Plasma HIV viral load in patients with hemophilia and late-stage HIV disease: a measure of current immune suppression. Multicenter Hemophilia Cohort Study

Utilizing electronic health record data to understand comorbidity burden among people living with HIV: a machine learning approach

OBJECTIVES

An understanding of the predictors of comorbidity among people living with HIV (PLWH) is critical for effective HIV care management. In this study, we identified predictors of comorbidity burden among PLWH based on machine learning models with electronic health record (EHR) data.

METHODS

The study population are individuals with a HIV diagnosis between January 2005 and December 2016 in South Carolina (SC). The change of comorbidity burden, represented by the Charlson Comorbidity Index (CCI) score, was measured by the score difference between pre- and post-HIV diagnosis, and dichotomized into a binary outcome variable. Thirty-five risk predictors from multiple domains were used to predict the increase in comorbidity burden based on the logistic least absolute shrinkage and selection operator (Lasso) regression analysis using 80% data for model development and 20% data for validation.

RESULTS

Of 8253 PLWH, the mean value of the CCI score difference was 0.8 ± 1.9 (range from 0 to 21) with 2328 (28.2%) patients showing an increase in CCI score after HIV diagnosis. Top predictors for an increase in CCI score using the LASSO model included older age at HIV diagnosis, positive family history of chronic conditions, tobacco use, longer duration with retention in care, having PEBA insurance, having low recent CD4+ cell count and duration of viral suppression.

CONCLUSION

The application of machine learning methods to EHR data could identify important predictors of increased comorbidity burden among PLWH with high accuracy. Results may enhance the understanding of comorbidities and provide the evidence based data for integrated HIV and comorbidity care management of PLWH.

Association of CD4+ T-cell Count, HIV-1 RNA Viral Load, and Antiretroviral Therapy With Kaposi Sarcoma Risk Among HIV-infected Persons in the United States and Canada

BACKGROUND

Kaposi sarcoma (KS) remains common among HIV-infected persons. To better understand KS etiology and to help target prevention efforts, we comprehensively examined a variety of CD4 T-cell count and HIV-1 RNA viral load (VL) measures, as well as antiretroviral therapy (ART) use, to determine independent predictors of KS risk.

SETTING

North American AIDS Cohort Collaboration on Research and Design.

METHODS

We followed HIV-infected persons during 1996-2009 from 18 cohorts. We used time-updated Cox regression to model relationships between KS risk and recent, lagged, trajectory, and cumulative CD4 count or VL measures, as well as ART use. We used Akaike's information criterion and global P values to derive a final model.

RESULTS

In separate models, the relationship between each measure and KS risk was highly significant (P < 0.0001). Our final mutually adjusted model included recent CD4 count [hazard ratio (HR) for <50 vs. ≥500 cells/μL = 12.4; 95% confidence interval (CI): 6.5 to 23.8], recent VL (HR for ≥100,000 vs. ≤500 copies/mL = 3.8; 95% CI: 2.0 to 7.3), and cumulative (time-weighted mean) VL (HR for ≥100,000 vs. ≤500 copies/mL = 2.5; 95% CI: 1.0 to 5.9). Each P-trend was <0.0001. After adjusting for these measures, we did not detect an independent association between ART use and KS risk.

CONCLUSIONS

Our results suggested a multifactorial etiology for KS, with early and late phases of development. The cumulative VL effect suggested that controlling HIV replication promptly after HIV diagnosis is important for KS prevention. We observed no evidence for direct anti-KS activity of ART, independent of CD4 count and VL.

Immunologic and virologic predictors of AIDS-related non-hodgkin lymphoma in the highly active antiretroviral therapy era

