Abstract
There were 1,311 newly-diagnosed HIV cases in San Francisco between 2005 and 2011 that were linked to care at publicly-funded facilities and had viral sequences available for analysis. Of the 214 cases characterized as recently-infected with HIV at time of diagnosis, 25% had a recent sexually transmitted infection (STI) diagnosis (vs. 10% among longer-standing HIV infections, p<0.001) and 57% were part of a phylogenetic transmission cluster (vs. 42% among longer-standing HIV infection, p<0.001). The association observed between recent HIV infection and having a STI diagnosis during the interval overlapping likely HIV acquisition points to potential opportunities to interrupt HIV transmission.
Keywords: HIV, recent infection, acute infection, sexually transmitted infections, phylogenetics, transmission clusters
Introduction
The majority of new HIV infections are thought to be transmitted by individuals who are unaware they are infected.1,2 With approximately 20% or fewer persons living with HIV in the United States unaware of their infection status, the leading edge of the epidemic may be concentrated in a few transmission chains or clusters at any point in time.3 Transmission clusters are defined as HIV infections that share sufficient viral genetic similarity to suggest a recent common source of infection or a chain of transmission. Prior studies have suggested that transmission clusters may contribute disproportionately to the rate of ongoing transmission.4-7 Meanwhile, the association between HIV and other sexually transmitted infections (STIs) has been proposed to be causally related to acquisition or transmission or as a marker for engaging in condomless sexual intercourse and having multiple partners in a short period of time.8-11 Therefore, characterizing the intersection of recent HIV infections, transmission clusters and incident STI may point to particularly efficient opportunities for prevention efforts. We examined the correlates associated with the stage of infection at time of diagnosis among new HIV cases in San Francisco.
Methods
We characterized San Francisco residents who were newly-diagnosed with HIV from 2005 through 2011, linked to care at publicly-funded facilities and had viral sequences available. Study data were obtained from the ARI-UCSF Laboratory of Clinical Virology (LCV), the San Francisco Department of Public Health (SFDPH) Laboratory, the SFDPH HIV/AIDS Case Registry and the SFDPH STD Case Registry. The LCV conducts HIV-1 drug resistance testing for publicly-funded and community-based clinics in San Francisco. The LCV database was matched to the HIV/AIDS Case Registry to obtain demographic and risk characteristics information. Matches to the STD Case Registry and SFDPH Laboratory dataset were conducted to obtain documentation of STI diagnoses and acute and recent HIV infection status, respectively.
Data on the demographic and risk characteristics of cases included gender, age, race/ethnicity and HIV transmission category. Clinical characteristics included stage of HIV infection at time of diagnosis and history of STIs. Recent HIV infection was defined as: 1) having a negative antibody test result within 6 months of HIV diagnosis based on self-reported data from HIV testing intake forms; and/or 2) presenting with acute infection; that is, being HIV antibody negative and HIV-RNA positive at time of HIV diagnosis. STIs reportable to the case registry included Chlamydia, gonorrhea, syphilis and non-gonococcal urethritis. Recent STI diagnosis was defined as being diagnosed with a reportable STI within the 6 months preceding HIV diagnosis. Cases could be diagnosed with a specific STI more than once and could be diagnosed with one or more STIs concurrently.
HIV transmission cluster membership was determined based on viral RNA sequences. Viral population sequencing of plasma viral RNA generated full protease and portions of the reverse transcriptase reading frames (TRUGENE HIV-1 Genotyping Assay, Siemens, Malvern, PA). All available viral sequences were included in the phylogenetic analysis. Viral sequences were aligned using ClustalW (in BioEdit 7.1.3.0) and adjusted manually. Sixty-seven codons known to be associated with drug resistance were excluded.12 A phylogeny was reconstructed using FastTree 2.1.5, with the generalized-time-reversible model of nucleotide substitution and a single substitution rate per cite (GTR+CAT). Transmission clusters were defined as having Shimodaira-Hasegawa node support greater than 0.90 and mean pairwise genetic distance less than 0.03 substitutions per site, a definition that is consistent with other similar published studies.13-15
Associations between recent HIV infection and demographic factors, HIV transmission risk, STI diagnosis and being part of a transmission cluster were evaluated using multivariate logistic regression. Data matching and statistical analyses were performed using SAS 9.3. The study received approval from the Institutional Review Board at the University of California, San Francisco.
