Regina C. LaRocque’s research while affiliated with Harvard Medical School and other places

Mass oral cholera vaccination campaigns targeted at subnational areas with high incidence are central to global cholera elimination efforts. Serological surveillance offers a complementary approach to address gaps in clinical surveillance in these regions. However, similar immune responses from vaccination and infection can lead to overestimates of incidence of infection. To address this, we analyzed antibody dynamics in infected and vaccinated individuals to refine seroincidence estimation strategies for partially vaccinated populations. We tested 757 longitudinal serum samples from confirmed Vibrio cholerae O1 cases and uninfected contacts in Bangladesh as well as vaccinees from Bangladesh and Haiti, using a multiplex bead assay to measure IgG, IgM, and IgA binding to five cholera-specific antigens. Infection elicited stronger and broader antibody responses than vaccination, with rises in cholera toxin B-subunit (CTB) and toxin-coregulated pilus A (TcpA) antibodies uniquely associated with infection. Previously proposed random forest models frequently misclassified vaccinated individuals as recently infected (over 20% at some time points) during the first four months post-vaccination. To address this, we developed new random forest models incorporating vaccinee data, which kept false positive rates among vaccinated (1%) and unvaccinated (4%) individuals low without a significant loss in sensitivity. Simulated serosurveys demonstrated that unbiased seroincidence estimates could be achieved within 21 days of vaccination campaigns by ascertaining vaccination status of participants or applying updated models. These approaches to overcome biases in serological surveillance enable reliable seroincidence estimation even in areas with recent vaccination campaigns enhancing the utility of serological surveillance as an epidemiologic tool in moderate-to-high cholera incidence settings. Significance statement Serological surveillance can improve how we monitor cholera in high burden areas where clinical surveillance is limited. However, vaccination can produce immune responses similar to infection, leading to overestimates in seroincidence. This study extends seroincidence estimation techniques using machine learning models to partially vaccinated populations. We analyzed antibody dynamics from vaccinated and infected individuals to develop methods that reduce misclassification of vaccinated individuals as recently infected. These methods enable reliable seroincidence estimates in areas with recent vaccination campaigns, providing a step toward better epidemiologic monitoring in the context of global cholera control initiatives.

