As members of Congress grapple over ways to reduce the U.S. budget, the publication on December 27 in the American Journal of Clinical Oncology of a complex model analyzing the costs of radiation therapy for prostate cancer seems particularly timely.
The model simulates the clinical history and treatment of a 65-year-old man diagnosed with low-risk prostate cancer. Based on Medicare reimbursement, it shows that stereotactic body radiation therapy (SBRT) costs between one and two times less than intensity-modulated radiation therapy (IMRT). On average, IMRT costs $27,564, whereas robotic SBRT costs $8,289 less and nonrobotic SBRT costs $17,455 less per patient.
Lead author Dr. David Sher, assistant professor of radiation oncology at Rush University Medical Center, and co-authors designed a Markov model to simulate outcomes of a cohort of hypothetical patients after radiotherapy. For the analysis, patients were initially considered healthy with no evidence of disease after receiving treatment. Patients could experience recurrence and/or distant metastasis, potentially leading to death (death could also result from other causes). All toxicities were generally assumed to occur within five years of therapy (Am J Clin Oncol, December 27, 2012).
The hypothetical patients received either IMRT to 75.6 Gy in 1.8-Gy daily fractions or five fractions of SBRT. Because short hypofractionated treatment can lead to an increased rate of late morbidity, the authors conducted sensitivity analyses to see if an increased risk of late toxicity would affect the cost-effectiveness of SBRT. The model included costs for managing proctopathy, bladder toxicity, and impotence, as well as treatments for cancer recurrence. However, the costs of routine surveillance visits were not included because these would be incurred regardless of the type of radiotherapy.
The quality-adjusted life expectancy after IMRT was slightly higher than after SBRT because worse toxicity was assumed with the latter treatment. However, the incremental cost-effectiveness ratios for IMRT over robotic SBRT and nonrobotic SBRT were $285,000 and $591,000, respectively, per quality-adjusted life year. When sensitivity analyses were conducted, the model showed that SBRT was almost always the most cost-effective therapy.
Clinical trial results comparing the effectiveness of IMRT and SBRT are still ongoing, the authors noted. However, with approximately 200,000 men being diagnosed with prostate cancer each year -- most in the early stages of the disease -- Sher and colleagues recommended that SBRT be studied even more aggressively, particularly regarding late toxicities and their current treatment costs.
The premium paid for robotic SBRT
The model also questions whether the U.S. Centers for Medicare and Medicaid Services (CMS) should pay a premium when robotic SBRT is used. Medicare is the only payor that distinguishes between robotic and nonrobotic SBRT, according to the authors. The two treatments are differentiated only by the technical capabilities of the treatment couch.
The authors suggest that the nearly 100% increase in average payment when a robotic SBRT system is used "is somewhat arbitrary." They were unable to find data on improved efficacy or increased marginal cost, noting that the presumption that more precise couch movement allows for smaller patient target margins and less toxicity has not been proved.
"Billing for a 'robotic system' does not imply an improved granularity in patient setup, only that the treatment table was moved outside of the treatment room," they wrote. Even the American Society for Radiation Oncology's (ASTRO) definition of robotic radiosurgery and radiotherapy does not include any specifications of how precisely the movements of a treatment couch can be controlled.
These factors question the efficacy of robotic couchtops, the authors concluded. They recommended that the efficacy of robotic couchtops and their increased cost be aggressively studied.