Evaluation of DNA synthesis with carbon-11-labeled 4′-thiothymidine

Abstract

In the cancer research field, the preferred method for evaluating the proliferative activity of cancer cells in vivo is to measure DNA synthesis rates. The cellular proliferation rate is one of the most important cancer characteristics, and represents the gold standard of pathological diagnosis. Positron emission tomography (PET) has been used to evaluate in vivo DNA synthetic activity through visualization of enhanced nucleoside metabolism. However, methods for the quantitative measurement of DNA synthesis rates have not been fully clarified. Several groups have been engaged in research on 4′-[methyl-¹¹C]-thiothymidine (¹¹C-4DST) in an effort to develop a PET tracer that allows quantitative measurement of in vivo DNA synthesis rates. This mini-review summarizes the results of recent studies of the in vivo measurement of cancer DNA synthesis rates using ¹¹C-4DST.

Key Words: 4′-[methyl-¹¹C]-thiothymidine; DNA synthesis; Cell proliferation; Tumor; Positron emission tomography

Core tip: There is a continuous demand to measure in situ DNA synthesis rates in living human cancer. The thymidine derivative 4′-[methyl-¹¹C] thiothymidine (¹¹C-4DST) has the potential to visualize in vivo DNA synthesis rates with positron emission tomography (PET). To confirm whether ¹¹C-4DST is a valid DNA synthesis marker, clinical and basic research is being conducted at several PET centers in Japan, European Union, and the United States. This mini-review summarizes the progress of recent studies involving the in vivo imaging of cancer DNA synthesis using ¹¹C-4DST PET.

INTRODUCTION

Rationale for DNA synthesis imaging

The basic principle of DNA synthesis measurement using positron emission tomography (PET) and the rationale for tracer development for DNA synthesis imaging were extensively reviewed by Bading et al[1], and Toyohara et al[2], respectively.

Briefly, thymidine is the only nucleoside that is exclusively incorporated into DNA. Therefore, DNA incorporation using [methyl-³H]-thymidine is used as the gold standard for a cell proliferation marker. The challenge in visualizing in vivo DNA synthesis with ¹¹C-thymidine has been pursued since 1972[3]. Extensive developments in the 1990s and early 2000s realized the estimation of thymidine flux from blood to DNA in somatic and brain tumors[4-7]. From these studies, it became clear that the routine use of ¹¹C-thymidine has several limitations, including issues related to the use of radiolabeled catabolites, the short half-life of ¹¹C, and relatively difficult synthesis. While ¹¹C-thymidine is effectively incorporated into DNA, it is also rapidly catabolized by thymidine phosphorylase, which complicates image analysis. Thymidine is a substrate for mitochondrial thymidine kinase 2 as well as cytosolic thymidine kinase 1, which leads to the absence of cell proliferation related uptake in tissues with high mitochondria content, such as the heart. Therefore, the ideal tracer for DNA synthesis imaging requires resistance to catabolism by thymidine phosphorylase, selective phosphorylation by thymidine kinase 1, and ready incorporation into DNA (Figure 1).

Figure 1 Rationale for designing an ideal DNA synthesis tracer for cell proliferation imaging. The ideal tracer must block catabolism by thymidine phosphorylase, while retaining anabolism by the salvage pathway. ENT1: Equilibrative nucleoside transporter 1; TDP: Thymidine di-phosphate; TMP: Thymidine monophosphate; TTP: Thymidine tri-phosphate; TK1: Thymidine kinase 1; TK2: Thymidine kinase 2; TP: Thymidine phosphorylase.

4′-[methyl-¹¹C] thiothymidine

4′-[methyl-¹¹C] thiothymidine (¹¹C-4DST) is a derivative of thymidine in which the 4′-oxygen is replaced with sulfur and the 5-methyl group is radiolabeled with ¹¹C by the C-C cross-coupling reaction (Figure 2)[8,9]. The first synthesis conditions reported involved using 5-tributylstannyl-4′-thio-2′-deoxyuridine (precursor)/tris (dibenzylideneacetone) dipalladium (0) [Pd₂(dba)₃]/tri(o-tryl) phosphine [P(o-CH₃C₆H₄)₃] (1.5:1:3.9 in molar ratio) at 130 °C for 5 min in N,N-dimethylformamide (DMF), which gave the desired products at only 30% decay-corrected yields based on [¹¹C]CH₃I[9]. Subsequently, slight modifications of the conditions, using precursor/Pd₂(dba)₃/P(o-CH₃C₆H₄)₃/CuCl/K₂CO₃ (1.5:1:3.9:4:3.6) at 80 °C, resulted in greatly improved decay-corrected yields (70%) based on [¹¹C]CH₃I[10]. Other reported conditions using precursor/Pd₂(dba)₃/P(o-CH₃C₆H₄)₃/CuCl/K₂CO₃ (25:1:32:2:5) at 80 °C also gave improved decay-corrected yields (42%-60%) using the two-pot method, and 65% by the one-pot method based on ¹¹C-CH₃I[11,12].

