572 
CYP2C19 Genetic Variants Affect Nelfinavir and M8 Pharmacokinetics and Virologic Response in HIV-1 infected Children Receiving HAART
Akihiko Saitoh*1, E Sarles1, E Capparelli1, F Aweeka2, A Kovacs3, S Buchett4, A Wiznia5, S Nachman6, T Fenton7, and S Spector1
1Univ of California, San Diego, US; 2Univ of California, San Francisco, US; 3Univ of Southern California, Los Angeles, US; 4Harvard Med Sch, Boston, MA, US; 5Jacobi Med Ctr, Bronx, NY, US; 6State Univ of New York at Stony Brook, US; and 7Harvard Sch of Publ Hlth, Boston, MA, US
Background: Nelfinavir (NFV) is transported
via P-glycoprotein and metabolized to active metabolite (M8) by cytochrome P450
(CYP) 2C19. We previously reported that genetic variants encoding for
P-glycoprotein (ABCB1) were associated with NFV pharmacokinetics and
virologic response in HIV-1-infected children (n = 71) receiving HAART. This
study extended the number of subjects from different cohorts to investigate the
association between genetic variants of CYP and ABCB1 on NFV pharmacokinetics
and the response to HAART in HIV-1-infected children.
Methods: We evaluated 152 HIV-1-infected
children from Pediatric AIDS Clinical Trials Group (PACTG) 366 and 377
receiving NFV as a component of HAART. Concomitant antiretrovirals included NRTI
alone (n = 23), ritonavir (RTV, n = 40), nevirapine (NVP, n
= 60), and RTV + NVP (n = 29). Genomic DNA from peripheral blood mononuclear
cells was tested for ABCB1 and CYP genetic variants by real-time polymerase
chain reaction (RT-PCR). Intensive (70%) and sparse (30%) pharmacokinetics were
used to calculate NFV and M8 levels. CD4+ T cell and HIV-1 RNA were
measured during HAART.
Results: The parameters associated with NFV oral
clearance (CL/F) included age (p <0.001), RTV use (p = 0.001),
CYP2C19-G681A (p = 0.001), and ABCB1-C3435T (p = 0.03)
by univariate analyses. The NFV CL/F (L/h/kg) in children with the CYP2C19-681-G/G
(wild type, 1.86, n = 102) was higher than for those with the G/A (heterozygous
variant, 1.33, n = 42, p = 0.001) and A/A (homozygous variant, 1.29,
n = 8, p = 0.02). In contrast, the ratio of M8 to NFV (M8:NFV)
was higher in children with the CYP2C19-681-A/A than for those with the
G/A (p = 0.003) and the G/G (p <0.001). Furthermore, the
children with the CYP2C19-681-G/A or A/A achieving HIV-1 RNA <400 copies/mL
(68%) was higher than those with the G/G (46%) at week 24 (p = 0.01). The
immunologic recovery was not associated with the genotypes at weeks 12 or 24 (p
>0.44). The allelic frequencies of the CYP2C19-681-A were higher
in blacks (0.61) than whites (0.15), or Hispanics (0.12) (p <0.001). The
multivariate analyses demonstrated that the CYP2C19-G681A, age, and RTV
use were independently associated with NFV CL/F (p <0.001), but not ABCB1-C3435T
(p = 0.13).
Conclusions: This is the first study demonstrating
the association between the CYP2C19 genotypes and NFV pharmacokinetics, and
the virologic response to HAART in HIV-1-infected children. These findings
suggest that CYP2C19 genetic variants play an important role in NFV pharmacokinetics
and virologic response in HIV-1-infected children.
|