Doxorubicin-loaded silk films: drug-silk interactions and in vivo performance in human orthotopic breast cancer

Abstract

Breast cancer is the most common of all malignant diseases in women. Systemic chemotherapy provides low clinical benefit for locoregional control of the disease, while localised chemotherapy may provide a therapeutic advantage. In this study, doxorubicin-loaded silk films were directly applied to tumours. Affinity binding studies demonstrated that the adsorption of doxorubicin onto silk was partially dependent on crystallinity. By manipulating silk crystallinity, or β-sheet content, the doxorubicin release rate could be controlled ranging from immediate release to prolonged release over >4 weeks. The therapeutic impact of doxorubicin-loaded silk films on primary tumour growth and metastasis was assessed in mice using a humanised orthotopic breast cancer model (adenocarcinoma). Both soluble and stabilised silk films loaded with doxorubicin had a significantly greater primary tumour response than the equivalent dose of doxorubicin administered intravenously in the absence of the silk film carrier. In addition to reducing primary tumour growth, stabilised silk films loaded with doxorubicin also reduced metastatic spread and autopsy indicated that these films were not associated with any local or systemic toxicities. Collectively, these results suggest that the future use of this approach for localised chemotherapy is promising.

Figure 1

Preparation of doxorubicin-loaded silk films. (A) Strategies to generate doxorubicin-loaded films that are water-soluble or (B) water-insoluble with variable β-sheet content. (C) Film thickness of dry and hydrated silk films (± SD; n = 8). (D) Free-standing doxorubicin-loaded silk films in the dry state (scale bar 7 mm). (E) Free-standing hydrated films following surface contours (scale bar 7 mm). (F) Silk films can be loaded with a range of doxorubicin concentrations.

Figure 2

Doxorubicin interactions and film stability. (A) Typical adsorption kinetics of doxorubicin to silk monitored by following doxorubicin depletion from solution using UV-Vis spectroscopy. (B) Equilibrium binding studies of doxorubicin to silk films. Graph depicts data for 4°C water vapour annealed silk films. (C) Cumulative release of doxorubicin into PBS. Significant differences between 4°C water vapour-annealed silk films and other groups was determined by ANOVA followed by Dunnett’s multiple comparison post hoc test, *P < 0.05; ±SD; n = 4. (D) In vitro degradation of silk films with α-chymotrypsin. Significant differences between 121°C water vapour annealed silk films and other groups was determined by ANOVA followed by Dunnett’s multiple comparison post hoc test, ***P < 0.0001; ± SD; n = 4).

Figure 3

In vitro toxicity of doxorubicin-loaded silk films towards human breast cancer cells. (A) MDA-MB-231 cell viability following a 72-h exposure to soluble or stabilised silk films ± doxorubicin and freely diffusible doxorubicin at the equivalent dose. Significant differences between control cells under optimum growth conditions (plasma-treated polystyrene, PTP) and other groups were determined by ANOVA followed by Dunnett’s multiple comparison post hoc test, ns = not significant, ***P < 0.0001; ± SD; n = 3). (B) Long-term toxicity of film preparations and diffusible doxorubicin. Culture medium was replaced at the indicated time point. With the exception of control wells or wells containing stabilised silk, wells were reseeded with MDA-MB-231 cells at day 6. Significant differences between diffusible doxorubicin and the other doxorubicin treatment groups was determined by ANOVA followed by Dunnett’s multiple comparison post hoc test, ns = not significant, ***P < 0.0001; ± SD; n = 4).

Figure 4

In vivo response of orthotopic breast tumours to doxorubicin-loaded silk films. (A) Cancer cell-associated bioluminescence at week 3, (B) 5 and (C) 6 weeks. (D) Weight of primary tumours at the end of the study (week 6). (E) In vivo tumour cell-specific bioluminescence of representative mice from each treatment group at week 6. Plot symbols are defined in panel (C). (F) Metastatic spread of cancer cells to organs at week 6. Plot symbols are defined in panel (C). Significant differences between control animals and the other treatment groups was determined by ANOVA followed by Dunnett’s multiple comparison post hoc test, ns = not significant, **P < 0.001; ± SEM; n = 5 except for stabilised silk films where n = 4).

Figure 5

In vivo toxicity following systemic and local doxorubicin application. (A) Image depicting tumour and stabilised silk film at the time of necroscopy (week 6, scale bar 5 mm). (B) Heart weights of animals treated with localised or systemic doxorubicin and the respective controls. Significant differences between control animals and the other treatment groups was determined by ANOVA followed by Dunnett’s multiple comparison post hoc test, ns=not significant, **P < 0.001, ***P < 0.0001; ± SEM; n = 5 except for stabilised silk films where n = 4. (C) Histological section and H&E staining of the abdominal wall exposed to stabilised or soluble silk film ± doxorubicin. (Scale bar 200 μm).