Purpose: We introduce here the most comprehensive mathematical model to date for high-grade serous ovarian cancer patient treatment response and use it to predict the effect of novel combination therapies.

Introduction: Platinum-based chemotherapy constitutes the backbone for clinical care of advanced solid cancers, such as high-grade serous ovarian cancer (HGSOC). The vast majority of patients, however, become resistant to platinum, which eventually leads to generally fatal refractory disease. As HGSOC tumors are heterogeneous, cancer cells often have activated many resistant mechanisms, and targeting a single mechanism at a time is unlikely to result in significant improvement in survival rates. A systems biology approach, where resistance mechanisms are comprehensively considered, is required in efforts to overcome platinum resistance.

Methods: Platinum resistance is a multifactorial process and the most important resistance mechanisms are: 1) reduced intake or increased efflux leading to reduced platinum bioavailability in a cell, 2) DNA repair mechanisms that overcome platinum-induced DNA adducts, and 3) dysfunctional apoptosis machinery.

The herein established mathematical model is based on three major resistance mechanisms and it models tumor growth and evolution as a deterministic continuous time, multitype branching process. Importantly, we have used comprehensive pre- and postoperative clinical data, including high-resolution 18F-fluorodeoxyglucose positron emission tomography/computed tomography images, from a HGSOC patient cohort. This allows accurate calibration of the model parameters, which ensures that the model is able to capture clinical course of HGSOC patients. In addition to quantifying the effect of a platinum-taxane to tumor burden of an HGSOC patient, we have modeled the impact of debulking surgery. This enables us to use the mathematical model in simulating “virtual HGSOC patients” who undergo standard-of-care (SOC) therapy, i.e., debulking surgery and platinum-taxane therapy.

Results: Our model-generated virtual patients show almost identical treatment responses with two validation HGSOC cohorts that together contain data from 105 HGSOC patients. The average virtual HGSOC patient has platinum-free interval (PFI) of six months, which is identical to the mean PFI in the training and independent validation cohorts. The ability to create in silico HGSOC patients whose responses correspond to real-life HGSOC cohorts was supported statistically by the Kolmogorov-Smirnov test.

As our model can capture the heterogeneous responses of individual HGSOC patients, we utilized it to evaluate the added value of combining novel targeted therapies to the current first-line SOC therapy. The results from the virtual clinical trials with selected targeted agents are in line with the results from early-phase clinical trials and pinpoint the most effective combination strategies. Further, our results show that targeting the three most important platinum-resistant mechanisms simultaneously with clinically available agents provides a significant survival advance to HGSOC patients.

Conclusions: High failure rate in translating preclinical results to successful clinical trials is a major issue in oncology. The herein introduced mathematical framework enables creating virtual patient cohorts that capture the response variation in real-life patient cohorts. This allows for rapid and reliable estimate of the added value of the novel therapy in comparison to SOC therapy to optimize clinical trial design. Here we focused on overcoming platinum resistance in HGSOC and our results show substantial added value of combining three early-phase clinical trial agents, which target the three major resistant mechanisms and have been individually combined with SOC.

Citation Format: Sampsa Hautaniemi, Emilia Kozlowska, Anniina Färkkilä, Tuulia Vallius, Olli Carpen, Seija Grenman, Rainer Lehtonen, Johanna Hynninen, Sakari Hietanen. Mathematical model quantifies the effect of novel combination therapies in high-grade serous ovarian cancer. [abstract]. In: Proceedings of the AACR Conference: Addressing Critical Questions in Ovarian Cancer Research and Treatment; Oct 1-4, 2017; Pittsburgh, PA. Philadelphia (PA): AACR; Clin Cancer Res 2018;24(15_Suppl):Abstract nr B70.