Background:  In a pediatric cohort receiving HAART with sustained virologic suppression for >104 weeks, slow responders (plasma HIV RNA >50 copies/mL after week 8 on treatment) had higher plasma HIV RNA (RNA) and HIV intracellular DNA (DNA) at baseline than rapid responders (RNA <50 copies/mL by week 8), and DNA levels remained higher in slow responders through week 104. Despite these differences, slow responders had higher CD4⁺ counts than rapid responders at baseline and through week 104. This study attempted to explain the apparent inconsistency between CD4⁺ T cells and persistent DNA by examining the association of TREC in rapid responders and slow responders with CD4⁺ T cells, and HIV RNA and DNA in this cohort.

Methods:  A total of 31 children (median age 5.6 years) were treated with efavirenz, nelfinavir, and 1 or 2 NRTI (PACTG 382) with sustained RNA <50 copies/mL during >104 weeks of HAART. Patients were naïve to PI and NNRTI. HIV DNA and plasma RNA, and TREC in PBMC were quantified by PCR at baseline, weeks 2, 4, 8, 20, 48, 80, and 104.

Results:  Although all subjects had a significant decline of RNA during the early stage of HAART and a gradual increase of CD4⁺ T cells throughout HAART, median TREC did not change significantly between baseline (7821/10⁵ PBMC) and each time point during HAART (p = 0.08 to 0.83, n = 21 to 31). TREC correlated significantly with CD4⁺ percentage (r = 0.43 to 0.53, p <0.03, n = 24 to 27) and a marginally significant correlation was observed with CD4⁺ counts (r = 0.37 to .44, p <0.08, n = 28 to 30) at weeks 2 to 8. Additionally, there was a persistent negative association between TREC and age throughout the study (r = -0.50 to 0.18, p = 0.03 to 0.38, n = 21 to 31). TREC and DNA correlated throughout HAART, most notably when all children reached RNA <50 copies/mL at weeks 48 to 80 (r = 0.61, p <0.01, n = 21 to 26). Also, the HIV DNA:TREC ratio was not significantly different for both groups at each time point (p >0.42, n = 21 to 31). TREC levels were significantly higher in slow responders (n = 15) at weeks 2 to 80 (p <0.05) despite this group having consistently higher DNA; thus suggesting how slow responders maintain higher CD4⁺ counts compared to rapid responders (n = 16).

Conclusions:  TREC levels are strongly associated with CD4⁺ percentages/counts and DNA in children with sustained, undetectable plasma RNA during HAART. The ability to maintain CD4⁺ T cells in the presence of an increased HIV burden is associated with an increase in TREC.  The equivalent ratio of HIV DNA:TREC in slow responders and rapid responders suggests that the release of TREC into the circulation results in a relatively fixed ratio of cells able to maintain HIV DNA despite sustained virologic suppression.

Keywords: TREC; HIV-1 DNA; child