Arcs vs step and shoot for h&n

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
2-4 arcs > 1 arc

1 arc roughly = 7 field step and shoot.

Previously 2 arcs had been our standard, but now with a more faster TPS, dosimetrists seem to have settled on 3 arcs as the sweet spot.
 
temujim said:
2-4 arcs > 1 arc

1 arc roughly = 7 field step and shoot.

Previously 2 arcs had been our standard, but now with a more faster TPS, dosimetrists seem to have settled on 3 arcs as the sweet spot.
Click to expand...
Have you seen greater hotspots, though, or higher parotid doses for instance, than doing 9ish field imrt? Arc therapy seems much easier from a creation and implementation pov. Otoh, a lot of docs I respect are still opting for multi field imrt and not vmat.
 
Ray D. Ayshun said:
Have you seen greater hotspots, though, or higher parotid doses for instance, than doing 9ish field imrt? Arc therapy seems much easier from a creation and implementation pov. Otoh, a lot of docs I respect are still opting for multi field imrt and not vmat.
Click to expand...
Been doing VMAT for almost 10 years now. Don't accept plans with hot spots >105%, though have had to relax that to 107% with implementation monte carlo dose calc. Parotid mean doses are lower with 2+ arcs, at least in published dosimetric comparison papers. Delivery of course is significantly faster. Always felt that it was a win all around.
 
Can't say any hard and fast things on planning between either VMAT vs static beam, because it all depends how savvy one's planner is, but I think it's been years since I used static beam because VMAT has the enormous advantage of much faster treatment time and safeguarding against intrafractional changes. That one factor has basically outlawed static beam in my practice.
 
FrostyHammer said:
Can't say any hard and fast things on planning between either VMAT vs static beam, because it all depends how savvy one's planner is, but I think it's been years since I used static beam because VMAT has the enormous advantage of much faster treatment time and safeguarding against intrafractional changes. That one factor has basically outlawed static beam in my practice.
Click to expand...

Planning savviness is an important point. But this can be measured and graded. I won't look at a dosimetrist's CV unless they have a portfolio of plan challenges, meeting an 80-90th+ percentile score.
 
Some MR linacs are VMAT compatible but only cleared for step and shoot at present. I was very surprised how similar the plans are. Step and shoot is always a little hotter but only a couple percent. Don’t have to accept anything unreasonable with it. But, when I have the choice, I’ll take arcs every time. Little better and a little faster.
 
Definitely see a difference in some of the advanced lungs where the low dose bath from a vmat ends up being worse on the lung metrics than a step and shoot. Haven't seen as big of a difference with the h&n other than that treatment time is shorter, obviously (which is nice).
 
FrostyHammer said:
much faster treatment time and safeguarding against intrafractional changes.
Click to expand...

To quote the hammer: bingo. This point gets lost on a lot of proponents of techniques like MR or Cyberknife for SBRT. They may give you more confidence at what you are initially shooting at, but if you take a 10 min treatment up to 30-60 min what does that do to your uncertainty throughout the rest of the fraction? Hint: nothing good. Everything has pluses and minuses that need to be considered.
 
VMAT for everything, including OP's questions. Step n Shoot is archaic, IMO. 3 arcs seems to be the sweet spot at my institution. If you're accepting higher hotspots and parotid dose just because of VMAT, then your dosimetrist isn't doing it right. One can vector beam angles to avoid entry through parotid on top of obviously giving importance to homogeneity at the expense of sharp dose fall offs all over the place (as is common place in H&N)

I see utility of Step n Shoot over VMAT in one clinical scenario - known preoperative (not definitive) management of esophageal cancer, where it seems to improve V5 (which is of importance in pre-op esophageal cancer).

In definitive lung RT, I do not even look at V5 anymore. V20 is my preferred driver of toxicity.
 
medgator said:
Definitely see a difference in some of the advanced lungs where the low dose bath from a vmat ends up being worse on the lung metrics than a step and shoot. Haven't seen as big of a difference with the h&n other than that treatment time is shorter, obviously (which is nice).
Click to expand...

evilbooyaa said:
I see utility of Step n Shoot over VMAT in one clinical scenario - known preoperative (not definitive) management of esophageal cancer, where it seems to improve V5 (which is of importance in pre-op esophageal cancer).
Click to expand...

Partial arcs
 
I use VMAT for all H&N now but had markedly worse skin toxicity than fixed angle when I started using it. Crop PTV back 5mm from skin (admittedly skin contour is auto and includes some of mask) and will let volumetric cropping cut into my CTV on axials at level of neck angle if need be. (Seems like this is where skin toxicity is worst.)

