Aims: Optimize the dosing regimen for an antibody-based therapy of psoriasis using Phase 1 and 2 data.
Methods: Optimizing dosing regimens is challenging in situations where the data is generated under non-steady-state conditions, with a delayed response, and there is uncertainty about the best route (IV vs SC) of administration. Under such circumstances, traditional methods of dose finding are often inefficient or inadequate. An iterative approach was used to sequentially build a PK/PD model of the 'regimen-response space' using emerging Phase 2 data: "forward" model-based predictions were compared to the new data at each iteration, and the model was then incrementally refined using the cumulative data. Model assessment was performed using standard goodness of fit diagnostics, visual predictive checks, and repeated external validation by prospectively predicting outcomes of the next iteration read-outs. Regimen optimization was performed using predictions based on the final model compared to competitor data.
Results: PK was described by a two compartment disposition model with first order absorption for SC administration. PD was described by a turnover model with stimulatory effect on the decrease of disease activity (Kout), driven by a sigmoidal Emax model as a function of drug concentration. This model captured the main trends across doses, dosing regimens and routes well. Using this model, different regimens were evaluated by simulation. Two previously untested SC regimens were finally selected as optimal for assessment in phase 3: 150mg and 300mg administered at weeks 0, 1, 2, 3, and 4, followed by dosing every 4 weeks from week 8. Both of these regimens were predicted to improve response over the subcutaneous regimens used in phase 2 and other available treatments. The Phase 3 results provided data for external validation of the model predictions.
Conclusions: An iterative PK/PD modelling approach integrating Phase 1 and 2 data following IV and SC administrations was used prospectively to recommend dosing regimens for phase 3. The recommended regimens deemed optimal were different from any of the previous clinically-tested regimens. The phase 3 studies confirmed the efficacy and safety of the new regimens, and supported regulatory approval of a product with optimized treatment schedules. This analysis highlighted:
the value of comprehensive informative data to underpin exploration of the 'input-output space'
the requirement of using appropriate model-based methods that can exploit this data
the necessity of generating confidence that approach is predictive and robust
Focusing on systematically modeling the 'regimen-response space', as opposed to the traditional focus on responses to a limited number fixed regimens, is a powerful means of optimization of the treatment especially when the treatment options are complex and the response is temporally dislocated from the treatment.
Langley RG, Elewski BE, Lebwohl M, Reich K, Griffiths CE, Papp K, Puig L, Nakagawa H, Spelman L, Sigurgeirsson B, Rivas E, Tsai TF, Wasel N, Tyring S, Salko T, Hampele I, Notter M, Karpov A, Helou S, Papavassilis C; ERASURE Study Group; FIXTURE Study Group. Secukinumab in plaque psoriasis–results of two phase 3 trials. N Engl J Med. (2014) 371(4):326-38.