Is pharmacogenetic-guided treatment cost-effective? No one size fits all!

A recently published article by M. Verbelen and colleagues in The Pharmacogenomics Journal is called, “Cost-effectiveness of pharmacogenetic-guided treatment: are we there yet?” As Betteridge’s law of headlines states, any headline that ends in a question mark can be answered by the word No. Regrettably, although that article presents useful information, it ends up by answering that question with a qualified Yes. We argue that applications of pharmacogenetics (PGx) are too diverse, with sparse epidemiological evidence of effectiveness and clinical utility, for assertions of the overall cost-effectiveness of PGx to be of value to decision makers. We suggest that a better way to frame the question is, “When is pharmacogenomic testing cost-effective?”

The sample space for Verbelen’s study was limited to the FDA Table of Pharmacogenomic Biomarkers in Drug Labeling, which at the time contained 137 PGx associations. Of these, only germline associations were considered, so that, for example, tumor biomarkers were not included. Verbelen and colleagues reviewed 44 economic evaluations relating to 10 drugs, of which 57% considered PGx testing to be either cost-effective or cost-saving. That conclusion is similar to previously published reviews on this topic, which is not surprising since most of the studies were included in previous reviews. The authors acknowledge that positive findings of cost-effectiveness are more likely to be published. We argue that generalizing about cost-effectiveness based on results from a limited set of economic evaluations that are unlikely to be representative of PGx applications is probably not helpful and could be misleading.

Cost-effectiveness depends crucially on evidence of effectiveness, i.e., improved health outcomes. As we have pointed out previously, including an invited commentary on a previous review by Phillips et al., lack of high quality evidence is the Achilles heel in PGx and genetic testing. The absence of clear evidence of effectiveness should call into question published findings of cost-effectiveness. If something is not effective, it cannot be cost-effective! Verbelen et al. acknowledge that most economic evaluations in genomics and PGx are not informed by high-quality clinical evidence. In the absence of such evidence, it is not difficult to make something appear cost-effective, if that is one’s objective, through the careful selection of assumptions.

Cost-effectiveness is a fluid concept. Like beauty, cost-effectiveness is in the eyes of the beholder. Stakeholders who have different definitions of cost and cost-effectiveness may reach different conclusions. Also, an intervention may be calculated to be cost-effective relative to doing nothing but not cost-effective compared with another treatment. Or, it may be cost-effective at one set of prices but not cost-effective at another. Verbelen and colleagues suggest that the price of genetic testing is influential in the cost-effectiveness of PGx, pointing out that PGx is more likely to be cost-effective if the cost of testing were to reach zero. However, real-world data summarized by the authors did not confirm the importance of testing costs. Specifically, although genetic testing costs are substantially higher in North America than in Europe, PGx tests were just as likely to be found to be cost-effective in U.S. and Canadian studies.

The authors helpfully discuss in detail the example of PGx-guided warfarin dosing, which is the most commonly studied PGx application. They note that most economic evaluations have not found this application to be clearly effective or cost-effective. First, Verbelen et al. note that PGx-guided warfarin dosing does not appear to be cost-effective relative to novel oral anticoagulants. Second, that application appears unlikely to be cost-effective even relative to standard warfarin dosing without PGx, since, “the clinical advantage of genetic-guided dosing over standard dosing appears to be small or even non-existent.”

Not all PGx applications have received the same critical scrutiny as warfarin dosing. We urge that stakeholders resist the temptation to reach conclusions about the cost-effectiveness of PGx without both clear evidence of clinical utility as well as thorough documentation of epidemiologic and cost assumptions. It is important that epidemiology not be relegated to a “black box” in economic evaluations of genetic testing. Also, cost-effectiveness analyses should not be seen as “shortcuts” to evidence-based genomic medicine.