Acidosis in tumor microenvironment could indicate tumor aggressiveness and treatment effects. The extracellular pH (pHe) of the tumor microenvironment is in the range of 6.5 to 6.9, whereas the pHe of normal tissue is approximately 7.2 to 7.5. Acidosis is the result of increased lactic acid production by aerobic glycolysis in cancer cells, which associated with hypoxia, poor perfusion of tumors and poor buffering in the microenvironment. This resultant acidity can enhance tumor aggressiveness, metastasis and angiogenesis. Furthermore, low pHe also leads to resistance to immunotherapy and specific chemotherapy. Thus, it has become increasingly important to measure pHe in diagnosis and treatment of cancers.

Chemical Exchange Saturation Transfer (CEST) MRI has shown promises in imaging pH alterations in tumors. It enables the detection of natural exchangeable protons of molecules, such as the amide protons of proteins, without the need of metallic or radioactive labeling. This proton exchange also enhances the sensitivity of detecting endogenous molecules at low concentrations, such as proteins, glucose, creatine and glutamate. Based on the proton exchange mechanism, CEST contrast is sensitive to pH, temperature and concentration of exchanges protons. Endogenous amide proton transfer weighted (APTw) CEST calculated by MTR_asym at 3.5 ppm has been widely studied in brain tumors. It could identify radiation necrosis from tumor recurrence. This is because of the high protein concentration in tumor and tumor is acidic. Since APTw indicates multiple contributions, such as pH, concentration and cellularity, it would be rather hard to use it to image pHe independently. Therefore, exogenous CEST contrast agents have been studied for pH imaging. Iodinated CT contrast agents, such as iopamidol, iopromide and iohexol, could be exogenous CEST contrast agents. These CT contrast agents have amide protons that are CEST detectable. Hence, it can be repurposed as CEST contrast agents for pH imaging. CEST pH imaging using CT contrast agents have demonstrated high sensitivity in detecting the low pHe in both preclinical mouse models and cancer patients. This approach requires the administration of CT contrast agents, it might limit the frequency assessment of pHe in clinical applications due to the following challenges. First, the CT contrast agent will be washed out eventually in tumor microenvironment, thus the imaging window will be limited to several hours after contrast administration. Second, high dose or repeated administration of CT contrast agents would lead to nephrotoxicity in patients. Thus, the design of biomaterials that enables the maintenance of high dose of CT agents in the tumor microenvironment and non-invasive imaging readout are required to facilitate the clinical applications of pHe imaging for cancer diagnosis and treatment monitoring. Here, we aim to design CT contrast agent-loaded liposomal hydrogel prepared by microfluidics to address these issues.

First, we designed a microfluidic device to facilitate the preparation of liposomal alginate microbeads. Numerous approaches are available to prepare liposomal alginate microbeads, such as extrusion, electrospray and microfluidics. For CEST imaging of pHe, we will need to fabricate microbeads with small size at tenth micron to provide high surface-to-volume ratio and with spherical shape to alleviate the immune response. Thus, microfluidics with flow-focusing mode and two cross-junctions were designed for liposomal alginate microbead preparation coupled with internal gelation. The first cross-junction with 50 μm width and 50 μm height was designed for precursor droplet formation, and the second cross-junction was used for gelation of hydrogel microbeads. The optimized flow rate of oil/ acid phase and concentration of acetic acid in oil/acid phase were 1500 μL/h and 1.5% v/v, respectively. The encapsulation efficiency of contrast agents in liposomal alginate microbeads increased to 40% with optimized parameters. Monodispersed pure alginate microbeads (without liposome) with spherical shape were prepared with microfluidic device. Diameter of microbeads decreased from 46.6 to 31.1 μm with oil flow rate increased from 200 to 600 μL/h.

Second, we prepared CT contrast agents (iopamidol and iohexol) loaded liposomes by thin film hydration and optimized the CEST contrast. The dynamic average size of iopamidol liposome was ∼200 nm with PDI (polydispersity index) of 0.19 and Zeta potential was -1.38±1.1 mV measured by Zetasizer. Liposome particle concentration for iopamidol liposome and iohexol liposome were (2.9±0.6)× 10¹⁷/mL and (2.8±0.8)× 10¹⁷/mL, respectively. The EE_Iop-lipo and EE_Ioh-lipo were 50.6% and 41.2%, respectively. CEST contrast of iopamidol liposome and iohexol liposome were 42.4% and 37.0% at B₁=1.6 μT, respectively, with optimized liposome formulation. This high level of contrast was necessary to ensure a decent level of contrast in resulted alginate microbeads.

