Quantifying diffusivity of ISFIs in vivo using DWI: A novel method to evaluate drug release
Nicole Vike1, Xin Li2, Kelsey Hopkins2, Luis Solorio2, and Joseph Rispoli2,3

1Basic Medical Sciences, Purdue University, West Lafayette, IN, United States, 2Biomedical Engineering, Purdue University, West Lafayette, IN, United States, 3Electrical and Computer Engineering, Purdue University, West Lafayette, IN, United States


Medication effectiveness relies on patient adherence to a given treatment regimen. Often, patients do not adhere to the temporal guidelines set by their physicians and treatments therefore remain less effective. In situ forming implants (ISFIs) eliminate the need for patient adherence and release an effective dose of drug overtime. However, no methods exist to noninvasively and temporally validate drug release in vivo. We conducted in vivo experiments that validated the use of DWI to monitor ISFI diffusivity overtime. This has enormous implications in pharmaceutical research as this method can robustly quantify diffusivity in ISFIs post-implantation to ensure effective drug release.


Despite their popularity, oral dosage forms can lead to patient adherence problems.1,2 Oral medications often require repeated administration over time to maintain a therapeutic window. Because of frequent dosing requirements, many patients fail to maintain their dosing regimen, thereby decreasing treatment efficacy. This can lead to undesirable side effects and/or disease persistence.1,2 In situ forming implants (ISFIs) are an alternative drug delivery system which can eliminate the patient adherence problem.3 When administered using standard needle injection, ISFIs form a small drug-containing implant in tissue. Over time, drug diffuses across the shell of the implant via exchange with water in surrounding tissue.4 ISFI synthesis can be adjusted so medications are released at varying rates. However, no noninvasive and quantitative methods exist to image ISFIs rigorously and continuously over time, in vivo. By using DWI, we can characterize the diffusivity in ISFIs in vivo over time. Results from these experiments can be used to validate dosing effectiveness. This is critical since tissue environment can drastically affect dosing effectiveness.5


Animal studies were performed following protocols approved by the Purdue Animal Care and Use Committee. Four four-week old male C57Bl/6 WT mice were used. 52 kDa poly(lactic-co-glycolic) acid (PLGA), N-methyl-2-pyrrolidone (NMP), and fluorescein were combined in a 39:60:1 ratio. Mice were anesthetized using 1.5% isoflurane with an oxygen flow rate of 2 L/min. 100 L of solution was injected subcutaneously over the right and left flank using a 23-gauge needle. For imaging, mice were anesthetized with 2.5% isoflurane and an oxygen flow rate of 250 mL/min. Standard DWI imaging was performed (TE=17.5 ms, TR=2500 ms, FOV=30x30 mm2, slice thickness=0.80 mm, b=0,1000 s/mm2) at set time points (1h, 6h, 24h, 72h, 120h post-injection) using a Bruker BioSpec 70/30 USR 7T Preclinical MRI system and a RF RES 300 1H 075/040 QSN TR volume coil. Trigger was used to gate respirations. Using a custom Matlab code, two users manually selected ROIs from each slice containing the implant to calculate MD; this was performed in triplicate. MD was averaged across slices and between users; the average value was used for further analyses. After imaging at 120h, mice were euthanized and implants were dissected out for scanning electron microscopy (SEM). Implants were freeze-fractured over dry ice and lyophilized for 4-5 days. They were then mounted on aluminum stubs and sputtercoated with palladium. SEM was completed using a NovaNanoSEM and Quanta 3D FEG SEM. ANOVA test with Tukey multiple comparisons was used to test for statistical significance between groups. Values are reported as mean±standard deviation.


Previous phantom results were presented at ISMRM 2018.6 These in vitro data are shown in Figure 1 which displays ADC maps for 52 kDa implants over the full time course. After 14d, the implant was fragmented because of degradation. Figure 2 shows representative ADC maps from one mouse over the course of five days. For these images, the slice containing the largest implant area was selected. Figure 3 compares MD between in vitro and in vivo experiments. In general, a similar trend in MD was observed for both in vitro and in vivo analyses. However, at timepoints 6h and 3d, MD was significantly different between in vitro and in vivo data. SEM images from in vitro and in vivo experiments are shown in Figure 4. In contrast with in vitro analyses, the in vivo implants were more misshapen due to constraints of the subcutaneous space and it was more difficult to observe the distinction between the core and shell of the implant.


DWI is a noninvasive and sensitive technique shown to provide valuable insight into drug release mechanisms of ISFIs. Specifically, MD is a useful metric for the quantification implant diffusivity in vivo over time. In general, MD showed similar trends between in vitro and in vivo analyses. Differences can be attributed to the dynamic environment of the tissue. Increased pressure from surrounding tissue in vivo can explain the significantly larger MD in vivo versus in vitro at 3d.


ISFIs are a promising alternative drug delivery depot as they eliminate the patient adherence problem. However, methods to noninvasively and quantitatively assess ISFIs in vivo remain elusive. Because tissue environments likely alter drug release profiles post-implantation, it is critical observe these changes. Based on our results, DWI proves to be a sensitive tool to evaluate ISFI diffusivity in vivo. DWI provides robust information regarding the time course and magnitude of diffusion in ISFIs. These metrics can be used to improve ISFI design and ultimately, treatment efficacy.


We would like to thank Dr. Gregory Tamer for continued maintenance and assistance with the 7T system. We would also like to thank Dr. Sarah Calve for the mice used in these experiments.


1. Bose B, Jose J. Patient Medication Adherence: Measures in Daily Practice. Oman Med J. 2011;26(3):155-159.

2. Brown M, Bussell J. Medication Adherence: WHO Cares? Mayo Clin Proc. 2011;86(4):304-314.

3. Wertheimer A, Santella T, Finestone A, Levy R. Clinical and Economic Advantages of Modern Dosage Forms: Improving Medication Adherence. https://www.npcnow.org/system/files/research/download/Clinical-and-Economic-Advantages-of-Modern-Dosage-Forms-Improving-Medication-Adherence.pdf. Published 2006. Accessed Novemeber 6, 2018.

4. Solorio L, Olear AM, Hamilton JI, et al. Noninvasive Characterization of the Effect of Varying PLGA Molecular Weight bBends on In Situ Forming Implant Behavior Using Ultrasound Imaging. Theranostics. 2012;2(11):1064-1077.

5. Solorio L, Exner AA. Effect of the Subcutaneous Environment on Phase-Sensitive in Situ-Forming Implant Drug Release, Degradation, and Microstructure. J Pharm Sci. 2015;104(12):4322-4328.

6. Vike N, Li X, Hopkins K, Solorio L, Rispoli J. A Pilot Study of Correlations Between Mean Diffusivity and Drug Release Over Time For an Implantable Drug Delivery System. In: ISMRM. 1036; 2018.


Figure 1: Apparent diffusion coefficient (ADC) maps of a selected implant at each time point for 52 kDa implants.

Figure 2: Apparent diffusion coefficient (ADC) maps for in vivo implants injected into the right and left flank.

Figure 3: MD plotted over five days to compare implants formed in vivo (n=6) versus those formed in vitro (n=3). (error bars are mean ± st.dev; * indicates p < 0.05)

Figure 4: SEM images from in vitro and in vivo analyses of 52 kDa implants. (A) SEM images from in vitro implants inset with corresponding ADC maps at 4d, 6d, and 7d. (B) SEM images of in vivo implants from three mice at 5d.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)