Evaluating the efficacy of Rose Bengal-PVA combinations within PCL/PLA implants for sustained cancer treatment

Abstract

The current investigation aims to address the limitations of conventional cancer therapy by developing an advanced, long-term drug delivery system using biocompatible Rose Bengal (RB)-loaded polyvinyl alcohol (PVA) matrices incorporated into 3D printed polycaprolactone (PCL) and polylactic acid (PLA) implants. The anticancer drug RB's high solubility and low lipophilicity require frequent and painful administration to the tumour site, limiting its clinical application. In this study, RB was encapsulated in a PVA (RB@PVA) matrix to overcome these challenges and achieve a localised and sustained drug release system within a biodegradable implant designed to be implanted near the tumour site. The RB@PVA matrix demonstrated an RB loading efficiency of 77.34 ± 1.53%, with complete RB release within 30 min. However, when integrated into implants, the system provided a sustained RB release of 75.84 ± 8.75% over 90 days. Cytotoxicity assays on PC-3 prostate cancer cells indicated an IC50 value of 1.19 µM for RB@PVA compared to 2.49 µM for free RB, effectively inhibiting cancer cell proliferation. This innovative drug delivery system, which incorporates a polymer matrix within an implantable device, represents a significant advancement in the sustained release of hydrosoluble drugs. It holds promise for reducing the frequency of drug administration, thereby improving patient compliance and translating experimental research into practical therapeutic applications.

Keywords: 3D printing; Cancer; Implant; Polyvinyl alcohol; Rose Bengal; Sustained drug release.

Fig. 1

Illustration of the RB-PVA (RB@PVA) matrix preparation using the casting-solvent evaporation technique and its inclusion into the 3D-printed implant. RB and PVA powders were dissolved in deionised water, and the viscous solution was cast into a glass petri dish. The dried matrix was removed and milled. The milled RB@PVA matrix was included in the 3D-printed implant. For implant fabrication details, refer to Korelidou et al., 2022 [37]. Symbols: RB- Rose Bengal, PVA- Polyvinyl Alcohol

Fig. 2

Long-term stability of Rose Bengal (RB) in buffer solution (pH 7.4) assessed by absorbance (Abs) at 549 nm (A) and converted in mg (B) over time. Results are expressed as mean ± SD of three samples, each analysed in triplicate (n = 9). Statistical significance was determined using Dunnett multiple comparison tests with a single pooled variance, comparing each point with the baseline (week 0). Significance is indicated as ****p < 0.0001

Fig. 3

Morphology and size distribution of the RB-loaded PVA (RB@PVA) matrix after milling. (A) RB powder under optical microscopy (scale bar 0.5 mm, magnification 25x). (B) PVA powder under optical microscopy (scale bar 1 mm, magnification 16x). (C) Milled RB@PVA matrix under optical microscopy (scale bar 0.5 mm, magnification 25x). (D1-3) Scanning electron microscopy (SEM) images of milled RB@PVA matrix (D1: scale bar: 2 mm, magnification: 50x; D2: scale bar: 1 mm, magnification: 100x; D3: scale bar: 300 μm, magnification: 250x). (E) Size measurement of 16 RB@PVA particles

Fig. 4

Physicochemical and mechanical characterisation of RB-loaded PVA (RB@PVA) matrix. (A) FTIR spectra of RB powder, PVA powder, RB/PVA physical mixture, and RB@PVA matrix showing characteristic peaks (transmittance over wavenumber range (4000–600 cm^− 1)). (B) DSC thermograms of RB powder, PVA powder, RB/PVA physical mixture, and RB@PVA matrix (heat flow over temperature (°C)). (C) Elastic modulus. (D) Offset yield strength. (E) Ultimate tensile strength of pure PVA matrix and RB@PVA matrix. C-E: Results are expressed as mean ± SD of four samples (n = 4). Statistical analysis was performed using an unpaired t-test (two-tailed). Significance is indicated as *p < 0.05, **p < 0.01

Fig. 5

In vitro dissolution study of the milled Rose Bengal-loaded PVA (RB@PVA) matrix in buffer medium (pH 7.4, 37 °C). (A) Percentage RB released over time (min) from RB@PVA matrix compared to RB/PVA physical mixture. (B) Percentage of RB released relative to matrix weight loss. Corresponding time (min) is indicated above each symbol. (C) Correlation between RB release and matrix weight loss. Corresponding time (min) is indicated above each symbol. (D) Visual representation of residual RB@PVA particles on the filter at various time points (initial amount: 10 mg). Results are expressed as mean ± SD of four samples (n = 4)

Fig. 6

Rose Bengal-PVA (RB@PVA) matrix-loaded implant before in vitro release study. A1-2. Empty implant without PCL sealing. B1-2. Sealed RB@PVA matrix-loaded implant before the release study, showing PCL sealing (B1) and the permeable membrane (B2). C1-2 SEM micrographs of the permeable membrane before the release study (D1: scale bar: 0.2 mm, magnification: 500x; D2, D2: scale bar: 0.1 mm, magnification: 1.0kx)

Fig. 7

In vitro release profile of Rose Bengal (RB) from RB@PVA matrix-loaded implants compared to RB/PVA physical mixture (RB/PVA PM) implants in buffer solution (pH 7.4, 37 °C). (A) Amount (mg) of RB released over time. (B) Percentage of RB released over time. (C-D) Detailed view of RB release during the first 30 days. Results are expressed as mean ± SD of four samples (n = 4). Statistical analysis was performed using two-way ANOVA with the Geisser-Greenhouse correction and Fisher’s LSD test for multiple comparisons. Significance is indicated as *p < 0.05

Fig. 8

Rose Bengal-PVA (RB@PVA) matrix-loaded implant after a 7-month in vitro release study. (A1) residual RB@PVA matrix in the implant; (A2) cylinder structure after washing; (A3) removed PCL sealing; (A4) removed permeable membrane. (B1-2) SEM micrographs of the permeable membrane after the release study (B1: scale bar: 0.2 mm, magnification: 500x; B2: scale bar: 0.1 mm, magnification: 1.0kx)

Fig. 9

Cytotoxicity assay of free RB and RB@PVA matrix on PC-3 prostate cancer cells using clonogenic assays. (A) Survival efficiency of PC-3 cells treated with PVA at 30 µM. (B) Survival efficiency of PC-3 cells treated with different concentrations of free RB and RB@PVA matrix. Results are expressed as mean ± SD of three independent samples (n = 3). Statistical analysis was performed using a two-way ANOVA, with a highly significant (p < 0.0001) increase in direct cytotoxicity observed between RB@PVA matrix over free RB (n = 3)