Epithelial ovarian cancer (EOC), particularly in its advanced and recurrent stages, presents a significant challenge due to its high recurrence rates and dissemination throughout the peritoneum. Despite achieving greater than 75% disease remission with first-line treatments, the presence of significant occult micrometastases contributes to poor outcomes. Immunotherapies, while showing promise in other cancers, have proven largely ineffective against EOC in clinical trials. The reasons for this failure include the characteristically immunosuppressive tumor microenvironment in EOC and the cytotoxic effects of high-dose chemotherapy. Photodynamic therapy (PDT) and its targeted variant, photoimmunotherapy (PIT), offer potential alternatives for treating disseminated disease due to their safe intraperitoneal delivery. PDT involves the activation of a photosensitizer by light, leading to the production of cytotoxic reactive oxygen species. PIT leverages cell-activatable antibody-photosensitizer conjugates for targeted delivery to tumor cells, enhancing selectivity and minimizing damage to normal tissues, including immune cells. Previous research has indicated that low-dose, non-ablative PDT can stimulate an anti-tumor immune response, whereas ablative doses do not. However, the optimal treatment regimens for PDT and PIT, considering factors like photosensitizer dose, irradiance, and total light dose, remain to be determined. The complex, non-linear interactions within the tumor-immune system necessitate mathematical modeling to investigate the various treatment combinations and explore the resulting dynamics. This study aimed to integrate in vitro experimental data on EOC tumor cell and T cell responses to chemotherapy and phototherapies (PDT and PIT) into a mathematical model to evaluate the potential of fractionated PIT to control tumors via the stimulation of an anti-tumor immune response. This approach was deemed crucial in order to overcome the limitations of exploring many dosing and sequencing schemes.

The authors reviewed the literature on advanced EOC treatment, highlighting the high recurrence rates and challenges in achieving long-term remission despite effective first-line therapies. The ineffectiveness of immunotherapies in EOC clinical trials was discussed, attributing this to the immunosuppressive tumor microenvironment and the cytotoxic effects of high-dose chemotherapy. Photodynamic therapy (PDT) and its targeted version, photoimmunotherapy (PIT), were presented as promising alternatives for treating disseminated disease due to their safe intraperitoneal delivery and potential for immune modulation. Studies on the dose-dependent effects of PDT and PIT on the immune system were reviewed, revealing the potential for low-dose regimens to stimulate anti-tumor immunity. The authors then discussed the limitations of experimental approaches for exploring diverse treatment strategies and emphasized the importance of mathematical oncology for modeling complex tumor-immune dynamics and predicting treatment efficacy.

The study used two human EOC cell lines, Ovcar3 and Ovcar5 (and their fluorescent protein-expressing derivatives), and mouse T cells (OT-1) or human Jurkat T cells for in vitro experiments. 3D co-cultures of EOC and T cells were treated with chemotherapy (cisplatin), PDT (using unconjugated benzoporphyrin derivative, BPD), and PIT (using cetuximab-conjugated BPD). Different doses of each treatment were administered, followed by a 3-day incubation period. Cell viability was assessed using high-content imaging and a custom MATLAB code. Dose-response curves were fitted to the data using least-squares regression. These in vitro data were then incorporated into a mathematical model of tumor-immune dynamics based on an ordinary differential equation (ODE) system describing the interactions between tumor cells (C) and immune effector cells (E). The model incorporated parameters representing logistic tumor growth, immune cell influx and recruitment, immune cell exhaustion and clearance, and tumor cell death due to C-E interactions. The model parameters were primarily adopted from previous literature. The ODE system was solved numerically in MATLAB using the ode45 solver. Dose-response curves were mapped onto the C-E phase plane to visualize treatment effects on tumor-immune dynamics. Simulations of fractionated PIT were performed by dividing a total light dose into multiple smaller fractions, simulating the effect of each fraction on the (C, E) values via application of the derived survival fractions. The minimum number of PIT fractions needed for tumor control was calculated for a range of dose/fraction values and initial conditions.

In vitro experiments revealed distinct dose-response patterns for chemotherapy, PDT, and PIT. Chemotherapy primarily reduced T cell survival at all doses. PDT showed comparable reductions in tumor and T cell survival at higher doses, while at lower doses, T-cell survival was comparable to tumour cell survival. PIT demonstrated two distinct regimes: at low light doses (≤10 J/cm²), there was no effect on tumor cells but a significant increase in T cell numbers; at high light doses (>10 J/cm²), both tumor and T cells experienced cytotoxicity, but T cell survival was consistently higher than tumor cell survival. Mapping treatment responses onto the C-E phase plane showed that, for the chosen initial conditions, all treatments resulted in tumor escape. However, simulations of fractionated PIT demonstrated that delivering a large total light dose in multiple smaller fractions could shift the tumor-immune dynamics into a regime of immune-mediated tumor control, even though a single large dose failed to do so. The minimum number of PIT fractions required for tumor control increased with increasing fraction size.

The findings support the potential of fractionated PIT as a novel therapeutic strategy for EOC. The immunostimulatory effect of low-dose PIT, as evidenced by increased T cell numbers, is a key mechanism underpinning the success of the fractionated regimen. The study highlights the importance of considering the complex interplay between tumor and immune cells when designing cancer therapies. The model suggests that, while large doses of PIT might not be successful alone, the careful application of fractionated PIT may be able to achieve tumor control through immune system stimulation. This mechanism differs from the previously described uses of fractionated PDT, which primarily focused on improving efficacy by enabling re-oxygenation of the tissue. The study emphasizes the importance of further investigation to refine model parameters for EOC specific tumor-immune interactions, and includes the dynamics of different immune cell types.

This study demonstrates the potential of fractionated PIT to stimulate an anti-tumor immune response and achieve tumor control in EOC. The integrated mathematical and in vitro approach provides valuable insights into the optimal design of PIT regimens. Future work should focus on validating the model predictions in vivo and refining the model to account for additional immune cell types, as well as additional biomarker targets for more precise and effective targeting of the tumor.

The study's limitations include the use of a non-syngeneic in vitro model, which does not fully capture the complexity of the tumor-immune microenvironment in human patients. The mathematical model, while incorporating key dynamics, is a simplification of the intricate interactions within the system. The model's parameters were primarily based on previous literature and may not be perfectly calibrated for EOC. The absence of immune suppressor cells in the model represents a simplification which could be important in determining the final outcome.