1. Introduction
Since immunotherapy began in 2011, immune checkpoint inhibitor (ICI) therapy based on the programmed death protein 1 (PD-1)/programmed death protein ligand 1 (PD-L1) signaling pathway has played an increasingly important role in cancer treatment [
1]. Recent clinical studies have demonstrated that PD-L1 expression has a significant influence on therapeutic efficacy, and positive responses to PD-1/PD-L1 ICI therapy are only possible in patients with tumors that contain high levels of expressed PD-L1 [
2,
3]. The main method used in the clinic to assess PD-L1 expression in the tumor microenvironment is invasive biopsy in conjunction with immunohistochemistry (IHC). However, due to the high heterogeneity of PD-L1 expression within both primary tumors and metastases, the ability of IHC detection to accurately evaluate the PD-L1 expression status in real-time and predict treatment response is limited, especially for patients with metastatic diseases [
4]. Compared with IHC, nuclear medicine techniques allow real-time, noninvasive visualization of tumor PD-L1 expression in the whole body, which can overcome the shortcomings of IHC methods.
The first-generation PD-L1-targeted radiotracers developed for positron emission tomography (PET) or single photon emission computed tomography (SPECT) imaging were mainly based on anti-PD-L1 antibodies [
5,
6,
7,
8,
9,
10,
11,
12,
13,
14,
15,
16]. For these radiotracers, a longer clearance time is often needed to obtain the optimal contrast of immune imaging. Sometimes, to match the slow clearance rate of antibodies, long half-life nuclides (e.g.,
89Zr: T
1/2 = 78.4 h;
111In: T
1/2 = 2.3 d) have to be used to prepare antibody-based radiotracers, which may lead to an increased internal radiation-absorbed dose in patients [
17]. Therefore, recent studies have focused mainly on nonantibiotic radiotracers with better pharmacokinetic properties, such as affibodies [
18,
19], nanobodies [
20,
21,
22], antibody fragments [
23], peptides [
24,
25,
26,
27,
28], and small molecules [
29,
30,
31]. Among these radiotracers, peptide-based PD-L1-targeted radiotracers have attracted increasing interest due to their small size, ease of synthesis and modification, high affinity, and good tumor penetration [
32].
The peptide WL12 is a macrocyclic peptide with the sequence cyclo[AcTyr-NMeAla-Asn-Pro-His-Leu-Hyp-Trp-Ser-Trp(Me)-NMeNle-NMeNle-Orn-Cys]-Gly-NH2, which can efficiently inhibit the PD-1/PD-L1 interaction (reported IC
50 of 22 nM) [
33,
34]. Recently, several WL12-conjugated derivatives have been radiolabeled with
64Cu,
68Ga, and
18F as positron emission tomography (PET) imaging agents [
34,
35,
36,
37,
38,
39,
40,
41]. [
64Cu]WL12 is the first reported WL12-based radiotracer. It displayed PD-L1 specificity both in vitro and in vivo, demonstrating the feasibility of introducing a chelating group into the -
Orn domain of WL12 [
34]. However, high radioactivity accumulation in the kidneys and liver was also observed with [
64Cu]WL12. Subsequent studies on
68Ga- [
38] and
18F-labeled [
36] WL12 peptides were performed to generate PET radiotracers with improved pharmacokinetic properties and image contrast. Among them,
68Ga-NOTA-WL12 was evaluated in a first-in-human study [
42]. The results demonstrated the feasibility of using
68Ga-NOTA-WL12 to detect tumor PD-L1 expression levels in patients with non-small cell lung cancer and its potential to guide ICI therapy. However,
68Ga-NOTA-WL12 also displayed high hepatic uptake and a slower clearance rate from the hepatobiliary system, resulting in higher background levels in the abdomen. Compared with PET imaging agents, few WL12-based radiotracers for single-photon emission computed tomography (SPECT) imaging have been published. To date, only Ferro-Flores G et al. reported the preclinical and clinical evaluation of two [
99mTc]Tc radiolabeled cyclic peptides ([
99mTc]Tc-iPD-L1 and [
99mTc]Tc-WL12) in 2023 [
43]. Compared to [
99mTc]Tc-WL12, [
99mTc]Tc-iPD-L1 exhibited greater HCC827 tumor uptake, as well as higher uptake in the liver and kidneys.