Is anybody doing anything different in terms of cropping PTV?

I've found VMAT plans to be very robust. In the few cases I go for adaptive because of weight loss or response, plan usually still within reason without modifications.
 
evilbooyaa said:
VMAT for everything, including OP's questions. Step n Shoot is archaic, IMO. 3 arcs seems to be the sweet spot at my institution. If you're accepting higher hotspots and parotid dose just because of VMAT, then your dosimetrist isn't doing it right. One can vector beam angles to avoid entry through parotid on top of obviously giving importance to homogeneity at the expense of sharp dose fall offs all over the place (as is common place in H&N)

I see utility of Step n Shoot over VMAT in one clinical scenario - known preoperative (not definitive) management of esophageal cancer, where it seems to improve V5 (which is of importance in pre-op esophageal cancer).

In definitive lung RT, I do not even look at V5 anymore. V20 is my preferred driver of toxicity.
Click to expand...
Yup. IMRT = VMAT, to me now. I wouldn't do any step and shoot for the reasons mentioned - intrafractional motion and with multiple arcs better plans. I don't know the physics of it, but if 5 field IMRT is better than a 4 field box, and 7 field IMRT is better than 5 field, and 9 field is better than 7 field, than infinite fields is probably better than 9 fields.

The concern for V5 for esophagus cancer trimodality is over stated. If you believe it, 41.4 with AP/PA + carbo/taxol is standard of care for trimodality and will lead to a lower v5 then step and shoot IMRT 😉 so, if that's a concern, you'll stay below tolerance for basically every structure (including cord and not have to do a cord block), and you will be "safe" during surgery. I don't know, tho. Maybe it really does matter, but the idea that this low dose bath is killing patients seems sus.

Another thing that is great is conformal arcs, which is billed as 3D, but deliver beautiful plans. Great for tubular/spherical targets, palliation for bone mets. You can even get SBRT-esque plans for bone mets and give 12-16 Gy in 1, if you're having an issue getting SBRT approved (if you believe in higher doses for single fx bone mets - many do not).
 
evilbooyaa said:
I see utility of Step n Shoot over VMAT in one clinical scenario - known preoperative (not definitive) management of esophageal cancer, where it seems to improve V5 (which is of importance in pre-op esophageal cancer).

In definitive lung RT, I do not even look at V5 anymore. V20 is my preferred driver of toxicity.
Click to expand...
Precisely

And even in preoperative cases, you can utilize one ap and one pa VMAT partial arc, works quite well too.
 
I presume...
"Step and shoot" ≠ step and shoot IMRT = segmental MLC IMRT
"Step and shoot" = static beam, dynamic MLC IMRT

(I personally would say/write dMLC IMRT instead of "step and shoot," because that nomenclature is already taken.)

I still like dMLC vs VMAT in HN because I hate beaming through the shoulders (I kick couch on the ~lateral beams). You don't have to beam through shoulders if you "half beam block" the VMAT, but then I don't like the dose distribution inferior to the VMAT as much. You can do partial VMAT arcs with couch kicks, but then it's like 4-6 partial arcs and not a lot of treatment time vs dMLC IMRT is saved. In general, VMAT will "beat" dMLC every time if only DVH metrics are looked at. Again, it's about, sometimes, "I just don't want an entrance beam through that at all." Couch kicks don't waste a lot of time if you give therapists ability to remote couch kick.

This is one of those situations where there should be a test-case DICOM set, and there should be a "contest" of some sort.
 
communitydoc13 said:
I use VMAT for all H&N now but had markedly worse skin toxicity than fixed angle when I started using it. Crop PTV back 5mm from skin (admittedly skin contour is auto and includes some of mask) and will let volumetric cropping cut into my CTV on axials at level of neck angle if need be. (Seems like this is where skin toxicity is worst.)

Is anybody doing anything different in terms of cropping PTV?

I've found VMAT plans to be very robust. In the few cases I go for adaptive because of weight loss or response, plan usually still within reason without modifications.
Click to expand...

Agreed, this was drilled into my head as part of my training and I do it for all elective nodal volumes in H&N. I usually crop anywhere between 3 to 5mm away from skin depending on the dose - if 70Gy then 5mm, 63Gy usually 5mm, 56 Gy usually 3mm.

If I have gross disease going right up to skin/dermis (even without obvious involvement) then I'll sometimes leave the PTV uncropped at skin, such as an intact parotid case (cutaneous SCC met to parotid deemed unresectable)
 
Fixed beam IMRT (or dMLC) may be preferable when you want non-coplanar beams (which can be added to a VMAT plan if you want) or when you really want to pick beam angles (i.e. to avoid a kidney). Of course with VMAT you can make an OAR (i.e kidney) sparing a huge priority in which case the beam would more likely than not be 'modulated off' (my own words and probably too simplistic) when entering or exiting the OAR. For HNC, perhaps some nasal/paranasal sinus cancers might warrant consideration for a non-coplanar beam.
 
