Arguing against the Proposition is Cedric X. Yu, D.Sc. Dr. Yu received his M.S. and D.Sc. degrees from Washington University, St. Louis and, after working in industry for three years, moved to William Beaumont Hospital, Royal Oak, MI, as a medical physicist and Assistant Professor at Oakland University. In 1997 he moved to the Department of Radiation Oncology, University of Maryland as Director of Medical Physics and became the endowed Carl M. Mansfield, M.D. Professor. Dr. Yu has served on many task groups and committees of the AAPM, including the Board of Directors, as well as President of the Mid-Atlantic Chapter. His major research interest is conformal radiotherapy such as intensity modulated photon therapy (IMRT) and intensity modulated arc therapy, which he invented in 1995. He holds 20 patents and has published over 100 peer-reviewed papers, and is certified in Radiation Oncology Physics by the ABMP. Dr. Yu is the founder and CEO of Xcision Medical Systems, LLC, and declares that he is a shareholder in the company. We see an increase in multimodality treatments such as combinations of radiation with targeted therapies or immunotherapy. These strategies will benefit from advanced control of the dose to organs at risk, tumor dose escalation, or dose painting. While a prescribed target dose can be often achieved with protons as well as photons, proton therapy will always deliver a factor of 2–7 less overall dose to critical structures, independent of photon delivery or optimization techniques.5 This adds leverage when optimizing multimodality treatment strategies. Most proton therapy patients are still treated with passive scattering. In the next few years, the majority of patients will be treated with proton beam scanning allowing treatment optimization with protons that is far superior to what is achievable with photons. Intensity modulated proton therapy (IMPT, which due to energy modulation offers one additional degree of freedom compared to IMRT) and robust optimization offer unprecedented dose shaping capabilities.6 Proton therapy is expected to catch up with photon therapy as far as in-room imaging is concerned. Even proton-MR systems are being designed. Most importantly though, there are novel imaging techniques for adaptive radiation therapy and dose delivery verification that are unique to proton beams (such as prompt gamma imaging) thus offering proton specific advancements in dose confirmation and image-guidance.7 Compared to these, further improvements in photon therapy are either marginal or can be utilized on the proton side as well. In summary, both photon as well as proton therapy technology will continue to improve but the ceiling for proton therapy is significantly higher. Consequently, in the future we will see an even bigger dosimetric advantage for protons. Although clinical significance has to be shown in clinical trials, it is likely that this will have a profound impact on treatment outcomes. In the two decades since the advent of IMRT, we have seen tremendous improvements in delivery efficiency with rotational IMRT and in targeting accuracy with on-line and on-board image guidance. However, the dose distributions have not shown much improvement in spite of the intense efforts spent on optimization methods. This fact has misled many in the field to think that photon radiotherapy has reached its limits set by the physics of dose deposition and the future of radiotherapy is in protons and heavy ions. In a review paper on IMRT published eight years ago,8 the authors concluded the following: “Based on 10 years of experience with IMRT, we have learned that the opportunities in improving plan quality are limited within the constraint of present linac/MLC delivery.” It is important to note that these conclusions were based on the (then) current designs of the linear accelerator (linac) and multileaf collimator (MLC) and current delivery methods. If a new system design or new treatment method injects new degrees of freedom into plan optimization, better treatment plans can be realized. An example of exploring additional freedom is the 4πRT proposed by Ke Sheng and his colleagues. They have demonstrated that significant improvements in dose distribution can be achieved for liver,9 lung,10 head and neck,11 and prostate12 by extensive use of noncoplanar IMRT fields. Photons should also include gamma rays emitted from radionuclides. The small source sizes allow design of less compromising, site-specific solutions. A prime example is the Gamma Knife for intracranial radiosurgery. With the possibility of focusing hundreds of beams to a single spot, convenient and effective treatments can be delivered. The relative smaller sizes of teletherapy machines and linacs also allow simpler integration with imaging. The MRIdian system developed by ViewRay and the Atlantic system developed by Elekta exemplify the advantages of smaller photon machines. While the dosimetric characteristics of protons have certain advantages over those of photons, we must also recognize their disadvantages. In addition to high cost, some of these disadvantages include broader penumbra, proton