Nuclear Oncology, 1 Ed.

CHAPTER 41

RADIONUCLIDE IMAGING OF ANNEXINS

Christel Vangestel • Leonie Wyffels • Marc Peeters • Steven Staelens • Sigrid Stroobants

INTRODUCTION

Cancer, one of the world’s leading causes of death, still lacks validated biomarkers for early detection of treatment response. To be able to provide “personalized medicine”: To identify the drug that works best in a specific patient at a specific time, in order to provide patient-specific tailored therapy to potentially improve treatment success, more rapid and reliable markers of treatment efficacy are needed. In oncology, nuclear imaging of programmed cell death (PCD) or apoptosis using radiolabeled Annexin A5 may help to evaluate the efficacy of anticancer therapy shortly after the start of the therapy.

Criteria used to define objective remission, stabilization, or progression of human malignancies following chemotherapy and/or radiotherapy are based on morphologic imaging and hence on computed tomography (CT) or magnetic resonance imaging (MRI)-based volumetric changes with a significant reduction in volume usually ensuing at the earliest 6 to 12 weeks following effective treatment. During this period, nonresponders experience unnecessary toxic side effects related to the treatment without other potential beneficial treatment. Accordingly, methods that allow for a more rapid assessment of response to chemotherapy and/or radiotherapy in cancer patients are needed.

Imaging extracellular phosphatidylserine (PS), a marker of PCD, with radiolabeled Annexin A5 is an ideal method to monitor tumor response to radiotherapy and chemotherapy. As cells are undergoing apoptosis, PS, a lipid normally facing the cytoplasm, flips and faces the extracellular side of the cell membrane. The protein Annexin A5 binds in the presence of calcium ions with high affinity and high specificity to PS. PS is a very attractive target because PS-expression is a near-universal event that occurs within a few hours of the cell death stimulus; it represents a very abundant target that is readily accessible on the extracellular face of the cell membrane. Available preclinical and clinical data support the notion that many of the effects of radiotherapy and chemotherapy are mediated by rapid induction of apoptosis, peaking within the first 24 hours following treatment administration. Many forms of PCD, including apoptosis, necrosis, mitotic catastrophe, or autophagic cell death are accompanied by the surface expression of PS. Recently, the role of other forms of PCD (e.g., autophagy, mitotic catastrophe) in determining the overall tumor response to anticancer therapy has become more and more highlighted. Thus, a noninvasive imaging technique to monitor baseline and therapy-induced PCD in real time could allow for assessment of response to chemotherapy and radiotherapy as early as 24 to 48 hours following treatment. The advantages of imaging with radiotracers not only include the ability to detect tumor response much earlier, but also the noninvasive nature of nuclear imaging techniques and the fact that different parameters can be imaged with different probes. Numerous radiolabeled Annexin A5 variants are in various stages of preclinical development. Some have already been applied in clinical trials. Hopefully, some will be approved for routine clinical use in the near future.

APOPTOSIS AND ITS ROLE IN CANCER

Apoptosis, a form of PCD, allows the organism to tightly control cell numbers and tissue size, maintaining overall homeostasis. It is involved in many physiologic processes, for example, development, proliferation/homeostasis, differentiation, regulation, and function of the immune system. Apoptosis is a highly regulated process that allows a cell to self-degrade in order for the body to eliminate unwanted or dysfunctional cells. Damaged or unneeded cells are removed by neighboring cells without causing inflammation from leakage of the cell content. Defects in apoptotic cell death regulation, leading to an excessive or deficient cell death, contribute to a variety of pathologic conditions including cancer, neurodegenerative disorders, cardiovascular disease, and autoimmune disease.1 Cancer can occur when the balance between cell proliferation and apoptosis is disturbed, either by an increase in cell proliferation or a decrease in tumor cell apoptosis. The goal of anticancer treatment is to induce cell death in tumor cells leaving healthy cells undamaged.2 Although apoptosis is a key factor influencing therapeutic outcome, tumor cells can also die by nonapoptotic mechanisms including autophagy, mitotic catastrophe, and necrosis, influencing the overall tumor response to anticancer therapy. Therapeutically induced apoptotic as well as nonapoptotic cell death enhances the effect of anticancer treatment.3 The morphologic characteristics of the apoptotic cell include chromatin condensation, nuclear fragmentation (internucleosomal DNA cleavage), plasma membrane blebbing, and cell shrinkage. Ultimately, the cell breaks into small membrane-surrounded structures, called apoptotic bodies. The externalization of the phospholipid, PS (normally found on the inner cell membrane leaflet to the exterior cell membrane leaflet) is one of the first events of apoptosis and serves as a signal for the phagocytes to recognize and engulf the apoptotic bodies, without causing an inflammatory response.4 This is in sharp contrast to necrotic cell death in which cells release their cellular content in the cell’s environment resulting in damage to neighboring cells followed by a strong inflammatory response in the surrounding tissue.5 In comparison to apoptosis, necrotic cells die accidentally in response to an acute injury. Cells undergoing necrosis have PS on the inside of their cell membranes, which may be as well a target for phagocytosis.6

Two main apoptotic pathways have emerged: The mitochondrial pathway (intrinsic pathway) and the death receptor pathway (extrinsic pathway). The intrinsic pathway is mediated by the Bcl-2 superfamily members that interact with the mitochondrial membrane, whereas the extrinsic pathway is governed by the tumor necrosis factor (TNF) superfamily of ligands and receptors, interacting at the cytoplasmic membrane. Both pathways converge on a common cascade of cysteine aspartate-specific proteases, known as the caspases, which upon activation cross-link and cleave specific intracellular proteins involved with apoptosis (Fig. 41.1).7

