SBRT lung margins

[radoncle]

What margins do you guys use for lung SBRT assuming you create an ITV from MIP?

---

Members don't see this ad.

 

[Bequerel]

5mm

 

[fiji128]

same, 5mm

 

[thecarbonionangle]

5mm, rarely more

 

[evilbooyaa]

5mm in all directions is the standard in 2021.

Just remember hat you can't always create an ITV from just the MIP. Tumors abutting mediastinum/chest wall require evaluation of all 10 phases of the 4DCT.

I think once in residency I had a pretty poor 4DCT that had lots of jumping, in part due to patient not breathing evenly. The attending did 7mm sup/inf off the ITV which I did not feel was unreasonable.

Seems like there's a story behind this question...

 

[metallica81788]

MIP is garbage for ITV on SBRT, IMO it's just a way to check yourself after contours are done. SBRT ITVs contoured solely on MIP may only be accurate 60-70% of the time.

4D CT all phases FTW

 

[evilbooyaa]

MIP is garbage for ITV on SBRT, IMO it's just a way to check yourself after contours are done. SBRT ITVs contoured solely on MIP may only be accurate 60-70% of the time.

4D CT all phases FTW

Eh, works fine for disease that is constantly surrounded by low density lung parenchyma throughout all the 4DCT phases, IMO.

That being said, I've even seen vessels tracking near the tumor make a MIP at least focally difficult to interpret, so I would agree to closely evaluate the 4DCT in 100% of cases all phases, whether that be contouring on each phase individually or doing a sanity check after you contour on MIP.

 

[elementaryschooleconomics]

Just to pile on - 5mm!

 

[Lamount]

same, 5mm

5 mm

 

[Ray D. Ayshun]

I just do 3 breath holds and boolean them to create itv. Then 5 mm. I breath hold all my sbrts if possible.

 

[TheWallnerus]

I just do 3 breath holds and boolean them to create itv. Then 5 mm. I breath hold all my sbrts if possible.

A very good method. Unfortunately, not everyone as you know can be reliable breath holders (although people reliably "hit the mark" with breath holding way more than I ever expected).

I always thought this would have been an intriguing option to completely eliminate motion but maybe too fidgety:

High-frequency jet ventilation for complete target immobilization and reduction of planning target volume in stereotactic high single-dose irradiation of stage I non-small cell lung cancer and lung metastases - PubMed

In all patients, SBRT under HFJV provided a reliable immobilization of the GTVs and achieved a reduction in PTVs, regardless of patient compliance. Tentative planning facilitated the selection of patients who could better undergo radiation in respiratory standstill, both with greater accuracy...

pubmed.ncbi.nlm.nih.gov

 

[medgator]

I just do 3 breath holds and boolean them to create itv. Then 5 mm. I breath hold all my sbrts if possible.

We must not get the same kind of SBRT pts lol. We're not talking healthy left sided breast pts here

 

[radiaterMike]

I just do 3 breath holds and boolean them to create itv. Then 5 mm. I breath hold all my sbrts if possible.

Do you do inhaled. exhaled and relaxed breath holds as your 3. Or are the 3 the same (with boolean taking care of inherent errors?)?

either would seem to work.

 

[thecarbonionangle]

If someone is so healthy and can do breath hold and are a surgical candidate, probably will not be seeing patient. More likely Dlco 40% and FEV 45%. BH out of the question, 4DCT with prolock

 

[Ray D. Ayshun]

Do you do inhaled. exhaled and relaxed breath holds as your 3. Or are the 3 the same (with boolean taking care of inherent errors?)?

either would seem to work.

Same 3. All dibh. Fuse bone to bone. Re confidence in setup we use variant rpm plus cbct plus c-rad surface guidance/monitoring. Beam off if rpm transponder moves out of range. Overall feel pretty confident about setup.

