Timmerman Editorial and Discussion

[deleted1111261]

Yah, it’s not the ethics …

It’s considering the counter factual …
If we had many more grade 5s, then this cheeky editorial doesn’t appear as cute.

On the other hand - as a specialty should we be more accepting of grace 5 toxicity ?

---

Members don't see this ad.

 

[evilbooyaa]

I mean I think there are lots of things that have been 'learning experiences' such as all the 25 x1 early pancreas SBRT data from stanford. Similarly, 54/3 for central tumors was a learning experience.

He got to post his thoughts that became lore because of how well renowned he was. Is this a problem? Sure. I don't think he was really as laissez-faire as he seems to have been back then, given he loaded the boat with other folks who felt that it seemed reasonable. While the editorial is written from his perspective, and thus will be about him, there were lots of chances for someone to basically go "whoa whoa whoa, idk about that" on any of these numbers.

I dont' even know what "not worked out" would've meant in this space - I guess if it was killing like 30-50% of all patients. Guess he'd have ended up like the neutron folks.

 

[deleted1111261]

Our mortality rates are exceedingly low and we have patients with terminal illness.

One could say that Chris Crane/Parag Parikh are doing something similar in pancreas space - 67.5/15 or 75/25 was unheard of, but now even this Parikh does it.

Thinking about it more… pushing the envelope is probably the best we can do with our patients. I still feel like RTOG NSCLC set us back so far. Instinctually, 60 Gy isn’t enough. Instead of re-litigating PORT, why not 70-80 Gy with VMAT, tight margins, gating and very restrictive cardiac constraints? Let’s get a Ph3 for pancreas cancer with these ablative doses. Let’s turn rectal cancer into a non-operatively managed disease.

Anyway… a man can dream

 

[jondunn]

there was another negative dose esclation study in NSCLC, this time guided by PET. RTOG 1106, abstract this year, i havent seen paper yet.

sorry - that to me is a lost cause, there is no reason to do another dose escalation lung (or esophagus!) trial.

2 negative dose esclation esophagus trials came out in the last year.

way more interesting things to pursue in lung, IMO

 

[elementaryschooleconomics]

I don't think the experts who make the sausage are at all surprised how the sausage is made. After all, the table itself proclaims widely, "Mostly unvalidated!"

Totally agree. That's part of why this captured my attention. I don't want to go so far as to say "this is the part you don't say out loud" - because I'm glad he did, so we can think and talk about it - but I don't know if I would have written something like this before I retired.

The man is bold, that's for sure!

 

[Lamount]

p

Our mortality rates are exceedingly low and we have patients with terminal illness.

One could say that Chris Crane/Parag Parikh are doing something similar in pancreas space - 67.5/15 or 75/25 was unheard of, but now even this Parikh does it.

Thinking about it more… pushing the envelope is probably the best we can do with our patients. I still feel like RTOG NSCLC set us back so far. Instinctually, 60 Gy isn’t enough. Instead of re-litigating PORT, why not 70-80 Gy with VMAT, tight margins, gating and very restrictive cardiac constraints? Let’s get a Ph3 for pancreas cancer with these ablative doses. Let’s turn rectal cancer into a non-operatively managed disease.

Anyway… a man can dream

For many, a big obstacle to dose escalation is an unwillingness to utilize dose painting. I have become a big fan of this for central lung mets

We were all taught in residency that you need homogeneous dose coverage… but when you have a radioresistant histology next to a sensitive OAR, something’s gotta give.

 

[wandering star]

Yah, it’s not the ethics …

It’s considering the counter factual …
If we had many more grade 5s, then this cheeky editorial doesn’t appear as cute.

On the other hand - as a specialty should we be more accepting of grace 5 toxicity ?

I also remember reading a book chapter by Timmerman, where he wrote something like, "We were expecting [list of toxicities]. But, to our surprise, they never materialized!"

The reality is that some of our best ideas are surprises and can't have been easily expected. CRISPR-Cas9 was kind of an accident of more basic research, not a purposeful campaign for gene editing. The blue (or white) LED was pursued by many companies for many years and the guy who invented it and who received a Nobel Prize later was using an approach that was widely unexpected to work. The inventor of the mRNA vaccine did not get NIH funding while at UPenn and as a result had to leave academia.

In the counterfactual universe where lung SBRT was much more toxic, we never would have gotten to this stage. They would have tried it on a few dozen patients with serious illness and no realistic alternatives. It would have been just another treatment regimen with risks that exceeded the benefits. Some academics would read the paper and say "oh dear, what were they thinking, of course you would kill a few patients," perhaps not realizing the potential upside if it DOES work, which was thankfully the case for Timmerman. But mostly it would be forgotten.