HIV-infected persons treated with highly active antiretroviral therapy (HAART) continue to have elevated risk for non-Hodgkin lymphoma (NHL). We conducted a retrospective cohort study of NHL among patients at an urban HIV clinic (N = 3025). Proportional hazards models identified immunologic and virologic predictors of NHL. Sixty-five NHLs arose during 1989 to 2006. NHL incidence declined over time. Nonetheless, 51 NHLs (78%) occurred within the HAART era (1996-2006). NHL risk increased with declining CD4 count (P trend < 0.0001) and increasing HIV viral load (P trend = 0.005). In a multivariable model, NHL risk was independently associated with both current CD4 count (hazard ratios 7.7 and 3.8, respectively, for CD4 counts 0-99 and 100-249 vs. 250+ cells/mm(3); P trend < 0.0001) and prior time spent with a viral load above 5.00 log(10) copies/mL (hazard ratios of 3.4, 2.6, and 6.8, respectively, for 0.1-0.4, 0.5-1.4, and 1.5+ yr vs. 0 yr; P trend = 0.004). Although serum globulin levels were elevated compared with the general population, NHL risk was unrelated to this B-cell activation marker (P = 0.39). Among HIV-infected individuals in the HAART era, NHLs are linked to immunosuppression and extended periods of uncontrolled HIV viremia. The association with high-level viremia could reflect detrimental effects on immune function related to incompletely effective HAART or direct effects on B cells.

AIDS denialism and public health practice

In this paper, we respond to AIDS denialist arguments that HIV does not cause AIDS, that antiretroviral drugs are not useful, and that there is no evidence of large-scale deaths from AIDS, and discuss the key implications of the relationship between AIDS denialism and public health practice. We provide a brief history of how the cause of AIDS was investigated, of how HIV fulfills Koch's postulates and Sir Bradford Hil's criteria for causation, and of the inconsistencies in alternatives offered by denialists. We highlight clinical trials as the standard for assessing efficacy of drugs, rather than anecdotal cases or discussions of mechanism of action, and show the unanimous data demonstrating antiretroviral drug efficacy. We then show how statistics on mortality and indices such as crude death rate, life expectancy, child mortality, and population growth are consistent with the high mortality from AIDS, and expose the weakness of statistics from death notification, quoted by denialists. Last we emphasize that when denialism influences public health practice as in South Africa, the consequences are disastrous. We argue for accountability for the loss of hundreds of thousands of lives, the need to reform public health practice to include standards and accountability, and the particular need for honesty and peer review in situations that impact public health policy.

Living with HIV/AIDS and Use of Online Support Groups

The present study explores the association between online support group use and health status, coping and social support among 640 individuals with HIV/AIDS. Results suggested that frequent users tended to be female, younger, single, and in a more advanced disease stage. After controlling for covariates, frequent users reported poorer health than non-users. In addition, both frequent and infrequent users scored higher in planning, active coping, instrumental support and emotional support coping from the brief COPE. No significant difference was found for social support. Our results suggest that online support groups may potentially offer some benefits for those living with HIV/AIDS.

Risk of human papillomavirus-associated cancers among persons with AIDS

BACKGROUND

Although risk of human papillomavirus (HPV)-associated cancers of the anus, cervix, oropharynx, penis, vagina, and vulva is increased among persons with AIDS, the etiologic role of immunosuppression is unclear and incidence trends for these cancers over time, particularly after the introduction of highly active antiretroviral therapy in 1996, are not well described.

METHODS

Data on 499 230 individuals diagnosed with AIDS from January 1, 1980, through December 31, 2004, were linked with cancer registries in 15 US regions. Risk of in situ and invasive HPV-associated cancers, compared with that in the general population, was measured by use of standardized incidence ratios (SIRs) and 95% confidence intervals (CIs). We evaluated the relationship of immunosuppression with incidence during the period of 4-60 months after AIDS onset by use of CD4 T-cell counts measured at AIDS onset. Incidence during the 4-60 months after AIDS onset was compared across three periods (1980-1989, 1990-1995, and 1996-2004). All statistical tests were two-sided.

