Nothing
R/est.nca.R
In ncappc: NCA Calculations and Population Model Diagnosis
Defines functions
est.nca
Documented in
est.nca
# R function to estimate NCA metrics
# Chayan, 12/2014
# roxygen comments
#' Estimate individual NCA metrics.
#'
#' Estimates a comprehensive set of NCA metrics
#' for a given individual using concentration vs. time data.
#'
#' \pkg{est.nca} estimates a comprehensive set of NCA metrics using the
#' concentration-time profile of an individual. NCA metrics are estimated
#' according to traditional PK calculations. The names of the various NCA
#' metrics estimated in this package are assigned mainly following the names
#' used in WinNonlin. This package accepts any of the three different types of
#' drug administration, (i) iv-bolus, (ii) iv-infusion and (iii) extravascular;
#' \pkg{ncappc} also can accept both non-steady state and steady-state data.
#' The NCA metrics that are estimated and reported by \pkg{ncappc} are listed
#' below.
#' \itemize{
#' \item \strong{C0} is the initial concentration at the dosing time. It is
#' the observed concentration at the dosing time, if available. Otherwise
#' it is approximated using the following rules.
#' \item \strong{Cmax, Tmax and Cmax_D} are the value and the time of maximum
#' observed concentration, respectively. If the maximum concentration is not
#' unique, the first maximum is used. For steady state data, The maximum value
#' between the dosing intervals is considered. Cmax_D is the dose normalized
#' maximum observed concentration.
#' \item \strong{Clast and Tlast} are the last measurable positive
#' concentration and the corresponding time, respectively.
#' \item \strong{AUClast} is the area under the concentration vs. time curve
#' from the first observed to last measurable concentration.
#' \item \strong{AUMClast} is the first moment of the concentration vs. time
#' curve from the first observed to last measurable concentration.
#' \item \strong{MRTlast} is the mean residence time from the first observed
#' to last measurable concentration.
#' \item \strong{No_points_Lambda_z} is the number of observed data points
#' used to determine the best fitting regression line in the elimination
#' phase.
#' \item \strong{AUC_pBack_Ext_obs} is the percentage of AUCINF_obs that is
#' contributed by the back extrapolation to estimate C0.
#' \item \strong{AUC_pBack_Ext_pred} is the percentage of AUCINF_pred that is
#' contributed by the back extrapolation to estimate C0.
#' \item \strong{AUClower_upper} is the AUC under the concentration-time
#' profile within the user-specified window of time provided as the
#' "AUCTimeRange" argument. In case of empty "AUCTimeRange" argument,
#' AUClower_upper is the same as AUClast.
#' \item \strong{Rsq, Rsq_adjusted and Corr_XY} are regression coefficient
#' of the regression line used to estimate the elimination rate constant, the
#' adjusted value of Rsq and the square root of Rsq, respectively.
#' \item \strong{Lambda_z} is the elimination rate constant estimated from the
#' regression line representing the terminal phase of the concentration-time
#' data.
#' \item \strong{Lambda_lower and Lambda_upper} are the lower and upper limit
#' of the time values from the concentration-time profile used to estimate
#' Lambda_z, respectively, in case the "LambdaTimeRange" is used to specify
#' the time range.
#' \item \strong{HL_Lambda_z} is terminal half-life of the drug.
#' \item \strong{AUCINF_obs and AUCINF_obs_D} are AUC estimated from the first
#' sampled data extrapolated to infinity and its dose normalized version,
#' respectively. The extrapolation in the terminal phase is based on the last
#' observed concentration Clast_obs.
#' \item \strong{AUC_pExtrap_obs} is the percentage of the AUCINF_obs that is
#' contributed by the extrapolation from the last sampling time to infinity.
#' \item \strong{AUMCINF_obs} is AUMC estimated from the first sampled data
#' extrapolated to infinity. The extrapolation in the terminal phase is based
#' on the last observed concentration.
#' \item \strong{AUMC_pExtrap_obs} is the percentage of the AUMCINF_obs that
#' is contributed by the extrapolation from the last sampling time to infinity.
#' \item \strong{Vz_obs} is the volume of distribution estimated based on
#' total AUC
#' \item \strong{Cl_obs} is total body clearance.
#' \item \strong{AUCINF_pred and AUCINF_pred_D} are AUC from the first sampled
#' data extrapolated to infinity and its dose normalized version, respectively.
#' The extrapolation in the terminal phase is based on the last predicted
#' concentration obtained from the regression line used to estimate Lambda_z
#' (Clast_pred).
#' \item \strong{AUC_pExtrap_pred} is the percentage of the AUCINF_pred that
#' is contributed by the extrapolation from the last sampling time to infinity.
#' \item \strong{AUMCINF_pred} is AUMC estimated from the first sampled data
#' extrapolated to infinity. The extrapolation in the terminal phase is based
#' on the last predicted concentration obtained from the regression line used
#' to estimate Lambda_z (Clast_pred).
#' \item \strong{AUMC_pExtrap_pred} is the percentage of the AUMCINF_pred that
#' is contributed by the extrapolation from the last sampling time to infinity.
#' \item \strong{Vz_pred} is the volume of distribution estimated based on
#' AUCINF_pred.
#' \item \strong{Cl_pred} is the total body clearance estimated based on
#' AUCINF_pred.
#' \item \strong{MRTINF_obs} is the mean residence time from the first
#' sampled time extrapolated to infinity based on the last observed
#' concentration (Clast_obs).
#' \item \strong{MRTINF_pred} is the mean residence time from the first
#' sampled time extrapolated to infinity based on the last predicted
#' concentration obtained from the regression line used to estimate Lambda_z
#' (Clast_pred).
#' \item \strong{Tau} is the dosing interval for steady-state data.
#' \item \strong{Cmin and Tmin} are the minimum concentration between 0 and
#' Tau and the corresponding time, respectively.
#' \item \strong{Cavg} is the average concentration between 0 and Tau for
#' steady-state data.
#' \item \strong{AUCtau and AUMCtau} are AUC and AUMC between 0 and Tau for
#' steady-state data.
#' \item \strong{Clss} is an estimate of the total body clearance for
#' steady-state data.
#' \item \strong{Vss_obs and Vss_pred} are estimated volume of distribution at
#' steady-state based on Clast_obs and Clast_pred, respectively.
#' \item \strong{p_Fluctuation} is the percentage of the fluctuation of the
#' concentration between 0 and Tau for steady-state data.
#' \item \strong{Accumulation_Index} is \eqn{1/(1-e^(-\lambda_z*\tau))}
#' }
#' @param time Numeric array for time
#' @param conc Numeric array for concentration
#' @param backExtrp If back-extrapolation is needed for AUC (TRUE or FALSE)
#' (\strong{FALSE})
#' @param negConcExcl Exclude -ve conc (\strong{FALSE})
#' @param doseType Steady-state (ss) or non-steady-state (ns) dose
#' (\strong{"ns"})
#' @param adminType Route of administration
#' (iv-bolus,iv-infusion,extravascular) (\strong{"extravascular"})
#' @param doseAmt Dose amounts (\strong{"NULL"})
#' @param method Method to estimate AUC. The \code{"linear"} method applies the linear
#' trapezoidal rule to estimate the area under the curve. The \code{"log"} method
#' applies the logarithmic trapezoidal rule to estimate the area under the
#' curve. The \code{"linearup-logdown"} method applies the linear trapezoidal rule to
#' estimate the area under the curve for the ascending part of the curve and
#' the logarithmic trapezoidal rule to estimate the area under the curve for
#' the descending part of the curve.
#' @param AUCTimeRange User-defined window of time used to estimate AUC
#' (\strong{"NULL"})
#' @param LambdaTimeRange User-defined window of time to estimate elimination
#' rate-constant (\strong{"NULL"})
#' @param LambdaExclude User-defined excluded observation time points for
#' estimation of elimination rate-constant (\strong{"NULL"})
#' @param Tau Dosing interval for steady-state data (\strong{"NULL"})
#' @param doseTime Dose time prior to the first observation for steady-state
#' data (\strong{\code{NULL}})
#' @param TI Infusion duration (\strong{"NULL"})
#' @param simFile Name of the simulated concentration-time data if present
#' (\strong{"NULL"})
#' @param dset Type, i.e., observed or simulated concentration-time data set
#' ("obs" or "sim") (\strong{"obs"})
#' @param onlyNCA If \code{TRUE} only NCA is performed and ppc part is ignored
#' although simFile is not \code{NULL}. Default is \strong{\code{FALSE}}
#' @param extrapolate Should the function extrapolate from the last observation to infinity?
#' @param sparse_compute Should NCA metrics be computed even with only one sample?
#' @param force_extrapolate Extrapolate AUC_inf with sparse data. Sparse data is defined as
#' fewer than 3 points after Cmax for non-bolus input. In that case, Cmax is included in the extrapolation
#' and only one extra point in required for extrapolation.
#' @param ... Arguments passed from other functions. Not used.
#'
#' @return An array of estimated NCA metrics
#' @export
#'
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
# function to estimate NCA parameters
est.nca <- function(time,
conc,
backExtrp=FALSE,
negConcExcl=FALSE,
doseType="ns",
adminType="extravascular",
doseAmt=NULL,
method="linearup-logdown",
AUCTimeRange=NULL,
LambdaTimeRange=NULL,
LambdaExclude=NULL,
doseTime=doseTime,
Tau=NULL,
TI=NULL,
simFile=NULL,
dset="obs",
onlyNCA=FALSE,
extrapolate=FALSE,
sparse_compute=FALSE,
force_extrapolate = FALSE,...){
"tail" <- "head" <- "lm" <- "coef" <- "arsq" <- NULL
rm(list=c("tail","head","lm","coef","arsq"))
# Set Tau to NA for non steady-state data
Tau <- ifelse(doseType=="ns", NA, Tau)
ntime <- as.numeric(time) # Save time to ntime
nconc <- as.numeric(conc) # Save conc to nconc
# check that method is defined as it should be
if(!(method %in% c("linear","log", "linearup-logdown")))
stop(paste(method,"is not one of 'linear','log', or 'linearup-logdown'"))
# Exclude zero or negative concentration from calculations
if(negConcExcl){
zid <- which (nconc < 0)
if(length(zid) > 0){ntime <- ntime[-zid]; nconc <- nconc[-zid]}
}
# Check if the number of observations is at least 3, otherwise skip the operation
if(length(nconc) >= 2 | sparse_compute){
# Steady state data
if(doseType == "ss"){
if(is.null(doseTime)){
ssIdx1 <- min(which(nconc>0)) # Index for the first observation with non-zero conc
}else{
ssIdx1 <- min(which(abs(round(ntime,0)-doseTime)==0)) # Index for the first observation within the steady state interval
}
sc1 <- nconc[ssIdx1] # Steady state 1st observed conc
st1 <- ntime[ssIdx1] # Steady state 1st observed time
ssIdx2 <- which.min(abs(ntime-(st1+Tau))) # Index for the last ss time (Tau added to first time and find the nearest time to that number)
sc2 <- nconc[ssIdx2] # Steady state last observed conc
st2 <- ntime[ssIdx2] # Steady state last observed time
# steady state observations
nconc <- nconc[which(ntime>=st1 & ntime<=st2)]
ntime <- ntime[which(ntime>=st1 & ntime<=st2)]
ssnPt <- length(ntime)
if(ssnPt != 0){
mIdx <- which(nconc == min(nconc[nconc!=0]))[1]
Tmin <- ntime[mIdx]
Cmin <- nconc[mIdx]
AUCtau <- 0.
