Background In this study, we aimed to evaluate the effects of tocilizumab in adult patients admitted to hospital with COVID-19 with both hypoxia and systemic inflammation. Methods This randomised, controlled, open-label, platform trial (Randomised Evaluation of COVID-19 Therapy [RECOVERY]), is assessing several possible treatments in patients hospitalised with COVID-19 in the UK. Those trial participants with hypoxia (oxygen saturation <92% on air or requiring oxygen therapy) and evidence of systemic inflammation (C-reactive protein ≥75 mg/L) were eligible for random assignment in a 1:1 ratio to usual standard of care alone versus usual standard of care plus tocilizumab at a dose of 400 mg–800 mg (depending on weight) given intravenously. A second dose could be given 12–24 h later if the patient's condition had not improved. The primary outcome was 28-day mortality, assessed in the intention-to-treat population. The trial is registered with ISRCTN (50189673) and ClinicalTrials.gov ( NCT04381936 ). Findings Between April 23, 2020, and Jan 24, 2021, 4116 adults of 21 550 patients enrolled into the RECOVERY trial were included in the assessment of tocilizumab, including 3385 (82%) patients receiving systemic corticosteroids. Overall, 621 (31%) of the 2022 patients allocated tocilizumab and 729 (35%) of the 2094 patients allocated to usual care died within 28 days (rate ratio 0·85; 95% CI 0·76–0·94; p=0·0028). Consistent results were seen in all prespecified subgroups of patients, including those receiving systemic corticosteroids. Patients allocated to tocilizumab were more likely to be discharged from hospital within 28 days (57% vs 50%; rate ratio 1·22; 1·12–1·33; p<0·0001). Among those not receiving invasive mechanical ventilation at baseline, patients allocated tocilizumab were less likely to reach the composite endpoint of invasive mechanical ventilation or death (35% vs 42%; risk ratio 0·84; 95% CI 0·77–0·92; p<0·0001). Interpretation In hospitalised COVID-19 patients with hypoxia and systemic inflammation, tocilizumab improved survival and other clinical outcomes. These benefits were seen regardless of the amount of respiratory support and were additional to the benefits of systemic corticosteroids. Funding UK Research and Innovation (Medical Research Council) and National Institute of Health Research.

We have previously found that interleukin (IL)-2, IL-10, interferon (IFN)-gamma, RANTES, and tumor necrosis factor (TNF)-alpha mRNA transcription remain elevated in the trigeminal ganglia (TG) of herpes simplex virus type 1 (HSV-1) latently infected mice up to 120 days postinoculation (p.i.). To determine if this phenomenon was dependent on HSV-1 DNA replication after the establishment of latency (i.e., reactivation), cytokine gene expression was compared in TG of acyclovir-treated and untreated latently infected mice. Oral acyclovir treatment (begun 16 days p.i.) had no effect on serum levels of total anti-HSV-1 antibodies. However, there was a significant reduction in the titer of antibody specific for glycoprotein D and glycoprotein B but not glycoprotein H/L 120 days PI in the acyclovir-treated compared to vehicle-treated mice. These differences were not significant at earlier time points (i.e., days 34 and 60 p.i.). Consistent with these findings, acyclovir had no effect on cytokine gene expression in latently infected TG 35 and 60 days p.i. However, 120 days p.i., IFN-gamma and TNF-alpha mRNA were approaching baseline levels in TG of acyclovir-treated mice, but remained significantly elevated in untreated controls (i.e., IFN-gamma mRNA levels were sixfold higher in TG of untreated mice). Therefore, viral DNA replication appears to provide an antigenic stimulus for persistent cytokine gene expression in latently infected TG.

No abstract

Relative to wild-type herpes simplex virus type 1 (HSV-1), ICP0-null mutant viruses reactivate inefficiently from explanted, latently infected mouse trigeminal ganglia (TG), indicating that ICP0 is not essential for reactivation but plays a central role in enhancing the efficiency of reactivation. The validity of these findings has been questioned, however, because the replication of ICP0-null mutants is impaired in animal models during the establishment of latency, such that fewer mutant genomes than wild-type genomes are present in latently infected mouse TG. Therefore, the reduced number of mutant viral genomes available to reactivate, rather than mutations in the ICP0 gene per se, may be responsible for the reduced reactivation efficiency of ICP0-null mutants. We have recently demonstrated that optimization of the size of the ICP0 mutant virus inoculum and transient immunosuppression of mutant-infected mice with cyclophosphamide can be used to establish wildtype levels of ICP0-null mutant genomes in latently infected TG (W. P. Halford and P. A. Schaffer, J. Virol. 74: 5957-5967, 2000). Using this procedure to equalize mutant and wild-type genome numbers, the goal of the present study was to determine if, relative to wild-type virus, the absence of ICP0 function in two ICP0-null mutants, n212 and 7134, affects reactivation efficiency from (i) explants of latently infected TG and (ii) primary cultures of latently infected TG cells. Although equivalent numbers of viral genomes were present in TG of mice latently infected with either wild-type or mutant viruses, reactivation of n212 and 7134 from heat-stressed TG explants was inefficient (31 and 37% reactivation, respectively) relative to reactivation of wild-type virus (KOS) (95%). Similarly, n212 and 7134 reactivated inefficiently from primary cultures of dissociated TG cells plated directly after removal from the mouse (7 and 4% reactivation, respectively), relative to KOS (60% reactivation). The efficiency and kinetics of reactivation of KOS, n212, and 7134 from cultured TG cells (treated with acyclovir to facilitate the establishment of latency) in response to heat stress or superinfection with a nonreplicating HSV-1 ICP4 ؊ mutant, n12, were compared. Whereas heat stress induced reactivation of KOS from 69% of latently infected TG cell cultures, reactivation of n212 and 7134 was detected in only 1 and 7% of cultures, respectively. In contrast, superinfection with the ICP4 ؊ virus, which expresses high levels of ICP0, resulted in the production of infectious virus in nearly 100% of cultures latently infected with KOS, n212, or 7134 within 72 h. Thus, although latent mutant viral genome loads were equivalent to that of wild-type virus, in the absence of ICP0, n212 and 7134 reactivated inefficiently from latently infected TG cells during culture establishment and following heat stress. Collectively, these findings demonstrate that ICP0 is required to induce efficient reactivation of HSV-1 from neuronal latency.Unlike productive infection, which is initiated by...