 

[elementaryschooleconomics]

In the counterfactual universe where lung SBRT was much more toxic, we never would have gotten to this stage. They would have tried it on a few dozen patients with serious illness and no realistic alternatives. It would have been just another treatment regimen with risks that exceeded the benefits. Some academics would read the paper and say "oh dear, what were they thinking, of course you would kill a few patients," perhaps not realizing the potential upside if it DOES work, which was thankfully the case for Timmerman. But mostly it would be forgotten.

An excellent point, and probably a very realistic version of alternative reality.

I would not call publication and dissemination of these tables unethical, not in any sense.

The ethical gray zone is entered when he plainly spells out his intent with publishing the constraints in an editorial is to give the constraints an official reference while avoiding scrutiny (and the need for validation/references). He clearly never lied about their validity (the language in the RTOG 0236 protocol is imbued with authority but makes no mention of validity; the published table explicitly calls them "mostly unvalidated").

If you read the passage from the 2008 editorial about the table, he is again clear about his constraints:

Based on my understanding of the timeline, by 2008 there should have been enough observational data and mathematical models to publish them in a traditionally peer-reviewed manner, yes? It might not have been in a journal with a Nature-level Impact Factor, but still.

Sitting here on December 17th, 2021, do I think Timmerman's decisions were "good", in the sense that they produced a good outcome? Yes, of course, and I think everyone else should think that too. I wish we had 100 clones of him. He explicitly states his constraints are not validated by long term follow-up, and he hopes they will be replaced by constraints which are actually valid. To @sueyom's point, I would consider myself in alliance with Timmerman and everything he's done.

It is because I admire Dr Timmerman and his work that I find myself disquieted with my own answer to that question. Because if you abstract out actions with stated intent, you are left with:

"A physician believed he was right but did not feel he could support his belief with the scientific method, so he exploited the academic system to publish something which he could subsequently reference to add authority to future work."

He literally writes "who says cheaters never win" in the editorial.

If you then abstract out the consequence of that, you're left with "A physician publishes his beliefs not supported by the scientific method in a way that other physicians could use to treat their own patients".

I don't think there's a real answer here. In a vacuum, it appears unethical. In the context of reality and history, we're (as a species) better off because of his actions, which you can't really say about many people, living or historical.

I would use this scenario as a medical school interview question, just to see what people would say.

 

[Radonky]

An excellent point, and probably a very realistic version of alternative reality.


The ethical gray zone is entered when he plainly spells out his intent with publishing the constraints in an editorial is to give the constraints an official reference while avoiding scrutiny (and the need for validation/references). He clearly never lied about their validity (the language in the RTOG 0236 protocol is imbued with authority but makes no mention of validity; the published table explicitly calls them "mostly unvalidated").

If you read the passage from the 2008 editorial about the table, he is again clear about his constraints:

View attachment 346967

Based on my understanding of the timeline, by 2008 there should have been enough observational data and mathematical models to publish them in a traditionally peer-reviewed manner, yes? It might not have been in a journal with a Nature-level Impact Factor, but still.

Sitting here on December 17th, 2021, do I think Timmerman's decisions were "good", in the sense that they produced a good outcome? Yes, of course, and I think everyone else should think that too. I wish we had 100 clones of him. He explicitly states his constraints are not validated by long term follow-up, and he hopes they will be replaced by constraints which are actually valid. To @sueyom's point, I would consider myself in alliance with Timmerman and everything he's done.

It is because I admire Dr Timmerman and his work that I find myself disquieted with my own answer to that question. Because if you abstract out actions with stated intent, you are left with:

"A physician believed he was right but did not feel he could support his belief with the scientific method, so he exploited the academic system to publish something which he could subsequently reference to add authority to future work."

He literally writes "who says cheaters never win" in the editorial.

If you then abstract out the consequence of that, you're left with "A physician publishes his beliefs not supported by the scientific method in a way that other physicians could use to treat their own patients".

I don't think there's a real answer here. In a vacuum, it appears unethical. In the context of reality and history, we're (as a species) better off because of his actions, which you can't really say about many people, living or historical.

I would use this scenario as a medical school interview question, just to see what people would say.

The thing is he was upfront about it. He worked the system in the sense that he knew what would happen once he put his work in a citation, but he was upfront in the title and commentary. It would be different had he pretended there was other than extrapolation behind his numbers, then I would agree it would be unethical. His commentary is very frank!

 

[taserlaser]

An excellent point, and probably a very realistic version of alternative reality.


The ethical gray zone is entered when he plainly spells out his intent with publishing the constraints in an editorial is to give the constraints an official reference while avoiding scrutiny (and the need for validation/references). He clearly never lied about their validity (the language in the RTOG 0236 protocol is imbued with authority but makes no mention of validity; the published table explicitly calls them "mostly unvalidated").