AUMCtau <- 0.
}
}
# Calculation of C0
noBackTime <- TRUE
if(backExtrp){
if(ntime[1]==0){
C0 <- as.numeric(nconc[1])
}else{
if(adminType=="extravascular" | adminType=="iv-infusion"){
if(doseType=="ss" && ssnPt!=0){
C0 <- as.numeric(min(nconc)) # Min obsreved for SS data
}else{
C0 <- 0 # C0 is set to 0 for single dose data
nconc <- c(C0, nconc)
ntime <- c(0, ntime)
noBackTime <- FALSE
}
}else if(adminType == "iv-bolus"){
slope <- (nconc[2]-nconc[1])/(ntime[2]-ntime[1])
tconc <- nconc[1:2]
ttime <- ntime[1:2]
lmEx <- FALSE
if(!is.null(LambdaExclude)){
for(i in 1:length(LambdaExclude)){
if(grepl("^[-]?[0-9]*[.]?[0-9]*[eE]?[-]?[0-9]*[.]?[0-9]*$", LambdaExclude[i])){
# Check if the LambdaExclude is numeric
if(any(ttime%in%LambdaExclude[i])) lmEx <- TRUE
}else{
# Check if the LambdaExclude is logical
if(any(eval(parse(text=paste0("ttime",LambdaExclude[i]))))) lmEx <- TRUE
}
}
}
if(slope >= 0 | nconc[1]==0 | nconc[2]==0 | lmEx){
C0 <- as.numeric(nconc[nconc>0][1])
}else{
C0 <- as.numeric(exp((ntime[1]*log(nconc[2])-ntime[2]*log(nconc[1]))/(ntime[1]-ntime[2])))
nconc <- c(C0, nconc) # extrapolation via log-linear regression
ntime <- c(0, ntime)
noBackTime <- FALSE
}
}
}
}
# Some of the NCA metrics including Cmax and Tmax
otime <- ntime - min(ntime) # Time is off-set to zero for estimation of the 1st moment
nPt <- length(nconc) # No. of data points
Cmax <- max(nconc)
mxId <- tail(which(nconc == Cmax),1)
Tmax <- ntime[mxId]
Cmax_D <- ifelse(!is.null(doseAmt), nconc[mxId]/doseAmt, NA)
lIdx <- max(which(nconc == tail(nconc[nconc>0],1))) # Index for last positive concentration
Tlast <- ntime[lIdx]
Clast <- nconc[lIdx]
if(extrapolate){
###### Prepare for extrapolation for the terminal elimination ######
# Determine the lower and upper indeces of the time vector used for Lambda calculation
llower <- ifelse(adminType!="iv-bolus", mxId+1, mxId) # Exclude Cmax from elimination phase extrapolation for non-bolus dose
if(force_extrapolate) llower <- mxId
lupper <- nPt
# Conc and Time for the elimination phase
lconc <- nconc[llower:lupper]
ltime <- ntime[llower:lupper]
# exclude points with zero or negative concentration
zid <- which(lconc <= 0)
if(length(zid) > 0){
lconc <- lconc[-zid]
ltime <- ltime[-zid]
}
# Re-define llower and lupper if time range is specified for Lambda calculation
if(!is.null(LambdaTimeRange)){
Lambda_z_lower <- sort(LambdaTimeRange)[1]
Lambda_z_upper <- sort(LambdaTimeRange)[2]
if(length(ltime[ltime>=Lambda_z_lower & ltime<=Lambda_z_upper]) >= 3){
llower <- head(which(ltime >= Lambda_z_lower),1)
lupper <- tail(which(ltime <= Lambda_z_upper),1)
lconc <- lconc[llower:lupper]
ltime <- ltime[llower:lupper]
}else{
lconc <- 0
print("Note: LambdaTimeRange does not include 3 or more elimination phase observations; hence, extrapolation for the elimination phase will not be performed.\n")
}
}
# exclude time points as defined by LambdaExclude
if(!is.null(LambdaExclude) & length(lconc)>=3){
for(i in 1:length(LambdaExclude)){
if(grepl("^[-]?[0-9]*[.]?[0-9]*[eE]?[-]?[0-9]*[.]?[0-9]*$", LambdaExclude[i])){
# Check if the LambdaExclude is numeric
lconc <- lconc[!ltime%in%LambdaExclude[i]]
ltime <- ltime[!ltime%in%LambdaExclude[i]]
}else{
# Check if the LambdaExclude is logical
lconc <- eval(parse(text=paste0("lconc[!ltime",LambdaExclude[i],"]")))
ltime <- eval(parse(text=paste0("ltime[!ltime",LambdaExclude[i],"]")))
}
}
}
# Determine the log-linear regression coefficients to calculate Lambda
num_points_needed <- 3
if(force_extrapolate) num_points_needed <- 2
if(length(lconc) >= num_points_needed){
# make sure force_extrapolate does not change results for "good" profiles
if(force_extrapolate & adminType!="iv-bolus"){
if(length(lconc) > (num_points_needed + 2)){
if(lconc[1]==Cmax){
lconc <- lconc[-1]
ltime <- ltime[-1]
num_points_needed <- 3
}
}
}
lconc <- log(lconc)
lnPt <- length(lconc)
infd <- data.frame(np=numeric(0),rsq=numeric(0),arsq=numeric(0),m=numeric(0),inpt=numeric(0))
for (r in 1:(lnPt-(num_points_needed -1))){
mlr <- lm(lconc[r:lnPt]~ltime[r:lnPt])
if(is.na(coef(mlr)[2])) next
if(coef(mlr)[2] >= 0) next
n <- (lnPt-r)+1
rsq <- summary(mlr)$r.squared
trsq <- 1-((1-rsq)*(n-1)/(n-2)) # adjusted r^2
tmp <- cbind(np=n,rsq=rsq,arsq=trsq,m=(coef(mlr)[2]),inpt=(coef(mlr)[1]))
infd <- rbind(infd,tmp)
}
if(nrow(infd) != 0){
infd <- dplyr::arrange(infd, -arsq)
for (r in 1:nrow(infd)){
if(r == 1){
Rsq <- infd$rsq[r]
Rsq_adjusted <- infd$arsq[r]
No_points_Lambda_z <- infd$np[r]
slope <- infd$m[r]
intercept <- infd$inpt[r]
}else{
tarsq <- infd$arsq[r]
tDPt <- infd$np[r]
if(is.nan(tarsq)) break
if(Rsq_adjusted - tarsq > 0.0001) break
if(No_points_Lambda_z > tDPt) next
Rsq <- infd$rsq[r]
Rsq_adjusted <- tarsq
No_points_Lambda_z <- tDPt
slope <- infd$m[r]
intercept <- infd$inpt[r]
}
}
Corr_XY <- -1*sqrt(Rsq)
Lambda_z <- (-1*slope)
HL_Lambda_z <- log(2)/Lambda_z
lastPt <- exp((slope*ltime[lnPt])+intercept)
}
} else {
message(num_points_needed," or more elimination phase observations were not identified\n",
"extrapolation for the elimination phase will not be performed.\n")
}
} # end if(extrapolate)
# Interpolation for AUClower_upper, if needed
# Intpol11 = TRUE if start time within observed data but does not coincide
# Intpol12 = TRUE if start time outside observed data
# Intpol21 = TRUE if end time within observed data but does not coincide
# Intpol22 = TRUE if end time outside observed data
Intpol11 <- Intpol12 <- Intpol21 <- Intpol22 <- FALSE
if(!is.null(AUCTimeRange)){
TR1 <- sort(AUCTimeRange)[1] # lower bound of the time range
TR2 <- sort(AUCTimeRange)[2] # upper bound of the time range
if(!TR1%in%ntime){
if(TR1<=Tlast & length(ntime[ntime<TR1])!=0 & length(ntime[ntime>TR1])!=0){
# If the start time falls within the range of the data but does not coincide with an observed data point
Intpol11 <- TRUE
ind1 <- tail(which(ntime==max(ntime[ntime<TR1])),1) # index just below time interpolation point
ind2 <- head(which(ntime==min(ntime[ntime>TR1])),1) # index just above time interpolation point
c1 <- nconc[ind1]; t1 <- ntime[ind1] # conc and time just below time interpolation point
c2 <- nconc[ind2]; t2 <- ntime[ind2] # conc and time just above time interpolation point
if(method=="linear" | (method=="linearup-logdown" & c2>=c1)) CR1 <- c1 + abs((TR1-t1)/(t2-t1)) * (c2-c1)
if(method=="log" | (method=="linearup-logdown" & c2<c1)) CR1 <- exp(log(c1) + abs((TR1-t1)/(t2-t1)) * log(c2/c1))
nconc <- c(nconc, CR1)
ntime <- c(ntime, TR1)
}else if(TR1>Tlast & exists("lastPt") & exists("Lambda_z")){
Intpol12 <- TRUE
CR1 <- lastPt * exp(-Lambda_z * (TR1-Tlast))
CR1 <- ifelse(CR1==0, 0.0001, CR1)
nconc <- c(nconc, CR1)
ntime <- c(ntime, TR1)
}
}
if(!TR2%in%ntime){
if(TR2<=Tlast & length(ntime[ntime<TR2])!=0 & length(ntime[ntime>TR2])!=0){
# If the end time falls within the range of the data but does not coincide with an observed data point
Intpol21 <- TRUE
ind1 <- tail(which(ntime==max(ntime[ntime<TR2])),1) # index just below time interpolation point
ind2 <- head(which(ntime==min(ntime[ntime>TR2])),1) # index just above time interpolation point
c1 <- nconc[ind1]; t1 <- ntime[ind1] # conc and time just below time interpolation point
c2 <- nconc[ind2]; t2 <- ntime[ind2] # conc and time just above time interpolation point
if(method=="linear" | (method=="linearup-logdown" & c2>=c1)) CR2 <- c1 + abs((TR2-t1)/(t2-t1)) * (c2-c1)
if(method=="log" | (method=="linearup-logdown" & c2<c1)) CR2 <- exp(log(c1) + abs((TR2-t1)/(t2-t1)) * log(c2/c1))
nconc <- c(nconc, CR2)
ntime <- c(ntime, TR2)
}else if(TR2>Tlast & exists("lastPt") & exists("Lambda_z")){
Intpol22 <- TRUE
CR2 <- lastPt * exp(-Lambda_z * (TR2-Tlast))
CR2 <- ifelse(CR2==0, 0.0001, CR2)
nconc <- c(nconc, CR1)
ntime <- c(ntime, TR1)
}
}
}
# If interpolation is executed, redefine time and conc data
if(Intpol11 | Intpol21){
pkData <- data.frame(time=ntime, conc=nconc)
pkData <- dplyr::arrange(pkData, time)
nconc <- pkData$conc
ntime <- pkData$time
otime <- ntime - min(ntime) # Time is off-set to zero for estimation of the 1st moment
nPt <- length(nconc) # No. of data points
Cmax <- max(nconc)
mxId <- which(nconc == Cmax)[length(which(nconc == Cmax))]
Tmax <- ntime[mxId]
Cmax_D <- ifelse(!is.null(doseAmt), nconc[mxId]/doseAmt, NA)
lIdx <- max(which(nconc == tail(nconc[nconc>0],1))) # Index for last positive concentration
Tlast <- ntime[lIdx]
Clast <- nconc[lIdx]
}
##### Estimate NCA metrics from T0 to Tlast #####
# Initiate vectors for AUClast & AUMClast
AUClast <- AUMClast <- AUCnoC0 <- AUClower_upper <- 0.
