Immunologic and Clinical Responses to Highly Active Antiretroviral Therapy in Patients with HIV Infection Aged >50 Years

Clinicians will encounter increasing numbers of older HIV-infected patients in the coming years. Recent data from the French Hospital Database on HIV, a nationwide hospital-based cohort of >100,000 HIV-infected patients, showed that 19% of the patients were >50 years old in 2003 [1]. This population is increasing over time because of the normal ageing of HIV-infected patients since HAART availability and because of late diagnosis of HIV infection in older patients [2]. Older patients have already been shown to be at high risk for late testing, because they are unaware that they are at risk in HIV infection, physicians are less likely to consider a diagnosis of HIV infection for older patients, and routine HIV testing is uncommon in this population [3].

The immunovirological efficacy of antiretroviral therapy in this population has been debated, but since 1996, several studies have shown that age is neither related to treatment failure nor related to poorer immunologic restoration [3,4,5–6]. Nevertheless, most of these studies provided data on small populations and did not take into account the “late testing” population, defined by either an AIDS-defining event or by a CD4 cell count <200 cells/mm³ at the time of diagnosis, which is known to experience poorer immunovirological evolution of infection in response to HAART [7, 8]. This characteristic could, in part, explain the poorer immunovirological evolution of infection in response to HAART among older patients in some studies [7, 9].

Furthermore, data on treatment tolerability in older patients is scarce [10]. Treatment tolerance could be worse in this population because of comorbidities and treatment interactions.

The aim of our study was to analyze the effect of age on HAART efficacy and tolerance in a large cohort of patients, taking late-testing situations into account.

Patients and methods. Information was collected from a large HIV reference center in Toulouse, France. This hospital (Hopital Purpan, Toulouse, France) maintains a prospective cohort of all patients with HIV infection who seek care in the center and provide written consent. The patients enter the cohort when they seek care in the center, regardless of their HIV disease history, and all previous clinical events, as well as a therapeutic history, are collected with appropriate dates. Demographic details, clinical events, a history of antiretroviral therapy, viral load data, and CD4 cell counts are collected at regular 3–6-month intervals during routine clinical assessment.

We selected antiretroviral-naive patients who initiated HAART from 1 January 1996 through 16 January 2006. The date of first HAART prescription was used as a baseline. HAART was defined as a combination of 2 nucleoside reverse-transcriptase inhibitors plus either a protease inhibitor, with or without pharmacological booster, or a nonnucleoside reverse-transcriptase inhibitor, or 3 nucleoside reverse-transcriptase inhibitors. We used an age cut off value of 50 years (<50 and ⩾50 years old at baseline), as has previously been used by others [3, 4, 6, 7].

Collected variables were sex, age at treatment initiation, most probable route of infection, date of HIV infection diagnosis, Centers for Disease Control and Prevention stage, date of first AIDS-defining event, hepatitis B or C virus coinfection, year of HAART initiation, type of first HAART (protease inhibitor, boosted protease inhibitor, nucleoside reverse-transcriptase inhibitor, or nonnucleoside reverse-transcriptase inhibitor), date of and reason for first treatment discontinuation (clinical, virologic failure, or poor tolerance), and CD4 cell count and viral load at the time of HIV infection diagnosis and at baseline (±3 months). Late HIV testing was defined as a CD4 cell count ⩾200 cells/mm³ at HIV infection diagnosis or an AIDS-defining event in the first year following the diagnosis.

A CD4 cell count >350 cells/mm³ and an HIV RNA load <2.3 log copies/mL after 6 months of HAART were chosen as indicators of immunologic reconstitution and virological efficacy, respectively. Proportions of patients reaching these end points were compared between groups. Clinical progression was defined as any new AIDS-defining event or death, whatever the cause of death. Treatment tolerability, estimated by the proportion of treatment discontinuations for poor tolerance, and time to treatment discontinuation, as well as time to clinical progression, were compared between groups.

Comparisons between the 2 groups were computed using a χ² test for categorical variables and nonparametric Wilcoxon rank-sum test for continuous variables. Multivariate logistic regression was performed to test the association between baseline characteristics and immunologic reconstitution, as well as the association between virological efficacy and clinical progression. Median duration of the first course of HAART was estimated with Kaplan-Meier analysis and compared between the 2 groups with the log-rank test. Multivariate logistic regression was also performed to analyze factors associated with treatment discontinuation for poor tolerance.

All analyses were intention-to-continue-treatment analyses, and only the first treatment regimen was considered. Multivariate final models included all variables with P < .2 in bivariate analysis and present only variables that were statistically significant to the threshold of .05.