If you read the passage from the 2008 editorial about the table, he is again clear about his constraints:

View attachment 346967

Based on my understanding of the timeline, by 2008 there should have been enough observational data and mathematical models to publish them in a traditionally peer-reviewed manner, yes? It might not have been in a journal with a Nature-level Impact Factor, but still.

Sitting here on December 17th, 2021, do I think Timmerman's decisions were "good", in the sense that they produced a good outcome? Yes, of course, and I think everyone else should think that too. I wish we had 100 clones of him. He explicitly states his constraints are not validated by long term follow-up, and he hopes they will be replaced by constraints which are actually valid. To @sueyom's point, I would consider myself in alliance with Timmerman and everything he's done.

It is because I admire Dr Timmerman and his work that I find myself disquieted with my own answer to that question. Because if you abstract out actions with stated intent, you are left with:

"A physician believed he was right but did not feel he could support his belief with the scientific method, so he exploited the academic system to publish something which he could subsequently reference to add authority to future work."

He literally writes "who says cheaters never win" in the editorial.

If you then abstract out the consequence of that, you're left with "A physician publishes his beliefs not supported by the scientific method in a way that other physicians could use to treat their own patients".

I don't think there's a real answer here. In a vacuum, it appears unethical. In the context of reality and history, we're (as a species) better off because of his actions, which you can't really say about many people, living or historical.

I would use this scenario as a medical school interview question, just to see what people would say.

Oh I’m definitely bookmarking this thread for a good coffee ethics discussion with some residents I have rotating with me in the next few months. It could make for a good journal club discussion too, in the right light. Regardless, it is food for thought and insight into how one of the most important toolkits in our arsenal came to be.

 

[communitydoc13]

What perhaps IS "unethical" is if these unvalidated tables get some undeserved respect and attention from the non-sausage making experts.

Great point. I would be interested in a list of dose constraints that are common, may impact clinical decision making away from meaningful dose escalation and remain very unvalidated.

I'll start with one. Great vessel constraints for thoracic SBRT.

 

[elementaryschooleconomics]

Great point. I would be interested in a list of dose constraints that are common, may impact clinical decision making away from meaningful dose escalation and remain very unvalidated.

I'll start with one. Great vessel constraints for thoracic SBRT.

Now I think we're wandering into what era did you train, and who/where trained you.

I strongly agree about Great Vessel constraints being problematic. The person I would consider most influential in my understanding and eventual independent practice of SBRT thought that Great Vessel "injury" from SBRT was very unlikely, and while they tried to minimize dose out of the principles of ALARA, never sacrificed coverage if vessel constraints were violated.

I'm not aware of any sequelae of that practice pattern. As such, I practice the exact same way. While I give Dosi GV constraints, and try to meet them...I generally won't send plans back if they don't.

 

[communitydoc13]

Thinking about it more… pushing the envelope is probably the best we can do with our patients. I still feel like RTOG NSCLC set us back so far. Instinctually, 60 Gy isn’t enough. Instead of re-litigating PORT, why not 70-80 Gy with VMAT, tight margins, gating and very restrictive cardiac constraints? Let’s get a Ph3 for pancreas cancer with these ablative doses. Let’s turn rectal cancer into a non-operatively managed disease.

I don't know. Our dose is contextualized by systemic therapy. The most obvious example of this is HL where doses went from 50+ -->30-->20-->0.

Significant among decreased risk of recurrence in the Pacific trial was intrathoracic recurrence. I still don't know how to interpret this. Often listed as new intrathoracic lesions, I don't know what the impact on actual true "local" as in XRT treated lesion failure was?

Timmerman did pick the right animal with early stage lung. Lots of it in poor performance status patients. Poor local control with std. fractionation. A parallel organ where small, high dose region likely acceptable.

I don't think the experts who make the sausage are at all surprised how the sausage is made.

No doubt. There are rare 30+ year careers building things up from basic insights to clinical applications (see Jim Allison) but in clinical medicine it's mostly just going for it with tons of unknowns and hoping for the best....then preferentially publishing the success stories.

 

[wandering star]

"A physician believed he was right but did not feel he could support his belief with the scientific method, so he exploited the academic system to publish something which he could subsequently reference to add authority to future work."

He literally writes "who says cheaters never win" in the editorial.

I read this paragraph not as a "Timmerman deathbed confessional: I cheated the system!!" but more in support of his main thesis: just because it's published and we did it this way in the past, doesn't mean we can't find a better way in the future. This is how he got to SBRT (being "unconstrained") and he wants to release these constraints on the next generation as well. I also agree that Timmerman could have published his tables elsewhere at a lower impact factor, or as a supporting table or an appendix saying "these were the constraints we used," and this would be enough to get the publication ball rolling.