for(r in 1:(nPt-1)){
if(method == "linear"){
delauc <- 0.5*(nconc[r:(r+1)])*(ntime[r+1]-ntime[r])
delaumc <- 0.5*((nconc[r+1]*otime[r+1])+(nconc[r]*otime[r]))*(otime[r+1]-otime[r])
AUClast <- sum(AUClast, delauc)
AUMClast <- sum(AUMClast, delaumc)
if(backExtrp){
if(noBackTime){
AUCnoC0 <- sum(AUCnoC0, delauc)
}else if(!noBackTime & r>1){
AUCnoC0 <- sum(AUCnoC0, delauc)
}
}
if(!is.null(AUCTimeRange)){if(ntime[r] >= TR1 & ntime[r+1] <=TR2){AUClower_upper <- sum(AUClower_upper, delauc)}}
if(doseType == "ss" && ssnPt != 0){
AUCtau <- sum(AUCtau, delauc); AUMCtau <- sum(AUMCtau, delaumc)
}
}else if(method == "log"){
delauc <- (nconc[r+1]-nconc[r])*(ntime[r+1]-ntime[r])/log(nconc[r+1]/nconc[r])
delaumc <- ((((nconc[r+1]*otime[r+1])-(nconc[r]*otime[r]))*(otime[r+1]-otime[r]))/(log(nconc[r+1]/nconc[r]))) -
((nconc[r+1]-nconc[r])*((otime[r+1]-otime[r])**2)/((log(nconc[r+1]/nconc[r]))**2))
AUClast <- sum(AUClast, delauc)
AUMClast <- sum(AUMClast, delaumc)
if(backExtrp){
if(noBackTime){
AUCnoC0 <- sum(AUCnoC0, delauc)
}else if(!noBackTime & r>1){
AUCnoC0 <- sum(AUCnoC0, delauc)
}
}
if(!is.null(AUCTimeRange)){if(ntime[r] >= TR1 & ntime[r+1] <= TR2){AUClower_upper <- sum(AUClower_upper, delauc)}}
if(doseType == "ss" && ssnPt != 0){
AUCtau <- sum(AUCtau, delauc); AUMCtau <- sum(AUMCtau, delaumc)
}
}else if(method == "linearup-logdown"){
if(nconc[r+1]>=nconc[r] | nconc[r+1]<=0 | nconc[r]<=0){
delauc <- mean(nconc[r:(r+1)])*(ntime[r+1]-ntime[r])
delaumc <- 0.5*((nconc[r+1]*otime[r+1])+(nconc[r]*otime[r]))*(otime[r+1]-otime[r])
AUClast <- sum(AUClast, delauc)
AUMClast <- sum(AUMClast, delaumc)
if(backExtrp){
if(noBackTime){
AUCnoC0 <- sum(AUCnoC0, delauc)
}else if(!noBackTime & r>1){
AUCnoC0 <- sum(AUCnoC0, delauc)
}
}
if(!is.null(AUCTimeRange)){if(ntime[r] >= TR1 & ntime[r+1] <= TR2){AUClower_upper <- sum(AUClower_upper, delauc)}}
if(doseType == "ss" && ssnPt != 0){
AUCtau <- sum(AUCtau, delauc); AUMCtau <- sum(AUMCtau, delaumc)
}
}else{
delauc <- (nconc[r+1]-nconc[r])*(ntime[r+1]-ntime[r])/log(nconc[r+1]/nconc[r])
delaumc <- ((((nconc[r+1]*otime[r+1])-(nconc[r]*otime[r]))*(otime[r+1]-otime[r]))/(log(nconc[r+1]/nconc[r]))) -
((nconc[r+1]-nconc[r])*((otime[r+1]-otime[r])**2)/((log(nconc[r+1]/nconc[r]))**2))
AUClast <- sum(AUClast, delauc)
AUMClast <- sum(AUMClast, delaumc)
if(backExtrp){
if(noBackTime){
AUCnoC0 <- sum(AUCnoC0, delauc)
}else if(!noBackTime & r>1){
AUCnoC0 <- sum(AUCnoC0, delauc)
}
}
if(!is.null(AUCTimeRange)){if(ntime[r] >= min(AUCTimeRange) & ntime[r+1] <= max(AUCTimeRange)){AUClower_upper <- sum(AUClower_upper, delauc)}}
if(doseType == "ss" && ssnPt != 0){
AUCtau <- sum(AUCtau, delauc); AUMCtau <- sum(AUMCtau, delaumc)
}
}
}
}
if(is.null(AUCTimeRange)){AUClower_upper <- AUClast}
if(backExtrp){
AUC_pBack_Ext_obs <- 100*(AUClast-AUCnoC0)/AUClast
AUC_pBack_Ext_pred <- AUC_pBack_Ext_obs
}
# MRTlast
if(adminType != "iv-infusion"){
MRTlast <- ifelse(doseType != "ss", AUMClast/AUClast, AUMCtau/AUCtau)
}else{
MRTlast <- (AUMClast/AUClast)-(TI/2)
}
########################################
# Extrapolation for the elimination phase
if(exists("lastPt") & exists("Lambda_z")){
AUCINF_obs <- exp(lconc[lnPt])/Lambda_z
AUMCINF_obs <- (exp(lconc[lnPt])/(Lambda_z**2))+(ltime[lnPt]*exp(lconc[lnPt])/Lambda_z)
AUCINF_pred <- lastPt/Lambda_z
AUMCINF_pred <- (lastPt/(Lambda_z**2))+(ltime[lnPt]*lastPt/Lambda_z)
# Calculate AUClower_upper if time range is outside last observation
if(Intpol12 | Intpol22){
uptime <- TR2
lowtime <- ifelse(Intpol12, TR1, max(ltime))
upconc <- CR2
lowconc <- ifelse(Intpol12, CR1, exp(lconc[lnPt]))
delauc <- (upconc-lowconc)*(uptime-lowtime)/log(upconc/lowconc)
AUClower_upper <- AUClower_upper + delauc
}
# Calculate AUCtau and AUMCtau if Tau is greater than Tlast
if(doseType == "ss" && (Tau > max(ltime) & AUCtau != 0)){
uptime <- Tau
lowtime <- max(ltime)
upconc <- ifelse(exp(slope*uptime+intercept)==0, 0.0001, exp(slope*uptime+intercept))
lowconc <- exp(lconc[lnPt])
delauc <- (upconc-lowconc)*(uptime-lowtime)/log(upconc/lowconc)
delaumc <- ((((upconc*uptime)-(lowconc*lowtime))*(uptime-lowtime))/(log(upconc/lowconc))) - ((upconc-lowconc)*((uptime-lowtime)**2)/((log(upconc/lowconc))**2))
AUCtau <- sum(AUCtau, delauc)
AUMCtau <- sum(AUMCtau, delaumc)
}
if(AUClast != 0 & AUCINF_obs != 0){
AUCINF_obs <- AUClast+AUCINF_obs; AUCINF_D_obs <- ifelse(!is.null(doseAmt),AUCINF_obs/doseAmt,NA); AUC_pExtrap_obs <- 100*(AUCINF_obs-AUClast)/AUCINF_obs
AUCINF_pred <- AUClast+AUCINF_pred; AUCINF_D_pred <- ifelse(!is.null(doseAmt),AUCINF_pred/doseAmt,NA); AUC_pExtrap_pred <- 100*(AUCINF_pred-AUClast)/AUCINF_pred
if(doseType=="ns"){
Vz_obs <- ifelse(!is.null(doseAmt),doseAmt/(Lambda_z*AUCINF_obs),NA); Cl_obs <- ifelse(!is.null(doseAmt),doseAmt/AUCINF_obs,NA)
Vz_pred <- ifelse(!is.null(doseAmt),doseAmt/(Lambda_z*AUCINF_pred),NA); Cl_pred <- ifelse(!is.null(doseAmt),doseAmt/AUCINF_pred,NA)
}else{
Vz_obs <- ifelse(!is.null(doseAmt),doseAmt/(Lambda_z*AUCtau),NA); Cl_obs <- ifelse(!is.null(doseAmt),doseAmt/AUCtau,NA)
Vz_pred <- NA; Cl_pred <- NA
}
}
# Calculate AUMC parameters
if(AUMClast != 0 & AUMCINF_obs != 0){
AUMCINF_obs <- AUMClast+AUMCINF_obs
AUMCINF_D_obs <- ifelse(!is.null(doseAmt), AUMCINF_obs/doseAmt, NA)
AUMC_pExtrap_obs <- 100*(AUMCINF_obs-AUMClast)/AUMCINF_obs
AUMCINF_pred <- AUMClast+AUMCINF_pred
AUMCINF_D_pred <- ifelse(!is.null(doseAmt), AUMCINF_pred/doseAmt, NA)
AUMC_pExtrap_pred <- 100*(AUMCINF_pred-AUMClast)/AUMCINF_pred
}
if(AUCINF_obs != 0 & AUMCINF_obs != 0){
MRTINF_obs <- ifelse(adminType!="iv-infusion", AUMCINF_obs/AUCINF_obs, ((AUMCINF_obs/AUCINF_obs)-(TI/2)))
}
if(AUCINF_pred != 0 & AUMCINF_pred != 0){
MRTINF_pred <- ifelse(adminType!="iv-infusion", AUMCINF_pred/AUCINF_pred, ((AUMCINF_pred/AUCINF_pred)-(TI/2)))
}
if(doseType == "ns" && (adminType=="iv-bolus" | adminType=="iv-infusion")){
if(exists("MRTINF_obs")) Vss_obs <- MRTINF_obs*Cl_obs
if(exists("MRTINF_pred")) Vss_pred <- MRTINF_pred*Cl_pred
}
if(doseType == "ss" && ssnPt != 0){
Cavg <- AUCtau/Tau
Clss <- ifelse(!is.null(doseAmt),doseAmt/AUCtau,NA)
Cmax <- max(nconc)
p_Fluctuation <- 100*(Cmax-Cmin)/Cavg
if(complete.cases(Lambda_z)) Accumulation_Index <- 1/(1-exp(-Lambda_z*Tau))
# re-define MRTINF for steady state and calculate Vss
if(complete.cases(AUCINF_obs) & complete.cases(AUMCINF_obs) & AUCtau != 0){
if(adminType == "iv-infusion"){
MRTINF_obs <- ((AUMCtau+Tau*(AUCINF_obs-AUCtau))/(AUCtau)-(TI/2))
MRTINF_pred <- ((AUMCtau+Tau*(AUCINF_pred-AUCtau))/(AUCtau)-(TI/2))
}else{
