Results A total of 12,675 sequenced cases were included in the analysis of which 11,621 had B.1.1.7 detected and 1,054 had B.1.617.2 detected. Characteristics of cases by variant are shown in table 1. Key differences with the B.1.617.2 variant include a higher proportion with a history of foreign travel, a higher proportion of cases in the most recent weeks (calendar weeks 17 and 18), a higher proportion of females and a higher proportion of cases in the North West region and in London, and a higher proportion in the ‘Indian or Indian British’ or ‘Any Other Asian background’ ethnic groups.
Among sequenced samples that were originally tested using the TaqPath assay, there was a high correlation between S-gene target status and the two variants under investigation with 87.5% of S-gene positive cases identified as B.1.617.2 and 99.7% of S-gene target negative cases identified as B.1.1.7 (supplementary table 2 and table 3). Results of the TNCC analysis are shown in table 2. In the ‘any vaccine’ analysis, effectiveness was notably lower after 1 dose of vaccine with B.1.617.2 cases 33.5% (95%CI: 20.6 to 44.3) compared to B.1.1.7 cases 51.1% (95%CI: 47.3 to 54.7). Results for dose 1 were similar for both vaccines. Following dose 2, the reduction in vaccine effectiveness was much smaller and non-significant: 86.8% (95%CI: 83.1 to 89.6) with B.1.1.7 and 80.9 (70.7 to 87.6) with B.1.617.2. With BNT162b2 there was a small reduction in effectiveness post dose 2 from 93.4% (95%CI: 90.4 to 95.5) with B.1.1.7 to 87.9% (95%CI:78.2 to 93.2) with B.1.617.2. Numbers vaccinated with 2 doses of ChAdOx1 were smaller and the overall 2 dose vaccine effectiveness was lower than with BNT162b2 however the difference in vaccine effectiveness between B.1.1.7 and B.1.617.2 was small and non-significant: 66.1% (95% CI: 54.0 to 75.0) and 59.8% (95%CI: 28.9 to 77.3) respectively.
Table 3 shows the adjusted odds ratios for detection of B.1.617.2 relative to B.1.1.7 in vaccinated compared to unvaccinated individuals odds of cases having B.1.617.2 detected in vaccinated individuals was higher than in unvaccinated individuals for dose 1 of any vaccine (OR 1.38; 95% CI 1.10-1.72) and dose 2 of any vaccine (OR 1.60; 0.87-2.97). Given that vaccine effectiveness against symptomatic disease with B.1.1.7 is estimated at approximately 60% after dose 1 and 85% after dose 2 (10, 27) these results would indicate effectiveness of 45% and 76% respectively for B.1.617.2. By vaccine type the reduction in vaccine effectiveness appeared to be greater with ChAdOx1 (OR 1.48; 95%CI 1.18-1.87) than BNT162b2 (OR 1.17; 95%CI 0.82-1.67) though confidence intervals overlapped. The sensitivity analysis comparing to the 0 to 13 day post-dose 1 period gave a similar pattern of results though the odds ratios were smaller and not statistically significant (supplementary table 1). This was also the case with the matched analysis (supplementary table 2).
Discussion Main findings We found an absolute reduction of one dose vaccine effectiveness against symptomatic disease with the B.1.617.2 variant of approximately 20% when compared to the B.1.1.7 variant. However, reductions in vaccine effectiveness after two doses were very small. This was the case for both the BNT162b2 and ChAdOx1 vaccines. Using a TNCC analysis, estimated vaccine effectiveness against symptomatic disease with B.1.617.2 for a single dose of either vaccine is approximately 33%, for two doses of BNT162b2 is approximately 88% and for two doses of ChAdOx1 is approximately 60%.
Interpretation These findings suggest a modest reduction in vaccine effectiveness. Nevertheless, a clear effect of both vaccines was noted with high levels of effectiveness after two doses. Vaccine effects after two doses of ChAdOx1 vaccine were smaller than for BNT162b2 against either variant. This is consistent with reported clinical trial findings. However, rollout of second doses of ChAdOx1 was later than BNT162b2 and the difference may be explained by the limited follow-up after two doses of ChAdOx1 if it takes more than two weeks to reach maximum effectiveness with this vaccine. Consistent with this, 74% of those who had received 2 doses of ChAdOx1 had done so between 2 and 4 weeks prior to symptom onset compared to 46% with BNT162b2 (supplementary figure 1).
