What is PAGE?

We represent a community with a shared interest in data analysis using the population approach.


2001
   Basel, Switzerland

Population PD(-PK) modeling and clinical trial simulation, characterizing schedule dependence of hematotoxicity and other Phase II trial design features for a new oral anticancer drug.

Timothy Goggin1,2Philippe Jacqmin4, Ronald Gieschke1,3, Goonaseelan (Colin) Pillai1,2, Eric Snoeck4, Pascal Girard4, Jean-Louis Steimer1,3.

1Modeling and Simulation Team, 2Clinical Pharmacology and 3Biostatistics,

In 2 phase 1 studies, neutropenia was identified as the major dose limiting toxicity of a new anticancer drug. We used the time course of neutrophil counts in 61 patients (about 70% of the final data set) given 4 diverse treatment regimens (on / off cycles of treatment) and the K-PD model (Jacqmin et al PAGE 2001) to characterize the time dependent hematotoxicity induced by the drug. Parameter estimates and standard errors / CV% are shown in the table. The final model included dependency of age and clearance/F of the drug on EDK50.

Final Run K-PD Model (Objective Function=174.5)

Parameter#

Units

Mean

SE

Inter-individual CV%

ALAG

h

102

10.2

-

KDE

1/h

0.00405

0.000749

101

KS

N/h/mcL

479

32.4

34.2

KD

1/h

0.103

Fixed to known physiological value

 

EDK50

mg/day

305*

43.9

31.5

Gamma

-

2.90

0.645

101

Variance

SE

CV%

   

Residual Variability

0.368

0.00978

60.7

 

# ALAG -lag time; KDE -effect site equilibrium rate; KS (or Rin) zero order rate of synthesis of neutrophils; KD (or Kout)- first order rate of elimination of neutrophils to tissues; EDK50 -dose rate resulting in 50% inhibition of neutrophil synthesis; Gamma -Hill coefficient for shape of dose response * for Clearance/F of 73.7 L/h and age of 58 years

Through implementation of this model, together with a predefined dose modification scheme (increase, decrease, withholding), in PharsightTSTM 2.1, we explored the distribution of outcomes for the 'base design study' as proposed by the development project team. The results indicated that the 2 proposed intermittent treatment regimens (a 4 week cycle with 14 days treatment and a 3 week cycle with 4 days treatment) produced very similar hematotoxicity and dropout rates. Predefined dose modifications were very successful in maintaining patients in the study. Additional 'simulation scenarios' indicated that controllable design features such as the pre-defined dose increase in selected patients, after completion of the first cycle of treatment, could be adjusted to increase the cumulative exposure to the drug without impacting hematotoxicity. Extrapolating beyond the range of tested regimens to continuous administration, it was predicted, that due to schedule dependence of the neutropenia induced by the drug, the same cumulative dose would result in less hematotoxicity and that a higher cumulative dose could be administered without worsening the hematotoxicity profile.



Top