RSV, Influenza, and SARS-CoV-2 Outcomes in Severely Immunosuppressed Adults

Background and rationale. Respiratory syncytial virus (RSV) is increasingly recognized as a significant cause of morbidity and mortality in adults, particularly those over 65 years of age and those with underlying comorbidities. In immunocompromised adults, RSV infection can progress from an upper respiratory tract infection (URTI) to a lower respiratory tract infection (LRTI), with reported progression rates of 19%-36% depending on the degree of immunosuppression, presence of lymphopenia, and associated comorbidities. In recipients of allogeneic hematopoietic stem cell transplantation (HSCT), RSV-related LRTI mortality has been reported to range from 10% to 40%, particularly among patients with graft-versus-host disease and prolonged neutropenia. Among solid organ transplant recipients, complications appear less frequent overall, though lung transplant recipients and those experiencing acute rejection or graft dysfunction remain at elevated risk. Despite this burden, there is no antiviral treatment specifically approved for RSV in adults, and management remains largely supportive. Influenza and SARS-CoV-2 pose analogous risks in severely immunosuppressed populations, and the recent approval of RSV vaccines for adults over 60 and for adults aged 18-59 with immunocompromising conditions makes it clinically important to better characterize the burden and clinical course of these three viral infections in this population, to inform vaccination policy and clinical management. Setting. The study is conducted within the Hospital Italiano de Buenos Aires (HIBA) network, an integrated, high-complexity, university-affiliated healthcare system in the Autonomous City of Buenos Aires comprising a main campus and a second site (San Justo), with 789 beds (246 critical care beds), approximately 2.5 million outpatient visits per year, and roughly 41,000 discharges annually. All clinical care across the network is documented in a single, unified electronic health record (EHR) system, in which comorbidities and diagnoses are automatically mapped to the SNOMED controlled vocabulary through an institutional terminology server. Design and population. This is a retrospective cohort study of all adults with severe immunosuppression who had a laboratory-confirmed diagnosis of RSV, influenza, or SARS-CoV-2, obtained by multiplex PCR of a nasopharyngeal swab or bronchoalveolar lavage sample, between November 1, 2011 and March 20, 2020, and from May 1, 2022 to December 31, 2025, within the Hospital Italiano de Buenos Aires (HIBA) network. Enrollment was interrupted between March 20, 2020 and May 1, 2022, corresponding to the period of most significant restructuring of hospital services during the COVID-19 pandemic response. During this interval, admission thresholds and practice patterns changed substantially - at times nearly all diagnosed patients were hospitalized regardless of clinical severity - which would have introduced systematic bias into severity-based outcome measures such as the ordinal clinical status scale; this period was therefore excluded. The total observation window spans approximately 140 non-consecutive months. Severe immunosuppression is defined broadly and includes, among other conditions: active leukemia or lymphoma, disseminated malignancy, aplastic anemia, graft-versus-host disease, congenital immunodeficiencies, treatment with immune checkpoint inhibitors, solid organ transplantation on immunosuppressive therapy, hematopoietic stem cell transplantation within two years or ongoing immunosuppressive therapy, high-dose systemic corticosteroids, alkylating agents or antimetabolites at specified thresholds, TNF-alpha inhibitors, lymphocyte-depleting or B-cell depleting biologic agents, and HIV infection with a CD4 count below 200 cells/mm³, an AIDS diagnosis, or symptomatic HIV. Diagnostic methods for viral detection evolved over the study period, from an in-house multiplex PCR platform with capillary electrophoresis (validated in November 2011, detecting a broad panel of respiratory viruses including RSV) to a commercial FTD Respiratory Pathogens 21 panel used in 2019, and the BIOFIRE FILMARRAY Respiratory 2.1 Panel implemented from January 2020 onward. Patients vaccinated against RSV prior to their acute respiratory infection episode are excluded from the cohort (applicable to the later study period only, coinciding with RSV vaccine availability in Argentina); RSV vaccination status is recorded to enable transparent reporting of this exclusion. Data source and handling. All data will be extracted from the institutional EHR using automated queries, without manual chart abstraction, and will be fully de-identified before analysis. The resulting dataset will be stored on a password-protected computer accessible only to the study's principal investigator and will be permanently deleted upon completion of the statistical analysis. Planned analysis. Baseline characteristics will be summarized using medians and interquartile ranges for continuous variables and absolute/relative frequencies for categorical variables. The association between candidate predictors and dichotomous outcomes (progression from URTI to LRTI, ICU admission, and 30-day mortality) will be assessed using bivariate and multivariable logistic regression, reporting odds ratios with 95% confidence intervals. An ordinal logistic regression model for proportional odds will be used to assess predictors of the ordinal clinical status outcome at days 0, 7, and 30, with the proportional odds assumption tested using the Brant test; models for clinical status at days 7 and 30 will be adjusted for baseline status. Each virus (RSV, influenza, and SARS-CoV-2) will be analyzed using a separate model; no formal statistical comparison between viruses will be performed. While the primary follow-up window for clinical status, hospitalization, and mortality outcomes is 30 days, patients will be followed for up to 366 days to ascertain major cardiovascular events (acute coronary syndrome or cerebrovascular accident) as a secondary outcome. A p-value below 0.05 will be considered statistically significant. All analyses will be performed using STATA version 16 MP. Ethical considerations. As a minimal-risk, retrospective study using fully de-identified data with no patient contact and no manual chart review, the study was approved with a waiver of informed consent by the Hospital Italiano de Buenos Aires Research Ethics Committee (Comité de Ética en Protocolos de Investigación, CEPI; approval No. 7738), consistent with CIOMS 2019 Guideline 10 on modifications and waivers of informed consent. The study is conducted in accordance with Argentine Ministry of Health Resolution 1480/2011, Law 3301/09 of the City of Buenos Aires, the Declaration of Helsinki, the ICH E6 Good Clinical Practice guidelines, and Argentine data protection law (Law 25.326, Habeas Data). The study is conducted under a research agreement with, and is funded by, Pfizer (NISAMR 2025010848).