Any rumors on whether there will be a savior for MRIdian users?
Viewray‘s Demise
- Thread starter Palex80
- Start date
D
deleted1111261
I doubt it.
They seem to have zero institutional memory of this. People bought that stupid gammapod thing for millions. In academics, you’re more likely to lose a job for offending someone compared to losing millions.
- Joined
- Apr 3, 2019
- Messages
- 4,269
- Reaction score
- 9,945
Well these are box and whisker plots not CIs. (Stats nerd sorry)
- Joined
- May 7, 2014
- Messages
- 1,657
- Reaction score
- 3,473
Agreed. TBH not sure what inference we are supposed to draw from the figure (haven't read the paper and probably won't)
- Joined
- Jun 21, 2022
- Messages
- 944
- Reaction score
- 3,286
Anyone doing a phase II of LDRT for endocrine therapy arthritis?
Maybe after the Lord Knea data comes out
- Joined
- Dec 2, 2011
- Messages
- 1,487
- Reaction score
- 3,010
The stats are not the issue for me. Bottom line, the incidence of G2+ GU toxicity was insanely high in the CT arm. I have an MR (Unity thank goodness…for now at least) and I do SBRT on both that and conventional CT. I don’t see a 50% difference in GU toxicity. It might be a smidge better, but this doesn’t match my real world experience. I smile and nod when my superiors get super excited about this stuff and then go about my day 🙂
Of course it was high- their margins were trash on the CT arm. This was a trial designed to show superiority of the MRI arm, and it did just that.
- Joined
- Aug 4, 2020
- Messages
- 1,967
- Reaction score
- 3,354
Lol, if this isn't the classic All-American way to make a buck, I don't know what is..
- Joined
- Oct 4, 2017
- Messages
- 5,017
- Reaction score
- 9,784
What margins do you use?
- Joined
- Apr 16, 2004
- Messages
- 4,661
- Reaction score
- 5,080
3mm all around for SBRT
- Joined
- Dec 2, 2011
- Messages
- 1,487
- Reaction score
- 3,010
This is going to "shock" everyone but they are not very different. Usually 3 mm on CT and 2 mm on MR.
The imaging advantages of MRI over CT are much greater in the upper abdomen than the pelvis. My bias for prostate (and most sites really) is that plan adaptation is far more advantageous than margin reduction. So, not only did Mirage use substantially different margins between groups, one was adaptive and one was not. Anyone want to bet if we compared MR vs Ethos with reasonable margins that MR would win? Didn't think so.
And for those who have not seen my posts before, I am actually pro adaptive MR-IGRT for the right situations. It lets me get away with pretty high dose isotoxic plans abutting bowel which can be really useful in the re-irradiation setting or other very specific instances. But the question has to be asked...how many of these cases are there? And the ever important follow up...how many of these machines then need to exist?
I don't want to see MR-linacs cease to exist. But I also don't want to see them used to gain a "competitive advantage" or become and endless source of half-hearted "trials" which are incremental even in the very best case scenario. Alas, the universe doesn't care about what I want
- Joined
- Jun 21, 2022
- Messages
- 944
- Reaction score
- 3,286
This is going to "shock" everyone but they are not very different. Usually 3 mm on CT and 2 mm on MR.
The imaging advantages of MRI over CT are much greater in the upper abdomen than the pelvis. My bias for prostate (and most sites really) is that plan adaptation is far more advantageous than margin reduction. So, not only did Mirage use substantially different margins between groups, one was adaptive and one was not. Anyone want to bet if we compared MR vs Ethos with reasonable margins that MR would win? Didn't think so.
And for those who have not seen my posts before, I am actually pro adaptive MR-IGRT for the right situations. It lets me get away with pretty high dose isotoxic plans abutting bowel which can be really useful in the re-irradiation setting or other very specific instances. But the question has to be asked...how many of these cases are there? And the ever important follow up...how many of these machines then need to exist?
I don't want to see MR-linacs cease to exist. But I also don't want to see them used to gain a "competitive advantage" or become and endless source of half-hearted "trials" which are incremental even in the very best case scenario. Alas, the universe doesn't care about what I want
I dont think MIRAGE used on table adaptive, but not totally sure.
