The short answer is that diagnostics are low-risk low-reward and therapeutics are high-risk high-reward. These risk terms are relative – in either field, most startups fail.
The cost of bringing a diagnostic device to market is $25–50M. Novel or highly instrumented technologies are on the high side of that range. The cost of bringing a new drug to market is nearly 100X more – $2.6B.
Diagnostics are less risky because the critical problems to be solved tend to be at the level of chemistry and physics. These disciplines are far more lawful and well-behaved than human disease biology. If a diagnostic test meets all its specifications for analytical performance in the lab, there is a very good chance that it will work as expected in the clinic. In contrast, despite years of preclinical testing in cell and animal models, less than 20%[1] of new drugs survive clinical testing and are ultimately approved.
But the financial upside in diagnostics is limited. As a society, we are far more willing to pay for potential cures than we are for tests that prevent or detect or manage disease.
We have a fairly effective system of price control in diagnostics. The American Medical Association has developed a set of Current Procedure Technology (CPT) codes, which cover all kinds of medical procedures, including lab tests[2] . Medicare and Medicaid have set up a schedule of reimbursement that they are willing to pay for each of these tests. Looking at the current schedule, I see that the median reimbursement for some 1400 lab tests is $19.69.
That’s reimbursement to the hospital. What hospitals – the buyers of the tests – are willing to pay is generally less than half that amount. Even for tests with sales of more than a million units per year the revenue upside is limited.
Medicare is moving from CPT-based reimbursement to Diagnosis Related Group coding, in which hospitals get a flat fee per patient based on their diagnosis. If you can make a compelling case to a hospital that your test will lower their treatment costs and thus increase their profit margin, then you may be able to command a premium price for your test. But hospital administrators are skeptical of economic models, distinguishing between “light-green dollars” of potential costs avoided vs “dark-green dollars” of actual and immediate cost reductions (eg, replacing an old expensive test with a newer cheaper test). They much prefer dark-green dollars.
By contrast, therapeutics are essentially free of price pressure in the US, and command impressive margins. Cancer drugs in particular are untethered to any linkage between cost and utility[3] . The cost per year of additional survival gained by anticancer drugs has gone from about $50K in 1996 to over $200K in 2014. The same clinical benefit costs four times as much now as it did 20 years ago.
From Pricing in the market for anticancer drugs
Bottom line – if you want to build a business that can make a modest profit and provide substantial benefit to society, start a diagnostics firm. If you want to make a billion dollars, start a cancer therapeutics firm.
Footnotes
[1] Trends in Risks Associated With New Drug Development: Success Rates for Investigational Drugs