Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

826 Abstracts 42254-42267

12th World AIDS Conference

Conclusions: These results provide further evidence that HU has the potential
to enhance the activity of the nucleoside analogues 3TC and ZDV, by increasing
their activation in relation to their respective endogenous triphosphates. In addition, although the activation of ddl was unaltered by HU, it is well documented
that HU reduces dATP levels, resulting in an increased ddATP/dATP ratio (Gao
et al., Proc. Natl. Acad. Sci. USA., 1995. 92; 8333-8337). These changes in
intracellular phosphorylation are linked to enhanced antiviral activity.
S42263 The pharmacokinetics and oral bioavailability of
lodenosine (F-ddA), a uniquely stable anti-HIV drug, in
adults with symptomatic HIV infection
James A. Kelley1, H. Ford, Jr.1, J.S. Roth1, L. Welles2, N.M. Malinowski1,
R.F. Little2, J.A. Leitzau2, L.A. Gillim2, J.P. Davignon3, J.S. Driscoll1,
R. Yarchoan2. Laboratory of Medicinal Chemistry NCI, Bldg 37 Room 5C02
NIH, Bethesda, MD 20892; 2HIV and AIDS Malignancy Branch NCI, Bethesda,
MD; 3US Bioscience Inc, W Conshohoche, PA, USA
Background: Lodenosine (2'-/-Fluoro-2', 3'-dideoxyadenosine; F-ddA) is a synthetic, sugar-fluorinated purine dideoxynucleoside analogue with excellent acid
and enzymatic stability, minimal in vitro cross-resistance to other reverse transcriptase inhibitors and in vitro anti-HIV activity and potency comparable to that
of didanosine (ddl). Oral F-ddA is currently undergoing Phase I clinical evaluation
in adult AIDS patients (see Abstract of R. F. Little et al.) where the objectives of
the pharmacokinetic studies are to determine F-ddA plasma kinetics, distribution,
metabolism, oral bioavailability (%F) and oral dosage form bioequivalance.
Methods: After an initial intravenous (iv) dose, F-ddA was administered orally
to patients twice daily as a 12-week course of treatment at 5 levels ranging from
0.2 to 3.2 mg/kg per dose. Serial blood samples and a complete 12-hr urine
collection were obtained for at least 3 patients at each level for the initial iv dose
and several oral doses. Resultant plasma concentrations (Cp) and the urinary
excretion of both F-ddA and its inosine metabolite F-ddl, which has equivalent
anti-HIV activity, were determined by high-performance liquid chromatographic
methods. Pharmacokinetic parameters were then calculated by noncompartmental methods.
Results: For a 90-min iv infusion, F-ddA steady-state Cp and F-ddl maximum
Cp were proportional to dose, achieving levels of 2.32 ptM and 12.2 pM, respectively, at 3.2 mg/kg. F-ddA was rapidly metabolized to F-ddl, accounting for 88 ~
3.5% (n = 9) of plasma F-ddA equivalents at doses of 0.8, 1.6 and 3.2 mg/kg. For
these dose levels, 92 ~ 12% of the administered dose was recovered in the urine.
When F-ddA was administered orally as a liquid while fasting, a mean %F of 67
~ 21% (n = 9) and a maximum F-ddl Cp -80% that of the iv dose was observed.
F-ddA capsules, administered while fasting or with food, were bioequivalent to
the liquid formulation.
Conclusions: Lodenosine administered orally while fasting or with food possesses excellent bioavailability, and preliminary results indicate that its capsule
dosage form exhibits apparent bioequivalence to the liquid formulation. The
greater enzymatic stability and superior bioavailabilty of F-ddA relative to ddl
(N.R. Hartman et al., Clin. Pharm. Ther. 1990; 47: 647-654) are reflected in its
production of a three-fold greater area-under-the-curve for equivalent 0.8 and 1.6
mg/kg doses.
42264 Control of zidovudine phosphorylation: The role of
increasing thymidine kinase activity versus reduction of
thymidine monophosphate (dTMP) levels
David J. Back, Danabhand Phiboonbanakit. Pharmacology University of
Liverpool, Ashton Street, Liverpool L69 3GE, UK
Aims: The cellular interaction between the phosphate intermediates of thymidine and zidovudine (ZDV) is not clearly understood. For example, methotrexate
(MTX), reduces intracellular thymidine triphosphate (dTTP) by interruption of the
thymidine de novo pathway prior to thymidine monophosphate (dTMP) formation,
and causes an upregulation of thymidine kinase (TK) activity. When MTX and
ZDV are combined an increase in ZDV triphosphate (ZDVTP) is seen. Competitive inhibition of thymidylate kinase (TMPK) by ZDV monophosphate (ZDVMP)
inhibits dTMP phosphorylation in vitro (Furman et al., Proc Natl Acad Sci USA
1986, 83: 8333-7). The aim was to study if the level of dTMP, rather than TK
activity, controls the extent of ZDV phosphorylation.
Methods: ZDV phosphorylation in the presence and absence of MTX (1 /iM)
was compared in U937 cells under two physiological conditions. U937 cells were
cultured in i) medium replenished daily and ii) medium exhausted of nutrients.
Exhausted medium causes a depletion of cellular kinases, therefore ZDV phosphorylation is not regulated by increased TK activity secondary to the depletion
of dTTP.
Results: In replenished culture medium, MTX significantly suppressed intracellular dTTP levels and significantly increased the intracellular concentrations of
all ZDV metabolites (p < 0.0005) compared to controls. In exhausted medium,
a significant reduction is dTTP and ZDVMP (p < 0.005) was evident compared
to replenished medium. MTX further reduced dTTP concentrations (p < 0.05)