Figure 2 Radiosynthesis of 4’-[methyl-¹¹C]-thiothymidine. The imaging agent ¹¹C-4DST was synthesized by methylation of 5-tributylstannyl-4'-thio-2'-deoxyuridine via a palladium-mediated Stille cross-coupling reaction with ¹¹C-methyl iodide. ¹¹C-4DST: 4’-[methyl-¹¹C]-thiothymidine; DMF: N,N-Dimethylformamide; P(o-tolyl)₃: Tri(o-tryl)phosphine; Pd₂(dba)₃: Tris(dibenzylideneacetone)dipalladium(0).

Preclinical evaluation indicated that the stability of ¹¹C-4DST within the body is higher than that of thymidine, and unlike 3′-deoxy-3′-[¹⁸F]fluorothymidine (¹⁸F-FLT), it is taken up into DNA as a substrate for DNA synthesis. Therefore, Toyohara et al[8,9] postulated that ¹¹C-4DST might be used as a valid in vivo DNA synthesis marker. ¹¹C-4DST was developed at the National Institute of Radiological Sciences (NIRS; Chiba, Japan) and has been approved for clinical use by the committee for medical use of cyclotron-produced radiopharmaceuticals of NIRS, as well as the medical use of cyclotron-produced radiopharmaceuticals and ethics committees of Tokyo Metropolitan Institute of Gerontology (TMIG; Tokyo, Japan). The first-in-human study was performed at TMIG in March 2010. ¹¹C-4DST was also described as ¹¹C-S-dThd[8,9], however, the names have since been unified to ¹¹C-4DST with the commencement of its clinical use[10].

INITIAL CLINICAL TRIAL

An initial clinical trial of ¹¹C-4DST, which was equivalent to a phase 1 trial, was performed according to guidelines approved in January 2008 by the institutional committee of TMIG for first-in-human use of novel radiopharmaceuticals. Briefly, five brain tumor patients volunteered to participate in a study of the side effects associated with administration of ¹¹C-4DST, which were assessed by clinical symptoms, physical findings, and blood tests. Concurrently, ¹¹C-4DST dynamic PET measurements were performed to assess the efficacy with which the desired function (DNA synthesis) can be measured. In addition, radiation dosimetry was estimated through whole-body PET measurement in three healthy volunteers.

The results of the initial clinical trial revealed no adverse events accompanying administration of ¹¹C-4DST, and the effective dose was calculated as 4.2 μSv/MBq. Whole-body PET indicated ¹¹C-4DST accumulations in bone marrow and spleen, where DNA synthesis is active in the adult. These observations suggest that ¹¹C-4DST uptake reflects the dynamics of DNA synthesis activity (Figure 3). In addition, physiological accumulation of radioactivity was observed in the liver, kidney, and salivary glands, which are components of the metabolic excretion pathway. The levels of ¹¹C-4DST accumulation in the mediastinum, cerebral parenchyma, lungs, cardiac muscles, and skeletal muscles were very low.

Figure 3 Representative whole-body decay-corrected maximum intensity-projection images of 4’-[methyl-¹¹C]-thiothymidine positron emission tomography images. ¹¹C-4DST showed high uptake in the excretory organs, such as the kidneys, liver, and urinary bladder. Moderate uptake was observed in the proliferative organs, such as bone marrow, spleen, and small intestine. The lowest uptake was observed in the non-proliferating tissues, such as muscle and lungs. This research was originally published in JNM[10]. ¹¹C-4DST: 4’-[methyl-¹¹C]-thiothymidine; SUV: Standardized uptake value.

The distribution of ¹¹C-4DST within the human body was different from that in rodents, and a high level of ¹¹C-4DST accumulation was observed in the human liver. This may have been due to species differences in ¹¹C-4DST metabolism; the metabolism of ¹¹C-4DST in humans is faster than that in rodents and produces hydrophilic metabolites. Treatment of the main component of the hydrophilic metabolites with β-glucuronidase results in the formation of ¹¹C-4DST. Therefore, the high level of ¹¹C-4DST accumulation in the human liver is due to conjugation of ¹¹C-4DST.