THE PS-ANNEXIN A5 INTERACTION: THE HEART OF NUCLEAR IMAGING OF PCD

The phospholipid, PS, is redistributed in the cell membrane as the tumor cells are undergoing PCD. PS is normally restricted to the inner leaflet of the cell lipid bilayer (thus facing the cytosol) by an ATP-dependent enzyme called translocase. Upon activation of PCD, PS becomes exposed rapidly (within minutes) to the outer leaflet of the plasma membrane. This process is facilitated by a calcium-dependent deactivation of translocase and activation of floppase and a third enzyme, scramblase.8 This selective exposure of PS by cells undergoing PCD forms the basis of Annexin A5 binding in vitro and in vivo. Annexin A5 (≈36 kDa) is an endogenous protein that binds in the presence of calcium ions with high affinity (Kd = 10−9 M) and high selectivity to PS. The Annexin family is a multiprotein family of over 160 proteins that share the property of Ca2+-dependent binding to negatively charged phospholipid surfaces. The binding of Annexin A5 to PS requires all four calcium-binding sites and involves the binding of eight PS molecules per molecule Annexin A5 at low levels of membrane occupancy (with respect to protein).9 PS presents a very abundant target (millions of binding sites per cell) that is readily accessible on the extracellular face of the cell membrane. The precise physiologic function of Annexin A5 is uncertain. Although Annexin A5 was first discovered as a protein with anticoagulant properties,10 advantages include no immunogenicity, lack of in vivo toxicity and the possibility of recombinant synthesis in Escherichia coli. When Annexin A5 is labeled, detection of cells undergoing PCD is possible (Fig. 41.2). Depending on the label (fluorescent, radionuclide, MRI contrast agents) this can be done in tissue sections, cell cultures, or in vivo. After the successful use of fluorescent-labeled Annexin A5 in vitro and in vivo, Annexin A5 was radiolabeled with different radionuclides for noninvasive assessment of PCD. From a clinical point of view, radiolabeled Annexin A5 can have many applications including determination of the severity of cardiac ischemia, determination of cardiac allograft rejection in transplant patients, and in neurology to identify patients at risk of undergoing stroke and/or transient ischemic attacks. In oncology, the main applications of radiolabeled Annexin A5 are to image and determine the nature and biology of intracardiac tumors without the need for biopsy and to evaluate early treatment response in solid tumors. Annexin A5 has been labeled with various radionuclides. Clinical trials among patients with various tumor types demonstrated that Annexin A5 nuclear imaging is a promising technique to confirm the onset of PCD induced by chemotherapy and radiotherapy.

FIGURE 41.1. Two main pathways to apoptosis. The extrinsic pathway is governed by the tumor necrosis factor (TNF) superfamily of ligands and receptors, interacting at the cytoplasmatic membrane, whereas the intrinsic pathway is mediated by the Bcl-2 superfamily members that interact with the mitochondrial membrane. Both pathways eventually converge on a common cascade of cysteine aspartate-specific proteases, caspase-3, which upon activation cross-link and cleave specific intracellular proteins involved with apoptosis (FADD, Fas-associated death domain; TRADD, TNFR-associated death domain; DISC, death-inducing signaling complex; cytc, cytochrome c; IAP, inhibitor of apoptosis protein).

FIGURE 41.2. Presentation of the principle underlying cell death imaging with radiolabeled Annexin A5. Living cells actively confine phosphatidylserine (PS) to the inner leaflet of the cell membrane. Upon activation of programmed cell death (PCD), PS becomes actively externalized. Annexin A5, labeled with a radioisotope, binds with high affinity to PS and in vivo radionuclide imaging of PCD is feasible.

ANNEXIN A5 SPECT RADIOTRACERS

Radiolabeling of Annexin A5 with various single photon emitters results in the synthesis of imaging agents for use with single photon emission computed tomography (SPECT). SPECT is a noninvasive imaging modality that uses radionuclides such as Technetium-99m (99mTc) which emits γ-photons (one photon/radioactive decay) that can be detected by a γ-camera. It is the most extensively used SPECT radionuclide. 99mTc has a favorable low γ-energy (140 keV), is low cost (convenient availability from commercial 99Mo/99mTc generator) compared to other radionuclides and easy radiochemistry. The 6-hour half-life is sufficiently long for radiopharmaceutical preparation and accumulation in target tissue in vivo, but yet short enough to minimize the radiation absorbed dose for patients.

99mTc-BTAP-Annexin A5

The first human trials with radiolabeled Annexin A5 were conducted with 99mTc-4,5-bis (thioacetamido)pentanoyl (BTAP or 99mTc-Apomate)-Annexin A5. A biodistribution and dosimetry study in humans demonstrated that radioactivity vigorously accumulated in the kidneys and liver (28 ± 8% and 20 ± 4% of the injected dose [ID] at 70 minutes post injection [p.i.], respectively). The fast appearance of activity in the intestine (27 ± 6%ID, 20 hours p.i.) may interfere with imaging in this region.11 The pharmacokinetics and imaging properties of 99mTc-BTAP-Annexin A5 were compared to those of 99mTc-(n-1-imino-4-mercaptobutyl)-Annexin A5 (99mTc-i-Annexin A5) in healthy volunteers and in patients suffering from different malignancies. 99mTc-BTAP-Annexin A5 proved to be the most stable with good and reproducible labeling characteristics and it had the advantage of faster clearance from the body. However, the uptake of 99mTc-BTAP-Annexin A5 in the kidneys and liver is even slightly higher than the other radiopharmaceuticals. In addition, activity is excreted considerably faster into the gut. The effective dose (ED) of 99mTc-BTAP-Annexin A5 (7.6 ± 0.5 μSv/MBq) was somewhat lower than that for 99mTc-i-Annexin A5 (9.7 ± 1 μSv/MBq).12 Belhocine et al. successfully conducted a study of 15 patients with lung cancer (n = 10), lymphoma (n = 3), and breast cancer (n = 2) before and within 3 days after the first course of chemotherapy. In all cases, no tracer uptake was observed before treatment. However, 24 to 48 hours after the first course of chemotherapy, in seven patients who showed 99mTc-BTAP-Annexin A5 uptake at tumor site, there was a complete (n = 4) or partial (n = 3) response. Six out of eight patients who showed no significant posttreatment tumor uptake had progressive disease. Two patients with metastatic breast cancer had a clinically significant partial response to taxol-based chemotherapy but no radiotracer uptake was observed after treatment. The optimal timing for scintigraphic imaging of apoptosis in this study was about 20 to 24 hours after the second injection (48 hours after chemotherapy). In two patients (one non–small-cell lung cancer [NSCLC] and one small-cell lung cancer [SCLC]) an increase in Annexin A5 uptake was observed 20 to 24 hours after therapy, indicating the significant importance of optimal timing of scanning as these results suggested variability with regard to cancer type and therapy.13,14 However, although 99mTc-BTAP-Annexin A5 may be produced with high radiochemical purity, the labeling method turned out to be very laborious with low radiochemical yields. Because of the slow blood clearance, meaningful planar images of low image quality could only be obtained 24 hours following radiopharmaceutical injection. Given the high radiotracer uptake in kidney, liver, and abdomen, this radiotracer was not used in subsequent clinical trials.

