In this study, we combined prospective genomic surveillance with historical datasets to characterize long-term changes in the population structure and antimicrobial resistance profiles of S. Typhi and S. Paratyphi A in Nepal, and to assess how the size and composition of these bacterial populations evolved across three key epidemiological periods: before the COVID-19 pandemic, during the pandemic, and following the introduction of typhoid conjugate vaccines (TCVs). Our findings reveal a marked recent decline in the effective population size of both pathogens, lineage-specific compositional shifts, and persistently high levels of fluoroquinolone resistance. Together, these results provide new insight into how major public health interventions and large-scale social disruptions may influence patterns of typhoidal Salmonella circulation in endemic settings.

Phylodynamic analyses revealed a consistent pattern of population expansion followed by decline across major lineages, but the timing of these declines was asymmetric in a way that allows the contributions of the pandemic and TCV introduction to be partially separated. S. Typhi H58 contracted sharply during the pandemic and remained depressed thereafter, whereas lineage 3.3 was nearly flat during the pandemic and reached its steepest decline only post-TCV. S. Paratyphi A, which is not targeted by the vaccine, showed only a modest pandemic-era decline and no further change post-TCV, providing a within-study negative control for vaccine-associated effects. This pattern was independently corroborated by classical population-genetic statistics: nucleotide diversity in lineage 3.3 dropped 36% post-TCV, H58 equilibrated after its pandemic-era contraction, and S. Paratyphi A diversified rather than contracting. Together, these three independent genomic lines of evidence (per-lineage growth-rate trajectories, cross-pathogen contrasts, and population-genetic diversity statistics) suggest that the pandemic-era decline was most pronounced in the most actively transmitting lineage (H58), whereas the post-TCV decline was concentrated in lineage 3.3. Although observational genomic data cannot establish causality directly, the contrasting trajectories between vaccine-targeted S. Typhi lineages and non-targeted S. Paratyphi A are consistent with TCV contributing to the observed post-2022 lineage 3.3 contraction.

These trends are consistent with emerging epidemiological data from Nepal and other endemic settings documenting substantial reductions in typhoid incidence during the COVID-19 pandemic [9,10,35,36]. The post-vaccine period, however, was more heterogeneous: clinical surveillance in Nepal documented continued declines among vaccine-eligible children but a concurrent rise among adults, who comprise the majority of cases, such that overall typhoid incidence rose following 2022 [10]. Our findings extend these observations by demonstrating parallel declines in the effective population size of both S. Typhi and S. Paratyphi A, providing genomic evidence supporting a true reduction in pathogen circulation rather than changes in surveillance or healthcare utilization alone. The post-TCV genotype shift observed among unvaccinated adults represents an indirect, herd-mediated signal of vaccine impact, in which the genotype pool to which the unvaccinated adult population is exposed is reshaped by the suppression of transmission in the vaccinated paediatric age group [6,7]. Our findings are consistent with the published TCV-effectiveness studies. Randomized trials in Malawi [6] and Bangladesh [7], a programmatic evaluation in Navi Mumbai [8], and a recent real-world meta-analysis [9] have all documented substantial reductions in paediatric typhoid following TCV introduction, while parallel genomic surveillance from Zimbabwe has shown contraction of the dominant H58 lineages after vaccine rollout [35]. Our data extend this landscape by demonstrating that, even when TCV was deployed immediately after a major pandemic-era contraction, an additional post-TCV signal remained detectable at the genomic level, and was lineage-specific, supporting the generalizability of TCV impact across diverse endemic backgrounds.

The decline in S. Paratyphi A, which is not targeted by current vaccines [11], suggests that factors beyond vaccination likely contributed to reduced transmission. These include pandemic-associated behavioral and environmental changes, such as reduced population mobility, altered healthcare-seeking behavior, and improvements in hygiene practices during the COVID-19 pandemic [3]. A genotype-level analysis restricted to Kathmandu, where most isolates were sampled, further suggested a marked shift in the S. Paratyphi A genotype mix in the post-TCV period, with the historically dominant genotype 2.4.3 contracting and minor genotypes accounting for the majority of recent isolates. Whether this reflects a true compositional turnover or sampling variation in a declining population will require longer follow-up. However, we did not observe clear evidence of serovar replacement, as the overall S. Paratyphi A population itself also contracted, indicating that vaccination has not produced an immediate expansion of the non-vaccine serovar. Sustained circulation of fluoroquinolone-non-susceptible S. Paratyphi A raises concern that successful typhoid control could progressively unmask paratyphoid fever as a proportionally larger enteric fever burden. Continued surveillance will be important to monitor longer-term changes in serovar distribution and to inform the case for a bivalent conjugate vaccine targeting both serovars [11].

A central concern when introducing a polysaccharide-conjugate vaccine into an endemic setting is the possibility of vaccine-driven escape at the antigenic target [6,7]. We found no non-synonymous variant exclusive to post-TCV isolates and no confirmed Vi-loss isolates in any setting. The absence of detectable vaccine escape across endemic backgrounds, together with the strict monophyly and 2008 origin of the high-level fluoroquinolone-resistant clade, provides reassurance that the post-TCV residual transmission reflects local persistence of the existing population under vaccine pressure rather than the emergence of an antigenically distinct escape lineage.

The results of this study should be interpreted within the context of some limitations. First, the number of isolates collected during the pandemic period was relatively small, which may limit the precision of temporal inferences. Second, our dataset is largely derived from population-based surveillance in specific geographic regions, which may not fully capture transmission dynamics across Nepal. Third, district-stratified analyses are anchored on Kathmandu, which dominates the cohort; signals in peri-urban and rural districts rest on small numbers of isolates, particularly in the post-TCV period for S. Paratyphi A. Fourth, phylodynamic estimates of effective population size are influenced by sampling density and may not directly reflect true incidence. Finally, although our cross-pathogen contrasts using S. Paratyphi A as a within-study negative control provide a partial causal anchor for the asymmetric vaccine-versus-pandemic interpretation, formal causal inference will require longer post-TCV follow-up and ideally individual-level vaccine histories.

In conclusion, our integrated genomic and phylodynamic analysis demonstrates that both S. Typhi and S. Paratyphi A populations in Nepal have undergone substantial recent declines in effective population size, alongside lineage-specific compositional shifts, and widespread fluoroquinolone resistance. These findings highlight the potential impact of vaccination and large-scale social disruptions on enteric fever epidemiology and underscore the importance of sustained genomic surveillance. Expanding vaccine strategies to include S. Paratyphi A and strengthening antimicrobial stewardship can contribute importantly to long-term control of enteric fever.