Folate Receptor Targeted Photodynamic Therapy: A Novel Way to Stimulate Anti-Tumor Immune Response in Intraperitoneal Ovarian Cancer

Abstract

Photodynamic therapy (PDT) has shown improvements in cancer treatment and in the induction of a proper anti-tumor immune response. However, current photosensitizers (PS) lack tumor specificity, resulting in reduced efficacy and side effects in patients with intraperitoneal ovarian cancer (OC). In order to target peritoneal metastases of OC, which overexpress folate receptor (FRα) in 80% of cases, we proposed a targeted PDT using a PS coupled with folic acid. Herein, we applied this targeted PDT in an in vivo mouse model of peritoneal ovarian carcinomatosis. The efficacy of the treatment was evaluated in mice without and with human peripheral blood mononuclear cell (PBMC) reconstitution. When mice were reconstituted, using a fractionized PDT protocol led to a significantly higher decrease in the tumor growth than that obtained in the non-reconstituted mice (p = 0.0469). Simultaneously, an immune response was reflected by an increase in NK cells, and both CD4+ and CD8+ T cells were activated. A promotion in cytokines IFNγ and TNFα and an inhibition in cytokines TGFβ, IL-8, and IL-10 was also noticed. Our work showed that a fractionized FRα-targeted PDT protocol is effective for the treatment of OC and goes beyond local induction of tumor cell death, with the promotion of a subsequent anti-tumor response.

Keywords: folate coupled photosensitizer; ovarian cancer; photodynamic therapy.

Figure 1

In vitro sensibility of OVCAR3-Luc cell line to PS-PDT over time: normalized viability for OVCAR3-Luc cell line at 1 h, 24 h, 48 h, and 72 h post-illumination. Cells are treated with 9 µM of PS upon 60 min of illumination (1 mW/cm², 3.7 J/cm²). NT: non-treated, PS: photosensitizer only, Illu: illumination only, PDT: illumination in the presence of PS. Results are presented as means of three independent experiments expressed in % of viability and normalized compared to NT condition. Two-way ANOVA statistical test was performed, all quoted p-values are two-sided, with p ≤ 0.01 (**), and p ≤ 0.0001 (****) being considered statistically significant for the first and highly significant for the others. n = 3.

Figure 2

Development of the in vivo humanized SCID model of intraperitoneal ovarian cancer: (a) evaluation of the tumor growth by bioluminescence after injection of 100 µL of D-luciferin over an IVIS LUMINA XR reader 38 days after tumor cell injection; (b) OVCAR3-Luc cells injected into SCID mice resulted in tumor growth 38 days after cell injection.

Figure 3

PS uptake: (a) evaluation of the uptake of the PS within ovaries and tumor in an in vivo humanized SCID mice model of intraperitoneal ovarian cancer (scale bar = 200 µm); (b) semi fluorescence quantification of the signal by ImageJ software (version 2.9.0) (NC: negative control). Results are expressed in fluorescence intensity 4 h, 6 h, and 24 h after PS injection and presented as means of three independent experiments. Two-way ANOVA statistical test was performed, all quoted p-values are two-sided, with p ≤ 0.001 (***) and p ≤ 0.0001 (****) being considered statistically highly significant. n = 3.

Figure 4

Determination of the most effective fractionized illumination protocol for PDT in humanized non-reconstituted SCID mice model of intraperitoneal ovarian cancer: mice were subjected to two different fractionized illumination protocols: 1 min illumination followed by 2 min of pause (1–2) vs. 15 min illumination followed by 10 min of pause (15–10), both delivering 27 J/cm² red light at 10 mW/cm². Comparison with NT: non treated group was performed. Arrow: day of illumination. Results are presented as means of two independent experiments per protocol. n = 2.

Figure 5

Tumor progression upon PDT for non-reconstituted SCID mouse model of intraperitoneal ovarian cancer: mice were subjected to a fractionized illumination protocol: 1 min illumination followed by 2 min of pause (10 mW/cm², 27 J/cm²) with NT: non treated, PS: photosensitizer only, Illu: illumination only, PDT: illumination in the presence of PS, arrow: day of illumination. PDT efficacy was examined over 56 days and gauged by bioluminescence quantification. Results are presented as means of three independent experiments and expressed in log(10) of bioluminescence. Two-way ANOVA statistical test was performed, all quoted p-values are two-sided, with p ≤ 0.001 (***) being considered statistically significant. n = 3.