¹¹C-4DST is a nucleoside derivative and therefore does not readily pass through the blood-brain barrier. Thus, accumulation of radioactivity in normal brain is low. However, the majority of brain tumor lesions could be imaged, and the accumulation of ¹¹C-4DST in brain tumor lesions varied depending on the treatment conditions. In some cases, the accumulation of ¹¹C-4DST was clearly different from ¹¹C-methionine (¹¹C-MET) images acquired at the same time. The images in Figure 4A shows a case that was resistant to the anticancer agent temozolomide, while the Figure 4B shows a case that was responsive to temozolomide. As can be seen in this figure, ¹¹C-4DST sensitively captures the responsiveness of the tumor cells toward radiotherapy or chemotherapy.

Figure 4 Brain tumor imaging in temozolomide-resistant (A) and -responsive (B) patients. DNA synthesis images provide completely different information from those obtained with an amino acid transport agent. In the case of recurrent anaplastic oligodendroglioma (A), both ¹¹C-MET and ¹¹C-4DST showed high uptake in the gadolinium-enhanced region of the MRI. However, the distribution pattern of each tracer in the tumor region was not identical. The tumor in this patient showed progressive enlargement despite continuous treatment with the DNA alkylating agent temozolomide. In the case of recurrent anaplastic astrocytoma (B), the patient received one course of treatment with temozolomide 3 d before PET examinations, and it was found that the ¹¹C-4DST uptake was negligible in the gadolinium-enhanced region where high uptake of ¹¹C-MET was observed. The enhanced tumor mass in this patient remained unchanged over 6 mo after commencement of temozolomide treatment. These observations suggest that DNA synthesis in the tumor was suspended by temozolomide treatment. Reprinted from Nariai et al[14], with the permission of Springer. ¹¹C-4DST: 4’-[methyl-¹¹C]-thiothymidine; ¹¹C-MET: ¹¹C-Methionine; Gd-MRI: Gadolinium-enhanced magnetic resonance image; PET: Positron emission tomography.

KINETIC ANALYSIS

Time-activity curves (TACs) of ¹¹C-4DST accumulation in brain tumors indicated irreversible kinetics, suggesting that ¹¹C-4DST might be metabolically trapped. In addition, Patlak graph analysis showed a linear plot, which supports the above suggestion[8,9]. The results of the preliminary kinetic analysis indicated that ¹¹C-4DST was irreversibly taken up into the DNA because ¹¹C-4DST fits the two-tissue three-compartment model, and k₄ is negligibly small. The value of k₃ (k₄ = 0) obtained by the analysis showed a strong correlation with K_i (k₄ = 0), which reflects the rate of ¹¹C-4DST incorporation flux (Pearson’s r = 0.925, P = 0.001), and no correlation was observed with K₁ (k₄ = 0), which reflects intracellular transport. In addition, the standardized uptake value (SUV) showed the best correlation with K_i (Patlak) and also showed good correlations with K_i (k₄ = 0) and k₃ (k₄ = 0) (Pearson’s r = 0.942, P = 0.0005; r = 0.857, P = 0.0065, respectively). The above results suggest that images of DNA synthesis might be obtained through non-quantitative analysis using SUV images.

In addition, we performed a basic validation study of ¹¹C-4DST accumulation kinetics into the DNA. The time course of the DNA uptake ratio in proliferating tissues in rats was illustrated by a Michaelis-Menten-type hyperbolic curve, and 50% of the radioactivity was taken into DNA at 5 and 8 min after administration in the duodenum and spleen, with maximum uptake ratios of 99% and 94%, respectively. The TACs of these tissues as a whole were almost identical to the TACs of DNA incorporated radioactivity (DNA fraction) and were different from the soluble non-DNA fraction (unchanged form + phosphorylated form) (Figure 5). On the other hand, in the AH109A rat liver cancer transplantation model, significant variation in DNA uptake rates between sampling sites was due to heterogeneity in the tumor tissues, with some regions exhibiting 60% DNA uptake at 1 min after administration (where cell proliferation is active). The above results indicate that the process of DNA uptake is the rate-limiting step in ¹¹C-4DST accumulation.

Figure 5 Time courses of 4’-[methyl-¹¹C]-thiothymidine activity in the mouse duodenum. The data is presented as the SUVmean in the indicated fractions (n = 3-5). Rapid DNA separation was performed using the DNAzol reagent[30]. The increase in total tissue radioactivity paralleled that of DNA incorporated radioactivity. The soluble fraction (4DST itself, phosphorylated 4DST, and other metabolites) was low and stable at 30 min after tracer injection. The contribution of the DNA incorporation process overshadows both the phosphorylation and dephosphorylation rates at later phases of the PET scan. PET: Positron emission tomography; SUV: Standardized uptake value.

