Simultaneous modelling of PSA production in Prostatic Benign Hyperplasia (PBH) and prostatic adenocarcinoma patients treated by prostate surgery
Alexandre SOSTELLY (1, 2), Emilie HÉNIN (1), Benoit YOU (2), Pascal GIRARD (2), Mats O KARLSSON (1)
(1) Department of Pharmaceutical Biosciences, Uppsala University, Sweden , (2) Université Lyon 1, EA3738, CTO, Faculté de Médecine Lyon-Sud, Oullins, France
Objectives: Prostate Specific Antigen (PSA) is mainly produces by prostate and is used as a biomarker of prostatic diseases. Prostatic Benign Hyperplasia (PBH) is a benign adenoma developed within transitional zone and treated by its removal [1]. Prostatic adenocarcinoma initially develops in peripheral zone and is treated by radical prostatectomy [2]. The aim of this work is to characterise PSA production from each part of the prostate (transition zone, peripheral zone, cancer zone) using a semi-mechanistic model in order to better characterize patient relapse.
Methods:
Patients: Postoperative PSA concentrations for 149 patients with prostatic diseases: 81 PBH patients treated by Millin´s adenomectomy and 68 cancer patients treated by radical prostatectomy were assessed [3]. Patients’ characteristics were similar in both cohorts and a mean of 2.5 assays per PBH patient and 9 per cancer patient were available. No preoperative PSA concentration was available. Prostate volume, prostate residual volume and tumor weight were assessed after surgery but information was missing in 20% of patients.
Model: Non linear mixed effects model including IIV was fitted to the PSA plasmatic concentration. Different rates of PSA production for peripheral zone, central zone and cancer zone and PSA elimination from plasma were described by first order constant. Prostate volume was treated as observations in order to handle missing data. Model was built firstly on PBH and cancer data separately and then simultaneously on both subpopulations using NONMEM VI with FOCE.
Results: PSA concentrations were best described by a two-compartment model with elimination from the plasma compartment. Parameter estimation was supported by both datasets. PSA elimination was “directly” observed in cancer patients after radical prostatectomy (no more PSA production) and PSA production from transition zone was observed in PBH patients after Millin’s adenomectomy (residual PSA production). Rates of PSA production were in accordance to their description in the literature: higher rate in cancer zone than in peripheral zone than in central zone. Model evaluation was performed using classical goodness of fit and visual predictive check.
Conclusions: A model for PSA production, distribution and elimination was built in PBH and cancer patients after surgery. In the future, this model may be used to assess the quality of prostate surgery and to help the prediction of relapse risk after prostate surgery.
References:
[1] Linton, H. J., L. S. Marks, et al. (2003). Benign prostate-specific antigen (BPSA) in serum is increased in benign prostate disease. Clin Chem 49(2): 253-9.
[2] Stamey, T. A., N. Yang, et al. (1987). Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med 317(15): 909-16.
[3] You, B., P. Perrin, et al. (2008). Advantages of prostate-specific antigen (PSA) clearance model over simple PSA half-life computation to describe PSA decrease after prostate adenomectomy. Clin Biochem 41(10-11): 785-95.