He is totally right that there is a different level of scrutiny when you are trying to do something truly new versus sticking with what someone else did and citing it. So today, it is very easy to just copy #TimmermanTables and plead tradition, but in writing "who says cheaters never win" Timmerman is trying to get us to innovate and find a way to come up with something better in the future.

 

[elementaryschooleconomics]

He is totally right that there is a different level of scrutiny when you are trying to do something truly new versus sticking with what someone else did and citing it. So today, it is very easy to just copy #TimmermanTables and plead tradition, but in writing "who says cheaters never win" Timmerman is trying to get us to innovate and find a way to come up with something better in the future.

The core of my argument has nothing to do with Timmerman or the constraints, only the ethics of scientific publishing.

Without this editorial narrative, we would never have known that his intent of publishing the constraints in the way that he did was to avoid providing data to support them. Prior to his admission, the fact that his 2008 invited editorial - meant to frame the journal's special issue on SBRT - included the table, which could then subsequently be referenced, appeared to be just a fortunate byproduct. But stating his intent in doing this was to avoid editorial and peer-review changes the nature entirely. The "intent" of actions is incredibly important - it literally differentiates the legal definitions of manslaughter (voluntary or involuntary), for example.

While I strongly believe that innovation is fundamental to medicine (and is why we're not currently diagnosing people with "dyscrasia of the humors"), I doubt that hacking the scientific publishing process is the lesson Timmerman wanted us to learn. Again, if you read anything he's ever written about his original constraints (in protocols, papers, and presentations you can find on the internet), he's crystal clear about their lack of validity and never ever lies about it.

But the road to hell is paved with good intentions. It's our responsibility as physicians and scientists to carefully consider the ramifications of what we introduce into the literature. The authors of the 1980 NEJM letter on addiction had no idea it would be twisted to get a special label for OxyContin from the FDA. More recently, using citations without checking the strength/status of the source material lead to hundreds of papers citing the retracted Surgisphere/"hydroxychloroquine treats COVID" paper with negative consequences for public health in an ongoing global pandemic.

I am absolutely certain that there were (and are) doctors out there who truly believed hydroxychloroquine treats COVID. If the authors of the original Surgisphere paper believed they were doing the right thing and knowingly took liberties with scientific publishing to get their work out there, that's an extreme example of the ethical theme I'm talking about. What if it turned out that hydroxychloroquine REALLY DID cure COVID, and it's only 20 years from now we learn that the Surgisphere database didn't exist as it was described in the paper?

 

[communitydoc13]

I credit the candor of Timmerman's piece. I don't know him personally but he did spend some time at my department when I was in training, visiting and giving multiple talks.

I agree with ESE that something strikes me as off-putting. But I think it's the tone not the action. Timmerman is a little flippant and gloating (his words) and is speaking from the position of a winner (big time).

He could have simply stated, "There was not substantial data to drive meaningful constraints, but there was a large clinical need (the medically unresectable early stage lung cancer patient) and new technology that made previous considerations of safety likely obsolete. We (ethically here) had to go forward, and we needed to formalize (even quantify) our thought process so that others would have a template for jumping in and something explicit to establish reasonableness (a medical/legal standard to this day). So we did the best we could."

I would have felt better about it.

But he references watching TV, and out of touch luddites, and protestations that turned out to be wrong, and the privilege that goes along with being well connected (publishing that editorial for instance).

I think the candor is good for many reasons, as is his frequent use of the irksome term "thought leader". Irksome, because I'm not one. It speaks to the smallness of our field and how much personalities matter, and knowing people matters, and taking risks matters if you are going to be a somebody. This is not a bad story, it is just a very true story, and for those of us with a little less ambition and a little more of the "do no harm, because we are not inventing anything here" approach, we have to come to terms with it.

Three things happened roughly simultaneously during my training roughly 10 years ago that all dovetail with these issues.

1. Timmerman came and gave his talks. He was fairly prescient about ENI, stating that it was hardly ever going to be done in the future. This made me sad because learning how and what to contour in terms of elective nodal regions was a source of pride to me and most of my peers. He was kind. He was candid (I remember him stating that he could do many 3 fraction treatments in 1 fraction but he needed to pay his nurse). Like most trailblazers, he himself at times represented what would in time turn out to be the conservative position ("don't segment through a moving target!", even though that's what most of us do now).

2. A professor in my department had initiated and then closed early a moderately hypofractionated prostate trial (little bit more hypo than 60/20). If you were on call, you knew something was up, because you were getting calls about acute toxicity. I remember this attending saying, "is this really worth it?". Of course, this is a different disease than NSCLC.

