Rad Onc Twitter

[NotMattSpraker]

---

Members do not see this ad.

View attachment 411416

This does not look very "real world" to me...
Or am I the only one?

A lot of people are doing Lattice SBRT treatments and/or other techniques like GRID in the community and internationally in all kinds of settings. I talk to them all the time. The Wash U protocol was designed for generalizability to the community, and you can do it pretty easily if you have decent physics and an SBRT program. My (community hospital) group even created an AI tool for automated contouring in Aria that we would be happy to share.

I've done Lattice and also PUSLAR type treatments in the community, but only in the palliative setting where there there is not a clear radiotherapy standard of care.

I know some are investigating using spatially fractionated boosts for bulky tumors in the curative setting. I would not deviate from an established standard off trial, it is way too early for that IMO.

 

[madchemist89]

We use spatially fractionated treatment for bulky tumors in the palliative setting. I can’t say that I’ve noticed a difference, but when you are treating a 13 cm melanoma metastasis what else are you going to do.

 

[taserlaser]

Yeah, I would love to do this but our physics department might cry. If there was some comparative evidence showing it was better than usual or just dose escalated palliation, would be an easier sell. I imagine it might come at some point; some of the anecdotal responses are quite impressive.

 

[RickyScott]

Yeah, I would love to do this but our physics department might cry. If there was some comparative evidence showing it was better than usual or just dose escalated palliation, would be an easier sell. I imagine it might come at some point; some of the anecdotal responses are quite impressive.

I have 2 sarcoma reiteration pts out 4 years.

 

[OTN]

Yeah, I would love to do this but our physics department might cry. If there was some comparative evidence showing it was better than usual or just dose escalated palliation, would be an easier sell. I imagine it might come at some point; some of the anecdotal responses are quite impressive.

That's why I haven't pushed to start a program yet. While anecdotal data is impressive, anecdotal data also isn't data. I hope we will see more robust data in the future, but until then devoting so many resources to something that is yet unproven to be better doesn't make sense in our clinic.

 

[Palex80]

 

[TheWallnerus]

View attachment 411767

View attachment 411768

I feel like we are harming people long term with huge single fraction doses (versus the normofract alternatives). Time will tell. Hope there is 10 and 20 year follow up and comparisons to normofract (there won’t be). I saw a single fraction breast preop trial of like 35 Gy or something. That’s f**cking insane in terms of late effect 2 Gy per fraction equivalent. The “young” have not been taught to respect hard won decades old rad bio lessons.

 

[radiation]

I feel like we are harming people long term with huge single fraction doses (versus the normofract alternatives). Time will tell. Hope there is 10 and 20 year follow up and comparisons to normofract (there won’t be). I saw a single fraction breast preop trial of like 35 Gy or something. That’s f**cking insane in terms of late effect 2 Gy per fraction equivalent. The “young” have not been taught to respect hard won decades old rad bio lessons.

yeah someone should really teach that Timmerman guy about rad bio Ablative Preoperative Single-Fraction Radiation Dose Escalation Among Patients With Breast Cancer: A Phase 1 Nonrandomized Clinical Trial - PMC

 

[OTN]

yeah someone should really teach that Timmerman guy about rad bio Ablative Preoperative Single-Fraction Radiation Dose Escalation Among Patients With Breast Cancer: A Phase 1 Nonrandomized Clinical Trial - PMC

That trial only reported 3-month toxicity, so maybe someone actually should?

Edit: I know Zimmerman knows a ton about radbio, but we do need longer follow up before we breathlessly state single-fraction SBRT is equivalent to multi-fraction.

 

[TheWallnerus]

yeah someone should really teach that Timmerman guy about rad bio Ablative Preoperative Single-Fraction Radiation Dose Escalation Among Patients With Breast Cancer: A Phase 1 Nonrandomized Clinical Trial - PMC

Yeah maybe

Because if you take time/convenience out of the equation, then unless you’re anti-science or adhere to some psueudoscience, you’d be ok with a phase one trial of 300 Gy in 150 fx preop for breast cancer

 

[taserlaser]

This is a take I agree with. Lumpectomy is so well tolerated at baseline, it will take substantial improvement to move the needle. To reduce the risk of local recurrence for breast cancer requires elective treatment into normal breast parenchyma and ablative RT seems to me counterproductive to reasonable normal tissue effects there. Reasonable to explore as palliation in non operative patients I think, but preop or definitive SBRT for routine pts does not seem like a winning strategy to me. Good for them for advancing the field and doing good science, but just not something that I see panning out long term.

 

[Ray D. Ayshun]

Yeah maybe

Because if you take time/convenience out of the equation, then unless you’re anti-science or adhere to some psueudoscience, you’d be ok with a phase one trial of 300 Gy in 150 fx preop for breast cancer

View attachment 411804

try it, and I'll immediately move-in and undercut you with a more convenient 140 fx regimen.

