Development of a Human Immunodeficiency Virus Risk Prediction Model Using Electronic Health Record Data From an Academic Health System in the Southern United States.

BACKGROUND: Human immunodeficiency virus (HIV) pre-exposure prophylaxis (PrEP) is underutilized in the southern United States. Rapid identification of individuals vulnerable to diagnosis of HIV using electronic health record (EHR)-based tools may augment PrEP uptake in the region. METHODS: Using machine learning, we developed EHR-based models to predict incident HIV diagnosis as a surrogate for PrEP candidacy. We included patients from a southern medical system with encounters between October 2014 and August 2016, training the model to predict incident HIV diagnosis between September 2016 and August 2018. We obtained 74 EHR variables as potential predictors. We compared Extreme Gradient Boosting (XGBoost) versus least absolute shrinkage selection operator (LASSO) logistic regression models, and assessed performance, overall and among women, using area under the receiver operating characteristic curve (AUROC) and area under precision recall curve (AUPRC). RESULTS: Of 998 787 eligible patients, 162 had an incident HIV diagnosis, of whom 49 were women. The XGBoost model outperformed the LASSO model for the total cohort, achieving an AUROC of 0.89 and AUPRC of 0.01. The female-only cohort XGBoost model resulted in an AUROC of 0.78 and AUPRC of 0.00025. The most predictive variables for the overall cohort were race, sex, and male partner. The strongest positive predictors for the female-only cohort were history of pelvic inflammatory disease, drug use, and tobacco use. CONCLUSIONS: Our machine-learning models were able to effectively predict incident HIV diagnoses including among women. This study establishes feasibility of using these models to identify persons most suitable for PrEP in the South.

Feasibility, Acceptability, and Preliminary Efficacy of a Gamified Mobile Health Contingency Management Intervention for PrEP Adherence Among Black MSM.

Oral HIV pre-exposure prophylaxis (PrEP) is effective at preventing HIV. However, low adherence is common and undermines these protective effects. This is particularly relevant for groups with disproportionately higher rates of HIV, including Black men who have sex with men (MSM). The current study tested the feasibility, acceptability, and preliminary efficacy of a gamified mobile health contingency management intervention for PrEP adherence-called mSMART (Mobile App-Based Personalized Solutions for Medication Adherence of Rx Pill Tool). Fifteen Black MSM already prescribed PrEP in the community completed baseline and follow-up assessments separated by 8 weeks of using mSMART. Regarding feasibility, there was no study attrition, no mSMART functional difficulties that significantly interfered with use, and a mean rate of 82% daily mSMART use. Acceptability ratings were in the moderately to extremely satisfied range for factors such as willingness to recommend mSMART to others and user-friendliness, and in the low range for ratings on difficulty learning how to use mSMART. Scores on a system usability measure were in the acceptable range for 73% of the sample. Qualitative analysis of follow-up interviews identified individual components of mSMART that could be modified in future iterations to make it more engaging. PrEP composite adherence scores from biomarkers indicated an improvement from baseline to follow-up with a medium effect size, as well as a decrease in the number of perceived barriers to medication adherence. Findings indicate a future efficacy trial is needed to examine the effects of this gamified mobile health contingency management intervention on PrEP adherence.

A virtual clinic improves pneumococcal vaccination coverage among patients living with HIV at a Veterans Affairs Medical Center.

People living with the human immunodeficiency virus (HIV) should receive pneumococcal vaccinations as part of their routine health maintenance. Our goal was to create a "virtual clinic" to help increase rates of pneumococcal vaccination among people living with HIV without adding substantially to the workload of primary providers. We used administrative data from our Veterans Affairs (VA) medical center to identify a cohort of veterans living with HIV who were not current with either the 13-valent pneumococcal conjugate vaccine (PCV13), the 23-valent pneumococcal polysaccharide vaccine (PPSV23) or both. We enrolled these individuals (n = 99) into a virtual clinic, notified providers via the electronic medical record and mailed letters to the veterans recommending they receive a pneumococcal vaccine. We also wrote orders for the appropriate pneumococcal vaccine that expired after 90 days. Among the virtual clinic cohort, 38% (38/99) of patients received the recommended vaccine within 180 days. Concurrent with our intervention, the Veterans Health Administration deployed a system-wide pneumococcal vaccine clinical reminder that incorporated recent PCV13 recommendations. To discern any effect of the virtual clinic beyond that of the clinical reminder, we compared the rate of PCV13 vaccinations among all HIV-positive veterans at our institution to the equivalent population from 2 other VA medical centers in Ohio. With consideration of the VHA's system-wide clinical reminder, the proportion of HIV-positive patients who received PCV13 in the first 90 days following the virtual clinic intervention was greater at our facility compared to another Ohio VA medical center (P < 0.05). The virtual clinic improved the pneumococcal vaccine coverage among HIV-positive veterans. These outcomes suggest that even in conjunction with a system-wide clinical reminder, the virtual clinic strategy improves vaccination rates among a high-risk population.