MRTINF_obs <- (AUMCtau+Tau*(AUCINF_obs-AUCtau))/AUCtau
MRTINF_pred <- (AUMCtau+Tau*(AUCINF_pred-AUCtau))/AUCtau
}
if(exists("MRTINF_obs")) Vss_obs <- MRTINF_obs*Clss
if(exists("MRTINF_pred")) Vss_pred <- MRTINF_pred*Clss
}
}
}
}
if(!exists("C0")) C0 <- NA
if(!exists("Tmax")) Tmax <- NA
if(!exists("Cmax")) Cmax <- NA
if(!exists("Cmax_D")) Cmax_D <- NA
if(!exists("Tlast")) Tlast <- NA
if(!exists("Clast")) Clast <- NA
if(!exists("AUClast")) AUClast <- NA
if(!exists("AUMClast")) AUMClast <- NA
if(!exists("MRTlast")) MRTlast <- NA
if(!exists("No_points_Lambda_z")) No_points_Lambda_z <- NA
if(!exists("AUC_pBack_Ext_obs")) AUC_pBack_Ext_obs <- NA
if(!exists("AUC_pBack_Ext_pred")) AUC_pBack_Ext_pred <- NA
if(!exists("AUClower_upper")) AUClower_upper <- NA
if(!exists("Rsq")) Rsq <- NA
if(!exists("Rsq_adjusted")) Rsq_adjusted <- NA
if(!exists("Corr_XY")) Corr_XY <- NA
if(!exists("Lambda_z")) Lambda_z <- NA
if(!exists("Lambda_z_lower")) Lambda_z_lower <- NA
if(!exists("Lambda_z_upper")) Lambda_z_upper <- NA
if(!exists("HL_Lambda_z")) HL_Lambda_z <- NA
if(!exists("AUCINF_obs")) AUCINF_obs <- NA
if(!exists("AUCINF_D_obs")) AUCINF_D_obs <- NA
if(!exists("AUC_pExtrap_obs")) AUC_pExtrap_obs <- NA
if(!exists("Vz_obs")) Vz_obs <- NA
if(!exists("Cl_obs")) Cl_obs <- NA
if(!exists("AUCINF_pred")) AUCINF_pred <- NA
if(!exists("AUCINF_D_pred")) AUCINF_D_pred <- NA
if(!exists("AUC_pExtrap_pred")) AUC_pExtrap_pred <- NA
if(!exists("Vz_pred")) Vz_pred <- NA
if(!exists("Cl_pred")) Cl_pred <- NA
if(!exists("AUMCINF_obs")) AUMCINF_obs <- NA
if(!exists("AUMC_pExtrap_obs")) AUMC_pExtrap_obs <- NA
if(!exists("AUMCINF_pred")) AUMCINF_pred <- NA
if(!exists("AUMC_pExtrap_pred")) AUMC_pExtrap_pred <- NA
if(!exists("MRTINF_obs")) MRTINF_obs <- NA
if(!exists("MRTINF_pred")) MRTINF_pred <- NA
if(!exists("Tmin")) Tmin <- NA
if(!exists("Cmin")) Cmin <- NA
if(!exists("Cavg")) Cavg <- NA
if(!exists("AUCtau")) AUCtau <- NA
if(!exists("AUMCtau")) AUMCtau <- NA
if(!exists("Clss")) Clss <- NA
if(!exists("Vss_obs")) Vss_obs <- NA
if(!exists("Vss_pred")) Vss_pred <- NA
if(!exists("p_Fluctuation")) p_Fluctuation <- NA
if(!exists("Accumulation_Index")) Accumulation_Index <- NA
if(!is.null(simFile) & dset == "obs"){
if(!onlyNCA){
NCAprm <- as.numeric(c(C0,Tmax,0,0,0,Cmax,0,0,0,Cmax_D,Tlast,Clast,AUClast,0,0,0,AUMClast,0,0,0,MRTlast,No_points_Lambda_z,AUClower_upper,0,0,0,Rsq,Rsq_adjusted,Corr_XY,Lambda_z,Lambda_z_lower,Lambda_z_upper,HL_Lambda_z,0,0,0,AUCINF_obs,0,0,0,AUCINF_D_obs,AUC_pExtrap_obs,AUC_pBack_Ext_obs,Vz_obs,Cl_obs,AUCINF_pred,0,0,0,AUCINF_D_pred,AUC_pExtrap_pred,AUC_pBack_Ext_pred,Vz_pred,Cl_pred,AUMCINF_obs,AUMC_pExtrap_obs,AUMCINF_pred,AUMC_pExtrap_pred,MRTINF_obs,MRTINF_pred,Tau,Tmin,Cmin,Cavg,AUCtau,AUMCtau,Clss,Vss_obs,Vss_pred,p_Fluctuation,Accumulation_Index))
names(NCAprm) <- c("C0","Tmax","simTmax","dTmax","npdeTmax","Cmax","simCmax","dCmax","npdeCmax","Cmax_D","Tlast","Clast","AUClast","simAUClast","dAUClast","npdeAUClast","AUMClast","simAUMClast","dAUMClast","npdeAUMClast","MRTlast","No_points_Lambda_z","AUClower_upper","simAUClower_upper","dAUClower_upper","npdeAUClower_upper","Rsq","Rsq_adjusted","Corr_XY","Lambda_z","Lambda_z_lower","Lambda_z_upper","HL_Lambda_z","simHL_Lambda_z","dHL_Lambda_z","npdeHL_Lambda_z","AUCINF_obs","simAUCINF_obs","dAUCINF_obs","npdeAUCINF_obs","AUCINF_D_obs","AUC_pExtrap_obs","AUC_pBack_Ext_obs","Vz_obs","Cl_obs","AUCINF_pred","simAUCINF_pred","dAUCINF_pred","npdeAUCINF_pred","AUCINF_D_pred","AUC_pExtrap_pred","AUC_pBack_Ext_pred","Vz_pred","Cl_pred","AUMCINF_obs","AUMC_pExtrap_obs","AUMCINF_pred","AUMC_pExtrap_pred","MRTINF_obs","MRTINF_pred","Tau","Tmin","Cmin","Cavg","AUCtau","AUMCtau","Clss","Vss_obs","Vss_pred","p_Fluctuation","Accumulation_Index")
}else{
NCAprm <- as.numeric(c(C0,Tmax,0,Cmax,0,Cmax_D,Tlast,Clast,AUClast,0,AUMClast,0,MRTlast,No_points_Lambda_z,AUClower_upper,0,Rsq,Rsq_adjusted,Corr_XY,Lambda_z,Lambda_z_lower,Lambda_z_upper,HL_Lambda_z,0,AUCINF_obs,0,AUCINF_D_obs,AUC_pExtrap_obs,AUC_pBack_Ext_obs,Vz_obs,Cl_obs,AUCINF_pred,0,AUCINF_D_pred,AUC_pExtrap_pred,AUC_pBack_Ext_pred,Vz_pred,Cl_pred,AUMCINF_obs,AUMC_pExtrap_obs,AUMCINF_pred,AUMC_pExtrap_pred,MRTINF_obs,MRTINF_pred,Tau,Tmin,Cmin,Cavg,AUCtau,AUMCtau,Clss,Vss_obs,Vss_pred,p_Fluctuation,Accumulation_Index))
names(NCAprm) <- c("C0","Tmax","simTmax","Cmax","simCmax","Cmax_D","Tlast","Clast","AUClast","simAUClast","AUMClast","simAUMClast","MRTlast","No_points_Lambda_z","AUClower_upper","simAUClower_upper","Rsq","Rsq_adjusted","Corr_XY","Lambda_z","Lambda_z_lower","Lambda_z_upper","HL_Lambda_z","simHL_Lambda_z","AUCINF_obs","simAUCINF_obs","AUCINF_D_obs","AUC_pExtrap_obs","AUC_pBack_Ext_obs","Vz_obs","Cl_obs","AUCINF_pred","simAUCINF_pred","AUCINF_D_pred","AUC_pExtrap_pred","AUC_pBack_Ext_pred","Vz_pred","Cl_pred","AUMCINF_obs","AUMC_pExtrap_obs","AUMCINF_pred","AUMC_pExtrap_pred","MRTINF_obs","MRTINF_pred","Tau","Tmin","Cmin","Cavg","AUCtau","AUMCtau","Clss","Vss_obs","Vss_pred","p_Fluctuation","Accumulation_Index")
}
}else{
NCAprm <- as.numeric(c(C0,Tmax,Cmax,Cmax_D,Tlast,Clast,AUClast,AUMClast,MRTlast,No_points_Lambda_z,AUClower_upper,Rsq,Rsq_adjusted,Corr_XY,Lambda_z,Lambda_z_lower,Lambda_z_upper,HL_Lambda_z,AUCINF_obs,AUCINF_D_obs,AUC_pExtrap_obs,AUC_pBack_Ext_obs,Vz_obs,Cl_obs,AUCINF_pred,AUCINF_D_pred,AUC_pExtrap_pred,AUC_pBack_Ext_pred,Vz_pred,Cl_pred,AUMCINF_obs,AUMC_pExtrap_obs,AUMCINF_pred,AUMC_pExtrap_pred,MRTINF_obs,MRTINF_pred,Tau,Tmin,Cmin,Cavg,AUCtau,AUMCtau,Clss,Vss_obs,Vss_pred,p_Fluctuation,Accumulation_Index))
names(NCAprm) <- c("C0","Tmax","Cmax","Cmax_D","Tlast","Clast","AUClast","AUMClast","MRTlast","No_points_Lambda_z","AUClower_upper","Rsq","Rsq_adjusted","Corr_XY","Lambda_z","Lambda_z_lower","Lambda_z_upper","HL_Lambda_z","AUCINF_obs","AUCINF_D_obs","AUC_pExtrap_obs","AUC_pBack_Ext_obs","Vz_obs","Cl_obs","AUCINF_pred","AUCINF_D_pred","AUC_pExtrap_pred","AUC_pBack_Ext_pred","Vz_pred","Cl_pred","AUMCINF_obs","AUMC_pExtrap_obs","AUMCINF_pred","AUMC_pExtrap_pred","MRTINF_obs","MRTINF_pred","Tau","Tmin","Cmin","Cavg","AUCtau","AUMCtau","Clss","Vss_obs","Vss_pred","p_Fluctuation","Accumulation_Index")
}
return(NCAprm)
}
Try the ncappc package in your browser
Any scripts or data that you put into this service are public.
ncappc documentation built on May 1, 2019, 7:31 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions est.nca
Documented in est.nca
# R function to estimate NCA metrics
# Chayan, 12/2014
# roxygen comments
#' Estimate individual NCA metrics.