beam range uncertainties, dose calculation uncertainties (e.g., CT artifacts), distal-end RBE uncertainties, high dosimetric sensitivity to anatomical changes (e.g., nasal cavity filling), and the limitations in spot size. The argument that proton therapy is still in its infancy and, therefore, it will have more room to advance is both untrue and invalid. Generally, the more complex the technology, the more restricted and difficult it is to advance. For example, intensity modulated proton arcs would be harder, if not impossible, to achieve with the current spot scanning technology. While the available freedom given by the current photon machine design and treatment methods has largely been exhausted, new designs and treatment techniques can inject new freedom and drastically improve the quality of plans. These dosimetry improvements can be achieved without losing the benefits of efficiency and image guidance. It would be shortsighted to think that radiotherapy with photons has reached its limits. Photon therapy has advanced leading to improved dose conformity. But these advances, instead of reducing the integral dose, mainly re-distribute the dose. This certainly has clinical merits. New techniques (e.g. 4π-RT) will increase dose conformity further. Yet, except at depths greater than about 15 cm (e.g., for lateral prostate fields), where scattering of protons causes the penumbra to be worse than with photons, the dose conformity can never reach that of a proton plan (which can be proven mathematically). Clinical significance is another matter. This is not the argument we are trying to settle. This debate is whether the gap between photons and protons is likely to narrow or to widen. Protons offer more potential in improvements compared to photons as outlined in my opening statement. Range and RBE uncertainties as well as sensitivity to anatomical changes are indeed proton-specific obstacles. Range uncertainties are currently reduced by advanced dose calculation and imaging, while RBE uncertainties are currently being assessed experimentally and clinically. This research will increase the current dosimetric gap between protons and photons, not decrease it. Anatomical changes are addressed with on-board imaging and adaptive therapy (for both photon and proton therapy). The lack of intensity modulated proton arcs is not a limitation because the technique is not even necessary for protons given the advanced dose shaping capabilities and small spot sizes. Photon machines are significantly smaller than proton machines (even single-room solutions). Whether efforts to make proton delivery systems more compact will result in machines with the same size as a photon system is debatable. But size is related to cost, not to treatment quality. I find my opponents’ statement that “new designs and treatment techniques can inject new freedom and drastically improve the quality of plans” unconvincing. While I agree that photon radiation therapy can be improved, it does have a lower ceiling than proton therapy. There is a consensus that the utmost goal in radiotherapy is delivering the best possible dose distribution to achieve the highest therapeutic ratio. The disagreement is which radiation type, photon and proton, will have a better long-term prospect to reliably deliver a more conformal dose distribution to the target while providing better sparing of the surrounding structures? Photon radiotherapy will more likely be the long-term winner based on the following facts. Most of the known drawbacks of proton beams as listed in my Opening Statement are rooted in the physics of particle transport in the medium and, therefore, are not easily overcome with technology. These drawbacks, coupled with the complexity and size of proton accelerators, will set the ceiling on the quality of plans and on the reliability in realizing such plans in the patient. There are many advanced treatment delivery and image guidance techniques being practiced today or emerging in practice for photons that will be hard for protons to follow. These include tracking tumor motion, rotational intensity modulation, and MRI imaging during treatment delivery. Prompt gamma imaging is unique to protons, but it merely overcomes one of the dosimetric uncertainties unique to proton beam delivery. Therefore, it should not be viewed as an advantage but rather an additional, required, imaging procedure that may further reduce patient throughput. Although proton beam therapy produces better dose distributions than photon beams today, it would be premature to assume that photon beam radiotherapy will not catch up or even surpass protons in dosimetric conformity. For example, 4π radiotherapy has shown dose distributions that rival IMPT for many sites.10,11 There is no reason that “the ceiling for proton therapy is significantly higher” than photons. It is often true that the more complex and cumbersome the technology, the harder it is to advance. Economic reasons aside, the relative reliability associated with photon beam treatment delivery and the associated treatment efficiency make advancing photon beam radiotherapy a better investment.