Purpose: To evaluate the incremental benefits and costs of adding magnetic resonance (MR) imaging to digital mammography (DM) screening in BRCA carriers.

Method: We used a Markov Monte Carlo model to compare four screening strategies to clinical surveillance (no imaging): 1) annual DM beginning at age 25 [DM25], 2) annual DM beginning at age 30 [DM30], 3) DM/MR beginning at age 25 [DM/MR25], and 4) DM/MR beginning at age 30[DM/MR30]. For combined strategies, we examined DM/MR alternating at 6-month intervals.  An excess relative risk model was used to incorporate radiation risk from DM. The primary outcomes were quality adjusted life years (QALYs), lifetime costs (2010 USD) and incremental cost-effectiveness ratios (ICERs).

Result: Adding MR to DM increased QALYs and costs in both BRCA1 and BRCA2 carriers (Table 1). The DM/MR25 and DM/MR30 strategies were equally effective; DM/MR30 was less costly. Compared to DM30, DM/MR30 resulted in 0.12 and 0.06 additional QALYs at a cost of $117,754 and $114,539 in BRCA1 and BRCA2 carriers, respectively. The ICERs for DM/MR30 vs DM30 were $70,105 (BRCA1) and $209,818 (BRCA2). For BRCA1 carriers, these results were most sensitive to MRI cost, lifetime breast cancer risk, age at prophylactic oophorectomy, and MR test performance.  Varying MR cost in BRCA1 carriers resulted in the widest range of ICER values.  As MR cost increased to $842 (base case: $619), the ICER for DM/MR30 vs. DM30 exceeded $100,000/QALY. As MR cost decreased to $363, the ICER fell below $50,000/QALY.  The results in BRCA2 carriers were stable across the range of parameters examined in sensitivity analysis.

Conclusion: Combined DM/MR screening alternating at six month intervals beginning at age 30 is considerably more cost-effective in BRCA1 carriers than in BRC2 carriers.  

Table 1. Incremental cost-effectiveness of screening in BRCA1 and BRCA2 carriers.
		Clinical Surveillance	DM30	DM25	DM/MR30	DM/MR25
BRCA1	Lifetime costs	$104,490	$109,006	$110,420	$117,754	$122,557
	QALYs (y)	43.96	44.25	44.25	44.37	44.38
	ICER ($/QALY)	-	$15,294	Eliminated	$70,105	$480,300
BRCA2	Lifetime costs	$97,121	$102,204	$103,726	$114,539	$119,678
	QALYs (y)	45.22	45.52	45.51	45.58	45.58
	ICER ($/QALY)	-	$17,078	Eliminated	$209,818	Eliminated

Purpose: Mammography screening has been found to reduce breast cancer mortality, but is also accompanied by harms, such as overdiagnosis. Overdiagnosis refers to the detection of tumors that would not have been detected in a woman’s lifetime in the absence of screening. Estimates of the amount of overdiagnosis vary widely. The aim of the present study is to estimate the amount of overdiagnosis for invasive breast cancer and ductal carcinoma in situ associated with screening women after age 74 years.

Method: The microsimulation model MISCAN-Fadia was used to simulate a cohort of women born in 1960. All women received biennial screening starting at age 50 with varying stopping ages of screening. First, we simulated the screening currently recommended, i.e., biennial screening from age 50 to 74 years, and determined the benefits and harms of the last screen at age 74 years. Then, the additional benefits and harms of adding one screen were estimated with increasing stopping ages. We estimated the number of life years gained, quality-adjusted life years, breast cancer deaths averted, false positives and number of overdiagnosed women for each screening scenario.

Result: The model predicted that screening after age 74 years resulted in benefits in terms of breast cancer deaths averted and life years gained with no upper age limit. The number of quality-adjusted life-years gained increased for screening up to age 90 years. The number of overdiagnosed women increased steeply with increasing upper age of screening. For screening women between age 50 and 74 years 4% of the invasive breast cancers that were detected were overdiagnosed, increasing to 13% for a screen at age 80 years, and 30% for a screen at age 90 years.

Conclusion: Screening women after age 74 years results in a less favorable balance of benefits and harms than screening women between the ages of 50 and 74 years, because of the increasing amount of overdiagnosis at older ages. Decisions on the appropriate upper age depend on individual preferences. Estimates of overdiagnosis are crucial to inform women about the balance of benefits and harms of mammography screening at higher ages.

