Discussion: Cost Effectiveness of Prophylactic Mastectomy and Autologous Flap Reconstruction in BRCA1/2-Positive Patients

In the study entitled “Cost Effectiveness of Prophylactic Mastectomy and Autologous Flap Reconstruction in BRCA1/2-Positive Patients: A Markov Chain Monte Carlo Simulation Analysis,”1 Smith et al. leverage the capabilities of Markov chain Monte Carlo simulations to model the cost effectiveness of bilateral prophylactic mastectomy and autologous reconstruction for BRCA1/2-positive patients. Their cost-effectiveness analysis is based on quality-adjusted life-year (QALY) calculations and prior Medicare fee schedules used by Klifto et al.2 in their cost-effectiveness comparison across immediate breast reconstruction modalities. The QALYs in the study by Klifto et al. were derived from average utility values determined based on a review of studies in the Tufts University Cost-Effectiveness Analysis Registry. These source studies, in turn, varied in both scientific rigor and techniques to define health utility.3 While models are very helpful in exploring situations that would be prohibitive to investigate from a practical or ethical feasibility standpoint, the models are only approximations of reality and are only as strong as their assumptions and inputs, which, in this case, have several layers. As statistician George Box once proclaimed, “All models are wrong … some are useful.”4 We applaud the authors for their attempt to model a complex and nuanced topic; however, several assumptions hamper the applicability of the model. The base case assumes diagnosis of the BRCA1/2 pathogenic mutation at age 25 years, although there are no data to support this average age. More commonly, BRCA mutations are indentified after an early diagnosis of breast cancer, typically BRCA1 for women in their 40s and BRCA2 for women in their 50s.5 In addition, it may not be representative to assume that all women at age 35 years or younger are candidates for autologous flap reconstruction, as many women have not completed childbearing at this age or do not have adequate abdominal fat reserves to support this technique. The model does not include the costs associated with magnetic resonance imaging surveillance for implant integrity, risk of recurrent disease after mastectomy, and partial or complete flap loss. It is also not accurate to assume that all reconstructions for the surveillance group would be implant-based and all for the prophylactic group would be autologous. As an alternative to varying life expectancy in the sensitivity analysis, the authors could have modeled variations in reconstructive modalities to better approximate real-world circumstances. Improvements to the model assumptions may not change the overall direction of the incremental cost-effectiveness ratio; however, it would make the model more generalizable. We have similar observations about the dependence of the model’s validity on the accuracy of its input data. Incomplete input data may explain the counterintuitive trend observed with gluteal flap reconstruction (ie, the incremental cost-effectiveness ratio increases from 30 to 35, but decreases from 35 to 40). A look at the source data shows that fewer studies were available for gluteal flaps, likely resulting in underreporting of or biased outcomes, as Klifto et al.2 acknowledge. Similarly, it is counterintuitive that the standard of care (surveillance with implant-based reconstruction upon diagnosis) was associated with decreased QALYs compared with prophylactic mastectomy and autologous reconstruction, irrespective of the age at prophylactic mastectomy. Considering that women retain their natural sensate breast with surveillance, and sensation is positively correlated with quality of life,6 we would expect surveillance to yield higher QALYs than mastectomy with autologous reconstruction. We suspect that breast sensation, as well as donor-site morbidity, was not factored into the input health utility calculation. Lastly, as the authors allude to in their Discussion, the viewpoint of the stakeholder is critical to cost-effectiveness analyses. The authors propose a maximum cost such that the deep inferior epigastric perforator flap with prophylaxis would be favorable from a societal perspective; however, unlike countries in Europe, Australia, and Canada, which heavily fund health care through taxation and incorporate cost-effectiveness analyses into their health care decision-making, the United States lacks the input of societal willingness that accompanies fully or mixed socialized health care systems. In the United States, health care costs are largely passed onto patients indirectly through higher premiums and deductibles. This makes a willingness-to-pay threshold of $50,000 less applicable in the United States and challenging to interpret. Is this the amount that a patient is willing to pay out of pocket, or the total amount that payors and patients are willing to distribute among themselves? How do we determine the ratio of this distribution? These questions are not unique to this study, but rather relate to cost-effectiveness studies within the United States more generally. We encourage future authors to consider and attempt to answer these difficult but pertinent questions. ACKNOWLEDGMENT This research was supported in part by the National Institutes of Health through the Cancer Center Support Grant P30 CA008748 that supports the research infrastructure at Memorial Sloan Kettering Cancer Center. DISCLOSURE The authors have no financial relationships or conflicts of interest to disclose.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.