RESULTS

Among persons with AIDS, we observed statistically significantly elevated risk of all HPV-associated in situ (SIRs ranged from 8.9, 95% CI = 8.0 to 9.9, for cervical cancer to 68.6, 95% CI = 59.7 to 78.4, for anal cancer among men) and invasive (SIRs ranged from 1.6, 95% CI = 1.2 to 2.1, for oropharyngeal cancer to 34.6, 95% CI = 30.8 to 38.8, for anal cancer among men) cancers. During 1996-2004, low CD4 T-cell count was associated with statistically significantly increased risk of invasive anal cancer among men (relative risk [RR] per decline of 100 CD4 T cells per cubic millimeter = 1.34, 95% CI = 1.08 to 1.66, P = .006) and non-statistically significantly increased risk of in situ vagina or vulva cancer (RR = 1.52, 95% CI = 0.99 to 2.35, P = .055) and of invasive cervical cancer (RR = 1.32, 95% CI = 0.96 to 1.80, P = .077). Among men, incidence (per 100 000 person-years) of in situ and invasive anal cancer was statistically significantly higher during 1996-2004 than during 1990-1995 (61% increase for in situ cancers, 18.3 cases vs 29.5 cases, respectively; RR = 1.71, 95% CI = 1.24 to 2.35, P < .001; and 104% increase for invasive cancers, 20.7 cases vs 42.3 cases, respectively; RR = 2.03, 95% CI = 1.54 to 2.68, P < .001). Incidence of other cancers was stable over time.

CONCLUSIONS

Risk of HPV-associated cancers was elevated among persons with AIDS and increased with increasing immunosuppression. The increasing incidence for anal cancer during 1996-2004 indicates that prolonged survival may be associated with increased risk of certain HPV-associated cancers.

Immune activation set point during early HIV infection predicts subsequent CD4+ T-cell changes independent of viral load

Although generalized T-cell activation is an important factor in chronic HIV disease pathogenesis, its role in primary infection remains poorly defined. To investigate the effect of immune activation on T-cell changes in subjects with early HIV infection, and to test the hypothesis that an immunologic activation “set point” is established early in the natural history of HIV disease, a prospective cohort of acutely infected adults was performed. The median density of CD38 molecules on CD4+ and CD8+ T cells was measured longitudinally in 68 antiretroviral-untreated individuals and 83 antiretroviral-treated individuals. At study entry, T-cell activation was positively associated with viremia, with CD8+ T-cell activation levels increasing exponentially at plasma HIV RNA levels more than 10 000 copies/mL. Among untreated patients, the level of CD8+ T-cell activation varied widely among individuals but often remained stable within a given individual. CD8+ T-cell activation and plasma HIV RNA levels over time were independently associated with the rate of CD4+ T-cell loss in untreated individuals. These data indicate that immunologic activation set point is established early in HIV infection, and that this set point determines the rate at which CD4+ T cells are lost over time.

Short-term risk of AIDS according to current CD4 cell count and viral load in antiretroviral drug-naive individuals and those treated in the monotherapy era

BACKGROUND

One key piece of information required when deciding whether to initiate antiretroviral therapy is the risk of AIDS before the next clinic visit. Information on the short-term (6-month) risk of AIDS according to the current viral load and CD4 cell count in untreated individuals and those treated in the zidovudine monotherapy era (i.e., pre-September 1995), especially in those with CD4 cell count > 200 x 10 cells/l, is lacking.

METHODS

Risk of AIDS was assessed in 3226 subjects with viral load and CD4 cell count known before initiation of antiretroviral therapy or during the zidovudine monotherapy era. These were from CASCADE Collaboration in which data from 20 cohorts of individuals with known dates of seroconversion to HIV, based in clinics in Europe and Australia, have been combined.

RESULTS

During a total 5126.0 person-years of follow-up, 219 individuals developed AIDS. In those with current CD4 cell count < 200 x 10 cells/l, 6-month risks were 4.9, 12.7, 17.7 and 22.4% for viral load groups < 10 000, 10 000-29 999, 30 000- 99 999 and > or = 100 000 copies/ml, respectively. For CD4 cell counts 200-349 x 10 cells/l risks were 0.5, 1.6, 3.2 and 4.7%, respectively, for the four viral load groups while the corresponding values for group with CD4 cell count > or = 350 x 10 cells/l were 0.2%, 0.5%, 0.9% and 2.2%, respectively. Results were similar when analysis was restricted to those with no antiretroviral drug experience. Older people had a higher risk of AIDS for a given CD4 cell count and viral load than younger people.

CONCLUSION

Combined with consideration of other issues, these estimates should prove useful information when deciding whether to initiate antiretroviral therapy in HIV-infected individuals.