Results
A total of 3,699 San Francisco residents were newly-diagnosed with HIV in San Francisco from 2005 through 2011, of whom 3,060 were linked to care at publicly-funded facilities and 1,311 had viral sequences from clinical resistance testing available for phylogenetics analysis. The 1,311 newly-diagnosed cases were mostly male, white, men who have sex with men (MSM), as detailed in Table 1. Median age was 36 years old (IQR 28, 44), ranging from 17 to 73 years old. Four hundred and seventeen cases (32%) had history of a STI diagnosis, with the most common being Chlamydia (n=384), gonorrhea (n=338), syphilis (n=201) and non-gonococcal urethritis (n=56). Among the 161 cases diagnosed with a STI within six months prior to HIV diagnosis, the most common STIs were Chlamydia (n=80), gonorrhea (n=64), syphilis (n=23) and non-gonococcal urethritis (n=6). The median time interval between the date of diagnosis and date of the first available blood sample for viral sequencing was 52 days (IQR 18, 364). There were 587 cases (45%) characterized as being part of a HIV transmission cluster; 350 (27%) were in a cluster of 2 or 3 cases and 237 (18%) were in a cluster of 4 or more cases.
Table 1.
Demographic characteristics, HIV transmission category and any sexually transmitted infection (STI) history of newly-diagnosed HIV persons, San Francisco, 2005-2011 (N=1,311)
| N | % | |
|---|---|---|
| Gender | ||
| Male | 1,098 | 84 |
| Female | 150 | 11 |
| Transgender | 63 | 5 |
| Age | ||
| ≤ 24 | 178 | 14 |
| 25-29 | 233 | 18 |
| 30-39 | 399 | 30 |
| 40-49 | 151 | 12 |
| ≥ 50 | 151 | 12 |
| Race/Ethnicity | ||
| White | 562 | 43 |
| Hispanic/Latino | 356 | 27 |
| Black | 250 | 19 |
| Asian/Pacific Islander | 96 | 7 |
| Other/Unknown | 47 | 4 |
| HIV Transmission Risk | ||
| Men who have sex with men | 1,009 | 76 |
| People who inject drugs | 147 | 11 |
| Heterosexual | 120 | 9 |
| Other/Unknown | 35 | 3 |
| Any STI History | ||
| None prior to HIV diagnosis | 894 | 68 |
| ≤ 6 months prior to HIV diagnosis | 161 | 12 |
| > 6 months prior to HIV diagnosis | 256 | 20 |
| HIV Transmission Cluster | ||
| Member of a cluster with 2-3 cases | 350 | 27 |
| Member of a cluster with ≥ 4 cases | 237 | 18 |
| Not member of a cluster | 724 | 55 |
There were 214 cases characterized as recent HIV infections; 27 cases were based on viral load testing for acute infection, 34 cases on a documented negative antibody test result within 6 months of HIV diagnosis and 153 cases on patient self-report of a last HIV negative test date within 6 months of HIV diagnosis. Of the 214 recent HIV infections cases, 109 (51%) had a STI history, with the most common being Chlamydia (n=105), gonorrhea (n=89), syphilis (n=43) and non-gonococcal urethritis (n=11). Among the 53 recent HIV infection cases that also had a STI diagnosis within 6 months prior to their HIV diagnosis, the most common STIs were Chlamydia (n=26), gonorrhea (n=24), syphilis (n=6) and non-gonococcal urethritis (n=2). Any history of Chlamydia (OR=2.8 [2.1-3.8]; p<0.0001), gonorrhea (OR=2.4 [1.8-3.3]; p<0.0001) and syphilis (OR=1.5 [1.0-2.2]; p=0.04) was associated with recent HIV infection. Persons with a recent diagnosis of Chlamydia (OR=2.7 [1.6-4.4]; p<0.0001) and gonorrhea (OR=3.3 [2.0-5.7]; p<0.0001) were more likely to be recently-infected with HIV whereas persons with a recent diagnosis of syphilis were not (OR=1.8 [0.7-4.7]; p=0.21). One hundred twenty-three cases with recent HIV infection (57%) were part of a transmission cluster, of which 64 (30%) were in a cluster with 2 or 3 cases and 59 (28%) were in a cluster with 4 or more cases.