- Joined
- Dec 15, 2014
- Messages
- 306
- Reaction score
- 580
Amazing that you can write this long of a post yet fail to realize mirage did not use adaptive therapy at all. It’s “shocking” to see people spout off definitively without knowing the basics. Pro-tip: try reading papers before criticizing.This is going to "shock" everyone but they are not very different. Usually 3 mm on CT and 2 mm on MR.
The imaging advantages of MRI over CT are much greater in the upper abdomen than the pelvis. My bias for prostate (and most sites really) is that plan adaptation is far more advantageous than margin reduction. So, not only did Mirage use substantially different margins between groups, one was adaptive and one was not. Anyone want to bet if we compared MR vs Ethos with reasonable margins that MR would win? Didn't think so.
And for those who have not seen my posts before, I am actually pro adaptive MR-IGRT for the right situations. It lets me get away with pretty high dose isotoxic plans abutting bowel which can be really useful in the re-irradiation setting or other very specific instances. But the question has to be asked...how many of these cases are there? And the ever important follow up...how many of these machines then need to exist?
I don't want to see MR-linacs cease to exist. But I also don't want to see them used to gain a "competitive advantage" or become and endless source of half-hearted "trials" which are incremental even in the very best case scenario. Alas, the universe doesn't care about what I want
- Joined
- Apr 3, 2019
- Messages
- 4,269
- Reaction score
- 9,945
Also fine for seed marker based kV!
- Joined
- Oct 24, 2010
- Messages
- 3,509
- Reaction score
- 6,356
My margins are better then your margins, but just slightly.
D
deleted1111261
Amar said it was allowed but no one actually got adapted, if I recall correctly
- Joined
- Dec 15, 2014
- Messages
- 306
- Reaction score
- 580
This is going to "shock" everyone but they are not very different. Usually 3 mm on CT and 2 mm on MR.
The imaging advantages of MRI over CT are much greater in the upper abdomen than the pelvis. My bias for prostate (and most sites really) is that plan adaptation is far more advantageous than margin reduction. So, not only did Mirage use substantially different margins between groups, one was adaptive and one was not. Anyone want to bet if we compared MR vs Ethos with reasonable margins that MR would win? Didn't think so.
And for those who have not seen my posts before, I am actually pro adaptive MR-IGRT for the right situations. It lets me get away with pretty high dose isotoxic plans abutting bowel which can be really useful in the re-irradiation setting or other very specific instances. But the question has to be asked...how many of these cases are there? And the ever important follow up...how many of these machines then need to exist?
I don't want to see MR-linacs cease to exist. But I also don't want to see them used to gain a "competitive advantage" or become and endless source of half-hearted "trials" which are incremental even in the very best case scenario. Alas, the universe doesn't care about what I want
Also it is interesting that you believe adaptive is so much more advantageous even though inherently the purpose of adaptive, like margin reduction, is to reduce dose to OARs while maintaining dose to the target. Even more interesting that you believe this despite literally 0 randomized data to support this. You're an academic with an MR-LINAC, surely, you are running a randomized trial (adaptive vs no adaptive) if this is something you want to actually prove? Didn't think so.
- Joined
- Jun 21, 2022
- Messages
- 944
- Reaction score
- 3,286
The purpose of on table adaptive is to allow you to adapt the radiation plan on the table.
Remember the MRIdian was an MR linac, and allowed gating, and allowed on table adaptive. It did all of those things.
Those features can be used in different ways. The isotoxic dose escalation strategy used in pancreas is not likely to be helpful in prostate.
- Joined
- Dec 15, 2014
- Messages
- 306
- Reaction score
- 580
Ramses was the one who said definitively that adaptation would work in prostate. Has he studied it? Is anyone studying it? How do we "know" that this is true? He was 100% confident in multiple things and 100% wrong.
- Joined
- Dec 2, 2011
- Messages
- 1,487
- Reaction score
- 3,010
What? I never said adaptive therapy "definitely" works for prostate cancer. I said I thought that plan adaptation is probably more valuable than simple margin reduction. More specifically, I said this was my bias which inherently means it is an opinion, not a fact. If you disagree, cool.
As I posted above, my experience using both MR and CT based SBRT for prostate cancer is that I don't see an obvious difference between them in my own hands. I won't be studying anything involving MR guidance for prostate cancer because I don't think it is worth the time.
- Joined
- Dec 2, 2011
- Messages
- 1,487
- Reaction score
- 3,010
Not at all. A very different beast indeed.
