This was a single-centre, retrospective, observational study of outpatients with a confirmed diagnosis of COVID-19, referred to Ospedale Luigi Sacco, in Milan from December 2021 until March 2023.

Medical records of adult patients diagnosed with COVID-19 through a SARS-CoV-2 positive RT-PCR or a rapid antigen test from nasal swabs and consequently evaluated for administration of early treatments, were pseudo-anonymized and abstracted on standardized data collection forms.

All participants signed informed written consent, and the study was approved by the Ospedale Luigi Sacco Institutional Review Board (n.prot 2020/ST/049).

We extracted the demographic, virological, and clinical data of all COVID-19 patients eligible for early treatments according to the Italian Medicines Agency recommendations [5].

Specifically, these patients had paucisymptomatic COVID-19, namely not-hospitalized and without any oxygen requirement. Patients were included if they had both mild COVID-19 (symptoms of SARS-CoV-2 infection but without pneumonia and without requiring oxygen-therapy) and fitted in the AIFA (Agenzia Italiana del Farmaco) definition of being at high-risk for progression into the severe disease. Specifically, they must have at least one of the following conditions: age over 65 years old, presence of solid or haematological cancer, chronic kidney disease, chronic liver disease, chronic lung disease, uncontrolled diabetes, neurological disease, cardiovascular disease, mental health condition, obesity, cerebrovascular disease and being immunocompromised (AIDS, solid organ or blood stem cell transplantation, and all those conditions requiring the use of corticosteroids or other immunosuppressive medications). [https://www.aifa.gov.it/emergenza-covid-19].

Patients were followed up for 28 days after the end of the drug administration to determine whether any side effects had occurred and whether they had been hospitalized for COVID-19.

Virological data consisted of the date of the first positive and first negative diagnostic test for COVID-19, and the genotyping of the SARS-CoV-2 isolates.

Clinical data comprised the onset symptoms, the vaccination status, and the underlying comorbidities.

Treatment data concerned which antiviral (Remdesivir, Nirmatrelvir/Ritonavir, Molnupiravir) or mABs treatment (Sotrovimab, Casirivimab/Imdevimab, Tixagevimab/Cilgavimab, Bamlanivimab/Etesevimab) was administered and any potentially related adverse event.

Treatments were administered according to the AIFA guidelines available at that time, and the decision to prescribe antivirals or monoclonal antibodies was a nuanced process, considering factors such as oral vs. intravenous administration, potential drug interactions, treatment efficacy, and patient convenience. We believe this approach allowed for tailored and patient-centered care based on the evolving circumstances of each case. [https://www.aifa.gov.it/emergenza-covid-19]

The primary outcome of this study was to evaluate the impact of mAbs and antiviral drugs on the length of SARS-CoV-2 viral shedding, identifying a potential drug which might more rapidly lead to a SARS-CoV-2 negative nasal swab.

Continuous variables were described as median and 25th and 75th percentile, while categorical variables as frequency and percentage.

The time of negativization of the SARS-CoV-2 nasal swab was calculated in days, as the difference between the first positive swab, executed at the onset of symptoms, and the first negative swab.

Correlation between days of negativization and continuous variables were investigated through Spearman’s correlation test, while differences between dichotomous categorical variables through Mann-Whitney U test.

The comparison of variables across treatments was conducted using the Kruskall-Wallis test and Chi-squared test with Yates’ continuity correction. Dunn’s test with Bonferroni adjustment was performed for post-hoc comparisons.

All mAbs were taken together in account as “mAbs group”, while antiviral drugs were individually considered.

A negative binomial regression adjusted for age, sex, total number of comorbidities, immunosuppression, and SARS-CoV-2 vaccination status was implemented.

Statistical significance threshold was set at 0.05.

Analyses were performed in R-Studio software (v. 4.2.3).