Hippocampal avoidance WBRT

[Ray D. Ayshun]

Doing first true HA WBRT (I do SRS, WBRT, or HA-GBM in general) case and wondering how to prescribe/constrain wrt the avoidance structure. Based on 0933, the PTV should be the brain minus the hippocampal avoidance structure, which is the bilateral hippocampi plus 5 mm. The constraints however are for the bilateral hippocampi. I'm just confirming that the hippocampal avoidance structure exists only for dosimetric reasons (to give 5 mm separation bt the hippocampi and PTV) and should not be constrained in any way,

---

Members don't see this ad.

 

[deleted1116990]

You contour the hippocampus, expand it by 5mm to create the PRV, then subtract the PRV out the brain to create your PTV and set your target criteria to the PTV. The hippocampal constraints are set to the hippocampus, not the hippocampus PRV. Your dosimetrist would have some questions if you tried to constrain to the PRV with 0933 Rx.

 

[Ray D. Ayshun]

You contour the hippocampus, expand it by 5mm to create the PRV, then subtract the PRV out the brain to create your PTV and set your target criteria to the PTV. The hippocampal constraints are set to the hippocampus, not the hippocampus PRV. Your dosimetrist would have some questions if you tried to constrain to the PRV with 0933 Rx.

No doubt. Just confirming. Thanks

 

[evilbooyaa]

You are correct. It's just an avoidance structure that is very borderline in terms of whether it is sufficient. It is also the 5mm margin around hippocampus that the patient shouldn't have gross disease in, because the entire PRV will be getting less than Rx dose in an attempt to meet hippocampus constraints.

You contour the hippocampus, expand it by 5mm to create the PRV, then subtract the PRV out the brain to create your PTV and set your target criteria to the PTV. The hippocampal constraints are set to the hippocampus, not the hippocampus PRV. Your dosimetrist would have some questions if you tried to constrain to the PRV with 0933 Rx.

I'm sure there are a number of dosimetrists out there with questions about constraining anything in the hippocampus... and therapists... just gotta go to the place with a boomer doc for the past 30 years.

 

[brushbeamscanning]

From what I remember from a talk by one of the study leaders the selection of the dmax and d100% was intended to simplify planning parameters for the trial since it mirrored other volumetric constraints for the hippocampus they had previously derived from glioma patients.

 

[deleted1111261]

I’m so glad you asked that.
First few cases I did, I was so confused. Makes me feel less dumb.

 

[elementaryschooleconomics]

I'm sure there are a number of dosimetrists out there with questions about constraining anything in the hippocampus... and therapists... just gotta go to the place with a boomer doc for the past 30 years.

*sigh*

You could win the lottery like I did and find a place where constraining anything except the spinal cord is unusual.

These places are pretty easy to find, if you know where to look:

1) Small community hospital or freestanding center
2) The RadOnc(s) graduated residency before 2000 and have practiced in the same location since that time
3) They have on-site Dosimetry, also at that same site for 10-20 years

It will be spun as "low turnover" and a good thing. It's a bad thing. Either run away or prepare to teach everyone about QUANTEC (after "reminding" them about Emami).

 

[deleted1111261]

Ah yes, Emami, the fifth flavor …

 

[CodeRedDew]

*sigh*

You could win the lottery like I did and find a place where constraining anything except the spinal cord is unusual.

These places are pretty easy to find, if you know where to look:

1) Small community hospital or freestanding center
2) The RadOnc(s) graduated residency before 2000 and have practiced in the same location since that time
3) They have on-site Dosimetry, also at that same site for 10-20 years

It will be spun as "low turnover" and a good thing. It's a bad thing. Either run away or prepare to teach everyone about QUANTEC (after "reminding" them about Emami).

I feel attacked

 

[elementaryschooleconomics]

I feel attacked

We should start our own professional society.

