Nirmatrelvir-ritonavir versus usual care in at-risk adults with early SARS-CoV-2 infection in the UK, 2022-23: virological and immunological results of an open-label randomised trial (PANORAMIC).

BACKGROUND: Nirmatrelvir-ritonavir is widely used for COVID-19, but effectiveness was established in unvaccinated people with SARS-CoV-2 variants before omicron. Within the randomised Platform Adaptive Trial of Novel Antivirals for Early Treatment of COVID-19 in the Community (PANORAMIC) trial, we did a substudy to evaluate virological and immunological outcomes. METHODS: Adults (aged ≥50 years, or ≥18 years with comorbidities) with early COVID-19 were recruited across the UK in an open-label trial and randomly assigned to receive nirmatrelvir-ritonavir twice daily for 5 days plus usual care (nirmatrelvir-ritonavir group) or usual care alone (supportive treatment only with the use of antipyretics as required; usual care group). Before randomisation, participants were also invited to enter the virology substudy, and those who consented were sent swabs and blood spot testing kits for self-sampling with instructions. We performed viral quantitative PCR, viral culture, and sequencing on self-collected combined nasal and pharyngeal swabs submitted daily for 7 days and at day 14 (intensive sampling group) or at baseline, day 5, and day 14. We measured blood spot SARS-CoV-2 spike antibody titre and C-reactive protein (CRP) at baseline, day 5, and day 14 in all participants. We undertook mixed-effects modelling to identify predictors of viral load or spike antibody rate of change and assessed viral rebound by multiple definitions. The trial was registered with EudraCT, 2021-005748-31. FINDINGS: From Sept 10, 2022, to Oct 23, 2023, 649 people agreed to participate in the virology substudy, of whom 27 were excluded per the prespecified analysis plan, resulting in 622 participants in the analysis: 326 in the nirmatrelvir-ritonavir group and 296 in the usual care group. Nirmatrelvir-ritonavir treatment reduced viral load from day 5 onwards and increased viral clearance. The estimated half-life was 0·78 days (95% CI 0·74-0·83) with usual care versus 0·66 days (95% CI 0·56-0·68) with nirmatrelvir-ritonavir; p<0·0001). Treatment reduced culturable virus at any timepoint after treatment initiation (three [3%] of 92 culture-positive samples vs 18 [15%] of 119 culture-positive samples; p=0·0046). Spike antibody doubling time over 14 days correlated positively with viral area under the curve and was significantly slower with nirmatrelvir-ritonavir treatment compared with usual care (doubling time 12·80 days [95% CI 11·11-14·83] with nirmatrelvir-ritonavir treatment vs 10·36 days [9·08-11·83] with usual care, p<0·0001); CRP at days 5 and 14 was similar between groups. There was no increase in viral rebound, mutagenesis, or resistance with nirmatrelvir-ritonavir treatment. Baseline spike antibody concentrations were high on average (mean 3·68 log10 U/mL [SD 0·48]), and baseline viral loads were low (mean 6·42 log10 copies per mL [1·49]) with the two correlating inversely (slope -0·63; p<0·0001). INTERPRETATION: Nirmatrelvir-ritonavir is effectively antiviral against omicron SARS-CoV-2 variants, but the magnitude of viral load reduction might be limited by effective population-wide immunity in the immunocompetent population. FUNDING: UK National Institute for Health and Care Research.