Reconsidering Adjuvant Versus Salvage Radiation Therapy for ...

28 downloads 6896 Views 52KB Size Report
salvage radiation therapy (SRT) for prostate cancer (PC) patients with adverse .... Results: Data were used to generate tables incorporating the differential.
Poster Viewing Abstracts S585

Volume 90  Number 1S  Supplement 2014

2951 Reconsidering Adjuvant Versus Salvage Radiation Therapy for Prostate Cancer Using an Individualized Decision Analysis With Genomic Classifier Scores D. Trifiletti,1 J. Mason,2 R. Den,3 A. Dicker,3 C. Buerki,4 E. Davicioni,4 A. Crisan,4 and T. Showalter1; 1University of Virginia Department of Radiation Oncology, Charlottesville, VA, 2University of Virginia Department of Public Health Sciences, Charlottesville, VA, 3Thomas Jefferson University Department of Radiation Oncology, Philadelphia, PA, 4 GenomeDx Biosciences Inc., Vancouver, BC, Canada

University in St. Louis, St. Louis, MO, 6Massachusetts General Hospital, Boston, MA Purpose/Objective(s): Proton therapy has been endorsed as a radiation therapy (RT) modality with the potential to avert RT associated comorbidities. Currently, relatively few proton centers exist, and the vast majority of countries do not have a proton center. No clear evidence-based data yet exist to adequately inform allotment of care. Childhood cancer survivors of central nervous system (CNS) malignancies have high prevalences of growth hormone deficiency (GHD) after hypothalamic exposure; further, GH replacement is notable for its high cost, greater than $10,000 per year. We propose a methodology to help guide proton referral of pediatric patients with CNS tumors through cost-effective analysis comparisons for estimated proton and photon hypothalamic dose. Materials/Methods: A Markov cohort model was designed to assess the expected costs and effectiveness for specific dosages of radiation to the hypothalamus. Patients entered the model receiving proton or photon RT for CNS tumors at four or twelve years of age, and were followed for 60 more years of life. Costs were measured in USD and captured cost of GHD and cost of RT at treatment. Cost of proton RT was estimated at $160K more than photon RT. The risk of GHD was based on data by Merchant, et al. Effectiveness was measured in quality-adjusted life years (QALYs). The main outcome measure used for comparison was the incremental costeffectiveness ratio (ICER). We assumed a societal willingness-to-pay threshold of $50,000/QALY. Results: Data were used to generate tables incorporating the differential cost of proton RT to project ICERs for different combinations of hypothalamic radiation dose as displayed in the table below. One-way sensitivity analyses were performed to test the robustness of the assumptions. Proton RT was cost-effective for some scenarios based on the difference in hypothalamic sparing. While some scenarios were not cost-effective, others demonstrated proton RT to be cost-saving when compared to photon therapy. Conclusions: These results provide the first evidence-based guide for identifying children with CNS malignancies who may benefit the most from proton therapy with respect to endocrine dysfunction. Our results support the hypothesis that proton therapy may be more cost-effective for scenarios in which radiation dose to the hypothalamus can be spared, but they also show that proton therapy may not be cost-effective when tumors are involving or in close proximity to the hypothalamus and proton plans also deliver high dose to the hypothalamus.