"Society Of Scholars And Doctors teaching Boomers and Others Irradiation"

SO SAD BOI​

 

[radiation]

Doing first true HA WBRT (I do SRS, WBRT, or HA-GBM in general) case and wondering how to prescribe/constrain wrt the avoidance structure. Based on 0933, the PTV should be the brain minus the hippocampal avoidance structure, which is the bilateral hippocampi plus 5 mm. The constraints however are for the bilateral hippocampi. I'm just confirming that the hippocampal avoidance structure exists only for dosimetric reasons (to give 5 mm separation bt the hippocampi and PTV) and should not be constrained in any way,

Vinai Gondi published a pretty good "how-to" a while ago

DEFINE_ME

Its a little dated now. Most places are not doing the couch kicks much anymore just cause its so painful. Varian has eclipse templates through rapidplan, but I believe thats an extra cost. If you pubmed hippocampal sparing techinques, various groups have tried different things to make the treatment easier to plan (elevated head angle etc).

One thing your dosimetrists may give you a hard time about is the pockets of hotspots all over the brain. Even on their papers the DVHs show hotspots going up to 35-36 gy. Your planner could spend a lot of time cooling those down but I haven't seen any data showing that is a big deal and the hotspot constraints on the protocol are pretty lax.

 

[Ray D. Ayshun]

Anybody doing 5 fx HA-WBRT?

 

[deleted1116990]

*sigh*

You could win the lottery like I did and find a place where constraining anything except the spinal cord is unusual.

These places are pretty easy to find, if you know where to look:

1) Small community hospital or freestanding center
2) The RadOnc(s) graduated residency before 2000 and have practiced in the same location since that time
3) They have on-site Dosimetry, also at that same site for 10-20 years

It will be spun as "low turnover" and a good thing. It's a bad thing. Either run away or prepare to teach everyone about QUANTEC (after "reminding" them about Emami).

4) Staff members are married to each other
5) On your first day, dosimetry tells you how they treat prostate cases there (the answer is 20 fractions -- small prostate, big prostate, high risk disease, low risk disease, huh what's an IPSS score? All 20 fractions. Our patients come from far away. We tell them they will get 20 treatments and have their CT sim when they come to see you for the first time. There is a big gas bubble in the rectum when the patient was simmed? No big deal, the computer will just shape dose around it. Wait, you want to be present for sim? What are you going to make us redo it or something?).

Don't waste your time with places like this. You will be universally hated if you try to change anything dear Dr. Boomer did. So your option is to try and practice good medicine and fight every day and have everyone hate you and go behind your back and tell admin what a horrible doctor you are, or just check out, die inside, and basically just babysit physics and dosimetry while they run the practice and make sure they don't do anything that will result in a lawsuit.

 

[seper]

Anybody doing 5 fx HA-WBRT?

4 Gy X 5? That's a bit of under-treatment, IMO

 

[jondunn]

Anybody doing 5 fx HA-WBRT?

I mean you certainly could. I haven't done it, but I guess if there was someone you wanted to get done quickly (inpatient perhaps).

My usual approach is SRS, then HA-WBRT, then if I still want to treat but don't think will live 6 months, standard WBRT. In that standard WBRT group I think 30/10 or 20/5 are fine

so that doesn't usually enter my own algorithm as if I feel strongly about HA, I will probably not be considering 20/5.

but no reason not to.

 

[Ray D. Ayshun]

Saw an 81 yo. Travel constraints etc. Diagnosed with denovo brain mets without symptoms. All sub 5 mm. Just started io alone, which could, but might not, help the brain. Am considering reimaging in 6 wks vs wbrt, and would prefer shorter course but also wondering how to mitigate side effects as much as possible. She's not really in bad shape overall. Kps 85

 

[deleted1116990]

4 Gy X 5? That's a bit of under-treatment, IMO

You could SIB the gross disease to 30 Gy in 5 fractions while sparing the hippocampus. This would be a unique situation where the patient really could not come in for an extra week, was not a candidate for SRS (no tech? small cell?), and had a limited number of mets.

I have never seen 5 fraction WBRT done in the US, although from the European literature I see no reason why the above would not be reasonable.

 

[deleted1111261]

You could SIB the gross disease to 30 Gy in 5 fractions while sparing the hippocampus. This would be a unique situation where the patient really could not come in for an extra week, was not a candidate for SRS (no tech? small cell?), and had a limited number of mets.

