COVID-19 Rebound: To Retreat or to Re-Treat
November 1, 2025
By Stan Deresinski, MD, FIDSA
Synopsis: Re-treatment of non-immunocompromised patients with mild-moderate COVID-19 rebound with nirmatrelvir/ritonavir provided no significant benefit.
Source: Zhang W, Terra SG, Weinstein EA, et al. Retreatment with nirmatrelvir/ritonavir following return of COVID-19 symptoms and SARS-CoV-2 positivity. Clin Infect Dis. 2025;Sep 30:ciaf548. [Online ahead of print].
The recurrence of symptoms and detectable virus after resolution of COVID-19 (i.e., rebound) occurs in a small proportion of patients regardless of whether they have received antiviral therapy. It has been suggested by some that despite an absence of supporting evidence, a second course of antiviral treatment should be administered to patients with rebound.
Zhang and colleagues examined the efficacy of a second course of nirmatrelvir/ritonavir (NR) in the treatment of patients with COVID-19 rebound. A total of 436 patients with mild-moderate symptoms were randomized (2:1) to receive either a second five-day course of NR or placebo with ritonavir. Patients were at least 12 years of age, had at least one risk factor for severe disease, and had recurrence of symptoms and positive antigen test within 14 days after completion of initial NR treatment. Reasons for exclusion included immunocompromise as well as a requirement for hospitalization during the initial COVID-19 infection or within 24 hours of randomization.
Of the total analysis set, 292 received NR and 144 received placebo plus ritonavir. All were infected with the Omicron variant of SARS-CoV-2. The median patient age was 55.0 years and 72% were younger than 65 years of age. At baseline, the median nasopharyngeal SARS-CoV-2 level was 5.51 log10 copies/mL (range, 0.00-8.96). Most patients (80.5%) were vaccinated. The median interval from completion of their initial treatment course to randomization was 5.0 days (range, 0-18). The primary efficacy endpoint was the change in viral SARS-CoV-2 ribonucleic acid (RNA) level from baseline to day 5 of repeat treatment.
Active treatment was associated with a significantly greater reduction in SARS-CoV-2 levels from a mean of 5.80 log10 copies/mL to 1.74 log10 copies/mL, while with placebo/ritonavir the change was from 5.6210 copies/mL to 2.4610 copies/mL (P = 0.0004). In the modified intent-to-treat (mITT) population, the median time to achieving two consecutive negative antigen tests 24 hours apart was four days in the NR group and five days in those randomized to receive placebo/ritonavir, a difference that was not statistically significant. Also not statistically significant was the median time to sustained alleviation or resolution of all targeted signs/symptoms, which was eight days in the nirmatrelvir/ritonavir group vs. nine days in the placebo/ritonavir treatment group.
Viral RNA rebound (rebound of the rebound?) through day 34 of retreatment was detected in 9/217 participants (4.2%) in the NR group and in 2/113 (1.8%) of placebo/ritonavir recipients. There was no association with emergence of Mpro substitutions or of infectious virus.
Treatment was well tolerated, although 10.4% and 1.4% of NR and placebo recipients, respectively, reported dysgeusia — a symptom that could affect patient perception of their assigned treatment group.
Commentary
COVID-19 rebound is defined as return of symptoms and/or viral positivity after resolution of the initial infection manifestations. It occurs in a small proportion after treatment with NR but also in the absence of any treatment. It is suggested that it is a result of incomplete antiviral response, including of the innate immune system.1 Nirmatrelvir inhibits SARS-CoV-2 replication but does not directly eliminate the virus, thus allowing reemergence of viral replication in the absence of complete immunological control. It should be noted that ritonavir is considered a placebo in this trial since it is not active against SARS-CoV-2 but rather interferes with the metabolism of nirmatrelvir and, as a result, increases exposure to this active agent.
The data indicate that a five-day retreatment of COVID-19 rebound in non-immunocompromised patients with mild to moderately severe disease is safe, well-tolerated, and significantly reduces nasopharyngeal viral load. While it reduced both the mean time to viral clearance and the median time to targeted symptom alleviation by one day, this difference was not statistically significant. Rebound symptoms were mild, and there were no hospitalizations or deaths in either treatment group. Thus, there is no evidence of clinical benefit from retreatment with NR in this non-immunocompromised, largely vaccinated patient population with mild-moderate disease. Of note, however, is that a recent study suggests that prolonging initial therapy to 10 or 15 days may possibly reduce the risk of occurrence of virological rebound in immunocompromised patients.2
Stan Deresinski, MD, FACP, FIDSA, is Clinical Professor of Medicine, Stanford University.
References
1. Phan T, Ribeiro RM, Edelstein GE, et al. Modeling suggests SARS-CoV-2 rebound after nirmatrelvir-ritonavir treatment is driven by target cell preservation coupled with incomplete viral clearance. J Virol. 2025;99(3):e0162324.
2. Weinstein E, Paredes R, Gardner A, et al. Extended nirmatrelvir-ritonavir treatment durations for immunocompromised patients with COVID-19 (EPIC-IC): A placebo-controlled, randomised, double-blind, phase 2 trial. Lancet Infect Dis. 2025; Jul 14:S1473-3099(25)00221-X. [Online ahead of print].
ABSTRACT & COMMENTARY