HIV and ID Observations

Over in the British Medical Journal, there’s a provocative editorial entitled, “Is widespread screening for hepatitis C justified?”

Based on the title alone, you can guess the authors’ answer to that question — a resounding “No!”

By taking this position, of course, they are opposing some very data-driven and well-respected arbiters of policy and clinical practice. These include not only the Centers for Disease Control — which started recommending universal screening in 2012 to people born between 1945–1965 (the years of peak incidence) — as well as the World Health Organization, and, for good measure, those staggeringly data-driven folks from the U.S. Preventive Services Task Force (USPSTF).

Remember, these USPSTF guys are a scrupulous bunch, opposing many forms of screening for insufficient evidence. Just the very fact that they endorse HCV screening is almost reason enough to accept it as worthwhile. Indeed, they reversed their earlier opposition to HCV screening with their latest review — and did so even before interferon-free therapies became standard of care!

(Why is it so hard to say and write that combination of letters? — “USPSTF” reminds me of the confirmation codes that the airlines use for flights.)

The key points of the of the BMJ opinion piece go like this, roughly in order:

1. The natural history of hepatitis C suggests it is generally benign.
2. Achieving a “sustained virology response,” or SVR, is only a surrogate marker for the important clinical endpoint of liver disease. Proof:  Some people who achieve SVR still go on to have decompensated disease and hepatocellular carcinoma.
3. SVR does not mean HCV is cured anyway.
4. Screening will be dangerous because the treatment could be more harmful than the disease; we also need long-term safety data on the newer drugs.
5. Before we adopt widespread HCV screening, a large randomized trial should be done comparing broad to risk-based screening, with clinical outcomes as the primary endpoint.

There’s always room for some dialogue in any policy statement, and the editorial does raise some important points.

But let’s take these above positions one at a time:

1. Most people with HCV will never develop clinically significant liver disease. Yes, this is true, but HCV is still the leading single cause of end-stage liver disease leading to liver transplantation in the USA; HCV can furthermore cause numerous non-hepatic complications; and a chronic infection with high rates of ongoing replication is probably bad for general health, which almost certainly accounts for the higher rates of fatigue, nausea, pain, and decreased quality of life in those with HCV versus those without.
2. SVR is only a surrogate marker. If you’re a real skeptic, you’d argue that the better clinical outcomes observed in patients who achieve SVR versus those that don’t might be confounded by their better health at baseline. On the other hand, many believe that the benefits of SVR have been convincingly shown through clinical cohorts, adjustment of baseline characteristics, and improvement in liver histology among those who achieve SVR. And what about those patients with SVR who still get liver-related illness? Most had advanced liver disease before they were treated — one could argue that if they’d been diagnosed decades earlier and cured, their liver disease never would have happened.
3. “Sustained virological response is not a cure,“ writes the paper confidently. This is certainly a minority view, and seems to be based on the existence of case reports of late relapses, with a frequency of no greater than 3% (and probably lower). Isn’t the very existence of these papers good evidence that most of the time SVR is a cure? Why would any of them be published unless they were exceedingly rare events?
4. Treatment is harmful — in most patients, more harmful than the disease. For currently relevant treatments, they cite this paper, where “3% of participants taking sofosbuvir experienced serious adverse events compared with 1% in the peginterferon plus ribavirin arm.” But these data are for any serious adverse event, not just those that are drug-related. Bad things can happen to patients who are in clinical trials, especially trials that include patients with chronic diseases — it doesn’t mean the drugs caused them. In trials of newer HCV drugs, especially those that don’t include interferon or ribavirin, serious adverse events that are conclusively drug-related are extraordinarily rare. And one last thing: No review of HCV therapy today should list side effects of telaprevir or boceprevir as a reason for avoiding treatment (both are cited). These drugs are no longer recommended — and the former isn’t even available. Even interferon is rarely indicated today.
5. We need to do a large randomized strategy trial of broad versus targeted HCV screening with clinical outcomes. It’s true we do not definitively know that broad screening will prevent liver disease or death beyond a risk-based approach. But unfortunately, patients can be infected with HCV (and unaware of that fact) for decades before showing up with advanced liver disease. Seems that a cheap, reliable blood test is highly justified when we now have such safe and effective treatments.

I agree with the authors that longer-term follow-up data of treated patients are needed, especially using our currently available drugs.

The good news is that with thousands taking of advantage of HCV therapy — including many patients who could never be treated in the interferon era — and with over 90% of these patients cured (there, I said it), we shouldn’t have to wait long.