 

[Gfunk6]

3 mm, no ITV. I treat on CK - don't hate me

 

[Ray D. Ayshun]

Breath-holding capability of adults. Implications for spiral computed tomography, fast-acquisition magnetic resonance imaging, and angiography​

S B Gay 1, C L Sistrom, C A Holder, P M Suratt
Affiliations expand

• PMID: 7995705

Abstract​

Purpose: The breath-holding capabilities of various groups of individuals were evaluated to develop protocols so that patients undergoing spiral computed tomography (CT), digital angiography, and breath-hold magnetic resonance imaging (MRI) can be studied successfully.
Methods: Twenty-five outpatients and 25 inpatients (all adults) were studied before undergoing body CT. Each subject was asked to hold his or her breath for as long as possible. Then each patient was asked to perform as many repetitive 12-second breath holds as possible. These data were correlated with demographic and historical information.
Results: The maximum breath-hold time for inpatients and those outpatients who were heavy smokers or had chronic obstructive pulmonary disease (COPD) or congestive heart failure (CHF) was 18 to 32 seconds (95% confidence interval) with a mean of 25 seconds. For all other outpatients, breath-hold time was 38 to 56 seconds (mean = 45 seconds). The 95% confidence interval for the number of 12-second breath holds for these two groups was 4 to 6 breath holds (mean = 4.9) and 6 to 7 breath holds (mean = 6.6), respectively. One inpatient could not hold his breath at all and three others were only able to hold their breath once for short periods. The sex and age of the patient had no significant effect on breath-holding performance.
Conclusions: Breath-holding protocols must account for the diminished capabilities of most inpatients, and outpatients who are heavy smokers or have COPD or CHF. Most outpatients who are not heavy smokers or without COPD or CHF can achieve a single breath hold of 38 seconds, or up to six 12-second breath holds.

I don't disagree about capabilities, but my toughest breath holds have been otherwise healthy breast patients. I really don't need Michael Phelps.

 

[metallica81788]

Breath holding is more mental than physical. Patients will psych themselves out over it. That's why I love the visual coaching device in the treatment room.

 

[dieABRdie]

3 mm, no ITV. I treat on CK - don't hate me

You have the M6? Just curious what you think about it...

 

[dieABRdie]

I just do 3 breath holds and boolean them to create itv. Then 5 mm. I breath hold all my sbrts if possible.

Even if it doesn't move (like apex)?

I actually used to breath hold every lung SBRT patient who could hold their breath. My rationale was that it always drops the lung V20 and therefore toxicity. But then I was thinking... if you are using coplanar arcs... how much tissue are you actually sparing when the tumor is in the apex of the lung? Yes the V20 drops.... but most of the movement of the lung is inferior so you aren't really radiating that tissue anyway. So the other partners made me stop...

 

[Gfunk6]

You have the M6? Just curious what you think about it...

We upgraded to S7. I think it's the bee's knees. We can treat tiny primaries and mets in the lung without very large target expansion.

 

[Ray D. Ayshun]

Even if it doesn't move (like apex)?

I actually used to breath hold every lung SBRT patient who could hold their breath. My rationale was that it always drops the lung V20 and therefore toxicity. But then I was thinking... if you are using coplanar arcs... how much tissue are you actually sparing when the tumor is in the apex of the lung? Yes the V20 drops.... but most of the movement of the lung is inferior so you aren't really radiating that tissue anyway. So the other partners made me stop...

We don't really have any throughput issues. I'm comfortable with reproducibility. However minimal, the dosimetry is better. It standardizes everything from a therapist pov, and is another way to make sure they stay sharp doing breath hold. Most of my lungs are stage iii or iv, so keeping them comfortable with lung sbrt dibh for the one a month I do is worthwhile. May scrap in the future as well.

 

[Neuronix]

We upgraded to S7. I think it's the bee's knees. We can treat tiny primaries and mets in the lung without very large target expansion.

What does it gate based on? Fiducials? Surface tracking?

