Resistance Analyses of Patient Viral Samples from the Remdesivir Phase 3 Adaptive COVID-19 Treatment Trial-1 (ACTT-1)

Background. Remdesivir (RDV) is a broad-spectrum nucleotide analog prodrug approved for the treatment of COVID-19 in non-hospitalized and hospitalized adult as well as pediatric patients with clinical benefit demonstrated in multiple Phase 3 trials. Here we present SARS-CoV-2 resistance analyses from the Phase 3 ACTT-1 placebo-controlled clinical trial in hospitalized adults. Methods. Oro- or nasopharyngeal swab samples in ACTT-1 study were collected on Day 1, 3, 5, 8, 11, 15, and 29. All participants with >80th and 50% of participants with < 20th percentile of cumulative viral shedding underwent resistance analysis in both the RDV and placebo arm. The SARS-CoV-2 genome was sequenced using next generation sequencing. Phenotyping was conducted using virus isolation from clinical samples or generation of select site-directed mutants (SDMs) in a SARS-CoV-2 replicon system. Results. The majority of the sequencing data were obtained from participants with 80th percentile of cumulative viral shedding from the RDV and placebo arms as shown in Table 1. Among participants with both baseline and postbaseline sequencing data, emergent substitutions in nsp12 were observed in 12 of 31 participants (38.7%) treated with RDV and 12 of 30 participants (40.0%) in the placebo arm. The nsp12 substitutions that emerged in the RDV arm were only observed in one participant each, and the majority were present as mixtures with wildtype sequence. The following nsp12 mutations emerged in the RDV treatment group and were successfully phenotyped as clinical isolates or SDMs with low to no fold change in RDV susceptibility: A16V (0.8-fold), P323L+V792I (2.2-fold), C799F (2.5-fold), K59N (1.0-fold), and K59N+V792I (3.4-fold). V792I and C799F were identified previously in vitro in resistance selection experiments (Stevens Sci Transl Med 2022). In addition, for D684N and V764L identified in the RDV arm, the recovery of neither clinical isolates nor SDMs for phenotypic analysis were successful. Conclusion. The similar rate of emerging nsp12 substitutions in participants treated with RDV compared to placebo and the minimal to no change in RDV susceptibility among the treatment-emergent nsp12 substitutions support a high barrier to RDV resistance development in COVID-19 patients.

Viral and Immunologic biomarkers of COVID-19 improve risk stratification and identify patients most likely to benefit from therapy with remdesivir

Background. ACTT-1 demonstrated clinical efficacy of remdesivir (RDV) in hospitalized patients with COVID-19;subgroup analyses suggested those most likely to benefit presented with milder clinical illness. To further clarify what subsets of hospitalized patients might benefit from RDV, we analyzed virological and immunological biomarkers in this previously reported cohort. Methods. Serum and upper respiratory tract (URT) swabs were collected on Day 1, 3, 5, 8, and 11 while hospitalized;Day 15 and 29 as able were collected and tested for quantitative RNA (URT and plasma), serum nucleoprotein (NPR), IL-6, CRP through Day 6, and serostatus (baseline only). Participants with a baseline and at least one subsequent sample were used in this analysis. Associations of all these biomarkers with clinical outcomes (mortality, recovery) and response to therapy were assessed. Of the 1062 participants in ACTT-1, 642 had baseline and at least one subsequent sample within 6 days of randomization (Fig 1, Table 1). Results. RDV-treated patients with moderate/severe disease who had elevated baseline NPR levels recovered faster (RRR 1.95 vs 1.04, p = 0.01);similar trends were noted for plasma and URT RNA levels (Fig 2A);mortality treatment effects by viral load subgroups (high or low) were not seen (Fig 2B). In patients with less severe illness, RDV treatment was associated with an accelerated decline in NPR (difference -0.062 log10 pg/ml per day, p = 0.003) and plasma RNA levels (difference -0.040 log10 pg/ml per day, p = 0.004. Fig 3A), and a decrease in the proportion of patients with increasing and/or persistent viral loads (Fig 3B). Patients with increasing/persistent viral loads also took longer to recover than those with decreasing viral loads, irrespective of disease severity: RRR for plasma RNA 0.45, 95% CI 0.28-0.73, RRR for NPR 0.44, 95% CI 0.22-0.88 for moderate/severe disease;RRR for plasma RNA 0.26, 95% CI 0.10 - 0.70 , RRR for NPR n.e. (no recoveries) for critical disease (Fig 4). Conclusion. Our study demonstrates a systemic antiviral effect of remdesivir, shows the prognostic value of viral and immunologic biomarkers for mortality and failure to recover, and identifies a group of hospitalized patients with COVID-19 most likely to benefit from remdesivir treatment. (Figure Presented).