Furthermore, Plotnik et al[13] evaluated the kinetics of ³H-4DST transport and metabolism in the human adenocarcinoma cell line A549 under exponential-growth conditions. ³H-4DST behaved qualitatively similar to endogenous thymidine in terms of equilibrative nucleoside transporter dependent cellular transport, shapes of cellular uptake curves, and relative DNA incorporation levels. As 4DST closely mimics thymidine metabolism, ¹¹C-4DST should provide a robust measurement of DNA synthesis. However, overall ³H-4DST metabolism was significantly lower than that of thymidine, which may reflect its lower affinity toward thymidine kinase 1. This slower metabolism might limit its usefulness because of the relatively short half-life of the ¹¹C label, especially in the case of slow-growing tumors like prostate cancer.

CLINICAL STUDY IN VARIOUS TUMOR TYPES

Although the presence of metabolites in blood was not desirable in terms of quantitative measurement, the results of the ¹¹C-4DST early clinical trials indicated that its usability exceeded the drawbacks. In addition, as high usability of ¹¹C-4DST was expected from the basic data obtained in animal experiments[8,9], collaborative research was initiated with the National Center for Global Health and Medicine (NCGM; Tokyo, Japan) and Kagawa University (Kagawa, Japan). In this collaborative research, technology transfer of radiosynthesis was first performed and then an application was submitted to the ethics committee. The clinical trial was started at approximately the same time as the early clinical trial performed at TMIG. A study to fundamentally support the clinical data is also being conducted in collaboration with the University of Groningen Medical Center (UMCG; Groningen, The Netherlands) and the University of Washington (Seattle, WA).

BRAIN TUMORS

Initial clinical studies of ¹¹C-4DST in brain tumors were performed at TMIG[14]. Fourteen patients with brain tumors (11 malignant gliomas, 2 metastatic tumors, 1 craniopharyngioma, and 1 malignant lymphoma) were included and the uptake of ¹¹C-4DST and ¹¹C-MET into gadolinium-enhanced lesions on T1 weighted magnetic resonance image (MRI) was evaluated. In this study, Nariai et al[14] observed a unique characteristic of ¹¹C-4DST in one case. They observed a marked decrease of ¹¹C-4DST uptake into metastatic lung cancer 5 d after gamma knife therapy. Although ring form enhancement with cystic formation did not disappear completely, tumor growth ceased with this treatment and the patient has maintained favorable activity of daily living 1 year after treatment (Figure 6). In another case, ¹¹C-4DST could differentiate oligodendroglioma from malignant transformation (Figure 7). However, ¹¹C-MET showed increased uptake, which indicates malignant transformation. An increased microvessel surface of oligodendroglioma causes the increased uptake of ¹¹C-MET despite low proliferation of the tumor cells. This case study suggests that ¹¹C-4DST accumulates in growing tumors but not in tumors stabilized by treatment. ¹¹C-4DST uptake into tumors was not influenced by increased transport from blood to tissue, as observed for ¹¹C-MET.

Figure 6 Early evaluation of the effectiveness of radiosurgery with 4’-[methyl-¹¹C]-thiothymidine positron emission tomography. A marked decrease in ¹¹C-4DST uptake into metastatic lung cancer was observed 5 d after gamma knife radiosurgery. Although ring form enhancement with cystic formation did not disappear completely, tumor growth was halted by this treatment and the patient has maintained favorable activity of daily living 1 year after treatment. These observations indicated that the effectiveness of gamma knife therapy was correctly monitored by ¹¹C-4DST only 5 d after treatment. Reprinted from Nariai et al[14], with the permission of Springer. ¹¹C-4DST: 4’-[methyl-¹¹C]-thiothymidine; Gd-MRI: Gadolinium-enhanced magnetic resonance image.

Figure 7 4’-[methyl-¹¹C]-thiothymidine uptake in oligodendroglioma. A patient was diagnosed with grade 2 astrocytoma during the initial operation. At the 1-year follow-up, an enhanced lesion appeared on MRI and ¹¹C-MET uptake increased around the initially resected lesion. These findings suggested that the malignant transformation of astrocytoma occurred. However, ¹¹C-4DST uptake into the tumor was not different from basal brain levels. The pathological diagnosis at the second operation was grade 2 oligodendrocytoma with a newly appearing oligodendroglioma component, but without malignant transformation. Reprinted from Nariai et al[14], with the permission of Springer. ¹¹C-4DST: 4’-[methyl-¹¹C]-thiothymidine; ¹¹C-MET: ¹¹C-Methionine; Gd-MRI: Gadolinium-enhanced magnetic resonance image.

