To the Editor—The current outbreak of COVID-19 in India, which started in early March 2021, has created a new world record even beyond the outbreaks in the UK, the United States and Brazil. Prior to March 2021, less than 0.7% of the Indian population was infected with COVID-19. This current second wave took only 2 months to infect an additional ~0.36% of the population, and India is now recording over 0.4 million new cases per day (as of 23 April 2021). The true number is probably even higher, with some estimates putting the number of daily new cases at over 1 million, more than five times the officially recorded number1.

The sudden surge in COVID-19 cases in India coincides with high prevalence of more-transmissible variants, associated with diagnostic test failures and antibody escape2. These coronavirus SARS-CoV-2 variants of concern—B.1.1.7 (501Y.V1), B.1.351 (501Y.V2) and B. (501Y.V3; also known as P.1)—were observed during the sudden surge in COVID-19 cases in the UK, South Africa and Brazil, respectively, with subsequent local transmission across the world2,3 (Fig. 1a).

Fig. 1: Distribution of emerging variants of concern and interest and COVID-19 burden in India.
figure 1

a, Global map indicating where local transmission of variants of concern (501Y.V1, 501Y.V2 and 501Y.V3; left key) has been reported (adapted from the Global Report Investigating Novel Coronavirus Haplotypes website ( The ‘pinned’ locations show nations that have reported the presence of variant of concern B.1.617 and variant of interest B.1.618 (right key), identified through sequence analysis. b, Apparent cumulative prevalence of various variants of concern (key) in India, presented as a 7-day rolling average of the percentage of total sequences; shading shows 95% confidence intervals. c, Estimated prevalence (bottom key) of variants of concern and the variant of interest B.1.618 (below plots) in various states of India. Data were plotted with a minimum size of 20 sequences within 130 days before 11 May 2021, with GISAID data10. Plots and maps in b,c adapted from the database3. d, Plot of the ratio of daily new recoveries to daily new cases, daily confirmed COVID-19 cases per million population and case-fatality rate for India (key; matches axis colors). Data source: Ministry of Health and Family Welfare, India, and COVID19 INDIA ( e, Share of the global population that has received all vaccination doses, as prescribed by national vaccination protocols. Data adapted from COVID-19 data of the Johns Hopkins University Center for Systems Science and Engineering8. f, Contour plot of the change in the current reproduction number (Rt) of SARS-CoV-2 (key) for all of India (In) and various Indian states (left margin), from January 2021 to April 2021: Uttar Pradesh (UP), Uttrakhand (Uk), Odisha (Od), Maharashtra (Mh), Kerala (Kl), Gujrat (Gj), Delhi (Dl), Bihar (Br) and Andhra Pradesh (AP). Data obtained from the Ministry of Health and Family Welfare for States and Union territories, adapted from the COVID-19 India-Timeline website ( Publ. note: Springer Nature is neutral about jurisdictional claims in maps.

In India, the frequency of 501Y.V1 is higher than that of 501Y.V2 and 501Y.V.3 (Fig. 1b). The recently designated variant of concern B.1.617 and variant of interest B.1.618 have also been gaining attention in India3 (Fig. 1a–c). Variant B.1.617.1 shows co-occurrence of three key mutations in sequence encoding the viral spike protein: L452R, E484Q and P681R. L452R raised concerns in the United States as part of the California variants B.1.427 and B.1.429 and conferred resistance to the neutralizing monoclonal antibodies X593 and P2B-2F64. E484Q shares antibody-escape features similar to those of mutation E484K, seen in variants 501Y.V2, 501Y.V3 and B.1.6182. P681R may enhance processing by host proteases by extending the polybasic ‘RRAR’ motif, which results in a greater viral load and the potential for increased transmission. Another similar variant of concern, B.1.617.2, with mutations L452R, T478K and P681R, is highly prevalent in the state of Gujrat in India3. There was enrichment for the mutation T478K or T478R when SARS-CoV-2 was subjected to weak neutralizing antibodies, which indicates this mutation may lead to antibody escape. Variant B.1.618, which has the E484K mutation, is prevalent in the state of West Bengal, India3 (Fig. 1c).

New variants are thought to be responsible for re-infections, either after natural infection or after vaccination, as observed in Brazil and the United States, respectively5,6. There is evidence that re-infections are already happening in India; a recent survey identified a re-infection proportion of 4.5% from a pool of 1,300 participants infected between January 2020 and October 2020 (ref. 7).