- Joined
- Dec 15, 2014
- Messages
- 306
- Reaction score
- 580
Way to own up to being completely wrong about adaptive being done on the trial. Didn't think you would. And by the way, anyone can clearly read between the lines of "So, not only did Mirage use substantially different margins between groups, one was adaptive and one was not." that you assume that the use of adaptive (which was in fact not used at all) contributed to the toxicity difference. Hiding behind "well that's just my opinion" is meaningless, obviously it's your opinion if you posted it. The fact is you were wrong about half of it (adaptive was used), and you have no evidence for the other half to back up your certainty (adaptive would have benefited).
- Joined
- Oct 24, 2010
- Messages
- 3,509
- Reaction score
- 6,356
- Joined
- Aug 4, 2020
- Messages
- 1,967
- Reaction score
- 3,354
RamsestheNice gonna be RamsesTheNasty pretty soon bruh
- Joined
- Dec 2, 2011
- Messages
- 1,487
- Reaction score
- 3,010
Yes, I was wrong about them not using adaptive therapy. It feels like doing 3D on a new true beam, but apparently its what they did.
But that doesn't change the fact that you are wrong about reading between my lines. Let me say it one more time. I don't believe the difference they report has anything to do with any aspect of MRI-guided radiation. In my hands using CT guided SBRT with more reasonable margins, I don't see the crazy high GU toxicity they claimed to see in their CT arm. I don't think they did a fair comparison.
- Joined
- Aug 4, 2020
- Messages
- 1,967
- Reaction score
- 3,354
- Joined
- Dec 15, 2014
- Messages
- 306
- Reaction score
- 580
I didn't say you said the benefit was due to MRI, I said you claimed adaptive had something to do with the outcomes, which you literally did say.
What PTV dose do you use for your SBRT?
- Joined
- Jun 21, 2022
- Messages
- 944
- Reaction score
- 3,286
Does it? I thought you said you don't think it helps haha. I think doing an IMRT plan on the MRIdian is very similar to an IMRT plan on the Truebeam.
(I am assuming you agree IMRT is better than 3D for prostate)
I do think it's fair, but people saying it's unfair is reasonable in my opinion. You have to remember this was an institutional study and it has to be completed in real life with real physicians enrolling to the trial.
The experimental arm margins were not standard at their center. Everyone wanted the same margins in both arms, but if the physicians were not comfortable enrolling to a small-margin CBCT based treatment, then the trial is DOA. If you say "well I treat on my linac with the MR-margins all the time!"... cool. You are awesome. Unless you work at UCLA it is irrelevant in this discussion.
This is the same discussion as that trial that compared proton CSI to spine IFRT with photons. One could do VMAT based CSI, but many dont or wont. The investigating center did not consider it standard, so therefore that is the trial you get at that center.
I think it is great people keep a watchful eye on industry for rigged trials, but despite wild accusations from many in the field, MIRAGE was an institutional study funded by UCLA with no design involvement from Viewray haha. The results had basically zero impact on our discussions of maybe buying a Viewray.
Im not sure how that is capitalist other than doing an IIT that enrolls patients to a new expensive machine is great synergy of priorities for the university and cancer center.
This is great discussion but it's important to have the facts straight.
- Joined
- May 7, 2014
- Messages
- 1,657
- Reaction score
- 3,473
I think using Grade2 physician reported as the primary endpoint is horribly prone to bias since the "raters" knew which patients were treated with MR. Acute Gr2 is usually adding tamsulosin for frequency. Lots of room for manipulation. The criteria say things like "medical management indicated"; indicated according to whom.
I have been doing prostate SBRT with CT since 2009 (on trial initially) and I put almost 100% men on tamsulosin prophylactically since frequently symptoms peak after the patient has completed treatment.
I have been doing prostate SBRT with CT since 2009 (on trial initially) and I put almost 100% men on tamsulosin prophylactically since frequently symptoms peak after the patient has completed treatment.
Getting MDs to change their practice just to manipulate a trial? Tough enough to get them to change practice for an arm of the trial in a consistent manner. In your example, once someone is placed on prophylactic, that medication wouldn't count as a grade 2; only an increase in dosage or a switch.
Most of the secondary endpoints were positive in mirage as well, including some of the patient reported outcomes. Unless UCLA rotten to the core, doubt there was bias strung throughout (just the margin difference -- addressed by @NotMattSpraker above).
