Extended nirmatrelvir-ritonavir for persistent COVID-19: Systematic review with individual patient data.

OBJECTIVES

To summarize individual patient data (IPD) on extended-duration nirmatrelvir-ritonavir (NMV-r) for persistent SARS-CoV-2 infection in immunocompromised adults and describe outcomes by regimen.

METHODS

We conducted a systematic review with IPD synthesis (PRISMA-IPD; PROSPERO CRD42025642455), searching databases through December 2025 and restricting eligible reports to January 1, 2022, through December 31, 2025 (Omicron-era focus). IPD were obtained for 39 patients from four low-risk-of-bias cohort studies (n = 30) and institutional cases (n = 9) meeting predefined criteria for persistent infection. We summarized outcomes after last-line extended NMV-r and report exploratory, unadjusted subgroup descriptions for monotherapy (n = 27) and combination regimens (n = 12).

RESULTS

Median age was 63 years. Patients had prolonged viral replication (median 58 days) before receiving extended NMV-r (median 10 days). Persistent infection after last-line extended therapy occurred in 5/38 (13.2%) evaluable patients. Persistence was observed in 2/26 (7.7%) evaluable monotherapy and 3/12 (25.0%) combination-therapy recipients; because regimen selection was nonrandom and baseline risk differed between groups, these subgroup proportions are descriptive and should not be interpreted as comparative effectiveness. All-cause mortality and adverse events (AEs) were rare (each 1/39; 2.6%).

CONCLUSION

In this selected observational IPD cohort, clearance after last-line extended NMV-r-containing therapy was commonly reported, and serious AEs were uncommon. Comparative inferences are limited by small sample size, imprecision, and confounding by indication; prospective studies are needed. PROSPERO registration CRD42025642455.