Further in-depth research of ¹¹C-4DST in brain tumors was conducted at Kagawa University. Toyota et al[15] directly compared ¹¹C-4DST and ¹⁸F-FLT in the same subjects. Twenty patients with primary (n = 9) and recurrent (n = 11) gliomas underwent ¹¹C-4DST and ¹⁸F-FLT PET/CT scans. In the normal brain, ¹¹C-4DST uptake was significantly higher than ¹⁸F-FLT (SUVmean; 0.34 ± 0.06 vs 0.19 ± 0.04, P < 0.001 by paired t-test). Individual ¹¹C-4DST SUVmax in the tumor was very similar to ¹⁸F-FLT. Therefore, the average tumor-to-normal tissue uptake (T/N) ratio of ¹⁸F-FLT in the tumor was significantly higher than that of ¹¹C-4DST (10.55 ± 5.45 vs 5.96 ± 3.86, P < 0.001 by paired t test), resulting in better tumor visualization with ¹⁸F-FLT. Both of these tracers did not show significant differences in T/N ratio among different glioma grades. Linear regression analysis showed a significant correlation between proliferative activity indicated by the Ki-67 labeling index and the T/N ratios of ¹¹C-4DST (r = 0.50, P < 0.05) and ¹⁸F-FLT (r = 0.55, P < 0.05). A highly significant correlation was observed between the individual T/N ratios of ¹¹C-4DST and ¹⁸F-FLT (r = 0.79, P = 0.0001). These results indicate that the uptake pattern and uptake values of ¹¹C-4DST in gliomas are similar to those of ¹⁸F-FLT. Two exceptions were observed, one non-enhanced primary diffuse astrocytoma and one recurrent glioblastoma with an oligodendroglioma component, in which only ¹⁸F-FLT could detect the tumor well, with ¹¹C-4DST showing no and faint uptake in the tumor, respectively.

In a follow-up study, Tanaka et al[16] retrospectively evaluated ¹¹C-4DST uptake in 23 patients with newly diagnosed gliomas, and correlated the results with the Ki-67 index and tumor grade in comparison with ¹¹C-MET[16]. ¹¹C-4DST PET/CT showed a slightly lower detection rate for gliomas than ¹¹C-MET (87% vs 96%); however, the difference was not statistically significant. The tracer uptake for normal brain tissue of ¹¹C-4DST was significantly lower than that of ¹¹C-MET (0.48 ± 0.19 vs 1.52 ± 0.36, P < 0.001). The tracer uptake of ¹¹C-4DST in the tumor was also significantly lower than that of ¹¹C-MET (2.14 ± 1.58 vs 5.39 ± 2.22, P < 0.001). Therefore, no significant difference in T/N ratio or metabolic tumor volume (MTV: volume with a threshold of 40% of SUVmax) was observed between ¹¹C-4DST and ¹¹C-MET. A weak correlation was observed between ¹¹C-4DST and Ki-67 index for SUVmax (r = 0.46, P < 0.03), for T/N ratio (r = 0.43, P < 0.05), and for MTV (r = 0.68, P < 0.001) and between ¹¹C-MET MTV and Ki-67 index (r = 0.43, P < 0.04). Among them, the correlation coefficient between ¹¹C-4DST MTV and Ki-67 index was the highest. There was a significant difference in SUVmax of ¹¹C-4DST between grades II and IV (P < 0.03) and in MTV between grades II and IV (P < 0.0009) and grades III and IV (P < 0.02).

HEAD AND NECK CANCER

Ito et al[17] prospectively compared the diagnostic value of ¹¹C-4DST PET/CT and ¹⁸F-FDG PET/CT in patients with head and neck squamous cell carcinoma (HNSCC)[17]. Thirty-eight patients with advanced HNSCC underwent ¹¹C-4DST PET/CT and ¹⁸F-FDG PET/CT before treatment. All patients were followed for 13.5 ± 7.5 mo to monitor recurrence. The total lesion glycolysis (TLG) for ¹⁸F-FDG and total lesion proliferation (TLP) for ¹¹C-4DST were used as outcome measures of the combination of functional information and volumetric data to predict patient prognosis[18-21]. Nine of the 38 patients with post-treatment recurrence were identified. Receiver operating characteristic curves for TLG_3.0 (sensitivity: 89%; specificity: 72%) and TLP_2.5 (sensitivity: 89%; specificity: 55%) showed the highest prognostic ability for recurrence. There was no significant correlation between the Ki-67 index and whether ¹⁸F-FDG SUVmax (P = 0.81) or ¹¹C-4DST SUVmax (P = 0.49) was observed in the primary tumor. One possible reason for this finding may be that the Ki-67 index was obtained mainly from biopsy specimens, not from resected specimens.

