ABSTRACT
The modern response to pandemics, critical for effective public health measures, is shaped by the availability and integration of diverse epidemiological outbreak data. Genomic surveillance has come to the forefront during the coronavirus disease 2019 (COVID-19) pandemic at both local and global scales to identify variants of concern. Tracking variants of concern (VOC) is integral to understanding the evolution of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in space and time. Combining phylogenetics with epidemiological data like case incidence, spatial spread, and transmission dynamics generates actionable information. Here we discuss the genome surveillance done in Pune, India, through sequencing 10,496 samples from infected individuals and integrating them with multiple heterogeneous outbreak data. The rise and fall of VOCs along with shifting transmission dynamics in the time interval of December 2020 to March 2022 was identified. Population-based estimates of the proportion of circulating variants indicated the second and third peak of infection in Pune to be driven by VOCs Kappa (B.1.617.1), Delta (B.1.617.2), and Omicron (B.1.1.529) respectively. Integrating single nucleotide polymorphism changes across all sequenced genomes identified C (Cytosine) > T (Thymine) and G (Guanine) > T (Thymine) substitutions to dominate with higher rates of adaptive evolution in Spike (S), RNA-dependent RNA polymerase (RdRp), and Nucleocapsid (N) genes. Spike Protein mutational profiling during and pre-Omicron VOCs indicated differential rank ordering of high-frequency mutations in specific domains that increased the charge and binding properties of the protein. Time-resolved phylogenetic analysis of Omicron sub-lineages identified specific recombinant X lineages, XZ, XQ, and XM. BA.1 from Pune was found to be highly divergent by global sequence alignment and hierarchical clustering. Our ''band of five'' outbreak data analytics which includes the integration of five heterogeneous data types indicates that a strong surveillance system with comprehensive high-quality metadata was critical to understand the spatiotemporal evolution of the SARS-CoV-2 genome in Pune. We anticipate the use of such integrated workflows to be critical for pandemic preparedness in the future.
Subject(s)
Coronavirus Infections , COVID-19ABSTRACT
Background The recent second wave in India in April-May 2021 placed an unprecedented burden on the Indian health systems. However, limited data exist on the epidemiology of the COVID-19 pandemic from the first wave through the second wave in India. With detailed epidemiologic data, we aimed to assess trends in incident cases and case fatality, its risk between pandemic waves in Pune, an epicenter of COVID-19 cases in India, a country with the second-largest absolute burden worldwide. Methods Programmatic COVID-19 data from Pune city between the first wave (March 09 th 2020-October 31 st , 2020), maintenance phase (November 01 st 2020-February 14 th , 2021), the second wave (February 15 th , 2021-May 31 st , 2021) were assessed for trends of incident cases, time-to-death, and case fatality rate (CFR). In addition, Poisson regression models adjusted for age and gender were used to determine the independent effect of pandemic waves on mortality. Results Of 465,192 COVID-19 cases, 162,182 (35%) were reported in the first wave, and 4,146 (2.5%) died among them; Maintenance period registered 27,517 (6%) cases with 590 (2.1%) deaths; Second wave reported 275,493 (59%) cases and 3184 (1.1%) deaths (p<0.01). The overall CFR was 1.16 per 1000 person-days (PD), which declined from 1.80 per 1000 PD during the first wave to 0.77 per 1000 PD in the second wave. The risk of death was 1.49 times higher during the first wave (adjusted case fatality rate ratio-aCFRR,1.49; 95% CI: 1.37–1.62) and 35% lower in the second wave (aCFRR, 0.65; 95% CI: 0.59 – 0.70), compared to the maintenance phase. Interpretation The absolute burden of COVID-19 cases and deaths were more significant in the second wave in Pune, India; however, the CFR declined as the pandemic progressed. Nevertheless, investigating newer therapies and implementing mass vaccinations against COVID-19 are urgently needed.
Subject(s)
COVID-19ABSTRACT
BackgroundReal-world data assessing the impact of lockdowns on COVID-19 cases remain limited from resource-limited settings. We examined growth of incident confirmed COVID-19 cases before, during and after lockdowns in Pune, a city in western India with 3.1 million population that reported the largest COVID-19 burden at the peak of the pandemic. MethodsUsing anonymized individual-level data captured by Punes public health surveillance program between February 1st and September 15th 2020, we assessed weekly incident COVID-19 cases, infection rates, and epidemic curves by lockdown status (overall and by sex, age, and population density) and modelled the natural epidemic using the 9-compartmental model INDSCI-SIM. Effect of lockdown on incident cases was assessed using multilevel Poisson regression. We used geospatial mapping to characterize regional spread. FindingsOf 241,629 persons tested for SARS-CoV-2, the COVID-19 disease rate was 267.0 (95% CI 265.3 - 268.8) per 1000 persons. Epidemic curves and geospatial mapping showed delayed peak of the cases by approximately 8 weeks during the lockdowns as compared to modelled natural epidemic. Compared to a subsequent unlocking period, incident COVID-19 cases 43% lower (IRR 0.57, 95% CI 0.53 - 0.62) during Indias nationwide lockdown and 22% (IRR 0.78, 95% CI 0.73 - 0.84) during Punes regional lockdown and was uniform across age groups and population densities. ConclusionLockdowns slowed the growth of COVID-19 cases in population dense, urban region in India. Additional analysis from rural and semi-rural regions of India and other resource-limited settings are needed.