LAUSR

Control of hepatitis B virus (HBV) infection requires functional virusspecific T cells, yet clinical management of patients with chronic HBV infection (CHB) relies exclusively on the assessment of virological (HBVDNA, HBsAg) and biochemical (alanine transaminase (ALT)) biomarkers. There is, however, a growing recognition of the necessity to categorise CHB patients based on their profile of HBVspecific immunity. Such immunological biomarkers might guide when to start or stop nucleos(t)ide analogue (NA) therapy, and/or identify patients who would benefit from novel therapeutic strategies designed to modify host–virus interaction by restoring HBVspecific immunity, either directly (therapeutic vaccines or check point inhibitors) or indirectly (antisense nucleotides, siRNA) (figure 1). In Gut, Ferrari and Boni’s group present findings which could potentially help address this unmet clinical need. Their study in treatment naïve CHB patients shows that simple phenotypic analysis of total circulating CD8T cells can predict HBVspecific CD8 T cell response to immunomodulatory compounds. For three decades, we have known that HBVspecific T cell immunity differs between patients with acute and CHB (reviewed in Chisari and Ferrari ). More recent works have shown that HBVspecific T cells recover, even though often only partially, 4 after functional cure. 6 Such data support the idea that restoration of functional HBVspecific T cell immunity can achieve HBV control but fails to take into account a key characteristic of HBVspecific T cells in CHB patients: their extreme heterogeneity. 7–9 Although extremely low levels of HBsAg and HBcrAg can predict superior responsiveness to PDL1 blockade, it is now clear that the HBVspecific T cell response does not preferentially segregate into the classical CHB definitions (ie, immune tolerant, chronic active, asymptomatic carriers), which are based on virological and biochemical profiles. 12 HBsAg quantity is not associated with frequency and/or function of HBVspecific T cells. 13 Levels of transaminases, the other biomarker used to categorise CHB patients, do not correlate with HBVspecific T cell responses; ALT levels are not proportional to HBVspecific T cell frequency both in the periphery and the liver. Indeed, hepatic damage in CHB is primarily caused by inflammatory and necrotic events supported by myeloid and nonantigenspecific CD8T cells. Thus, since virological and clinical biomarkers cannot predict the level of HBVspecific T cell immunity, its direct assessment is necessary. Unfortunately, such assessment is highly complex, and has thus far prevented its application in the clinic. It is therefore of great practical importance that Rossi et al showed that a phenotypic analysis of circulating total CD8T cells, performed by analysing exhaustion (PD1, TOX) and memory (CD127, BCL2) markers, can generate an ‘exhaustion score’ that predicts HBVspecific T cell responsiveness to immunomodulatory therapy. The authors first measured the expression of exhaustion and memory markers in selected core and polymeraseHLAA2 restricted CD8T cells in treatment naïve CHB patients with liver inflammation (HBeAg+ and HBeAg− chronic hepatitis). In line with previously described HBVspecific CD8T cell heterogeneity, they found variable levels of expression among patients. The expression of these markers was then quantified to calculate an exhaustion score (EI, a numeric value of the zstandard medium fluorescent values of PD1, TOX, CD39, CD127, BCl2). The authors found that EI scores of HBcorespecific CD8T cells varied across the cohort. The EI score was independent of viral load and HBsAg quantity but positively correlated with ALT. Importantly, the score was inversely correlated with the ability of HBcorespecific CD8T cells to produce cytokines directly ex vivo. To expand on this observation, the authors measured the capacity of the HBcorespecific CD8T cells to proliferate in vitro after peptidespecific stimulation, either in the presence of antioxidant, polyphenolic or PD1/PDL1 targeting compounds or of a tolllike receptor 8 (TLR8) agonist—all agents proposed to restore exhausted T cell function. The ability of the HBcore CD8T cells to proliferate in vitro in the presence of these