The Opportunity for Protons in Head and Neck Cancer.
Head and Neck cancer is the proton therapy industry's greatest opportunity - here is why I think that is the case.
With 3 prospective trials on the horizon, proton therapy has the opportunity for the big win that it must show.
In my last post, we covered where I believed we were about 3 years ago.
Proton Therapy vs. IMRT for Head and Neck Cancer: A look back
Today, we’ll look at some data regarding what has transpired since that period and where I believe we are today. This article will briefly lay out why protons can and should be able to be shown superior in this patient subset and how, in a way, presents both real opportunity and risk for the industry.
Protons deliver far less integral dose:
First let’s just start with context that many radiation oncologist (I almost guarantee) under appreciate. Integral dose delivered to the patient is DRAMATICALLY less with proton therapy. Here is a pretty typical example for a unilateral proton plan compared to a pharyngeal constrictor optimized IMRT plan. In this case, in order to equalize the dose to the patient, you must deliver 200 cGy via a 10x10 field to DMax 35 times. That is simply a staggering number.
Here is a bit more precise breakdown of where those 35 treatments must be delivered - 15 can be delivered in the low neck - below the larynx. This treatment region is a relatively “low risk” area and so while they contribute to toxicity, that would not be the main region of toxicity difference between the two approaches.
Of the remaining 20 fractions, 10 can be delivered posterior to the parotids. So assuming spinal cord dosing is safe under either approach (which should happen), this is a relatively safe region to deliver 20 Gy. Again, similar to the low neck, this leads to what I would anticipate to be a limited difference in patient toxicity.
That leaves 10 fractions to the anterior oral cavity/ oropharynx /parotid region. So a 20 Gy difference to a terribly sensitive / toxicity related region to deliver treatment. And this is where there is a real potential difference to see large difference. While the low neck and posterior neck dose may contribute some toxicity, the differences (if any) will be due to the additional dose to the anterior oral cavity region (ant to parotids and from the larynx superiorly)
But at least until more recently, there was/is a downside that is seen in the image above but not often discussed. In the image above, you can see that the high dose volume is actually larger on the passive scattering / uniform scanning proton plan (example is technically a USPT). That difference, where the high dose extends near the surface of the patient will offset some of the integral dose gains making difference to between IMRT and older proton plans.
Within pencil beam, we still tend to be more conservative (my words) with the treatment volume due to range uncertainty and physics concerns but the high dose volumes are much more similar along with a similar massive reduction in integral dose delivered to the patient. This data is taken specifically from my head and neck work referenced below (REF 1) but it is well know that integral dose reductions are around 50% for many cancers treated with proton therapy compared to IMRT (REF 2). That is the common fact that people might know, but no one understands that it would equal 35 extra 2Gy treatments via a 10x10 to dmax delivered into the head and neck region. That difference alone would, in almost every situation, be enough for physician adoption of a new standard of care as no one would willingly deliver that much additional dose to anyone (but equipment price and access argue against this path).
The transition from passive scanning approaches to pencil beam is important.
In both the lung trial (100% of patients) and esophageal trial (80% of patients), the nearly all patients were treated with passive scatter approaches. In the simplest terms, passive scatter allows the shaping of either the distal or proximal beam edge to tumor (almost always distal end) whereas pencil beam allows for the shaping of both the distal and proximal beam edge to the tumor.
It is pretty easy to visually see above in the comparisons where the red volume is actually larger on the proton plan then the IMRT plan. I think that has to contribute to clear lack of a large win in either lung or esophagus cancer (both are further complicated by tissue / air interfaces - but again, we’ll stay on the main points).
Retrospective data remains robust in support of proton therapy.
Here is a publication from November of 2022 and something outside of MDACC. (I trained at MDACC so I tend to be data heavy from there but MSKCC has published strong retrospective outcomes as well)
Memorial Sloan Kettering Head and Neck Proton Data:
Published November 11, 2022. (REF 3)
The trial is a retrospective trial as we wait for randomized data, but consistent with a large volume out of MD Anderson, the MSKCC has published their data showing a significant reduction in toxicity with no LRR differences.
LRR 5% vs 4%
Grade 2 pain - 72% vs 93% p<0.001
Grade 2 dysgeusia 28% vs. 57% p<0.001
Grade 3 mucositis 53% vs 70% p=0.003
Grade 2 nausea 0% vs 18% p=0.04
Grade 2 wt loss 37% vs 59% p<0.001
Chronic toxicity was similar however with only xerostomia being worse in the IMRT arm.
Below is the discussion from the abstract. I think it is very well written so I have included it here in its entirety.
What this trial shows, as well as the MDACC data and essentially every other published toxicity paper, is that proton therapy in the treatment of head and neck cancer reduces toxicity. Significantly. But perhaps more subtlety if you look a bit closer.
I think this trial is interesting to note and come back to upon publication of the randomized trial. Here we see many Gr 2 toxicities fade to similar outcomes at 90 days and in a way , that is not that impressive. The p values are impressive and toxicities appear to go stepwise down but not from terrible to zero. This study, to me, fits in very well with the prospective Randomized Esophageal data (REF 4) - across a number of measured toxicities, you see a difference in outcomes (mainly acute toxicities). On whole and at a later time period, differences (OS, PFS, Late toxicities differences) have been more difficult to prove.
Changes in toxicity due to technique differences can be measured in head and neck cancer.
