Avian Flu Diary: EID: Highly Divergent SARS-CoV-2 Alpha Variant in Chronically Infected Immunocompromised Person

#16,931

Thousands of SARS-CoV-2 variants have been detected around the world since the virus emerged in 2019, but only a couple of dozen or so have been classified as VOIs (Variants of Interest), and fewer still achieved VOC (Variant of Concern) status.

Some variants, after initially raising concerns - have passed quietly into obscurity (e.g. Denmark's `mink variant') - unable to successfully compete against more aggressive VOCs like Delta and Omicron .

Other, more successful VOCs - like Alpha and Delta - reigned supreme for a time, only to be supplanted by newer, more `fit' Omicron variants.

But gone doesn't necessarily mean extinct.

Last February, in Preprint: Evolutionary Trajectories of SARS-CoV-2 Alpha and Delta Variants in White-Tailed Deer in Pennsylvania, we looked at a report finding that a highly divergent Alpha variant was still circulating in deer long after Alpha had been supplanted by Delta in the human population.

The authors wrote:

The alpha lineage persisted in deer after its displacement by delta in humans, and deer-derived alpha variants diverged significantly from those in humans, consistent with a distinctive evolutionary trajectory in deer.

(SNIP)

Our findings of alpha persistence in deer after replacement of alpha by delta in humans, and the divergence seen between our deer and human alpha genomes, are all consistent with long-term persistence and spread of the alpha variant in deer.

Since we've seen SARS-CoV-2 successfully spillover into other species (see here, here, and here), there may very well be other examples out there we aren't aware of.

Chinese researchers have continually warned that these types of spillover events are capable of dramatically altering the evolutionary trajectory of future COVID variants (see China CDC Weekly Perspective: The “Wolf” Is Indeed Coming: Recombinant “Deltacron” SARS-CoV-2 Detected).

But there is also a human reservoir capable of harboring older variants, although it doesn't appear to happen very often. Immunocompromised individuals have been known to carry the virus for weeks, and it has been widely suggested (see below) that these persistent infections may contribute to the creation of new variants.

A year ago the NEJM carried an article warning of the potential risk of new variants emerging from long-term infections among immunocompromised individuals (see SARS-CoV-2 Variants in Patients with Immunosuppression).

While in December of 2020, the ECDC Threat Assessment Brief On UK SARS-CoV-2 Variant speculated that the large number of concurrent mutations in the B.1.1.7 variant may have arisen from a single chronically-infected, likely immunocompromised, individual.

All of which brings us to a fascinating report in the EID Journal from the Erasmus Medical Center in the Netherlands on an immunocompromised patient who appears to have carried the Alpha variant for 42 weeks, testing positive for it long after the virus had been supplanted by Delta.

Follow the link to read the report in its entirety:

Research Letter
Highly Divergent SARS-CoV-2 Alpha Variant in Chronically Infected Immunocompromised Person

Bas B. Oude Munnink , Roel H.T. Nijhuis, Nathalie Worp, Marjan Boter, Babette Weller, Babs E. Verstrepen, Corine GeurtsvanKessel, Maarten L. Corsten, Anne Russcher, and Marion Koopmans

Abstract

We detected a highly divergent SARS-CoV-2 Alpha variant in an immunocompromised person several months after the latest detection of the Alpha variant in the Netherlands. The patient was infected for 42 weeks despite several treatment regimens and disappearance of most clinical symptoms. We identified several potential immune escape mutations in the spike protein.

Persons with an immune deficiency can be infected with viral pathogens for a prolonged period. This occurrence has been reported for noroviruses (1) but also has been documented for SARS-CoV-2 (2). We report a patient with type 2 diabetes mellitus and chronic lymphocytic leukemia who had been infected for 42 weeks with SARS-CoV-2. The patient was hospitalized on April 23, 2021, and received optiflow treatment with dexamethasone, tocilizumab, and remdesivir. After May 11, 2021, the patient recovered and experienced no residual symptoms. Almost 9 months later, on February 3, 2022, the patient was readmitted to the hospital for leukemia-related anemia and tested positive for SARS-CoV-2 once again. A month later, the patient died of causes unrelated to SARS-CoV-2.

