Chess is usually won and lost by key moves made in the early stages of the game. So too is the strategic battle between a virus and the host immune system. Host–virus interactions in the acute phase of infection are thought to be crucial in determining the outcome — ranging from rapid viral clearance to persistent infection. Hepatitis C virus (HCV) can establish a chronic infection that leads to liver disease, but the acute phase of infection is most commonly asymptomatic. By the time liver damage alerts clinicians and immunologists to a case of HCV infection, it is already checkmate.

But now, a study published in the Journal of Experimental Medicine provides the first insight into immune interactions in the acute phase of human HCV infection. Robert Thimme and co-workers measured virus levels, liver pathology and immune responses in five individuals who developed acute infection after accidental exposure to HCV through needlestick injuries. Within this group, three patterns of infection were observed: symptomatic acute infection leading to chronic infection (two subjects); asymptomatic acute infection leading to chronic infection (two subjects); and asymptomatic acute infection that was cleared (one subject).

The presence of alanine aminotransferase in the blood was used as an indicator of liver damage. In all subjects, viraemia quickly rose to high levels within the first 2–3 weeks after exposure — but the onset of liver damage was delayed by several weeks. This indicates that the virus itself is not directly cytopathic. Later, the virus levels decline, but there was no simultaneous increase in liver damage, showing that acute-phase HCV infection is partially controlled by non-cytopathic mechanisms.

Studies in chimpanzees (the only animal model of HCV infection) have implicated early CD8+ T-cell responses in viral clearance. Is this true in humans? Several HLA-A2-restricted HCV epitopes have been characterized, so HLA-A2 tetramers loaded with specific HCV peptides can be used to track specific CD8+ T-cell responses. The one individual who cleared infection was HLA-A2 positive. HCV-specific CD8+ T cells with an activated (CD38+) phenotype were readily detectable throughout the course of infection, but, in general, these cells did not produce the effector cytokine interferon-γ (IFN-γ). Only at 12 weeks post-infection did a subset of IFN-γ+, HCV-specific T cells appear, coinciding with viral clearance and resolution of liver disease. Intriguingly, these cells were CD38−. A vigorous CD4+ T-cell proliferative response to HCV was also observed, and this also correlated with viral clearance.

By contrast, the two subjects with asymptomatic infection leading to chronic infection had very weak CD4+ T-cell responses. One of these individuals was also HLA-A2 positive, but no HCV-specific CD8+ T-cell responses were detected using the tetramers. In the two patients that developed chronic infection after symptomatic acute infection, only the CD4+ T-cell responses could be measured. Both individuals had strong HCV-specific CD4+ T-cell responses that developed later. The function of this response is not clear, but it did correlate with a partial control of viraemia.

Clearly, there are different patterns of early host–virus interaction that lead to chronic HCV infection. These results indicate that HCV persistence might be due to a fundamental failure in both CD4+ and CD8+ T-cell responses early in infection. In particular, the emergence of an IFN-γ+, CD38−, non-cytopathic CD8+ T-cell response might be key to the host's victory over HCV.