99mTc-EC-Annexin A5

Preclinical Data

Another method to prepare 99mTc-labeled Annexin A5 is the ­conjugation of ethylenedicysteine (EC) to Annexin A5 using sulfo-N-hydroxysuccinimide and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide·HCl (EDC) as coupling agents. The radiochemical yield and purity obtained approached 100%. Biodistribution study demonstrated the kidneys and liver as the main organs of tracer uptake. Breast tumor–bearing rats were treated with paclitaxel to demonstrate in vivo tumor cell PCD because of chemotherapy. Significantly increased tracer uptake in tumors was observed 3 days after paclitaxel treatment. At 5 days, however, post treatment, tumor uptake was slightly lower compared to control animals not receiving therapy.15

Clinical Data

A preliminary clinical study of 10 primary breast cancer patients (stage II–III) before treatment or during the first cycle of induction chemotherapy was performed with 99mTc-EC-Annexin A5 to image tumor cells undergoing PCD. Five patients received induction therapy consisting of paclitaxel in two patients, 5-fluorouracil, doxorubicin, and cyclophosphamide in one patient and bcl-2 antisense oligonucleotide therapy in the remaining two patients, 16 hours before the imaging session. Whole-body planar images were obtained at 0.5, 2 to 4, 18 to 24 hours after injection. Computer outlined regions of interest (ROI) (counts/pixel) were used to determine tumor-to-background count density ratios (T/N). The ratios were used to compare dynamic tumor uptake pre- and posttreatment. In nine patients, detectable 99mTc-EC-Annexin A5 uptake corresponded to the area of palpable invasive disease. In patients receiving induction chemotherapy, the T/N was higher than in those that did not receive chemotherapy (2.6 ± 0.5 versus 1.5 ± 0.2). Patients with a high T/N tended to show a better response. From the clinical image–based dosimetric study, the total effective dose equivalent (EDE) for single-injection dose of 925 to 1,037 MBq of 99mTc-EC-Annexin A5 was 6.80 to 7.89 mSv which is reasonable for clinical use.16 Although this method appears to be efficient and fast to label Annexin A5, more clinical research is warranted.

99mTc-MAG3-Annexin A5

In an effort to decrease the high kidney and liver accumulation observed with 99mTc-hydrazinonicotinamide (HYNIC)-Annexin A5, Annexin A5 was conjugated to mercaptoacetyltriglycine (MAG3). Biodistribution data in normal mice showed a significant decrease in kidney and liver uptake 1 hour p.i. of 99mTc-MAG3-Annexin A5 compared to 99mTc-HYNIC-Annexin A5 (from 24 to 4%ID for the liver, and 45 to 15%ID for the kidneys, respectively). The biodistribution was also characterized by a lower retention of radioactivity in the whole body. However, accumulation in the small intestine was five-fold higher than observed with 99mTc-HYNIC-Annexin A5 at 1 hour p.i.17 This increased uptake in the small intestine is problematic for the in vivo detection of ongoing PCD in the lower abdomen. No additional studies in humans have been performed.

99mTc-HYNIC-Annexin A5

The most widely applied radiolabeled Annexin A5 is 99mTc-HYNIC-Annexin A5 using SPECT imaging. 99mTc-HYNIC-Annexin A5 offers a simple and fast preparation at room temperature, while providing high radiochemical yields (>90%) and purities (>90%). Labeling is a one-step reaction, with 30 minutes of incubation at room temperature, no additional purification is required. All of these parameters together, make this radiotracer very suitable for routine production and clinical applications.