Purpose: To assess the cost-effectiveness of epidermal growth factor receptor (EGFR) gene mutation testing to guide the selection of gefitinib as first-line therapy in patients with advanced non-small cell lung cancer (NSCLC) in Ontario.

Method: A decision analytic model was developed to conduct this cost-effectiveness analysis from the perspective of the Ontario Ministry of Health and Long-Term Care (MOHLTC). Under EGFR gene mutation testing strategy, tumour tissues from biopsy were assessed for EGFR gene mutation status. Patients with EGFR gene mutation would receive gefitinib as first-line therapy and conventional chemotherapy (platinum based chemotherapy and docetaxel (or pemetrexed)) before best supportive care (BSC). Patients without EGFR gene mutation would receive conventional chemotherapy and BSC. The other patients with undetermined EGFR gene mutation status would receive the same care as the patients under no testing strategy, who would receive conventional chemotherapy, erlotinib, and BSC. Literature review was conducted to estimate the epidemiology and natural history of advanced NSCLC, failure rate of EGFR gene mutation testing, and efficacy of treatments. A regression analysis on utility of patients with advanced NSCLC was applied to estimate utility variables. The estimation of cost variables was based on two Ontario cost studies for advanced NSCLC. Both benefits and costs were discounted at 5% per annum.

Result: Compared to no testing strategy, EGFR gene mutation testing strategy would need $46,021 for one additional life year or $81,071 for one additional quality adjusted life year (QALY). One-way sensitivity analysis indicated that the cost-effectiveness of EGFR gene mutation testing was highly sensitive to the efficacy and cost of gefitinib. Probabilistic sensitivity analysis suggested that the chance for EGFR gene mutation testing to be cost-effective would not be over 50% until willingness-to-pay (WTP) per QLAY increased to $93,340. Budget impact analysis predicted that the adaption of EGFR gene mutation testing would increase the annual direct medical costs by $4.6M, $7.0M, $7.9M, $8.1M, and $8.1M from 2011 to 2015 respectively on the Ontario health care system.

Conclusion: Applying EGFR gene mutation testing to guide the use of gefitinib as first-line therapy for patients with advanced NSCLC would not be considered cost-effective until WTP of MOHLTC was over $81,071 per QALY. The cost-effectiveness of EGFR gene mutation testing was highly sensitive to the efficacy and cost of gefitinib.

Purpose: To assess the cost-effectiveness of early prostate cancer detection with a novel prostate cancer detection index* added to serum prostate-specific antigen (PSA) compared with PSA alone test from a managed care organization perspective.

Method: The prostate cancer detection index is a mathematical formula combining Access Hybritech PSA, free PSA, and a PSA precursor form [-2]proPSA, to predict prostate cancer. It is used as an aid in distinguishing prostate cancer from benign prostatic conditions in men with a PSA test result ≥2 or ≥4 ng/mL and nonsuspicious digital rectal exam. A Markov model was constructed to estimate the expected costs and utilities of prostate cancer detection and consequent treatment under four testing strategies in men aged 50 through 75 years.  The testing strategies varied in test thresholds (PSA ≥2 or ≥4 ng/mL) and methods (PSA alone vs. PSA plus the index) to recommend a prostate biopsy. The transition probabilities were from the electronic medical records analysis for male members in Kaiser Permanente Southern California (KPSC) during 1998-2007.  Health state utilities and prostate cancer treatment costs were derived from the published literature. The model’s cycle length was 1.5 years based on KPSC’s usual practices.

Result: The most cost-effective strategy is to use PSA plus the index at PSA 2-10 ng/mL to estimate the probability of prostate cancer and recommend a biopsy, which has the lowest costs and highest effectiveness [cost/effectiveness (C/E)=13,650/12.416, $1,099/QALY].  Next is PSA plus the index at PSA 4-10 ng/mL [C/E=14,095/12.364, $1,140/QALY), followed by PSA test alone using PSA threshold ≥4 ng/mL [C/E=15,256/12.304, $1,240/QALY), and finally, PSA ≥2 ng/mL [C/E=15,789/12.287, $1,285/QALY).  The strategy of PSA plus the index at PSA 2-10 ng/mL displays a 74% to 86% probability of being cost-effective at a willingness-to-pay range of 0 to $150,000/QALY gained.  Variables including discount rate, starting or stopping age for PSA screening, and health utility of cancer have the most impact on the model.