#'
#' Estimates a comprehensive set of NCA metrics
#' for a given individual using concentration vs. time data.
#'
#' \pkg{est.nca} estimates a comprehensive set of NCA metrics using the
#' concentration-time profile of an individual. NCA metrics are estimated
#' according to traditional PK calculations. The names of the various NCA
#' metrics estimated in this package are assigned mainly following the names
#' used in WinNonlin. This package accepts any of the three different types of
#' drug administration, (i) iv-bolus, (ii) iv-infusion and (iii) extravascular;
#' \pkg{ncappc} also can accept both non-steady state and steady-state data.
#' The NCA metrics that are estimated and reported by \pkg{ncappc} are listed
#' below.
#' \itemize{
#' \item \strong{C0} is the initial concentration at the dosing time. It is
#' the observed concentration at the dosing time, if available. Otherwise
#' it is approximated using the following rules.
#' \item \strong{Cmax, Tmax and Cmax_D} are the value and the time of maximum
#' observed concentration, respectively. If the maximum concentration is not
#' unique, the first maximum is used. For steady state data, The maximum value
#' between the dosing intervals is considered. Cmax_D is the dose normalized
#' maximum observed concentration.
#' \item \strong{Clast and Tlast} are the last measurable positive
#' concentration and the corresponding time, respectively.
#' \item \strong{AUClast} is the area under the concentration vs. time curve
#' from the first observed to last measurable concentration.
#' \item \strong{AUMClast} is the first moment of the concentration vs. time
#' curve from the first observed to last measurable concentration.
#' \item \strong{MRTlast} is the mean residence time from the first observed
#' to last measurable concentration.
#' \item \strong{No_points_Lambda_z} is the number of observed data points
#' used to determine the best fitting regression line in the elimination
#' phase.
#' \item \strong{AUC_pBack_Ext_obs} is the percentage of AUCINF_obs that is
#' contributed by the back extrapolation to estimate C0.
#' \item \strong{AUC_pBack_Ext_pred} is the percentage of AUCINF_pred that is
#' contributed by the back extrapolation to estimate C0.
#' \item \strong{AUClower_upper} is the AUC under the concentration-time
#' profile within the user-specified window of time provided as the
#' "AUCTimeRange" argument. In case of empty "AUCTimeRange" argument,
#' AUClower_upper is the same as AUClast.
#' \item \strong{Rsq, Rsq_adjusted and Corr_XY} are regression coefficient
#' of the regression line used to estimate the elimination rate constant, the
#' adjusted value of Rsq and the square root of Rsq, respectively.
#' \item \strong{Lambda_z} is the elimination rate constant estimated from the
#' regression line representing the terminal phase of the concentration-time
#' data.
#' \item \strong{Lambda_lower and Lambda_upper} are the lower and upper limit
#' of the time values from the concentration-time profile used to estimate
#' Lambda_z, respectively, in case the "LambdaTimeRange" is used to specify
#' the time range.
#' \item \strong{HL_Lambda_z} is terminal half-life of the drug.
#' \item \strong{AUCINF_obs and AUCINF_obs_D} are AUC estimated from the first
#' sampled data extrapolated to infinity and its dose normalized version,
#' respectively. The extrapolation in the terminal phase is based on the last
#' observed concentration Clast_obs.
#' \item \strong{AUC_pExtrap_obs} is the percentage of the AUCINF_obs that is
#' contributed by the extrapolation from the last sampling time to infinity.
#' \item \strong{AUMCINF_obs} is AUMC estimated from the first sampled data
#' extrapolated to infinity. The extrapolation in the terminal phase is based
#' on the last observed concentration.
#' \item \strong{AUMC_pExtrap_obs} is the percentage of the AUMCINF_obs that
#' is contributed by the extrapolation from the last sampling time to infinity.
#' \item \strong{Vz_obs} is the volume of distribution estimated based on
#' total AUC
#' \item \strong{Cl_obs} is total body clearance.
#' \item \strong{AUCINF_pred and AUCINF_pred_D} are AUC from the first sampled
#' data extrapolated to infinity and its dose normalized version, respectively.
#' The extrapolation in the terminal phase is based on the last predicted
#' concentration obtained from the regression line used to estimate Lambda_z
#' (Clast_pred).
#' \item \strong{AUC_pExtrap_pred} is the percentage of the AUCINF_pred that
#' is contributed by the extrapolation from the last sampling time to infinity.
#' \item \strong{AUMCINF_pred} is AUMC estimated from the first sampled data
#' extrapolated to infinity. The extrapolation in the terminal phase is based
#' on the last predicted concentration obtained from the regression line used
#' to estimate Lambda_z (Clast_pred).
#' \item \strong{AUMC_pExtrap_pred} is the percentage of the AUMCINF_pred that
#' is contributed by the extrapolation from the last sampling time to infinity.
#' \item \strong{Vz_pred} is the volume of distribution estimated based on
#' AUCINF_pred.
#' \item \strong{Cl_pred} is the total body clearance estimated based on
#' AUCINF_pred.
#' \item \strong{MRTINF_obs} is the mean residence time from the first
#' sampled time extrapolated to infinity based on the last observed
#' concentration (Clast_obs).
#' \item \strong{MRTINF_pred} is the mean residence time from the first
#' sampled time extrapolated to infinity based on the last predicted
#' concentration obtained from the regression line used to estimate Lambda_z
#' (Clast_pred).
#' \item \strong{Tau} is the dosing interval for steady-state data.
#' \item \strong{Cmin and Tmin} are the minimum concentration between 0 and
#' Tau and the corresponding time, respectively.
#' \item \strong{Cavg} is the average concentration between 0 and Tau for
#' steady-state data.
#' \item \strong{AUCtau and AUMCtau} are AUC and AUMC between 0 and Tau for
#' steady-state data.
#' \item \strong{Clss} is an estimate of the total body clearance for
#' steady-state data.
#' \item \strong{Vss_obs and Vss_pred} are estimated volume of distribution at
#' steady-state based on Clast_obs and Clast_pred, respectively.
#' \item \strong{p_Fluctuation} is the percentage of the fluctuation of the
#' concentration between 0 and Tau for steady-state data.
#' \item \strong{Accumulation_Index} is \eqn{1/(1-e^(-\lambda_z*\tau))}
#' }
#' @param time Numeric array for time
#' @param conc Numeric array for concentration
#' @param backExtrp If back-extrapolation is needed for AUC (TRUE or FALSE)
#' (\strong{FALSE})
#' @param negConcExcl Exclude -ve conc (\strong{FALSE})
#' @param doseType Steady-state (ss) or non-steady-state (ns) dose
#' (\strong{"ns"})
#' @param adminType Route of administration
#' (iv-bolus,iv-infusion,extravascular) (\strong{"extravascular"})
#' @param doseAmt Dose amounts (\strong{"NULL"})
#' @param method Method to estimate AUC. The \code{"linear"} method applies the linear
#' trapezoidal rule to estimate the area under the curve. The \code{"log"} method
#' applies the logarithmic trapezoidal rule to estimate the area under the
#' curve. The \code{"linearup-logdown"} method applies the linear trapezoidal rule to
#' estimate the area under the curve for the ascending part of the curve and
#' the logarithmic trapezoidal rule to estimate the area under the curve for
#' the descending part of the curve.
#' @param AUCTimeRange User-defined window of time used to estimate AUC
#' (\strong{"NULL"})
#' @param LambdaTimeRange User-defined window of time to estimate elimination
#' rate-constant (\strong{"NULL"})
#' @param LambdaExclude User-defined excluded observation time points for
#' estimation of elimination rate-constant (\strong{"NULL"})
#' @param Tau Dosing interval for steady-state data (\strong{"NULL"})
#' @param doseTime Dose time prior to the first observation for steady-state
#' data (\strong{\code{NULL}})
#' @param TI Infusion duration (\strong{"NULL"})
#' @param simFile Name of the simulated concentration-time data if present
#' (\strong{"NULL"})
#' @param dset Type, i.e., observed or simulated concentration-time data set
#' ("obs" or "sim") (\strong{"obs"})
#' @param onlyNCA If \code{TRUE} only NCA is performed and ppc part is ignored
#' although simFile is not \code{NULL}. Default is \strong{\code{FALSE}}
#' @param extrapolate Should the function extrapolate from the last observation to infinity?
#' @param sparse_compute Should NCA metrics be computed even with only one sample?
#' @param force_extrapolate Extrapolate AUC_inf with sparse data. Sparse data is defined as
#' fewer than 3 points after Cmax for non-bolus input. In that case, Cmax is included in the extrapolation
#' and only one extra point in required for extrapolation.
#' @param ... Arguments passed from other functions. Not used.
#'
#' @return An array of estimated NCA metrics
#' @export
#'
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
# function to estimate NCA parameters
est.nca <- function(time,
conc,
backExtrp=FALSE,
negConcExcl=FALSE,
doseType="ns",
adminType="extravascular",
doseAmt=NULL,
method="linearup-logdown",
AUCTimeRange=NULL,
LambdaTimeRange=NULL,
LambdaExclude=NULL,
doseTime=doseTime,
Tau=NULL,
TI=NULL,
simFile=NULL,
dset="obs",
onlyNCA=FALSE,
extrapolate=FALSE,
sparse_compute=FALSE,
force_extrapolate = FALSE,...){
"tail" <- "head" <- "lm" <- "coef" <- "arsq" <- NULL
rm(list=c("tail","head","lm","coef","arsq"))
# Set Tau to NA for non steady-state data
Tau <- ifelse(doseType=="ns", NA, Tau)
ntime <- as.numeric(time) # Save time to ntime
nconc <- as.numeric(conc) # Save conc to nconc
# check that method is defined as it should be
if(!(method %in% c("linear","log", "linearup-logdown")))
stop(paste(method,"is not one of 'linear','log', or 'linearup-logdown'"))
# Exclude zero or negative concentration from calculations
if(negConcExcl){
zid <- which (nconc < 0)
if(length(zid) > 0){ntime <- ntime[-zid]; nconc <- nconc[-zid]}
}
# Check if the number of observations is at least 3, otherwise skip the operation
if(length(nconc) >= 2 | sparse_compute){
# Steady state data
if(doseType == "ss"){
if(is.null(doseTime)){
ssIdx1 <- min(which(nconc>0)) # Index for the first observation with non-zero conc
}else{
ssIdx1 <- min(which(abs(round(ntime,0)-doseTime)==0)) # Index for the first observation within the steady state interval
}
sc1 <- nconc[ssIdx1] # Steady state 1st observed conc
st1 <- ntime[ssIdx1] # Steady state 1st observed time
ssIdx2 <- which.min(abs(ntime-(st1+Tau))) # Index for the last ss time (Tau added to first time and find the nearest time to that number)
sc2 <- nconc[ssIdx2] # Steady state last observed conc
st2 <- ntime[ssIdx2] # Steady state last observed time
# steady state observations
nconc <- nconc[which(ntime>=st1 & ntime<=st2)]
ntime <- ntime[which(ntime>=st1 & ntime<=st2)]
ssnPt <- length(ntime)
if(ssnPt != 0){
mIdx <- which(nconc == min(nconc[nconc!=0]))[1]
Tmin <- ntime[mIdx]
Cmin <- nconc[mIdx]
AUCtau <- 0.