Preclinical Data

Blankenberg et al.18,19 derivatized Annexin A5 with HYNIC and coupled it to 99mTc before intravenous administration in rodents. The first preclinical study evaluated 99mTc-HYNIC-Annexin A5 in two models: Fulminant hepatic apoptosis induced by anti-Fas antibody injection in BALB/c mice and cyclophosphamide treatment of transplanted murine B-cell lymphomas. The treatments resulted in a two- to six-fold increase in the uptake of 99mTc-HYNIC-Annexin A5 at sites of apoptosis compared to the nontreated control animals.18 Biodistribution of 99mTc-HYNIC-Annexin A5 in untreated BALB/c mice demonstrated highest concentration of the radiotracer in the kidneys (52%ID) 1 hour p.i., followed by the liver (12.8%ID). The brain, heart, and thymus had the lowest uptake of radiolabeled Annexin A5 (less than 0.2%ID). One hour after injection, residual blood activity was about 3% of the ID.19 In a similar study, mice were treated with dexamethasone to induce thymic apoptosis and a 40-fold increase in 99mTc-HYNIC-Annexin A5 concentration in apoptotic thymocytes was observed. A correlation of r = 0.78 was found between radioactivity and flow cytometric and histologic evidence of apoptosis.20 To determine the optimal time point for detection of apoptosis in vivo with 99mTc-HYNIC-Annexin A5 after chemotherapy,21 rats inoculated with allogenic hepatoma cells were treated with a single dose of cyclophosphamide and 99mTc-HYNIC-Annexin A5 was injected in the rats 4, 12, and 20 hours after the treatment. Radioactivity in tissue was determined 6 hours after 99mTc-HYNIC-Annexin A5 injection. The accumulation of 99mTc-HYNIC-Annexin A5 in tumors significantly increased at 20 hours, but not at 4 or 12 hours after a single dose of cyclophosphamide, which was consistent with the rate of apoptotic cells determined by terminal deoxynucleotidyl transferase biotin-dUTP nick-end labeling (TUNEL) and caspase-3 immunostaining.22 Two similar preclinical studies using rats bearing allogenic hepatoma cells and treated with chemotherapy, demonstrate the usefulness of 99mTc-HYNIC-Annexin A5 to predict therapy outcomes and for prognosis at early stage of chemotherapy.20,22 In a more recent study, the time course of apoptosis induced by a single dose of paclitaxel in a model of virus-induced murine breast cancer was determined. The biodistribution of 99mTc-HYNIC-Annexin A5 was evaluated at baseline and at 1, 3, 6 and 24 hours after paclitaxel treatment. Apoptotic markers in tumor samples obtained at the same time points were also evaluated, including DNA breaks (TUNEL), active caspase-3, and apoptosis-inducing factor (AIF). Baseline uptake of 99mTc-HYNIC-Annexin A5 in breast tumors was about two-fold higher than uptake in normal breast tissue; tracer uptake increased at 1 and 3 hours after paclitaxel administration and then declined back to baseline. The fraction of TUNEL-positive cells also peaked at 3 hours after treatment, whereas active caspase-3 had a small peak at 1 hour and a much higher peak at 6 hours after paclitaxel administration. AIF did not show any significant positivity. As different markers of apoptosis (PS-expression, caspase-3 activation, AIF activation) occurs at different time points during the apoptotic process, difference in time expressing are to be expected. Although this study confirms that in vivo uptake of 99mTc-HYNIC-Annexin A5 reflects the degree of apoptosis, the study also demonstrates that the response to anticancer therapy may differ from tumor type to tumor type and that the selection of an appropriate interval between the administration of the chemotherapy and the imaging with radiolabeled Annexin A5 is crucial.23 Two recent preclinical studies focused on the feasibility of in vivo 99mTc-HYNIC-Annexin A5 imaging of radiation-induced apoptosis in tumor-bearing mice.24,25 In lymphoma-bearing mice the 99mTc-HYNIC-Annexin A5 uptake significantly increased as the radiation dose was escalated from 0 to 8 Gy, and significantly correlated with the number of TUNEL-positive cells (r = 0.892, p < 0.001). These authors conclude that the early phase apoptosis induced by radiation is dose dependent and that 99mTc-HYNIC-Annexin A5 imaging can reflect the dose–response relationship.25

Methodologic Studies in Patients

Safety, biodistribution, and dosimetry of 99mTc-HYNIC-Annexin A5 in six healthy, male volunteers were determined in a phase I study. Highest uptake was observed in the kidneys (49.7 ± 8.1%ID) 3 hours p.i. Liver (13.1 ± 1.0%ID), the red marrow (9.2 ± 1.8%ID), and spleen (4.6 ± 1.6%ID) also showed considerable uptake whereas the gut showed no uptake even at 24 hours after injection. An example of biodistribution of 99mTc-HYNIC-Annexin A5 is shown in Figure 41.3. More than 90% of blood activity was cleared with a short half-life of 25 minutes. Because of the long biologic half-life, the ED of 99mTc-HYNIC-Annexin A5 is in the high range of values for commonly used 99mTc-radiopharmaceuticals (11 ± 0.8 μSv/MBq).26 Several phase I/II studies subsequently evaluated the potential of 99mTc-HYNIC-Annexin A5 for imaging of apoptosis in human patients. In a pilot study by Vermeersch et al., 18 patients suffering from primary head and neck cancer underwent a spiral CT scan and 99mTc-HYNIC-Annexin A5 scintigraphy within 1 week, followed by resection of the lesion. 99mTc-HYNIC-Annexin A5 imaging allowed visualization of all primary head and neck tumors identified by CT scan. Involved lymph nodes, however, were not identified. The low resolution of the γ-camera and related partial volume effect (lymph nodes in these patients were smaller than 15 mm in size) may explain the failure to detect the lymph nodes. Furthermore, because involved lymph nodes mostly show focal tumor uptake, the uptake of 99mTc-HYNIC-Annexin A5 may have been insufficient to detect in these very small lesions.27 In a comparable protocol, 20 patients with primary or recurrent head and neck carcinomas underwent spiral CT, 99mTc-HYNIC-Annexin A5 tomography, and subsequent surgical resection of the tumor. Quantitative 99mTc-HYNIC-Annexin A5 tumor uptake correlated well with the number of apoptotic cells assessed by TUNEL assays performed on the tumor slides, if only samples with no or minimal amounts of necrosis were considered. Including tumor samples with increasing amounts of necrosis resulted in a progressive decrease of correlation between 99mTc-HYNIC-Annexin A5 uptake and number of apoptotic cells and loss of statistical significance.28 For clinical application, sufficient reproducibility of 99mTc-HYNIC-Annexin A5 uptake has to be demonstrated to allow a study of cell death changes induced by chemotherapy over time and intersubject. Thirteen patients with a clinically suspected and postscintigraphy, histologically confirmed squamous head and neck carcinoma were included in this study. All patients underwent a spiral CT, and 99mTc-HYNIC-Annexin A5 scintigraphy which was performed twice (identified as Day 1 and Day 2) within 3 to 5 days from each other. The percentage uptake of the ID of 99mTc-HYNIC-Annexin A5 in tumor lesion divided by the tumor volume (derived from CT) was determined twice within an interval of 2 weeks by observer 1, and once by observer 2 on Day 1 and once on Day 2 by observer 1. Intraobserver (–3.4%), interobserver (2.4%), and day-to-day (–6%) reproducibility of quantitative 99mTc-HYNIC-Annexin A5 quantitative uptake measurements proved suitable for clinical applications.29 Another clinical finding, reported by the same research group, is that quantitative uptake of 99mTc HYNIC-Annexin A5 in normal tissues such as the kidney, spleen, bone marrow, is not significantly changed with chemotherapy or prior administration of radiolabeled Annexin A5 within a 48-hour period. Eleven cancer patients were investigated. Five cancer patients received no treatment (control group) and six cancer patients received chemotherapy and underwent 99mTc-HYNIC-Annexin A5 scintigraphy (TAVS) pretreatment as well as 5 to 7 and 40 to 44 hours after treatment initiation. In addition, biodistribution of 99mTc HYNIC-Annexin A5 in the whole body, kidneys, and liver remained unchanged after chemotherapy or prior administration of tracer.30