 

[Gfunk6]

What does it gate based on? Fiducials? Surface tracking?

Both fiducials and surface tracking. In cases of larger tumors, it can visualize internal movement in both cameras which can be done without fiducials.

CK has a program called Synchrony which takes the internal movement and external movement and builds a breathing model which allows the robotic head to actively track movement and modify based on breathing.

 

[evilbooyaa]

3 mm, no ITV. I treat on CK - don't hate me

The phrase in 2021 is "don't @ me"

And you didn't say it, so now i'mma @ you

@Gfunk6

CK still requires Fiducials right? Never made sense to me to advocate for that in LUNG SBRT of all places.

What has been the complication rate of fiducial placement necessary for CK treatment in your patient population? Do you think that rate is worth the dosimetric benefits CK gives you, across all patients?

I presume no empiric SBRT patients being treated on CK, right?

 

[jondunn]

treating lung SBRT on a CK (if it requires fiducials, I don't know if that is true always still?)bto me is as backwards as when people who get treated with proton for prostate cancer need a rectal balloon or a spacer.

keep it simple stupid!

 

[Gfunk6]

The phrase in 2021 is "don't @ me"

And you didn't say it, so now i'mma @ you

@Gfunk6

CK still requires Fiducials right? Never made sense to me to advocate for that in LUNG SBRT of all places.

What has been the complication rate of fiducial placement necessary for CK treatment in your patient population? Do you think that rate is worth the dosimetric benefits CK gives you, across all patients?

I presume no empiric SBRT patients being treated on CK, right?

60% of cases overall require fiducials, 40% can be treated without.

Since we have a dedicated thoracic surgeon using Super Dimensional Bronchoscopy to place fiducials we haven't had a single complication.

 

[elementaryschooleconomics]

60% of cases overall require fiducials, 40% can be treated without.

Since we have a dedicated thoracic surgeon using Super Dimensional Bronchoscopy to place fiducials we haven't had a single complication.

Fiducials for lung SBRT was something never done at my residency institution, but a couple of the docs in my current practice will occasionally do it.

Does anyone have a sense for how common it is, overall? If @Gfunk6 is doing it 60% of the time, that's a lot higher than I would have guessed. Similar to the SpaceOAR debate, it seems like you're doing something invasive prophylactically for a potential (but debatable) clinical benefit down the road.

I experience a lot of angst over this. Have I done SBRT without fiducials with good outcomes? Yes. Have I done SBRT with fiducials with good outcomes? Yes. Have I seen toxicity happen because fiducials were placed, or have I seen toxicity happen without fiducials that might have been avoided had they been placed? Yes and yes. Same with SpaceOAR.

Angst!

 

[medgator]

60% of cases overall require fiducials, 40% can be treated without.

Since we have a dedicated thoracic surgeon using Super Dimensional Bronchoscopy to place fiducials we haven't had a single complication.

Enb has been huge in reducing complications, namely pneumothorax with CT-guided bx which were fairly significant. The reps have asked us about using fiducials/placement during bronchoscopy but i feel fairly comfortable with 4Dct and cbct at this point

 

[TheWallnerus]

Fiducials for lung SBRT was something never done at my residency institution, but a couple of the docs in my current practice will occasionally do it.

Does anyone have a sense for how common it is, overall? If @Gfunk6 is doing it 60% of the time, that's a lot higher than I would have guessed. Similar to the SpaceOAR debate, it seems like you're doing something invasive prophylactically for a potential (but debatable) clinical benefit down the road.

I experience a lot of angst over this. Have I done SBRT without fiducials with good outcomes? Yes. Have I done SBRT with fiducials with good outcomes? Yes. Have I seen toxicity happen because fiducials were placed, or have I seen toxicity happen without fiducials that might have been avoided had they been placed? Yes and yes. Same with SpaceOAR.

Angst!