Numbers of cases and follow-up periods are currently insufficient to estimate effectiveness against severe disease, including hospitalisation and mortality, however, previous vaccine effectiveness estimates with other variants have shown higher levels of effectiveness against more severe outcomes (10, 14, 28). Therefore, higher levels of effectiveness against severe disease may be anticipated with the B.1.617.2 vaccine.
Comparison with other studies This is the first study that we are aware of to report on vaccine effectiveness against the B.1.617.2 variant. We were also unable to find any neutralisation data for this variant. One study from India has reported neutralisation data with the broader B.1.617 variant category suggests that both convalescent sera of COVID-19 cases and sera from recipients of the BBV152 (Covaxin) vaccine were able to neutralise B.1.617 (29). Assuming that a significant proportion of these were B.1.617.2 and that effectiveness the impact on different vaccines is similar, this would support our findings. Compared to recent findings from Qatar comparing the effectiveness of BNT162b2 against the B.1.1.7 and B1.351 variants, our findings would suggest that effectiveness against B.1.617.2 after a full course lies somewhere between these two (17)
Strengths and limitations The large scale of testing and whole genome sequencing in the UK as well as the recording of vaccination status in a national vaccination register has allowed us to analyse vaccine effectiveness within a few weeks of the variant first emerging in the UK. We use two distinct analytical approaches which give broadly similar results and findings with our control analysis (using B.1.1.7) are consistent with those previously reported (7, 8, 10, 17). Findings were also similar when comparing to the first 2 weeks post the first dose of vaccine (supplement), which helps to exclude unmeasured confounders associated with both the likelihood of being vaccinated and the likelihood of being exposed to a variant. Using a TNCC design also helped us to control for differences in health seeking behaviour between vaccinated and unvaccinated individuals.
There are also limitations to this study. These are observational findings and should be interpreted with caution. There may be factors that could increase the risk of COVID-19 in vaccinated individuals, for example if they adopt more risky behaviours following vaccination, however, this would be likely to affect analysis of both variants. Low sensitivity or specificity of PCR testing could also result in cases and controls being misclassified which would attenuate vaccine effectiveness estimates. This could affect one variant more than another though this might be expected to affect B.1.1.7 more than B.1.617.2 given that B.1.1.7 results in S-gene target failure. While we control for ethnicity, region and level of deprivation, differences in vaccine coverage in population groups that may have more or less exposure to B.1.617.2 may have affected our first analysis, but should not have affected the TNCC design. There may also be differences in the populations that received each vaccine, for example among younger age groups, more healthcare workers are likely to have received BNT162b2 whereas more individuals in clinical risk groups are likely to have received ChAdOx1, while we control for these factors in the analysis, we cannot exclude residual confounding (11). Timing of rollout of different vaccines varied, the reduced follow-up post two doses of ChAdOx1 could have attenuated these results, as discussed above. Furthermore, numbers who had received the Moderna mRNA-1273 vaccine were too small to be able to estimate vaccine effectiveness for this vaccine. As such, it is important that these findings are triangulated with emerging in vitro data on immune response in vaccinated individuals.
Conclusions Overall, we found high levels of vaccine effectiveness against symptomatic disease after two doses. These estimates were only modestly lower than vaccine effectiveness against the B.1.1.7 variant. It is likely that vaccine effectiveness against more severe disease outcomes will be greater. Our findings would support maximising vaccine uptake with two doses among vulnerable groups in the context of circulation of B.1.617.2.
The Pfizer vaccine is slightly less effective but appears to still protect against the more transmissible Indian strain of the virus that causes COVID-19, according to a study by France's Pasteur Institute.
"Despite slightly diminished efficacy, the Pfizer vaccine probably protects" against the Indian variant, according to laboratory test results, said Olivier Schwartz, the institute's director and co-author of the study that was published on the BioRxiv website ahead of peer review.