compared to exhausted medium alone. In the absence of MTX, ZDVDP and
ZDVTP levels were similar in replenished and exhausted medium. Levels were
significantly increased (p < 0. 05) upon addition of MTX to exhausted medium
when compared to both media alone.
Conclusions: Increased TK activity may be partly responsible for the increase in ZDV phosphorylation, in particular ZDVMP levels. However, ZDVDP
and ZDVTP are not directly controlled by TK and the data suggest the increases

in ZDVDP and ZDVTP are due to a direct effect of reduced dTMP formation
rather than increase TK activity brought about by reduced dTTP. The results
support the hypothesis that dTMP plays an important role in the control of ZDV
phosphorylation.
42265 | Steady-state pharmacokinetic interactions between
ritonavir (RTV), nelfinavir (NFV), and the nelfinavir active
metabolite M8 (AG1402)
Charles Flexner12, A. Hsu3, B. Kerr4, C. Wong3, J. Gallant1, R. Anderson4,
M. Health - Chiozzi3. 1Osler 524, 600 N. Wolfe Street Baltimore, MD 21 287 -
5554; 2Johns Hopkins Baltimore MD; 3Abbott Laboratories Abbott Park IL;
4Agouron Pharmaceuticals Torrey Pines CA, USA
Objectives: To assess the pharmacokinetic interactions between RTV, NFV, and
NFV's major metabolite, M8, in HIV-infected subjects. M8 has anti-HIV activity
comparable to NFV in vitro.
Design: Inpatient pharmacokinetic (PK) component of a 48-week treatment
trial.
Methods: 10 subjects receiving RTV 400 mg b.i.d. + NFV 500 mg b.i.d. (Group
I, 4/10 female), and 10 subjects receiving RTV 400 mg b.i.d. + NFV 750 mg
b.i.d. (Group II, 1/10 female), had 12 hour PK sampling conducted after 2 weeks
and 5 weeks of dosing. Subjects took no other antiretrovirals or drugs known to
affect RTV or NFV PK. Data were not normally distributed, and non-parametric
comparisons were performed.
Results: AUC for NFV, M8, and RTV did not differ significantly between Weeks
2 and 5. Table shows median (IQR) for Week 5 AUCO-12 (mg*h/L); historical NFV
and M8 AUCO-8 for t.i.d. are scaled to 12 h for comparison.