FIGURE 41.3. The biodistribution of 99mTc-HYNIC-Annexin A5 in a healthy human. High uptake in the kidneys, less uptake in the liver and spleen, and low uptake in the bone marrow are demonstrated.

Chemo/Radiotherapy Response Monitoring

Kartachova et al. performed several clinical studies to test the usefulness of TAVS as a predictive test for treatment response in cancer patients. In a first study, a series of 33 patients suffering from various malignant diseases (lymphoma, leukemia, NSCLC, head and neck squamous cell carcinoma [H&NSCC]) underwent radiotherapy (n = 27), platinum-based chemotherapy (n = 5), or concurrent chemoradiation (n = 1). Planar and SPECT images were visually examined to assess changes in tumor 99mTc-HYNIC-Annexin A5 uptake (before and within 72 hours after start of the therapy) and were compared with treatment response determined according to Response Evaluation Criteria in Solid Tumors (RECIST). Complete or partial tumor response was associated with a marked increase of 99mTc-HYNIC-Annexin A5 accumulation during treatment compared to baseline values. In stable or progressive disease, pretreatment scans demonstrated predominantly low 99mTc-HYNIC-Annexin A5 tumor uptake and no significant increase early after treatment.31 In a follow-up study, 16 patients with advanced NSCLC, scheduled for platinum-based chemotherapy underwent TAVS before and 48 hours after start of the therapy. Chemotherapy-induced changes in tumor Annexin A5 uptake, calculated as maximum count per pixel and expressed as percentage to baseline value, were compared with treatment response according to RECIST. In all lung cancer patients, with complete or partial response a notably increase in tumor 99mTc-HYNIC-Annexin A5 was observed. Stable or progressive disease correlated with less prominently increased or decreased uptake. This study demonstrated a rapid increase of Annexin A5 tumor uptake within 48 hours after the first cisplatin administration, whereas no, or less, increase in tumor Annexin A5 uptake is associated with less successful treatment.32 The same research group confirmed the reliability of visual analysis of 99mTc-HYNIC-Annexin A5 tumor uptake and quantitative tracer uptake analysis in 38 patients with various malignancies. A highly significant correlation was found between the changes in 99mTc-HYNIC-Annexin A5 uptake and therapy outcome: r = 0.97 (p < 0.0001) and r = 0.99 (p < 0.0001) for visual and quantitative analyses, respectively.33 In a heterogeneous patient population (n = 20) receiving chemotherapy or biphosphonate treatment, changes in relative 99mTc-HYNIC-Annexin A5 tumor uptake were evaluated at baseline and at 5 to 7 and 40 to 44 hours after treatment initiation. Imaging results were related to clinical outcome, as defined by RECIST and related to observed changes in the ratios of tumor-to-background activity for the largest known lesion; values exceeding 25% to baseline value on either the 5- to 7-hour scan or the 40- to 44-hour scan were considered significant. Responders to treatment could be separated from nonresponders with 94% accuracy. Results obtained suggested that sequential 99mTc-HYNIC-Annexin A5 imaging might allow for assessment of response to chemotherapy within 3 days after initiations of treatment.34

To evaluate the potential of TAVS to monitor radiation-induced apoptotic cell death, patients with follicular lymphoma underwent 99mTc-HYNIC-Annexin A5 imaging before and after the last radiotherapy fraction of the 2 × 2 Gy involved radiotherapy regimen. Fine-needle aspiration cytology was performed on 5 consecutive days to determine the optimal time window for apoptosis detection. The increase of 99mTc-Annexin A5 after 2 × 2 Gy on day 4 was concordant with the presence of apoptotic cells determined by cytology and correlated with clinical outcome.35 In a more recent study, TAVS was performed before and within 48 hours following the first course of cisplatin-based chemoradiation in patients with H&NSCC. The radiation dose to the tumor at the time of post-treatment TAVS was 6 to 8 Gy. Several ROIs (primary tumor, involved lymph nodes, and salivary glands) were drawn to calculate the radiation dose at posttreatment TAVS. TAVS showed a radiation-dose–dependent uptake in the parotid glands, indicative of early apoptosis during treatment. However, treatment-induced Annexin A5 uptake in primary tumors did not predict outcome. An explanation for this lack of correlation might be that in this study advanced H&NSCC patients with large tumors harboring more necrosis were included. This finding suggests cellular response to radiation is complex and that additional large-scale studies are needed to clarify the role of TAVS in predicting early treatment response.36 Illustrations of 99mTc-HYNIC-Annexin A5 SPECT in human patients are given in Figures 41.4 and 41.5.