Currently, developing a novel WL12-based radiotracer with optimal excretion kinetics is highly desirable to obtain a high target-to-background ratio for PD-L1-positive tumor imaging. In this study, we focused on developing
99mTc-labeled HYNIC-conjugated WL12 for SPECT imaging, which is expected to provide a simple, convenient, and inexpensive diagnostic tool for assessing the status of PD-L1 in cancer patients.
99mTc is the most widely used radionuclide due to its excellent nuclide properties, low cost, and high availability through a
99Mo/
99mTc generator. Herein, 6-hydrazino nicotinamide (HYNIC) was used as the bifunctional chelating group to conjugate the -
Orn of WL12 for the following reasons: (1) A high
99mTc-labeling efficiency can be achieved at very low concentrations of HYNIC-conjugated biomolecules [
44], which is beneficial for obtaining radiotracers with high molar activity through a kit preparation process while following the requirements of good manufacturing practice (GMP). (2) During the
99mTc-HYNIC radiolabeling process, coligands are essential because they occupy the remaining sites of the
99mTc coordination sphere to form
99mTc-HYNIC complexes with good stability. Due to the significant effect of coligands on their physicochemical properties, such as hydrophilicity, charge, and stability [
44,
45,
46,
47,
48], a strategy is convenient for optimizing the pharmacokinetic properties of
99mTc-HYNIC complexes via the selection of coligands. In this study, lyophilized kits containing different coligands, tricine/M (M = triphenylphosphine-3,3′,3″-trisulfonic acid trisodium salt (TPPTS), isonicotinic acid (ISONIC), 3,5-pyridine dicarboxylic acid (PDA), or 4-pyridinesulfonic acid (4-PSA), respectively), were developed to prepare
99mTc-WL12 complexes with high radiochemical purity and suitable molar activity. In vitro and in vivo evaluations of these radiotracers were performed, and the results were compared to develop a favorable PD-L1-targeted tumor imaging agent with optimal pharmacokinetic properties for SPECT imaging.
3. Discussion
A simple, efficient, and reproducible kit-based radiolabeling process is essential for the clinical application of
99mTc-radiolabeled radiopharmaceuticals. In this study, a kit formulation was developed for the routine preparation of [
99mTc]Tc-HYNIC-WL12-tricine/M (M = TPPTS, PDA, ISONIC, and 4-PSA). During the process of optimizing the kit formulation, a high radiolabeling yield (>97%) of the radiotracers could be obtained with 5 μg of the cold peptide HYNIC-WL12 (the lowest amount tested,
Figure S5). However, significant glass surface absorption of these radiotracers was also observed under these low levels of cold peptide. After all the [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS solution was removed from the common glass vial, more than 80% of the radioactivity remained on the glass wall. The glass-surface absorption of the radiotracer could be effectively reduced by utilizing silanized glass vials and adding more HYNIC-WL12 peptide as a carrier. However, the addition of cold ligands decreased the molar activity of radiotracers. Generally, high molar radioactivity is needed for receptor-targeted probes due to the limited binding sites and low concentration of biomarkers (usually at the nanomolar level). To evaluate the impact of excessive cold HYNIC-WL12 ligand, a comparative biodistribution experiment was conducted between [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS injection with or without excess cold ligand. The results showed that both tumor uptake and tumor-to-background ratios were significantly reduced when excess cold HYNIC-WL12 was removed by further HPLC purification. This result suggested that excessive mass of cold HYNIC-WL12 exerts a positive effect on [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS sensitivity in PD-L1-positive tumors. As shown in the biodistribution data of [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS with different molar activities (Max: >3 TBq/μmol by HPLC purification, Min: 1.5 GBq/μmol), tumor uptake exhibited a bell-shaped trend with decreasing molar activity. A similar phenomenon was also found for several reported peptide-based radiotracers [
49,
50,
51] and PD-L1-targeted radiolabeled antibodies [
52,
53]. This was probably because cold HYNIC-WL12 could occupy nonspecific or PD-L1 binding sites in nontarget tissues [
8,
54], allowing more “free state” radiotracers to accumulate in tumors with high PD-L1 expression. We concluded that a radiotracer with an
Am ranging from 100.5 GBq/μmol to 300 GBq/μmol yielded the best tumor uptake and tumor-to-background contrast. At the typical radiopharmaceutical dose (740–1110 MBq), a kit containing 15 μg of HYNIC-WL12 peptide in a silanized glass vial would be suitable for routine clinical
99mTc radiolabeling.
