Introduction

Thermoresponsive carriers that release their load upon temperature stimuli are intriguing smart drug delivery systems^1-2. For an effective and safe thermal drug delivery, an accurate spatial guidance and temperature control is essential. Utilizing RF transmission for MRI, controlled RF heating and MR temperature monitoring in an integrated applicator^3-4 is conceptually appealing for advancing the capabilities of thermal targeted delivery of drugs, MR sensitive probes or other loads. Here we demonstrate the applicability of an RF heating system integrated into a 7.0T whole body MR scanner for drug release using a pH- and thermoresponsive polymer carrier loaded with fluorescent model drug.

Materials and methods

Based on electromagnetic field simulations⁵, an agarose phantom (dimensions=(18x28x9)cm³; σ=1.12±0.03S/m, ε=84.4±1.4) was constructed with seven sample holders and 15 PTFE tubes for fiber optic temperature sensors (FOTS) (Fig.1a-d). The phantom was positioned in a water bath (28°C) thermally regulated by a bath circulator⁶. An extraction system enables sample extraction during experiments without repositioning the patient table (Fig.1b). For RF heating, a bow tie dipole antenna³ (Fig.1e) was located above the central sample holder (Fig.1a) which was filled with a pH7.4 polymer⁷ solution. A control sample of the same solution was positioned outside of the RF heated area (Fig.1a). A custom build transmit/receive switch⁸ was used to support high peak power (MR imaging) and high average power (RF heating). Temperature maps were calculated based on the proton resonance frequency shift⁹ and a dual gradient-echo technique (FOV=(300×300)mm2, TR=61ms, TE1/TE2=6.34ms/11.44ms, spatial resolution=(1.5×1.5×4.0)mm3)^10-11. Three FOTS¹² were used as external references. The thermoresponsive polymer P(DEGMA-co-OEGMA-b-[TMSPMA-co-VBA])⁷ was used (Fig.2a). The lower critical solution temperature (LCST) was determined by turbidity measurents at 520nm¹³. Fluoresceinamine was loaded to the polymers as a drug model by formation of Schiff base bonds in the presence of triethylamine catalyst⁷. Prior to experiment in the MR scanner, the temperature induced release of fluoresceinamine was determined to be later used as reference. The polymer solutions, prepared at pH5.5 and pH7.4, were held at 25°C or heated at 40°C in a water bath for 60min. During this time small samples were extracted from the polymer solutions and passed through a size exclusion chromatography column¹⁴ where the released fluoresceinamine (347Da) is retained while the polymers (15kDa) with the attached fluoresceinamine were collected and analyzed in a fluorescence reader¹⁵. ¹H-NMR spectroscopy was performed at 300MHz¹⁶. RF heating experiments were performed in a B₀=7.0T MR system¹⁷. The heating paradigm consisted of 6x5min RF heating (P_avg=100W) interleaved with MR thermometry. After reaching 40°C in the heated polymer solution, the temperature was maintained for 120min with subsequent 2min RF heating periods. During this time, small samples of the polymer solutions were extracted, passed through size exclusion chromatography columns and analyzed in a fluorescence reader¹⁵.

Results

Turbidity measurement showed a LCST=34°C (Fig.2b). Water bath heating experiments showed little or no release of fluoresceinamine from polymer solutions held at T<LCST, except the initial 40% release from pH5.5 solution due to Schiff base bond breakage in an acidic environment (Fig.2c). Solutions at T>LCST showed almost complete release over 60min with a faster release for pH5.5 versus pH7.4 solutions. The aldehyde peak at 10ppm of ¹H-NMR spectrum of the polymer (Fig.2d) is reduced when Schiff base bond is formed between the polymer and fluoresceinamine, which potentially allows the determination of drug release using ¹H-MRS. Temperature in the heated polymer solution increased from 28°C to 40°C after 30min of RF heating, while the control sample maintained at 28°C (Fig.3). MRTh measurements were consistent with FOTS readings, with 0.2-0.9°C of deviation (Fig.3c). The release of fluoresceinamine from the polymers was verified in the heated solution. The fluoresceinamine concentration curve used to quantify these results has an R²=0.998 for linear fit indicating good fidelity (Fig.4a). After 30min at 40°C, 50% of loaded fluoresceinamine was detached from the polymers (Fig.4b). After 120min, the release was 89% (Fig.4b). The control solution did not show apparent fluoresceinamine release (Fig.4b).

Conclusion

This work demonstrated the applicability of an RF heating system integrated in a 7.0T MR scanner for temperature induced release of a model drug attached to a thermoresponsive polymer. Our approach adds a thermal intervention dimension to an MR imaging device and provides an ideal testbed for the study of temperature induced release of drugs, MR probes and other agents from thermoresponsive polymers, nanogels and other carriers.