Skip to main content

Detecting Rapid Spread of SARS-CoV-2 Variants, France, January 26-February 16, 2021

Our take —

This study described the spread of SARS-CoV-2 variants throughout France from late January to early February 2021. Variant spread in the region was more rapid than expected, with 50% higher rates of transmission when compared to wild-type strains. Most variant-positive samples were classified as B.1.1.7, and variants were identified more frequently in samples from younger persons tested in the community. This study demonstrates the effective use of targeted RT-PCR assays for variant detection. As these assays are cheaper and easier to deploy than other genomic surveillance methods (i.e., whole genome sequencing), they may provide a valuable approach for population-level monitoring of SARS-CoV-2 variants of concern.

Study design

Retrospective Cohort

Study population and setting

This study describes the rapid spread of SARS-CoV-2 variants of concern (largely B.1.1.7) in France in early 2021. Variant-specific RT-PCR assays (VirSNiP SARS-CoV-2 Spike del+501 and ID SARS-CoV-2/UK/SA Variant Triplex) can discriminate between variant and wild-type samples by using probes that target the variant-specific mutations delta 69–70 and N501Y. The 501Y.V2/B.1.351 and P.1 variants harbor the N501Y mutation, and both mutations are present in the B.1.1.7 variant; neither mutation is commonly present in wild-type samples. These assays were used to test unique SARS-CoV-2–positive samples (n=35,208) collected largely from community-based testing (92%) vs. hospital samples (7.9) in twelve regions in France between January 26 and February 16, 2021. From February 5 onward, all positive samples were evaluated for variants per the above description. All samples with the N501Y mutation were considered to be variant-positive. A logistic growth model was used to estimate variant transmission rates.

Summary of Main Findings

The N501Y mutation was detected in 10,212 (29%) samples, which were classified as variant-positive. Results for 6,702 (19%) samples were considered uninterpretable (due to amplification failure for the positive control); these samples were treated as missing. Positive variant identification was significantly associated with younger age, later sampling date, and community(vs. hospital) samples. Logistic growth models indicated that the variant strains were 50% more transmissible when compared to wild-type.

Study Strengths

Unlike other studies relying on S-gene target failure (SGTF) to identify variant samples, this study employs targeted RT-PCR assays to positively identify variant samples. Next generation sequencing data was consistent with assay results for a subset of samples with confirmatory testing (n=48).


Samples were analyzed retrospectively, which may introduce bias if variant-associated transmission networks were oversampled. Also, not all regions were represented equally in the analysis. A large number of samples (n=5,569) were excluded prior to the analysis because data were missing or age was outside the target range of 5-80 years; additional samples (n=6,702) were excluded due to uninterpretable assay results. The high proportion of missing data may have also introduced sample bias. Associations and growth models were determined using aggregate data for all variant samples, rather than according to specific variant strains.

Value added

This study demonstrates the utility of variant-specific RT-PCR assays for monitoring the spread of SARS-CoV-2 variants of concern. This methodology is simpler and less expensive than whole genome sequencing, but more specific than SGTF.

This review was posted on: 23 April 2021