Dose
RTV 400/NFV 500
RTV 400/NFV 750
NFV 750 tid.
RTV 600 b.i.d.

NFV AUC
28 (15-47)
26 (22-38)
20 (24-30)

M8 AUC
19(15-27)
25 (22-31)
6 (8-10)

RTV AUC
60 (57-91)
46 (42-58)
87(70-115)

historical data for standard regimens: N = 41 patients for NFV; N = 29 patients for RTV.
Differences in AUC between Group I and Group II were not significant (MannWhitney U test) except for RTV (p =.049, Week 2, and.059, Week 5); p-value
for M8 was 0.13.
Conclusion: Steady-state pharmacokinetic interactions between RTV and NFV
are likely to involve both cytochrome P450 inhibition and induction. Concurrent
RTV significantly increased concentrations of M8. In this small study, there was
little difference in the AUC of NFV, M8, or RTV between the two dose groups. The
optimal dose of this b.i.d. combination, especially if used with other antiretroviral
drugs, must be determined in long-term treatment trials.
|42266 Relationships between indinavir (IDV) pharmacokinetics
and antiviral activity in phase 1/11 trials
Jeffrey Chodakewitz, P. Deutsch, R. Leavitt, J. McCrea, M. Nessly, A. Sterrett,
G. Winchell. 1 Merck Research Laboratories, West Point Pennsylvania, USA
Background: The antiviral activity of indinavir has been clearly demonstrated to
be dose-dependent in early dose-finding studies at doses between 0.8 g/day to
2.4 g/day. In this analysis, we investigate the existence of a relationship between
antiviral activity and individual pharmacokinetic (PK) data at doses of 1.8-3.2
g/day, closer to the therapeutic dose range.
Methods: Data were pooled from all patients treated with IDV monotherapy in
Phase I and Phase II clinical protocols for which there were measurements of
both serum viral RNA decline at weeks 4 and/or 24 and plasma PK at steady
state. Total daily IDV dose ranged from 1.8 g/day to 3.2 g/day, given q8h or q6h.
There were 95 patients eligible for analyses from Merck sponsored protocols. The
majority (75/95) were receiving IDV at doses >2.4 g/day. Viral RNA (vRNA) at
baseline ranged from 13,000 to 649,840 copies/mL (median 74,320 copies/mL).
Results: There was no statistically significant correlation between vRNA decline (week 4 or 24) and either daily AUC (AUC over one dosing interval x doses
per day) or Ctrough. All correlation coefficients were between -0.15 and +0.1, and
p values >0.15. Daily AUC and Ctrough values were not distinguishable between
the groups with vRNA > 500 copies/mL versus <500 copies/mL.
Conclusions: Near the therapeutic dose range, the poor correlation of antiviral activity and PK suggests that factors other than PK play important roles
in inter-patient variability in response. Intra-patient variability in plasma PK may
also obscure small correlations. Additional investigations for correlations are warranted, but available data do not identify relevant PK target values to guide dose
modifications for individual patients on therapeutic doses of indinavir.
42267 Population pharmacokinetic analysis of ViraceptM
(nelfinavir mesylate) in HIV-infected patients enrolled in a
phase III clinical trial
Kimberley A. Jackson', S.E. Rosenbaum1, G. Yuen2, R. Anderson2
M.N. Dudley1. 1University of Rhode Island, Kingston, RI; 2AG134 3-511 Clinical
Study Group; Agouron Pharmaceuticals Inc., La Jolla, CA, USA
Objectives: To derive a population pharmacokinetic model for nelfinavir in HIVinfected patients and to determine the influence of patient characteristics and
concomitant medications on the pharmacokinetics of nelfinavir.