Conclusion:   From a managed care payer prospective, using the index as an aid to distinguish prostate cancer from benign prostatic conditions at PSA 2-10 ng/mL dominated other strategies, and was optimal in all strategies under the willingness-to-pay of $150,000/QALY. This strategy could be an important method of prostate cancer detection and improving men’s health outcome. ^*Not currently available in the U.S.

Purpose: To coherently integrate multiple sources of available evidence to project lifetime outcomes for newly diagnosed men considering immediate treatment or active surveillance (AS).

Method: Lifetime estimates of time from treatment to progression (T-P) and time from progression to mortality (P-M) were estimated for the 11,347 men diagnosed in 2004-2006 in the SEER cancer registry with low-risk disease (≤ grade 6  and ≤ stage T2a) who were treated immediately with surgery.  Over 38% of all patients diagnosed during these years had low-risk disease. Outcomes under an alternative scenario of active surveillance were estimated for this cohort. Estimates for this scenario integrated a model for diagnosis to delayed treatment (D-T) including parameters for grade and PSA progression, with the same T-P and P-M models.  Estimates of the potential harm of surveillance were based on advanced disease characteristics at time of delayed treatment.   Large cohorts from CaPSURE, Johns Hopkins, and Mayo Clinic were used to inform the models.

Result: With immediate surgery, 26% of low-risk patients will experience biochemical failure and 1.8% will die from prostate cancer. The sampled surgery included only men who were low grade and only 37% had a PSA ≥6. The surveillance scenario resulted in 58% of patients going on to be treated, with 34% upgraded at time of treatment  and 59% having a PSA ≥6. Although disease characteristics were more advanced at the time of delayed treatment, there were no additional deaths due to prostate cancer with surveillance and only a total of 18% of men experienced biochemical failure.  Mean life expectancy for low-risk men treated with surgery between 2004-2006 is projected to be 19.5 years. Under the surveillance scenario, had this cohort selected surveillance they would have experienced only 11.3 treated-person years.

Conclusion:  Although active surveillance is associated with more advanced disease characteristics for some low-risk men who go on to be treated, projections of mortality based on these upgraded disease states did not result in any additional deaths.  Surveillance offers a substantial reduction in the number of treated-person years. Models can help integrate multiple sources of data to help overcome the extremely long time required to observe outcomes in prospective studies between diagnosis and prostate cancer mortality.

Purpose: Quantitative fecal occult blood tests (FOBTs) allow to specify the positivity threshold. We compared colorectal cancer (CRC) screening strategies with quantitative FOBTs, varying the positivity threshold and adapting the screening interval accordingly (longer intervals for better sensitivity).

Method: We used a Markov state-transition model of CRC to calculate life-years and the lifetime number of screening-related tests (FOBTs and follow-up/surveillance colonoscopies; the number of both procedures were combined using their US cost ratio as weighting factor) for a cohort of US 50-year-olds to whom FOBT screening is offered. We compared 2 strategies: 1) FOBT with per-test specificity of 95% in combination with a screening interval of 1 year (FOBT95-1y) and 2) FOBT with per-test specificity of 80% in combination with a screening interval of 5 years (FOBT80-5y). We selected specificity and screening interval combinations such that both strategies had a similar chance that an individual would experience a false positive FOBT result during the screening program. Per-test sensitivities for FOBT95-1y and FOBT80-5y were assigned according to ROC curve projections (15% and 30% for small precursor lesions, 35% and 50% for large precursor lesions, and 70% and 90% for CRC, respectively). We assumed perfect adherence in the base-case analyses. In sensitivity analyses, we used recent US data on longitudinal adherence with FOBT screening in community practice that showed that of those who attended FOBT screening, 42%, 44% and 14%, respectively, received 1, 2-3, and 4 or more FOBTs during the 5-year study period.   

Result: In the base-case analyses, FOBT95-1y saved 22% more life-years and the number of screening-related tests was 29% higher compared to FOBT80-5y. When using observed data on longitudinal adherence with FOBT, the effectiveness of FOBT95-1y decreased by one third compared with the base-case analyses. The FOBT80-5y strategy was now more effective (given the higher per-test sensitivity), saving 19% more life-years, while requiring 17% more screening-related tests compared with FOBT95-1y.   

Conclusion: Taking into account that regular adherence with yearly FOBT screening is low, the potential benefit of this screening strategy is not realized in practice. Lowering the positivity threshold of quantitative FOBT (yielding a higher per-test sensitivity and a lower per-test specificity) in combination with an extended screening interval could be a pragmatic approach to optimize FOBT screening in view of real-life adherence patterns.