Rubella virus (RuV) has recently been found in association with granulomatous inflammation of the skin and several internal organs in patients with inborn errors of immunity (IEI). The cellular tropism and molecular mechanisms of RuV persistence and pathogenesis in select immunocompromised hosts are not clear. We provide clinical, immunological, virological, and histological data on a cohort of 28 patients with a broad spectrum of IEI and RuV-associated granulomas in skin and nine extracutaneous tissues to further delineate this relationship. Combined immunodeficiency was the most frequent diagnosis (67.8%) among patients. Patients with previously undocumented conditions, i.e., humoral immunodeficiencies, a secondary immunodeficiency, and a defect of innate immunity were identified as being susceptible to RuV-associated granulomas. Hematopoietic cell transplantation was the most successful treatment in this case series resulting in granuloma resolution; steroids, and TNF-α and IL-1R inhibitors were moderately effective. In addition to M2 macrophages, neutrophils were identified by immunohistochemical analysis as a novel cell type infected with RuV. Four patterns of RuV-associated granulomatous inflammation were classified based on the structural organization of granulomas and identity and location of cell types harboring RuV antigen. Identification of conditions that increase susceptibility to RuV-associated granulomas combined with structural characterization of the granulomas may lead to a better understanding of the pathogenesis of RuV-associated granulomas and discover new targets for therapeutic interventions.

A strong association between rubella virus (RuV) and chronic granulomas, in individuals with inborn errors of immunity, has been recently established. Both the RA27/3 vaccine and wild-type RuV strains were highly sensitive to a broad-spectrum antiviral drug, nitazoxanide (NTZ), in vitro. However, NTZ treatment, used as a salvage therapy, resulted in little or no improvements of RuV-associated cutaneous granulomas in patients. Here, we report investigations of possible causes of treatment failures in two ataxia-telangiectasia patients. Although a reduction in RuV RNA in skin lesions was detected by real-time RT-PCR, live immunodeficiency-related vaccine-derived rubella viruses (iVDRV) were recovered from granulomas, before and after the treatments. Tizoxanide, an active NTZ metabolite, inhibited replications of all iVDRVs in cultured A549 cells, but the 50% and 90% inhibitory concentrations were 10–40 times higher than those for the RA27/3 strain. There were no substantial differences in iVDRV sensitivities, neither before nor after treatments. Analysis of quasispecies in the E1 gene, a suspected NTZ target, showed no effect of NTZ treatments on quasispecies’ complexity in lesions. Thus, failures of NTZ therapies were likely due to low sensitivities of iVDRVs to the drug, and not related to the emergence of resistance, following long-term NTZ treatments.

A young man with X-linked severe combined immunodeficiency developed a persistent vaccine-derived rubella virus (VDRV) infection, with the emergence of cutaneous granulomas more than fifteen years after receipt of two doses of measles-mumps-rubella (MMR) vaccine. Following nasopharyngeal swab (NP) collection, VDRV was detected by real-time polymerase chain reaction (RT-qPCR) and sequencing, and live, replication-competent VDRV was isolated in cell culture. To assess duration and intensity of viral shedding, sequential respiratory samples, one cerebrospinal fluid sample, and two urine samples were collected over 15 months, and VDRV RNA was detected in all samples by RT-qPCR. Live VDRV was cultured from nine of the eleven respiratory specimens and from one urine specimen. To our knowledge, this was the first reported instance of VDRV cultured from respiratory specimens or from urine. To assess potential transmission to close contacts, NP specimens and sera were collected from all household contacts, all of whom were immunocompetent and previously vaccinated with MMR. VDRV RNA was not detected in any NP swabs from the contacts, nor did serologic investigations suggest VDRV transmission to any contacts. This report highlights the need to understand the prevalence and duration of VDRV shedding in granuloma patients and to estimate the risk of VDRV transmission to immune and non-immune contacts.