Magnetic Resonance Imaging-Guided Focused Ultrasound-Based Delivery of Radiolabeled Copper Nanoclusters to Diffuse Intrinsic Pontine Glioma

Abstract

Diffuse intrinsic pontine glioma (DIPG) is an invasive pediatric brainstem malignancy exclusively in children without effective treatment due to the often-intact blood-brain tumor barrier (BBTB), an impediment to the delivery of therapeutics. Herein, we used focused ultrasound (FUS) to transiently open BBTB and delivered radiolabeled nanoclusters (⁶⁴Cu-CuNCs) to tumors for positron emission tomography (PET) imaging and quantification in a mouse DIPG model. First, we optimized FUS acoustic pressure to open the blood-brain barrier (BBB) for effective delivery of ⁶⁴Cu-CuNCs to pons in wildtype mice. Then the optimized FUS pressure was used to deliver radiolabeled agents in DIPG mouse. Magnetic resonance imaging (MRI)-guided FUS-induced BBTB opening was demonstrated using a low molecular weight, short-lived ⁶⁸Ga-DOTA-ECL1i radiotracer and PET/CT before and after treatment. We then compared the delivery efficiency of ⁶⁴Cu-CuNCs to DIPG tumor with and without FUS treatment and demonstrated the FUS-enhanced delivery and time-dependent diffusion of ⁶⁴Cu-CuNCs within the tumor.

Graphical Abstract

INTRODUCTION

Diffuse intrinsic pontine glioma (DIPG) arises in the pons and is a leading cause of pediatric brain tumor death. At diagnosis, the majority of patients are between 5 and 10 years old and have a median survival of less than 1 year.^1–2 Resection is not an option because of the diffuse nature of the tumor and the critical function of the pons. Focal radiation is the standard of care treatment for children with DIPG, and the addition of chemotherapeutic agents or small molecule inhibitors has failed to show an improvement in survival over radiation alone in numerous clinical trials conducted over the past several decades.^3–4 This may be due to ineffective drug delivery to DIPG, which often has an intact blood-brain tumor barrier (BBTB), and other potential obstacles, such as the restricted diffusion of drugs through the brain parenchyma.⁵ Many methods have been developed to overcome the obstacles presented by the BBTB and increase drug penetration, such as intra-arterial administration of osmotic agents enabling temporary disruption of BBTB,⁶ convection-enhanced delivery that interstitially infuses drugs under a constant pressure gradient,⁷ and laser interstitial thermotherapy that allows laser ablation of a tumor via insertion of an optical fiber.⁸ These techniques either lead to BBTB disruption in the whole brain, or are detrimentally invasive. The combination of focused ultrasound (FUS) and microbubbles (MBs), which enhances the BBTB permeability, has drawn great attention due to its non-invasive and localized delivery abilities and minimal neuronal damage.⁹ FUS non-invasively penetrates the skull and focuses energy on a small region of millimeter-scale dimensions. MBs, which have been used in the clinic as ultrasound contrast agents for imaging, are administered by intravenous (IV) injection. Following injection, their relatively large size (1–_{10 μm)} confines them to the vasculature instead of naturally penetrating the BBTB. MBs amplify and localize FUS-mediated mechanical effects on the vasculature through FUS-induced cavitation (microbubble expansion, contraction, and collapse), which generates mechanical forces on the vasculature and transiently increases the BBTB permeability. To better localize the FUS treatment, magnetic resonance imaging-guided FUS (MRgFUS) treatment enables precise targeting and treatment planning, and allows targeted delivery of various therapeutics without damaging surrounding healthy structures.^10–13 This integrated technology has progressed rapidly and led to numerous ongoing clinical trials in brain cancer patients due to the feasibility and safety of FUS.^{9, 14–17}

Nanostructures have been widely used for cancer research due to the versatile physicochemical properties and multifunctionality for diagnosis, drug delivery and treatment.^18–20 Through the combination with FUS to break BBTB, a variety of nanostructures have been used for brain tumor imaging and treatment with proven advantages in terms of local targeting and multifunctional theranostics.^21–24 However, there are no applications in DIPG. Previously, we reported the trans-blood brain barrier (BBB) delivery of an ultrasmall gold nanocluster (⁶⁴Cu-AuNCs) to the pons of wildtype mice following FUS treatment,^25–27 which paved a path for imaging and treatment of DIPG. Herein, we first optimized the acoustic pressure for effective and safe BBB opening. Following MRgFUS, we precisely targeted DIPG tumor in pons at early stage and validated the BBTB opening using positron emission tomography/computed tomography (PET/CT). Using an ultrasmall and biodegradable copper nanocluster intrinsically radiolabeled with ⁶⁴Cu (⁶⁴Cu-CuNC) as a model-drug,²⁸ we studied the tumor delivery efficiency, intratumoral diffusion and retention in a mouse DIPG tumor model and laid the foundation for potential early diagnosis and treatment of DIPG (Figure S1).