3. In talking to a recently hired (and since very successful) physician scientist about "thinking hard about problems", he told me (as my memory permits), "Those days are over, sitting in a room and thinking hard. It is about collaboration and hustle, getting funding and taking chances".

Timmerman just letting us know how ballers ball.

 

[elementaryschooleconomics]

Three things happened roughly simultaneously during my training roughly 10 years ago that all dovetail with these issues.

1. Timmerman came and gave his talks. He was fairly prescient about ENI, stating that it was hardly ever going to be done in the future. This made me sad because learning how and what to contour in terms of elective nodal regions was a source of pride to me and most of my peers. He was kind. He was candid (I remember him stating that he could do many 3 fraction treatments in 1 fraction but he needed to pay his nurse). Like most trailblazers, he himself at times represented what would in time turn out to be the conservative position ("don't segment through a moving target!", even though that's what most of us do now).

2. A professor in my department had initiated and then closed early a moderately hypofractionated prostate trial (little bit more hypo than 60/20). If you were on call, you knew something was up, because you were getting calls about acute toxicity. I remember this attending saying, "is this really worth it?". Of course, this is a different disease than NSCLC.

3. In talking to a recently hired (and since very successful) physician scientist about "thinking hard about problems", he told me (as my memory permits), "Those days are over, sitting in a room and thinking hard. It is about collaboration and hustle, getting funding and taking chances".

Timmerman just letting us know how ballers ball.

Ah, the painful truth of these stories resonates deeply with me, and probably has a lot to do with how I interpreted the editorial as well. I'm really glad you brought this up.

1. While I don't recall ever meeting Timmerman in person (I could have, who knows), the timeline of my involvement with Radiation Oncology is such that he is an oft-discussed figure. Everyone describes his personality in the same way, which - it is what it is, we all have our own quirks. I only bring this up because I want to be clear that my personal thoughts about the piece have nothing to do with his - or others - personality quirks.

2. The timeline of training in modern Radiation Oncology has A LOT to do with how we perceive things. I have, and continue to, work with people who graduated residency from the late 1980s to now. The people who graduated 10 years ago, like @communitydoc13, have a very interesting perspective, depending on where they trained. While IMRT was ~10 years old (in a commercial, adoption sense - it was "invented" earlier, obviously), contouring and how we think about many things were notably different. Same with the adoption of SBRT. While I entered RadOnc from the mostly research side 15 years ago, my primary clinical experience started 5 years ago. For me, the Timmerman constraints were often treated like gospel. However, by then, we had A LOT more data about SBRT, and I didn't find it strange. But now, seeing the origin of these constraints, and remembering my personal experience with them...it's weird to juxtapose.

I guess I can't say this with authority, because I only asked a few people this question over the last week from a few institutions, but to me - it seems like those of us who trained in the modern SBRT era were taught that adhering to these constraints was of the utmost importance, which left me thinking, at least, that they "were right".

3. I have literally heard, and probably said, the same thing about thinking hard. It's actually what drove me to SDN to post anonymously. In real life, my accomplishments no longer felt real, they were often a product of who I knew and what I had done in the past. Here, we can at least (sometimes) discuss ideas in an egalitarian sense. Do you think I'd bring up the problematic Timmerman ethics using my real name? No way.

Anyway, I think everyone from different eras has their own Dovetail Stories which influences their feelings about this, which I find fascinating.

 

[Lamount]

Ah, the painful truth of these stories resonates deeply with me, and probably has a lot to do with how I interpreted the editorial as well. I'm really glad you brought this up.

1. While I don't recall ever meeting Timmerman in person (I could have, who knows), the timeline of my involvement with Radiation Oncology is such that he is an oft-discussed figure. Everyone describes his personality in the same way, which - it is what it is, we all have our own quirks. I only bring this up because I want to be clear that my personal thoughts about the piece have nothing to do with his - or others - personality quirks.

2. The timeline of training in modern Radiation Oncology has A LOT to do with how we perceive things. I have, and continue to, work with people who graduated residency from the late 1980s to now. The people who graduated 10 years ago, like @communitydoc13, have a very interesting perspective, depending on where they trained. While IMRT was ~10 years old (in a commercial, adoption sense - it was "invented" earlier, obviously), contouring and how we think about many things were notably different. Same with the adoption of SBRT. While I entered RadOnc from the mostly research side 15 years ago, my primary clinical experience started 5 years ago. For me, the Timmerman constraints were often treated like gospel. However, by then, we had A LOT more data about SBRT, and I didn't find it strange. But now, seeing the origin of these constraints, and remembering my personal experience with them...it's weird to juxtapose.

