Turgay Ayer

OR Forum---A POMDP Approach to Personalize Mammography Screening Decisions

Breast cancer is the most common nonskin cancer and the second leading cause of cancer death in U.S. women. Although mammography is the most effective modality for breast cancer screening, it has several potential risks, including high false-positive rates. Therefore, the balance of benefits and risks, which depend on personal characteristics, is critical in designing a mammography screening schedule. In contrast to prior research and existing guidelines that consider population-based screening recommendations, we propose a personalized mammography screening policy based on the prior screening history and personal risk characteristics of women. We formulate a finite-horizon, partially observable Markov decision process POMDP model for this problem. Our POMDP model incorporates two methods of detection self or screen, age-specific unobservable disease progression, and age-specific mammography test characteristics. We solve this POMDP optimally after setting transition probabilities to values estimated from a validated microsimulation model. Additional published data is used to specify other model inputs such as sensitivity and specificity of test results. Our results show that our proposed personalized screening schedules outperform the existing guidelines with respect to the total expected quality-adjusted life years, while significantly decreasing the number of mammograms and false-positives. We also report the lifetime risk of developing undetected invasive cancer associated with each screening scenario.

First- and Second-Line Bevacizumab in Addition to Chemotherapy for Metastatic Colorectal Cancer: A United States–Based Cost-Effectiveness Analysis

PURPOSE The addition of bevacizumab to fluorouracil-based chemotherapy is a standard of care for previously untreated metastatic colorectal cancer. Continuation of bevacizumab beyond progression is an accepted standard of care based on a 1.4-month increase in median overall survival observed in a randomized trial. No United States-based cost-effectiveness modeling analyses are currently available addressing the use of bevacizumab in metastatic colorectal cancer. Our objective was to determine the cost effectiveness of bevacizumab in the first-line setting and when continued beyond progression from the perspective of US payers. METHODS We developed two Markov models to compare the cost and effectiveness of fluorouracil, leucovorin, and oxaliplatin with or without bevacizumab in the first-line treatment and subsequent fluorouracil, leucovorin, and irinotecan with or without bevacizumab in the second-line treatment of metastatic colorectal cancer. Model robustness was addressed by univariable and probabilistic sensitivity analyses. Health outcomes were measured in life-years and quality-adjusted life-years (QALYs). RESULTS Using bevacizumab in first-line therapy provided an additional 0.10 QALYs (0.14 life-years) at a cost of

Informatics in Radiology: Comparison of Logistic Regression and Artificial Neural Network Models in Breast Cancer Risk Estimation

59,361. The incremental cost-effectiveness ratio was

Hepatitis C Disease Burden in the United States in the era of oral direct-acting antivirals

571,240 per QALY. Continuing bevacizumab beyond progression provided an additional 0.11 QALYs (0.16 life-years) at a cost of

Economic Burden of Chronic Lymphocytic Leukemia in the Era of Oral Targeted Therapies in the United States

39,209. The incremental cost-effectiveness ratio was

Cost-Effectiveness Analysis of Regorafenib for Metastatic Colorectal Cancer

364,083 per QALY. In univariable sensitivity analyses, the variables with the greatest influence on the incremental cost-effectiveness ratio were bevacizumab cost, overall survival, and utility. CONCLUSION Bevacizumab provides minimal incremental benefit at high incremental cost per QALY in both the first- and second-line settings of metastatic colorectal cancer treatment.

Breast cancer risk estimation with artificial neural networks revisited: discrimination and calibration.

Computer models in medical diagnosis are being developed to help physicians differentiate between healthy patients and patients with disease. These models can aid in successful decision making by allowing calculation of disease likelihood on the basis of known patient characteristics and clinical test results. Two of the most frequently used computer models in clinical risk estimation are logistic regression and an artificial neural network. A study was conducted to review and compare these two models, elucidate the advantages and disadvantages of each, and provide criteria for model selection. The two models were used for estimation of breast cancer risk on the basis of mammographic descriptors and demographic risk factors. Although they demonstrated similar performance, the two models have unique characteristics-strengths as well as limitations-that must be considered and may prove complementary in contributing to improved clinical decision making.

Optimal timing of hepatitis C treatment for patients on the liver transplant waiting list

Oral direct‐acting antivirals (DAAs) represent a major advance in hepatitis C virus (HCV) treatment. Along with recent updates in HCV screening policy and expansions in insurance coverage, treatment demand in the United States is changing rapidly. Our objective was to project the characteristics and number of people needing antiviral treatment and HCV‐associated disease burden in the era of oral DAAs. We used a previously developed and validated Hepatitis C Disease Burden Simulation model (HEP‐SIM). HEP‐SIM simulated the actual clinical management of HCV from 2001 onward, which included antiviral treatment with pegylated interferon (Peg‐IFN)‐based therapies as well as the recent oral DAAs, risk‐based and birth‐cohort HCV screening, and the impact of the Affordable Care Act. We also simulated two hypothetical scenarios—no treatment and treatment with Peg‐IFN‐based therapies only. We estimated that in 2010, 2.5 (95% confidence interval [CI], 1.9‐3.1) million noninstitutionalized people were viremic, which dropped to 1.9 (95% CI, 1.4‐2.6) million in 2015, and projected to drop below 1 million by 2020. A total of 1.8 million HCV patients will receive HCV treatment from the launch of oral DAAs in 2014 until 2030. Based on current HCV management practices, it will take 4‐6 years to treat the majority of patients aware of their disease. However, 560,000 patients would still remain unaware by 2020. Even in the oral DAA era, 320,000 patients will die, 157,000 will develop hepatocellular carcinoma, and 203,000 will develop decompensated cirrhosis in the next 35 years. Conclusions: HCV‐associated disease burden will still remain substantial in the era of oral DAAs. Increasing HCV screening and treatment capacity is essential to further decreasing HCV burden in the United States. (Hepatology 2016;64:1442‐1450)

Association of periodic and rhythmic electroencephalographic patterns with seizures in critically ill patients

Purpose Oral targeted therapies represent a significant advance for the treatment of patients with chronic lymphocytic leukemia (CLL); however, their high cost has raised concerns about affordability and the economic impact on society. Our objective was to project the future prevalence and cost burden of CLL in the era of oral targeted therapies in the United States. Methods We developed a simulation model that evaluated the evolving management of CLL from 2011 to 2025: chemoimmunotherapy (CIT) as the standard of care before 2014, oral targeted therapies for patients with del(17p) and relapsed CLL from 2014, and for first-line treatment from 2016 onward. A comparator scenario also was simulated where CIT remained the standard of care throughout. Disease progression and survival parameters for each therapy were based on published clinical trials. Results The number of people living with CLL in the United States is projected to increase from 128,000 in 2011 to 199,000 by 2025 (55% increase) due to improved survival; meanwhile, the annual cost of CLL management will increase from

Necitumumab in Metastatic Squamous Cell Lung Cancer: Establishing a Value-Based Cost

0.74 billion to