- Joined
- May 7, 2014
- Messages
- 1,657
- Reaction score
- 3,473
I disagree with your statement that prophylactic not Grade 2. No tamsulosin at baseline but added just in case after 1-4 fractions is Grade 2. I am not suggesting rotten to the core just that the endpoints is subjective. I recall that all differences evaporated at 3 months and most PRO werent different
- Joined
- May 18, 2019
- Messages
- 50
- Reaction score
- 72
Saw that WashU and Wisconsin shut down their ViewRay systems (per twitter). Is this a sign because worry about service contracts/parts? Just trying to figure out the implications for other centers with the technology…
- Joined
- Dec 2, 2011
- Messages
- 1,487
- Reaction score
- 3,010
Oh, I think my joke may have been missed. Was trying to imply that using a true beam to routinely do 3D feels like a waste. You know, like going the speed limit in a Diablo on an empty highway.
As for my thoughts on adaptation for prostate, I hadn’t planned on getting this nuanced but my 6 week old is sleeping well tonight so here goes. I don’t think adaptation for prostate cancer is likely to help with high grade GU toxicity. Full, empty, or otherwise, there is a significant interface between the prostate and posterior/inferior bladder wall that is unavoidable. The absolute volume of that interface is relatively fixed. Bladder filling will only change the percentage of the bladder in that high dose region which almost certainly matters at the extreme. If it’s exceedingly empty on CT, you take them down and have them fill more. It’s the same on MRI. You can’t adapt yourself out of an unacceptable set up. But the more common situation is that the bladder is more like 70% the volume it was at simulation. I think we can all agree that the return of adapting the plan here is probably limited as both plans would be acceptable. Further, the margin story only accounts for the bladder contribution to GU toxicity but does nothing for the urethral component.
The rectum though, I feel differently about. The volume of rectum receiving rx (or near RX) dose often is not fixed and manipulating the rectal volume/distention is less predictable than with the bladder. Again, if they have a big one locked and loaded, you can’t adapt your way out of that. But if the rectum is modestly distorted and if you know the true volume getting rx dose is significantly higher than the initial plan would suggest, you can do something about it in a way that has at least a reasonable chance of reducing the risk of late bleeding in some people.
Before anyone says anything, this is an opinion that I have no interest in proving because it’s almost unprovable. Chronic proctitis can be very stubborn to treat but fortunately, it’s pretty uncommon at baseline. On top of that, a lot of people just set up well and I don’t do any adapt to shape on about half of our SBRT patients. Bottom line, I don’t think that adaptation provides a huge benefit for prostate and I do not advocate buying an MR linac to treat prostate cancer. I do it for individual patients for whom I think it might help because we happen to have one. I do not uniformly do all prostate SBRT on the MR.
What I meant earlier was that I think adaptation is a better rational for MR IGRT because it’s something you can’t do with most CT based systems. Even as an MR user, I don’t think MR is needed for significant margin reduction, especially in the pelvis. Most CBCTs are good enough now. And as I strongly implied above, if you do a real apples to apples comparison with something like Ethos, I am not sure what endpoint MR would have a realistic chance of winning even with an unlimited number of patients.
As for the UCLA group, I don’t disagree with you. I suspect you are right about how they got there nor do I see any capitalist shenanigans afoot. You have to start somewhere and there will always be realities to deal with. But it doesn’t make it any less unfortunate for us. Were this differences they saw because of the modality or the systematic differences between the arms that may not even be necessary? We unfortunately can’t know for sure.
Last edited:
- Joined
- Jun 21, 2022
- Messages
- 944
- Reaction score
- 3,286
Very interesting points on tamsulosin and adaptive. Maybe the med should be controlled in the trial setting. I disagree on value of adaptive for prostate, but thats okay, we only go off personal experience here there is no data really.
I don't think anyone should buy an MRL for prostate SBRT, but if I had one I would use the MIRAGE protocol and offer it to eligible patients.
I also wonder if CT based on table adaptive will make MRL "obsolete" for most practices. Those new CBCTs from Varian looked pretty good and it is a way more versatile and faster machine.
I don't think anyone should buy an MRL for prostate SBRT, but if I had one I would use the MIRAGE protocol and offer it to eligible patients.
I also wonder if CT based on table adaptive will make MRL "obsolete" for most practices. Those new CBCTs from Varian looked pretty good and it is a way more versatile and faster machine.
