Wondering what your guys' approach is to patients with a histologically confirmed lung primary with PET positive nodes in the mediastinum. How often are you requesting confirmatory pathological sampling of the mediastinum...always, sometimes, never??? There seems to be a split at my institution with the surgeons, of course, always recommending path confirmation, while the med and rad oncs are largely relying on PET only to call these patients IIIA-B.
PET and need for mediastinal node sampling
- Thread starter Reaganite
- Start date
- Joined
- Apr 16, 2004
- Messages
- 4,923
- Reaction score
- 6,119
PET is specific but not sensitive, false-positive rate is ~21% due to inflammation. Therefore, at my institution we routinely perform confirmatory mediastinoscopy in patients with PET positive N2 LNs. However, if LNs are PET negative there is only a ~9% false negative rate so we can then proceed to less aggressive treatment options as appropriate.
See here for the quintessential paper on the subject.
See here for the quintessential paper on the subject.
You swapped the terms Gfunk. PET is sensitive but not specific. But the take home message is the same.
- Joined
- Jul 20, 2007
- Messages
- 213
- Reaction score
- 26
I think if there is a mediastinum chock full of large, PET+ nodes, with a large path proven primary and/or N1 nodes raising the pre-test probability of N2 dz, path confirmation of the nodes via another procedure is overkill, IMO. Why do an expensive invasive test to tell you what you already 'know'? And if you're not gonna believe a negative result (med sensitivity ain't 100%) then don't do an invasive, morbid test is my thinking...
On the other hand, if there are a few, small or questionable PET+ LNs, I would advocate for staging, but not via med. We almost never see med at our institution, almost always EBUS, though there are some limitations in terms of accessible LN stations (as there are for med). I think this RCT should be putting the mediastinoscopy largely in the past, as EBUS has a NPV of 85% (equivalent to med), sensitivity of 79%, with less morbidity.
http://www.ncbi.nlm.nih.gov.ezproxy.library.wisc.edu/pubmed/21098770
EBUS alone might lead to a few more thoracotomies in pIIIa pts, but if the bulk of N2 dz is so low that it's in question and can't be confirmed via EBUS, I'm not sure surgery + PORT isn't the right treatment anyway. Our surgeons seem to think so (often resect single station N2 dz) and these pts have superior prognosis in the latest AJCC staging paper (5 yr OS 34% vs 20% multistation).
- Joined
- Apr 16, 2004
- Messages
- 4,923
- Reaction score
- 6,119
I completely agree with Cancerdancer's far more comprehensive and nuanced answer.
ditto x 2
D
deleted4401
Hmm...
I just got a mock oral on lung and now am getting a little confused.
Let me try to get this straight:
1) Big, bulky, FDG avid N2 nodes = obvious cancer = no need for further assessment by med or EUS, can proceed to CRT
2) No enlarged/FDG avid nodes on CT and PET = N0/N1 = good to go for surgery if operable or CRT if not
3) Minimally enlarged/FDG avid nodes on CT and PET = ?? = EBUS alone is enough, if positive consider CRT or preop CRT --> surgery and if negative, proceed with surgery if operable
Does location of tumor (peripheral/central) and/or enlarged N1 nodes change anything?
Is there any role for med or just go with above algorithm?
S
I just got a mock oral on lung and now am getting a little confused.
Let me try to get this straight:
1) Big, bulky, FDG avid N2 nodes = obvious cancer = no need for further assessment by med or EUS, can proceed to CRT
2) No enlarged/FDG avid nodes on CT and PET = N0/N1 = good to go for surgery if operable or CRT if not
3) Minimally enlarged/FDG avid nodes on CT and PET = ?? = EBUS alone is enough, if positive consider CRT or preop CRT --> surgery and if negative, proceed with surgery if operable
Does location of tumor (peripheral/central) and/or enlarged N1 nodes change anything?
Is there any role for med or just go with above algorithm?
S
In major centers, for situations described in 2) and 3), mediastinoscopy is still performed. That's because EBUS obviously has a sampling error/operator dependent, and sensitivity of PET for mediastinal mets is fact not nearly 90%, as some nuc.med studies claim.
- Joined
- Apr 21, 2011
- Messages
- 3,749
- Reaction score
- 9,579
I usually send for EBUS to confirm N2 disease, then treat any PET positive node.
- Joined
- Dec 17, 2007
- Messages
- 3,719
- Reaction score
- 5,244
If you can confirm N2 via EBUS you normally don't really "need" a mediastinoscopy.
You do need the mediastinoscopy after a N2-positive-EBUS however if you have suspicious contralateral lymph nodes, which you could not prove to have been negative via EBUS.
A combined negative PET & EBUS gives you a very high negative predictive value > 90%. Therefore IMHO it would have been an overkill to perform a media in such a situation.
Tricky is the part where PET is positive and EBUS negative, but there are some uncertainties concerning how representative the EBUS-sample war. In these cases we generally do a media.
You do need the mediastinoscopy after a N2-positive-EBUS however if you have suspicious contralateral lymph nodes, which you could not prove to have been negative via EBUS.
A combined negative PET & EBUS gives you a very high negative predictive value > 90%. Therefore IMHO it would have been an overkill to perform a media in such a situation.
Tricky is the part where PET is positive and EBUS negative, but there are some uncertainties concerning how representative the EBUS-sample war. In these cases we generally do a media.