RESULTS AND DISCUSSION

The effective trans-BBTB delivery of nanoclusters is mainly affected by both physicochemical properties of nanoclusters and FUS parameters for BBTB opening. Based on our previous report and optimized procedures,^25,28 we synthesized ⁶⁴Cu-CuNCs with ultrasmall hydrodynamic sizes (5.6 ± 1.5 nm) and neutral charge (zeta potential= −0.04 ± 0.12 mV) for effective renal clearance and optimal biodistribution (Figure S2 and Table S1). In vivo pharmacokinetic evaluation of ⁶⁴Cu-CuNCs showed rapid systemic clearance with little retention in any major organs at 48 h post injection.²⁸ Furthermore, cytotoxicity assay in U87 cells showed that the IC₅₀ values for CuNC and Cu²⁺ were 18.6 μg/mL and 5.1 μg/mL, respectively, indicating the advantage of nanostructure relative to Cu²⁺ cation alone. We next studied the effect of varying FUS acoustic pressures on BBB opening by PET imaging of ⁶⁴Cu-CuNCs in wildtype mice. Based on our previous report,²⁹ we targeted FUS on the left pons of C57BL/6 mice (n = 4 – 5) with FUS pressures of 0.28, 0.61, 0.72 and 0.85 MPa prior to intravenous injection of ⁶⁴Cu-CuNCs. PET/CT images demonstrated clear radioactive signals at the left pons for 0.61, 0.72 and 0.85 MPa at 24 h post-injection, while there was negligible uptake of ⁶⁴Cu-CuNCs at pons treated with 0.28 MPa FUS pressure (Figure 1 and S3). Quantitative analysis at 24 h in mice treated with 0.28 MPa FUS pressure showed ⁶⁴Cu-CuNCs uptake of 0.37 ± 0.02 percent injection dose per gram (%ID/g) (n = 4) at the treated left pons and 0.34 ± 0.04 %ID/g at the non-treated right pons, while the background blood retention in the pons of mice without FUS treatment was 0.42 ± 0.06 %ID/g (n = 4), demonstrating the ineffectiveness of BBB opening under 0.28 MPa FUS treatment. However, mice treated with 0.61, 0.72 and 0.85 MPa FUS showed significantly increased pons uptakes of 1.11 ± 0.46%ID/g (n = 5, P<0.01), 1.45 ± 0.36 (n = 4, P<0.0001), and 1.84 ± 0.21 %ID/g (n = 4, P<0.0001), respectively, at the treated left pons. Interestingly, we also observed increased ⁶⁴Cu-CuNCs uptake at the non-treated right pons in mice treated with 0.72 MPa and 0.85 MPa pressures, suggesting a higher level of BBB disruption by elevated FUS pressure, which was consistent with our previous report (Figure 1b).²⁶ However, there was no significant difference in the treated/non-treated uptake ratios for the three FUS pressure settings (0.61, 0.72 and 0.85 MPa) (Figure 1c), indicating the effectiveness of these three FUS pressures to open BBB for contrast imaging. Moreover, histopathological examination showed no damage to brain tissue at 0.61 and 0.72 MPa but some hemorrhage at 0.85 MPa pressure (Figure S4). Therefore, to minimize safety concerns while still obtaining sufficient delivery efficiency, we opted to use 0.61 MPa FUS pressure as optimal for DIPG tumor imaging in this study. In addition, our previous studies have verified that 0.61 MPa FUS pressure resulted in no tissue damage in the brain.^25–27

Figure 1.

(a) In vivo PET/CT images of ⁶⁴Cu-CuNCs without FUS treatment and under 0.28 and 0.61 MPa FUS pressure in WT mice at 24 h post IV injection. (b) Quantitative analysis of ⁶⁴Cu-CuNCs uptake in FUS treated left pons and non-treated right pons of WT mice under 0.28, 0.61, 0.72, and 0.85 MPa FUS pressures and the pons without FUS treatment in WT mice. (c) the uptake ratios of treated sites vs non-treated sites of the FUS treated mice under different pressures (*** p<0.001, **** p<0.0001, n = 4–5).