 

[TheWallnerus]

This is a take I agree with. Lumpectomy is so well tolerated at baseline, it will take substantial improvement to move the needle. To reduce the risk of local recurrence for breast cancer requires elective treatment into normal breast parenchyma and ablative RT seems to me counterproductive to reasonable normal tissue effects there. Reasonable to explore as palliation in non operative patients I think, but preop or definitive SBRT for routine pts does not seem like a winning strategy to me. Good for them for advancing the field and doing good science, but just not something that I see panning out long term.

Nixon said if the president does it, it can’t be illegal

Rad oncs applying same line of reasoning to Timmerman, or any “leader” …. bad!

Anyone testing 38 Gy preop with EBRT for breast cancer needs heaps and heaps of opprobrium

 

[TheWallnerus]

Edit: I know Zimmerman knows a ton about radbio

But does he. I can tell you that if *I* (a nobody) tried 38 Gy in one fraction to a not so small volume inside a healthy woman’s boob you would think “he does not know a ton about rad bio.” And I don’t even know you. Even the smartest minds have a hard time thinking exponentially instead of linearly, and above 5 Gy per fraction or so exponential dominates.

 

[Gfunk6]

1. Most of the ultra-hypofractionated prostate and breast work is coming from national health systems where there’s strong pressure for shorter, cheaper regimens. They’re meticulous about acute toxicity, but they’re clearly more willing than most US practices to accept uncertainty about very-late, moderate toxicity if the efficiency gains are large.
2. Using LQ/BED to convert SBRT regimens to a ‘conventional EQD2’ is increasingly unreliable once you’re above about 6–8 Gy per fraction. Past that, ‘equivalence’ is more fiction than physics.

 

[TheWallnerus]

I think the BED formula breaks ~7 Gy per fraction or so.

I'm ready to have a complete come to Jesus talk about this.

First, extraordinary claims require extraordinary evidence. What evidence is there that "the BED formula breaks ~7Gy per fraction or so." And which way does it break; the alpha/beta ratio bends upward or downward and resultant BEDs bend oppositely? And it breaks for acute or late effects... or both? Citations desperately needed.

This has become a weird persistent delusion in rad onc. In other words, we just have NO model of how our treatments work past ~7 Gy per fraction??? Because I have heard of no other model than the LQ formalism (the "BED formula"). Every single cell survival curve in Hall plots the single fraction doses out past 7 Gy. And every single one matches a linear-quadratic equation. If DNA were a single strand, or a triple helix, it would not, but linear-quadratic matches DNA double helix-ness. (In fact, you can have a much firmer ground to stand on if you say LQ doesn't work as well in the sub-1-Gy-per-fraction range... that I'm open to hearing.)

The ENTIRETY of current trends to lessen prostate and breast fractions, even down to five fractions, is based on LQ. The ENTIRETY of "guesses" for 5-fraction dose regimen variability is based on LQ. Every single SBRT paper worth any substance/salt I've ever seen references LQ, especially for tumor effects. I have not seen a new model proposed.

"I think the BED formula breaks ~7 Gy" is simply inherently un-falsifiable. In other words, it's... poppycock, scientifically speaking.

 

[TheWallnerus]

The people who run single fraction trials in prostate and breast are probably not as overly concerned with late toxicity as you.

They actually totally are. You see LQ mentioned in every paper... an example from a 19 Gy/1 fx prostate paper:

A 38 Gy/1 fx calculation, like showed above or what I did on the EQD2 website, would/should scare the bejesus out of a rad onc... likely why LQ or radiation biology of any sort is not mentioned in their paper as far as I could tell.

But maybe Timmerman is just too smart for BED or LQ etc and/or has data falsifying LQ. Anything's possible.

Good for them for advancing the field and doing good science

I can't understand why testing 38 Gy/1 fx preop for breast cancer is "good science" ... I suppose in scenarios like this my desire for obsequiousness goes out the window

 

[Gfunk6]

I'm ready to have a complete come to Jesus talk about this.

First, extraordinary claims require extraordinary evidence. What evidence is there that "the BED formula breaks ~7Gy per fraction or so." And which way does it break; the alpha/beta ratio bends upward or downward and resultant BEDs bend oppositely? And it breaks for acute or late effects... or both? Citations desperately needed.

This has become a weird persistent delusion in rad onc. In other words, we just have NO model of how our treatments work past ~7 Gy per fraction??? Because I have heard of no other model than the LQ formalism (the "BED formula"). Every single cell survival curve in Hall plots the single fraction doses out past 7 Gy. And every single one matches a linear-quadratic equation. If DNA were a single strand, or a triple helix, it would not, but linear-quadratic matches DNA double helix-ness. (In fact, you can have a much firmer ground to stand on if you say LQ doesn't work as well in the sub-1-Gy-per-fraction range... that I'm open to hearing.)