Vaccination with Live Attenuated Simian Immunodeficiency Virus (SIV) Protects from Mucosal, but Not Necessarily Intravenous, Challenge with a Minimally Heterologous SIV.

UNLABELLED: Few studies have evaluated the impact of the viral challenge route on protection against a heterologous simian immunodeficiency virus (SIV) challenge. We vaccinated seven macaques with a live attenuated SIV that differed from SIVmac239Δnef by 24 amino acids, called m3KOΔnef. All animals were protected from an intrarectal SIVmac239 challenge, whereas only four animals were protected from subsequent intravenous SIVmac239 challenge. These data suggest that immune responses elicited by vaccination with live attenuated SIV in an individual animal can confer protection from intrarectal challenge while remaining insufficient for protection from intravenous challenge. IMPORTANCE: Our study is important because we show that vaccinated animals can be protected from a mucosal challenge with a heterologous SIV, but the same animals are not necessarily protected from intravenous challenge with the same virus. This is unique because in most studies, either vaccinated animals are challenged multiple times by the same route or only a single challenge is performed. An individually vaccinated animal is rarely challenged multiple times by different routes, so protection from different challenge routes cannot be measured in the same animal. Our data imply that vaccine-elicited responses in an individual animal may be insufficient for protection from intravenous challenge but may be suitable for protection from a mucosal challenge that better approximates human immunodeficiency virus (HIV) exposure.

Acute-phase CD8 T cell responses that select for escape variants are needed to control live attenuated simian immunodeficiency virus.

The overall CD8 T cell response to human/simian immunodeficiency virus (HIV/SIV) targets a collection of discrete epitope specificities. Some of these epitope-specific CD8 T cells emerge in the weeks and months following infection and rapidly select for sequence variants, whereas other CD8 T cell responses develop during the chronic infection phase and rarely select for sequence variants. In this study, we tested the hypothesis that acute-phase CD8 T cell responses that do not rapidly select for escape variants are unable to control viral replication in vivo as well as those that do rapidly select for escape variants. We created a derivative of live attenuated SIV (SIVmac239Δnef) in which we ablated five epitopes that elicit early CD8 T cell responses and rapidly accumulate sequence variants in SIVmac239-infected Mauritian cynomolgus macaques (MCMs) that are homozygous for the M3 major histocompatibility complex (MHC) haplotype. This live attenuated SIV variant was called m3KOΔnef. Viremia was significantly higher in M3 homozygous MCMs infected with m3KOΔnef than in either MHC-mismatched MCMs infected with m3KOΔnef or MCMs infected with SIVmac239Δnef. Three CD8 T cell responses, including two that do not rapidly select for escape variants, predominated during early m3KOΔnef infection in the M3 homozygous MCMs, but these animals were unable to control viral replication. These results provide evidence that acute-phase CD8 T cell responses that have the potential to rapidly select for escape variants in the early phase of infection are needed to establish viral control in vivo.

Epidemiology and outcomes of bacterial meningitis in the neonatal intensive care unit.

OBJECTIVE: Examine pathogen distribution, antibiotic resistance patterns, and hospital outcomes of infants with bacterial meningitis in neonatal intensive care units (NICUs) in the US from 2013-2018. STUDY DESIGN: Infants were divided into 2 groups based on age at the time of meningitis: early-onset (0-3 days) and late-onset (>3 days). We compared demographics, clinical characteristics, epidemiology, hospital outcomes, distribution of organisms and resistance, and blood culture timing relative to cerebrospinal fluid culture. RESULTS: From 345 NICUs, 659 infants were diagnosed with bacterial meningitis. The cumulative incidence was 1.1-1.3 cases/1000 NICU discharges. Median gestational age was 33 weeks, median birth weight was 1910 grams, 12% failed hearing screening, and 9% died prior to discharge. Of 141 cases of E. coli meningitis, 53% were resistant to ampicillin. CONCLUSIONS: Significant morbidities occur in infants with culture-proven meningitis in NICUs. Culture and subsequent discernment of sensitivity are crucial to guide definitive therapy.