Although 99mTc-HYNIC-Annexin A5 is one of the most applied radiolabeled Annexin A5 tracers, it also has some disadvantages. HYNIC is a bifunctional chelator capable of binding to the terminal amine (–NH2) of lysine residues of proteins on the one hand and sequestering 99mTc on the other hand. As any of the -NH2 groups of the 21 lysines present in Annexin A5 can be targeted, this labeling method is rather nonspecific. Also, the labeled complexes with tricine as coligands are often inhomogeneous, consisting of multiple isomeric forms in aqueous solution with various numbers of coligands attached. Other limitations include high uptake in the kidneys which may limit imaging in the abdominal region and may lead to high radiation burden. The persistent localization of radioactivity in the liver and kidneys is attributed to the slow elimination rate of the final radiometabolite 99mTc (HYNIC-lysine)-(tricine)2 from the lysozymes, the main site of intracellular catabolism of proteins. The dissociation of one of the tricine coligands at the low pH environment in the lysozymes and the absence of excess coligands lead to binding of this complex to proteins present in the organelles.37

FIGURE 41.4. A three-dimensional rendered image of the thorax (s = sternum) of a female patient with uptake of 99mTc-HYNIC-Annexin A5 in a primary breast tumor (arrow).

99mTc-Tricarbonyl (CO)3-Annexin A5

The organometallic [99mTc(H2O)3(CO)3]+ core contains three fixed CO ligands and three water molecules, bound to the remaining coordination sites, which can undergo ligand exchange with appropriate donor ligands like biomolecules or simple ligands for the preparation of 99mTc radiopharmaceuticals. The major advantages of the 99mTc-(CO)3 core is the wide variety of ligands which efficiently bind to Tc(I). Complex stabilities are governed by kinetic stability or inertness, are highly robust and do not compose in serum in vivo. The [99mTc(H2O)2(CO)3]+ precursor can be prepared by commercial available kits and takes only 20 to 30 minutes.38 The 99mTc(I)-(CO)3 labeling of Annexin A5 via thiol functions was reported by Biechlin et al.39 and high radiochemical yield and stabile complexes were obtained. Using the 99mTc-(CO)3 core, stable complexes can also be obtained via binding to terminal histidine (His) amino acids in the so-called His-tagged proteins.40 Using this labeling method, 99mTc-(CO)3-labeled His-tagged Annexin A5 may most likely retain its PS-binding affinity because site-specific labeling occurs at specifically introduced N-terminal His tags, not involved in the binding region of the protein.

FIGURE 41.5. A three-dimensional rendered image of the thorax (s = sternum) of a male patient with uptake of 99mTc-HYNIC-Annexin A5 in a squamous tumor cell mass located above the clavicula (arrow).

The radiolabeling of 99mTc-(CO)3 His-Annexin A5 has been characterized and used to determine the value of this Annexin A5 radiotracer to detect PCD in vitro and in vivo in a colorectal tumor model. His-tagged Annexin A5, with an N-terminal extension containing six His residues, was successfully labeled with 99mTc-(CO)3. Radiochemical yields of 70% to 85% and purities of 95% to 99% were obtained with the remainder being free 99mTc-pertechnetate (1% to 5%). Following intravenous (i.v.) injection,99mTc-(CO)3 His-Annexin A5 was rapidly cleared from the blood (7.02 ± 0.44%ID during the first 10 minutes and at 1 hour p.i. only 1.39 ± 0.13%ID remained in the blood). The biodistribution study performed in nude mice demonstrated a predominant uptake of the radiotracer in the kidneys, followed by the liver, indicating a urinary excretory pathway of the tracer. High accumulation of the radiotracer in the renal cortex may be attributed to high concentrations of PS in that region, or because of the nonspecific uptake of low–molecular-weight proteins by proximal renal tubule cells, as previously described.41,42 The biodistribution did not reveal any substantial bowel excretion, resulting in good imaging conditions in the abdominal region; this is in contrast to other 99mTc-labeled Annexin A5 variants such as 99mTc-BTAP-Annexin A5 and 99mTc-labeled Annexin A5 in the presence of SDH, PDTA, tricine, and nicotinic acid.11,43 Overall, the pharmacokinetics of 99mTc-(CO)3 His-Annexin A5 are somewhat similar to those of 99mTc-HYNIC-Annexin A5 in mice, rats, rabbits, and swine.19,20,44 However, for 99mTc-HYNIC-Annexin A5, other organs that concentrated radiolabeled Annexin A5 include spleen, bone marrow, stomach, and lungs.19,20 For 99mTc-(CO)3 His-Annexin A5, these organs did not show visible uptake. Biologic half-life of 99mTc-(CO)3 His-Annexin A5 determined in nude mice is approximately 5.5 hours which is less long compared to the biologic half-lives of 99mTc-HYNIC-Annexin A5 (69 ± 7 hours), 99mTc-BTAP-Annexin A5 (16 ± 7 hours), and 99mTc-i-Annexin A5 (62 ± 13 hours) determined in humans.11,26,45 The faster clearance of the 99mTc-(CO)3 His-Annexin A5 may result in lower background activity; however, it may also lead to a smaller probability of binding of the radiotracer to the PS-binding sites in vivo. Images of the biodistribution of 99mTc-(CO)3 His-Annexin A5 in mice are shown in Figure 41.6.