AUMCtau <- 0.
}
}
# Calculation of C0
noBackTime <- TRUE
if(backExtrp){
if(ntime[1]==0){
C0 <- as.numeric(nconc[1])
}else{
if(adminType=="extravascular" | adminType=="iv-infusion"){
if(doseType=="ss" && ssnPt!=0){
C0 <- as.numeric(min(nconc)) # Min obsreved for SS data
}else{
C0 <- 0 # C0 is set to 0 for single dose data
nconc <- c(C0, nconc)
ntime <- c(0, ntime)
noBackTime <- FALSE
}
}else if(adminType == "iv-bolus"){
slope <- (nconc[2]-nconc[1])/(ntime[2]-ntime[1])
tconc <- nconc[1:2]
ttime <- ntime[1:2]
lmEx <- FALSE
if(!is.null(LambdaExclude)){
for(i in 1:length(LambdaExclude)){
if(grepl("^[-]?[0-9]*[.]?[0-9]*[eE]?[-]?[0-9]*[.]?[0-9]*$", LambdaExclude[i])){
# Check if the LambdaExclude is numeric
if(any(ttime%in%LambdaExclude[i])) lmEx <- TRUE
}else{
# Check if the LambdaExclude is logical
if(any(eval(parse(text=paste0("ttime",LambdaExclude[i]))))) lmEx <- TRUE
}
}
}
if(slope >= 0 | nconc[1]==0 | nconc[2]==0 | lmEx){
C0 <- as.numeric(nconc[nconc>0][1])
}else{
C0 <- as.numeric(exp((ntime[1]*log(nconc[2])-ntime[2]*log(nconc[1]))/(ntime[1]-ntime[2])))
nconc <- c(C0, nconc) # extrapolation via log-linear regression
ntime <- c(0, ntime)
noBackTime <- FALSE
}
}
}
}
# Some of the NCA metrics including Cmax and Tmax
otime <- ntime - min(ntime) # Time is off-set to zero for estimation of the 1st moment
nPt <- length(nconc) # No. of data points
Cmax <- max(nconc)
mxId <- tail(which(nconc == Cmax),1)
Tmax <- ntime[mxId]
Cmax_D <- ifelse(!is.null(doseAmt), nconc[mxId]/doseAmt, NA)
lIdx <- max(which(nconc == tail(nconc[nconc>0],1))) # Index for last positive concentration
Tlast <- ntime[lIdx]
Clast <- nconc[lIdx]
if(extrapolate){
###### Prepare for extrapolation for the terminal elimination ######
# Determine the lower and upper indeces of the time vector used for Lambda calculation
llower <- ifelse(adminType!="iv-bolus", mxId+1, mxId) # Exclude Cmax from elimination phase extrapolation for non-bolus dose
if(force_extrapolate) llower <- mxId
lupper <- nPt
# Conc and Time for the elimination phase
lconc <- nconc[llower:lupper]
ltime <- ntime[llower:lupper]
# exclude points with zero or negative concentration
zid <- which(lconc <= 0)
if(length(zid) > 0){
lconc <- lconc[-zid]
ltime <- ltime[-zid]
}
# Re-define llower and lupper if time range is specified for Lambda calculation
if(!is.null(LambdaTimeRange)){
Lambda_z_lower <- sort(LambdaTimeRange)[1]
Lambda_z_upper <- sort(LambdaTimeRange)[2]
if(length(ltime[ltime>=Lambda_z_lower & ltime<=Lambda_z_upper]) >= 3){
llower <- head(which(ltime >= Lambda_z_lower),1)
lupper <- tail(which(ltime <= Lambda_z_upper),1)
lconc <- lconc[llower:lupper]
ltime <- ltime[llower:lupper]
}else{
lconc <- 0
print("Note: LambdaTimeRange does not include 3 or more elimination phase observations; hence, extrapolation for the elimination phase will not be performed.\n")
}
}
# exclude time points as defined by LambdaExclude
if(!is.null(LambdaExclude) & length(lconc)>=3){
for(i in 1:length(LambdaExclude)){
if(grepl("^[-]?[0-9]*[.]?[0-9]*[eE]?[-]?[0-9]*[.]?[0-9]*$", LambdaExclude[i])){
# Check if the LambdaExclude is numeric
lconc <- lconc[!ltime%in%LambdaExclude[i]]
ltime <- ltime[!ltime%in%LambdaExclude[i]]
}else{
# Check if the LambdaExclude is logical
lconc <- eval(parse(text=paste0("lconc[!ltime",LambdaExclude[i],"]")))
ltime <- eval(parse(text=paste0("ltime[!ltime",LambdaExclude[i],"]")))
}
}
}
# Determine the log-linear regression coefficients to calculate Lambda
num_points_needed <- 3
if(force_extrapolate) num_points_needed <- 2
if(length(lconc) >= num_points_needed){
# make sure force_extrapolate does not change results for "good" profiles
if(force_extrapolate & adminType!="iv-bolus"){
if(length(lconc) > (num_points_needed + 2)){
if(lconc[1]==Cmax){
lconc <- lconc[-1]
ltime <- ltime[-1]
num_points_needed <- 3
}
}
}
lconc <- log(lconc)
lnPt <- length(lconc)
infd <- data.frame(np=numeric(0),rsq=numeric(0),arsq=numeric(0),m=numeric(0),inpt=numeric(0))
for (r in 1:(lnPt-(num_points_needed -1))){
mlr <- lm(lconc[r:lnPt]~ltime[r:lnPt])
if(is.na(coef(mlr)[2])) next
if(coef(mlr)[2] >= 0) next
n <- (lnPt-r)+1
rsq <- summary(mlr)$r.squared
trsq <- 1-((1-rsq)*(n-1)/(n-2)) # adjusted r^2
tmp <- cbind(np=n,rsq=rsq,arsq=trsq,m=(coef(mlr)[2]),inpt=(coef(mlr)[1]))
infd <- rbind(infd,tmp)
}
if(nrow(infd) != 0){
infd <- dplyr::arrange(infd, -arsq)
for (r in 1:nrow(infd)){
if(r == 1){
Rsq <- infd$rsq[r]
Rsq_adjusted <- infd$arsq[r]
No_points_Lambda_z <- infd$np[r]
slope <- infd$m[r]
intercept <- infd$inpt[r]
}else{
tarsq <- infd$arsq[r]
tDPt <- infd$np[r]
if(is.nan(tarsq)) break
if(Rsq_adjusted - tarsq > 0.0001) break
if(No_points_Lambda_z > tDPt) next
Rsq <- infd$rsq[r]
Rsq_adjusted <- tarsq
No_points_Lambda_z <- tDPt
slope <- infd$m[r]
intercept <- infd$inpt[r]
}
}
Corr_XY <- -1*sqrt(Rsq)
Lambda_z <- (-1*slope)
HL_Lambda_z <- log(2)/Lambda_z
lastPt <- exp((slope*ltime[lnPt])+intercept)
}
} else {
message(num_points_needed," or more elimination phase observations were not identified\n",
"extrapolation for the elimination phase will not be performed.\n")
}
} # end if(extrapolate)
# Interpolation for AUClower_upper, if needed
# Intpol11 = TRUE if start time within observed data but does not coincide
# Intpol12 = TRUE if start time outside observed data
# Intpol21 = TRUE if end time within observed data but does not coincide
# Intpol22 = TRUE if end time outside observed data
Intpol11 <- Intpol12 <- Intpol21 <- Intpol22 <- FALSE
if(!is.null(AUCTimeRange)){
TR1 <- sort(AUCTimeRange)[1] # lower bound of the time range
TR2 <- sort(AUCTimeRange)[2] # upper bound of the time range
if(!TR1%in%ntime){
if(TR1<=Tlast & length(ntime[ntime<TR1])!=0 & length(ntime[ntime>TR1])!=0){
# If the start time falls within the range of the data but does not coincide with an observed data point
Intpol11 <- TRUE
ind1 <- tail(which(ntime==max(ntime[ntime<TR1])),1) # index just below time interpolation point
ind2 <- head(which(ntime==min(ntime[ntime>TR1])),1) # index just above time interpolation point
c1 <- nconc[ind1]; t1 <- ntime[ind1] # conc and time just below time interpolation point
c2 <- nconc[ind2]; t2 <- ntime[ind2] # conc and time just above time interpolation point
if(method=="linear" | (method=="linearup-logdown" & c2>=c1)) CR1 <- c1 + abs((TR1-t1)/(t2-t1)) * (c2-c1)
if(method=="log" | (method=="linearup-logdown" & c2<c1)) CR1 <- exp(log(c1) + abs((TR1-t1)/(t2-t1)) * log(c2/c1))
nconc <- c(nconc, CR1)
ntime <- c(ntime, TR1)
}else if(TR1>Tlast & exists("lastPt") & exists("Lambda_z")){
Intpol12 <- TRUE
CR1 <- lastPt * exp(-Lambda_z * (TR1-Tlast))
CR1 <- ifelse(CR1==0, 0.0001, CR1)
nconc <- c(nconc, CR1)
ntime <- c(ntime, TR1)
}
}
if(!TR2%in%ntime){
if(TR2<=Tlast & length(ntime[ntime<TR2])!=0 & length(ntime[ntime>TR2])!=0){
# If the end time falls within the range of the data but does not coincide with an observed data point
Intpol21 <- TRUE
ind1 <- tail(which(ntime==max(ntime[ntime<TR2])),1) # index just below time interpolation point
ind2 <- head(which(ntime==min(ntime[ntime>TR2])),1) # index just above time interpolation point
c1 <- nconc[ind1]; t1 <- ntime[ind1] # conc and time just below time interpolation point
c2 <- nconc[ind2]; t2 <- ntime[ind2] # conc and time just above time interpolation point
if(method=="linear" | (method=="linearup-logdown" & c2>=c1)) CR2 <- c1 + abs((TR2-t1)/(t2-t1)) * (c2-c1)
if(method=="log" | (method=="linearup-logdown" & c2<c1)) CR2 <- exp(log(c1) + abs((TR2-t1)/(t2-t1)) * log(c2/c1))
nconc <- c(nconc, CR2)
ntime <- c(ntime, TR2)
}else if(TR2>Tlast & exists("lastPt") & exists("Lambda_z")){
Intpol22 <- TRUE
CR2 <- lastPt * exp(-Lambda_z * (TR2-Tlast))
CR2 <- ifelse(CR2==0, 0.0001, CR2)
nconc <- c(nconc, CR1)
ntime <- c(ntime, TR1)
}
}
}
# If interpolation is executed, redefine time and conc data
if(Intpol11 | Intpol21){
pkData <- data.frame(time=ntime, conc=nconc)
pkData <- dplyr::arrange(pkData, time)
nconc <- pkData$conc
ntime <- pkData$time
otime <- ntime - min(ntime) # Time is off-set to zero for estimation of the 1st moment
nPt <- length(nconc) # No. of data points
Cmax <- max(nconc)
mxId <- which(nconc == Cmax)[length(which(nconc == Cmax))]
Tmax <- ntime[mxId]
Cmax_D <- ifelse(!is.null(doseAmt), nconc[mxId]/doseAmt, NA)
lIdx <- max(which(nconc == tail(nconc[nconc>0],1))) # Index for last positive concentration
Tlast <- ntime[lIdx]
Clast <- nconc[lIdx]
}
##### Estimate NCA metrics from T0 to Tlast #####
# Initiate vectors for AUClast & AUMClast
AUClast <- AUMClast <- AUCnoC0 <- AUClower_upper <- 0.