The characteristics of being recently-infected with HIV compared to having a longer-standing infection at time of diagnosis are presented in Table 2. Having a recent STI diagnosis (p<0.001), being part of a transmission cluster (p=0.003) and younger age at HIV diagnosis were associated with recent HIV infection. Heterosexuals were less likely to be recently-infected at time of HIV diagnosis (p=0.02) than MSM and people who inject drugs. There was no association detected between gender, race and ethnicity and being recently-infected with HIV.
Table 2.
Characteristics of newly-diagnosed persons with recent HIV infection compared to longer-standing HIV infection at time of diagnosis, San Francisco, 2005-2011 (N=1,311)
| All HIV Infections |
Recent HIV Infections |
Longer-standing HIV Infections |
Odds Ratio (95% CI) |
P value | Adjusted OR (95% CI) |
P value | |
|---|---|---|---|---|---|---|---|
| Total | N=1,311 | N=214 | N=1,097 | ||||
| Gender | |||||||
| Female | 11% | 9% | 12% | 0.7 (0.4-1.2) | 0.20 | 2.0 (0.9-4.7) | 0.10 |
| Transgender | 5% | 5% | 5% | 0.9 (0.5-1.9) | 0.84 | 0.8 (0.4-1.7) | 0.65 |
| Male | 84% | 86% | 83% | Referent | - | Referent | - |
| Age at HIV diagnosis (years) | |||||||
| ≤ 24 | 14% | 21% | 12% | 4.8 (2.3-9.8) | < 0.0001 | 3.6 (1.7-7.6) | 0.0009 |
| 25-29 | 18% | 23% | 17% | 3.8 (1.8-7.7) | 0.0003 | 2.7 (1.3-5.7) | 0.009 |
| 30-39 | 30% | 30% | 30% | 2.7 (1.4-5.5) | 0.004 | 2.2 (1.1-4.6) | 0.03 |
| 40-49 | 26% | 21% | 28% | 2.1 (1.0-4.2) | 0.04 | 1.8 (0.9-3.7) | 0.12 |
| ≥ 50 | 12% | 5% | 13% | Referent | - | Referent | - |
| Race/Ethnicity | |||||||
| Black | 19% | 14% | 20% | 0.7 (0.5-1.1) | 0.12 | 1.0 (0.6-1.6) | > 0.99 |
| Hispanic/Latino | 27% | 29% | 27% | 1.0 (0.7-1.5) | 0.87 | 1.0 (0.7-1.5) | 0.84 |
| Asian/Pacific Islander | 7% | 8% | 7% | 1.1 (0.7-2.0) | 0.63 | 1.1 (0.6-2.0) | 0.67 |
| Other/Unknown | 4% | 5% | 3% | 1.3 (0.6-2.8) | 0.43 | 1.4 (0.7-3.1) | 0.36 |
| White | 43% | 44% | 43% | Referent | - | Referent | - |
| HIV Transmission Risk | |||||||
| Men who have sex with men | 76% | 86% | 75% | 1.6 (1.0-2.7) | 0.08 | 1.9 (0.9-4.0) | 0.12 |
| Heterosexual | 9% | 3% | 11% | 0.4 (0.1-1.0) | 0.05 | 0.3 (0.1-0.9) | 0.02 |
| Other/Unknown | 4% | 3% | 3% | 1.5 (0.4-4.1) | 0.44 | 1.7 (0.6-4.9) | 0.36 |
| People who inject drugs | 11% | 8% | 12% | Referent | - | Referent | - |
| Recent STI Diagnosis | |||||||
| ≤ 6 months prior to HIV diagnosis | 12% | 25% | 10% | 3.0 (2.1-4.4) | < 0.0001 | 2.4 (1.7-3.6) | < 0.0001 |
| > 6 months/none prior to HIV diagnosis | 88% | 75% | 90% | Referent | - | Referent | - |
| HIV Transmission Cluster | |||||||
| Linked to a cluster with ≥ 2 cases | 45% | 57% | 42% | 1.8 (1.4-2.5) | < 0.0001 | 1.8 (1.3-2.4) | 0.0003 |
| Not linked to a cluster | 55% | 43% | 58% | Referent | - | Referent | - |
Discussions
Many newly-diagnosed HIV cases in San Francisco had a history of an STI diagnosis, were part of a transmission cluster and were younger in age. These characteristics were even more prevalent among those cases that were recently-infected at time of HIV diagnosis. A quarter of recent HIV infection cases were also recently diagnosed with a STI, particularly Chlamydia and gonorrhea. More than half of recent HIV infections were found within a phylogenetic transmission cluster. Nearly half of recently-infected individuals were less than 30 years old at the time of their HIV diagnosis, the vast majority were MSM. Along with the potential attributable fraction of HIV due to undiagnosed infection, these factors help narrow the target to interrupt on-going transmission.
