One of the most common criticisms of MRI is expense. This problem is especially visible in the setting of prostate cancer, the most common cancer in men in the US (1) that is commonly overdiagnosed and overtreated (2). Despite high sensitivity and specificity, there is a reluctance to incorporate MRI prior to biopsy into practice guidelines due to cost (3). We explored the underlying assumptions behind the current clinical state by examining the cost effectiveness of MRI in the management of prostate cancer, and discuss the implications of this analysis.

An extensive model of treatment algorithms for prostate cancer was created using base cases of a biopsy-naïve man in three age groups (41-50 years, 51-60 years, and 61-70 years), with a biopsy indicated based on clinical criteria. The following strategies were evaluated: 1) Detecting prostate cancer with transrectal ultrasound (TRUS) guided biopsy and using biopsy results, staging, and PSA levels to guide management. 2) Performing MRI before biopsy to guide targeted biopsy using one of three strategies: (a) Cognitive guidance (b) MRI-ultrasound fusion and (c) in-gantry MR biopsy. 3) Performing MRI before biopsy and proceeding with standard TRUS biopsy even when MR does not depict suspicious lesions. The cost effectiveness of a contrast enhanced exam was examined by further simulating each strategy with and without Gadolinium contrast.

Prevalence of prostate cancer, probability of detecting significant cancer on MRI and standard TRUS biopsy, sensitivity and specificity of each technique, the probability of a patient choosing a given treatment pathway (watchful waiting, active surveillance, androgen deprivation therapy, brachytherapy, external radiotherapy and radical prostatectomy), the cost of these pathways, complication rate, and Quality Adjusted Life Years (QALY) values for each strategy, procedure costs (MRI exam/TRUS biopsy/MRI-guided biopsy), pathology costs, cost of losing a day of work, and cost of complications were derived from the literature and the Bureau of Labor Statistics (4-8). A decision tree model was formed using TreeAge Pro Healthcare (Williamstown, MA) software. One-way sensitivity analysis was performed to determine the range of parameters in which a particular strategy would be cost effective. A willingness-to-pay (WTP) threshold of $50,000 per QALY was used. Incremental cost and incremental QALYs per strategy, and Incremental cost effectiveness ratio (ICER) per QALY gained were calculated using TRUS biopsy as the common comparator group.

Table 1 summarizes each MR strategy with associated incremental QALY and ICER– higher QALY and minimized ICER are desirable. MR in-gantry biopsy, which was dominated by cognitive biopsy in terms of cost, nevertheless yielded higher QALY. Table 2 depicts the cost effectiveness of this incremental gain in QALY. Table 3 summarizes one-way sensitivity analysis thresholds, below which a particular strategy ceases to be cost effective. MRI guided strategies were cost effective over the entire plausible range of most parameters used, with the exceptions noted in the Table 3.1. MRI performed before biopsy is cost effective, even if the patient proceeds to TRUS biopsy regardless of MR results. The cost effectiveness is dominated by the need to place patients in appropriate treatment algorithm, and the cost to the patient in terms of QALY or direct treatment costs can be quite high. 2. Biopsy targeted with simple cognitive guidance is the most cost effective strategy (Table 1). The simple strategy of performing MR to decide on whether to perform a biopsy at all, followed by simple cognitive guidance could be readily adopted clinically without purchase of specialized equipment or training. However, our analysis also shows that if the routine use of MR changes patient and physician behavior from the present norms to use more optimal treatment algorithms, then fusion or in-gantry biopsy could become the dominant strategies (data not shown). 3. The incremental gain in QALY by utilizing in-gantry biopsy is cost effective. This could serve as an impetus for innovation in MR-guided intervention. 4. Based on a meta-analysis published in the literature (9), adding gadolinium contrast to the exam does not increase sensitivity and specificity over structural and diffusion imaging. With this assumption, gadolinium contrast is not cost effective. The model shows that even a minimal (i.e. 1-2%) improvement would mean that the use of Gd is cost effective. This finding illustrates an open opportunity for technical innovation in MR –Improvement in detection of prostate cancer using MRI appears to provide substantial benefit to the patient as measured by QALY and is likely to outweigh the marginal increase in cost, as tested by cost effectiveness analysis.