|
|
||
 |
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 25, No 24 (August 20), 2007: pp. 3565-3566 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JCO.2007.11.9768
The Titanic and the Iceberg: Prostate Proton Therapy and Health Care EconomicsDepartment of Radiation Oncology, Massachusetts General Hospital, Boston, MA Almost 100 years ago, the Titanic and the iceberg each grew in slow splendor on opposite sides of the wide Atlantic. One, fashioned in the shipyards of Belfast, represented the quintessence of high technology and high finance. The other, patient and brooding in the cold northern waters, was the stern representative of immutable nature. None could guess their ultimate destiny. When the unpredictable yet completely avoidable collision came, it was the hubris of technology that fared worse. At the beginning of this century we, too, are about to witness a collision of two colossi; this time, although quite predictable, it seems unavoidable. The expansive vigor of medical innovation is heading inexorably toward the harsh reality of economic fact. The controversial treatment of prostate cancer epitomizes this clash and represents one sharp point of impact along its very broad front. Sailing forward, powered by the winds of advocacy, of market forces, and of high-stakes investment, is proton therapy: the proud vanguard of modern technology. Elsewhere, waiting patiently in the darkness, are the hard, cold, and unyielding laws of economics. The collision is yet to come, but, with many new proton centers breaking ground, it cannot be far away. Konski et al1 have made an illuminating and timely contribution to this debate by assessing the potential costs and benefits of proton therapy in contemporary prostate cancer treatment. I say timely because the business model employed by the emerging US proton therapy centers is prostate cancer, and it is on this disease that they will either sink or sail (Note: The author treats prostate cancer with both IMRT and proton beam and the Massachusetts General Hospital and the Francis H. Burr Proton Therapy Center, Boston, MA.). Proton beam therapy (PBT) has been in use for many decades; although its theoretical benefits are not in doubt, there is remarkably little published data strongly supporting its use over other forms of radiation, with the exception of ocular and skull-base tumors and malignancies of children. However, these cancers are rare and the patient demand would be insufficient for new proton therapy centers, or more recently the vendors, to recover their investment. From a commercial perspective, PBT needs to be used for the high-volume treatment of common cancers, and prostate cancer is a perfect fit. PBT has been used in prostate cancer for many decades and there is certainly a growing body of evidence confirming clinical efficacy, but as yet, apart from some comparative planning studies, there is no proof that it is superior to its alternatives. Ten or 15 years ago when the only widely used radiation alternative to proton beam in prostate cancer was two-dimensional photon therapy, the superiority of protons could be assumed, given that dose escalation was otherwise simply impossible. Since then, the landscape has changed; three-dimensional therapy and intensity-modulated radiation therapy (IMRT) have narrowed the gap to a degree that truly challenges the underlying premise. Undoubtedly, theoretical advantages still exist. The integral dose to the pelvis with proton beam is lower than that with IMRT, but lateral, the dose through the hips is substantially higher. What then matters more, a low dose to the pelvis or a higher dose to the hips—more rectal or bladder cancers or more hip fractures? If there is a meaningful clinical benefit to PBT in prostate cancer, it can come in one of two ways. Either more radiation can be administered to the prostate than with IMRT, thus increasing the chance for cure, or the same dose of radiation can be administered for a lower risk of adverse effects. Konski et al1 have examined the first of these propositions. They take the plentiful randomized data supporting dose escalation, speculate that PBT could be used to deliver 91.8 Gy, and calculate, using published dose-response curves, that this will yield a 10% improvement in 5-year freedom from biochemical failure for men with intermediate-risk prostate cancer. The downstream benefits are a reduction in risk of clinical recurrence and consequent use of costly hormonal and chemotherapy. The probability of this benefit being cost effective approaches 50% if the man lives 15 years after therapy and is even higher if he lives longer. The authors calculate that for younger men with this category of disease, the benefit anticipated may just fall within current standards of economic acceptability ($50,000 per quality-adjusted life-year). A number of issues, however, require clarification. First, intermediate-risk prostate cancer represents only 15% to 20% of the total incidence of this disease. Of these men, a large number are diagnosed at age older than 70 years, and with a more limited life expectancy, have relatively little to gain. A significant proportion of the younger patients are already managed more cost-effectively by either surgery or brachytherapy, which in relative terms seem to be a bargain. Unless it is anticipated that many younger men will switch to PBT, these exclusions do not leave too many potential beneficiaries. Of those who may benefit, the cost-benefit calculation is based on one critical assumption: that it is possible to administer 91.8 Gy safely with proton beam. I know of no institution that has even begun to test this. At the Massachusetts General Hospital (Boston, MA) and Loma Linda University Medical Center (Loma Linda, CA), we