I have never seen 5 fraction WBRT done in the US, although from the European literature I see no reason why the above would not be reasonable.

Did it all the time at Banner Health for inpatients.

 

[deleted1116990]

Did it all the time at Banner Health for inpatients.

Checks NCCN...

Dose range is 20-37.5 in 5-15 tx. 20/5 is reasonable for poor prognosis. Consider memantine if giving standard whole brain to better prognosis patients. HA-WBRT preferred for better prognosis. 30/10 preferred for better prognosis patients.

No mention of 20/5 with HA-WBRT but I think you can combine all the above and make a case.

Looks like they finally dropped the 40/20 boomer special option.

 

[deleted1111261]

The underlying assumption is that some people think 20/5 is worse than 30/10. It may well be but I don't think the data supports this.

But, I wasn't saying do HA-WBRT. The above comment said that no one does 20/5 and I didn't take it to mean HA.

 

[deleted1116990]

The underlying assumption is that some people think 20/5 is worse than 30/10. It may well be but I don't think the data supports this.

But, I wasn't saying do HA-WBRT. The above comment said that no one does 20/5 and I didn't take it to mean HA.

Correct. I have never seen or heard of that done in the US. You are the first. I don't think you are wrong.

 

[deleted1111261]

Correct. I have never seen or heard of that done in the US. You are the first. I don't think you are wrong.

I definitely wasn't first there. But, I was so surprised to see it. One of my partners was Canadian and the chief ran the British Columbia Radonc group - was common there, I guess.

 

[TheWallnerus]

The underlying assumption is that some people think 20/5 is worse than 30/10. It may well be but I don't think the data supports this.

But, I wasn't saying do HA-WBRT. The above comment said that no one does 20/5 and I didn't take it to mean HA.

Correct. I have never seen or heard of that done in the US. You are the first. I don't think you are wrong.

I believe the data for the most part says: every time 20/5 has been tested against 30/10, it’s never been shown to be worse. Agree 20 “sounds low” but it’s never lost a gun fight.

Technically, I think 20/5 is the dose you would use for PCI in ES SCLC eg; I think most on the trial got that dose. It’s very European.

 

[deleted1116990]

I definitely wasn't first there. But, I was so surprised to see it. One of my partners was Canadian and the chief ran the British Columbia Radonc group - was common there, I guess.

I mean, you are the first I have seen in the US doing it. I am sure there are others. I know it's common outside of the US. I have never heard a single good/evidence-based argument why it's wrong but when this was brought up in residency one time, the response was fear the higher dose per fraction to the whole brain would turn them into zombies or something.

 

[deleted1111261]

I mean, you are the first I have seen in the US doing it. I am sure there are others. I know it's common outside of the US. I have never heard a single good/evidence-based argument why it's wrong but when this was brought up in residency one time, the response was fear the higher dose per fraction to the whole brain would turn them into zombies or something.

It's certainly possible! But, anecdotally, I did not see any zombie transformation.

 

[Ray D. Ayshun]

I don't know what we're talking about at this point. 5 fx wbrt or 5 fx ha-wbrt? We did 20/5 wbrt all the time in training and I do it around 10% of the time in the real world.

 

[TheWallnerus]

I mean, you are the first I have seen in the US doing it. I am sure there are others. I know it's common outside of the US. I have never heard a single good/evidence-based argument why it's wrong but when this was brought up in residency one time, the response was fear the higher dose per fraction to the whole brain would turn them into zombies or something.

This is the classic American response

I tried this dose in a rural center once and they looked at me like I was trying to ban guns

Anyway, there is robust data imho that neurocognition is the same with 30/10 vs 20/5

 

[HolySmokes]

"This intubated patient with uncontrolled extracranial disease and 50+ brain mets can't get 20/5, think about the short term memory loss!"

- Every attending of mine in residency

 

[deleted1111261]

I don't know what we're talking about at this point. 5 fx wbrt or 5 fx ha-wbrt? We did 20/5 wbrt all the time in training and I do it around 10% of the time in the real world.

Interesting!
Must be regional thing. Never did until got to AZ, 8 years into my career.
I don't think I would do for an outpatient. I don't have good logic for that, though.