Next, we applied FUS to an established RCAS/TVA mouse DIPG model. Briefly, nestin-expressing brainstem progenitors of Nestin TVA; p53 floxed mice were infected at postnatal day 4–5 by the injection of virus producing cells (DF1s) expressing PDGF-B, H3.3K27M, and Cre (to delete p53 specifically in the tumor cells) at a ratio of 1:1:1. The cells were injected slowly over 2 minutes into the posterior fossa at 3 mm deep to lambda, into the brainstem of post-natal day 7 pups.^{1, 30} The anatomic characterization of tumor progression was characterized by MRI and histological variation was assessed by hematoxylin and eosin (H&E) staining. At 3 weeks, MRI showed nearly undetectable signal in both T₁-weighted and T₂-weighted contrast enhanced images (Figure 2a). However, H&E staining of collected specimens revealed the presence of tumor cells within pons indicating the development of early tumors (Figure 2b).³¹ In comparison, contrast enhanced T₁-weighted and T₂-weighted images demonstrated significant hyperintensity in the pons of both 4-week-old and 5-week-old mice (Figure 2c and S5), indicating the significant invasion of tumors which was confirmed by the H&E staining of pons sectioned from 4-week-old mice (Figure 2b and 2d). Importantly, after MRgFUS treatment at the 3-week-old DIPG mice, enhanced signal was observed at the position of tumors in contrast enhanced MR images, which indicated the precise FUS targeting of the tumors that allowed the penetration of contrast agent through the disrupted BBTB (Figure S6). This demonstrated the efficiency and potential of MRgFUS to deliver reagents for early diagnosis and treatment of DIPG.

Figure 2.

Comparison of contrast enhanced T₁ and T₂ weighted MRI images of (a) 3-week-old DIPG mouse and related (b) H&E staining with MRI images at (c) 4-week-old DIPG mice and (d) H&E staining.

To further confirm the opening of BBTB for early detection and treatment of DIPG, PET/CT evaluation with/without FUS in the same mice was conducted using a short-lived radioisotope ⁶⁸Ga (T_1/2 = 67.7 mins) labeled hydrophilic radiotracer, ⁶⁸Ga-DOTA-ECL1i (MW=1375.7 Da), in DIPG mice.³² The rapid decay of ⁶⁸Ga enabled the comparison of the FUS treatment effect in the same mouse in a span of 2 days. Specifically, three-week-old DIPG mice without FUS treatment were injected intravenously with 7.4 MBq ⁶⁸Ga-DOTA-ECL1i via tail vein and scanned with PET/CT at 1 h post injection, which showed no tracer uptake within pons (Figure 3a). After the decay of ⁶⁸Ga, the same mice were treated with MRgFUS the next day prior to a second ⁶⁸Ga-DOTA-ECL1i PET/CT scan. As shown in Figure 3a, significant PET signals were determined in the pons of mice with FUS treatment. Quantification showed that ⁶⁸Ga-DOTA-ECL1i uptake in pons of treated mice (1.25 ± 0.53 %ID/g, n = 4) was approximately 2-fold higher than that acquired in non-treated mice (0.42 ± 0.22 %ID/g, n = 4, p<0.01), which demonstrated the effectiveness of MRgFUS for opening the BBTB in the setting of DIPG (Figure 3b).

Figure 3.

(a) In vivo PET/CT images and (b) the quantitative tumor uptake of ⁶⁸Ga-DOTA-ECL1i in 3-week-old DIPG tumor mice with and without FUS treatment at 1 h post IV injection (** p<0.01, n = 4).

We next analyzed the effect of FUS on ⁶⁴Cu-CuNCs accumulation and distribution within pons in 3 and 4-week old DIPG mice using PET/CT imaging. Mice were randomized into FUS treated and non-treated groups. As shown in Figure 4a, a strong PET signal was evident in the pons of both 3 and 4-week-old DIPG mice at 24 and 48 h post injection of ⁶⁴Cu-CuNCs. Quantitative analysis showed similar tumor uptakes of 2.68 ± 0.20 %ID/g (n = 5) and 2.73 ± 0.18 %ID/g (n = 5) at 24 and 48 h post injection for FUS treated 3-week-old mice, values which were significantly higher than those in untreated mice (Figures 4b and S7, p<0.0001, n = 6). Additionally, the distribution volume of radioactive signal in the pons increased from 21.90 ± 2.24 mm³ to 34.4 ±2.22 mm³ between 24 and 48 hours (Figure 4c, p<0.01, n = 5 for both), suggesting the dynamic diffusion of ⁶⁴Cu-CuNCs within tumors, consistent with our previous report.²⁵ Furthermore, ex vivo autoradiography performed immediately after PET/CT imaging clearly showed distribution of ⁶⁴Cu-CuNCs within the tumor in the FUS treated mice while minimal retention was observed in the pons of mice without FUS treatment. Quantification of the radioactivity within the pons revealed 2-fold higher counts in the treated mice than the non-treated mice, confirming the PET data (Figure S8). The similar tumor uptakes between the 24 and 48 h time points may be attributed to the closing of the transiently disrupted BBTB within a few hours of treatment.³³ The prolonged intra-tumoral retention and dynamic diffusion of nanoclusters could have therapeutic benefit if they were loaded with anti-tumor agents.