I guess I can't say this with authority, because I only asked a few people this question over the last week from a few institutions, but to me - it seems like those of us who trained in the modern SBRT era were taught that adhering to these constraints was of the utmost importance, which left me thinking, at least, that they "were right".

3. I have literally heard, and probably said, the same thing about thinking hard. It's actually what drove me to SDN to post anonymously. In real life, my accomplishments no longer felt real, they were often a product of who I knew and what I had done in the past. Here, we can at least (sometimes) discuss ideas in an egalitarian sense. Do you think I'd bring up the problematic Timmerman ethics using my real name? No way.

Anyway, I think everyone from different eras has their own Dovetail Stories which influences their feelings about this, which I find fascinating.

As someone who trained in the SBRT era…

During training, we were taught the “2 cm rule”, and would consider CRT and rarely 7.5 Gy x 8 in these cases.

As an attending, I was handed the timmerman constraints. I tend to use the timmerman constraints as a starting point but would often try to find some empirical outcomes data (NOT just the constraints from a protocol) to hone in on a better number. I usually end up being a little more conservative with the exception of LV.

I never perceived his constraints as “gospel”. They are merely based on simple LQ calculations from a reasonable EQD2 constraint, assuming a/b 2 for nerves, and a/b 3 for everything else. That’s how they always came across. I tended to round down to account for the uncertainties in the LQ assumption, which I would caution everyone to do. I would also caution everyone to limit point max constraints in thorax and GI to a PRV, rather than the actual OAR.

 

[elementaryschooleconomics]

As someone who trained in the SBRT era…

During training, we were taught the “2 cm rule”, and would consider CRT and rarely 7.5 Gy x 8 in these cases.

As an attending, I was handed the timmerman constraints. I tend to use the timmerman constraints as a starting point but would often try to find some empirical outcomes data (NOT just the constraints from a protocol) to hone in on a better number. I usually end up being a little more conservative with the exception of LV.

I never perceived his constraints as “gospel”. They are merely based on simple LQ calculations from a reasonable EQD2 constraint, assuming a/b 2 for nerves, and a/b 3 for everything else. That’s how they always came across. I tended to round down to account for the uncertainties in the LQ assumption, which I would caution everyone to do. I would also caution everyone to limit point max constraints in thorax and GI to a PRV, rather than the actual OAR.

I totally agree with you, and it sounds like we practice (and think about) SBRT the same way. I love PRVs and use them a lot, and think everyone else should, too.

But there were attendings I had in residency who were unwilling to appreciate nuance regarding constraints - in SBRT, mod hypo, or conventional. Sometimes it was clear they were doing it from a medicolegal point of view, other times...I think they honestly thought they were "real" (gospel).

 

[Lamount]

Great point. I would be interested in a list of dose constraints that are common, may impact clinical decision making away from meaningful dose escalation and remain very unvalidated.

I'll start with one. Great vessel constraints for thoracic SBRT.

Based upon this paper, it I think it is reasonably conservative to limit max Aorta EQD2 <140 Gy and PA < 120 (a/b=3). I don't think we know that exceeding these thresholds is dangerous (given how rare events are), but from the experience noted (mostly Japanese), it does seem that less than this is exceedingly unlikely to result in G3+ toxicity with GV.

 

[@JeffBradleyMD]

I have to admit that I am a Timmerman disciple. I read this SDN conversation with great interest. I'm glad that the editorial has generated so much interest! Basically, I think maintaining OAR dose constraints is one of the most important issues in Radiation Oncology. They continue to evolve with experience. Today most clinics have OAR dose limit templates for various fractionation schemes (i.e. for 1, 3, 5, 10 fractions, etc.) for most tumor locations. These are still mostly based in Bob's educated estimates. To add a bit of historical context, prior to Timmerman's hypofractionation OAR constraint tables, there were the Emami dose constraints. These were based on a meeting of experts led by Dr. Bahman Emami when 3DCRT was rolling out. Dr. Emami admits that these were based on experience and best estimates, not actual clinical data. I venture to say that these two contributions are two of the most important in the history of our field. At a time when Timmerman and McGarry brought lung SBRT to Indiana (circa 2000), the rest of us were using 70 Gy in 35 fractions with 2 or 3 fields and settling for a 30% local control rate. The risk they took, along with several others (including Joern Wulf, Hiroshi Onishi, Yasushi Nagata, etc), increased local control rates for the same population to >90%. It was a risk.... and it produced a profound result. ...and yes, the RTOG knew that these were best estimates when we rolled out RTOG 0236. Local control and survival rates were terrible at the time. Timmerman, Choy, Curran, myself and others felt it was worth the risk as long as we did it on a carefully controlled multi-institutional prospective clinical trial.

