Treating 20 brain mets

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
alwaysbeIMRTing said:
One time I wanted to do SRS to 9 NSCLC brain mets. One of the partners told me it was "an inappropriate use of resources" and, again quoting verbatim from chart rounds to "whole brain that sh-t"

F that place.
Click to expand...

Your treatment paradigm probably required less dosi, therapy, and on treatment table time if you did single iso multi lesion than if that old timer was SRSing somebody with 3 brain mets....
 
alwaysbeIMRTing said:
One time I wanted to do SRS to 9 NSCLC brain mets. One of the partners told me it was "an inappropriate use of resources" and, again quoting verbatim from chart rounds to "whole brain that sh-t"

F that place.
Click to expand...

Time to take this trial back to him and give him the business

edit: I should have been more clear and linked, but I was referencing the data recently presented at ASCO
 
Last edited:
Sorry, but as someone who has a lot of problems with the way trials are designed and reported, particularly in our field, I have a couple issues with what I've seen, though maybe there are more data that I don't have access to. I'm referring to the study concerning SRS vs WBRT in patients with 5-20 mets. One consequence of course is that if srs to 20 mets is better, then more patients should probably leave their community for treatment. This doesn't mean I'll never believe the studies, and I did contour 40+ mets for SRS during training, but there's certainly a built-in incentive for Harvard to justify treating 20 mets with SRS. That said, my problems in the reporting and ignorance of the treated population is as follows.

The title of the presentation was as follows:

Stereotactic radiation versus hippocampal avoidance whole brain radiation in patients with 5-20 brain metastases: A multicenter, phase 3 randomized trial.​

On the other hand, the study details note that the treatment will be:

Radiation: Whole brain radiation
  • Treatment of the whole brain with radiation. When possible the hippocampus will be spared from radiation.
Which is it/what proportion got HA-WBRT? I'm sure this known but didn't see it mentioned in the abstract or youtube vid.

It's also fair to know how many mets were present, which I didn't see? Was this a trial of SRS vs opposed lats in patients with 6 mets or SRS vs HA-WBRT in patients with 15?
 
Ray D. Ayshun said:
Sorry, but as someone who has a lot of problems with the way trials are designed and reported, particularly in our field, I have a couple issues with what I've seen, though maybe there are more data that I don't have access to. I'm referring to the study concerning SRS vs WBRT in patients with 5-20 mets. One consequence of course is that if srs to 20 mets is better, then more patients should probably leave their community for treatment. This doesn't mean I'll never believe the studies, and I did contour 40+ mets for SRS during training, but there's certainly a built-in incentive for Harvard to justify treating 20 mets with SRS. That said, my problems in the reporting and ignorance of the treated population is as follows.

The title of the presentation was as follows:

Stereotactic radiation versus hippocampal avoidance whole brain radiation in patients with 5-20 brain metastases: A multicenter, phase 3 randomized trial.​

On the other hand, the study details note that the treatment will be:

Radiation: Whole brain radiation
  • Treatment of the whole brain with radiation. When possible the hippocampus will be spared from radiation.
Which is it/what proportion got HA-WBRT? I'm sure this known but didn't see it mentioned in the abstract or youtube vid.

It's also fair to know how many mets were present, which I didn't see? Was this a trial of SRS vs opposed lats in patients with 6 mets or SRS vs HA-WBRT in patients with 15?
Click to expand...

This may be controversial, but by now I would expect community practices to be able to do both intra- and extracranial radiosurgery.
 
OTN said:
This may be controversial, but by now I would expect community practices to be able to do both intra- and extracranial radiosurgery.
Click to expand...
To 20 mets? I'd argue there's a chasm between what it takes wrt machine, dosimetry and physics capabilities to SRS a solitary met vs 20. I don't think a TrueBeam without micro MLCs should think about doing 5-20, and I'm not sure how many of us single linac docs have the ability to be setup for this. In any case, this isn't about that per se, point being that academic centers have a built in financial incentive to design trials that ignore a lot to push shorter treatment courses, which we point out in virtually every other disease site. I'm not arguing SRS isn't better, but it's being talked about as if it's a foregone conclusion regardless of the number of mets, and the argument is now being made that it's superior to HA-WBRT based on a trial that didn't mandate HA-WBRT.
 
Ray D. Ayshun said:
I don't think a TrueBeam without micro MLCs should think about doing 5-20
Click to expand...
Physics support and planning expertise. Fractionate (9x3 or 6x5 for big ones). Integral brain dose still way lower and LC probably higher than WBRT. Should pretty much never have to go above 2-3 isos even with large intracranial distances.

