That is a really great question, one that my wife often asks me, or a variant of it - "Are protons really worth the added cost?" Or as I think of it, are protons "better enough"?
The answer is a long one..one that I've been pondering for over 5 years.
I suppose it depends on what we value. If we only value local control, which is a function of dose to tumor, the answer is no, protons are not more valuable. 3D, or even 2D radiation, can give 70 or 80 Gy to any tumor.
However, if value is a function of reducing side effects to a socially acceptable level, then I think the answer is yes, or at least, in certain cases - yes, protons are "better enough."
The definition of socially acceptable varies from nation to nation and over time. The side effects that patients endured in the cobalt and pre-cobalt eras makes me wonder how we even survived as a field, and yet it was the best thing we had at the time, at least until Linacs got better and now mass produced.
The Netherlands
has been rationing their proton access in its 2 centers according to whether or not a patient's anatomy is predicted by a model to benefit from protons. Eg, in head and neck, will OAR constraints be unmet by IMRT, or "better enough" by protons? If so, then they will cover it. The computer model improves over time with more cases.
I apply a similar mental model in my clinic every day. Some patients want protons, no matter what, even if I don't know they will benefit. Others don't want to travel for protons, even I know it might save them severe toxicity. Most often I tell them that I won't know for sure if they'll need protons until we simulate them and make a comparison plan.
Typically now I can predict by exam and imaging who will be "better enough," ie, not able to meet breast/heart/LAD constraints with DIBH, or H&N constraints with IMRT/VMAT. However, it's taken me 5 years and a mid-career fellowship to get there. AI and autoplanning will hopefully pick it up more quickly than that, because there is a shortage of experienced people at all levels in the proton world, and I don't believe that a weekend course is good enough for a human to confidently make the switch at this point.
It's important to realize that in terms of patients treated, protons are only at 250k patients globally. That may sound like a lot, but it isn't a lot per doctor or center. IMRT probably got to that number in 30, because the number of annual Xray patients is about 99 times higher than protons.
We are still in the early days and the steep part of the learning curve, like when IMRT was burning people's lips unexpectedly and SBRT was causing rib fractures, chestwall necrosis, and a no-fly zone lung had not been identified. Some people I met didnt want to learn 3D contouring skills or prescribe 20 Gy per fraction, and never picked it up. IMRT even ended up on the front page of the New York times for accidentally killing a guy. I don't think any of us gave up IMRT over that, but it was a learning moment for our field and reinforced the need for QA before beaming on.
New indications will continue to emerge for particles, and side effects will go down, just as they have for IMRT and SBRT in the last 15 years. There are still about 100 metro areas in the US with >500k population, i.e. larger and economically healthier than Flint, with higher expectations for their healthcare, for whom protons will be "better enough."