Results. A total of 639 patients were included in our study; of these, 99 (15.5%) were aged >50 years, and 257 (40.2%) were late testers. Comparisons between groups at baseline are shown in table 1. Late testing was more frequent in older patients than in younger ones (56.2% vs. 44.9%; P = .05). Median duration of follow-up was 51 months (interquartile range, 25–81.7 months) in the older group and 56.6 months (interquartile range, 30.4–90 months) in the younger group (P = .25). Among the 639 patients included in the study, 463 (72.5%) had an available CD4 cell count and 438 patients (68.5%) had a viral load measurement after 6 months of HAART.

View larger version:

• In this window
• In a new window

• Download as PowerPoint Slide

Table 1

Baseline demographic and clinical characteristics, by age.

Median absolute increase in CD4 cell count at month 6 was similar in the 2 groups (+100 cells/mm³ vs. +104 cells/mm³; P = .43), as was the proportion of patients who had a CD4 cell count of >350 cells/mm³ at 6 months (54.6% vs. 50.5%; P = .52), but late testers reached this threshold less frequently than did other patients (19.3% vs. 75.3%; P < .001). In the multivariate analysis, the probability of having a CD4 cell count of <50 cells/mm³ after 6 months of HAART was only associated with late testing (OR, 3.3; 95% CI, 1.6–6.7).

The proportion of patients who had a viral load ⩽2.3 log copies/mL was similar in the 2 groups (66.7% vs. 68.6%). In the multivariate analysis, no characteristic was independently associated with this end point.

The proportion of clinical progression was similar in the 2 groups (8.1% vs. 5.7%), as was the median time before the clinical event (10.8 months [range, 4.6–32.8 months] and 7.1 months [range, 0.7–56.9 months] for older patients and younger patients, respectively). In the multivariate analysis, late testing was the only characteristic associated with clinical progression of infection (OR, 3.4; 95% CI, 1.5–8).

First HAART discontinuation was observed for 428 patients (67%) during follow-up, with no difference in the rate of discontinuation between the groups (older patients, 66.7%; younger patients, 67%). The median time before discontinuation was 6.4 months (interquartile range, 2.5–13.7 months) for older patients and 14.1 months (interquartile range, 4.9–31.0 months) for younger patients (P < .01, by log-rank test). In the multivariate analysis, discontinuation of the first HAART regimen because of poor tolerance was independently associated with age (OR, 2; 95% CI, 1.1–3.6), with late testing (OR, 0.6; 95% CI, 0.4–0.9), and with year of treatment initiation (OR, 2 [95% CI, 1.1–3.8], and OR, 1.1 [95% CI, 0.7–1.9], for 2003–2006 and 2000–2002, respectively, compared with 1996–1999). Treatment discontinuations as the result of neurologic or psychiatric adverse effects were more frequent among older patients than younger patients (9.1% vs. 3%; P = .03), as were hematological adverse effects (13.6% vs. 5.5%; P = .03).

Discussion. This study of an open-label prospective cohort included 639 antiretroviral-naive patients at the time of initiation of their first HAART regimen. Ninety-nine patients (15.5%) were aged >50 years. After 6 months of HAART, immunologic and virological evolution of infection did not differ between the 2 age groups, with 52.5% of the patients having a CD4 cell count >350 cells/mm³ and 67.7% having undetectable viral loads. Late testing was more frequent among the older group and was significantly associated with less frequent immunologic reconstitution and with clinical progression. First discontinuation of HAART was not more frequent among older patients, but when it occurred, it was earlier and more frequently related to poor tolerance.

The comparable efficacy of HAART in older patients has already been described in recent studies [4,5–6, 10, 11]. By contrast, some studies have shown that immunologic reconstitution was less frequent with increasing age [9, 12], but these studies did not take the effect of late testing into account.

Our study has some limitations. Because our institution does not collect quantitative data on treatment adherence, this important information could not be analyzed. For this study, we selected patients initiating HAART in 1996 or after to maximize the number of patients. This long period is in part responsible for the missing values, because for those patients who were previously treated in another center, we may not have been able to collect pretherapeutic information. The missing data were equally distributed between groups, and we think that this should not be responsible for major biases.

The high frequency of late testing in the older population has already been described [1, 11], and we agree, as do others [6], that earlier diagnosis of HIV infection is mandatory in older patients, because the use of HAART allows them to achieve immunovirological responses similar to those in younger patients.

• Received February 16, 2007.
• Accepted May 1, 2007.

• © 2007 Infectious Diseases Society of America