Développement d'un implant solide biodégradable à base d'amidon réticulé à teneur élevée en amylose, pour la libération contrôlée d'un principe actif

Cross-linked high amylose starch (Contramid^®) is a hydrophilic swelling matrix that was originally developed for controlled release solid oral dosage forms. The purpose of this doctorate work was to evaluate Contramid^® as a carrier of drug delivery implant.

Initially, Contramid^® was investigated for evaluation of host response in mice and of hexenoyl-trans-3-hGRF(1–44)NH₂ (Hex-hGRF) delivery in pigs. Mice were administered subcutaneously one placebo Contramid^® implant and host reaction was evaluated over six months. Pigs were administered 15 mg Hex-hGRF: (1) one pure Hex-hGRF implant, (2) four 30/70 w/w Hex-hGRF/Contramid^® implants or (3) eight 15/85 w/w Hex-hGRF/Contramid^® implants. A fourth group was injected twice daily with 10 μg/kg of Hex-hGRF over five days. Serum insulin-like growth factor-I (IGF-I) was monitored over one month. In mice, macroscopic and microscopic inflammatory reactions were always localized. Polymorphonuclear cells and macrophages predominated within and around implants, respectively. Thin fibrovascular septa eventually subdivided Contramid^® implants, which were progressively phagocytosed by macrophages. In pigs, serum IGF-I concentrations were increased over a ten day period in all implanted groups. The initial IGF-I peak observed in the daily injected group was avoided in both Contramid^® implant groups but not in the pure Hex-hGRF implant group.

The Contramid^® biocompatibility and degradation characteristics were further investigated over four months with subcutaneous and intramuscular implants in rats. Macroscopic observations of implantation sites were performed over time and tissue samples were removed for histologic examination. No macroscopic inflammatory reaction was observed. Microscopically, inflammatory reaction was moderate and restricted to implantation sites. Degradation of Contramid ^® implants was characterized with fragmentation by fibrovascular septa and phagocytosis by macrophages. Contramid^® was mostly absorbed by the end of the four-month period.

Then, Contramid^® was assessed for in vitro ciprofloxacin (CFX) delivery. Twelve formulations were prepared: control Contramid^®; Contramid^® with 1% hydrogenated vegetable oil (HVO); and Contramid^® with 10 or 20% hydroxypropylmethylcellulose (HPMC), each of them with three CFX loadings (2.5, 5.0 and 7.5%). All implants were used for 24-hour dissolution tests to evaluate swelling, erosion, water uptake and CFX release. Additionally, 1%-HVO Contramid^® implants were used for an extended dissolution test. The presence of HPMC increased CFX release rate, swelling, erosion and water uptake in a concentration-dependent manner whereas HVO had no effect. With increasing drug loading, a decrease of cumulative CFX percent release was observed in both 24-hour and extended dissolution tests. Of the different formulations tested, 7.5%-CFX/1%-HVO implants provided the longest period of drug delivery (>three weeks) without any initial burst effect.

At last, various CFX loadings of Contramid^® implants were tested in vivo. Rabbits were administered six 2.5, two 7.5, one 15.0 or one 20.0%-CFX implants along the femur to determine systemic versus local CFX concentrations over one month. After implantation, blood, muscle, femur and residual polymer were collected after over time for CFX assay. Serum CFX concentrations were low regardless of implant loading. Increased drug loading resulted in a higher and longer release of CFX with local concentrations largely in excess of the minimal inhibition concentration over 28 days for 20.0%-CFX implants.

Therefore, it has been demonstrated that Contramid^® is a biocompatible and absorbable material, which can be used as a sustained antimicrobial delivery implant for local prevention and/or treatment of osteomyelitis.