Background Between 200–300 severe malaria cases are diagnosed annually in the U.S. Intravenous artesunate (IVAS), the FDA-approved 1st-line therapy for severe malaria since 2020, is commercially available but at high cost. Uncertainty regarding where people with severe malaria will seek medical care poses supply chain challenges. We identified hospitals where people with severe malaria were treated and mapped proximity to Level 1 trauma centers as a proxy for hospitals that provide complex clinical care, which could assist with planning for IVAS distribution. Location identification and manual adjudication process for hospitals that hospitalized people with severe malaria as reported to CDC from 2012–2018. Of the original 1,912 cases of severe malaria reported to the CDC from 2012-2018, we excluded 379 cases that were not admitted or were admitted but missing a hospital name. Of the remaining 1,533 severe malaria cases that were eligible for adjudication, 1,237 were an exact match with an existing hospital location. Of the 296 cases that were not an exact match, we manually adjudicated 229 cases by clarifying a hospital name that was abbreviated, incomplete, or had a change in name; only 49 cases could not be identified given multiple possible sites with the same name, and 18 cases with no name that matched a hospital site. Six additional cases were excluded because of occurring at a site with no Level 1 trauma center within driving distance (i.e., Alaska and US Virgin Islands). Methods Using 2012–2018 data reported to the Centers for Disease Control and Prevention (CDC), we analyzed free text of U.S. hospital names where people were treated as inpatients for severe malaria. We identified these hospitals using Google Maps Geocoding Application Programming Interface (API) followed by manual adjudication. We excluded cases when data were incomplete or ambiguous or if a Level 1 trauma center was inaccessible (e.g., Alaska or U.S. Virgin Islands). We then mapped hospitals to U.S. counties and estimated travel times between hospitals and the nearest Level 1 trauma center as per 2014–2016, using Google Maps Distance Matrix API. Average numbers of severe malaria cases at identified hospitals per US county annually as reported to the CDC during 2012–2018. Panel A shows the geographic distribution and average number of severe malaria cases at identified hospitals per year by county reported to CDC during 2012–2018; darker color shows counties with more severe malaria cases reported. Panel B shows that the distribution is highly skewed to 2 counties that reported more than 10 inpatient cases annually; 2,824 counties (pale peach) reported no cases of severe malaria admitted to an identified hospital during 2012–2018. Results After excluding 379 cases with no recorded hospital name from 1,912 severe malaria cases, we successfully adjudicated hospitals for 1,460/1,533 cases (Fig 1). Two counties reported >10 inpatient cases per year (Bronx County, NY and New York County, NY), and 2,824 counties never reported a hospitalized severe malaria case (Fig 2). Almost 35% of inpatient cases occurred at Level 1 trauma centers (508/1,460 [34.8%]) (Fig 3). Of the 952/1,460 (65.2%) inpatient cases not at Level 1 trauma centers, 891/952 (93.5%) were at hospitals located < 2h estimated driving time of a Level 1 trauma center. Only 61/1,460 (4.2%) cases were at locations >2h from a Level 1 trauma center. Limitations include that the analysis did not consider the original site of clinical presentation with malaria or locations of people with severe malaria not hospitalized. Histogram of modeled travel times between each severe malaria case at an identified hospital and the nearest Level 1 trauma center. Of severe malaria cases reported to CDC in 2012–2018 and hospitalized at an identified hospital, 34.8% were treated at Level 1 trauma centers (red bar); for severe malaria cases that were not at Level 1 trauma centers, we found that modeled transit times between treatment site and Level 1 trauma center were less than 2 hours except in 4.2% of severe malaria cases at an identified hospital (right of black dashed vertical line). Conclusion More than 95% of people admitted for severe malaria at identified hospitals reported to the CDC in 2012–2018 were at or within 2h estimated driving time of a Level 1 trauma center, which may suggest priorities for implementing IVAS distribution. Disclosures Alison Ridpath, MD, MPH, Abbvie Inc.: Stocks/Bonds (Public Company)|Amarin Corperation PLC: Stocks/Bonds (Public Company)|Amgen Inc: Stocks/Bonds (Public Company)|BioNTech: Stocks/Bonds (Public Company)|Immunic Inc: Stocks/Bonds (Public Company)|infinity Pharmaceutical Companies: Stocks/Bonds (Public Company)|IQVIA Holdings Inc: Stocks/Bonds (Public Company)|Johnson and Johnson: Stocks/Bonds (Public Company)|Merck and Co inc: Stocks/Bonds (Public Company)|Pfizer Inc.: Stocks/Bonds (Public Company)|Protagonist Therapeutics: Stocks/Bonds (Public Company)

Antimicrobial resistance (AMR) is an urgent threat to public health, but gaps in surveillance limit the detection of emergent novel threats and knowledge about the global distribution of AMR genes. International travelers frequently acquire AMR organisms, and thus may provide a window into AMR dynamics in otherwise poorly monitored regions and environments. To assess the utility of travelers as global AMR sentinels, we collected pre- and post-travel stool samples from 608 travelers, which were screened for the presence of extended-spectrum beta-lactamase producing Enterobacterales, carbapenem-resistant Enterobacterales, and mcr-mediated colistin-resistant Enterobacterales. A total of 307 distinct AMR organisms were sequenced in order to determine genotypic patterns and their association with travel region and behavior. Travel-associated AMR organisms were overwhelmingly E. coli, which exhibited considerable phylogenetic diversity regardless of travel region. However, the prevalence of resistance genes varied by region, with bla CTX-M-55 and bla CTX-M-27 significantly more common in travelers returning from South America and South-Eastern Asia, respectively. Hybrid assembly and plasmid reconstruction revealed the genomic neighborhood of bla CTX-M-55 frequently matched a motif previously linked to animal populations. Contact with animals was also associated with virulence factors in acquired AMR organisms, including carriage of the ColV plasmid, a driver of avian pathogenic E. coli. We identified novel variants of the mcr-1 gene in strains acquired from Western Africa, highlighting the potential for traveler surveillance to detect emerging clinical threats. Ongoing efforts to track travel-acquired organisms could complement existing global AMR surveillance frameworks.