Application of 99mTc-(CO)3 His-Annexin A5 to patients would result in an ED of 7 ± 0.28 μSv/MBq corresponding to a total ED of 3.5 ± 0.14 (ED per organ) mSv for a patient dose of 500 MBq. These values are in the low range compared to those for 99mTc-i-Annexin A5 (9.7 μSv/MBq), 99mTc-BTAP-Annexin A5 (7.6 μSv/MBq), and 99mTc-HYNIC-Annexin A5 (11 μSv/MBq).11,45,46 However, allometric scaling from the laboratory animals to humans was performed on the basis of body/organ weight which assumes that the biokinetics of compounds mainly depend on the metabolic rate of the animal which in turn is a function of the body weight or body surface area. The ED obtained provides insight into the possible ED in patients. A detailed study in humans, however, is still needed. To image spontaneous levels of apoptosis in vivo, Colo205-bearing mice were imaged on a U-SPECT-II (tomographic, acquisition time of 20 minutes) 3.5 hours after injection of 500-μCi 99mTc-(CO)3 His-Annexin A5 to allow sufficient blood pool clearance. A heterogeneous uptake of the radiotracer was seen in all tumors corresponding to a heterogeneous distribution of spontaneous apoptosis. A high and significant correlation between 99mTc-(CO)3 His-Annexin A5 tumor uptake and number of caspase-3–positive cells in the tumors was demonstrated (Fig. 41.7).47 In a subsequent study, the changes in 99mTc-(CO)3 His-Annexin A5 were examined to determine the dynamics of apoptosis in the same colorectal tumor model, in response to chemotherapy (5-fluorouracil, irinotecan, oxaliplatin), a monoclonal antibody against vascular endothelial growth factor (VEGF), bevacizumab and a monoclonal antibody against epithelial growth factor receptor (EGFR), panitumumab. 99mTc-(CO)3 His-Annexin A5 was administered 4, 8, 12, 24, and 48 hours after start of different treatments and μSPECT was ­performed. Immunostaining of caspase-3 was performed on the tumor slides to correlate with 99mTc-(CO)3 His-Annexin A5 SPECT data. In general, for the three chemotherapeutics tested in this colorectal xenograft model, caspase-3 expression level was concordant with the accumulation of 99mTc-(CO)3 His-Annexin A5 in the tumor.48

FIGURE 41.6. Posterior view of scintigraphic images showing in vivo biodistribution of 99mTc-(CO)3 His-Annexin A5 in healthy mice. Mice were under isoflurane anesthesia after administration of 99mTc-(CO)3 His-Annexin A5 (1 mCi), during first 10 minutes, 4 and 8 hours after injection of the radiotracer (acquisition time of 10 minutes). Rapid clearance from the blood is demonstrated. Kidneys are the main organs of radiotracer uptake, followed by the liver. p.i., post injection.

FIGURE 41.7. On the left, micro-SPECT image of a mouse treated with irinotecan (24 hours) and imaged 3.5 hours after 99mTc-(CO)3 His-Annexin A5 injection. Kidneys are the organs with most uptake and uptake in bone marrow is also observed. Tumor in the right hind leg shows heterogeneous uptake of the radiotracer. On the right, caspase-3 immunostaining performed on the same tumor demonstrates presence of apoptotic cells (brown).

In a recent study, the 99mTc-(CO)3 His-Annexin A5 tracer was used to detect the vascular normalization window in the same colorectal cancer model. Sequential treatments of bevacizumab and irinotecan were administered to normalize the tumor vasculature and to induce PCD. The timing between bevacizumab and irinotecan that would lead to the most effective killing of the cancer cells was evaluated. Microvessel density (MVD), pericyte coverage (α-smooth muscle actin immunostaining), collagen-covered tumor vessels (Masson’s Trichrome staining), and tumor hypoxic fraction (pimonidazole staining) were determined at the three different time points following treatment of bevacizumab. Four days after bevacizumab administration, MVD decreased significantly, α-smooth muscle actin and collagen-covered vessels were increased compared to control tumors, suggesting normalization of the tumor vasculature. Hypoxic fraction was slightly reduced 4 days after treatment with bevacizumab. SPECT analyses demonstrated a significant increase in tumoral 99mTc-(CO)3 His-Annexin A5 uptake 4 days after bevacizumab treatment and 24 hours after irinotecan administration (232.78 ± 24.82%ID/tumor weight (g)/body weight (kg), p < 0.05) compared to each monotherapy demonstrating a synergistic effect of both therapies. 99mTc-(CO)3 His-Annexin A5 μSPECT demonstrates increased antitumor activity of irinotecan during the transient vascular normalization period caused by bevacizumab. The data suggest the importance of timing of combined anti-VEGF treatment with chemotherapy.49 Measurement of the normalization window by noninvasive imaging of chemotherapy-induced PCD could be a powerful tool for clinicians to make decisions regarding pretreatment with drugs affecting the tumor vasculature. In a recent study, 99mTc-(CO)3 His-Annexin A5 was investigated as effective imaging probe to detect apoptosis of internal tissue subjected with high radiation dose (10 Gy) in a γ-irradiated mouse model. Significant higher 99mTc-(CO)3 His-Annexin A5 uptake in the spleen (three- to five-fold) and intestine (two- to three-fold) was found in the irradiated mice as compared to control animals at time points between 45 and 165 minutes p.i. An imaging method to detect and quantify the severity of acute radiation–induced organ or tissue toxicities is still not available. 99mTc-(CO)3 His-Annexin A5 SPECT, to detect radiation-induced organ or tissue damage (2 hours after exposure), prior to the development of severe clinical syndromes (10 to 14 days), may therefore also be of clinical importance.50 So far, no clinical trials have been performed with the 99mTc-(CO)3 His-Annexin A5 radiotracer.

Iodine-123 (123I)-Annexin A5 and Iodine-125 (125I)-Annexin A5

Annexin A5 has been labeled with halogens, including 123I and 125I, thereby providing a broad range of imaging applications in PCD research. Iodinated Annexin A5 can be produced, by either direct or indirect radiolabeling methods. Direct iodination of proteins and peptides are commonly performed using the electrophilic method with mild oxidative reagents and conditions like IodoGen, IodoBeads, chloramine-T (CAT), or oxidative enzymatic lacto-bromo- or myeloperoxidase methods. Indirect labeling has been performed using the Bolton–Hunter reagent (radioiodolabeled N-succinimidyl-3-(4-hydroxyphenyl)propionate), using the prelabeled reagent N-succinimidyl 3-124I-iodobenzoate (124I-m-SIB), and by using 124I-N-hydroxysuccinimidyl-4-iodobenzoate prepared by iododestannylation of N-hydroxysuccinimidyl-4-(tributylstannyl)benzoate (creating 124I-4IB). 123I with γ-energy of 160 keV and physical half-life of 13.2 hours may represent a promising alternative to 99mTc for SPECT imaging of PCD in vivo. 125I has a long physical half-life (59.4 days) and low γ-energy (35 keV).