LUNG CANCER

At NCGM, the first study focused mainly on lung cancer cases[22]. The subjects were 18 primary lung cancer patients (19 lesions) who underwent segmental resection and lymph node dissection. ¹⁸F-FDG PET/CT and ¹¹C-4DST PET/CT were performed in these patients during the same period and SUVmax was measured at the lesioned region in each patient. The results and histopathological evaluation (e.g., Ki-67 index) were compared. The observed histological types included 16 adenocarcinomas (including bronchioloalveolar carcinoma), 2 squamous cell carcinomas, and 1 large cell carcinoma, and the average tumor size was 27.2 mm. The average levels of SUVmax in the primary tumors were 2.9 ± 1.0 and 6.2 ± 4.5 for ¹¹C-4DST and ¹⁸F-FDG, respectively; SUVmax of ¹¹C-4DST was approximately half that of ¹⁸F-FDG. Figure 8 shows representative images of ¹¹C-4DST in patients with lung adenocarcinoma. ¹¹C-4DST clearly visualized hilar lymph node metastasis because of low physiologic accumulation in the mediastinum blood pool. The correlation coefficients between SUVmax and Ki-67 index were 0.81 and 0.71 for ¹¹C-4DST and ¹⁸F-FDG, respectively, and that of ¹¹C-4DST was significantly higher. ¹¹C-4DST appears to be effective as a PET tracer in primary lung cancer indicating DNA synthetic activity (cell proliferation activity).

Figure 8 4’-[methyl-¹¹C]-thiothymidine images in non-small cell lung cancer. Axial images of CT (A), ¹¹C-4DST PET (B), and ¹⁸F-FDG PET (C) in a 58-year-old man with lung adenocarcinoma in the right lower lobe. Radioactivity of ¹¹C-4DST in the ascending aorta (representing blood pool) is lower than that of ¹⁸F-FDG. Both ¹¹C-4DST and ¹⁸F-FDG clearly visualize lung lesions. Right hilar lymph node metastasis is confirmed on CT (D), and uptake of both ¹¹C-4DST PET (E) and ¹⁸F-FDG PET (F) identifies the lesion (arrow). However, ¹¹C-4DST images are clearer than ¹⁸F-FDG images because of low physiologic ¹¹C-4DST in the mediastinum (blood pool). This research was originally published in JNM[22]. ¹¹C-4DST: 4’-[methyl-¹¹C]-thiothymidine; PET: Positron emission tomography; ¹⁸F-FDG: 2-Deoxy-2-[¹⁸F]fluoro-D-glucose; CT: Computed tomography.

The diagnostic ability of ¹¹C-4DST for detecting regional lymph node metastasis of non-small cell lung cancer (NSCLC) was prospectively evaluated by the same group (NCGM)[23]. A total of 31 patients with NSCLC underwent ¹¹C-4DST PET/CT and ¹⁸F-FDG PET/CT. Patients were followed for up to 2 years to assess disease-free survival. A total of 123 nodal groups were defined for 27 patients, with proven malignancy in 17 nodal groups of 9 patients. The sensitivity on a per-node basis was significantly higher with ¹¹C-4DST than with ¹⁸F-FDG (82.5% vs 29.4%, P < 0.002). In contrast, the specificity on a per-node basis was significantly lower with ¹¹C-4DST than with ¹⁸F-FDG (71.7% vs 85.8%, P < 0.02). The disease-free survival rate with positive ¹¹C-4DST uptake in nodal lesions was 0.35, which was considerably lower than the rate of 0.83 with negative findings (P = 0.04). Interestingly, nodal staging by ¹¹C-4DST was the most influential prognostic factor (P = 0.05) among the factors tested. The high sensitivity of ¹¹C-4DST for nodal metastasis indicates that it may be efficient in detecting lymph node micrometastasis.

