ACGME accredited fellowship for rad onc

[Krukenberg]

As I mentioned, I did train with protons. Where I trained they do have a proton fellowship and I don't believe that they're doing any of the robustness calculations that you described.

PS: I contour oral cavity and lips routinely for H&N cases.

This is what I'm talking about: https://www.raysearchlabs.com/globa...e-paper-8---robust-omptimization-aug-2015.pdf

Robust optimization has come online and is now included in planning software packages probably in the last 2-4 years. If it's not standard yet at your center then it will be soon. It's most necessary when using pencil beam scanning with multi-field optimization (any one beam may not scan the whole tumor). If your center wasn't doing MFO when you were there then that's probably why. Whether your physicians even look at the uncertainty curves generated from RO is probably a leadership decision, but if you're evaluating a plan based on nominal PTV coverage then it logically follows that you should be evaluating how sensitive your CTV coverage and cord/brainstem doses will be to perturbations. The dosimetrist can play with beam angles, PBS spot sizes, spot placement, and soon PBS rescanning to increase robustness. If robustness still sucks then you switch to single-field optimization or IMRT.

Again, I'm not at all stating that any of this makes it better than IMRT. Just saying that these new concepts and technologies are what it takes to deliver dose via proton beam, safely.

---

Members don't see this ad.

 

[Neuronix]

Robust optimization has come online and is now included in planning software packages probably in the last 2-4 years. If it's not standard yet at your center then it will be soon. It's most necessary when using pencil beam scanning with multi-field optimization (any one beam may not scan the whole tumor). If your center wasn't doing MFO when you were there then that's probably why. Whether your physicians even look at the uncertainty curves generated from RO is probably a leadership decision, but if you're evaluating a plan based on nominal PTV coverage then it logically follows that you should be evaluating how sensitive your CTV coverage and cord/brainstem doses will be to perturbations. The dosimetrist can play with beam angles, PBS spot sizes, spot placement, and soon PBS rescanning to increase robustness. If robustness still sucks then you switch to single-field optimization or IMRT.

Again, I'm not at all stating that any of this makes it better than IMRT. Just saying that these new concepts and technologies are what it takes to deliver dose via proton beam, safely.

So... All the people who did proton fellowships 4+ years ago should do another one?

My argument was never that there aren't things specific to protons that you need to know. It was simply that you don't need a fellowship to learn those things.

 

[Krukenberg]

This is what I'm talking about: https://www.raysearchlabs.com/globa...e-paper-8---robust-omptimization-aug-2015.pdf

Robust optimization has come online and is now included in planning software packages probably in the last 2-4 years. If it's not standard yet at your center then it will be soon. It's most necessary when using pencil beam scanning with multi-field optimization (any one beam may not scan the whole tumor). If your center wasn't doing MFO when you were there then that's probably why. Whether your physicians even look at the uncertainty curves generated from RO is probably a leadership decision, but if you're evaluating a plan based on nominal PTV coverage then it logically follows that you should be evaluating how sensitive your CTV coverage and cord/brainstem doses will be to perturbations. The dosimetrist can play with beam angles, PBS spot sizes, spot placement, and soon PBS rescanning to increase robustness. If robustness still sucks then you switch to single-field optimization or IMRT.

Again, I'm not at all stating that any of this makes it better than IMRT. Just saying that these new concepts and technologies are what it takes to deliver dose via proton beam, safely.

Also evaluating target coverage with robust optimization is a total paradigm shift from what we do with IMRT (evaluating PTV coverage). In RO the planner robustly plans for CTV coverage. There's no "PTV" concept because the setup error and range uncertainty error are built into the robust optimization. The planner hands you a plan that could be just as conformal as IMRT in some parts of the target (the parts that are highly robust to uncertainty) and potentially LESS conformal in areas that are not robust to setup or range uncertainty.

 

[RO2019]

I agree with Krukenberg that treating proton does legitimately take extra training, for all the reasons he mention

the thing is that you don't need to do a year fellowship to learn it. If you're a smart well-trained rad onc, places will hire you and will be willing to send you to training courses to learn, and the rest you learn on the job.

But people making light of all that goes into proton planning and plan eval in 2019 aren't that sharp

 

[deleted969641]

A lot of questionable statements in this thread and notably overconfident statements from current 2nd or 3rd year residents (and I'm certainly no expert in a lot of things, but I always try to remember to know what I don't know).

However, I will say that the term "protons" is being thrown out like it's one type of treatment.

Are we walking about passively scattered protons, actively dynamic spot scanned protons also using single field uniform dose technique, or true "intensity-modulated" proton therapy (IMPT -- albeit somewhat of a misnomer as dynamic proton therapy is inherently "intensity" modulated). Are you just doing "field patching," the proton equivalent of IMRT? Or are you doing real IMPT? And if it's IMPT, is it distal edge tracking, 2-D, "2.5-D", or a 3-D modulation technique?

Remember, typically in IMPT you are modulating both fluence AND beam energy. You have more issues with homogeneity that can lead to problems with hot spot and cold spot areas in the PTV.

