Prolonged viral shedding of SARS-CoV-2 in two immunocompromised patients, a case report

The need to determine the duration of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) shedding in infected individuals and thereby estimate the period of infectiousness is of vital importance in containing the coronavirus disease 2019 (COVID-19) pandemic.

Measured by real time reverse transcription polymerase chain reaction (RT-PCR), infected patients shed virus for up to 83 days after symptom onset, but replicative virus were not detected by culture after 9 days of symptoms [1]. A study including 21 hospitalized COVID-19 patients showed that the median time from symptom onset to viral clearance in culture was 7 days [2]. These findings underlie the isolation guidelines used by the World Health Organization (WHO) which state that SARS-CoV-2 infected persons can discontinue isolation 10 days after symptom onset plus at least 3 additional days without symptoms [3]. Regarding asymptomatic cases with SARS-CoV-2 infection, the duration of infectiousness is still uncertain. Asymptomatic patients appear to shed virus for longer periods than symptomatic patients, measured by real time RT-PCR, and this may reflect a weaker degree of immune activation [4]. Jiang et al. [5] reported shortened IgG sero-conversion times among asymptomatic compared to symptomatic patients, whereas Shirin et al. reported the opposite [6]. These deviating findings could be attributable to differences in study groups, methods of symptoms recollection, viral loads and sensitivity of immune assays or underlying immunity. Immune markers might be useful as determinants for prolonged SARS-CoV-2 shedding. A limited number of case reports indicate protracted viral shedding in immunocompromised patients observed in both real time RT-PCR and in viral culture [7‐9] but these studies provided no detailed description of cellular immune status.

Here we report two cases of immunocompromised patients in which one patient had a severe course of COVID-19 whereas the other patient was asymptomatic.

For detailed description of laboratory methods, see Additional file 3.

Despite the clinical differences in these two immunocompromised patients, both continuously shed SARS-CoV-2 as measured by real time RT-PCR and had high viral loads in culture. To our knowledge, this is the first report describing the duration and amount of viable virus in an asymptomatic immunocompromised adult patient infected with SARS-CoV-2. Furthermore, not many studies have quantified viable virus. The significance of the amount of viable virus is still uncertain, but it must be assumed that higher viral load will mean greater infectivity.

Patient 1 shed viable virus until day 13 after symptom onset and 25 days after the first positive real time RT-PCR test. The real time RT-PCR was positive as far as 42 days after first positive test and remained positive during admission. This emphasizes that a positive real time RT-PCR not necessarily reflects replicative virus with implications for the estimation of SARS-CoV-2 transmissivity especially in immunocompromised patients. The asymptomatic patient had viable virus 12 days after initial positive real time RT-PCR. At that time point, he was highly contagious based on plaque assay analysis showing a high viral load of 55,944 PFU/swab. According to WHO, asymptomatic patients can discontinue isolation after 10 days of isolation from the first positive test [3]. This case shows that the immunocompromised patient, though asymptomatic, can potentially shed viable SARS-CoV-2 after 10 days and therefore this patient deviates from the guidelines.

Whole genome sequencing showed two different virus lineages and whether this can explain the different clinical courses of COVID-19 is unknown.

As IgG seroconversion times in asymptomatic/mildly symptomatic COVID-19 patients diverge [5, 6], cellular immune markers may represent underlying determinants for seroconversion and viral shedding. Both patients displayed markedly reduced fractions of allo-responsive CD4+ T cells. Allo-responsive T cells are dependent upon cognate TCR-MHC-peptide interactions as is the generation of cT_FH [10]. Patient 1 received a T cell repressive regimen while patient 2 was lymphopenic and received lenalidomid which suppresses CD4+ T cell proliferation in-vitro [11]. Patient 2 also suffered from a disseminated Nocardia infection, indicating compromised CD4+ T cell immunity. Hence, both patients reduced allo-reactive CD4+ T cell fractions were consistent with their impaired cT_FH formation versus that of an immunocompetent COVID-19 patient (day 20: 7.14%) [12]. CT_FH are strong indicators for the generation of antibodies to protein antigens [10]. Expansion of viral sensitive CD8+ CD38+ HLA-DR+ T cells [12] was observed in both patients, consistent with their preserved fractions of allo-reactive CD8+ T cells [Additional file 2 Table 1]. Frequencies of ASC were slightly elevated in patient 2 with myeloma but still lower than the ASC frequencies reported in symptomatic COVID-19, also during convalescence [12].

However, the surface markers CD27+, CD19+, and CD38+ are phenotypic characteristics associated with myeloma [13, 14], thereby questioning the immunological relevance of patient 2’s increased ASC frequencies. Compromised humoral immunity also characterizes myeloma patients. The strength of this case report is the inclusion of analyzes and description of cellular immune markers in both the asymptomatic and the symptomatic immunocompromised patient infected with SARS-CoV_2. Furthermore, the quantification of viable virus with plaque assay gives a more accurate picture of the duration of infectiveness. As a limitation, whole genome sequencing was only performed once during admission for both patients, and therefore potential mutations during hospitalization were not characterized.

In conclusion, both immunocompromised patients displayed marked expansions of virus reactive CD8+ T cells but cellular markers indicative of attenuated humoral immunity. Both patients shed high amounts of viable virus, despite completely different clinical courses and our cases indicate the importance of individual considerations relating to isolation for immunocompromised patients.