With little data from randomized controlled trials demonstrating better efficacy or lower adverse effects, medical centers continue to build and open new proton beam radiation therapy (PBRT) facilities. At costs between $100 million and $225 million each, more research into the relative efficacy and safety of this modality compared with current photon (x-ray) therapies is needed and is planned. It is somewhat of a “chicken and egg” situation: How many more proton beam facilities are needed and justified before the data are fully in, to generate the data on their usefulness? At present, seven facilities are operating in the United States with four more planned or under construction.
In a report in Annals of Internal Medicine last fall (Terasawa T et al: 2009;151:556–565), researchers presented their systematic review of 243 studies of any design that described clinical outcomes or adverse effects in 10 or more patients treated with charged particle radiation therapy. Of the 243 studies, only eight were randomized, and nine were nonrandomized trials, comparing treatments with or without charged particles. The majority (185) were single-group retrospective studies. Particle therapy was delivered either alone or in combination with other interventions for common or uncommon types of cancers.
“No comparative study reported statistically significant or important differences in overall or cancer-specific survival or in total serious adverse events,” the authors concluded, noting that many of the studies were not independent and included overlapping patient populations.
In an e-mailed response to questions for this article, first author Teruhiko Terasawa, MD (now a clinician in Japan) and senior author Thomas Trikalinos, MD, PhD, of Tufts Medical Center in Boston said that as of December 2007, they were aware of at least 61,800 patients who had received particle beam radiotherapy around the world for various cancers and other diseases. The vast majority (approximately 54,000 or 87%) received protons, and the rest received helium or carbon ions.
The authors concluded that more research is needed on the comparative effectiveness and safety of charged-particle radiation therapy in cancer to assess the benefits, risks, and costs of this modality and treatment alternatives. The topic is also a concern to the United States Agency for Healthcare Research and Quality (AHRQ), which contracted the Tufts Evidence-based Practice Center to prepare a Technical Brief on the subject. The Annals paper was based on that brief, although the authors noted that the opinions are their own and not those of the AHRQ.
Funding of these very expensive facilities has come from various sources. The one at the University of Texas M. D. Anderson Cancer Center was funded by an investment bank, a developer of health care facilities, public employee pension systems, and manufacturers of radiation therapy and medical imaging systems.
MDACC secured a philanthropic foundation grant to educate clinicians, patients, and the public about proton therapy and subsidize transportation and housing costs for patients who could not otherwise travel to Houston.
The newly opened Roberts Proton Therapy Center of the University of Pennsylvania was built using revenues from Penn Medicine, a large regional patient care, education, and research system.
How Proton Beam RT works
Conventional x-ray or gamma ray therapy delivers electromagnetic ionizing radiation (mass-less photons) to the target site, depositing most of its energy at the skin surface and in normal tissues going in, hitting the target, and still affecting normal tissues when coming out past the target. Clinicians often try to minimize the effects on healthy tissue by varying the delivery paths, shaping the beam, or modulating the beam intensity.
Proton beam radiation therapy (PBRT) uses heavier charged particles (hydrogen nuclei) to irradiate a tumor or other target tissue. These particles release all their energy at a specific depth depending on their speed, which can be adjusted. Jim Metz, MD, Vice Chair for Clinical Operations in the Department of Radiation Oncology at the University of Pennsylvania, explained that the particles pass through intervening tissue without depositing much of their energy, and very little of it remains to affect tissue past the target (Figure), raising the potential for dose escalation.
The high costs of proton beam facilities include a cyclotron to generate and accelerate the protons, treatment rooms involving very large gantries for targeting the beams, treatment planning computers, and imaging equipment, as well as substantial real estate to house the facility and beamlines.
Using the Roberts Proton Therapy Center as an example, the cyclotron weighs 220 tons—around the weight of a fully loaded Boeing 747 jetliner; each of four gantries is three stories high and weighs 90 tons, and the facility occupies 75,000 square feet.
The beamline runs about the length of a football field. Along the beamline, powerful magnets direct the proton beam into the treatment rooms, where a multi-leaf collimator shapes the beam and tunes its energy according to the depth and shape of the tumor target.