Third, iopamidol liposome and iohexol liposome were incorporated into alginate hydrogel microbeads with microfluidics, respectively. The optimized flow rate of oil/acid phase and concentration of acetic acid were 150 μL/h and 1.5% v/v, respectively. The EE_{Iop-lipobeads} and EE_{Ioh-lipobeads} were 40.8% and 40.5%, respectively. Average diameter of iopamidol-loaded liposomal alginate microbeads (Iop-lipobeads) decreased from 54.5 μm to 35.6 μm with oil flow rate increased from 200 to 600 μL/h. We also optimized the acquisition parameter for the acquisition of CEST contrast of the liposomal hydrogel microbeads. Iop-lipobeads at 4.2 ppm increased from 4.2% to 11.5% when B₁ increased from 0.6 μT to 1.6 μT, and there was no further increase observed beyond 1.6 μT. Similarly, CEST contrast of iohexol-loaded liposomal alginate microbeads (Ioh-lipobeads) at 4.2 ppm increased from 5.2% to 9.8% when B₁ increased from 0.6 μT to 1.6 μT. Therefore, 1.6μT was the optimized B₁ power. Stable CEST contrast for 20 days were observed with daily replacement of saline at 37ºC in vitro. We further characterized the pH sensitivity of these microbeads at relevant pHs. We observed a 13.4% and 50.7% percentage increase in CEST contrast at 4.2 ppm in Iop-lipobeads and Ioh-lipobeads, respectively, which were observed when pH increased from 6.5 to 7.0, Moreover, their CEST contrast enables the identification of the hypoxic medium from the normoxic medium.

Fourth, we studied the pH sensitivity in a tumor mouse model with subcutaneous U87 tumor, and upon bicarbonate treatment. CEST contrast at 4.2 ppm of Iop-lipobeads in mice without tumor increased from 4.6% to 6.9% with B₁ power increased from 0.6 μT to 1.6 μT, then decreased to 6.2% at 2.0 μT, which was slightly different from in vitro due to the additional magnetization transfer contribution in vivo. Thus, the same B₁ power of 1.6 μT was applied for in vivo experiments. After injection to tumor-bearing mice, Iop-lipobeads showed stable CEST contrast at 6.4% for two days. CEST at 4.2 ppm of Iop-lipobeads increased to 7.5% and 7.3% during and after treatment, respectively, representing a percentage increase of 19.7% and 15.2%, respectively. In addition to the pH sensing in the microbead region, the endogenous CEST contrast of tumors could indicate the tumor responses. Thus, we compared the percentage changes of amide CEST (at 3.4 ppm) and rNOE CEST (at -3.4 ppm) of tumors, which were calculated by Lorentizian fitting, with that of microbeads. Amide CEST in tumors was 6.9%, 7.7% and 7.5% and rNOE in tumors was 6.3%, 6.9% and 6.5% before, during and after treatment, respectively. The percentage changes were only 11.7% and 9.1% for amide CEST during and after treatment, respectively; and 9.6% and 4.1% for rNOE CEST during and after treatment, respectively. As shown in our findings, the percentage changes of CEST contrast of Iop-lipobeads observed were in general higher than the endogenous CEST contrast (amide and rNOE CEST), and it is significantly different (P<0.05) during treatment and after treatment when compared with amide CEST of tumors.

In summary, microfluidic preparation of Iop-lipobeads is robust approach to generate liposomal microbeads at tenth of micron. CEST contrast of the Iop-lipobeads is sensitive to local pH changes, especially for sensing pHe during and after bicarbonate treatment. Moreover, CEST MRI provides multiple parameters, i.e. CEST contrast of Iop-lipobeads at 4.2 ppm, tumors at 3.5 ppm and -3.5 ppm, to assess the treatment effect and tumor responses. These findings provide an effective mean to image pHe of tumors during the course of treatment non-invasively, longitudinally and repeatedly.