MULTIPLE MYELOMA

In whole-body imaging diagnosis of multiple myeloma (MM), it was demonstrated that myeloma has various characteristics that become prominent at different stages, including energy metabolism (¹⁸F-FDG), production of M protein (¹¹C-MET), and cell proliferation (¹¹C-4DST). We hope to be able to accurately determine the periods where follow-up observation is sufficient and aggressive chemotherapy is necessary, to optimize treatment in such cases. Okasaki et al[24] prospectively evaluated the possibility of ¹¹C-MET and ¹¹C-4DST whole-body PET/CT when searching for bone marrow involvement in patients with MM, in comparison with those for ¹⁸F-FDG PET/CT and aspiration cytology[24]. A total of 64 patients with MM or monoclonal gammopathy of undetermined significance (MUGS) underwent three whole-body PET/CT scans with ¹⁸F-FDG, ¹¹C-MET, and ¹¹C-4DST within a period of 1 wk. The tracer accumulation was evaluated as positive, equivocal, or negative for 55 lytic lesions visualized using CT in 24 patients. To verify tracer uptake by lesions, 36 patients underwent bone aspiration cytology within 1 wk of the three PET/CT scans. The SUVmax of ¹¹C-4DST in lytic lesions were highest among the three tracers. ¹¹C-4DST and ¹¹C-MET provided clearer findings than ¹⁸F-FDG for lytic lesions. A typical MM patient with multiple active lesions is shown in Figure 9. ¹¹C-MET and ¹¹C-4DST detected positive lesions, whereas ¹⁸F-FDG detected an equivocal lesion. ¹⁸F-FDG was rarely able to detect skull lesions because of the high physiological accumulation in the brain, whereas ¹¹C-MET and ¹¹C-4DST were capable of clearly detecting skull lesions because of their low accumulation in the brain. Furthermore, ¹¹C-4DST and ¹¹C-MET had higher diagnostic accuracies than that of ¹⁸F-FDG, when compared with iliac crest biopsy. However, no significant difference was observed between the ¹¹C-MET and ¹¹C-4DST findings.

Figure 9 4’-[methyl-¹¹C]-thiothymidine images in multiple myeloma. ¹⁸F-FDG (A), ¹¹C-MET (B), and ¹¹C-4DST (C) PET images obtained in a 63-year-old man with multiple myeloma. Numerous active lesions are visible in the three maximum intensity projection images. The fusion images are for the cross-section at the level indicated by the red lines (D). The lesion in the right ischium (red arrow) was positive in all three PET scans. However, the lesion in the right pubis (white arrow) was only positive on the ¹¹C-MET and ¹¹C-4DST PET scans and was equivocal on the ¹⁸F-FDG PET scan. Reprinted from Okasaki et al[24], with the permission of Springer. ¹¹C-4DST: 4’-[methyl-¹¹C]-thiothymidine; ¹¹C-MET: ¹¹C-Methionine; ¹⁸F-FDG: 2-Deoxy-2-[¹⁸F]fluoro-D-glucose; PET: Positron emission tomography.

RENAL CELL CANCER

Minamimoto et al[25] evaluated the potential of ¹¹C-4DST PET/CT for imaging cellular proliferation in advanced clear cell renal cell cancer (RCC), compared with ¹⁸F-FDG PET/CT[25]. Five patients with a single RCC lesion were examined. The typical tumor accumulation pattern of ¹⁸F-FDG was diffuse or with a thick uptake layer, in contrast to outer surface-dominant uptake with ¹¹C-4DST (Figure 10). The SUVmax of ¹¹C-4DST (7.3 ± 12.2) was slightly higher than that of ¹⁸F-FDG (6.0 ± 12.8). The correlation between SUVmax and Ki-67 index was higher with ¹¹C-4DST (r = 0.61) than with ¹⁸F-FDG (r = 0.43). Tumor uptake of both tracers was correlated with the Fuhrman nuclear grading system. Interestingly, the ¹¹C-4DST uptake increased as the degree of pathological phosphorylation of mammalian target of rapamycin (pmTOR) increased. This might indicate that ¹¹C-4DST has potential for use in evaluating the therapeutic effect of mTOR inhibitors in patients with RCC.

Figure 10 4’-[methyl-¹¹C]-thiothymidine images in clear cell renal carcinoma. A 78-year-old male with a clear cell renal carcinoma (RCC) in the right kidney. Contrast-enhanced abdominal CT (A), fused ¹¹C-4DST PET/CT (axial image) (B), fused ¹⁸F-FDG PET/CT (axial image) (C), coronal ¹¹C-4DST PET (D), and coronal ¹⁸F-FDG PET (E) images are shown. The uptake of ¹¹C-4DST was predominantly near the outer surface of the RCC tumor; the uptake of ¹⁸F-FDG was more diffuse. Reprinted from Minamimoto et al[25], with the permission of Springer. ¹¹C-4DST: 4’-[methyl-¹¹C]-thiothymidine; ¹⁸F-FDG: 2-Deoxy-2-[¹⁸F]fluoro-D-glucose; PET: Positron emission tomography; CT: Computed tomography.