The study sampled 28 healthcare workers in the city of Orleans. Sixteen of them had received two doses of the Pfizer vaccine, while 12 had received one dose of the AstraZeneca vaccine.
People who had received two doses of Pfizer saw a three-fold reduction in their antibodies against the Indian variant, B.1.617, according to the study, but were still protected.
"The situation was different with the AstraZeneca vaccine, which induced particularly low levels of antibodies neutralising" the Indian variant, the study said.
Patients who had had COVID-19 within the past year and people vaccinated with two doses of Pfizer retained enough antibodies to be protected against the Indian variant, but three to six times less antibodies than against the UK variant, Schwartz said.
The study shows that "this variant.. has acquired partial resistance to antibodies," Schwartz said.
Since first emerging in late 2019 in China, the SARS-CoV-2 virus that causes COVID-19 has developed several variants, usually named for the places where it first appeared including the so-called South Africa and UK strains.
The variant first detected in India appears to be much more transmissible than earlier variations.
It has now been officially recorded in 53 territories, according to a World Health Organization report.
To try to curb its spread, France and Germany have re-introduced tighter rules on arrivals from affected countries, including the United Kingdom.
More information: Delphine Planas et al, Reduced sensitivity of infectious SARS-CoV-2 variant B.1.617.2 to monoclonal antibodies and sera from convalescent and vaccinated individuals, BioRxiv (2021). DOI: 10.1101/2021.05.26.445838
The B.1.617.2 coronavirus variant originally discovered in India last December has now become one the most — if not the most — worrisome strain of the coronavirus circulating globally. Recent research suggests it may the most transmissible variant yet and has fueled numerous waves of the pandemic around the world. B.1.617.2 has already spread to more than 60 countries, including the U.S., and undoubtedly contributed to the massive wave of cases that has inundated India in recent months. It also appears to have become the dominant strain infecting unvaccinated people in the U.K., and may be more likely to infect people who are only partially vaccinated than other strains. Below is what we know about B.1.617.2 — also known as the Delta variant.
How is B.1.617.2 different from other variants, and why may it be more dangerous? The Delta variant has multiple mutations that appear to give it an advantage over other strains. The most important apparent advantage is that the mutations may make the strain more transmissible, which would also make it the most dangerous variant yet. One study indicated B.1.617.2 may be up to 50 percent more transmissible than the B.1.1.7 (U.K./Alpha) variant, and B.1.1.7 is more transmissible than the original strain of the coronavirus, which emerged in China in late 2019.
The bottom line is that if this preliminary research is accurate, the Delta variant may soon become the most dominant COVID strain in the world and lead to rapid outbreaks in countries without high vaccination rates.
There are, so far, no indications that the Delta variant causes more severe illness than other variants, but again, more research is needed.
This chart, created by cardiologist Eric Topol, provides a simple breakdown of how B.1.617.2 compares to other variants of concern:
Why is the B.1.617.2 variant now being called ‘Delta’? On May 31, the World Health Organization announced that it would give new designations to COVID variants of concern using the Greek alphabet, both because of confusion over the “alphabet soup” names currently in wide use and to prevent variants from being referred to based on where they were first discovered (i.e., the U.K., South Africa, or India variants), a practice that runs the risk of creating harmful stigmas about specific countries and that may become confusing if more than one variant of concern originates in a single country. The WHO has designated the B.1.617.2 variant as Delta.
Scientists will continue to use the more complicated alphanumeric names for variants, as they always have, but the WHO hopes that the Greek-letter-based names will become the widely used ones among nonscientists.
Vaccines appear to be slightly less effective Recent research by the U.K. government has found that full vaccination is still largely effective against the Delta strain but may slightly be less effective than against other variants, and even less so after only one dose. The research found that two doses of a COVID vaccine provided 81 percent protection against the B.1.617.2 variant (compared with 87 percent protection against the B.1.1.7 variant). One dose only provided 33 percent protection against symptomatic infection from B.1.617.2 (compared with 51 percent protection against B.1.1.7). That means, according to a Financial Times analysis, that a single dose is 35 percent less effective against B.1.617.2 than it is against B.1.1.7.