"I used to be Snow White, but I drifted," said Mae West. I used to insist on gating. I used to insist on fiducials. I had ~90% local controls (yes, I measure and graph my outcomes over time). I then went to a place that had neither gating nor superstring theory bronchoscopy. I still wanted to do SBRT; maybe my case selection for 3-5 fx got a little more strict. You do use a slightly different workflow. Yet I still had ~90% local controls over time. Bob Timmerman has a quote too: "SBRT is unique in the realm of high-tech radiation oncology in that outcomes are dependent more on the training of staff and physicians than the style of technology used." We need to remember: Bob was doing SBRT for a while AP/PA on a non-IGRT linac and reporting good cancer outcomes. He had a couple toxicity disasters but that's neither here nor there.

 

[Neuronix]

I had some experience with the old CK (pre-MLC), but didn't know how much it had changed. Thanks for filling me in.

I try not to put fiducials in anything anymore. It's just an unnecessary procedure with its own costs and risks. Placement issues (logistics, complications), fiducials migrate, what do you do with patients on blood thinners (and complication of holding blood thinners) or other contra-indications or difficult to place fiducial location, etc.

In most soft tissues I think MRI-guidance helps a lot over CBCT without fiducials.

The exception is the lung. In the lung there's so much soft tissue/air contrast that you can usually see the target really well. So MRI-guidance isn't as interesting to me though sometimes it can be useful for high risk targets (ultra-central, involving chest wall close to cord, etc). I sim with 4D CT and position daily treatments with breath hold cone beam and gate with surface tracking.

If they can't hold their breath you can still just do an ITV +/- abdominal compression (if tolerated). MRI can gate for poor breathers (takes forever though depending on duty cycle) or can be helpful with ultra-central or certain chest wall invasion situations. It hasn't been a slam dunk in my experience.

I'm curious to know if there's ever a situation nowadays where lung fiducials are helpful. Maybe to give the referring doc something to do? I guess this is all in the eye of the beholder. People without MRI say MRI isn't helpful too. Shrug.

 

[Gfunk6]

There are many ways to do lung SBRT. At the end of the day, the efficacy is so high and the toxicity so low that I think we are splitting hairs to see what is better. Things that work for me with CK:

1. Free breathing, no breath hold or compression required

2. Arms down comfortably at side and no tattoos

3. Able to use very thick memory foam pad over hard, fiberglass couch

4. Patients don’t necessarily breathe the same way each day, a new Synchrony is model is built per fraction

5. Ultimately you need a biopsy to treat. Our local referring docs now drop a fiducial at the same time as biopsy because they know they will be sending to us for SBRT

 

[jondunn]

5. Ultimately you need a biopsy to treat.

 

[medgator]

5. Ultimately you need a biopsy to treat. Our local referring docs now drop a fiducial at the same time as biopsy because they know they will be sending to us for SBRT

I think there's a number of series out now treating without tissue. In Japan, it happens quite often supposedly.

 

[elementaryschooleconomics]

I think there's a number of series out now treating without tissue. In Japan, it happens quite often supposedly.

 

[radoncle]

There are many ways to do lung SBRT. At the end of the day, the efficacy is so high and the toxicity so low that I think we are splitting hairs to see what is better. Things that work for me with CK:

1. Free breathing, no breath hold or compression required

2. Arms down comfortably at side and no tattoos

3. Able to use very thick memory foam pad over hard, fiberglass couch

4. Patients don’t necessarily breathe the same way each day, a new Synchrony is model is built per fraction

5. Ultimately you need a biopsy to treat. Our local referring docs now drop a fiducial at the same time as biopsy because they know they will be sending to us for SBRT

I’ve treated without tissue on many occasions when there is clear progression on serial scans and either a biopsy is non-diagnostic/the risk of pneumothorax is significant.
The patients we see for lung SBRT are often frail non-surgical candidates, with COPD, on oxygen, multiple comorbidities, poor PS, etc. I definitely disagree that you always need a biopsy to treat.