It depends on: your LINAC, software version, disease site (OP vs. larynx and individual dosimetrist skill
 
evilbooyaa said:
VMAT for everything, including OP's questions. Step n Shoot is archaic, IMO. 3 arcs seems to be the sweet spot at my institution. If you're accepting higher hotspots and parotid dose just because of VMAT, then your dosimetrist isn't doing it right. One can vector beam angles to avoid entry through parotid on top of obviously giving importance to homogeneity at the expense of sharp dose fall offs all over the place (as is common place in H&N)

I see utility of Step n Shoot over VMAT in one clinical scenario - known preoperative (not definitive) management of esophageal cancer, where it seems to improve V5 (which is of importance in pre-op esophageal cancer).

In definitive lung RT, I do not even look at V5 anymore. V20 is my preferred driver of toxicity.
Click to expand...

Got a gal with synchronous limited stage SCLC of the left upper lobe to bulky hilum/mediastinum and locally advanced distal esophageal ca. Dosimetry is on the struggle bus trying to treat both. Trained at a place with big V5 belief but not seeming achievable for this case. What is highest lung V5 people would accept?
 
radoncopotamus said:
Got a gal with synchronous limited stage SCLC of the left upper lobe to bulky hilum/mediastinum and locally advanced distal esophageal ca. Dosimetry is on the struggle bus trying to treat both. Trained at a place with big V5 belief but not seeming achievable for this case. What is highest lung V5 people would accept?
Click to expand...
V5 is not a remotely well validated pneumonitis parameter. What is the V5 of WLI to 12 Gy in a pediatric patient again?

For lymphocyte preservation per MDACC data, V5 of the lung or heart (not both) may be important- I set the initial constraint at 50% if size/location/anatomy is permissible, but up to 65% is OK, but if not then you sacrifice all lung V5 and focus on heart V5 if that's easier, using similar constraints.
 
radoncopotamus said:
Got a gal with synchronous limited stage SCLC of the left upper lobe to bulky hilum/mediastinum and locally advanced distal esophageal ca. Dosimetry is on the struggle bus trying to treat both. Trained at a place with big V5 belief but not seeming achievable for this case. What is highest lung V5 people would accept?
Click to expand...
Chance of cure from LD-SCLC about 20% (bulky N2, so probably a bit less than the overall 25% in the trials who also had N0/N1).
Chance of cure from locally advanced distal esophageal Ca with primary radiochemotherapy about 35% (?).
--> Chance of cure from both thus well below 10%.
🤔
Don't push it.
 
radoncopotamus said:
Got a gal with synchronous limited stage SCLC of the left upper lobe to bulky hilum/mediastinum and locally advanced distal esophageal ca. Dosimetry is on the struggle bus trying to treat both. Trained at a place with big V5 belief but not seeming achievable for this case. What is highest lung V5 people would accept?
Click to expand...

Assuming she is not getting surgery for her esophageal cancer given LS-SCLC which is more likely to kill her.

In that case I would not care about V5 of lung at all.

I am skeptical of the V5 --> lymphopenia and driver of mortality at the current time. Your priorities shouljd be minimizing V20 first and foremost, which is going to be very difficult given dual primaries, in different locations of the chest, both requiring Rx dose in close proximity to normal lung. Worry about V20 first, second, and third in the lung, and everything else after that.
 
Palex80 said:
Chance of cure from LD-SCLC about 20% (bulky N2, so probably a bit less than the overall 25% in the trials who also had N0/N1).
Chance of cure from locally advanced distal esophageal Ca with primary radiochemotherapy about 35% (?).
--> Chance of cure from both thus well below 10%.
🤔
Don't push it.
Click to expand...
I don't do much lung. Do they do adjuvant IO for SCLC, or just NSCLC. I have 2 patients who had synchronous NSCLC and EC with a CPS > 5 that I treated with "definitive" CRT and then adjuvant IO (as per PACIFIC for their NSCLC) for both sites > 12 months ago and both are still NED. Its a small N and I don't think think these responses will be typical but I am interested to see where this goes.

Re: V5: agree with above. Is it a problem on its own or a metric for higher dosing? Hard to know since everything is correlated. If you are really worried about it could you consider SBRTing the SCLC? Its not something we typically do but patients don't typically present with 2 lethal cancers. It could be a consideration.
 