Purpose/Objective(s): There is controversy regarding the merits of adjuvant radiation therapy (ART) versus close observation with selective salvage radiation therapy (SRT) for prostate cancer (PC) patients with adverse pathologic features after radical prostatectomy (RP). A recent decision analysis model suggests SRT is preferred over ART; however, the model was based upon group-level average inputs. Our objective is to evaluate the comparative effectiveness of ART versus SRT using an analytical method that incorporates individualized estimates of cancer progression. Materials/Methods: We developed a Markov state transition model to estimate life expectancy, in life years (LYs), and quality-adjusted life expectancy, in quality-adjusted life years (QALYs), for a cohort of men with PC who previously received RP. The model design included assignment to one of two treatment options: either ART or close observation with selective SRT. Sources for the inputs were identified based upon the published literature. Individual subjects enter the model at the time of prostatectomy, and exit at the time of death or the end of a 10-year time horizon. Cohort simulation experiments were conducted for a cohort of patients. After developing the model with group-level inputs, we used the genomic classifier (GC) test from subjects in the cohort to assign individualized probabilities for development of distant metastases, with associated model estimates for biochemical recurrence and PC-related death probabilities. Next, the model was adapted to reflect GC riskdependent treatment decision probabilities based upon the DECIDE study. This scenario, with clinical decisions based upon GC scores, was compared to usual-care treatment decisions. Results: The use of GC scores to guide treatment decisions was associated with increase in ART utilization from 14% (usual care) to 48%. SRT was associated with more LYs than ART under usual care (9.58 vs 9.41 LYs) and GC-based decision (9.74 vs 9.37 LYs) scenarios. SRT was associated with more QALYs than ART in the GC-based decision setting (7.64 vs 6.98 QALYs), but fewer QALYs than ART in the usual care scenario (6.89 vs 6.97 QALYs). QALYs were higher in the GC-based decision simulations than with usual care. Findings were robust to sensitivity analyses. Conclusions: The use of GC risk estimates to inform individual treatment decisions after RP can influence projected outcomes after ART or SRT among PC patients with adverse pathological features. GC-defined heterogeneity among patients should be considered when comparing ART and SRT approaches. Author Disclosure: D. Trifiletti: None. J. Mason: E. Research Grant; GenomeDx Biosciences, Inc. R. Den: E. Research Grant; GenomeDx Biosciences, Inc. A. Dicker: E. Research Grant; GenomeDx Biosciences, Inc. C. Buerki: A. Employee; GenomeDx Biosciences, Inc. E. Davicioni: S. Leadership; GenomeDx Biosciences, Inc. A. Crisan: A. Employee; GenomeDx Biosciences, Inc. T. Showalter: E. Research Grant; GenomeDx Biosciences, Inc., Prostate Cancer Foundation Young Investigator Award, ASTRO Comparative Effectiveness Research Award.

Author Disclosure: R. Mailhot Vega: None. J. Kim: None. A. Hollander: None. J.A. Hattangadi: None. J. Michalski: None. N.J. Tarbell: None. T. Yock: None. M. Bussiere: None. S. MacDonald: None.

2952

2953

Cost-Effectiveness Analysis of Proton Versus Photon Therapy With Respect to Risk of Growth Hormone Deficiency R. Mailhot Vega,1 J. Kim,2 A. Hollander,3 J.A. Hattangadi,4 J. Michalski,5 N.J. Tarbell,6 T. Yock,6 M. Bussiere,6 and S. MacDonald6; 1Mount Auburn Hospital, Cambridge, MA, 2Harvard University School of Public Health, Boston, MA, 3Washington University in St. Louis, St. Louis, MO, 4 University of California San Diego, San Diego, CA, 5Washington

Cost-Effectiveness Analysis of Stereotactic Body Radiation Therapy for Pulmonary Oligometastases N.H. Lester-Coll, R.H. Decker, and J.B. Yu; Department of Therapeutic Radiology, Yale University School of Medicine, New Haven, CT

Scientific Abstract 2952; Table Projected ICERs ($/QALY) of proton vs photon dose to the hypothalamus for a 12yo patient Proton Dose (Gy) Photon Dose (Gy) 5 10 15 20 25 30 35 40

0

5

10

15

20

94095 26340 cost-saving cost-saving cost-saving cost-saving cost-saving cost-saving

305368 87424 18933 cost-saving cost-saving cost-saving cost-saving

253336 71556 12653 cost-saving cost-saving cost-saving

240666 68674 21253 cost-saving cost-saving

244745 95348 42912 20899

Purpose/Objective(s): Metastasectomy has historically been used for the management of pulmonary metastases, with retrospective data suggesting