It's been a nuclear year for hepatitis C drugs in the nuc class. Another one blew up this week.

At least it was only a small explosion.

Vertex Pharmaceuticals (VRTX 0.10%) said the Food and Drug Administration placed a partial clinical hold on its hepatitis C drug VX-135, because the biotech saw elevated liver enzymes in patients receiving 400 mg dose of VX-135 in combination with ribavirin in a phase 2 study in Europe.

Elevated liver enzymes can be an early sign of liver damage. The issue has probably led to the demise of more drugs than any other side effect -- the liver clears foreign substances from the blood, so the organ is a common off-target side effect -- but drug-induced liver damage is especially troubling for VX-135 because the hepatitis C virus attacks the liver.

The hold is only a partial hold because the FDA is allowing Vertex to continue studies of VX-135 at the 100 mg level, but not 200 mg (and certainly not 400 mg). The FDA is clearly being cautious here. A study of the 200 mg doses in combination with ribavirin in Europe didn't produce any issues in the 10 patients it was tested in.

Vertex is continuing a trial combining VX-135 at 100 mg and 200 mg with Bristol-Myers Squibb's daclatasvir in New Zealand. The FDA doesn't have any control of clinical trials outside the U.S., although regulators in New Zealand could presumably put the kibosh on all or part of the trial.

From the limited data, it appears the 200 mg dose works better than the 100 mg dose, producing undetectable hepatitis C virus in 80% and 70% of the patients tested respectively, but the lower dose might ultimately be OK, because VX-135 will be used in a cocktail of other drugs.

Even if their effects are only additive, a 70% cure rate by VX-135, plus an 80% cure rate from daclatasvir or Johnson & Johnson's simeprevir, which is also being tested in combination with VX-135, gets you to a 94% cure rate.

In reality, the effects are likely to be synergistic. That is, the 30% of patients who didn't respond to VX-135 likely have weakened virus that daclatasvir or simeprevir could more easily knock down than their individual cure rates would imply.

Ticking time bomb?
The nuc drugs mimic nucleotides that are required by the virus to replicate. This sounds like a promising target, in theory, but they've been plagued with side-effect issues. In the last year, Bristol Myers' BMS-986094 caused heart failure, and Idenix Pharmaceuticals' IDX184 and IDX19368 were nixed because their structure was too close to BMS-986094.

There aren't too many left: Gilead Sciences' (GILD -1.15%) sofosbuvir and Achillion Pharmaceuticals' ACH-3422, which are at opposite ends of the spectrum. Sofosbuvir is under review at the FDA, and ACH-3422 is expected to enter the clinic early next year.

While the class has been haunted, I don't think Gilead's investors have too much to worry about. The drug has been through multiple phase 3 trials; if we were going to see an issue, it would likely have surfaced already.

While the risk of a side effect is low at this point, it would be disastrous for the stock price if one was discovered -- or if the drug wasn't approved because problems with its brethren caused additional scrutiny from the FDA reviewers. Achillion's ACH-3422, on the other hand, could blow up in phase 1 or phase 2, and no one would even bother to take a look at the carnage.