Figure 6

Evaluation of the cytokine release upon PDT in humanized non-reconstituted SCID mice model of intraperitoneal ovarian cancer: ELISA were performed on mice serum at 72 h after PDT (a) transforming growth factor (TGF) ß; (b) interferon (IFN) γ with NT: non treated, PS: photosensitizer only, Illu: illumination only, PDT: illumination in the presence of PS. Results are presented as means of three independent experiments expressed in % of the NT. One-way ANOVA statistical test was performed, all quoted p-values are two-sided, with p ≤ 0.001 (**) being considered statistically significant. n = 3.

Figure 7

Evaluation of the PDT effect in PBMC reconstituted or not reconstituted humanized SCID mouse model of human intraperitoneal ovarian cancer: (a) bioluminescence in mice subjected to a fractionized illumination protocol: 1 min illumination followed by 2 min of pause repeated 45 times (10 mW/cm², 27 J/cm²) with R-NT: reconstituted non treated, R-PS: reconstituted subjected to photosensitizer only, R-Illu: reconstituted subjected to illumination only, R-PDT: reconstituted subjected to illumination in the presence of PS, NRNT: non-reconstituted non treated, and arrow: day of illumination. Results are presented as means of three independent experiments. Two-way ANOVA statistical test was performed, all quoted p-values are two-sided, with p ≤ 0.0001 (****) being considered highly statistically significant. n = 3; (b) the overall tumor progression throughout the experiment was evaluated with R-PDT: reconstituted subjected to illumination in the presence of PS and NR-PDT: non-reconstituted subjected to illumination in the presence of PS. Results are presented as means of three independent experiments. Mann–Whitney U test was used to compare the two conditions with p ≤ 0.01 (*) being considered statistically significant.

Figure 8

Evaluation of the immune response upon PDT in PBMC-reconstituted humanized SCID mouse model of peritoneal ovarian cancer: blood from not reconstituted (NR NT) or reconstituted mice subjected to illumination (R-PDT, R-Illu) or not (R-PS, R-NT) was examined and immune population was investigated 7 days (D7) and 30 days (D30) after PDT: (a) B cells (LB); (b) CD4+ T cells; (c) CD8+ T cells; (d) natural killers cells (NK); (e) myeloid cells; (f) monocytes with NRNT: non-reconstituted not treated, R-NT: reconstituted non treated, R-PS: reconstituted subjected to photosensitizer only, R-Illu: reconstituted subjected to illumination only, R-PDT: reconstituted subjected to illumination in the presence of photosensitizer. Two-way ANOVA statistical test was performed and comparisons with the R-NT conditions were done (at D7 and D30 separately), all quoted p-values are two-sided, with p ≤ 0.01 (*), p ≤ 0.01 (**), p ≤ 0.001 (***), and p ≤ 0.01 (****) being considered statistically significant for the first and (****) highly significant for the others.

Figure 9

Evaluation of the cytokine release upon PDT in PBMC-reconstituted humanized SCID mouse model of peritoneal ovarian cancer: blood from not reconstituted (NR NT) or reconstituted mice subjected to illumination (R-PDT, R-Illu) or not (R-PS, R-NT) was examined for cytokine release 7 days (D7) and 30 days (D30) after PDT: (a) transforming growth factor ß (TGFß); (b) interleukine 8 (IL-8); (c) interleukin 10 (IL-10); (d) interferon γ (IFNγ); (e): tumor necrosis factor α (TNFα) with NRNT: non–reconstituted non treated, R-NT: reconstituted non treated, R-PS: reconstituted subjected to photosensitizer only, R-illu: reconstituted subjected to illumination only, R-PDT: reconstituted subjected to illumination in the presence of photosensitizer. Results are presented as normalized values compared to NRNT. Two-way ANOVA statistical test was performed, all quoted p-values are two-sided, with p ≤ 0.001 (**) being considered statistically highly significant.

Figure 10

Structure of the Pyro-PEG-FA: folic acid conjugated to pyropheophorbide-a via a polyethylene glycol type spacer.