A primary aim of this analysis was to characterize the clustering of related viral sequences at a population level rather than to identify individual transmission linkages. HIV infections diagnosed soon after acquisition were more likely to be found within a detected phylogenetic transmission cluster. Infections not linked to a phylogenetic transmission cluster harbored a virus that was less closely related to other sequences in the study sample, possibly because persons with potentially related viruses were receiving care at private facilities in San Francisco or at facilities outside of San Francisco or are undiagnosed infections.
The study sample was limited to individuals who received care at publicly-funded facilities, which represented over 80% of all newly-diagnosed HIV cases during the study period. The associations observed among this sample may differ compared to individuals receiving care at private facilities or with private providers and persons who are not linked to care. Self-reports of a negative antibody test result within 6 months of HIV diagnosis could not be confirmed. However, our prior research found HIV-1 incidence estimates derived from laboratory-based assays and self-reported HIV testing history yielded comparable results.16 HIV transmission cluster linkages may have been underestimated since we would be unable to detect linkages that are mediated through a third party with viral sequences not available for analysis. Long intervals between time of infection and time of sampling for viral sequencing may affect cluster identification. However, given the median time interval between the date of diagnosis and date of the first available blood sample for viral sequencing was less than 2 months, the amount of genetic drift within an individual is likely to have limited impact on our cluster identification.
Our study findings point to intervention opportunities to interrupt HIV transmission. Recent HIV infection was associated with being diagnosed with a STI during the time interval when HIV transmission most likely occurred. Therefore, our finding highlights the value of integrating HIV and STI testing services, screening for acute HIV infection among newly-diagnosed STI cases, early initiation of antiretroviral treatment (eiART), increased use of STI partner services and use of HIV pre-exposure prophylaxis (PrEP) by uninfected members of these sexual networks where the risk for HIV infection is most acute.
The identification of HIV transmission clusters may further help to focus prevention efforts. The high rate of clustering observed among recently-infected persons indicates that recent infection may be associated with increased risk of onward transmission of HIV. This finding is consistent with the evidence of increased transmission risk related to unawareness of infection status and high viral load in acute and early untreated infection and does not support a paradigm of many infections arising sporadically from long-standing infections. Information on whether individuals were screened for syphilis and on negative test results were not available, which may partially account for the observed lack of association between recent infection and syphilis. Data were not available on whether or not individuals were screened for syphilis or if the screening results were negative, which may account in part for the observed lack of association between recent infection and syphilis. Recently-infected individuals may be reached through intervention programs such as partner services and risk reduction counseling for persons living with HIV to reduce the risk of further spread of infection. These findings also point towards the need for further studies to elucidate the factors associated with HIV transmission clusters and ways to reach them in real time with individual- and social-level interventions.