The ENTIRETY of current trends to lessen prostate and breast fractions, even down to five fractions, is based on LQ. The ENTIRETY of "guesses" for 5-fraction dose regimen variability is based on LQ. Every single SBRT paper worth any substance/salt I've ever seen references LQ, especially for tumor effects. I have not seen a new model proposed.

"I think the BED formula breaks ~7 Gy" is simply inherently un-falsifiable. In other words, it's... poppycock, scientifically speaking.

View attachment 411807

First citation attached.

Relevant quote: "However, in RTOG 0915, the capability of the LQ model to describe dose relation effects when the dose fraction is higher than 8 Gy has been questioned."

Second citation: https://journals.sagepub.com/doi/10.1177/1559325819828623?utm_source=chatgpt.com

Relevant quote: "A new EUD model was derived to compare treatment plans with different time-dose fraction schemes and dose distributions. The sectional function solves the problem that the LQ model is not accurate and applicable to high doses per fraction. Based on this mathematical model, a new EUD model derived is instructive to compare the radiation effects of volumes with various dose distributions."

 

[Gfunk6]

They actually totally are. You see LQ mentioned in every paper... an example from a 19 Gy/1 fx prostate paper:

Well yes, it is considered of course. However, my point is that what is considered acceptable late toxicity in countires with socialized medicine may not be the same as the US.

 

[drowsy12]

Well yes, it is considered of course. However, my point is that what is considered acceptable late toxicity in countires with socialized medicine may not be the same as the US.

do you have an example?

 

[Palex80]

However, my point is that what is considered acceptable late toxicity in countires with socialized medicine may not be the same as the US.

I question whether or not delivering one fraction of SBRT with 19 Gy, with all the precautions necessary in terms of immobilization, tracking, etc (as done in the trial) is less resource intensive than a 5 fraction course of "normal" SBRT, as in PACE.

EDIT: from the trial protocol

 

[Gfunk6]

do you have an example?

RTOG 0415 - low risk prostate cancer treated with conventional fractionation vs moderate hypofractionation. Late G2-G3 GI/GU events were increased in the hypofrac arm (~1.4) but was still viewed as non-inferior.

HYPRO - low risk prostate cancer treated with conventional fractionation with more accelrated hypofractionation. Moderately higher >= G3 late GU toxicity in hypofrac arm by about 7%.

 

[RickyScott]

I question whether or not delivering one fraction of SBRT with 19 Gy, with all the precautions necessary in terms of immobilization, tracking, etc (as done in the trial) is less resource intensive than a 5 fraction course of "normal" SBRT, as in PACE.

EDIT: from the trial protocol
View attachment 411816

Fractions after the first entail marginal resources. This type of effort is purely for careerism/publication overseas and to capture distant pts in the us.

 

[TheWallnerus]

First citation attached.

Relevant quote: "However, in RTOG 0915, the capability of the LQ model to describe dose relation effects when the dose fraction is higher than 8 Gy has been questioned."

Second citation: https://journals.sagepub.com/doi/10.1177/1559325819828623?utm_source=chatgpt.com

Relevant quote: "A new EUD model was derived to compare treatment plans with different time-dose fraction schemes and dose distributions. The sectional function solves the problem that the LQ model is not accurate and applicable to high doses per fraction. Based on this mathematical model, a new EUD model derived is instructive to compare the radiation effects of volumes with various dose distributions."

These are "cool story bro" links imho. RTOG 0915 was pretty small, didn't follow really long term, and the dose is so localized, and deep in the lung parenchyma away from anything our eyeballs would see, or could be a clinical issue (usually), you won't see the long term toxicities. (Already the 25 Gy in 1 fraction heart treatments are literally burning holes in stomachs... I can't find the pic/links but they're impressive). And for tumor effects, the RTOG 0915 outcomes match LQ.

The problem with "we've got a new model" papers I have seen so far is that some say LQ doesn't estimate cell kill enough (brain radiosurgeons like those papers), and some say it overestimates. The links above fall in to the latter category. But even so, 38 Gy/1 fx in any model is going to have not just twice the late effects of 19/1.

 

[drowsy12]

RTOG 0415 - low risk prostate cancer treated with conventional fractionation vs moderate hypofractionation. Late G2-G3 GI/GU events were increased in the hypofrac arm (~1.4) but was still viewed as non-inferior.

HYPRO - low risk prostate cancer treated with conventional fractionation with more accelrated hypofractionation. Moderately higher >= G3 late GU toxicity in hypofrac arm by about 7%.

yeah I guess I meant an example where the US did not adopt something that they did in Europe, to your statement.

moderate hypofrac has been widely adopted in the US.