The results of the IC
50 determination (as shown in
Table S1) displayed that the introduction of the HYNIC moiety in the -
Orn of WL12 has little influence on the affinity of HYNIC-WL12 for the PD-L1 protein. The results of in vitro cellular assays further demonstrated that four
99mTc-labeled HYNIC-WL12 radiotracers bind to tumor cells in a PD-L1 expression-dependent manner. The cellular uptake of the four radiotracers in MC38-B7H1 cells (PD-L1-positive) was approximately 2.38–6.73-fold higher than that in MC38 cells (PD-L1-negative), which could also be significantly blocked by the addition of the WL12 peptide (
p < 0.01). The uptake of [
99mTc]Tc-HYNIC-WL12-tricine/M (M = TPPTS or PDA or ISONIC or 4-PSA) in MC38-B7H1 tumors was 18.22 ± 4.57, 4.61 ± 1.32, 6.63 ± 0.80, and 6.96 ± 1.15%ID/g at 2 h p.i., respectively, which was 3.11–6.93-fold greater than that in MC38 tumors at the same time points (
Figure 3, 2.63 ± 0.98, 1.48 ± 0.55, 1.70 ± 0.27, and 1.22 ± 0.19%ID/g, respectively). The difference in radioactive uptake between the two tumor models was consistent with the IHC staining results, in which PD-L1 expression in the MC38-B7H1 tumors was higher than that in the MC38 tumors (
Figure S7). In addition, radioactive accumulation in MC38-B7H1 tumors was reduced by approximately 76.44–89.44% in the blocking group (
Figure 3,
p < 0.01). These results suggested that the tumor uptake of [
99mTc]Tc-HYNIC-WL12-tricine/M was PD-L1-specific and associated with the expression level of PD-L1.
Despite their high affinity and specificity for PD-L1, the four
99mTc-labeled HYNIC-WL12 radiotracers displayed significantly different pharmacokinetic properties due to their different coligands (
Table 3,
Table 4,
Table 5 and
Table 6). Among them, [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS exhibited the highest and most increased uptake in MC38-B7H1 tumors (11.81 ± 1.53%ID/g at 0.5 h p.i. and 18.22 ± 4.57%ID/g at 2 h p.i.). The tumor uptake of [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC and [
99mTc]Tc-HYNIC-WL12-tricine/4-PSA were comparable (7.93 ± 1.19%ID/g vs. 8.22 ± 1.92%ID/g at 0.5 h p.i., respectively), and over 80% of the radioactive accumulation in the MC38-B7H1 tumors still remained at 2 h post-injection. The tumor uptake of [
99mTc]Tc-HYNIC-WL12-tricine/PDA was the lowest (5.02 ± 1.61%ID/g at 0.5 h p.i. and 4.61 ± 1.32%ID/g at 2 h p.i.). As shown in
Table 7, all four radiotracers displayed the highest initial radioactivity accumulation in the kidneys. However, their kidney clearance rates were significantly different and decreased in the following order: [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC > [
99mTc]Tc-HYNIC-WL12-tricine/4-PSA > [
99mTc]Tc-HYNIC-WL12-tricine/PDA > [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS. After 4 h p.i., the kidney uptake of [
99mTc]Tc-HYNIC-WL12-tricine/M (M = TPPTS, PDA, ISONIC or 4-PSA) was 154.67 ± 21.76, 31.75 ± 5.16, 5.34 ± 0.73, and 10.80 ± 2.14%ID/g, respectively. The kidney uptake of [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS was 28.96-fold higher than that of [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC at 4 h p.i. In addition, the liver uptake of [
99mTc]Tc-HYNIC-WL12-tricine/M (M = TPPTS, PDA, ISONIC, or 4-PSA) was 12.00 ± 1.11, 6.45 ± 2.70, 2.54 ± 0.70, and 23.44 ± 1.44%ID/g at 2 h p.i., respectively, in the following order of coligand M: 4-PSA > TPPTS > PDA > ISONIC. Compared to [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS and [
99mTc]Tc-HYNIC-WL12-tricine/4-PSA, in which the coligand (M = TPPTS or 4-PSA) contains sulfonic acid groups, [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC and [
99mTc]Tc-HYNIC-WL12-tricine/PDA (containing 1–2 carboxyl groups in the coligand ISONIC or PDA) exhibited faster clearance in the liver, as well as other nontarget tissues, such as the lungs, spleen, muscle, and blood. Although its tumor uptake was lower than that of [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS, [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC exhibited the highest tumor-to-background ratios (
Figure 6) due to its faster clearance in nontarget tissues.
