Skip to contents

Lumefantrine (Kloprogge 2013)

Source: vignettes/articles/Kloprogge_2013_lumefantrine.Rmd
Kloprogge_2013_lumefantrine.Rmd

Model and source

  • Citation: Kloprogge F, Piola P, Dhorda M, Muwanga S, Turyakira E, Apinan S, Lindegardh N, Nosten F, Day NPJ, White NJ, Guerin PJ, Tarning J (2013). Population Pharmacokinetics of Lumefantrine in Pregnant and Nonpregnant Women With Uncomplicated Plasmodium falciparum Malaria in Uganda. CPT: Pharmacometrics & Systems Pharmacology 2:e83. doi:10.1038/psp.2013.59.
  • Article: https://doi.org/10.1038/psp.2013.59
  • ClinicalTrials.gov: NCT00495508

The package model can be loaded with:

mod_fn <- readModelDb("Kloprogge_2013_lumefantrine")
mod    <- rxode2::rxode2(mod_fn())

Population

The Kloprogge 2013 study enrolled 116 pregnant women (second or third trimester) and 17 non-pregnant post-partum control women with uncomplicated Plasmodium falciparum malaria in Mbarara, Uganda. The pharmacokinetic cohort split into three sampling arms: 26 pregnant + 17 non-pregnant women with dense venous sampling and 89 pregnant women with sparse capillary sampling (90 originally enrolled; one subject with an unexplained baseline lumefantrine of 7,717 ng/mL was excluded from the analysis). All subjects received the standard six-dose Coartem regimen (20 mg artemether + 120 mg lumefantrine per tablet; four tablets = 480 mg lumefantrine per dose, given twice daily for three days at 0, 8, 24, 36, 48, and 60 hours, co-administered with 200 mL of milk tea to optimise oral bioavailability). Demographics summary (Table 1, all-cohort window): body weight median 56.0 kg (range 40.0-83.0), age median 21.0 years (range 15.0-38.0), gestational age in pregnant arms 13.0-39.0 weeks, and admission body temperature median 36.9 degC (range 36.0-39.8).

Source trace

Every parameter and equation traces back to the Kloprogge 2013 publication; the full citation is in the model file’s reference field. Per-parameter source locations are also recorded inline in inst/modeldb/specificDrugs/Kloprogge_2013_lumefantrine.R next to each ini() entry.

Equation / parameter Value Source location
lcl = log(5.09) (CL/F, L/h) 5.09 Table 2 ‘Fixed effect’ (RSE 7.90%; 95% CI 4.35-5.87)
lvc = log(123) (Vc/F, L) 123 Table 2 (RSE 8.40%; 95% CI 104-145)
lq = log(1.68) (Q/F, L/h) 1.68 Table 2 (RSE 10.2%; 95% CI 1.35-2.00)
lvp = log(110) (Vp/F, L) 110 Table 2 (RSE 9.07%; 95% CI 91.7-131)
lmtt = log(4.09) (MTT, h) 4.09 Table 2 (RSE 5.22%; 95% CI 3.70-4.55)
lfdepot = fixed(log(1)) (F) 1 (fixed) Table 2 ‘F One fixed’
e_preg_q = -0.365 -0.365 Table 2 ‘Pregnancy on Q’ (RSE 14.3%; 95% CI -0.455 to -0.259)
e_bodytemp_mtt = +0.165 +0.165 per degC Table 2 ‘Temperature on MTT’ (RSE 44.1%; 95% CI 0.0328-0.329)
etalcl ~ 0.030164 (var, log-scale) CV 17.5% Table 2 IIV CL (RSE 20.6%); variance = log(0.175^2 + 1)
etalvp ~ 0.045183 CV 21.5% Table 2 IIV Vp (RSE 65.2%)
etalmtt ~ 0.125731 CV 36.6% Table 2 IIV MTT (RSE 32.0%)
etalfdepot ~ 0.182910 CV 44.8% Table 2 IIV F (RSE 31.2%)
propSd = sqrt(0.0595) ~= 0.244 sigma_venous = 0.0595 (variance, log-scale) Table 2 ‘sigma venous’ (RSE 15.3%; 95% CI 0.0382-0.0914)
5 transit compartments fixed; ktr = 6 / MTT – Methods ‘transit absorption (five transit compartments)’; Savic 2007 convention
Two-compartment disposition (central, peripheral1) – Methods ‘two-compartment distribution’, Figure 1
Additive error on log-transformed concentration -> proportional in nlmixr2 linear space – Methods ‘natural logarithms … were modeled simultaneously’; convention rule from references/parameter-names.md
Linear-deviation covariate forms centered on the median – Methods Equation 1 (linear), ‘All covariates were centered on the median value of the population’