Up until now, radiation has been dosed based on how much normal tissues will tolerate. Now, though, with PBRT, explains Stephen Hahn, MD, Professor and Chair of the Department of Radiation Oncology at Penn, it should be possible to give a higher dose to the tumor with less dose to the normal tissue; because of that targeting, increasing the chance to make it more effective.”
Examples of eligible tumors he cited are prostate, breast, bladder, lung, spine, brain and base of the skull, head and neck, and gastrointestinal such as esophageal, pancreas, and liver.
Because of more precise delivery of the ionizing radiation, PBRT may show some advantage in targeting tumors adjacent to critical structures, such as in the eye or the lungs. Dr. Metz said a high rate of local-regional failure occurs in cases of lung cancer, since doses of x-rays are limited by surrounding structures, and PBRT may therefore be indicated.
In a study at MDACC last year (Cox J et al: ASTRO IMRT and Proton Symposia, 1/09), lung cancer was treated with 3D conformal radiation, intensity-modulated radiation therapy (IMRT), or PBRT. The rates of Grade 3 or higher esophagitis were 16%, 40%, and 6%, respectively. Grade 3 or higher treatment-related pneumonitis rates were 32%, 9%, and 0%, respectively. No survival data were reported.
Randomized Trials Really Necessary?
In the approximately 20 years since the first hospital-based proton therapy center in the United States was opened at Loma Linda University Medical Center in California in 1990, several more have come on line in the US and around the world, but any clinical superiority for proton beam radiation therapy over other modern radiation techniques is yet to be proven. “Appropriate clinical utilization is controversial,” Dr. Terasawa and his colleagues wrote.
Researchers have published many studies since the first charged-particle therapy became clinically available in 1954. But Dr. Terasawa and his coauthors found that typically the studies did not provide detailed information on patients, cancer staging, or clinical context. Most of the studies reported overall survival and/or surrogate outcomes of survival, about half reported cancer-specific survival, and some included quality of life outcomes (see table on page 36). Because of the various designs and outcomes measured, comparisons across studies would be difficult or impossible.
From their review of the 243 publications, these authors concluded, “Although randomized evidence is lacking, nonrandomized comparative studies in general failed to demonstrate a survival advantage of charged-particle radiation therapy over conventional radiation therapy.”
Drs. Terasawa and Trikalinos told OT, “We think that there are specific types of research that should be performed—research that compares the effectiveness and safety of proton radiation therapy with contemporary alternatives, especially for common cancers.”
Determining the value of PBRT would require a decision analysis, first specifying a “decisional context” that asks: which patients, compared with what alternative, from whose perspective (patient, payer, physician, society), and for what time horizon?
Some authors have argued that further comparative effectiveness and safety studies are not needed. The rationale is that charged-particle therapy delivers superior dose distributions compared with photon beams, biological effects are already known, and that it should be self- evident that it is a good idea to spare normal tissues from radiation.
“However,” as Dr. Terasawa and coauthors noted, “this line of reasoning equates precision in radiation therapy delivery with clinical outcomes.”
Financial Aspect in Motivation for Clinical Trials?
One set of authors from the Department of Radiation Oncology at Massachusetts General Hospital and Harvard Medical School (Suit H et al: Radiother Oncol 2008; 86:148-153) even predicted: “Were proton therapy less expensive than x-ray therapy, there would be no interest in conducting Phase III trials.”
Based on similar rationale, Michael Goiten and James Cox of the Departments of Radiation Oncology at Harvard and M. D. Anderson (JCO 2008:26:175-176) argue that from what is already known, there can no longer be equipoise between the arms of randomized controlled comparative trials and therefore, it would be unethical to conduct them.
Interestingly, these authors, too, apparently see a financial aspect in the motivation to do the trials: “Can anyone seriously believe that, if protons were cheaper than x-rays, there would be similar objections raised as to their immediate and widespread use?” they asked. “We find it totally unacceptable to insist on what we judge to be unethical RCTs purely to establish the financial cost-effectiveness of an admittedly better technology—nor would patients, if fully informed, consent to participate in such studies.”