The effect of coligands on the pharmacokinetic properties of
99mTc-labeled HYNIC-conjugated biomolecules has been discussed in the literature. One popular explanation is related to the influence of different coligands on the lipophilicity of
99mTc-HYNIC complexes [
55,
56]. Generally, the increased hydrophilicity of radiotracers often leads to increased urinary excretion. In this study, the log
D values of [
99mTc]Tc-HYNIC-WL12-tricine/M (M = TPPTS, ISONIC, PDA, and 4-PSA) were −1.71 ± 0.09, −0.48 ± 0.04, −0.39 ± 0.03, and −0.23 ± 0.03, respectively. [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS, which has the highest hydrophilicity, displayed the highest kidney uptake. However, this hypothesis does not explain why [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS and [
99mTc]Tc-HYNIC-WL12-tricine/4-PSA exhibit significantly different hydrophilicities (log
D = −1.71 ± 0.09 vs. −0.23 ± 0.03) but display similar liver uptake (32.09 ± 1.50 and 32.75 ± 2.29%ID/g at 0.5 h p.i., respectively) and slower clearance. An alternative explanation may be related to the effect of the coligands on the stability of the
99mTc-HYNIC complexes [
32,
57,
58]. It was reported that the metabolic stability of [
99mTc(HYNIC tetramer)(tricine)(TPPTS)] was much better than that of [
99mTc(HYNIC tetramer)(tricine)(ISONIC)] and [
99mTc(HYNIC tetramer)(tricine) (PDA)], correlating well with the electron-donating capability of coligands, which decreased as follows: TPPTS > ISONIC > PDA. Compared to [
99mTc(HYNIC tetramer)(tricine)(TPPTS)], [
99mTc(HYNIC tetramer)(tricine)(PDA)] exhibited lower uptake in most organs of interest due to its poor in vivo stability [
57]. In this study, compared to [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC, [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS had greater tumor uptake and slower clearance rates in nontarget tissues, such as the kidneys, liver, and blood (
t1/2α = 8.55 min vs. 59.39 min,
Table 9). However, it does not correlate with the fact that both [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS and [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC displayed good in vitro stability and similar metabolic stabilities in tumor, blood, and urine (
Figure 4). Generally, higher hepatic uptake and longer blood retention often indicate greater protein binding of radiotracers [
55,
56]. It was reported that the sulfonic acid could act as an “albumin binder” to increase the protein binding of
64Cu-labeled PD-L1 small molecules [
59]. The coligands TPPTS and 4-PSA both contain a “-SO
3−” group. Therefore, we hypothesized that the higher radioactivity accumulation and retention in nontarget tissues of [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS and [
99mTc]Tc-HYNIC-WL12-tricine/4-PSA might be related to the “-SO
3−” group of coligands.