4DST IN BASIC RESEARCH

In cancer, differential diagnosis of benign vs malignant lesions is important. Toyohara et al[26] investigated whether 4DST is useful for differentially diagnosing tumors and inflammation in animal models[26]. C6 glioma cells were transplanted into the right shoulder of Wistar rats, and turpentine was injected into the left hind leg 9 d later to create a tumor and acute inflammation model. Dynamic ¹¹C-4DST PET scan was performed one day after turpentine injection. The biodistribution of ¹¹C-4DST was determined and compared with those of ¹⁸F-FLT, ¹⁸F-FDG, ¹¹C-choline, ¹¹C-MET, and two sigma receptor ligands 1-4-2′-¹⁸F-fluoroethoxy-3-methoxyphenethyl-4-3-4-fluorophenyl-propyl-piperazine (¹⁸F-FE-SA5845) and 1-3,4-¹¹C-dimethoxyphenylethyl-4-3-phenylpropyl-piperazine (¹¹C-SA4503). The results showed that ¹¹C-4DST had the highest tumor uptake among the tracers examined (SUV = 4.9), and the tumor muscle ratio was 13, which is equivalent to that of ¹⁸F-FDG. ¹¹C-4DST showed an extremely high tumor inflammation ratio of 49, which was comparable to that of ¹⁸F-FE-SA5845, and was more than 10-fold higher than that of ¹⁸F-FDG. ¹¹C-4DST showed a persistent increase in accumulation in the tumor and bone marrow. On the other hand, it was not retained in inflammatory tissues, and returned to the same level as in muscles 40 min after injection. As indicated here, ¹¹C-4DST is a useful PET tracer with high sensitivity and specificity, with completely different dynamics in tumors and inflammatory tissue. The above results strongly suggest that ¹¹C-4DST will be effective in clinical settings.

In a follow-up study, Toyohara et al[27] evaluated longitudinal changes in ¹¹C-4DST uptake in turpentine-induced acute, subacute, and chronic phases of inflammatory tissues[27]. They found that ¹¹C-4DST uptake in inflammatory tissue was transiently increased during the subacute phase and subsequently decreased to normal levels. The transient increase in ¹¹C-4DST uptake paralleled changes in the Ki-67 labeling index in inflammatory tissues. These data indicate that avoiding the subacute phase is required for proper evaluation of tumor responses using ¹¹C-4DST PET.

Tsuji et al[28] established a clinically relevant mouse model of mesothelioma[28] and applied ¹¹C-4DST as a tool to differentiate histological subtypes. In their mouse model, ¹¹C-4DST was suitable for imaging the epithelioid subtype, while ¹⁸F-FDG was more suitable for the sarcomatoid subtype.

Hasegawa et al[29] evaluated radiation induced carcinogenesis and its suppression by ¹¹C-4DST PET in a radiation-induced thymic lymphoma model[29]. They found initial higher ¹¹C-4DST uptake in irradiated thymus at 1 wk after fractionated whole-body X-ray irradiation. The increased ¹¹C-4DST uptake in the thymus was completely abolished by bone marrow transplantation. This study demonstrates the feasibility of cellular proliferation imaging with ¹¹C-4DST for the noninvasive monitoring of tumorigenic processes in animal models of radiation-induced cancer.

CONCLUSION

¹¹C-4DST is a newly developed PET tracer that can be used to image DNA synthesis, and its application in clinical settings is being studied at several facilities. In the future, ¹¹C-4DST will be applied to more types of cancer, which will further establish the significance of ¹¹C-4DST PET diagnoses through the accumulation of data on the physiological accumulation in lesions.

ACKNOWLEDGMENTS

The author thanks Mr. Kunpei Hayashi for the production of ¹¹C-4DST. Figures 4, 6 and 7 were reprinted from Nariai et al[14], with the permission of Springer. Figures 3 and 8 were reprinted from Toyohara et al[10] and Minamimoto et al[22], respectively, with the permission of the Society of Nuclear Medicine and Molecular Imaging, Inc. Figures 9 and 10 were reprinted from Okasaki et al[24] and Minamimoto et al[25], respectively, with the permission of Springer.