Further strengthening the argument for opportunity is that we KNOW that dose differences impact clinical outcomes. The ONLY randomized prospective OAR (organ at risk -fancy verbiage for normal tissue) data that exists was published back in May of 2020 (REF 5). It is the DARS trial looking at reducing dose to the pharyngeal constrictors. In the experimental arm the constrictor are prioritized OVER covering of the low dose neck volumes.
Note: It is a specific technique and not just circling the constrictors and optimizing them at some point. In my experience, it creates pretty consistent plans that have a certain look / feel and my guess is, that level of consistency helped in demonstrating a benefit to the approach, but back to the trial.
In the end, there was a significant improvement in outcomes. Reduced dose resulted in significantly higher MDADI (MD Anderson Dysphagia Inventory) scores at 77.7 vs 70.3 (p=0.016) in favor of the experimental / lower dose to the swallowing muscle arm of the trial.
This is important. If we believe this trial on any real level, it somewhat removes the opportunity for protons to fail the randomized trials - if they are better, the improvement should be measurable.
Standards Evolve in Medicine
Treatment with IMRT is not some defined “standard” - it evolves over time as does any medical treatment. And it far better today than 15 yrs ago. Therein lies a weakness of trying to define a piece of clear winning evidence moving forward - the bar to which one measures is always moving and big trials take years. Protons are fighting an uphill battle with respect to this aspect of medicine. IMRT has at least 50x the install and the patient treatment base of protons. It is constantly iterating and moving forward. Proton therapy has to win big enough to remove this benefit of scale and faster iteration. That has proven to be a difficult bar to surpass.
The two most clear examples in head and neck management related to this topic are 1) Do-IMRT (dysphagia optimized IMRT - proven beneficial above) and, as referenced in MSKCC discussion, 2) a continued push to attempt to reduce neck volumes and doses to minimize toxicity. So we continue to push for less toxicity within the IMRT and are making slow but steady stepwise progress.
But on the side of protons in this space is the following:
70 Gy in 35 fractions along with cisplatin remains a standard for a large subset of head and neck cancer patients.
And it seems pretty likely that this standard will remain in place for a number of years. So while IMRT treatment for head and neck cancers continues to evolve, it is doing so in relatively measured steps. It is not rapidly moving to hypofractionation or SBRT. The chemotherapy does not currently seem to be leaving for many patients. There is de-escalation work for primarily HPV+ cases, but even there it has been slow progress (REF 6).
So for the moment, I think there is great opportunity in this segment of the patient population for a jump forward and for that new approach with protons to have a significant time period to be “the standard”, but it has to win.
Protons still outperform - at least on paper.
Very shortly after release of the DARS trial data, I wanted to look at patients I had treated to see if USPT and or PBS would still outperform the most modern IMRT plans. This new planning method created a different version of IMRT treatment (I’ve done my own planning optimization for 20+ yrs - and it is truly different if you follow the rules), and I wanted to see if protons still held up as well. It was submitted a couple of times so close to “peer reviewed” but it isn’t (perhaps a different post on a different day).
The full article is linked here (REF 7).
In simple terms, even the older style passive scanning / USPT still outperformed IMRT and pretty handily based on the dosimetry and work I did. Pencil beam planning is a small step better than USPT as one would expect.
The point being, even against current iterations of IMRT that represent proven improvements to that treatment approach, proton therapy appears to maintain a significant advantage - dosimetrically.
Caveat being: patient selection especially in passive scanning environments.
The table is set.
So…
We see that protons deliver less dose.
The retrospective studies appear to demonstrate very clear cut with large - apparently repeatable - decreases in acute toxicity.
Paired with prospective data showing that changes in dose to normal structures affects outcome and those effects ARE measurable in HN cancer
A relatively long runway where, for many of the patients, the standard today will be the standard for some significant period as we move forward.
In the upcoming weeks, we’ll look at the impact of what the landscape of head and neck radiation in the US might react to both negative and positive trials. Protons have large hurdles, some of which are arguably larger than data. But the next step for the industry is clearly that protons need a “big win.”
REFRENCES:
Dysphagia-Optimized Unilateral Proton Radiation Therapy: A Comparative Study Evaluating Constrictor Muscle Dosimetry
Promise and Pitfalls of Heavy-Particle Therapy, Mitin, Zietman
https://ascopubs.org/doi/10.1200/JCO.2014.55.1945Toxicity Profiles and Survival Outcomes Among Patients With Nonmetastatic Oropharyngeal Carcinoma Treated With Intensity-Modulated Proton Therapy vs Intensity-Modulated Radiation Therapy
https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2798320Esophageal Cancer Randomized Trial:
https://ascopubs.org/doi/10.1200/JCO.19.02503Nutting C, Rooney K, Foran B, et al on behalf of the DARS Investigators. Results of a randomized phase III study of dysphagia-optimized intensity modulated radiotherapy (Do-IMRT) versus standard IMRT (S-IMRT) in head and neck cancer. J Clin Oncol May 20, 2020; 38(15_suppl) 6508-6508.
What is the future of treatment de-escalation for HPV-positive oropharyngeal cancer? A review of ongoing clinical trials
https://www.frontiersin.org/articles/10.3389/fonc.2022.1067321/fullDysphagia-Optimized Unilateral Proton Radiation Therapy: A Comparative Study Evaluating Constrictor Muscle Dosimetry. Mark R Storey MD, Suresh Rana MS. Linked above in article.