Sam Hellman used to tell us as residents that the Hippocratic Oath 'do no harm' mantra had a flaw. Sometimes we need to take a risk to do some good.

 

[elementaryschooleconomics]

I have to admit that I am a Timmerman disciple. I read this SDN conversation with great interest. I'm glad that the editorial has generated so much interest! Basically, I think maintaining OAR dose constraints is one of the most important issues in Radiation Oncology. They continue to evolve with experience. Today most clinics have OAR dose limit templates for various fractionation schemes (i.e. for 1, 3, 5, 10 fractions, etc.) for most tumor locations. These are still mostly based in Bob's educated estimates. To add a bit of historical context, prior to Timmerman's hypofractionation OAR constraint tables, there were the Emami dose constraints. These were based on a meeting of experts led by Dr. Bahman Emami when 3DCRT was rolling out. Dr. Emami admits that these were based on experience and best estimates, not actual clinical data. I venture to say that these two contributions are two of the most important in the history of our field. At a time when Timmerman and McGarry brought lung SBRT to Indiana (circa 2000), the rest of us were using 70 Gy in 35 fractions with 2 or 3 fields and settling for a 30% local control rate. The risk they took, along with several others (including Joern Wulf, Hiroshi Onishi, Yasushi Nagata, etc), increased local control rates for the same population to >90%. It was a risk.... and it produced a profound result. ...and yes, the RTOG knew that these were best estimates when we rolled out RTOG 0236. Local control and survival rates were terrible at the time. Timmerman, Choy, Curran, myself and others felt it was worth the risk as long as we did it on a carefully controlled multi-institutional prospective clinical trial.

Sam Hellman used to tell us as residents that the Hippocratic Oath 'do no harm' mantra had a flaw. Sometimes we need to take a risk to do some good.

Thank you for your perspective!

If the RTOG and the investigators who enrolled/treated patients knew that the constraints were "best estimates", and arrived at through the methods described in the editorial, it resolves the ethical concerns I had regarding that part of the narrative. The way the editorial is written, it is not clear to those of us who weren't involved how deep of a "behind the scenes" view the trialists had regarding the origin of the constraints.

I do agree that medicine cannot progress without risks, and our field has a long history of troubling conservatism in this area (compared to other specialties). These risks can be taken while still within the realm of classic medical ethics. There are a myriad of inspiring lessons we could all learn from the pioneering development of SBRT, though I would hope that engaging in a version of citation manipulation is not one of them. It is clear from some of the replies in this thread that publishing the tables in the way that he did is a trivial matter in the face of all the good his efforts have done for the world, and perhaps it is.

However, in an era where academic publishing is the "coin of the realm" for not only scientists but also clinically-oriented physicians, at a time when the rise of the publicly available "pre-print" can have immediate consequences (ref: all the COVID controversies), it is precisely the best of us - like Bob Timmerman - who should "show us the way", as it were.

What a blockbuster opening issue, @sueyom. What's on deck for next month?

 

[Radonky]

Sam Hellman used to tell us as residents that the Hippocratic Oath 'do no harm' mantra had a flaw. Sometimes we need to take a risk to do some good.

Love this

 

[ramsesthenice]

We were all taught in residency that you need homogeneous dose coverage… but when you have a radioresistant histology next to a sensitive OAR, something’s gotta give.

Bingo. The Brachytherapist in me calls bulli**** on this concept. We prescribe such that the D90 receives RX dose. Why in the world is it ok to under dose up to 10% of the CTV? Because you absolutely nuke a large portion of it. Do we completely understand the biology of why this works? No. But it does.

Same for panc dose escalation. It only works because you accept less coverage than normal. I almost never get above 80% coverage of a tight margin PTV when going to 50/5. But I do make sure that 95% gets 30 Gy. In effect, you could argue I am using conventional SBRT dosing but letting it get very hot. That, by definition, is dose painting.

 

[sueyom]

What a blockbuster opening issue, @sueyom. What's on deck for next month?

honestly this was a little overwhelming so I think we will need to take it easy for a while LOL
but I am happy we were able to bring this "forgotten history" to light and very grateful to Bob Timmerman for sharing it with a new generation, it is one of the important stories in our specialty
if there are any other controversies you want to know about, lmk and we'll get back to you in 2023

 

[sueyom]

Timmerman just letting us know how ballers ball.