As you can see, there are many different ways of bombarding patients with H+ particles, each of which carries a variety of different challenges for physicists, dosimetrists, engineers, and clinicians.

Does that mean an ACGME-required fellowship is required to properly train a clinician to know how to use the equipment he/she will be using in practice? No, of course not. That's stupid.

 

[Krukenberg]

So... All the people who did proton fellowships 4+ years ago should do another one?

My argument was never that there aren't things specific to protons that you need to know. It was simply that you don't need a fellowship to learn those things.

I think that in order to appropriately manage these issues when your name is on the line, you need to have managed real patients with these technical issues in a training environment, not an online module or simulated training course. If you're taking a job in an experienced center where all of this is standard, then you would have experienced faculty and physicists who can guide you. But even still...when you're in that learning curve your patients are still at risk and ultimately your name is on the line. If you're taking a job at a new center where you would have a lot of influence over how the place runs, then I think training is very important.

As I said earlier, "time" in fellowship is less important than # of cases across X, Y, Z disease sites. To see all the technical problems you have to solve with proton treatment you really need to manage the patient from contour approval to end of treatment. If you multiply that amount of time by multiple cases and multiple disease sites then that is going to be longer than a couple of months.

 

[Lamount]

Learning the biology and physics of proton radiotherapy is extremely important if you wish to use this modality... but as others have mentioned, this can be learned in a class, rather than a fellowship. Conversely, clinical training in protons is germane to learning what you can actually get away with... and what you can't. There are things you can do with protons that you wouldn't ever try with photons. Re-treatment comes much more feasible; nerves are easier to chase; the skull base is much more manageable; you can treat 79 Gy adjacent to the spine without causing myelopathy... But you can also really hurt people as well if you don't understand the uncertainties. There are just some things that you cannot read in a book.

 

[scarbrtj]

But people making light of all that goes into proton planning and plan eval in 2019 aren't that sharp

Are proton patients' clinical outcomes different in 2019 than they were in, say, 1999? Anecdotal is fine. Published proof would be even better.

 

[RickyScott]

Are proton patients' clinical outcomes different in 2019 than they were in, say, 1999? Anecdotal is fine. Published proof would be even better.

Latest publication as far as I know that when you use rectal spacer/balloons and pencil beam (state of art), 8% of pts need to have rectum lazered within 2 years.

 

[medgator]

Latest publication as far as I know that when you use rectal spacer/balloons and pencil beam (state of art), 8% of pts need to have rectum lazered within 2 years.

I think imrt had better outcomes a decade ago

 

[OTN]

Radoncs have been dealing with new technical developments which impact treatments for years. IMRT, HDR, SRS/SBRT...these all carry their own technical considerations, have a learning curve and, yes, you can hurt people if not done correctly. However, no one argued that a practicing radonc needed to go back to fellowship to learn those techniques. They were learned through symposiums, classes, ASTRO sessions, on-the-job training, etc. Protons are no different.

 

[RollTideRadOnc]

Are proton patients' clinical outcomes different in 2019 than they were in, say, 1999? Anecdotal is fine. Published proof would be even better.

This FTW.

No evidence of any improved outcome despite how many theoretical DVH curves are generated. Given the benefits are likely purely dosimetric and marginal outside of special cases like choroma, clinical gains will be extremely difficult to demonstrate in any prospective manner.

This is the bar none one of the biggest examples of medical waste until proven otherwise.

And as OTN and others have pointed out, it's ridiculous to think that this can't be learned on the job plus a short intensive boot camp.

 

[scarbrtj]

Latest publication as far as I know that when you use rectal spacer/balloons and pencil beam (state of art), 8% of pts need to have rectum lazered within 2 years.

Yes. And "state of art"... horrible term. If it's new and costs more and shows improvement on paper (and takes an extra physics FTE), in rad onc it's state-of-art. We worship too much at The Shrine of The State of The Art. (I'm no luddite. But I don't want to wind up being a "technology virtue signaler" by saying stuff like "I do 10-12 theoretical DVH curves in my planning" either.) Off the top of my head, I can remember when BAT was state-of-art for prostate IMRT. Like, if you didn't do BAT you weren't s**t. BAT actually hurt setup accuracy.

 

[scarbrtj]

you can treat 79 Gy adjacent to the spine without causing myelopathy

What is the clinical indication for prescribing 79 Gy of protons (or cobalt gray equivalents) to something close to the spine?

 

[scarbrtj]

2) also contouring the lips/oral cavity is very standard.

I trained at a pretty respectable program; we were not lip contourers. Knock on wood: zero lip-related toxicities from IMRT in my whole career.

I probably need a lip contouring fellowship now.

 

[scarbrtj]

prime example is the "learning curve" that was seen in the Liao trial:

If I am reading this right, and thanks for citing, they saw greater radiation pneumonitis rates with protons vs IMRT, but they got SIGNIFICANTLY better with their proton planning/dosimetry over time? Gosh, I wonder how bad the proton patients did in the beginning.

 

[Lamount]

What is the clinical indication for prescribing 79 Gy of protons (or cobalt gray equivalents) to something close to the spine?

Definitive chordoma. If resection is not feasible, can treat with 77.4-79.2