 

[dieABRdie]

There are many ways to do lung SBRT. At the end of the day, the efficacy is so high and the toxicity so low that I think we are splitting hairs to see what is better. Things that work for me with CK:

1. Free breathing, no breath hold or compression required

2. Arms down comfortably at side and no tattoos

3. Able to use very thick memory foam pad over hard, fiberglass couch

4. Patients don’t necessarily breathe the same way each day, a new Synchrony is model is built per fraction

5. Ultimately you need a biopsy to treat. Our local referring docs now drop a fiducial at the same time as biopsy because they know they will be sending to us for SBRT

How long does each fraction of SBRT take with the S7? That was my biggest gripe with the CK.

 

[Gfunk6]

How long does each fraction of SBRT take with the S7? That was my biggest gripe with the CK.

Depends on target size. However, if you use MLC head and VOLO (the latest iteration of CK's inverse planning software), beam on time is generally 10-15 min

 

[Krukenberg]

Depends on target size. However, if you use MLC head and VOLO (the latest iteration of CK's inverse planning software), beam on time is generally 10-15 min

Is there an advantage to having 1 linac and 1 CK vs. 2 Linacs with 1 of them being SBRT/SRS enabled?

 

[medgator]

Is there an advantage to having 1 linac and 1 CK vs. 2 Linacs with 1 of them being SBRT/SRS enabled?

I'll take two beam matched SBRT linacs all day over anything else, personally

 

[Krukenberg]

I'll take two beam matched SBRT linacs all day over anything else, personally

That definitely makes it easy during downtime, but isn’t it rather cost inefficient to have two sbrt enabled linacs? I suppose the gains of being able to flex patients during downtime might outweigh the costs of two sbrt linacs.

 

[Neuronix]

I think there's a number of series out now treating without tissue. In Japan, it happens quite often supposedly.

We do it in academics at the places I'm familiar with.

Is this not common in the community?

I have always thought that risk of pneumothorax or other complications in someone with unhealthy lungs is higher than risk of lung SBRT (assuming small target, not ultra-central, good technique, etc).

That definitely makes it easy during downtime, but isn’t it rather cost inefficient to have two sbrt enabled linacs? I suppose the gains of being able to flex patients during downtime might outweigh the costs of two sbrt linacs.

All the employed positions I see nowadays are only installing SBRT capable linacs. Some of the private groups or low resource environments are still running old setups to maximize efficiency, but those machines are getting so antiquated that when they get replaced it's usually with a new SBRT capable machine. For me it's just a question of which features you're buying with the machine (6D couch, surface tracking techniques, microMLC, etc).

This comes to a philosophical question that everyone seems to be asking nowadays. Is there anything you can do with a CK that you can't do with a well equipped "swiss army knife" linac like an Edge? I can't think of anything. So on this I'll agree with medgator. I just want one of those machines to have microMLCs, 6D couch, and OSMS.

 

[OTN]

We do it in academics at the places I'm familiar with.

Is this not common in the community?

I have always thought that risk of pneumothorax or other complications in someone with unhealthy lungs is higher than risk of lung SBRT (assuming small target, not ultra-central, good technique, etc).

I've done it several times

 

[medgator]

I've done it several times

Likewise, usually get pulmonary opinion and document in chart as well. I like to have multiple scans showing radiographic progression, including the pet

 

[medgator]

That definitely makes it easy during downtime, but isn’t it rather cost inefficient to have two sbrt enabled linacs? I suppose the gains of being able to flex patients during downtime might outweigh the costs of two sbrt linacs.

Precisely. CK will have its own planning system and you can't flex it to any other machine if it goes down, only issue with beam matching is the edge won't match with a standard truebeam given the micro MLC and field size differences

 

[Neuronix]

Likewise, usually get pulmonary opinion and document in chart as well. I like to have multiple scans showing radiographic progression, including the pet

In addition to serial scans, I sprinkle in some antibiotic like levofloxacin. Sometimes that clears up the finding.