 

[deleted1116990]

"This intubated patient with uncontrolled extracranial disease and 50+ brain mets can't get 20/5, think about the short term memory loss!"

- Every attending of mine in residency

Ok, glad it wasn't just me. Ray seems to have had a more unique experience with thinking about whole brain radiation in training and amongst peers. I wanted to slightly bump up the dose beyond 3 Gy x 10 for a patient with a radioresistant histology once and I was told I would turn him into a vegetable zombie. I think he ended up not making it through treatment anyway (so... we were both wrong).

 

[HolySmokes]

As it turns out, "memory" was just code for "RVU"

 

[TheWallnerus]

Ok, glad it wasn't just me. Ray seems to have had a more unique experience with thinking about whole brain radiation in training and amongst peers. I wanted to slightly bump up the dose beyond 3 Gy x 10 for a patient with a radioresistant histology once and I was told I would turn him into a vegetable zombie. I think he ended up not making it through treatment anyway.

These are ones I skip whole brain and try multi target SRS whenever possible

I do so much less whole brain nowadays than when I first got out of training

I think this has been a pretty big downward dip in RT utilization in this arena

 

[deleted1111261]

I think part of lack of utilization for 20/5 is QUARTZ trial showing no benefit vs BSC.
So, people say "I bet 30/10 is better than 20/5, so it's better than BSC"
Which, again, is not unreasonable, but I don't think data-driven.

 

[Ray D. Ayshun]

Comparison of short-course versus long-course whole-brain radiotherapy in the treatment of brain metastases - PubMed

Short-course WBRT with 20 Gy in five fractions is preferable for most patients, because it is associated with similar survival as longer programs and is less time-consuming.

pubmed.ncbi.nlm.nih.gov

the patients I treat with 20/5 fall into the group pulled out on MVA. I don't refer t this paper, per se, it just works out that way. The patient I asked about fits, but I wonder if avoiding the hippocampi might make the rest of her life better.

edit: not gonna be the first

 

[Palex80]

I think part of lack of utilization for 20/5 is QUARTZ trial showing no benefit vs BSC.
So, people say "I bet 30/10 is better than 20/5, so it's better than BSC"
Which, again, is not unreasonable, but I don't think data-driven.

No data exist, pointint out that 30/10 is better than 20/5.
One could speculate that the higher BED would result in more favorable intracranial control. Dose escalation trials "beyond" 30/10 using prolonged treatment courses or BID did show trends in that direction. We do not know however if that also applied to escalation beyond 20/5 with "just" 30/10. Those earlier escalation trials went for far higher doses.

The only randomized comparisons between 30/10 and 20/5 are from the 60s/70s, without any benefit observed.

The last point is what "better" means. Perhaps better intracranial control, but no way of telling if that would also translate into an OS benefit. There's always the "hybrid" option too:
Give a short course of WBRT now, scan in 3 moths and SRS any remaining big/problematic lesions, in stable patients.

 

[Palex80]

The patient I asked about fits, but I wonder if avoiding the hippocampi might make the rest of her life better.

The questions are:
1. Do you believe that hippocampal avoidance works --> improves functional outcomes.
2. Do you believe that decline in functional outcomes will happen within her life-span.

We have taken a look at a few hundred WBRT patients over the course of the past 10 years in our institutions and it is shocking to see how many of them died within 3 months of treatment. Those patients likely did not benefit from the WBRT at all and hippocampal avoidance would not have changed that.

 

[Ray D. Ayshun]

The questions are:
1. Do you believe that hippocampal avoidance works --> improves functional outcomes.
2. Do you believe that decline in functional outcomes will happen within her life-span.

We have taken a look at a few hundred WBRT patients over the course of the past 10 years in our institutions and it is shocking to see how many of them died within 3 months of treatment. Those patients likely did not benefit from the WBRT at all and hippocampal avoidance would not have changed that.

Hard to say, but maybe. I've seen older folks take a huge hit acutely with wbrt, but don't have enough experience with ha-wbrt to know if it's different in that population. Frustrating thing is there is potential upside, while the only real downside is the added financial hit.