Overall, the patient tested positive for SARS-CoV-2 in 6 nasal or pharyngeal swabs. We performed whole-genome sequencing on all specimens by using an amplicon-based sequencing approach, as previously described, with the updated ARTIC primers version 4.1 (ARTIC Network, https://artic.network/mcov-2019) (3). The sequencing was successful for 2 specimens from mid-2021 and 2 specimens from early 2022 (Table). Pangolin version 4.0.6 PLEARN-v1.8 classification using default settings demonstrated that all sequenced viruses belonged to the Alpha (B.1.1.7) variant of concern (VOC) (5), Nextclade version 1.14.1 strain 20I (6). The samples were run on flowcells containing 96 samples, including a positive and negative control (pangolin lineage B.1.77.50) to exclude potential contamination.

GISAID’s EpiCoV database (https://www.gisaid.orgExternal Link) showed that the latest isolate identified as Alpha in the Netherlands was collected on October 13, 2021, suggesting that the variant had not been circulating in the Netherlands since that time. Phylogenetic analysis by IQ-TREE (7) using all unique downsampled Alpha sequences available in GISAID (8) from the Netherlands showed that the viruses detected on January 31 and February 3, 2022, were identical but distinct from previously observed Alpha lineages in the Netherlands (Figure). A zoom-in of the phylogenetic tree showed that all sequences of the virus in the patient cluster together in a separate branch, suggesting that the patient was chronically infected with this specific variant of SARS-CoV-2 (Appendix Figure 1).

Over time, we identified 24 nucleotide mutations when we compared sequences from the earliest and latest timepoints. Of these mutations, 19 mutations were nonsynonymous, resulting in 13 amino acid mutations in open reading frame 1ab and 6 amino acid mutations in the spike protein (Table; Appendix Figure 2). Of the 6 mutations in the spike protein, 3 are located in the receptor-binding domain (G339D, N439K, and V483F), 2 are located in the N-terminal domain (W64R and G142V), and 1 is located in the transmembrane domain (P1263L). The mutation G339D can also be found in all Omicron VOCs.

G142V has coevolved independently in >1 immunocompromised person with a long-term Alpha variant infection (S.A.J. Wilkinson et al., unpub. data,(https://doi.org/10.1101/2022.03.02.22271697), and a mutation in the same position (G142D) has also been described in the Delta (B.1.617.2) and in all Omicron variants.

Our data imply that, despite receiving treatment with dexamethasone, tocilizumab, and remdesivir and being discharged without residual symptoms, the patient had not cleared the infection. Unfortunately, ex vivo rescuing of the viruses from the swabs to assess potential immune escape from circulating neutralizing antibodies was not successful, but some of the mutations we observed in this immunocompromised person with long-term SARS-CoV-2 infection could be linked to immune escape.

Previous studies suggest that the G339D mutation affects neutralization in a subset of neutralizing antibodies (9) and that the N439K mutation causes immune escape and enhances binding affinity for human angiotensin-converting enzyme 2 (10,11). In addition, the V483F mutation has been shown previously to cause immune escape (12).

The constellation of this particular set of mutations has not been found elsewhere yet despite active ongoing genomic surveillance, which indicates the virus did not spread in the population (Appendix Table). Nonetheless, the detection of an Alpha variant infection in an immunocompromised person in a time when Omicron was the primary circulating variant indicates that reinfection is unlikely, which is also supported by phylogenetic analysis.

This occurrence illustrates that this VOC did not completely disappear although it was last detected on October 13, 2021, in the Netherlands. In addition, several mutations were found that are also present in other VOCs, suggesting that VOCs might have emerged in long-term infected immunocompromised persons as suggested previously (13).

Our findings illustrate that in previously unidentified reservoirs, such as immunocompromised persons, virus variants might still be present even when these variants are regarded as extinct and are no longer circulating among the population. In addition, we show that several mutations associated with immune escape that maintain virulence and fitness can accumulate in such an immunocompromised person. Continuous genomic surveillance in long-term infected persons is essential to elucidate their potential role in the emergence of future VOCs.

Dr. Oude Munnink is a researcher at the Erasmus Medical Center. His research interests include genomic surveillance of genomic pathogens and viral zoonoses.

While this particular Alpha variant does not appear to have spread in the population, this is a reminder of the wide range of opportunities and tools SARS-CoV-2 has to reinvent itself.