Treating 20 punctate lesions? Yeah, dumb in most circumstances. There are likely 15 that you are not seeing. Treating 12 lesions varying from 6 mm through 25 mm? Can be done well with setup described IMO.
 
Found more on trial I think that was mentioned with median number of Mets of 14, so apologies on that, but like I said, I was trying to access data. It's true as community doc said, that for instance, 6x5 to all Mets it's relatively easy, but where I trained it was single iso 18-22 gy to 1-50 Mets. The former (6x5) seems like a slight variation of ha-wbrt while the latter is ultimately unsafe I think without an edge, cyber knife, gamma knife etc. I'd happily do the former, but I'm not sure there's a precedent. Is anyone doing this?
 
Neuronix said:
What's the best part about contouring 20 brain mets?
Click to expand...
Time Wine GIF
 
Ray D. Ayshun said:
Found more on trial I think that was mentioned with median number of Mets of 14, so apologies on that, but like I said, I was trying to access data. It's true as community doc said, that for instance, 6x5 to all Mets it's relatively easy, but where I trained it was single iso 18-22 gy to 1-50 Mets. The former (6x5) seems like a slight variation of ha-wbrt while the latter is ultimately unsafe I think without an edge, cyber knife, gamma knife etc. I'd happily do the former, but I'm not sure there's a precedent. Is anyone doing this?
Click to expand...

I’m confused how it matters whether you are doing 5 fx or 1 fx? How is the former more like WBRT?

The target is the target the technique is the technique. Regardless of dose

Also disagree that you need an Edge to do this. This seems a little inconsistent with modern community experience IMO
 
drowsy12 said:
I’m confused how it matters whether you are doing 5 fx or 1 fx? How is the former more like WBRT?

The target is the target the technique is the technique. Regardless of dose

Also disagree that you need an Edge to do this. This seems a little inconsistent with modern community experience IMO
Click to expand...
What's the dose/volume constraint you use for normal brain when doing 5 fx SRS?
 
I largely use 1 or 3, but typically regardless of dose it’s a per lesion analysis, for 5 fx too there are well published and accepted constraints. It’s better than WBRT by definition no matter what
 
drowsy12 said:
I’m confused how it matters whether you are doing 5 fx or 1 fx? How is the former more like WBRT?

The target is the target the technique is the technique. Regardless of dose

Also disagree that you need an Edge to do this. This seems a little inconsistent with modern community experience IMO
Click to expand...
Harder to treat smaller lesions well on a standard truebeam without micro mlc

Better on an edge, stx etc
 
Last edited:
drowsy12 said:
I largely use 1 or 3, but typically regardless of dose it’s a per lesion analysis, for 5 fx too there are well published and accepted constraints. It’s better than WBRT by definition no matter what
Click to expand...
Okay, so, for single fraction SRS, the per lesion adjacent normal brain constraints, at least for me, are V12 <8 cc and V10 <10 cc. With respect to prescription this is around 50% <10 cc. I wasn't really trained on the "accepted" 5 fx adjacent normal brain constraints but a quick google search suggests 20 Gy to <24 cc. With respect to prescription, this would be 66% <24 cc. Create a plan that meets those constraints and then prescribe 20 Gy in a single fx and lemme know if it's acceptable. I'm not sure if you're arguing there's no difference in the safety concerns wrt single vs 5 fx brain treatment, but it sounds like it. I'm arguing 30/5 to all lesions is similar to HA-WBRT because it's just again reducing the amount of normal brain being treated while having substantially fewer to no safety concerns, particularly for clinics without NSG support. Community doc mentioned something along these line, but I'm unaware of the precedent. I'd imagine the control and toxicity outcomes are similar to this trial and more safely achieved with a TrueBeam with 6 degree couch and no micro MLCs. I have no lack of confidence in My dosimetrists to create plans that meet constraints and they do this at the institution I trained. But there's a difference to me between how it looks on a computer screen and the entire picture of what happens in reality.
 