A survey of South Carolina pharmacists' readiness to prescribe human immunodeficiency virus pre-exposure prophylaxis.

Introduction: HIV pre-exposure prophylaxis (PrEP) is largely underutilized in the Southern United States. Given their community presence, pharmacists are well positioned to provide PrEP within rural, Southern regions. However, pharmacists' readiness to prescribe PrEP in these communities remains unknown. Objective: To determine the perceived feasibility and acceptability of prescribing PrEP by pharmacists in South Carolina (SC). Methods: We distributed a 43-question online descriptive survey through the University of SC Kennedy Pharmacy Innovation Center's listerv of licensed SC pharmacists. We assessed pharmacists' comfort, knowledge, and readiness to provide PrEP. Results: A total of 150 pharmacists responded to the survey. The majority were White (73%, n=110), female (62%, n=93), and non-Hispanic (83%, n=125). Pharmacists practiced in retail (25%, n=37), hospital (22%, n=33), independent (17%, n=25), community (13%, n=19), specialty (6%, n=9), and academic settings (3%, n=4); 11% (n=17) practiced in rural locales. Pharmacists viewed PrEP as both effective (97%, n=122/125) and beneficial (74% n=97/131) for their clients. Many pharmacists reported being ready (60% n=79/130) and willing (86% n=111/129) to prescribe PrEP, although over half (62% n=73/118) cited lack of PrEP knowledge as a barrier. Pharmacists described pharmacies as an appropriate location to prescribe PrEP (72% n=97/134). Conclusions: Most SC pharmacists surveyed considered PrEP to be effective and beneficial for individuals who frequent their pharmacy and are willing to prescribe this therapy if statewide statutes allow. Many felt that pharmacies are an appropriate location to prescribe PrEP but lack a complete understanding of required protocols to manage these patients. Further investigation into facilitators and barriers of pharmacy-driven PrEP are needed to enhance utilization within communities.

Transcriptionally abundant major histocompatibility complex class I alleles are fundamental to nonhuman primate simian immunodeficiency virus-specific CD8+ T cell responses.

Simian immunodeficiency virus (SIV)-infected macaques are the preferred animal model for human immunodeficiency virus (HIV) vaccines that elicit CD8(+) T cell responses. Unlike humans, whose CD8(+) T cell responses are restricted by a maximum of six HLA class I alleles, macaques express up to 20 distinct major histocompatibility complex class I (MHC-I) sequences. Interestingly, only a subset of macaque MHC-I sequences are transcriptionally abundant in peripheral blood lymphocytes. We hypothesized that highly transcribed MHC-I sequences are principally responsible for restricting SIV-specific CD8(+) T cell responses. To examine this hypothesis, we measured SIV-specific CD8(+) T cell responses in MHC-I homozygous Mauritian cynomolgus macaques. Each of eight CD8(+) T cell responses defined by full-proteome gamma interferon (IFN-γ) enzyme-linked immunospot (ELISPOT) assay were restricted by four of the five transcripts that are transcriptionally abundant (>1% of total MHC-I transcripts in peripheral blood lymphocytes). The five transcriptionally rare transcripts shared by these animals did not restrict any detectable CD8(+) T cell responses. Further, seven CD8(+) T cell responses were defined by identifying peptide binding motifs of the three most frequent MHC-I transcripts on the M3 haplotype. Combined, these results suggest that transcriptionally abundant MHC-I transcripts are principally responsible for restricting SIV-specific CD8(+) T cell responses. Thus, only a subset of the thousands of known MHC-I alleles in macaques should be prioritized for CD8(+) T cell epitope characterization.

Perceptions on HIV Pre-Exposure Prophylaxis Among Urgent Care Clinicians in the Southern United States.