for(r in 1:(nPt-1)){
if(method == "linear"){
delauc <- 0.5*(nconc[r:(r+1)])*(ntime[r+1]-ntime[r])
delaumc <- 0.5*((nconc[r+1]*otime[r+1])+(nconc[r]*otime[r]))*(otime[r+1]-otime[r])
AUClast <- sum(AUClast, delauc)
AUMClast <- sum(AUMClast, delaumc)
if(backExtrp){
if(noBackTime){
AUCnoC0 <- sum(AUCnoC0, delauc)
}else if(!noBackTime & r>1){
AUCnoC0 <- sum(AUCnoC0, delauc)
}
}
if(!is.null(AUCTimeRange)){if(ntime[r] >= TR1 & ntime[r+1] <=TR2){AUClower_upper <- sum(AUClower_upper, delauc)}}
if(doseType == "ss" && ssnPt != 0){
AUCtau <- sum(AUCtau, delauc); AUMCtau <- sum(AUMCtau, delaumc)
}
}else if(method == "log"){
delauc <- (nconc[r+1]-nconc[r])*(ntime[r+1]-ntime[r])/log(nconc[r+1]/nconc[r])
delaumc <- ((((nconc[r+1]*otime[r+1])-(nconc[r]*otime[r]))*(otime[r+1]-otime[r]))/(log(nconc[r+1]/nconc[r]))) -
((nconc[r+1]-nconc[r])*((otime[r+1]-otime[r])**2)/((log(nconc[r+1]/nconc[r]))**2))
AUClast <- sum(AUClast, delauc)
AUMClast <- sum(AUMClast, delaumc)
if(backExtrp){
if(noBackTime){
AUCnoC0 <- sum(AUCnoC0, delauc)
}else if(!noBackTime & r>1){
AUCnoC0 <- sum(AUCnoC0, delauc)
}
}
if(!is.null(AUCTimeRange)){if(ntime[r] >= TR1 & ntime[r+1] <= TR2){AUClower_upper <- sum(AUClower_upper, delauc)}}
if(doseType == "ss" && ssnPt != 0){
AUCtau <- sum(AUCtau, delauc); AUMCtau <- sum(AUMCtau, delaumc)
}
}else if(method == "linearup-logdown"){
if(nconc[r+1]>=nconc[r] | nconc[r+1]<=0 | nconc[r]<=0){
delauc <- mean(nconc[r:(r+1)])*(ntime[r+1]-ntime[r])
delaumc <- 0.5*((nconc[r+1]*otime[r+1])+(nconc[r]*otime[r]))*(otime[r+1]-otime[r])
AUClast <- sum(AUClast, delauc)
AUMClast <- sum(AUMClast, delaumc)
if(backExtrp){
if(noBackTime){
AUCnoC0 <- sum(AUCnoC0, delauc)
}else if(!noBackTime & r>1){
AUCnoC0 <- sum(AUCnoC0, delauc)
}
}
if(!is.null(AUCTimeRange)){if(ntime[r] >= TR1 & ntime[r+1] <= TR2){AUClower_upper <- sum(AUClower_upper, delauc)}}
if(doseType == "ss" && ssnPt != 0){
AUCtau <- sum(AUCtau, delauc); AUMCtau <- sum(AUMCtau, delaumc)
}
}else{
delauc <- (nconc[r+1]-nconc[r])*(ntime[r+1]-ntime[r])/log(nconc[r+1]/nconc[r])
delaumc <- ((((nconc[r+1]*otime[r+1])-(nconc[r]*otime[r]))*(otime[r+1]-otime[r]))/(log(nconc[r+1]/nconc[r]))) -
((nconc[r+1]-nconc[r])*((otime[r+1]-otime[r])**2)/((log(nconc[r+1]/nconc[r]))**2))
AUClast <- sum(AUClast, delauc)
AUMClast <- sum(AUMClast, delaumc)
if(backExtrp){
if(noBackTime){
AUCnoC0 <- sum(AUCnoC0, delauc)
}else if(!noBackTime & r>1){
AUCnoC0 <- sum(AUCnoC0, delauc)
}
}
if(!is.null(AUCTimeRange)){if(ntime[r] >= min(AUCTimeRange) & ntime[r+1] <= max(AUCTimeRange)){AUClower_upper <- sum(AUClower_upper, delauc)}}
if(doseType == "ss" && ssnPt != 0){
AUCtau <- sum(AUCtau, delauc); AUMCtau <- sum(AUMCtau, delaumc)
}
}
}
}
if(is.null(AUCTimeRange)){AUClower_upper <- AUClast}
if(backExtrp){
AUC_pBack_Ext_obs <- 100*(AUClast-AUCnoC0)/AUClast
AUC_pBack_Ext_pred <- AUC_pBack_Ext_obs
}
# MRTlast
if(adminType != "iv-infusion"){
MRTlast <- ifelse(doseType != "ss", AUMClast/AUClast, AUMCtau/AUCtau)
}else{
MRTlast <- (AUMClast/AUClast)-(TI/2)
}
########################################
# Extrapolation for the elimination phase
if(exists("lastPt") & exists("Lambda_z")){
AUCINF_obs <- exp(lconc[lnPt])/Lambda_z
AUMCINF_obs <- (exp(lconc[lnPt])/(Lambda_z**2))+(ltime[lnPt]*exp(lconc[lnPt])/Lambda_z)
AUCINF_pred <- lastPt/Lambda_z
AUMCINF_pred <- (lastPt/(Lambda_z**2))+(ltime[lnPt]*lastPt/Lambda_z)
# Calculate AUClower_upper if time range is outside last observation
if(Intpol12 | Intpol22){
uptime <- TR2
lowtime <- ifelse(Intpol12, TR1, max(ltime))
upconc <- CR2
lowconc <- ifelse(Intpol12, CR1, exp(lconc[lnPt]))
delauc <- (upconc-lowconc)*(uptime-lowtime)/log(upconc/lowconc)
AUClower_upper <- AUClower_upper + delauc
}
# Calculate AUCtau and AUMCtau if Tau is greater than Tlast
if(doseType == "ss" && (Tau > max(ltime) & AUCtau != 0)){
uptime <- Tau
lowtime <- max(ltime)
upconc <- ifelse(exp(slope*uptime+intercept)==0, 0.0001, exp(slope*uptime+intercept))
lowconc <- exp(lconc[lnPt])
delauc <- (upconc-lowconc)*(uptime-lowtime)/log(upconc/lowconc)
delaumc <- ((((upconc*uptime)-(lowconc*lowtime))*(uptime-lowtime))/(log(upconc/lowconc))) - ((upconc-lowconc)*((uptime-lowtime)**2)/((log(upconc/lowconc))**2))
AUCtau <- sum(AUCtau, delauc)
AUMCtau <- sum(AUMCtau, delaumc)
}
if(AUClast != 0 & AUCINF_obs != 0){
AUCINF_obs <- AUClast+AUCINF_obs; AUCINF_D_obs <- ifelse(!is.null(doseAmt),AUCINF_obs/doseAmt,NA); AUC_pExtrap_obs <- 100*(AUCINF_obs-AUClast)/AUCINF_obs
AUCINF_pred <- AUClast+AUCINF_pred; AUCINF_D_pred <- ifelse(!is.null(doseAmt),AUCINF_pred/doseAmt,NA); AUC_pExtrap_pred <- 100*(AUCINF_pred-AUClast)/AUCINF_pred
if(doseType=="ns"){
Vz_obs <- ifelse(!is.null(doseAmt),doseAmt/(Lambda_z*AUCINF_obs),NA); Cl_obs <- ifelse(!is.null(doseAmt),doseAmt/AUCINF_obs,NA)
Vz_pred <- ifelse(!is.null(doseAmt),doseAmt/(Lambda_z*AUCINF_pred),NA); Cl_pred <- ifelse(!is.null(doseAmt),doseAmt/AUCINF_pred,NA)
}else{
Vz_obs <- ifelse(!is.null(doseAmt),doseAmt/(Lambda_z*AUCtau),NA); Cl_obs <- ifelse(!is.null(doseAmt),doseAmt/AUCtau,NA)
Vz_pred <- NA; Cl_pred <- NA
}
}
# Calculate AUMC parameters
if(AUMClast != 0 & AUMCINF_obs != 0){
AUMCINF_obs <- AUMClast+AUMCINF_obs
AUMCINF_D_obs <- ifelse(!is.null(doseAmt), AUMCINF_obs/doseAmt, NA)
AUMC_pExtrap_obs <- 100*(AUMCINF_obs-AUMClast)/AUMCINF_obs
AUMCINF_pred <- AUMClast+AUMCINF_pred
AUMCINF_D_pred <- ifelse(!is.null(doseAmt), AUMCINF_pred/doseAmt, NA)
AUMC_pExtrap_pred <- 100*(AUMCINF_pred-AUMClast)/AUMCINF_pred
}
if(AUCINF_obs != 0 & AUMCINF_obs != 0){
MRTINF_obs <- ifelse(adminType!="iv-infusion", AUMCINF_obs/AUCINF_obs, ((AUMCINF_obs/AUCINF_obs)-(TI/2)))
}
if(AUCINF_pred != 0 & AUMCINF_pred != 0){
MRTINF_pred <- ifelse(adminType!="iv-infusion", AUMCINF_pred/AUCINF_pred, ((AUMCINF_pred/AUCINF_pred)-(TI/2)))
}
if(doseType == "ns" && (adminType=="iv-bolus" | adminType=="iv-infusion")){
if(exists("MRTINF_obs")) Vss_obs <- MRTINF_obs*Cl_obs
if(exists("MRTINF_pred")) Vss_pred <- MRTINF_pred*Cl_pred
}
if(doseType == "ss" && ssnPt != 0){
Cavg <- AUCtau/Tau
Clss <- ifelse(!is.null(doseAmt),doseAmt/AUCtau,NA)
Cmax <- max(nconc)
p_Fluctuation <- 100*(Cmax-Cmin)/Cavg
if(complete.cases(Lambda_z)) Accumulation_Index <- 1/(1-exp(-Lambda_z*Tau))
# re-define MRTINF for steady state and calculate Vss
if(complete.cases(AUCINF_obs) & complete.cases(AUMCINF_obs) & AUCtau != 0){
if(adminType == "iv-infusion"){
MRTINF_obs <- ((AUMCtau+Tau*(AUCINF_obs-AUCtau))/(AUCtau)-(TI/2))
