"....The biomarkers predicted response independent of conventional prognostic factors such as obesity and ethnicity, he added.
An estimated 75% to 80% of patients infected with HCV in the United States are genotype 1, which tends to confer the greatest resistance to treatment. Conventional clinical, demographic, and environmental prognostic factors have fairly limited utility for early identification of patients who will respond to treatment.
Dr. Younossi and colleagues performed mRNA profiling using six housekeeping genes and 317 mRNA transcripts from 160 genes. They performed logistic regression analysis to identify genes associated with response to therapy. They also evaluated patterns of gene expression in patients before and during treatment.
The researchers then developed predictive models that they applied to 44 patients chronically infected with HCV genotype 1. The study group comprised 19 treatment-naive patients and 25 who had not responded to prior therapy. Gene expression patterns were evaluated in each patient at baseline and then at days one, seven, 28, and 56.
Dr. Younossi reported data from analyses of gene expression models for predicting sustained virologic response before treatment, 24 hours after initiation of treatment, and seven days after starting treatment. Treatment consisted of pegylated interferon-alfa plus ribavirin.
In the treatment-naive cohort, a two-gene panel consisting of EP300 and SOC56 had a sensitivity of 86% and a specificity of 91% for predicting response before treatment.
During the first 24 hours after initiation of treatment, the combination of IL1B and ADAM9 yielded a sensitivity of 86% and a specificity of 91% for sustained virologic response.
At seven days a model focusing on changes in expression of PRKRIR yielded a sensitivity of 100% and a specificity of 75%.
Dr. Younossi said pretreatment gene expression in the treatment-experienced patients included upregulation of IFR-2 a negative regulator of interferon signaling. After initiation of therapy, changes in activity were observed in a variety of signaling cascades involved in apoptosis, proliferation, and lymphocyte metabolism.
IRF-2 provided the best pretreatment prognostic accuracy for the treatment-experienced patients. The model had a sensitivity of 62.5% and a specificity of 87.5%.
At 24 hours after the start of treatment, a model derived from expression patterns of two genes-IFIT2 and JAK1-resulted in a sensitivity of 100% and a specificity of 75%.
At seven days, a six-gene panel had a sensitivity of 100% and a specificity of 92.3% for predicting sustained virologic response.
Although the results are promising, Dr. Younossi said the prognostic strategy requires validation in additional studies involving more patients...
thanks to the two of you for your heads up....
Not to be missed is that this prelinary model can predict SVR with reasonable accuracy BEFORE treatment starts, after 24 hours of treatment and after seven days of treatment. The model appears to get incrementally more accurate in that order.
Are you able to explain what sensitivity and specificity mean?
Thank you for posting the link. If this is the real deal it could mean huge changes in treatment protocol. I'm withholding judgement, though, on anything having to do with genetics. I'm not really qualified to hold an opinion on this stuff, and there have been disappointments in the past - specifically the "18 gene predictive sequence" the Canadians announced three years ago and I've never heard anything about since.
If this is real it'll have consequences to everything from insurance, to who gets into trials, to you name it.
I'll take a stab based on the below. I'm sure someone will correct me if wrong. In any event, the results seem impressive.
For the seven day group -- After seven days they can predict/identify with 100% accuracy those who will not SVR and with 92.3% accuracy those who will SVR. I think that's it :)
Sensitivity is the proportion of patients with disease who test positive. In probability notation: P(T+|D+) = TP / (TP+FN).
Specificity is the proportion of patients without disease who test negative. In probability notation: P(T-|D-) = TN / (TN + FP).