Time Rules the Efficacy of Immune Checkpoint Inhibitors in Photodynamic Therapy

Abstract Lack of adequate effector T cells infiltrated in tumor is one of the main problems in the failure of immune checkpoint blockade therapy (ICBT). Photodynamic therapy (PDT) induced acute inflammation can sensitize tumors and activate T cells, thus assisting immune checkpoint inhibitors (ICI) against tumor growth and metastasis. T cells maturation and activation lag 3 to 7 days behind PDT. However, such timing in the combination therapy of ICI and PDT is commonly ignored in designing numerous multi‐functional integrated nanomedicines. Herein, the authors illustrate that intervention timing of ICI after PDT affects the anti‐tumor efficacy. A tumor‐targeting nanomedicine is prepared by encapsulating indocyanine green into CD44 specifically binding material, a hyaluronic acid conjugated lipid poly(ethylene glycol). The PDT nanomedicine is designed to induce a robust immune response in tumor. The optimal group (Combo‐STAR), ICI gave 5 days after PDT, significantly suppresses local tumor growth and eliminates metastasis. What should be highlighted is the time point of administration because if ICI is given too early, T cells are immature, otherwise, T cells are exhausted if ICI is given too late. This work presents theoretical guidance for raising awareness of intervention timing when augmenting ICBT with immune response inducers in clinic.


Synthesis and characterization of DSPE-PEG-HA
Following the synthesis route shown in Supplementary Fig. 1, HA (20 μmol) and EDC (5 μmol) were stirred and reacted in PBS solution (pH = 6.8) for 1 hour, then the methanol solution of HOBt (5 μmol) was dropwisely added to activate the carboxyl group on HA for different times (4, 10, 24 or 48 h), and then the PBS solution with a total of 10 μmol of DSPE-PEG-NH 2 was slowly added into the mixed solution, the polymerization was quenched 24 hours later. The resulting solution molecules were dialyzed with excess methanol for 24 hours through a dialysis bag with an MW cut-off of 7000 DA, followed by pure water : methanol (V : V) = 1 : 3 for one day and pure water: methanol (V : V) = 1 : 1 for another day. After lyophilization, the final product DSPE-PEG-HA was obtained as a white powder. 1 H NMR spectroscopy data of polymer DSPE-PEG-HA were obtained via a 400 MHz Bruker Advanced Spectrometer (BRUKER, Switzerland), and the chemical shifts were reported in ppm on the δ scale.
With the prolonging of activation time of carboxyl group, the yield also increased, but the water-solubility of the polymer also decreased gradually. As shown in Supplementary Fig. 2, the peak pattern of the chemical shift value of the polymer at 4.52 ppm becomes sharper and sharper with the increase of activation time, which was probably due to the increasing number of DSPE-PEG-NH 2 bonded on HA molecule leads to poor water-solubility of the product, resulting in a stronger and sharper peak of solvent D 2 O. After comprehensive consideration, we chose 10 hours as the optimal activation time of carboxyl for subsequent experiments. The yield of DSPE-PEG-HA was 36.29%. The 1 H NMR spectroscopy data shown in Supplementary Fig. 3 showed the substitution degree of HA on polymer was 15.80 %.

Determination of critical micelle concentration (CMC)
The critical micelle concentration (CMC) of DSPE-PEG-HA was estimated using fluorescence spectroscopy with pyrene as a probe. 1 mL of pyrene solution in acetone (5 × 10 -6 mol/L) in a brown bottle was stirred in a fume hood in dark until the acetone evaporated. DSPE-PEG-HA solution diluted with PBS at different concentrations ranging from 0.01 μg/mL to 1.0 mg/mL was added to each bottle, and the solution was stirred for 24 h in a constant temperature magnetic stirrer. The emission spectra of pyrene were monitored by a fluorescence spectrophotometer (E m =334 nm) with an excitation wavelength of 334 nm and a slit width of 5 nm. The intensity ratios of I 373 /I 384 were plotted against the DSPE-PEG-HA concentration to determine CMC. The CMC of DSPE-PEG-NH 2 was estimated by the same method and the result was shown in Supplementary Fig. 4.

Preliminary characterization of drug-loaded micelles
As shown in Supplementary Fig. 5, an ultraviolet-visible (UV) spectrophotometer was used to scan the absorbance of free ICG, IM, and HIM at a wavelength of 500-1000 nm and the UV absorption spectra were recorded. Similarly, their emission spectra at 780 nm excitation wavelength were monitored by a fluorescence spectrophotometer.
The standard curve of ICG aqueous solution at 779nm was determined by UV spectrophotometer (Supplementary Fig. 6). The regression equation was y = 0.2056x + 0.0661, R 2 = 0.9991, indicating that ICG had a good linearity in the concentration range of 1-5 μg/mL.