I have to add, you kind of gotta love that last line
thanks for making me laugh

 

[elementaryschooleconomics]

honestly this was a little overwhelming so I think we will need to take it easy for a while LOL
but I am happy we were able to bring this "forgotten history" to light and very grateful to Bob Timmerman for sharing it with a new generation, it is one of the important stories in our specialty
if there are any other controversies you want to know about, lmk and we'll get back to you in 2023

Well, I think this is an absolutely fascinating mix of factors that is probably pretty rare:

1) A still-active (not retired) person who was a core pioneer of

2) A controversial (for the time) technique which

3) Was ultimately demonstrated to have significantly safe and positive effects and

4) Changed how an entire specialty (and the discipline of Oncology) thinks about and approaches many diseases

5) Authoring, by themselves, an editorial about a part of the development of the technique which

6) Had (has) tremendous implications regarding patient safety

7) Within which, he candidly asks the field to produce/adhere to something better/more data driven

8) and in doing so, reveals that his process wasn't perfect from a few angles.

Since Bradley gave context on how the RTOG knew the ingredients of the "secret sauce" for the constraints, I think that resolves what I viewed as potentially problematic for 0236, namely, that all involved parties knew the theoretical nature of the constraints and consented to participate in the trial with that understanding.

Regarding the second issue, which I somewhat harshly label "citation manipulation": as part of my education in an MD-PhD program, when I entered the grad phase, I was required to take an entire course on the ethical conduct of research. All of the other students in my program were required to take that course, and they always had been, and (presumably) always will be. While I think everyone is aware that things like plagiarism and data fabrication are problematic (duh), there are SO MANY OTHER potential issues that I had no idea about until that course.

However, I don't recall ever being required to take a similar course in the medical side of my education. In talking with friends and colleagues, I think having MD-only students be exposed to something like that, formally, is rare - though maybe this has changed in the modern era.

Historically, Radiation Oncology has heavily incentivized participation in research by people who might not have formal instruction in conducting and disseminating research, mentored by faculty who might also lack formal training. While everyone is smart enough to understand the ethical engagement of research in broad strokes, it's the nuanced points that are worth highlighting.

Talking about these nuances in major publications (the Red Journal) and heavily viewed forums (SDN) is important to aid in the "informal" education for us all. I am deeply grateful to you and Timmerman for publishing something like this, and it adds to the Lore of Bob Timmerman.

 

[sueyom]

Well, I think this is an absolutely fascinating mix of factors that is probably pretty rare:

1) A still-active (not retired) person who was a core pioneer of

2) A controversial (for the time) technique which

3) Was ultimately demonstrated to have significantly safe and positive effects and

4) Changed how an entire specialty (and the discipline of Oncology) thinks about and approaches many diseases

5) Authoring, by themselves, an editorial about a part of the development of the technique which

6) Had (has) tremendous implications regarding patient safety

7) Within which, he candidly asks the field to produce/adhere to something better/more data driven

8) and in doing so, reveals that his process wasn't perfect from a few angles.

Since Bradley gave context on how the RTOG knew the ingredients of the "secret sauce" for the constraints, I think that resolves what I viewed as potentially problematic for 0236, namely, that all involved parties knew the theoretical nature of the constraints and consented to participate in the trial with that understanding.

Regarding the second issue, which I somewhat harshly label "citation manipulation": as part of my education in an MD-PhD program, when I entered the grad phase, I was required to take an entire course on the ethical conduct of research. All of the other students in my program were required to take that course, and they always had been, and (presumably) always will be. While I think everyone is aware that things like plagiarism and data fabrication are problematic (duh), there are SO MANY OTHER potential issues that I had no idea about until that course.

However, I don't recall ever being required to take a similar course in the medical side of my education. In talking with friends and colleagues, I think having MD-only students be exposed to something like that, formally, is rare - though maybe this has changed in the modern era.

Historically, Radiation Oncology has heavily incentivized participation in research by people who might not have formal instruction in conducting and disseminating research, mentored by faculty who might also lack formal training. While everyone is smart enough to understand the ethical engagement of research in broad strokes, it's the nuanced points that are worth highlighting.

Talking about these nuances in major publications (the Red Journal) and heavily viewed forums (SDN) is important to aid in the "informal" education for us all. I am deeply grateful to you and Timmerman for publishing something like this, and it adds to the Lore of Bob Timmerman.

yes as I progress in my career there are so many ethical issues that are not part of courses like that, this is where I think it's really important to have mentors and peers and research coordinators who can check you and keep you on track
I totally agree there is still a lot to do in the teaching of high quality research methods
and also agree, this is sort of a perfect storm we may never replicate - but kind of fun to start the New Year with some excitement, we'll try again for another splashy holiday

 

[sueyom]

i just noticed they put all the timmerman tables into radonc review, sheesh those guys are fast
poor Bob, i hope someone tells him what is happening on the Internet since he does not do much social media