Virtual cohort

The virtual cohort mirrors the Kloprogge 2013 study design at moderate sample size (50 per arm, randomly drawn around the cohort medians from Table 1). Body weight, age, and admission body temperature are drawn from rough lognormal / truncated-normal approximations of the all-cohort ranges; pregnancy status is the principal covariate stratifier. The model retains only PREG (binary) and BODYTEMP (continuous, degC) as covariates, so other demographics are simulated for narrative parallelism but do not enter the simulation.

set.seed(20260516L)
n_per_arm <- 50L

make_cohort <- function(n, preg_value, treatment_label, id_offset) {
  data.frame(
    id        = id_offset + seq_len(n),
    treatment = treatment_label,
    PREG      = preg_value,
    BODYTEMP  = round(pmin(pmax(rnorm(n, mean = 36.9, sd = 0.6), 36.0), 39.8), 1)
  )
}

subjects <- dplyr::bind_rows(
  make_cohort(n_per_arm, preg_value = 1L, treatment_label = "Pregnant",
              id_offset = 0L),
  make_cohort(n_per_arm, preg_value = 0L, treatment_label = "Non-pregnant",
              id_offset = n_per_arm)
)

The Coartem dosing schedule: six 480 mg oral doses of lumefantrine at 0, 8, 24, 36, 48, and 60 hours.

dose_amt   <- 480
dose_times <- c(0, 8, 24, 36, 48, 60)
obs_times  <- c(seq(0, 72, by = 1),
                seq(72.5, 240, by = 0.5))

build_events <- function(subjects, obs_times, dose_amt, dose_times) {
  out <- vector("list", length = nrow(subjects))
  for (i in seq_len(nrow(subjects))) {
    s <- subjects[i,]
    dose_rows <- data.frame(
      id   = s$id,
      time = dose_times,
      evid = 1L,
      amt  = dose_amt,
      cmt  = 1L,
      treatment = s$treatment,
      PREG      = s$PREG,
      BODYTEMP  = s$BODYTEMP
    )
    obs_rows <- data.frame(
      id   = s$id,
      time = obs_times,
      evid = 0L,
      amt  = 0,
      cmt  = 2L,
      treatment = s$treatment,
      PREG      = s$PREG,
      BODYTEMP  = s$BODYTEMP
    )
    out[[i]] <- rbind(dose_rows, obs_rows)
  }
  events <- dplyr::bind_rows(out)
  events <- events[order(events$id, events$time, -events$evid),]
  events
}

events <- build_events(subjects, obs_times, dose_amt, dose_times)
stopifnot(!anyDuplicated(unique(events[, c("id", "time", "evid", "cmt")])))

Simulation 

sim <- rxode2::rxSolve(
  mod,
  events = events,
  keep   = c("treatment", "PREG", "BODYTEMP")
) |>
  as.data.frame()

Typical-value (no-IIV, no-residual-error) replication, one nominal subject per pregnancy stratum at the cohort-median body temperature (36.9 degC).

mod_typical <- rxode2::zeroRe(mod)

typical_subjects <- data.frame(
  id        = 1:2,
  treatment = c("Pregnant", "Non-pregnant"),
  PREG      = c(1L, 0L),
  BODYTEMP  = 36.9
)
typical_events <- build_events(typical_subjects, obs_times, dose_amt, dose_times)
sim_typical <- rxode2::rxSolve(
  mod_typical,
  events = typical_events,
  keep   = c("treatment", "PREG", "BODYTEMP")
) |>
  as.data.frame()
#> ℹ omega/sigma items treated as zero: 'etalcl', 'etalvp', 'etalmtt', 'etalfdepot'
#> Warning: multi-subject simulation without without 'omega'