Acknowledgments
Funding Support: NIH R01 MH096642 (PI: H.M. Truong)
Footnotes
Presented in Part at: 21st Conference on Retroviruses and Opportunistic Infections, Paper 37, Boston, MA, March 3-6, 2014
References
- 1.Hall HI, Holtgrave DR, Maulsby C. HIV transmission rates from persons living with HIV who are aware and unaware of their infection. AIDS. 2012;26:893–896. doi: 10.1097/QAD.0b013e328351f73f. [DOI] [PubMed] [Google Scholar]
- 2.Truong HM, Berrey M, Shea T, Diem K, Corey L. Concordance between HIV source partner identification and molecular confirmation in acute retroviral syndrome. J Acquir Immune Defic Syndr. 2002;29:232–243. doi: 10.1097/00042560-200203010-00003. [DOI] [PubMed] [Google Scholar]
- 3.Chen M, Rhodes PH, Hall IH, Kilmarx PH, Branson BM, Valleroy LA. Prevalence of undiagnosed HIV infection among persons aged ≥13 years—National HIV Surveillance System, United States, 2005-2008. MMWR. 2012;61(Suppl):57–64. [PubMed] [Google Scholar]
- 4.Ambrosioni J, Junier T, Delhumeau C, et al. Impact of highly active antiretroviral therapy on the molecular epidemiology of newly diagnosed HIV infections. AIDS. 2012;26:2079–2086. doi: 10.1097/QAD.0b013e32835805b6. [DOI] [PubMed] [Google Scholar]
- 5.Brenner BG, Roger M, Stephens D, et al. Transmission clustering drives the onward spread of the HIV epidemic among men who have sex with men in Quebec. J Infect Dis. 2011;204:1115–1119. doi: 10.1093/infdis/jir468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lewis F, Hughes GJ, Rambaut A, Pozniak A, Brown AJL. Episodic sexual transmission of HIV revealed by molecular phylodynamics. PLoS Med. 2008;5:e50. doi: 10.1371/journal.pmed.0050050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Pao D, Fisher M, Hue S, et al. Transmission of HIV-1 during primary infection: relationship to sexual risk and sexually transmitted infections. AIDS. 2005;19:85–90. doi: 10.1097/00002030-200501030-00010. [DOI] [PubMed] [Google Scholar]
- 8.HIV prevention through early detection and treatment of other sexually transmitted diseases—United States. Recommendations of the Advisory Committee for HIV and STD prevention. MMWR Recomm Rep. 1998;47(RR-12):1–24. [PubMed] [Google Scholar]
- 9.Wasserheit JN. Epidemiological synergy: interrelationships between human immunodeficiency virus infection and other sexually transmitted diseases. Sex Transm Dis. 1992;19:61–77. [PubMed] [Google Scholar]
- 10.Dickerson MC, Johnston J, Delea TE, White A, Andrews E. The causal role for genital ulcer disease as a risk factor for transmission of human immunodeficiency virus. Sex Transm Dis. 1996;23:429–440. doi: 10.1097/00007435-199609000-00015. [DOI] [PubMed] [Google Scholar]
- 11.Kuiken CL, van Griensven GJP, de Vroome EMM, Coutinho RA. Risk factors and changes in sexual behavior in male homosexuals who seroconverted for human immunodeficiency virus antibodies. Am J Epidemiol. 1990;132:523–530. doi: 10.1093/oxfordjournals.aje.a115688. [DOI] [PubMed] [Google Scholar]
- 12.Johnson VA, Calvez V, Gunthard HF, et al. Update of the drug resistance mutations in HIV-1: March 2013. Top Antivir Med. 2013;21:6–14. [PMC free article] [PubMed] [Google Scholar]
- 13.Nguyen L, Hu DJ, Choopanya K, et al. Genetic analysis of incident HIV-1 strains among injection drug users in Bangkok: evidence for multiple transmission clusters during a period of high incidence. J Acquir Immune Defic Syndr. 2002;30:248–256. doi: 10.1097/00042560-200206010-00014. [DOI] [PubMed] [Google Scholar]
- 14.Chalmet K, Staelens D, Blot S, et al. Epidemiological study of phylogenetic transmission clusters in a local HIV-1 epidemic reveals distinct differences between subtype B and non-B infections. BMC Infect Dis. 2010;10:262. doi: 10.1186/1471-2334-10-262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Audelin AM, Gerstoft J, Obel N, et al. Molecular phylogenetics of transmitted drug resistance in newly diagnosed HIV Type 1 individuals in Denmark: a nation-wide study. AIDS Res Hum Retroviruses. 2011;27:1283–1290. doi: 10.1089/aid.2010.0368. [DOI] [PubMed] [Google Scholar]
- 16.Truong HM, Kellogg T, Louie B, Klausner J, Dilley J, McFarland W. Recent HIV-1 infection detection: comparison of incidence estimates derived by laboratory assays and repeat testing data. J Acquir Immune Defic Syndr. 2009;51:502–505. doi: 10.1097/QAI.0b013e3181a25383. [DOI] [PubMed] [Google Scholar]