BACKGROUND: Underutilization of HIV pre-exposure prophylaxis (PrEP) in the Southern United States (US) is well-documented. Urgent care (UC) centers are positioned as communityfacing access points to PrEP, but the feasibility of integrating PrEP services into this setting is unclear. We conducted a survey of UC clinicians in the Southern US to better understand their perceptions of the feasibility of providing PrEP in their practice setting. OBJECTIVE: The study aims to determine the feasibility and acceptability of providing PrEP services in the UC setting through a cross-sectional survey of UC clinicians. METHODS: We conducted a 48-item cross-sectional survey of UC clinicians in the Southern US, between July and September 2020. The survey was distributed through the Urgent Care Association (UCA) and American Academy of Urgent Care Medicine (AAUCM) professional listservs as well as directly to publicly listed e-mail addresses. RESULTS: Eighty-two clinicians responded to the survey. Most clinicians had familiarity with PrEP (97%). All respondents rated PrEP as an effective way to prevent HIV. However, less than half felt UC facilities were an appropriate place to prescribe PrEP. Few respondents (8%) expressed doubts that expansion of PrEP access would decrease the incidence of HIV in their community. CONCLUSION: These findings show UC clinicians are familiar with PrEP, and many believe it would benefit their patients; however, provider opinions on the appropriateness of providing PrEP in the UC setting differ. Further studies on PrEP implementation in UC centers are needed.

Understanding Retention in Pre-Exposure Prophylaxis Care in the South: Insights from an Academic HIV Prevention Clinic.

HIV pre-exposure prophylaxis (PrEP) is poorly utilized in the southern United States. We examined PrEP retention in care and sexually transmitted infections (STIs) through a retrospective review of the Duke University PrEP Clinic from January 1, 2015 to October 15, 2019. We evaluated short-term (3 months), long-term (additional 8-12 months), and longitudinal retention in care in our clinic. Adjusted odds ratios (aOR) were generated to explore demographics associated with retention. Kaplan-Meier curves were generated to view retention longitudinally. STIs were examined at baseline (1 year before initial PrEP visit) and while retained in care. Of a total of 255 patients; 88% were men, 37% were black, and 73% were men who have sex with men (MSM). Short- and long-term retention in care were met by 130/237 (55%) and 80/217 (37%) patients, respectively. MSM were more likely to be retained in the short term (aOR = 5.22, 95% confidence interval [CI] = 1.57-17.32). Self-referred patients were more likely to be retained in the long term (aOR = 2.18, 95% CI = 1.12-4.23). Uninsured patients were less likely to be retained in the long term (aOR = 0.32, 95% CI = 0.11-0.91). STI diagnoses include 42 infections at baseline and 69 infections during follow-up. STI diagnosed while in PrEP care was associated with longer retention in care over time. Patients discontinue PrEP care over time and STIs were frequently encountered. Additional studies are needed to determine the best way to retain patients in HIV preventative care.

Experimental analysis of sources of error in evolutionary studies based on Roche/454 pyrosequencing of viral genomes.

Factors affecting the reliability of Roche/454 pyrosequencing for analyzing sequence polymorphism in within-host viral populations were assessed by two experiments: 1) sequencing four clonal simian immunodeficiency virus (SIV) stocks and 2) sequencing mixtures in different proportions of two SIV strains with known fixed nucleotide differences. Observed nucleotide diversity and frequency of undetermined nucleotides were increased at sites in homopolymer runs of four or more identical nucleotides, particularly at AT sites. However, in the mixed-strain experiments, the effects on estimated nucleotide diversity of such errors were small in comparison to known strain differences. The results suggest that biologically meaningful variants present at a frequency of around 10% and possibly much lower are easily distinguished from artifacts of the sequencing process. Analysis of the clonal stocks revealed numerous rare variants that showed the signature of purifying selection and that elimination of variants at frequencies of less than 1% reduced estimates of nucleotide diversity by about an order of magnitude. Thus, using a 1% frequency cutoff for accepting a variant as real represents a conservative standard, which may be useful in studies that are focused on the discovery of specific mutations (such as those conferring immune escape or drug resistance). On the other hand, if the goal is to estimate nucleotide diversity, an optimal strategy might be to include all observed variants (even those at less than 1% frequency), while masking out homopolymer runs of four or more nucleotides.