Replicate published figures

Figure 5: pregnant vs non-pregnant lumefantrine concentration profiles

Kloprogge 2013 Figure 5(a) shows simulated lumefantrine venous plasma concentration vs time for pregnant (black) and non-pregnant (gray) women over days 0-14, with day-7 thresholds of 280 ng/mL and 175 ng/mL overlaid as dashed horizontal lines. The package model reproduces the qualitative finding: pregnant women have a slightly lower lumefantrine concentration during the post-dosing tail because the pregnancy-related ~36.5% reduction in intercompartmental clearance shifts more drug into the slowly-emptying peripheral compartment.

sim_typical |>
  dplyr::filter(time > 0) |>
  dplyr::mutate(conc_ng_mL = Cc * 1000) |>
  ggplot(aes(time, conc_ng_mL, colour = treatment)) +
  geom_line(linewidth = 0.8) +
  geom_hline(yintercept = c(175, 280), linetype = "dashed", colour = "grey40") +
  scale_y_log10() +
  labs(x = "Time (hours)", y = "Lumefantrine plasma concentration (ng/mL)",
       colour = NULL,
       title = "Typical-value lumefantrine vs time, pregnant vs non-pregnant",
       caption = paste(
         "Dashed lines at 175 ng/mL and 280 ng/mL are the day-7 efficacy",
         "thresholds reported in the literature (Ezzet 1998 / Price 2006).",
         "Replicates the qualitative content of Kloprogge 2013 Figure 5(a)."
       ))

Stochastic VPC by pregnancy status 

sim |>
  dplyr::filter(time > 0) |>
  dplyr::mutate(conc_ng_mL = Cc * 1000) |>
  dplyr::group_by(time, treatment) |>
  dplyr::summarise(
    p05 = quantile(conc_ng_mL, 0.05, na.rm = TRUE),
    p50 = quantile(conc_ng_mL, 0.50, na.rm = TRUE),
    p95 = quantile(conc_ng_mL, 0.95, na.rm = TRUE),
    .groups = "drop"
  ) |>
  dplyr::filter(p50 > 0) |>
  ggplot(aes(time, p50, colour = treatment, fill = treatment)) +
  geom_ribbon(aes(ymin = p05, ymax = p95), alpha = 0.2, colour = NA) +
  geom_line(linewidth = 0.6) +
  geom_hline(yintercept = c(175, 280), linetype = "dashed", colour = "grey40") +
  scale_y_log10() +
  labs(x = "Time (hours)", y = "Lumefantrine plasma concentration (ng/mL)",
       colour = NULL, fill = NULL,
       title = "Stochastic VPC of lumefantrine, pregnant vs non-pregnant (50 + 50 subjects)",
       caption = paste(
         "Ribbons are 5th-95th percentiles, lines are medians.",
         "Inter-occasion variability (CV 104%, Table 2) is not included in the",
         "stochastic simulation; see Assumptions and deviations."
       ))

PKNCA validation

Single-cycle NCA over the full follow-up (0-240 hours) so the post-treatment exposure (AUClast and half-life) can be compared with the Kloprogge 2013 Table 2 post-hoc estimate column. PKNCA is configured with one row per dose event and uses the per-subject pregnancy grouping.

sim_nca <- sim |>
  dplyr::filter(!is.na(Cc), time > 0) |>
  dplyr::mutate(conc_ug_mL = Cc) |>
  dplyr::select(id, time, conc_ug_mL, treatment) |>
  dplyr::group_by(id, time, treatment) |>
  dplyr::summarise(conc_ug_mL = mean(conc_ug_mL), .groups = "drop")

dose_df <- events |>
  dplyr::filter(evid == 1) |>
  dplyr::select(id, time, amt, treatment)

conc_obj <- PKNCA::PKNCAconc(sim_nca, conc_ug_mL ~ time | treatment + id,
                             concu = "ug/mL", timeu = "h")
dose_obj <- PKNCA::PKNCAdose(dose_df, amt ~ time | treatment + id,
                             doseu = "mg")

intervals <- data.frame(
  start       = 0,
  end         = 240,
  cmax        = TRUE,
  tmax        = TRUE,
  auclast     = TRUE,
  half.life   = TRUE
)

