A. Solms, M. Frei, A. Keller, S. Willmann
Bayer AG, Pharmacometrics/Modeling & Simulation
Introduction: Patients with end-stage kidney disease (ESKD) requiring hemodialysis (HD) have an increased risk of thrombotic events and bleeding. Reduction of factor XI (FXI) is a promising strategy in these patients to reduce the risk of thrombotic events while not increasing bleeding risk as shown in large epidemiological studies in FXI deficient (hemophilia C) subjects [1]. Additionally, it has been demonstrated that pharmacologically induced reduction of FXI prevents venous thromboembolism without increasing bleeding risk [2,3,4,5].
FXI-ASO (IONIS FXIRx, BAY 2306001) is an antisense oligonucleotide that inhibits FXI production in the liver and thereby pharmacologically mimics FXI deficiency. FXI-ASO was studied in two phase 2 studies in ESKD [6,7].
Objectives: The objective of this analysis was to utilize a previously developed PK/PD model of FXI-ASO [8] to quantify the extent of FXI inhibition achieved in these two studies and evaluate if reduced FXI levels are associated with an increased bleeding risk. Additional objectives were to investigate the relationship with efficacy and improvement in semi-quantitative clotting scores (SQCS) obtained from the dialyzer/circuits.
Methods: ASO-CS4 (CS4) and ASO-CS5 (CS5) were placebo-controlled Ph2 studies in ESKD [6,7]. FXI-ASO was administered once weekly subcutaneously at doses of 200 and 300mg for 14 weeks (CS4, N = 49) and of 200, 250 and 300mg for 26 weeks (CS5, N = 210). In CS5 a subgroup of ESKD patients with atrial fibrillation (Afib, N = 25) was included. In all patients FXI levels were measured repeatedly.
If bleeds occurred these were categorized as minor, non-major clinically relevant (NMCR) or major. Only the worst bleed per subject was included, procedure-related bleeds were omitted from this analysis.
SQCS were longitudinally assessed in all patients; SQCS were obtained visually and reported in descriptive categories from 1 to 4 [6]; for the analysis a binary variable was derived with merged scores of 1-2 (“clean”) vs. 3-4 (“clotted”).
Efficacy events of interest were myocardial infarction, stroke, systemic embolism and cardiovascular mortality and were only available for CS5.
A previously developed PK/PD model [8] for FXI-ASO was used as starting point to generate individual FXI predictions and metrices as input for an exposure-response analysis of bleeding events, efficacy events and SQCS.
Results: FXI levels could adequately be described by the FXI-ASO model [8] with minimal adaption due to the inclusion of new FXI-ASO data. Based on post-hoc estimates individual FXI trajectories were simulated. From these simulation metrices such as FXI pre-dose, average FXI and minimum FXI level per Cycle (defined as 4 administrations) were derived.
FXI reduction was achieved in a dose-dependent manner with steady state reached after approximately 6-12 weeks. In total, treatment with FXI-ASO in studies CS4 and CS5 lead to approximately 48 patient-years with considerably reduced FXI level (below 0.5 U/mL).
In total 9 (3) and 42 (11) subjects in CS4 and CS5 reported a bleeding (clinically relevant) after start of treatment. Clinically relevant bleed rates were neither significantly different between dose groups, nor showed any trend with dose. Data does not indicate an increased bleeding risk at reduced FXI level, although the number of events were small. An exploratory assessment including also minor bleeds showed a higher bleeding rate in the 300mg dose group, but with no apparent association with reduced FXI level.
Although most efficacy events occurred in normal FXI range (7 out of 9), the data remains inconclusive regarding prevention or absence of prevention of events with reduced FXI level due to the small number of events. Apparently, Afib-patients seem to have much higher risk compared to non-Afib patients (16% vs. 3%).
While the majority of SQCS were “clean” (n=4179) compared to “clotted” (n=882), a significant improvement (p<0.001) in risk of “clotted” SQCS could be identified with reduced FXI levels obtained by logistic regression analysis.
Conclusions: CS4 and CS5 generated approximately 8 and 40 patient-years data with reduced FXI below 0.2 U/mL and below 0.5 U/mL. Based on these data no sign of increased clinically relevant bleeding risk but a clear trend for improved SQCS with pharmacologically reduced FXI levels could be detected; as only limited number of efficacy events occurred, the impact of reduced FXI levels on prevention of thrombotic events could not be assessed reliably.
References:
[1] M Preis et al., Factor XI deficiency is associated with lower risk for cardiovascular and venous thromboembolism events, Blood 2017; 129 (9), https://doi.org/10.1182/blood-2016-09-742262
[2] HR Brüller et al., Factor XI Antisense Oligonucleotide for Prevention of Venous Thrombosis, NEJM 2015; 372 (3), https://doi.org/10.1056/NEJMoa1405760
[3] JI Weitz et al., Effect of Osocimab in Preventing Venous Thromboembolism Among Patients Undergoing Knee Arthroplasty – The FOXTROT Randomized Clinical Trial, JAMA 2020; 323 (2), https://doi:10.1001/jama.2019.20687
[4] JI Weitz et al., Milvexian for the Prevention of Venous Thromboembolism, NEJM 2021; 385, https://doi:10.1056/NEJMoa2113194
[5] P Verhamme et al., Abelacimab for Prevention of Venous Thromboembolism, NEJM 2021; 385, https://doi:10.1056/NEJMoa2105872
[6] Walsh et al., Phase 2 Study of the Factor XI Antisense Inhibitor IONIS-FXIRx in Patients With ESRD, Kidney International Reports 2022; 7, https://doi.org/10.1016/j.ekir.2021.11.011
[7] ClinicalTrials.gov, Identifier NCT03358030, A Study of ISIS 416858 Administered Subcutaneously to Participants With End-Stage Renal Disease (ESRD) on Hemodialysis (EMERALD), https://clinicaltrials.gov/ct2/show/NCT03358030
[8] Willmann et al, PK/PD modeling of FXI antisense oligonucleotides to bridge the dose-FXI activity relation from healthy volunteers to end-stage renal disease patients, CPT:PSP 2021; 10 (8), https://doi.org/10.1002/psp4.12663
Reference: PAGE 30 (2022) Abstr 10146 [www.page-meeting.org/?abstract=10146]
Poster: Drug/Disease Modelling - Other Topics