Skip to content

Non-Compartmental Analysis (NCA)

Non-compartmental analysis provides model-independent exposure metrics following FDA and EMA guidelines.

Overview

NCA calculates pharmacokinetic parameters directly from concentration-time data without assuming a specific compartmental model.

Key Metrics

Metric Symbol Description
Maximum Concentration Cmax Peak observed concentration
Time to Maximum Tmax Time of Cmax
AUC to Last AUC0-t Area under curve to last observation
AUC to Infinity AUC0-∞ Extrapolated total exposure
Terminal Half-life t½ ln(2)/λz
Terminal Rate λz Slope of terminal phase
Clearance CL/F Dose/AUC (apparent)
Volume Vz/F CL/(F·λz) (apparent)
Mean Residence Time MRT AUMC/AUC

Documentation

Quick Start

Basic NCA

using NeoPKPD

# Concentration-time data
times = [0.0, 0.5, 1.0, 2.0, 4.0, 8.0, 12.0, 24.0]
conc = [0.0, 1.8, 2.0, 1.5, 1.0, 0.5, 0.25, 0.06]
dose = 100.0

# Run NCA
result = run_nca(times, conc, dose)

# Access results
println("Cmax: ", result.cmax, " mg/L")
println("Tmax: ", result.tmax, " h")
println("AUC0-t: ", result.auc_0_t, " mg·h/L")
println("AUC0-∞: ", result.auc_0_inf, " mg·h/L")
println("t½: ", result.t_half, " h")
println("CL/F: ", result.cl_f, " L/h")

With Configuration

config = NCAConfig(
    method = :log_linear,           # AUC calculation method
    lambda_z_min_points = 3,        # Min points for λz
    lambda_z_r2_threshold = 0.9,    # Quality threshold
    extrapolation_max_pct = 20.0,   # Warning if >20% extrapolated
    blq_handling = :zero            # Handle BLQ as zero
)

result = run_nca(times, conc, dose; config=config)

# Check quality metrics
println("λz R²: ", result.lambda_z_r_squared)
println("AUC extrapolated: ", result.auc_extrapolated_pct, "%")

AUC Calculation Methods

Linear Trapezoidal

\[AUC_{t_1 \to t_2} = \frac{(C_1 + C_2)}{2} \cdot (t_2 - t_1)\]

Best for: Ascending portions of the curve

Log-Linear Trapezoidal

\[AUC_{t_1 \to t_2} = \frac{(C_1 - C_2)}{\ln(C_1/C_2)} \cdot (t_2 - t_1)\]

Best for: Descending (elimination) portions

  • Linear trapezoidal for ascending
  • Log-linear for descending
config = NCAConfig(method = :lin_log_mixed)

Terminal Phase Analysis

λz Estimation

The terminal elimination rate constant is estimated by log-linear regression:

# Manual lambda_z estimation
lambda_z, r_squared, n_points = estimate_lambda_z(times, conc)

# With custom parameters
lambda_z, r_squared, n_points = estimate_lambda_z(
    times, conc;
    min_points = 3,
    r2_threshold = 0.9
)

Half-life

\[t_{1/2} = \frac{\ln(2)}{\lambda_z}\]

NCA Result Structure

struct NCAResult
    # Primary metrics
    cmax::Float64
    tmax::Float64
    auc_0_t::Float64
    auc_0_inf::Float64
    t_half::Float64

    # Terminal phase
    lambda_z::Float64
    lambda_z_r_squared::Float64
    lambda_z_n_points::Int

    # Clearance and volume
    cl_f::Float64
    vz_f::Float64

    # Additional metrics
    mrt::Float64                    # Mean residence time
    auc_extrapolated_pct::Float64   # % AUC extrapolated

    # Metadata
    method::Symbol
    dose::Float64
end

FDA/EMA Compliance

NeoPKPD NCA calculations follow regulatory guidance:

  • FDA Guidance for Industry: Bioavailability and Bioequivalence Studies
  • EMA Guideline on the Investigation of Bioequivalence

Key Requirements

Requirement NeoPKPD Implementation
λz from ≥3 points Configurable minimum
R² > 0.9 for λz Configurable threshold
AUC extrapolation <20% Warning flag
Log-linear interpolation Supported method
BLQ handling Multiple options

Next Steps