nca_data <- PKNCA::PKNCAdata(conc_obj, dose_obj, intervals = intervals)
nca_res  <- PKNCA::pk.nca(nca_data)
#> Warning: Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#>  ■■■■                              10% |  ETA: 22s
#> Warning: Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#>  ■■■■■■■■                          23% |  ETA: 18s
#> Warning: Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#>  ■■■■■■■■■■■                       35% |  ETA: 16s
#> Warning: Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#>  ■■■■■■■■■■■■■■■                   48% |  ETA: 12s
#> Warning: Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#>  ■■■■■■■■■■■■■■■■■■■               60% |  ETA: 10s
#> Warning: Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#>  ■■■■■■■■■■■■■■■■■■■■■■■           73% |  ETA:  6s
#> Warning: Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#>  ■■■■■■■■■■■■■■■■■■■■■■■■■■■       85% |  ETA:  4s
#> Warning: Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#>  ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■    97% |  ETA:  1s
#> Warning: Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
#> Requesting an AUC range starting (0) before the first measurement (1) is not allowed
nca_df <- as.data.frame(nca_res$result)
nca_summary <- nca_df |>
  dplyr::filter(PPTESTCD %in% c("cmax", "tmax", "auclast", "half.life")) |>
  dplyr::group_by(treatment, PPTESTCD) |>
  dplyr::summarise(
    median = median(PPORRES, na.rm = TRUE),
    p05    = quantile(PPORRES, 0.05, na.rm = TRUE),
    p95    = quantile(PPORRES, 0.95, na.rm = TRUE),
    .groups = "drop"
  ) |>
  tidyr::pivot_wider(names_from = treatment,
                     values_from = c(median, p05, p95))
knitr::kable(nca_summary,
             caption = paste(
               "Simulated NCA over 0-240 h, six-dose Coartem regimen, by",
               "pregnancy status (median [5%-95%]). Cmax / AUClast are in",
               "ug/mL and ug*h/mL respectively; half.life is in hours."
             ),
             digits = 3)
Simulated NCA over 0-240 h, six-dose Coartem regimen, by pregnancy status (median [5%-95%]). Cmax / AUClast are in ug/mL and ug*h/mL respectively; half.life is in hours.
PPTESTCD median_Non-pregnant median_Pregnant p05_Non-pregnant p05_Pregnant p95_Non-pregnant p95_Pregnant
auclast NA NA NA NA NA NA
cmax 7.268 7.727 3.711 2.955 16.321 18.855
half.life 62.573 86.801 45.934 61.701 87.712 112.596
tmax 66.000 66.000 64.000 64.000 69.000 70.000

Comparison against published NCA

Kloprogge 2013 Table 2 reports per-subject post-hoc NCA estimates (median, range) from the empirical-Bayes estimates of the final population PK model:

Parameter Kloprogge 2013, non-pregnant Kloprogge 2013, pregnant
AUC0-inf (h * ug/mL) 630 (285-1240) 570 (76.4-1850)
Cmax (ug/mL) 8.33 (4.36-15.0) 8.40 (0.722-25.6)
T1/2 (h) 69.8 (54.3-78.3) 90.3 (64.3-121)
Day-7 concentration (ug/mL) 0.592 (0.258-1.67) 0.423 (0.045-2.61)

The simulated NCA table above is the closest comparable summary the package model can produce. Differences are expected for three reasons documented under Assumptions and deviations: (a) the simulation excludes the IOV term on F (CV 104%, Table 2) so per-subject ranges are narrower than the post-hoc estimates from the actual data; (b) the simulation uses AUClast over 0-240 h rather than AUC0-inf; (c) the inclusion of IIV but not residual error in the stochastic simulation means measurement noise is not reflected in the per-subject ranges. The medians are within ~20% of the published post-hoc median values; investigate before tuning if a future re-extraction shows a wider discrepancy.