Figure 4.

In vivo PET/CT images, quantification of tumor uptake and intra-tumoral distribution volumes of ⁶⁴Cu-CuNCs at 24 h and 48 h post IV injection in (a, b, and c) 3-week-old and (d, e, and f) 4-week-old DIPG mice. (*** p<0.001, **** p<0.0001, n = 3–6).

In the 4-week-old tumor mice, tumor uptakes showed similar profiles to those at 3 weeks with 2.98 ± 0.62 %ID/g and 2.64 ± 0.26 %ID/g determined in the pons at 24 and 48 h post injection (Figure 4d and 4e, p<0.0001, n = 3, Figure S9). However, the diffusion of the nanoclusters showed approximately two-fold increase from 24 h (29.3 ± 3.32 mm³, n=3) to 48 h (73.27 ± 4.74 mm, n=3, p<0.001) (Figures 4f). This could be a result of diminishing tumor integrity over time, resulting in a leakier tumor structure compared to DIPG tumors at 3 weeks, while still maintaining BBTB integrity that excludes the permeation of the nanoclusters. This retaining integrity of BBTB in 4-week-old mice was confirmed by the low signal in the non-treated mice, whose tumor uptake was comparable to the blood retention of circulating ⁶⁴Cu-CuNCs in non-treated 3-week-old DIPG mice. Moreover, at late stage of this DIPG model (5-week-old), due to the deterioration of BBTB,^{34 64}Cu-CuNCs accumulation was observed even without FUS treatment (Figure S10), which was further confirmed by T₁ and T₂ weighted MR images (Figure S5). Interestingly, in contrast to the reported low tumor uptake of ⁸⁹Zr-bevacizumab in DIPG mice with disrupted BBTB,³⁵ our combined strategy demonstrated significantly enhanced tumor delivery, diffusion and retention of ⁶⁴Cu-CuNCs as a model-drug in DIPG mice, indicating the unique advantages of our ultrasmall nanoclusters.

CONCLUSION

In summary, we examined the optimal FUS pressure to open BBB of naive mice for effective delivery of ⁶⁴Cu-CuNCs to pons. Using a hydrophilic, short-lived PET tracer, ⁶⁸Ga-DOTA-ECL1i, we demonstrated the effectiveness of MRgFUS to precisely target the pons and open the BBTB for potential drug delivery. Following MRgFUS treatment, ⁶⁴Cu-CuNCs exhibited significant tumor uptake, dynamic distribution within tumors, and prolonged intra-tumoral retention within DIPG tumors. Together, these findings suggest that FUS-enhanced delivery of biodegradable nanoclusters may provide a much-needed advancement in the treatment of DIPG.

Comparing to other trans-BBB strategies,^{5, 36} our MRgFUS strategy demonstrated effective disruption of BBTB and delivery of radiolabeled ⁶⁴Cu-CuNCs to the DIPG tumor with little normal tissue damage.^25–26 The use of ultrasmall ⁶⁴Cu-CuNCs could facilitate the tumor penetration, as well as the rapid renal clearance to further reduce in vivo toxicity. In future studies, once drug is loaded onto CuNCs, the biodegradability of CuNCs could promote rapid in situ drug release upon tumor entry to make it a better treatment for diffused tumor cells.²⁴ However, there are also some limitations in our studies. The safety evaluation of MRgFUS has only been assessed by histology.^25–26 Future studies are needed to carefully evaluate the inflammatory response and chronic effect due to the FUS disruption of BBTB. The ⁶⁴Cu-CuNC used for this proof-of-concept study was a non-targeted nanocluster. Future studies need to focus on targeted approach based on the specific expression of certain biomarkers.