Anne-Gaelle Dosne1, Nele Goeyvaerts1, Elodie Valade1, Peter De Porre1, Anjali Avadhani2, Lilian Y. Li2, Daniele Ouellet2, Juan Jose Perez Ruixo1
1Janssen Research & Development, Beerse, Belgium 2Janssen Research & Development, Spring House, PA, USA
Erdafitinib is an oral, pan-fibroblast growth factor receptor (FGFR) tyrosine kinase inhibitor, in development for the treatment of locally advanced or metastatic urothelial carcinoma (mUC) and other cancers with certain FGFR genetic alterations. Consistent with the erdafitinib mechanism of action, serum phosphate (PO4) is a pharmacodynamic (PD) marker of FGFR engagement, and proposed to be a biomarker for efficacy and safety. To maximize efficacy and limit toxicity, erdafitinib dosing in the pivotal, Phase 2, multicenter, open-label BLC2001 study (NCT02365597) was adapted for each individual based on their PO4 concentrations after 2-4 weeks of erdafitinib treatment.
Objectives:
To characterize the exposure-response (ER) relationship between PO4 and selected efficacy and safety endpoints based on data from study BLC2001, to confirm PO4 as a biomarker for erdafitinib dose individualization.
Methods:
The ER analysis was performed using data from patients randomized to one of two dosing regimens: 6 mg group (6 mg erdafitinib daily with PD-guided up-titration to 8 mg daily, based on PO4 at Day 28); and the 8 mg group (8 mg erdafitinib with PD-guided up-titration to 9 mg based on PO4 at Day 14). The relationship between erdafitinib exposure and PO4 concentrations was described earlier using a population PK-PD model[1], which was used to conduct simulations to support the proposed PD-guided dosing regimen[1]. Individual average PO4 values were derived based on posthoc PK and PK-PD parameter estimates, which were used as the exposure metric for ER analysis. Efficacy endpoints included objective response rate based on best response (ORR), progression-free survival (PFS), and disease control rate (DCR). Safety was assessed by incidence of selected clinical safety endpoints: eye disorders, central serous retinopathy (CSR), nail disorders, palmar-plantar erythrodysaesthesia (PPES) and skin disorders. The ER relationship between average PO4 and ORR, PFS, or adverse events (AE) was evaluated using either logistic regression or Cox models. Baseline prognostic factors were also evaluated in these models. The analysis was carried out using the R Project for Statistical Computing, version 3.4.1 or higher for Windows[2].
Results:
For the efficacy ER analysis, data from 156 chemotherapy-relapsed/refractory patients in the 6 mg group (n=68) and the 8 mg group (n=88) were available. The safety ER analysis included data from 177 patients (n=78 in the 6 mg group and n=99 in the 8 mg group). Higher average daily PO4 from the start of erdafitinib up to the first response assessment (planned at 6 weeks) was associated with significantly higher ORR, with the type of FGFR alterations included as a significant covariate (Odds Ratio (OR): 1.38; 95%, CI: 1.02-1.86 for 1 mg/dL increase in PO4; p=0.04), longer PFS (Hazard Ratio (HR): 0.80; 95%CI: 0.67-0.94 for 1 mg/dL increase in PO4; p=0.01), and greater DCR (by PO4 terciles, p<0.001). Results showed that achieving PO4 concentrations above 5.5 mg/dL following PD-guided individualized up-titration and recommended dose modification was expected to be associated with clinically relevant improvement in ORR and PFS in the target population. Further, the PK-PD model revealed an attenuation of the erdafitinib effect on PO4 over time, contributing to decreasing PO4 levels. Including PO4 as a time-dependent covariate in the Cox model for PFS led to a similar HR estimate compared to average PO4 up to 6 weeks, and showed that decreasing PO4 levels over time are likely not relevant for PFS. The incidence of key AEs of any grade, including eye, nail or skin disorders, as well as CSR and PPES, had a positive and statistically significant relationship with average daily PO4 up to the day of event (OR range 1.61-2.84 for 1 mg/dL increase in PO4). The magnitude of the association was strongest (OR>2 for 1 mg/dL increase PO4) for nail and eye disorders.
Conclusions:
The relationship between PO4 and efficacy/safety as characterized in the ER analysis justified the erdafitinib dose individualization to maximize the benefit/risk ratio of erdafitinib treatment.
References:
[1] Dosne et al PAGE 27 (2018) Abstr 8584 [www.page-meeting.org/?abstract=8584]
[2] R Development Core Team. R: A language and environment for statistical computing. R Foundation for statistical computing, Vienna, Austria, 2012. http://www.R-project.org/
Reference: PAGE 28 (2019) Abstr 8877 [www.page-meeting.org/?abstract=8877]
Poster: Drug/Disease Modelling - Oncology