For PD-L1-targeted diagnostic radiotracers, an important effort is to reduce radioactive accumulation in nontarget tissues, such as the liver, kidneys, and blood, to (1) obtain a high target-to-background ratio for PD-L1-positive tumor imaging and (2) decrease the radiation risk to the main source organs and the whole body of patients. High kidney and liver uptake are common shortcomings of most reported WL12-based radiotracers. For example, almost all
64Cu- and
68Ga-labeled WL12 radiotracers significantly accumulated in the kidneys, probably due to the renal clearance pathway of radiolabeled peptides. High hepatic uptake was observed for
64Cu-WL12 [
39] (24.2 ± 2.5%ID/g at 1 h p.i.) and [
18F]FPy-WL12 [
36] (more than 20%ID/g at 2 h p.i.). As seen in the PET/CT images,
68Ga-NOTA-WL12 and
68Ga-HBED-CC-WL12 accumulated at significantly high levels in the liver [
35,
42]. Rapid clearance from the liver was observed only with DOTAGA-conjugated [
68Ga] WL12 [
38], the liver uptake of which was 15.1 ± 7.6 at 1 h p.i. and 2.7 ± 0.2%ID/g at 2 h p.i., respectively. Compared with
64Cu-WL12 [
34] and [
18F]FPy-WL12 [
36], [
68Ga] WL12 [
38] cleared more quickly from blood and muscle, resulting in greater image contrast.
Table 10 compares the biodistribution data between [
68Ga] WL12 and [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC. Similar to [
68Ga] WL12, [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC showed rapid excretion in nontarget tissues, such as the liver, blood, and muscle. After 2 h p.i., the liver uptake, T/B ratio, and T/M ratio of [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC were comparable to those of [
68Ga] WL12 (liver uptake: 2.54 ± 0.70%ID/g vs. 2.7 ± 0.2%ID/g; T/B: 14.72 ± 2.77 vs. 16.02 ± 3.40; T/M: 40.42 ± 1.59 vs. 100.47 ± 61.23). Differences were observed in kidney uptake, with [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC exhibiting ~50% less kidney uptake than [
68Ga] WL12. The kidney uptake of [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC was 28.84 ± 4.63 and 11.91 ± 2.68%ID/g at 1 h and 2 h p.i., respectively, and for [
68Ga] WL12, it was 64.7 ± 12.1 and 27.67 ± 4.09%ID/g, respectively, at the same time points. The faster clearance of [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC in the hepatobiliary and renal systems may result in a lower radiation risk for patients. As shown in
Table 8, compared to the other [
99mTc]Tc-HYNIC-WL12 radiotracers, [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC exhibited the lowest organ doses for the liver and kidneys. The effective dose of [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC was 2.12 × 10
−3 mSv/MBq, which was much lower than that of
68Ga-NOTA-WL12 [
42] (1.85 × 10
−2 mSv/MBq), a WL12-based PET imaging probe applied for first-in-human evaluation.
Due to faster clearance in nontarget tissues and higher T/B, T/M, and tumor-to-liver ratios, [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC exhibited greater contrast between the tumor and background in SPECT/CT images. As shown in
Figure 5, the MC38-B7H1 tumor (PD-L1-positive) was clearly visualized at 2 h p.i. After coinjection with cold WL12 peptide, the tumor was almost invisible, which was consistent with the trend observed in the biodistribution results, suggesting that [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC is worthy of further preclinical development. However, [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC displayed significantly lower MC38-B7H1 tumor uptake at 2 h p.i. than that of [
99mTc]Tc-HYNIC-WL12-tricine/TPPTS (6.63 ± 0.80%ID/g vs. 18.22 ± 4.57%ID/g). An improvement in the tumor uptake of [
99mTc]Tc-HYNIC-WL12-tricine/ISONIC was warranted. As demonstrated in published literature, the modification of different linkers can improve the tumor uptake of radiotracers to varying degrees [
60,
61,
62]. Additionally, researchers have reported that introducing a PEG3 linker enhances the pharmacokinetic and pharmacodynamic characteristics of Al[
18F]F-labeled NOTA-PCP1 [
63]. In future studies, we will focus on improving the tumor uptake of the
99mTc-radiolabeled HYNIC-WL12 peptide by modifying peptides with different pharmacokinetic (PKM)-modifying linkers.