Background Infections by Streptococcus pneumoniae and influenza viruses are vaccine-preventable diseases causing great morbidity and mortality. We evaluated pneumococcal and influenza vaccination practices during pre–international travel health consultations. Methods We evaluated data on pretravel visits over a 10-year period (1 July 2012 through 31 June 2022) from 31 sites in Global TravEpiNet (GTEN), a consortium of US healthcare facilities providing pretravel health consultations. Data were collected using an online structured questionnaire utilized by GTEN providers. We obtained summary statistics and performed multivariable logistic regression models to identify characteristics associated with receiving the vaccinations. Results At 116 865 pretravel visits, 28 754 (25%) travelers were eligible to receive pneumococcal vaccination and 56 150 (48%) travelers were eligible to receive influenza vaccination. A total of 19 557 (68%) pneumococcal vaccine–eligible travelers were not offered the vaccine at the pretravel visit. Among influenza vaccine–eligible travelers, 8592 (15%) were not offered the vaccine, and an additional 16 931 (30%) travelers declined the vaccine. Influenza vaccine was not available for 8014 (14%) eligible travelers. Nonadministration of the influenza vaccine was most frequent in the months of April through September. Compared to nonacademic centers or centers in the South or Midwest, travelers seen in academic centers or centers in the Northeast were more likely to receive either vaccine. Conclusions Increasing awareness of global influenza transmission patterns and improving access to routine vaccines at the pretravel encounter may enhance vaccination for respiratory pathogens in departing US international travelers.

Memory lymphocytes are durable cells that persist in the absence of antigen, but few human B cell subsets have been characterized in terms of durability. The relative durability of eight non-overlapping human B cell sub-populations covering 100% of all human class-switched B cells was interrogated. Only two long-lived B cell populations persisted in the relative absence of antigen. In addition to canonical germinal center-derived switched-memory B cells with an IgD ⁻ CD27 ⁺ CXCR5 ⁺ phenotype, a second, non-canonical, but distinct memory population of IgD ⁻ CD27 ⁻ CXCR5 ⁺ DN1 B cells was also durable, exhibited a unique TP63 -linked transcriptional and anti-apoptotic signature, had low levels of somatic hypermutation, but was more clonally expanded than canonical switched-memory B cells. DN1 B cells likely evolved to preserve immunological breadth and may represent the human counterparts of rodent extrafollicular memory B cells that, unlike canonical memory B cells, can enter germinal centers and facilitate B cell and antibody evolution. Graphical Abstract

We evaluated Klebsiella pneumoniae (Kp) gut carriage in healthy, unrelated adults and children living in separate households in Dhaka, Bangladesh. Average Kp prevalence in stool samples ranged from 61% in young children (15/25) to 81% in adults (21/26), with significantly higher abundance in adults (p = 0.03, t-test). Kp was also prevalent in household water (64%, 21/33) and standing water (85%, 23/27). The presence of Kp in household water was not strongly linked to stool Kp abundance among household members. Antimicrobial resistance was notable: 9% (6/69) of stool and 16% (7/44) of water isolates exhibited multi-drug resistance. Carbapenem resistance was observed in 12% of stool isolates (8/69) and 14% of water isolates (6/44). These findings underscore the commonality of Kp in human and environmental reservoirs in Dhaka, Bangladesh, and highlight the emergence of drug-resistant Kp beyond healthcare settings.