Last edited:
‘I'm arguing 30/5 to all lesions is similar to HA-WBRT because it just again reducing the amount of normal brain being treated while having substantially fewer to no safety concerns, particularly for clinics without NSG support’

What im saying is that the plans are still nothing like HA-WBRT, and way better, even with 30/5. I’m not saying to not do 5 fx. I’m missing how it’s similar to HA WBRT? The target is the target and the technique is the technique
 
drowsy12 said:
‘I'm arguing 30/5 to all lesions is similar to HA-WBRT because it just again reducing the amount of normal brain being treated while having substantially fewer to no safety concerns, particularly for clinics without NSG support’

What im saying is that the plans are still nothing like HA-WBRT, and way better, even with 30/5. I’m not saying to not do 5 fx. I’m missing how it’s similar to HA WBRT? The target is the target and the technique is the technique
Click to expand...
30 Gy in 5 is similar to HA-WBRT (which allows 40 Gy in 10 fx to 2% of the brain) with respect to BED:

30/5 BED10: 48 Gy
30/10 BED10: 39 Gy
20/1 BED10: 60 Gy

30/5 BED2: 120 Gy
40/10 BED2: 120 Gy
20/1 BED2: 220 Gy

I did 40/10 for BED2 for toxicity consideration. The average brain is 1300 cc apparently, so 120 Gy2 is allowed to 26 cc in these plans. I very much understand how planning works, even as it regards hotspots and dose fall-off when planning SBRT vs standard IMRT. My point is that 30/5 to all lesions would seems to have a similar RN risk as HA-WBRT and substantially lower than single fraction SRS while having local control and toxicity profile similar to single fraction SRS and better than HA-WBRT. It appears you are suggesting the same thing. What I have not seen is a trial suggesting I have permission to do this. I've been on SDN for a minute, and have come to understand that @Neuronix treats CNS. I wonder his thoughts on a doc who only has access to a TrueBeam with 6 degree couch and no microMLCs or in-house NSG doing single fraction SRS to 20 mets. If this is not a safe, is there something between this and HA-WBRT or WBRT, or should I just send them elsewhere (which I am presently doing)?
 
Ray D. Ayshun said:
I wonder his thoughts on a doc who only has access to a TrueBeam with 6 degree couch and no microMLCs or in-house NSG doing single fraction SRS to 20 mets. If this is not a safe, is there something between this and HA-WBRT or WBRT, or should I just send them elsewhere (which I am presently doing)?
Click to expand...

There are a lot of pieces to linac SRS. It's not just the equipment. Is the QA solid? Is the MRI and fusion adequate? Is the planning technique correct? What about the immobilization and position verification during therapy?

The MLCs are a minor concern. UCSD apparently does all their linac radiosurgery on 5 mm MLCs. The dosimetry isn't as good when it comes to mets under 1 cm, but the clinical relevance is questionable.
 
Neuronix said:
There are a lot of pieces to linac SRS. Its not just the equipment. Is the QA solid? Is the MRI and fusion adequate? Is the planning technique correct? What about the immobilization and position verification during therapy?

The MLCs are a minor concern. UCSD apparently does all their linac radiosurgery on 5 mm MLCs. The dosimetry isn't as good when it comes to mets under 1 cm, but the clinical relevance is questionable.
Click to expand...
Absolutely. For most of the arc, lesions aren’t even seeing the micro mlcs? IMO, fusion is often suboptimal. I have also seen changes in the lesions and mass effect from steroids given after mri, but before ct sim
 
Ray D. Ayshun said:
What I have not seen is a trial suggesting I have permission to do this.
Click to expand...

so we both agree that 30/5 is lower dose than 20/1
We also agree that the risk of RN should be lower for 30/5 than 20/1

the trial shows that despite the increased risk of RN with SRS, the quality of life is still better with the RN risk that comes with multi-site single fraction SRS because of what we all know - fewer visits, less fatigue, less hair loss, less memory loss, less xerostomia, as compared to HA-WBRT

So I think your proposed approach is exactly what the trial is, just with an even more 'safe' dose. I don't think you need another trial.....

I will also add that I think Neuronix is exactly right that every other part of the process is more important than the MLC part. so I guess if one is concerned about their ability to do it, that would hold regardless of the dose/fx

All that said I don’t disagree with you at all that this should be done carefully and folks should consider sending out if they don’t think they have the infrastructure in place for it. Respect to you for being willing to do that, as we know some people won’t send out that they should
 
Last edited:
You must log in or register to reply here.

Similar threads

R
moving brains mets
Replies
10
Views
1K
elementaryschooleconomics
elementaryschooleconomics
Gfunk6
Recurrent brain met in left mid brain
Replies
28
Views
3K
deleted1180461
D
R
Sbrt may speed up Mets?l
Replies
21
Views
2K
SneakyBooger
SneakyBooger
S
Cryo for bone met, lung tx, etc?
Replies
19
Views
2K
evilbooyaa
evilbooyaa
S
Not radiating asymptomatic, limited brain mets
Replies
31
Views
3K
evilbooyaa
evilbooyaa
Top