Virtual Patient Cohorts using Modeling and Simulation to support drug development in rare diseases
M Savelieva (1), I Baltcheva (1)
(1) Novartis Pharma AG, Basel, Switzerland
Objectives: Rare diseases affect approximately 350 million people worldwide with only ~5% of diseases having approved treatments. Drug development in these indications is challenging due to sometimes a limited disease understanding, small patient numbers, and difficulties with recruitment. Single arm trials often used in rare disease drug development do not allow for a direct comparison of the treatment effects of investigated drug vs. placebo or active control. The need for new methodologies and inclusion of varied data sources to augment the information generated in a clinical trial setting has been acknowledged by regulators and payers [1]. Statistical approaches to augment the knowledge generated in a single arm clinical trial rely on synthetic control arms using historical or RWD and are often of limited use in a “small sample” setting. Here, we present a case study that serves as a proof-of-concept to illustrate how the use of pharmacometric Modelling and Simulation methodologies combined with RWD can strengthen the evidentiary standard of a trial, enhance drug positioning, and accelerate patient access to treatments in rare disease indications. We evaluate the use of PKPD models in combination with RWD in contextualizing the results from a single arm clinical trial in a rare disease indication on the example of a Phase 3 trial in Paroxysmal Nocturnal Hemoglobinuria (PNH) patients treated with iptacopan (NCT04820530, APPOINT-PNH) [2]. The challenges and limitations of the proposed approach are discussed.
Methods: The data collected throughout the development program of iptacopan in PNH in two Phase 2 trials (NCT03439839, NCT03896152), and two Phase 3 trials, APPLY-PNH (NCT04558918) [3] and APPOINT-PNH), were previously used to develop a PKPD model to characterize one of the components of the primary endpoint, haemoglobin (Hb) response. The pharmacokinetics and Hb levels under iptacopan treatment were described using a longitudinal indirect response model [4, 5, 6]. To contextualize the results of the uncontrolled APPOINT-PNH trial, we generated a “Virtual Patient Cohort” using model-based predictions of the Hb response for this population as if patients were treated with a Standard of Care (SoC) anti-C5 drug (eculizumab) [7, 8]. To do so, we developed a Hb model under SoC where the disease-specific model parameters (i.e., the rate of Hb production) were borrowed from the iptacopan model. The remaining (drug-specific) parameters were assessed using RWD [8, 9, 10]. This Virtual Patient Cohort allowed for an indirect comparison between the levels of Hb observed in the trial for patients treated with iptacopan and the anticipated one based on modelling and simulation under SoC for the same patient population.
Results: In the APPOINT-PNH trial patients treated with iptacopan achieved Hb levels of at least 12 g/dL sustained between day 126 and day 168 without the need for red blood cell (RBC) transfusions (primary endpoint) in 62.8% (95% CI, 47.5%-77.5%) of patients [11]. The longitudinal Hb levels under eculizumab simulated up to a steady-state for the patient population of iptacopan APPOINT-PNH (n=40, adjusted for sex and baseline Hb) showed that less than 10% of patients achieve the desired Hb threshold of 12 g/dL at any time point and after reaching Hb steady-state. This result provides evidence of a better anaemia control under iptacopan as compared to SoC and hence allowed to contextualize the results of this single arm trial.
Conclusions: We showcased how Modeling and Simulation techniques using clinical trial data and complemented with RWD allow to generate Virtual Patient Cohorts that substitute SoC information missing in single arm trials for rare disease indications. The approach presents several advantages: it allows to account for the underlying disease dynamics and components of clinical response (e.g., biomarkers) at individual and population levels; different data sources can be leveraged in a way that reduces bias and increases robustness of the inference in the situations where data is sparse. The challenges related to this approach encompass access to RWD and missing data relevant for modelling. The Virtual Patient Cohort generated with the help of PKPD and RWD can be a powerful tool allowing to contextualize the outcome of single arm trials and provide support in addressing potential questions of interest to a clinician, regulator, or payer.
References:
[1] Horizon INVENTS: https://ecrin.org/projects/invents
[2] Régis Peffault de Latour, Bing Han, Yasutaka Ueda, Yu Cheng, Georgina Bermann, Marion Dahlke, Antonio M. Risitano, CT-121: Phase 3 Study of the Efficacy and Safety of Iptacopan (LNP023), an Oral Factor B Inhibitor, in Adult Patients with Paroxysmal Nocturnal Hemoglobinuria (PNH) Naïve to complement Inhibitor Therapy, Clinical Lymphoma Myeloma and Leukemia, Volume 21, Supplement 1, 2021, p. S450, https://doi.org/10.1016/S2152-2650(21)01999-6.
[3] Regis Peffault de Latour, Alexander Roeth, et. al.; Oral Monotherapy with Iptacopan, a Proximal Complement Inhibitor of Factor B, Has Superior Efficacy to Intravenous Terminal Complement Inhibition with Standard of Care Eculizumab or Ravulizumab and Favorable Safety in Patients with Paroxysmal Nocturnal Hemoglobinuria and Residual Anemia: Results from the Randomized, Active-Comparator-Controlled, Open-Label, Multicenter, Phase III Apply-PNH Study. Blood 2022; 140 (Supplement 2): LBA–2. doi: https://doi.org/10.1182/blood-2022-171469
[4] I Baltcheva et al, The challenge of drug development in rare diseases: impact of pharmacometrics on the quantitative dose justification of iptacopan in PNH. Submitted to PAGE 2024
[5] Risitano AM, Peffault De Latour R, Jang JH et al. Exposure-Response Relationships between the Complement Factor B Inhibitor Iptacopan and Lactate Dehydrogenase (LDH) and Hemoglobin (Hb) in Patients (Pts) with Paroxysmal Nocturnal Hemoglobinuria (PNH). Blood 2023; 142:5643.
[6] Iptacopan monotherapy in patients with paroxysmal nocturnal hemoglobinuria: a 2-cohort open-label proof-of-concept study, J H Jang et. al., Blood Advances, 2022 Aug 9; 6(15):4450-4460
[7] Pharmacology, Pharmacokinetics and Pharmacodynamics of Eculizumab, and Possibilities for an Individualized Approach to Eculizumab, K L Wijnsma et al, Clinical Pharmacokinetics (2019) 58:859–874, https://doi.org/10.1007/s40262-019-00742-8
[8] Long-term outcomes of patients with paroxysmal nocturnal hemoglobinuria treated with eculizumab in a real-world setting, K Versmold et al, Eur J Haematol. 2023; 111:84–95, DOI: 10.1111/ejh.13970
[9] Soliris EPAR, https://www.ema.europa.eu/en/documents/scientific-discussion/soliris-epar-scientific-discussion_en.pdf
[10] Soliris FDA Review (2007), Center for Drug Evaluation and Research. Approval package for: Application Number: 125166. Clinical pharmacology biopharmaceutics review.
[11] Risitano AM, Han B, Ueda Y, et al. Oral Complement Factor B Inhibitor Iptacopan Monotherapy Improves Hemoglobin to Normal/Near-Normal Levels in Paroxysmal Nocturnal Hemoglobinuria Patients Naïve to Complement Inhibitors: Phase III APPOINT-PNH Trial. Presented at: 49th Annual Meeting of the European Society for Blood and Marrow Transplantation (EBMT); April 23-36, 2023; Paris, France.
