Perhaps the most anticipated Food and Drug Administration approval of the year has finally arrived. Today, the agency approved Gilead Sciences (GILD -1.42%) Harvoni to treat hepatitis C.

Source: Gilead Sciences

Harvoni is special because it's the first cocktail that can rid patients with genotype 1 infection of the hepatitis C virus without the use of interferon, which has to be injected, and causes flu-like symptoms. The oral drug combines ledipasvir with Gilead Sciences' already-approved Sovaldi.

Prescriptions for Sovaldi have slumped recently, but that was simply because doctors were waiting for Harvoni to gain approval. Now that it's approved, expect Sovaldi sales to plummet and be replaced with revenue from Harvoni. And then some.

Harvoni will cost $94,500, substantially more than Sovaldi's $84,000 price tag. While the sales will bring in more for Gilead Sciences, insurers won't pay more because the cost is essentially the same as Sovaldi plus interferon and a third drug, ribavirin, which was taken concurrently. The big losers here are Merck (MRK 1.37%) and Roche, which make the interferons.

Then again, maybe Gilead Sciences won't actually be making more...

Eight versus 12 weeks
The recommended treatment regimen is 12 weeks -- except for patients who have cirrhosis of the liver, and have failed previous treatments. They have to take Harvoni for 24 weeks, although we may see Gilead Sciences give insurers a break for those patients.

Source: Gilead Sciences.

But see that star next to the 12 weeks in the treatment-naive category? There's a side note that goes with it:

* HARVONI for eight weeks can be considered in treatment-naive patients without cirrhosis who have pre-treatment HCV RNA less than 6 million IU/mL [see Clinical Studies (14)].

In one of the phase 3 clinical trials that Gilead Sciences ran, the biotech compared eight weeks of treatment, which cured 94% of patients, to 12 weeks of treatment, which cured 96% of patients. When you just look at patients who started with viral RNA of less than 6 million IU/mL, adding the additional four weeks didn't do anything. Cure rates for eight and 12 weeks were 97% and 96%, respectively.

How many patients might fall in the less than 6 million IU/mL category? In the clinical trial, it was 59% of the patients. It's unclear whether the real-life number is similar. And, of course, adding in patients who have failed previous treatment will reduce the overall percentage of patients who take the drug for only eight weeks, which garners Gilead Sciences just $63,000 per patient.

The bigger question is, how many patients with higher viral loads might forgo the final four weeks of treatment, figuring that the difference between 94% and 96% isn't that big of a deal? But those numbers are misleading, because they include all the patients with low viral loads. If you back out those patients, the cure rates for patients with high viral loads treated for eight and 12 weeks are 90% and 96%, respectively.

While that still isn't that big of a difference given that the FDA recommends patients with high viral load take the full 12 weeks, I think most doctors will insist on patients taking the drug for 12 weeks.

Competition coming
(ABBV 0.11%) has its own interferon-free cocktail under review by the FDA. Assuming it gets approved later this year, we should see an interesting battle for patients. Harvoni is easier to take since it's just one pill per day, and doctors are familiar with Sovaldi, which will help make the transition to Harvoni easy.

AbbVie's management has said it doesn't want to get in a price war; but if it can't get market share at a price comparable to Harvoni, the company might have to undercut Harvoni. While insurers would love it, a price war won't be good for either company.