There’s a lot we don’t know about how and why hepatitis C treatment works in some people but not others. Roughly half of people treated with pegylated interferon and ribavirin achieve a sustained virologic response (SVR – hepatitis C is undetectable six months after ending treatment), the main measurement of treatment success. That makes treatment decisions something of a gamble – you can’t know in advance whether treatment will work for you and whether it’s worth the risk of side effects.
Complicating these questions is the fact that the actual mechanisms through which pegylated interferon and ribavirin act to suppress hepatitis C are not well understood. Both drugs have a range of direct and indirect effects, both in blocking the virus and stimulating or shifting immune responses. But it’s not clear which of these effects are actually important in fighting hepatitis C.
For now, speculation about the reasons for treatment success or failure fall into three categories: 1) it’s about the virus; 2) it’s about the person; 3) it’s about the drugs. I’ll talk separately about the evidence and rationale for each of these explanations here and in future entries.
It’s about the virus: the evidence
The bulk of evidence points to a major role for the hepatitis C virus in determining treatment outcomes. First, the biggest variation in treatment responses are seen between genotypes (strains of virus): about 80% or more of people with genotype 2 or 3 respond to treatment, but only about half of people with genotype 1 clear the virus. People with genotype 2 or 3 generally need only 6 months of treatment, but people with genotype 1 are typically treated for a full year (see current treatment guidelines: PDF file).
It’s not clear why genotypes differ in susceptibility to treatment, but these differences have been seen in all major treatment studies. For this reason, many doctors are more aggressive about treating people with genotype 2 or 3 because of the strong likelihood of an SVR. Some doctors would treat people with genotype 2 or 3 even without first performing a liver biopsy to assess liver damage, but virtually all doctors would want biopsy results for people with genotype 1 in order to assess the relative need and urgency for treatment.
Some other research has tried to determine whether minor genetic variations in the virus, or overall genetic diversity of the virus population, influences response to treatment, but the results of these studies have been mixed and somewhat controversial. As a result, doctors don’t look at these factors when talking to people about treatment.
How much virus someone starts off with also influences response to treatment. People with higher viral loads are less likely to achieve an SVR. A recently published study found that in people with genotype 1 treated for 48 weeks with pegylated interferon alfa-2a (Pegasys) and 1,000-1,200 mg of ribavirin, an SVR occurred in 65% of people with a low hepatitis C viral load (below 2 million copies/mL) but only 47% of people with a high viral load (Hadziyannis et al., Peginterferon-α2a and Ribavirin Combination Therapy in Chronic Hepatitis C, Annals of Internal Medicine vol. 140 issue 5, March 2 2004 – abstract, summary for patients).
This makes intuitive sense; the less virus you start with, the easier it should be to clear it. And a growing amount of evidence shows that treatment’s most likely to work you’re your viral load drops quickly and substantially after starting treatment. This field of research, called viral kinetics, has demonstrated that people who don’t experience at least a hundred-fold or 99% decrease in viral load (e.g., from 2 million down to 20,000 or lower; also called a 2-log drop) after the first 12 weeks of treatment virtually never achieve an SVR. Thus the 12-week stop rule: most doctors tend to recommend discontinuing treatment if your viral load isn’t undetectable, or hasn’t fallen a hundred-fold, by week 12 (3 months) of treatment.
Some research also suggests that changes in viral load at 4 weeks on treatment provide a good indication of the chance of achieving an SVR, and people who study viral kinetics are attempting to refine their models to help predict treatment outcomes at the earliest possible time. The goal is to help people who won’t get an SVR find out as soon as they can, so they don’t have to stay on treatment and experience unnecessary side effects. For now, the 12-week stopping rule seems the most reliable. (See this recent review by Peter Ferenci from the Journal of Antimicrobial Chemotherapy for more details on viral kinetics research).
(Keep in mind that some people who don’t get an SVR seem to experience some benefit to the liver from staying on treatment – this is called a histologic response, as measured by biopsies before and after treatment. That could provide a rationale for staying on treatment even though you don’t meet the 12-week criteria for virologic success—particularly for people with advanced liver disease or increased risk of progression. But not everybody sees a histologic benefit, and nobody knows how closely it’s related to decreases in viral load. More importantly, it’s not clear whether a histologic response without an SVR translates into slower disease progression or better long-term health and survival.)
Other researchers have recently looked at changes in viral load during treatment of people with genotype 1. George Drusano and Sandra Preston of the Ordway Research Institute in New York developed a mathematical model based on published hepatitis C treatment studies and found that people with genotype 1 may need to be treated for longer than they typical 48 weeks (abstract from Journal of Infectious Diseases vol. 189 issue 6, March 15 2004). They observe that it can take people with genotype 1 substantially longer to reach an undetectable viral load – on average 30 weeks – than it typically does for people with genotype 2 or 3.
Drusano and Preston suggest that SVR rates may increase if people with genotype 1 were treated for an additional 32-36 months after first reaching an undetectable viral load – for a total, on average, of 62-66 weeks of treatment (about 15 months), compared to the current standard of 48 weeks (12 months). This model needs to be tested in a study before gaining widespread acceptance.
So characteristics of the virus you’re starting with clearly make a difference, both for determining the overall chances of treatment success, the duration of treatment, and whether to continue treatment after the first 12 weeks. But you still see variation in response to treatment, even among people with the same genotype and viral load range. So what else is going on?
Next: Is it about the person?