 

[TheWallnerus]

The questions are:
1. Do you believe that hippocampal avoidance works --> improves functional outcomes.
2. Do you believe that decline in functional outcomes will happen within her life-span.

We have taken a look at a few hundred WBRT patients over the course of the past 10 years in our institutions and it is shocking to see how many of them died within 3 months of treatment. Those patients likely did not benefit from the WBRT at all and hippocampal avoidance would not have changed that.

The receipt of WBRT may be the rad onc subconsciously picking out near death patients
*shocked emoji*

On another note
I believe HA is abbreviation for hokum activity

However, hippocampus avoidance metrics would be much easier to achieve with Rx dose of 20 Gy

 

[jondunn]

I mean, you are the first I have seen in the US doing it. I am sure there are others. I know it's common outside of the US. I have never heard a single good/evidence-based argument why it's wrong but when this was brought up in residency one time, the response was fear the higher dose per fraction to the whole brain would turn them into zombies or something.

Yeah I’ve seen 20/5 commonly used in the US. Great for inpatients. May just be different crowds

 

[Ray D. Ayshun]

If wbrt turned them into zombies, how would that count wrt overall survival?

 

[evilbooyaa]

20/5 regular WBRT - sure, standard option for inpatients looking for quick D/C, IMO. Also not unreasonable for end of the line outpatients with uncontrolled extracranial disease.

20/5 HA-WBRT, nope never done it. I can't imagine the planning requirements (necesittating urgent treatment), and wouldn't use on end of the line patients without 6 months expected survival

 

[radiation]

A practical technique to avoid the hippocampus in prophylactic cranial irradiation for lung cancer - PubMed

A practical technique is presented to deliver hippocampus avoiding prophylactic cranial irradiation for lung cancer patients, using two lateral fields. For a prescribed dose of 12×2.5 Gy, sparing of the hippocampi to 6.1 Gy was achieved with a V95% of the brain of 81.7%.

pubmed.ncbi.nlm.nih.gov

People have done 3D hippocampal avoidance, could be feasible to do for 20 gy in 5. There is a huge area of brain that is undercovered, but for mostly palliative pts with the option for SRS salvage I dont think its a dealbreaker. I wouldnt do for lmd though

 

[seper]

20Gy /5 is good for inpatients, agree, but the clinical value of WBRT in those scenarios is doubtful. You are treating referrings and family, essentially

When you are doing HS-IMRT, you are presuming relatively good prognosis to justify the expense.
One, therefore, should try to give 30 Gy, to give the patient a running chance of lesion(s) control

 

[Lamount]

I think part of lack of utilization for 20/5 is QUARTZ trial showing no benefit vs BSC.
So, people say "I bet 30/10 is better than 20/5, so it's better than BSC"
Which, again, is not unreasonable, but I don't think data-driven.

Then again… QUARTZ didn’t actually demonstrate non-inferiority (but it didn’t stop the authors from claiming non-inferiority)

 

[jondunn]

quartz was mostly really bad PS people. I think even 30/10 is unlikely to benefit some of these people. we all treat them though, unless we dont want to, and then quartz is handy.

 

[Ray D. Ayshun]

interesting to find that 20/5 is so controversial. I thought this was more common in the less robust outpatient population.

 

[evilbooyaa]

interesting to find that 20/5 is so controversial. I thought this was more common in the less robust outpatient population.

20/5 for bone mets all the time. Not as high a percentage of whole brainers, especially those who aren't inpatient

 

[thecarbonionangle]

Very odd. Saw 20/5 WBRT in training routinely for some pts. I see it at chart rounds with 3d and HA with SIB to 30/5 in my department. Of course this is not the dose used for most people. This is not new folks!

 

[Ray D. Ayshun]

what's getting sib'd? mets >1 cm?

 

[thecarbonionangle]

what's getting sib'd? mets >1 cm?

If you can see it on an MRI and can do so safely why not hit it? Sib to 40/10 too with 10 fx

I do wonder though if someone here is thinking, “hmmmm if sdn thinks i can bill APBI as SBRT, maybe i should bill this too!”