Precisely. CK will have its own planning system and you can't flex it to any other machine if it goes down, only issue with beam matching is the edge won't match with a standard truebeam given the micro MLC and field size differences

A lot easier to re-calc between two Varian systems (especially Truebeams/Edges) assuming it's not a large field size than it is to jump between CK's planning system and Eclipse.

 

[Lamount]

Likewise, usually get pulmonary opinion and document in chart as well. I like to have multiple scans showing radiographic progression, including the pet

I feel comfortable treating without tissue when nodule is 1) enlarging on multiple scans + 2) PET-avid +3) pt a smoker + 4) either biopsy is considered 'unsafe' by pulm/IR or two attempts were made it was non-diagnostic. In rare instances, I have treated without all of these factors, but I usually prefer to have all of them.

 

[evilbooyaa]

There are many ways to do lung SBRT. At the end of the day, the efficacy is so high and the toxicity so low that I think we are splitting hairs to see what is better. Things that work for me with CK:

1. Free breathing, no breath hold or compression required

2. Arms down comfortably at side and no tattoos

3. Able to use very thick memory foam pad over hard, fiberglass couch

4. Patients don’t necessarily breathe the same way each day, a new Synchrony is model is built per fraction

5. Ultimately you need a biopsy to treat. Our local referring docs now drop a fiducial at the same time as biopsy because they know they will be sending to us for SBRT

I agree with multiple ways to skin the SBRT cat.

1. Why no compression? Even gentle compression is generally relatively well tolerated, even in those with COPD on baseline supplemental O2.

2. Fair enough - I could see a patient unable to lift arms use this as rationale, although I suppose static beam SBRT could be feasible.

3. We use a Vac-lok for immobilization - not a memory foam mattress I suppose, but why lay them straight on the couch?

4. Fair enough, this is somewhere tracking will help.

5. As multiple others have said, very reasonable to treat without biopsy. I do agree that if the pulm folks are thinking that far ahead then there is no/minimal added morbidity of dropping a fiducial at time of biopsy. I guess, what if biopsy is negative?

 

[Krukenberg]

We do it in academics at the places I'm familiar with.

Is this not common in the community?

I have always thought that risk of pneumothorax or other complications in someone with unhealthy lungs is higher than risk of lung SBRT (assuming small target, not ultra-central, good technique, etc).



All the employed positions I see nowadays are only installing SBRT capable linacs. Some of the private groups or low resource environments are still running old setups to maximize efficiency, but those machines are getting so antiquated that when they get replaced it's usually with a new SBRT capable machine. For me it's just a question of which features you're buying with the machine (6D couch, surface tracking techniques, microMLC, etc).

This comes to a philosophical question that everyone seems to be asking nowadays. Is there anything you can do with a CK that you can't do with a well equipped "swiss army knife" linac like an Edge? I can't think of anything. So on this I'll agree with medgator. I just want one of those machines to have microMLCs, 6D couch, and OSMS.

Yea to clarify I meant two modern linacs, only one of which has all the optional sbrt/SRS bells and whistles (micro MLC, 6D couch, OSMS, etc). E.g. Edge and true beam, or two true beams. I just don’t see the point of having two linacs with all the expensive extras, just so you can switch your sbrt patients back and forth during downtime. I do think there’s an advantage to being able to flex all your non-sbrt patients however

 

[medgator]

Yea to clarify I meant two modern linacs, only one of which has all the optional sbrt/SRS bells and whistles (micro MLC, 6D couch, OSMS, etc). E.g. Edge and true beam, or two true beams. I just don’t see the point of having two linacs with all the expensive extras, just so you can switch your sbrt patients back and forth during downtime. I do think there’s an advantage to being able to flex all your non-sbrt patients however

You cant treat large imrt/3D plans on an edge or truebeam stx