DIVERGENT ADAPTIVE IMMUNE RESPONSES DEFINE 2 TYPES of LONG COVID

Background: More than 10% of patients infected with SARS-CoV-2 experience a Long COVID syndrome, characterized by the persistence of a diverse array of symptoms where fatigue predominates. The role of the adaptive immune response in Long COVID remains poorly understood, with contrasting hypotheses suggesting either an insufficient antiviral response or an excessive immune response that would trigger autoimmune damage. To address this issue, we set to characterize humoral and cellular responses in Long COVID patients prior to SARS-CoV-2 vaccination. Methods: Long COVID patients (n=36) were included based on (1) an initial SARS-CoV-2 infection documented by PCR or the conjunction of two major signs of COVID-19 and (2) the persistence or resurgence of symptoms for over 3 months. They were compared to convalescent COVID patients with resolved symptoms (n=23) and uninfected control individuals (n=20). IgG and IgA antibodies specific to the SARS-CoV-2 spike were detected by a sensitive S-flow assay, which measures antibody binding to spike-expressing 293T cells. For CD4+ T cell response analyses, cytokine production was measured by intracellular staining on primary T cell lines stimulated by immunodominant peptides derived from the S, M, and N viral proteins. Results: Antibody analyses revealed either strong or very low/undetectable amounts of spike-specific IgG in sera from Long COVID patients, thus distinguishing a seropositive and a seronegative group. Seropositive Long COVID patients (n=21) showed strong CD4 responses that tended to be of higher magnitude than those of convalescents (P<0.05 for 2 immunodominant peptides). In contrast, seronegative Long COVID patients (n=15) showed low or undetectable CD4+ T cells responses, with 4/15 patients showing responses above those observed in healthy donors. CD4+ T cell responses correlated with spike-specific IgG responses in seropositive Long COVID patients (P≤0.002) but not in convalescents, pointing to differences in immune memory persistence. Conclusion: These findings highlight divergent adaptive immune responses among Long COVID patients, with a group characterized by seroconversion and particularly strong CD4+ T cell responses, and a second group characterized by low or undetectable antibody and cellular responses. Further studies are warranted to determine whether the etiology and the duration of symptoms differ in these two groups of Long COVID patients.

Respiratory Syncytial and Parainfluenza Virus Infection Increase the Risk of Cytomegalovirus Reactivation in Allogeneic Hematopoietic Cell Transplant Recipients

Background. Respiratory virus infections are associated with significant and specific local and systemic inflammatory response patterns, which may lead to reactivation of latent viruses. We examined whether viral upper (URTI) or lower respiratory tract infection (LRTI) with common respiratory viruses increased the risk of CMV viremia after allogeneic hematopoietic cell transplantation (HCT). Methods. We retrospectively analyzed patients undergoing allogeneic HCT between 4/2008 and 9/2018. CMV surveillance was performed weekly and the presence of upper and lower respiratory symptoms were evaluated by multiplex respiratory viral PCR. We used Cox proportional hazards models to evaluate risk factors for development of any CMV viremia or high level CMV viremia in the first 100 days post-HCT. Each respiratory virus infection episode was considered positive for 30 days beginning the day of diagnosis. Results. Among 2,545 patients (404 children, 2141 adults), 1,221 and 247 developed CMV viremia and high level CMV viremia, respectively, in the first 100 days post-HCT. Infections due to human rhinoviruses (HRV, N=476) were most frequent, followed by parainfluenza viruses 1-4 (PIV, N=139), seasonal human coronaviruses (COV, N=134), respiratory syncytial virus (RSV, N=77), influenza A/B (FLU, N=35), human metapneumovirus (MPV, N=37), and adenovirus (ADV, N=61). In adjusted models, RSV LRTI was associated with increased risk of developing CMV viremia at all levels (Figures 1 and 2), and PIV or RSV URTI increased the risk of high level CMV viremia;all other viruses showed no association in univariable models. Figure 1. Model estimates for associations between LRTI and development of any CMV viremia Figure 2. Model estimates for associations between LRTI and development of high level CMV viremia Conclusion. We demonstrated that RSV and PIV infections are associated with an increased risk for development of CMV viremia after allogeneic HCT. This novel association provides the rationale to explore virus-specific inflammatory pathways that may trigger CMV reactivation. CMV viremia may also serve as an endpoint in clinical trials that assess new preventative or therapeutic interventions of RSV or PIV infection.

Stability of SARS-CoV-2 in phosphate-buffered saline for molecular detection. (Special Issue: Diagnostic testing for severe acute respiratory syndrome coronavirus 2 and lessons from this pandemic.)

This study evaluated the stability of differing viral loads of SARS-CoV-2 over 28 days stored at room temperature, 4 degrees Celsius, -20 Celsius, or -80 Celsius. For the high concentration of SARS-CoV-2, regardless of storage conditions, 100% of samples were detected by qRT-PCR through day 28. At room temperature, median cycle threshold (CT) values for lower titers for both N1 and N2 targets remained consistent through day 28, fluctuating less than 1 median CT. For lower concentrations of virus, storage at room temperature was associated with reductions of positivity beginning at day 7, and by day 28, 0% of samples were detected for N1. Storage at room temperature was the least stable of all environmental conditions tested, with 54.2% of negative PCR results. At 4 degrees Celsius, there was minimal change in CTs over time at the higher viral concentration. For lower titers, CTs increased by 2.1 CTs for N1 and 2.6 CTs for N2 over the 28 days. At -20 degrees Celsius, lower titers of virus fluctuated slightly more, increasing by 3 CTs. Storage of SARS-CoV-2 in PBS at -20 degrees Celsius was the second least stable condition, accounting for 37.5% of negative PCR results. Storage at -80 degrees Celsius showed the greatest stability, with all samples detected throughout the 28 days and 1.5 median CTs for both N1 and N2 targets. Here, this study shows that the stability of SARS-CoV-2 can be maintained at 4 degrees Celsius for up to a month when -80 degrees Celsius storage is not available. At viral loads of >5,000 copies/ml corresponding to >75% of positive samples recovered in the clinical lab to date-different storage temperatures did not have a substantial impact on the ability to detect SARS-CoV-2 when stored in PBS.