have completed a pilot protocol delivering 82 Gy at 2 Gy per fraction, but these results have not been analyzed, let alone published. It is my personal impression that 82 Gy is pushing the envelope, although I may be proven wrong. No matter how accurate the radiation delivery, some volume of rectum, some volume of bladder neck, and all of the prostatic urethra will receive the full dose. At some point one of these normal tissues must become the dose-limiting factor. In an analysis of Massachusetts General Hospital patients treated with 77 Gy a median of 13 years ago, 50% had had at least one episode of hematuria by 10 years.2 No one has yet begun to measure the long-term functional consequences of high-dose radiation on the urinary sphincters of men treated decades earlier. The consequences as men age can only be guessed. If the cancer control benefit of proton beam dose escalation in intermediate-risk disease is slight and speculative, then a different benefit may be possible in another domain not evaluated by Konski et al.1 If the dose delivered by protons were no greater than what we currently feel to be a comfortable IMRT limit, say 79 to 81 Gy, could there be both clinical and economic gains through reduced morbidity? Again, we can only speculate. Reductions in photon beam morbidity have come about from improved targeting with image guidance, better planning, and improved delivery systems. The morbidity has decreased even though doses have gone up. The all-important late, grade 3 and higher rectal and bladder toxicities are now infrequent, and it will be a statistical challenge to detect any additional reduction. In addition, one of the principal patient concerns, the incidence of erectile dysfunction, may actually increase as high technology, be it protons or photons, emboldens the use of higher doses. The neurovascular bundles cannot be spared along their paraprostatic path by any contemporary technique. A recent comparative planning study by Trofimov et al shows clearly that though the low-dose volume is substantially lower with conventional (passively scattered) opposed lateral proton beams than with IMRT, the high-dose volume is actually a little larger because of uncertainties at the end of the beam range and the emergence of a penumbra at depth.3 Lateral overshoot in the region of the neurovascular bundles is therefore necessary to avoid the risk of underdosage. In the future, this potential disadvantage will likely be mitigated by the use of intensity-modulated proton therapy, although at present only one center in the world is using this. If doses with proton beam cannot be escalated over those that can be administered readily with currently available technologies, then the hypothesis that PBT reduces morbidity needs urgent evaluation in a prospective randomized trial with rigorous, validated, quality-of-life end points. Ideally, doses delivered would be identical, although this will be a challenge given that the proton dose distribution is less heterogeneous than with IMRT. The very best of each technology would have to be compared (IMRT v intensity-modulated proton therapy) to avoid a trial that is outdated by the time of reporting. The article by Konski et al1 represents a legitimate and laudable attempt to calculate the cost of proton therapy in prostate cancer based on current costs and the limited published clinical data available. Costs, charges, and reimbursements are notoriously volatile. The equation can be altered in either direction by changes in the price of proton center installation or the reimbursement paid by Medicare. If proton reimbursement were to be reduced at a rate faster than that for IMRT, then Konski et al could rewrite their article with new figures and perhaps a different result. As things currently stand, however, it is likely that IMRT reimbursement will decline first. The authors are at pains to point out that their calculations apply only to patients with intermediate-risk disease. No advantage would be calculated using this model for low-risk disease, in part because control rates are excellent already and in part because the majority of older men probably do not need treatment anyhow. This group of men, more than any, needs proof of superior quality of life before PBT use becomes more widespread. In conclusion, PBT is not an experimental therapy but a legitimate form of external radiation for prostate cancer that has some theoretical advantages. Dose escalation has clear advantages in intermediate-risk disease. If PBT cannot be used to escalate doses further than currently available alternatives, then its economic utility relies entirely on the demonstration of a clear and meaningful difference in subsequent quality of life. This should be an area investigated urgently by those with open minds and no commercial bias. If gains are seen, then we will accept them and work out how to meet the resultant demand. If there is no advantage, unless the costs change dramatically, proton therapy cannot be judged worth the price for localized prostate cancer. Beam time and physics expertise should then be redirected into clinical areas where benefits are more likely. Now is not the time to be rearranging the deck chairs but a time to consider redirecting the ship. If we fail to give this issue our attention, then we may have a 21st-century collision between technology and the iceberg, with the same sorry end. AUTHOR'S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The author(s) indicated no potential conflicts of interest. ACKNOWLEDGMENTS The inspiration for this Editorial was the Thomas Hardy (1912) poem, The Convergence of the Twain (Lines on the loss of the Titanic). REFERENCES
Related Article
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