Another concern about the emergence of new variants is the potential failure of RT-PCR tests for diagnostics. Failure to target the gene encoding the spike protein was observed during detection of the 501Y.V1 variant in the UK2. According to the latest guidelines of the Indian Council of Medical Research, approved test kits employ multiplex RT-PCR assays, as tests assessing only the spike protein may fail and thereby underestimate the true number of cases.

During mid-January 2021, vaccination programs commenced in India with the BBV152 (Covaxin) and ChAdOx1 (Covishield) vaccines. Although India is administering more than 3 million single doses of vaccine per day, an amount similar to that in the United States and China, the share of the population fully vaccinated against COVID-19 remains much lower than the global average and remains lower than that in some other nations with a high COVID-19 burden, such as Brazil (Fig. 1e). Adding to these worries, the new variants of concern and variants of interest may be able to escape immunity, induced by natural infection or vaccines, although more data on this are needed2. The vaccines BBV152 (Covaxin), ChAdOx1 (Covishield) and Gam-COVID-Vac (Sputnik V), which is soon to be administered, have all shown effectiveness in neutralizing the 501Y.V1 (UK) variant. The current vaccines also seem to provide protection from serious disease and death, but not necessarily from infection8.

It is important to note that the current prevalence estimates for variants 501Y.V1, 501Y.V2, B.1.617 and B.1.618 are limited to just a few Indian states, as little or no sequencing data have been available from many Indian states, including those in the north and east of the country (Fig. 1c). Despite India’s having ten national centers of excellence for genetic sequencing, with capacity for over 30,000 sequences per month, the surveillance of SARS-CoV-2 variants has been slow and has involved only ~2,700 sequenced samples collected from eight states between January 2021 and April 2021 (ref. 3). The potential for new variants to increase transmission and reduce the effectiveness of vaccines2,6 means that it is crucial to scale up Indian sequencing efforts and characterize the distribution of variants across all states.

This is especially important because in some Indian states, the current reproduction number (Rt; January 2021 to April 2021) has increased to more than 2, compared with an Rt of 1.5 for all of India (Fig. 1f). Prioritizing mass immunization, along with variant-guided vaccination drives, which prioritize vaccination in the states reporting new variants and/or a high Rt, will do the most to reduce the impact of COVID-19, in terms of morbidity, mortality and the economy. Early talk of a K-shaped recovery, with growth in the gross domestic product of 4–5% in 2021, now seems far-fetched.

Confirmed daily new cases in India have spiked from 53 per million population (up to March 2021) to >200 per million population (after mid-March 2021) (Fig. 1d). A glimmer of hope lies in an apparently low case-fatality ratio of <1.5%, compared with >3% between July 2020 and August 2020 (Fig. 1d), although many researchers question the official fatality rates. In contrast, the ratio of daily recoveries to daily new cases from the beginning of March 2021 have dropped steeply, reflective of the severe burden of the outbreak on the healthcare infrastructure in India (Fig. 1d). Leading hospitals in Delhi, Mumbai, Bangalore, Hyderabad and Chennai have shortages of essential supplies, and over 130 Indian cities have reported increased numbers of deaths of patients with COVID-19 who are in intensive care, due to a shortage of medical-grade oxygen and life-saving medicines. India can learn from the responses to COVID-19 of other nations such as France, Germany, Spain, Canada and Italy, where the number of hospitalized critically ill patients during the second wave was almost double that in the first wave9.

If this second wave of COVID-19 in India continues unabated, daily deaths in India could escalate above 15,000 before the second wave subsides and could account for 1.49 million deaths by September 2021 (according to data from the Institute of Health Metrics and Evaluation: Rising infections, hospitalizations and deaths can be mitigated through mass vaccination, better management of health resources and appropriate non-pharmaceutical interventions. The experience of other countries suggests that, apart from new variants, human behavior will also be major issue to worry about in the pandemic.

India has a population of 1.4 billion people, so the large number of cases is helping to drive global cases to unprecedented heights. This outbreak can be tackled only by the leadership in India, which should suspend mass gatherings, increase the vaccination of its population, and prioritize viral genome sequencing for surveillance and community tracking.