If that is accurate, it means that B.1.617.2 may be the variant that currently poses the biggest threat to partially vaccinated populations worldwide.
Again, as with every known variant, full vaccination works against the Delta strain — but there are still signs that the variant marks a worrisome evolution in the coronavirus, and it seems likely it could raise the stakes for countries that continue to struggle to vaccinate their populations.
The strain has already become dominant in the U.K. and may threaten the end of its lockdown As of last week, B.1.617.2 accounted for up to 75 percent of all new cases in the U.K., indicating it is outcompeting the B.1.1.7 (U.K.) strain — now also known as the Alpha strain — and has become the dominant variant in the country.
The rapid spread of B.1.617.2 is prompting warnings that a third wave may already be under way in the country among those who remain unvaccinated. The rise of the variant has led the U.K. to attempt to speed up its vaccination campaign, particularly the second doses that half of U.K. adults still have not received.
Data released on May 27 indicated that nearly 7,000 cases of the B.1.617.2 variant have been confirmed in the U.K., more than double the number that had been confirmed the previous week. U.K. public-health officials have also acknowledged that the increase in cases may be linked to increased testing.
Prime Minister Boris Johnson has already said that the country “may need to wait” to lift its lockdown as planned on June 21, though he has maintained there is not enough evidence to support that move just yet. Some scientists are warning that such a delay must remain an option, informed by evidence rather than a predesignated date.
The good news is twofold. First, because about 44 percent of the U.K. population is fully vaccinated, the number of people the B.1.617.2 strain is infecting remains small and the country is not seeing a surge in hospital admissions. Second, because the U.K. is a world leader in the genome sequencing of variants, it is providing the clearest picture yet of B.1.617.2 capabilities — to the benefit of scientists worldwide.
Why is a more transmissible variant more dangerous? In a May 28 op-ed for the New York Times, Zeynep Tufekci succinctly broke down the threat of increased transmissibility:
A variant with higher transmissibility is a huge danger to people without immunity either from vaccination or prior infection, even if the variant is no more deadly than previous versions of the virus. Residents of countries like Taiwan or Vietnam that had almost completely kept out the pandemic, and countries like India and Nepal that had fared relatively well until recently, have fairly little immunity, and are largely unvaccinated. A more transmissible variant can burn through such an immunologically naïve population very fast.
Increased transmissibility is an exponential threat. If a virus that could previously infect three people on average can now infect four, it looks like a small increase. Yet if you start with just two infected people in both scenarios, just 10 iterations later, the former will have caused about 40,000 cases while the latter will be more than 524,000, a nearly 13-fold difference.
This is why allowing the coronavirus to spread and evolve unchecked is so dangerous B.1.617.2 is yet more proof of both how SARS-CoV-2 continues to evolve and how that evolution is continuing to produce variants that are more dangerous than those that came before them. From the available evidence, B.1.617.2 may be the most transmissible variant to yet spread in the world, and thus poses the biggest risk to unvaccinated populations, and possibly also populations where most vaccine recipients have only received one dose. Scientists have good reasons to sound the alarm over it.
And the threat of any more dangerous COVID variant also raises the threat of more dangerous COVID variants which may evolve from it. Vietnam’s health ministry has announced that it has detected a variant which appears to be a hybrid of both B.1.1.7 and B.1.617.2 variants. The country has only been able to give at least one dose of a COVID vaccine to about 1 percent of its population thus far, leaving it highly vulnerable to the new variants despite faring much better than most of the world at preventing the spread of COVID-19. Now Vietnam is racing to do more testing to see how far the hybrid strain has spread and how it differs from its predecessors. (It should be noted that some scientists are urging restraint before jumping to any conclusions about how bad this — or any — new variant is.)
The best way to prevent new variants from evolving is to give the coronavirus fewer opportunities to evolve by preventing and containing outbreaks with effective precautions like face masks and proper ventilation, and by vaccinating people before they can be exposed to infection in the first place.