Vaginal delivery of paclitaxel via nanoparticles with non-mucoadhesive surfaces suppresses cervical tumor growth

Abstract

Local delivery of chemotherapeutics in the cervicovaginal tract using nanoparticles may reduce adverse side effects associated with systemic chemotherapy, while improving outcomes for early-stage cervical cancer. It is hypothesized here that drug-loaded nanoparticles that rapidly penetrate cervicovaginal mucus (CVM) lining the female reproductive tract will more effectively deliver their payload to underlying diseased tissues in a uniform and sustained manner compared with nanoparticles that do not efficiently penetrate CVM. Paclitaxel-loaded nanoparticles are developed, composed entirely of polymers used in FDA-approved products, which rapidly penetrate human CVM and provide sustained drug release with minimal burst effect. A mouse model is further employed with aggressive cervical tumors established in the cervicovaginal tract to compare paclitaxel-loaded poly(lactic-co-glycolic acid) (PLGA) nanoparticles (conventional particles, or CP) and similar particles coated with Pluronic F127 (mucus-penetrating particles, or MPP). CP are mucoadhesive and, thus, aggregated in mucus, while MPP achieve more uniform distribution and close proximity to cervical tumors. Paclitaxel-MPP suppress tumor growth more effectively and prolong median survival of mice compared with unencapsulated paclitaxel or paclitaxel-CP. Histopathological studies demonstrate minimal toxicity to the cervicovaginal epithelia, suggesting paclitaxel-MPP may be safe for intravaginal use. These results demonstrate the in vivo advantages of polymer-based MPP for treatment of tumors localized to a mucosal surface.

Keywords: biodegradable polymers; cancer; chemotherapy; controlled release; drug delivery.

Figure 1

Characterization of PTX/PLGA nanoparticles in vitro. Scanning electron micrographs of PTX/MPP (a) and PTX/CP (b); scale bar represents 1 µm. (c) Cumulative in vitro release of PTX from PTX/PLGA nanoparticles over time. Data represent the average values from two independent studies using different particle preparations. Error bars represent S.E.M.

Figure 2

Transport of PTX-loaded PLGA particles in fresh human cervicovaginal mucus. Representative trajectories for (a) PTX/MPP and (b) PTX/CP, with effective diffusivities within one S.E.M. of the mean. (c) Ensemble-averaged geometric mean square displacements (<MSD>) as a function of time scale. Error bars represent S.E.M. * indicates statistically significant difference across all time scales (p < 0.05). (d) Distributions of the logarithms of individual particle effective diffusivities (D_eff) at a time scale of 1 s. Data represent the average of at least three independent experiments, with n ≥ 100 per experiment for both PTX/MPP and PTX/CP.

Figure 3

Magnetic resonance imaging (MRI) and bioluminescence imaging of TC-1 cervical tumors implanted in the mouse cervicovaginal tract. Representative coronal (a), sagittal (b), and axial (c) anatomical MR images of the female mouse reproductive tract, with yellow arrows indicating the location of the TC-1 tumor and the green arrow indicating the location of CV tract. The size of the tumor shown was measured to be 4.3 mm × 1.5 mm × 4.7 mm in the coronal, sagittal and axial directions, respectively. (d) Correlation of total bioluminescence and TC-1 tumor weight. Bioluminescence signals were quantified as the total photon flux (p/s, photons per second) of the tumor regions. The top row of images shows the tumor location and relative biolumniscence intensity, and the bottom row shows the physical appearance of the excised tumors. Error bars represent S.E.M. Scale bars represent 5 mm.

Figure 4

Transverse 50 µm thick frozen sections of mouse vaginal tissue containing red fluorescent (a) PLGA CP or (b) PLGA MPP. Green corresponds to TC-1 cervical tumor expressing GFP. The epithelium is outlined in white to help distinguish the tissue surface. Arrows indicate the presence of large particle aggregates in the CV tract with PLGA CP in (a), while arrowheads indicate examples of well-dispersed PLGA MPP in (b). The same image acquisition and contrast settings were used for both images.

Figure 5

Amount of PTX retained in the mouse CV tract over time, as delivered by PTX/MPP or as free drug by vaginal administration (local Taxol^®) or by intraperitoneal administration (systemic Taxol^®). The fraction of drug retained in the vagina is expressed as a percentage of the total amount of drug administered. * indicates statistically significant difference compared to free drug at the same time point (p < 0.05). Data represent at least n = 3 mice per condition and time point. Error bars represent S.E.M.

Figure 6

Response of TC-1 tumors established in the mouse cervicovaginal tract to local treatments with untreated control, free Taxol^®, PTX/CP or PTX/MPP (n = 5–6 mice per treatment group). Tumors were inoculated on Day 0; treatments were given daily starting on Day 3. (a) Change in bioluminescence signal over the indicated time course. Signals were quantified as the total photon flux (p/s, photons per second) of the tumor regions. Error bars represent S.E.M. * indicates significantly improved tumor suppression (p < 0.05) compared to other treatments starting from Day 7. (b) Kaplan-Meier survival curves for mice bearing cervicovaginal TC-1 tumors. * indicates significantly improved survival rate (p < 0.05) compared to other treatments starting from Day 13.

Figure 7

Histological sections of cervicovaginal tissue from mice treated with systemic Taxol^®, local Taxol^®, PTX/CP or PTX/MPP at 24 hr and 48 hr post administration (n = 5 mice per treatment group). Note the presence of pyknotic and apoptotic cells (arrows, right side inset) in local treatment groups at 24 hr and 48 hr, and inflammatory cells (arrowheads, left side inset) in PTX/CP and PTX/MPP treated mice at 24 hr.