MRTINF_pred <- ((AUMCtau+Tau*(AUCINF_pred-AUCtau))/(AUCtau)-(TI/2))
}else{
MRTINF_obs <- (AUMCtau+Tau*(AUCINF_obs-AUCtau))/AUCtau
MRTINF_pred <- (AUMCtau+Tau*(AUCINF_pred-AUCtau))/AUCtau
}
if(exists("MRTINF_obs")) Vss_obs <- MRTINF_obs*Clss
if(exists("MRTINF_pred")) Vss_pred <- MRTINF_pred*Clss
}
}
}
}
if(!exists("C0")) C0 <- NA
if(!exists("Tmax")) Tmax <- NA
if(!exists("Cmax")) Cmax <- NA
if(!exists("Cmax_D")) Cmax_D <- NA
if(!exists("Tlast")) Tlast <- NA
if(!exists("Clast")) Clast <- NA
if(!exists("AUClast")) AUClast <- NA
if(!exists("AUMClast")) AUMClast <- NA
if(!exists("MRTlast")) MRTlast <- NA
if(!exists("No_points_Lambda_z")) No_points_Lambda_z <- NA
if(!exists("AUC_pBack_Ext_obs")) AUC_pBack_Ext_obs <- NA
if(!exists("AUC_pBack_Ext_pred")) AUC_pBack_Ext_pred <- NA
if(!exists("AUClower_upper")) AUClower_upper <- NA
if(!exists("Rsq")) Rsq <- NA
if(!exists("Rsq_adjusted")) Rsq_adjusted <- NA
if(!exists("Corr_XY")) Corr_XY <- NA
if(!exists("Lambda_z")) Lambda_z <- NA
if(!exists("Lambda_z_lower")) Lambda_z_lower <- NA
if(!exists("Lambda_z_upper")) Lambda_z_upper <- NA
if(!exists("HL_Lambda_z")) HL_Lambda_z <- NA
if(!exists("AUCINF_obs")) AUCINF_obs <- NA
if(!exists("AUCINF_D_obs")) AUCINF_D_obs <- NA
if(!exists("AUC_pExtrap_obs")) AUC_pExtrap_obs <- NA
if(!exists("Vz_obs")) Vz_obs <- NA
if(!exists("Cl_obs")) Cl_obs <- NA
if(!exists("AUCINF_pred")) AUCINF_pred <- NA
if(!exists("AUCINF_D_pred")) AUCINF_D_pred <- NA
if(!exists("AUC_pExtrap_pred")) AUC_pExtrap_pred <- NA
if(!exists("Vz_pred")) Vz_pred <- NA
if(!exists("Cl_pred")) Cl_pred <- NA
if(!exists("AUMCINF_obs")) AUMCINF_obs <- NA
if(!exists("AUMC_pExtrap_obs")) AUMC_pExtrap_obs <- NA
if(!exists("AUMCINF_pred")) AUMCINF_pred <- NA
if(!exists("AUMC_pExtrap_pred")) AUMC_pExtrap_pred <- NA
if(!exists("MRTINF_obs")) MRTINF_obs <- NA
if(!exists("MRTINF_pred")) MRTINF_pred <- NA
if(!exists("Tmin")) Tmin <- NA
if(!exists("Cmin")) Cmin <- NA
if(!exists("Cavg")) Cavg <- NA
if(!exists("AUCtau")) AUCtau <- NA
if(!exists("AUMCtau")) AUMCtau <- NA
if(!exists("Clss")) Clss <- NA
if(!exists("Vss_obs")) Vss_obs <- NA
if(!exists("Vss_pred")) Vss_pred <- NA
if(!exists("p_Fluctuation")) p_Fluctuation <- NA
if(!exists("Accumulation_Index")) Accumulation_Index <- NA
if(!is.null(simFile) & dset == "obs"){
if(!onlyNCA){
NCAprm <- as.numeric(c(C0,Tmax,0,0,0,Cmax,0,0,0,Cmax_D,Tlast,Clast,AUClast,0,0,0,AUMClast,0,0,0,MRTlast,No_points_Lambda_z,AUClower_upper,0,0,0,Rsq,Rsq_adjusted,Corr_XY,Lambda_z,Lambda_z_lower,Lambda_z_upper,HL_Lambda_z,0,0,0,AUCINF_obs,0,0,0,AUCINF_D_obs,AUC_pExtrap_obs,AUC_pBack_Ext_obs,Vz_obs,Cl_obs,AUCINF_pred,0,0,0,AUCINF_D_pred,AUC_pExtrap_pred,AUC_pBack_Ext_pred,Vz_pred,Cl_pred,AUMCINF_obs,AUMC_pExtrap_obs,AUMCINF_pred,AUMC_pExtrap_pred,MRTINF_obs,MRTINF_pred,Tau,Tmin,Cmin,Cavg,AUCtau,AUMCtau,Clss,Vss_obs,Vss_pred,p_Fluctuation,Accumulation_Index))
names(NCAprm) <- c("C0","Tmax","simTmax","dTmax","npdeTmax","Cmax","simCmax","dCmax","npdeCmax","Cmax_D","Tlast","Clast","AUClast","simAUClast","dAUClast","npdeAUClast","AUMClast","simAUMClast","dAUMClast","npdeAUMClast","MRTlast","No_points_Lambda_z","AUClower_upper","simAUClower_upper","dAUClower_upper","npdeAUClower_upper","Rsq","Rsq_adjusted","Corr_XY","Lambda_z","Lambda_z_lower","Lambda_z_upper","HL_Lambda_z","simHL_Lambda_z","dHL_Lambda_z","npdeHL_Lambda_z","AUCINF_obs","simAUCINF_obs","dAUCINF_obs","npdeAUCINF_obs","AUCINF_D_obs","AUC_pExtrap_obs","AUC_pBack_Ext_obs","Vz_obs","Cl_obs","AUCINF_pred","simAUCINF_pred","dAUCINF_pred","npdeAUCINF_pred","AUCINF_D_pred","AUC_pExtrap_pred","AUC_pBack_Ext_pred","Vz_pred","Cl_pred","AUMCINF_obs","AUMC_pExtrap_obs","AUMCINF_pred","AUMC_pExtrap_pred","MRTINF_obs","MRTINF_pred","Tau","Tmin","Cmin","Cavg","AUCtau","AUMCtau","Clss","Vss_obs","Vss_pred","p_Fluctuation","Accumulation_Index")
}else{
NCAprm <- as.numeric(c(C0,Tmax,0,Cmax,0,Cmax_D,Tlast,Clast,AUClast,0,AUMClast,0,MRTlast,No_points_Lambda_z,AUClower_upper,0,Rsq,Rsq_adjusted,Corr_XY,Lambda_z,Lambda_z_lower,Lambda_z_upper,HL_Lambda_z,0,AUCINF_obs,0,AUCINF_D_obs,AUC_pExtrap_obs,AUC_pBack_Ext_obs,Vz_obs,Cl_obs,AUCINF_pred,0,AUCINF_D_pred,AUC_pExtrap_pred,AUC_pBack_Ext_pred,Vz_pred,Cl_pred,AUMCINF_obs,AUMC_pExtrap_obs,AUMCINF_pred,AUMC_pExtrap_pred,MRTINF_obs,MRTINF_pred,Tau,Tmin,Cmin,Cavg,AUCtau,AUMCtau,Clss,Vss_obs,Vss_pred,p_Fluctuation,Accumulation_Index))
names(NCAprm) <- c("C0","Tmax","simTmax","Cmax","simCmax","Cmax_D","Tlast","Clast","AUClast","simAUClast","AUMClast","simAUMClast","MRTlast","No_points_Lambda_z","AUClower_upper","simAUClower_upper","Rsq","Rsq_adjusted","Corr_XY","Lambda_z","Lambda_z_lower","Lambda_z_upper","HL_Lambda_z","simHL_Lambda_z","AUCINF_obs","simAUCINF_obs","AUCINF_D_obs","AUC_pExtrap_obs","AUC_pBack_Ext_obs","Vz_obs","Cl_obs","AUCINF_pred","simAUCINF_pred","AUCINF_D_pred","AUC_pExtrap_pred","AUC_pBack_Ext_pred","Vz_pred","Cl_pred","AUMCINF_obs","AUMC_pExtrap_obs","AUMCINF_pred","AUMC_pExtrap_pred","MRTINF_obs","MRTINF_pred","Tau","Tmin","Cmin","Cavg","AUCtau","AUMCtau","Clss","Vss_obs","Vss_pred","p_Fluctuation","Accumulation_Index")
}
}else{
NCAprm <- as.numeric(c(C0,Tmax,Cmax,Cmax_D,Tlast,Clast,AUClast,AUMClast,MRTlast,No_points_Lambda_z,AUClower_upper,Rsq,Rsq_adjusted,Corr_XY,Lambda_z,Lambda_z_lower,Lambda_z_upper,HL_Lambda_z,AUCINF_obs,AUCINF_D_obs,AUC_pExtrap_obs,AUC_pBack_Ext_obs,Vz_obs,Cl_obs,AUCINF_pred,AUCINF_D_pred,AUC_pExtrap_pred,AUC_pBack_Ext_pred,Vz_pred,Cl_pred,AUMCINF_obs,AUMC_pExtrap_obs,AUMCINF_pred,AUMC_pExtrap_pred,MRTINF_obs,MRTINF_pred,Tau,Tmin,Cmin,Cavg,AUCtau,AUMCtau,Clss,Vss_obs,Vss_pred,p_Fluctuation,Accumulation_Index))
names(NCAprm) <- c("C0","Tmax","Cmax","Cmax_D","Tlast","Clast","AUClast","AUMClast","MRTlast","No_points_Lambda_z","AUClower_upper","Rsq","Rsq_adjusted","Corr_XY","Lambda_z","Lambda_z_lower","Lambda_z_upper","HL_Lambda_z","AUCINF_obs","AUCINF_D_obs","AUC_pExtrap_obs","AUC_pBack_Ext_obs","Vz_obs","Cl_obs","AUCINF_pred","AUCINF_D_pred","AUC_pExtrap_pred","AUC_pBack_Ext_pred","Vz_pred","Cl_pred","AUMCINF_obs","AUMC_pExtrap_obs","AUMCINF_pred","AUMC_pExtrap_pred","MRTINF_obs","MRTINF_pred","Tau","Tmin","Cmin","Cavg","AUCtau","AUMCtau","Clss","Vss_obs","Vss_pred","p_Fluctuation","Accumulation_Index")
}
return(NCAprm)
}
Try the ncappc package in your browser
Any scripts or data that you put into this service are public.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.