Prevalence of hepatitis C virus infection and risk for hepatocellular carcinoma and non-Hodgkin lymphoma in AIDS

Hepatitis C virus (HCV) infection is highly prevalent in some subpopulations with AIDS. HCV is linked to hepatocellular carcinoma (HCC) and possibly non-Hodgkin lymphoma (NHL), but the impact of AIDS on these associations is uncertain. We used U.S. registry data to study HCC and NHL risk in 304,411 adults with AIDS, comparing cohort subgroups with high prevalence (hemophiliacs and injection drug users) or low prevalence (homosexual men, heterosexuals, and others) of HCV infection. The ratio of observed to expected cancer cases (standardized incidence ratio [SIR]) measured risk relative to the general population. Sixty-one HCC cases were observed (SIR, 7.5; 95% confidence interval, 5.7-9.6). Risk for HCC was higher in subgroups with high prevalence of HCV infection than in subgroups with low prevalence of HCV infection (SIR: 11.4 versus 5.5, respectively; p =.004). Subjects developed the following NHL grades: low, 35 cases; intermediate, 1035 cases; high, 784 cases; and unspecified, 1395 cases. For each NHL grade, SIRs were highest in subgroups with low prevalence of HCV infection. These data suggest an effect of HCV infection on HCC risk among adults with AIDS. On the other hand, NHL risk was not higher for groups in whom HCV infection was prevalent.

Predictive value of immunologic and virologic markers after long or short duration of HIV-1 infection

Laboratory markers that predict HIV-1 disease progression include plasma viral burden, CD4+ T-cell count, and CD38 expression on CD8 T cells. To better understand whether the predictive value of these markers is dependent on how long an individual has been infected, we analyzed data from the Multicenter AIDS Cohort Study early (median = 2.8 years) and late (median = 8.7 years) in the course of infection. Overall, we found that HIV RNA and CD38 levels were similarly predictive of AIDS early on compared with a relatively weaker CD4 cell count signal. Later in the course of infection, CD38 level remained the strongest predictive marker and CD4 cell count registered a marked increase in prognostic power. Among untreated individuals, there was little difference in prognosis (median time to AIDS) associated with given marker values regardless of infection duration. The prognosis given a specific viral load level tended to deteriorate late in the course of infection among those undergoing treatment with monotherapy or combination therapy, however. These data provide a unique historical look at the predictive value and prognostic significance of HIV-1 disease markers at different stages of infection in a large cohort, with direct relevance to current patients who are untreated or for whom treatment is ineffective.

Do HIV type 1 RNA levels provide additional prognostic value to CD4(+) T lymphocyte counts in patients with advanced HIV type 1 infection?

Our objective was to assess whether HIV-1 RNA levels provide additional prognostic information beyond CD4(+) T lymphocyte counts in the prediction of subsequent HIV-1 disease progression among patients with advanced HIV-1 disease. In a nested case-control study conducted in patients with baseline CD4(+) T lymphocyte counts < 300 cells/mm(3) and receiving nucleoside reverse transcriptase inhibitors, 102 patients who progressed to an AIDS-defining event or death were matched within 10 CD4(+) T lymphocyte cells/mm(3) to patients who did not progress. The relationship between plasma HIV-1 RNA levels and HIV-1 disease progression was studied using conditional logistic regression analysis, which adjusts for the matching by baseline CD4(+) T lymphocytes. We observed a 0.10 log(10) copies/ml difference in baseline HIV-1 RNA levels between cases and their matched controls (p = 0.027). The relative risk for HIV-1 disease progression increased with increasing baseline HIV-1 RNA levels (odds ratio [OR] for a 3-fold higher HIV-1 RNA level, 1.42; 95% confidence interval [CI], 1.08--1.86), and remained important when also controlling for clinical status at baseline and CD4(+) T lymphocytes at 2 months (p = 0.038). Higher baseline HIV-1 RNA levels were associated with HIV-1 disease progression among patients with a baseline CD4(+) T lymphocyte count of 100 cells/mm(3) or greater (OR, 1.80; 95% CI, 1.15--2.81), but not among patients with a baseline CD4(+) T lymphocyte count < 100 cells/mm(3) (OR, 1.09; 95% CI, 0.73--1.63). We concluded that HIV-1 RNA levels predict subsequent HIV-1 disease progression independent of CD4(+) T lymphocyte counts. The magnitude and importance of the prognostic information contained in the HIV-1 RNA levels appear to depend on the CD4(+) T lymphocyte counts.