No data for adjuvant IO in LS-SCLC as of yet.

SCLC as per radoncopotamous has hilar/mediastinal dz so desn't make sense to do SBRT. If it was like cT1-2N0 then yeah it'd be reasonable...
 
evilbooyaa said:
No data for adjuvant IO in LS-SCLC as of yet.

SCLC as per radoncopotamous has hilar/mediastinal dz so desn't make sense to do SBRT. If it was like cT1-2N0 then yeah it'd be reasonable...
Click to expand...

I didn't think so. I know there are trials ongoing but hadn't heard of any approvals.

As to the SBRT there hypothetically may still be a role depending on where the primary is. If it is removed from the hilum/mediastinum you could still dosimetrically make some gains doing conventional to the esophagus and mediastinum and more focal therapy to the primary. Don't have enough details to know. Its really more of an out of the box suggestion if they are really on the struggle bus. Unless things are super bulky, I'd be shocked if they can't meet objectives with conventional. I have done this several times without issue (conventional to both I mean).
 
ramsesthenice said:
I don't do much lung. Do they do adjuvant IO for SCLC, or just NSCLC. I have 2 patients who had synchronous NSCLC and EC with a CPS > 5 that I treated with "definitive" CRT and then adjuvant IO (as per PACIFIC for their NSCLC) for both sites > 12 months ago and both are still NED. Its a small N and I don't think think these responses will be typical but I am interested to see where this goes.

Re: V5: agree with above. Is it a problem on its own or a metric for higher dosing? Hard to know since everything is correlated. If you are really worried about it could you consider SBRTing the SCLC? Its not something we typically do but patients don't typically present with 2 lethal cancers. It could be a consideration.
Click to expand...
IO for LS-SCLC is going to be the SOC at some point in the future...NRG LU005 is accruing and phase II data look really really good (https://www.jto.org/article/S1556-0864(20)30713-9/abstract)
 
ramsesthenice said:
I didn't think so. I know there are trials ongoing but hadn't heard of any approvals.

As to the SBRT there hypothetically may still be a role depending on where the primary is. If it is removed from the hilum/mediastinum you could still dosimetrically make some gains doing conventional to the esophagus and mediastinum and more focal therapy to the primary. Don't have enough details to know. Its really more of an out of the box suggestion if they are really on the struggle bus. Unless things are super bulky, I'd be shocked if they can't meet objectives with conventional. I have done this several times without issue (conventional to both I mean).
Click to expand...

Sbrt a good thought, but wouldnt have helped in this case. Lung V5 ended up being < 100% but I won't say by how much. The 3D/IMRT plans I saw looked much worse for other heart and lung metrics compared to VMAT which had higher lung V5. Thank you all for your input.
 
radoncopotamus said:
Sbrt a good thought, but wouldnt have helped in this case. Lung V5 ended up being < 100% but I won't say by how much. The 3D/IMRT plans I saw looked much worse for other heart and lung metrics compared to VMAT which had higher lung V5. Thank you all for your input.
Click to expand...

V5 was just removed from NCCN lung constraint table with 2021 update, FWIW.
 
In most cases, I think VMAT is just fine.

Step and shoot (or more specifically, sliding window) IMRT may be a little better to treat very oblong targets, and not as good at rounder targets.

With each arc of VMAT, you can only achieve one MLC configuration per beam. Your degrees of freedom and conformality come from the fact that you have SO many beam angles.

With IMRT, you have few beam angles, but with each beam angle, the complex motion of the MLCs provides you with many degrees of freedom to create complex fluence, helping make the plans high conformality.

With oblong targets, certain beam angles are much more helpful than others. In these cases, creating a sliding window IMRT plan using these beam angles may yield more conformal plans than VMAT. I use this in the chest with N3 cases where I am treating the bilateral hila and am bumping up against V20 constraints. In the H&N, I can see this being helpful in specific cases as well, like when you are dealing with a NPX Ca invading the clivus and are worried about BS dose.
 
You must log in or register to reply here.

Similar threads

Palex80
Adjuvant Reirradiation in H&N cancer
Replies
49
Views
4K
evilbooyaa
evilbooyaa
M
Bionix for H&N RT
Replies
18
Views
1K
medgator
M
M
Persistently enlarged, but PET negative cervical node post chemorads for H&N Cancer
Replies
17
Views
1K
evilbooyaa
evilbooyaa
R
oligomet H&N cases
Replies
23
Views
2K
jondunn
jondunn
R
HCC, SBRT vs RFA
Replies
13
Views
2K
evilbooyaa
evilbooyaa
Top