Association of CD4+ T-lymphocyte counts and new thymic emigrants in HIV-infected children during successful highly active antiretroviral therapy

Abstract

Background: In a cohort of children receiving highly active antiretroviral therapy (HAART) with sustained plasma HIV-1 RNA < 50 copies/mL, children who reached undetectable RNA after week 8 (slow responders, median: week 20) had higher HIV-1 intracellular DNA (HIV-1 DNA) and equal or greater CD4+ T-lymphocyte counts compared with children who reached undetectable plasma HIV-1 RNA by week 8 (rapid responders) throughout HAART.

Objective: To determine whether levels of T-cell receptor excision circles (TRECs) could explain the apparent inconsistency between the quantity of HIV-1 DNA and CD4+ T-lymphocyte counts in HIV-1-infected children receiving HAART with sustained virologic suppression.

Methods: T-cell receptor excision circles and HIV-1 DNA and plasma HIV-1 RNA were quantified longitudinally by PCR in 31 children (median age, 5.6 years) with sustained undetectable plasma HIV-1 RNA for >104 weeks of HAART.

Results: There was a positive correlation between TREC and HIV-1 DNA during HAART, notably at weeks 48 and 80 (P < .004). During the early stage of HAART, TREC levels positively correlated with CD4+ T-lymphocyte percentages (P < .02) and naive CD4+ T-lymphocyte counts (P < .001) and percentages (P = .05). Median TREC levels were consistently equal or higher in slow responders compared with rapid responders (P < .001) despite slow responders having consistently greater quantities of HIV-1 DNA.

Conclusion: To maintain adequate levels of CD4+ T-lymphocytes, children with high HIV-1 DNA maintain high levels of TREC while receiving HAART. Thus, a thymic control mechanism is required to maintain new CD4+ T lymphocytes in the presence of persistent virus.

Clinical implications: The TREC level is a useful marker of thymic function in HIV-infected children.

FIG 1

Positive correlation between TREC levels and naive CD4⁺ T-lymphocyte counts. There was a strong positive correlation between TREC levels and naive CD4⁺ T-lymphocyte counts (CD4⁺CD45RA⁺; r = .92; P < .001) at week 20.

FIG 2

Comparison of median HIV-1 intracellular DNA levels between rapid responders (HIV-1 RNA < 50 copies/mL by week 8; white bars) and slow responders (HIV-1 RNA < 50 copies/mL after week 8; gray bars). Slow responders had persistently higher median HIV-1 intracellular DNA levels compared with rapid responders during HAART (P < .001). Data shown are expressed as the median value ± 25th and 75th percentiles. *P < .05; **P < .01.

FIG 3

Comparison of CD4⁺ T-lymphocyte counts between rapid responders (HIV-1 RNA < 50 copies/mL by week 8; white bars) and slow responders (HIV-1 RNA < 50 copies/mL after week 8; gray bars). Slow responders had equal or higher median CD4⁺ T-lymphocyte counts throughout HAART compared with rapid responders (P = .12) that reached significance at week 80 (P = .05) and week 104 (P = .04). Data shown are expressed as the median value ± 25th and 75th percentiles. *P < .05.

FIG 4

Comparison of TREC levels between rapid responders (HIV-1 RNA < 50 copies/mL by week 8; white bars) and slow responders (HIV-1 RNA levels < 50 copies/mL after week 8; gray bars). TREC levels were higher in slow responders than in rapid responders during HAART (P < .001) and reached significance at week 48 (P = .004) and week 80 (P = .01). Data shown are expressed as the median value ± 25th and 75th percentiles. * P < .05.