Drs. Goiten and Cox posited that as more PBRT facilities are built, their costs will come down, as will the cost of treatment. They even suggest that once PBRT becomes more widely available, “is not the burden of proof on conventional x-ray therapy?, that the cost savings achieved by using x-rays are not accompanied by undesirable additional morbidity?”
But at this point, for many observers, questions still remain about the relative effectiveness, safety, and the most appropriate application of PBRT. For example, in a report from a National Cancer Institute workshop, (Vikram B et al: Oncology. 23(4): 1–8), the authors cited two papers by Shipley and co-workers (Shipley WU et al: Int J Radiat Oncol Biol Phys 1995;32:3–12; Gardner BG et al: J Urol 2002;167:123–126) demonstrating in a prospective randomized trial that there was no survival advantage for patients with advanced prostate cancer by adding proton radiotherapy to conventional photon therapy vs photons alone.
However, the combined radiation therapy group experienced more rectal bleeding (32% vs 12%), with a trend toward more urethral strictures (19% vs 8%). “In this study, the addition of an advanced technology unexpectedly resulted in an inferior result,” the workshop report concluded.
Terasawa et al raised a similar point, reminding readers that despite a strong pathophysiologic rationale for superior effectiveness, randomized clinical trials of antiarrhythmic drugs for premature ventricular contractions and of erythropoietin for the anemia of chronic kidney disease showed harm—results that were counterintuitive and not expected.
Similarly, trials of high-dose chemotherapy with stem cell transplant to treat advanced breast cancer showed no benefit.
In another systematic review of publications on proton therapy to treat a variety of cancers, Michael Brada and colleagues last summer (The Cancer Journal 2009;15(4): 319–324) reached conclusions similar to the Terasawa group, that the literature is “devoid of any clinical data demonstrating benefit in terms of survival, tumor control, or toxicity in comparison with best conventional treatment for any of the tumors so far treated including skull base and ocular tumors, prostate cancer, and childhood malignancies.”
Those authors recommended, “The future use of protons should be guided by clear evidence of benefit demonstrated in well-designed prospective studies…[and] should not be employed on the basis of belief alone and requires testing to avoid inappropriate use of potential detriment to future patients.”
Dr. Hahn agrees: Whatever the theoretical advantages of proton therapy, such as sparing normal tissues, the key question, he says, is “Does that translate into a clinical benefit for patients that you can measure?”
Specific research questions, he said, are:
- Which cancer patients are most likely to benefit from PBRT?
- Can radiation doses using protons be escalated to cure more patients?
- Can side effects be reduced?
- Will proton therapy produce a quality-of-life benefit?
- Can the course of treatment be shortened. and/or can protons be combined with other cancer treatments?
- Does PBRT provide a health economic benefit compared with other therapies?
- Can PBRT be further improved?
Given that proton beam radiation therapy will most often be used in conjunction with other cancer therapies, it may be difficult to ascribe a clinical outcome to a specific component of the overall treatment strategy. “What matters in the end, however, is whether a given patient-management strategy is more effective or safer that an alternative strategy,” Drs. Terasawa and Trikalinos said.
More Being Planned
Dr. Hahn said that he knows of as many as 25 trials of PBRT that are planned, and that he expects that more proton centers will be built before all the data are in, with new technologies coming on board even as data on older ones are coming in.
Both randomized clinical trials and registries can help sort out what treatment strategies are best and for whom. Penn's Roberts Proton Center participates in a group to perform a prospective registry of all patients treated with protons.
A registry can also monitor how the technology evolves as it becomes part of the mainstream of cancer therapy, provide utilization data, and evaluate cancer control and treatment toxicities.
But Dr. Hahn says that to control for patient selection and other potential sources of bias, head-to-head randomized trials are necessary. Now being planned at his center is a randomized controlled trial of IMRT vs protons for prostate cancer patients. The primary endpoint will be rectal symptoms, including proctitis, and the researchers will also look at bladder symptoms and erectile dysfunction.
Coming Next Time:
More to come in the next part of this series: Real-time PET scans, cost effectiveness, and medicine vs marketing.