Aurore Carrot* (1) ; Reza-Thierry Elaidi* (2) ; Olivier Colomban (1) ; Denis Maillet (3) ; Michel Tod (1) ; Benoit You** (1)(3) ; Stéphane Oudard** (2) *Equivalent role a first author **Equivalent role a last author
(1) EA3738 CICLY, UCBL - HCL Faculté de Médecine Lyon-Sud, Université Claude Bernard Lyon 1, univ Lyon, Oullins, France (2) Department of Medical Oncology, Georges Pompidou Hospital, 20 Rue Leblanc, 75908 Paris Cedex 15, France (3) Institut de cancérologie des Hospices Civils de Lyon (IC-HCL), Oncologie médicale, CITOHL, Lyon, France
Objectives: In patients with recurrent prostate cancer, the addition of docetaxel to androgen-deprivation therapy (ADT) is a standard treatment as it was associated with survival benefit in patients with hormone-naive metastatic prostate cancer [1]. However, this strategy was not found effective for improving the PSA-based progression-free survival (PFS) in patients presenting with rising prostate-specific antigen (PSA) levels after primary local therapy and high-risk factors but no evidence of metastatic disease (NCT00764166, Oudard JAMA Oncol 2019).
The objective of the present study was to assess the kinetics of PSA using a Kinetic-Pharmacodynamic (K-PD) model with a nonlinear mixed-effects population modelling approach, as a way of identifying a kinetic parameter of interest that could be used for further analyses meant to assess the prognostic/predictive value of PSA early kinetics regarding the benefit from docetaxel addition to ADT.
Methods: Data: The database comes from the phase III clinical trial (Oudard and al., 2019) [2], comparing ADT with or without docetaxel in 250 patients who had undergone primary local therapy for prostate cancer and who were experiencing rising PSA levels and at high risk of metastatic disease (N+, positive margins, Gleason ≥ 8, PSA velocity 0.75 ng/mL per year, PSA-DT ≤ 6 months, and time to PSA recurrence <12 months) in terms of PSA-progression free survival (PFS) PSA measurements were done every 3 weeks for patients treated with docetaxel, and every month for the other arm during the treatment duration. PSA levels were also measured after the end of the treatment until progression was noted, every 3 weeks. Potential clinical covariates are available.
Model: A K-PD model was used to fit the PSA kinetics during the first hundred days of treatment administration, to identify prematurely efficient predictive factors, due to the absence of available PK data [3]. In this model, PSA production was inhibited by the treatment through an indirect effect model. Model validation criteria were the NPDE (Normalised Prediction Distribution Errors), Relative Standard Errors (RSE), shrinkage values, classical Goodness Of Fits plots and Visual Predictive Checks using NONMEM 7.5.0.
Results: Of the 250 patients, 177 patients were selected for population and individual PSA kinetic modelling as they experienced PSA decline during the first hundred days of treatment. The best model was a three-compartment model with a depot compartment, a compartment for (unobserved) « treatment concentration », and a PSA compartment. PSA values were log-transformed. The error model took into account Lower Limit Of Quantification (LLOQ) values (=0.1 ng/mL) because 36.44% of PSA measurements were under the LLOQ. Five parameters were estimated: treatment kinetic rate constant K; PSA production rate KPROD; PSA elimination rate constant KELIM; initial PSA value PSA0E; treatment concentration to obtain 50% of maximal effect EC50. For all parameters, RSE were low (<25%) except for K and EC50 (<50%). Between-subjects variability was also estimated for each parameter (RSE<25% except for between-subjects variability of K and EC50: RSE<80%). No clinical covariate improved the objective function when added to the model. Visual Predictive Checks suggested good predictions of PSA kinetics.
Conclusions: This model allowed us to characterize the decrease of PSA in relapsed prostate cancer patients receiving hormonal therapy, with/without chemotherapy, during the first hundred days of administration. It leads to a better understanding of PSA decline. With this model, a series of individual indices related to PSA kinetics will be derived and tentatively correlated with OS and PFS such as the prognostic/predictive value of PSA kinetics regarding the benefit from docetaxel, which will be investigated.
References:
[1] M. Tucci et al., ‘Addition of Docetaxel to Androgen Deprivation Therapy for Patients with Hormone-sensitive Metastatic Prostate Cancer: A Systematic Review and Meta-analysis’, Eur. Urol., vol. 69, no. 4, pp. 563–573, Apr. 2016, doi: 10.1016/j.eururo.2015.09.013.
[2] S. Oudard et al., ‘Effect of Adding Docetaxel to Androgen-Deprivation Therapy in Patients With High-Risk Prostate Cancer With Rising Prostate-Specific Antigen Levels After Primary Local Therapy’, JAMA Oncol., vol. 5, no. 5, pp. 623–632, May 2019, doi: 10.1001/jamaoncol.2018.6607.
[3] B. You et al., ‘The strong prognostic value of KELIM, a model-based parameter from CA 125 kinetics in ovarian cancer: data from CALYPSO trial (a GINECO-GCIG study)’, Gynecol. Oncol., vol. 130, no. 2, pp. 289–294, Aug. 2013, doi: 10.1016/j.ygyno.2013.05.013.
Reference: PAGE 29 (2021) Abstr 9709 [www.page-meeting.org/?abstract=9709]
Poster: Drug/Disease Modelling - Oncology