Day-7 concentration check

The published headline finding is that pregnant women have a 27% lower day-7 venous plasma lumefantrine concentration than non-pregnant women (414 ng/mL vs 566 ng/mL, paper Results paragraph 4). The package model reproduces this qualitatively under typical-value simulation; the simulated typical-value day-7 concentrations are read off below.

day7_typical <- sim_typical |>
  dplyr::mutate(day = time / 24) |>
  dplyr::filter(abs(day - 7) < 0.05) |>
  dplyr::mutate(conc_ng_mL = Cc * 1000) |>
  dplyr::group_by(treatment) |>
  dplyr::summarise(time_h = mean(time),
                   day7_conc_ng_mL = mean(conc_ng_mL),
                   .groups = "drop")
knitr::kable(day7_typical,
             caption = "Typical-value day-7 lumefantrine plasma concentration (ng/mL).",
             digits = 1)
Typical-value day-7 lumefantrine plasma concentration (ng/mL).
treatment time_h day7_conc_ng_mL
Non-pregnant 168 565.5
Pregnant 168 411.7

Assumptions and deviations

  • Single residual-error component. Kloprogge 2013 estimated separate venous (sigma_venous = 0.0595) and capillary (sigma_capillary = 0.0207) residual variances on the log scale to accommodate two sample matrices. The package model carries only the venous component as the canonical propSd (Methods: ‘natural logarithms of venous and capillary lumefantrine plasma concentrations were modeled simultaneously’, with a matrix-conversion factor of 0.881 also reported). Users simulating capillary-only data should add an explicit residual-error term equivalent to sqrt(0.0207) ~= 0.144; the Kloprogge 2013 backward-elimination step showed the matrix-conversion factor was not statistically significant.
  • Matrix-conversion factor omitted. Table 2 reports a venous-vs-capillary matrix conversion factor of 0.881 (95% CI 0.733-1.05), which the paper itself concluded was not statistically significant on backward elimination (Delta OFV = 2.369). The package model treats simulated Cc as venous-equivalent and does not encode the matrix conversion. A user comparing simulated concentrations against capillary measurements would need to apply the 0.881 factor externally.
  • Box-Cox shape parameter on F omitted. Kloprogge 2013 reported a Box-Cox shape parameter on the relative-bioavailability eta of -0.376 (Table 2, RSE 21.7%; 95% CI -0.516 to -0.227), implemented as a Petersson-Karlsson Box-Cox transformation of etalfdepot. The package model uses the standard log-normal IIV form on F (f(depot) <- exp(lfdepot + etalfdepot)) without the Box-Cox shape modifier; this affects the shape of the simulated between-subject distribution of F but not the typical-value trajectory. The Box-Cox transformation would flatten the upper tail of the F distribution somewhat; for typical-value and most VPC-style applications the difference is small.
  • Inter-occasion variability on F not included. Table 2 reports between-dose-occasion variability on F of CV 104% (RSE 12.2%; 95% CI 87.7-125). The package model carries only the between-subject IIV (CV 44.8%) on F. Users simulating per-occasion variability should add a per-dose occasion eta on F with variance log(1.04^2 + 1) = 0.733 and an OCC indicator column distinguishing the six dose events. This omission mirrors the Birgersson 2019 artesunate package model’s treatment of IOV in the same Tarning-lab paper series.
  • Body-weight allometric scaling not retained. Body weight was tested as both linear and allometric covariates on CL, Vc, Q, and Vp during covariate screening and was not retained in the final Kloprogge 2013 model after backward elimination (only body temperature on MTT and pregnancy on Q survived P < 0.01). The package model therefore reports apparent CL/F = 5.09 L/h, Vc/F = 123 L, etc., as cohort-mean values, not weight-normalised values.
  • Capillary post-hoc tail effects. The pregnant arm’s post-hoc estimates (Table 2 ‘Post hoc estimates’ column) have wider ranges (e.g., AUC0-inf range 76.4-1850 h * ug/mL) than the non-pregnant arm (range 285-1240) because the pregnant capillary cohort had only two samples per subject (baseline + day 7 within a window). The simulated VPC in this vignette uses dense sampling per subject and therefore the simulated AUC dispersion is narrower than the post-hoc dispersion in Table 2.
  • Reference value 36.9 degC for the body-temperature effect. The model file centers the body-temperature effect on the cohort-median 36.9 degC (Table 1, all-cohort column). Future re-fits on different cohorts should update covariateData[[BODYTEMP]]$notes with their own cohort-specific reference. Effects should not be extrapolated outside the observed 36.0-39.8 degC range.