- Joined
- Dec 17, 2007
- Messages
- 3,390
- Reaction score
- 4,403
Meanwhile, it appears that ViewRay is still doing projects with US government labs?
Last edited:
- Joined
- Dec 2, 2011
- Messages
- 1,487
- Reaction score
- 3,010
Taking politics, billing, competitive marketing, and all that jazz out of the equation, this is a very fair question and I think you and I are (as usual
) on the same page. From a scientific/clinical perspective, I think someone nailed it above or in a recent thread with their knife analogy. Arguing that MR vs CT IGRT even remotely compares to CBCT vs portal imaging is ridiculous. We are rapidly approaching the wall of diminishing returns. On paper, MRaIGRT has some potential advantages over CTaIGRT. The soft tissue resolution is a bit better and it does enable continuous cine imaging and the opportunity for some pretty slick gaiting options. On some level, the plans delivered with MRaIGRT are probably more "accurate" than CTaIGRT. But different enough to be clinically meaningful or accept reduced throughput? I wouldn't personally bet on it. I have no issue with people doing well controlled comparisons because you never know. But Elekta is all in on the Unity and these are potentially existential questions that should be pretty anxiety provoking.totally agree. It is a useful tool for some situations, no question, but ultimately its just a cool toy that is likely marginally improving outcomes in some particular scenarios
there are things we do that clearly feel better, but are hard to prove.
Take 4D for lung. We all do it and it makes sense to do it. but if you didnt assess motion, with a 5 mm CTV and 5 mm PTV, you likely arent going to miss. it is hard to miss with a CBCT daily and essentially a 1 cm margin on tumor. but if people use bigger margins because they dont use 4D? Yeah the metrics will be higher and at population level the risk of pneumonitis should be higher, but would anyone ever be able to show that in a trial? very likely no.
We are talking about one awesome knife versus another knife that has slightly different capabilities. if we all had that knife in our arsenal, we wouldn't care and would know it for what it is.
there are things we do that clearly feel better, but are hard to prove.
Take 4D for lung. We all do it and it makes sense to do it. but if you didnt assess motion, with a 5 mm CTV and 5 mm PTV, you likely arent going to miss. it is hard to miss with a CBCT daily and essentially a 1 cm margin on tumor. but if people use bigger margins because they dont use 4D? Yeah the metrics will be higher and at population level the risk of pneumonitis should be higher, but would anyone ever be able to show that in a trial? very likely no.
We are talking about one awesome knife versus another knife that has slightly different capabilities. if we all had that knife in our arsenal, we wouldn't care and would know it for what it is.
Last edited:
- Joined
- Dec 17, 2007
- Messages
- 3,390
- Reaction score
- 4,403
- Joined
- Apr 21, 2011
- Messages
- 3,622
- Reaction score
- 9,248
Continuing with the "knife" analogy, If we say linac based photon radiation is open surgery, if feels like platforms such as MR adaptive and protons are the rough equivalent of robotic surgery. Been around forever but still "feels" "high-tech" and "cutting-edge." Seems intuitively like it "should" be better and more "precise", but very little evidence indicating that clinically. And the evidence that does support is mainly tortured statistics of endpoints of dubious relevance. Some evidence pointing to the exact opposite, that it may do more harm than good, particularly in certain hands is roundly ignored. Certainly much more expensive and nice to market directly to patients, so the beat goes on.
FWIW: there's MUCH more evidence supporting the use of robotic surgery than protons.
FWIW: there's MUCH more evidence supporting the use of robotic surgery than protons.
- Joined
- Apr 21, 2011
- Messages
- 3,622
- Reaction score
- 9,248
I'm sure Varian is working on it. Tomo was able to do some rudimentary version of this nearly 2 decades ago with the lousiest looking MVCTs ever seen. Additionally, if "adaptive planning" is omitted from ROCR, the demand for CBCT based adaptive planning will be so great, every center in America will purchase.
- Joined
- Dec 15, 2014
- Messages
- 306
- Reaction score
- 580
What ptv dose for prostate sbrt that you see less toxicity?
- Joined
- Dec 2, 2011
- Messages
- 1,487
- Reaction score
- 3,010
3 mm margins to 37.5 Gy in 5 fractions. A touch lower than MIRAGE. I suspect the smaller margin has more to do with it than the lower dose but it is a variable.
