unfortunate this poor patient has to deal with the mental anguish of this 35 years later after beating his brain cancer.

The concept of suing a radiation oncologist because the pathologist mis-diagnosed him is likely just frustration on the patient's behalf but would set an incredibly dangerous precedent.

 

[medgator]

.

The concept of suing a radiation oncologist because the pathologist mis-diagnosed him is likely just frustration on the patient's behalf but would set an incredibly dangerous precedent.

Don't think it would pass muster, doesn't mean that a settlement still couldn't happen and end up on their NPDB report

 

[ramsesthenice]

Don't think it would pass muster, doesn't mean that a settlement still couldn't happen and end up on their NPDB report

Absolutely this kind of thing could end up with a settlement. More often than not it would get tossed before getting to the settlement/trial phase but if it progressed to that point you could end up paying out. The goal of the patients council in malpractice cases is always the same: convince the juror's that whatever happened could happen to them too. Emotion is as important as fact and case law if the system allows it to progress.

That said, I would take this bet any day even with hind sight. I think if you had gone the other way and told the patient, nah, I don't trust the second report. Lets treat this like a pilocytic and they quickly recurred you would be much more likely to find yourself in hot water. At the end of the day, you should always be treating based on what you think is most appropriate and not what is most likely to get you sued (or not). Fortunately, the two usually go hand in hand for us.

 

[radiaterMike]

I am not a lawyer but I think the statute of limitations starts at the time of the event on which the case is based or discovery. However, in this case there was no discovery since the pathology slides are gone.

 

[ramsesthenice]

I am not a lawyer but I think the statute of limitations starts at the time of the event on which the case is based or discovery. However, in this case there was no discovery since the pathology slides are gone.

With no slides to prove there was an error it would be very difficult for someone to argue that an error exists which is part of the burden to bring a suite. Hypothetically if they were profoundly injured by the therapy and enough experts lined up to state that the over read should clearly have been discounted there might be a path forward but that doesn't sound applicable either. My hypothetical response to Gator assumed the slides in question were in existence. Even then it would be extremely hard to meet a legal burden of proof. You have 2 diagnoses that are absolutely feasible for a patient of that age and as long as both were made by board certified pathologists there is still no obvious evidence of an error. Diagnostic disagreement does not in and of its self support the existence of an error.

 

[yeasterbunny]

I practice at a small community hospital and patients go outside for second opinions, and we send our path off for second opinions all the time. If I got a path consult back from the Mayo that upgraded the interpretation of our local pathologists, I would have done the exact same thing this patient's treating physician did, for sure.

Maybe the lesson is to take what our academic sacred cows say with a grain of salt instead of taking their word as gospel. For sure there is a level of deference we little guys pay to the big academic powerhouses. In general it's coming from a justified place, after all, Mayo's pathologists see a lot more of these cases than our local pathologists do. And imagine the alternative: if this patient's physicians treated him more conservatively and then he had a bad outcome they'd be in a world of hurt.