J PKPD

J PKPD 31(6), December 2004. ようやく出た（ようやく見えるようになった？）

• Rombout, Aarons, Karlsson, Man, Mentré, Nygren, Racine, Schaefer, Steimer, et al., "Modelling and Simulation in the Development and use of Anti-Cancer Agents: An Underused Tool?"
  • 全然関係ないんですけど，Rombout さんの所属はドイツの Grünenthal．いや，事実を記したまで．
  • おりしも，多発性骨髄腫に対するサリドマイドの治験が始まる，というニュース（読売，日経，google ニュース検索）が,昨日あったばかり．
• Tornøe, ..., and Jonsson, "Pharmacokinetic/Pharmacodynamic Modellingof GnRH Antagonist Degarelix: A Comparisonof the Non-linear Mixed-Effects Programs NONMEM and NLME"
  • NONMEM と nlme の比較．ただし，S-Plus の nlme ではなく，R での話．その理由は，R 用に implement された nlmeODE パッケージを使うため．
  • 結論は次のとおり．

In general, NONMEM seems to be superior in accuracy, stability, flexibility, and speed compared to NLME but when performing graphical and statistical data analysis, NLME is preferred over NONMEM.
(...)
The comparison revealed that NLME should not be considered as a replacement for NONMEM but perhaps more as an alternative.

• • nlmeODE の論文は，Tornoe, ..., Jonsson, et al. (2004) Comput Methods Programs Biomed, "Non-linear mixed-effects pharmacokinetic/pharmacodynamic modelling in NLME using differential equations"
• Nestorov, Zitnik, and Ludden, "Population Pharmacokinetic Modeling of Subcutaneously Administered Etanercept in Patients with Psoriasis"
  • 普通の PPK 解析をしました，という論文．濃度の prediction interval とかシミュレーションで作ってます．CL と Vd に time-dependency が認められたので，（簡単に）モデル化した．
  • 1,000 回ブートストラップを実施して，930 回分を採用した．

One hundred and forty three out of 1000 bootstrap runs, used to determine the confidence parameters of the model parameters resulted in NONMEM rounding error at three significant digits. The latter were rerun with four significant digits and only 70 of them resulted in NONMEM rounding error. As a result, the 95% confidence intervals for each parameter were calculated over 930 bootstrap runs.