New drugs for Hepatitis-C – Cost-effective but too costly?

Until a few years ago, people infected with the Hepatitis-C virus (HCV) were in a bad way.  Treatments were available, but they included interferon which had nasty side effects.  Most patients experienced fatigue, headache, and muscle aches. A third or more had nausea, fever, depression, irritability and insomnia.  Even worse, cure rates averaged under 60%.

But then there was a major clinical breakthrough.  Gilead Sciences introduced a new drug –Sovaldi (sofosbuvir) – that provided cure rates of 95%.  While most patients had to take Sovaldi with interferon, many could be treated with an interferon-free regimen that avoided most side effects.  A year or so later Gilead introduced an improved product – Harvoni – that consisted of sofosbuvir and ledipasvir.  With Harvoni almost all HCV could be cured by a single 12-week, interferon-free regimen.  What a drug – ultra-high cure rates and minimal side effects.

But, as with most things in life, there was a down side.  The drugs are expensive.  The list price for the recommended 12-week treatment is $84,000 for Sovaldi or $94,500 for Harvoni.

This was a new situation for most third-party payers.  Sovaldi and Harvoni were not the first high-priced drugs to hit the market.  A number of very expensive treatments for cancer, rheumatoid arthritis, and multiple sclerosis were commonly used and reimbursed by third-party payers.  But these products treated diseases with relatively small patient populations.  Sovaldi and Harvoni treated a disease that affected 3 million people in the U.S.

Costs and Cost-effectiveness

When evaluating whether or not to provide coverage for a new drug, third-party payers typically look first at safety and effectiveness and, assuming the drug is sufficiently safe and effective, at its cost.  Sovaldi and Harvoni pass the safety and efficacy screen.  They appear to be safe and they are much more effective than the older treatments.

In situations where a new drug is more effective than the current standard of therapy, but also more expensive, payers frequently examine the new drug’s cost-effectiveness to determine whether or not to provide coverage.  This involves comparing the cost and effectiveness of the new drug to that of the current standard of therapy.  A general rule of thumb is that if the drug has a cost per quality adjusted life year (QALY) saved of less than $50,000 (or, for some payers, $100,000), then it’s cost-effective and should be covered.  But a drug that is cost-effective will not necessarily be cost-saving.  Treating HCV, or any disease, with a drug with a cost-effectiveness ratio of $50,000 per QALY saved will result in a lot of additional spending.

Studies by  Chhatwal, et al., Linas, et al., and Najafzadeh, et al. indicate that Sovaldi and Harvoni are cost-effective for most types of HCV.  This is a result of their high effectiveness, the much lower effectiveness of the prior treatments, and the high cost of the clinical outcomes of untreated HCV (cancer and liver transplants).

Cost-effective products can still break the bank

While Sovaldi and Harvoni are cost-effective, they are not cost saving.  A payer that covers them can expect to see a big increase in the amount it spends on drugs.  A back-of-the-envelope calculation shows the problem:

Assume it costs $94,500 to treat an HCV patient and that there are 3 million of them in the U.S. This means it would cost the U.S. $284 billion to treat all its HCV patients.  By comparison, all U.S. prescription expenditures amounted to $325 billion in 2013.  Providing Harvoni to every U.S. HCV patient would increase national drug expenditures by 47%.

A recent analysis by Chhatwal, et al. in the Annals of Internal Medicine presents a more sophisticated analysis.  It assumes that  about 1.8 million people with HCV in the U.S. are aware that they have the disease, that about 510,000 more will be diagnosed over the next 5 years, and that only these will be treated.   Their results indicate that it would cost $136 billion to treat these patients with Harvoni, that this would be $65 billion more than it would cost to treat them with the older therapies, and that use of Harvoni would prevent $16 million in costs – such as treatment for liver cancer and liver transplant – that would have occurred had patients not been effectively treated.  Even with these more conservative estimates the authors state that “despite the cost-effectiveness of HCV treatment, our analysis shows that it is unaffordable at the current price.”

Recent estimates indicate that Gilead has been providing payers with discounts of 40-60% on Sovaldi and Harvoni.  Chhatwal and colleagues only assumed an 11% discount.  But even assuming the higher end discount of 60%, treating known HCV patients would cost an additional $29 billion over the next 5 years. (Chhatwal’s estimate of $65 billion in additional spending assumed an 11% discount.  This means the undiscounted amount was $65/(1-0.11) = $73 billion.  A 60% discount off $73 billion would be $29 billion).

So, Sovaldi and Harvoni present the interesting case of drugs that are both cost-effective and too expensive.

What are payers to do?

This places payers in a tough spot.  On one hand, how can they justify denying patients treatment with a drug that is both highly effective and cost-effective?  (There is now another similarly effective product for HCV – Viekira Pak – but at a price of $83,320 for a 12-week course of therapy it’s not a lot cheaper.)  On the other hand, how can they treat all HCV patients without going broke?

And worse situations may be around the corner.  HCV is an infection that can be cured with a single 12-week treatment.  The New York Times  recently reported that the newest treatments for high cholesterol – the PCSK9 inhibitors – have estimated prices of $14,600 per month.  Most patients will be on these products for life.