An ongoing dialogue on HIV/AIDS, infectious diseases,
July 19th, 2017
Mystifying Cochrane Library Review on HCV Therapy Elicits Strong Response from IDSA
Last month, the Cochrane Review published a controversial paper on HCV therapy that left many ID doctors and hepatologists perplexed.
After reviewing 138 randomized clinical trials using directly acting, non-interferon based therapies, they came to the following conclusions:
- The use of sustained virologic response (“SVR”) — or “cure”, if you want to use plain English — as a valid endpoint for predicting clinical outcomes is questionable.
- There is currently insufficient evidence that treatment with DAA-based regimens improves clinical outcome.
- The studies reviewed were at high risk of bias, so tended to overestimate benefits and minimize harm.
- More randomized clinical trials are needed.
Anyone — clinician, researcher, or patient — who has experienced the miraculous advances in HCV therapy that started in 2014 could easily be scratching their heads at these conclusions.
The FDA might be surprised as well, since they have allowed SVR as an appropriate “surrogate” marker of the effectiveness of HCV therapy for some time.
Fortunately, we now have a focused, persuasive response by the IDSA, just published in Clinical Infectious Diseases.
I strongly encourage anyone who doubts the clinical benefits of curing HCV to read the full paper, but in essence the argument goes like this:
- The review was overly selective in the papers it included. Remember, many HCV trials could not include a control group since DAA therapies were so rapidly effective and well tolerated it would have been unethical. These non-controlled studies were not included in the review.
- HCV cure as an appropriate marker for treatment efficacy was established during the years of interferon-based therapy. Liver inflammation (as measured by biopsy or serial LFTs), fibrosis, portal hypertension, splenomegaly, even cirrhosis improved in those with SVR. And I would add that some surrogate markers are more intuitively obvious than others — and you can’t really get more obvious than curing the very infection that’s causing the disease. HCV RNA is not an obscure, indirect tumor marker (oncology), or a change in lipids (cardiology). It’s analogous to HIV RNA in HIV therapy, only better. And is there any plausible biologic reason why HCV cure with DAAs might be less effective in improving clinical outcomes than using interferon?
- The time horizon to see the full clinical benefit for HCV cure will take many years. We’ve only had these therapies widely available since 2015 — hardly enough time to see reductions in the incidence of long-term complications such as cirrhosis or hepatocellular carcinoma. Note that we’ve already seen benefits in HCV transmission from treatment in a clinical cohort of MSM from Europe.
- Despite this short time period of DAA availability, clinical benefits have already been observed with HCV cure. These include resolution of vasculitis, spontaneous remission of non-Hodgkin lymphoma, and — perhaps most remarkably — stabilization or improvement in those with the most advanced forms of HCV liver disease.
I will note that this isn’t the first time a “systematic review” of an Infectious Disease treatment under the Cochrane name ended up with a surprising conclusion.
Remember this one on HIV treatment with TDF/FTC/efavirenz? The one which stated there was insufficient evidence to support its use, despite numerous randomized clinical trials documenting its efficacy? And its widespread adoption in clinical guidelines?
It may be hard to find today, since it was later withdrawn.
It might be fortunate that we now have a focused, persuasive response by the IDSA, just published in Clinical Infectious Diseases but it is unfortunate we don’t have an analysis of the financial incentives of any of the recommendations (or at least any mention of such an analysis). Financial incentives are a powerful motivating force and often the origin of complex rationale justifying such recommendations.
What a wonderful description of the failure of “research” to translate into “practice”.
Remember, according to research, on average every human has one testicle.
I don´t see (or better saying, I should say we ALL see) the whys for such a strong response to the Cochrane metanalysis result pointed in their paper on HCV treatment with DAAs as they are clearly unbiased and seems to be very adequate.
They don´t claim that the drugs don´t work, but merely point out that there are no solid evidence of its clinical benefits.
This is something I´ve being alerting for a while, since one of the first papers published on NEJM about HCV treatment with interferon, in which it was recommended to use erythropoetin and granulokyne to circunvent adverse events of a therapy merely 30% effective at best.
From then on, what we have abundantly saw is probably the best exercise of publication planning ever.
Worse than that they succedded in convincing people that they don´t have to follow up patients that achived SVR so it will be more and more difficult to see if they really benefit on the long run.
I´m not saying that these drugs don´t work, but they have to be put to the real test and we, ID physicians should be more cautious when alerting our patients about the long term results of such treatments.