Preclinical Data

Lahorte et al.51 described optimization of the radiolabeling method for 123I-Annexin A5 using both the IodoGen and the IodoBead method. Radiochemical purities of >98%, and radiochemical yields of 75% to 85% were obtained. Binding experiments with blood platelets demonstrated that 123I-Annexin A5 binding was time- and Ca2+-dependent and indicated that the radiotracer retained its biologic activity.51 The biodistribution and dosimetry of 123I-Annexin A5 was evaluated in mice. 123I-Annexin A5 accumulated primarily in liver, kidney, stomach, and lung. Clearance was predominantly urinary.52 123I-Annexin A5 was prepared using the Iodogen method and tested in Tipifarnib (a selective farnesyltransferase inhibitor [FTI])-treated LoVo-bearing mice (human colon adenocarcinoma cell line). In vivo, LoVo inoculated athymic mice showed a 40% increase in 123I-Annexin A5 uptake 8 hours after a single oral administration of Tipifarnib as well as after 3 days of twice daily treatment with the same dose. Ex vivo TUNEL assays confirmed the presence of apoptotic cells in the tumors.53 In an animal model of dexamethasone-induced apoptosis in the thymus, the degree of apoptosis was determined in the same animal at 6 hours and 11 hours after a single administration of dexamethasone by 123I-Annexin A5 and by in vitro evaluation of the apoptotic index in the form of DNA fragmentation on tissue sections using in situ oligoligation (ISOL). 123I-Annexin A5 uptake significantly increased in the thymus of rats 11 hours after dexamethasone administration and significantly decreased at the 6-hour time point. On the other hand, ISOL demonstrated an increase in the apoptotic index at 6- and 11-hour time points. A positive correlation was determined between the 123I-Annexin A5 uptake as determined by in vivo biodistribution and apoptotic index in thymic tissue as determined by in vitro ISOL (r = 0.508, p = 0.006).54

Russell et al. iodinated (125I) Annexin A5 directly using IodoBeads (resulting in iodination of tyrosine residues) and indirectly by the Bolton–Hunter reagent, which binds to lysine. The efficiency of IodoBeadsiodination was just below 30% whereas with the Bolton–Hunter method 40% efficiency was achieved. When the IodoBeads-labeled Annexin A5 was injected into nude mice, activity accumulated rapidly in the thyroid, which is the result of the rapid deiodination of the protein in vivo. By contrast, when Annexin A5 was labeled with the Bolton–Hunter method, there was no evidence of activity accumulating in the thyroid. In mice given 5-Gy whole-body irradiation, a significant induction of apoptosis in the spleen was measured by TUNEL assay and also a four-fold increase of 125I-Annexin A5 uptake in the spleen above controls was determined. These authors conclude that direct iodination of Annexin A5 is a poor technique suffering from rapid deiodination in vivo, whereas the indirect iodination of Annexin A5 by the Bolton–Hunter method delivers a stable variant and binds to apoptotic cells in vivo.55

Clinical Data

In humans, 123I-Annexin A5 was characterized by a fast biexponential clearance from the blood (t1/2,α = 3.87 ± 0.90 minutes and t1/2,β = 4.13 ± 2.04 hours) and predominant uptake in the kidneys, liver, and gastrointestinal tract, followed by renal excretion. The average ED of 123I-Annexin A5 was estimated to be 0.02 mSv/MBq. The extensive bowel activity and rapid in vivo dehalogenation seriously compromise the use of 123I-Annexin A5 for imaging apoptosis in a clinical setting.52

Indium-111 (111In)-DTPA-PEG-Annexin A5

Annexin A5 has also been labeled with SPECT isotope 111In, which has a longer physical life (2.81 days) than 99mTc, making it possible to detect therapy-induced apoptosis over a prolonged period. Because apoptosis is a dynamic process and apoptotic cells are engulfed by neighboring cells within hours, it has been suggested that increasing the interval between radiotracer injection and imaging would allow detection of apoptotic cells over a longer period in time, enhancing detection sensitivity. However, the longer half-life of the isotope may cause a rather high radiation burden in patients.

Diethylenetriaminepentaacetic acid (DTPA) was used as chelator for 111In labeling of Annexin A5 and ovalbumin modified with polyethylene glycol (PEG). PEG-modified proteins have been shown to exhibit reduced liver uptake and increased blood circulation biologic half-lives. 111In-DTPA-PEG-Annexin A5 was obtained with high radiochemical purity (98%) and high radiochemical yield (92%). Mice bearing subcutaneously inoculated human mammary MDA-MB-468 tumors were treated with poly(L-glutamic acid)-paclitaxel, monoclonal antibody C225 (targeting EGFR), or a combination of both. Images were acquired 48 hours after the injection of the radiotracer. Autoradiography and TUNEL staining were performed on tumor slides to localize apoptotic cells. 111In-DTPA-PEG-Annexin A5 uptake increased in the gastrointestinal tract as well significantly in the tumor of mice treated with either poly(L-glutamic acid)-paclitaxel (10.76 ± 1.38%ID/g, p = 0.001) or combined poly(L-glutamic acid)-paclitaxel and C225 (9.84 ± 2.51%ID/g, p = 0.029) during 4 days compared to uptake of radioactivity in tumors of untreated mice (6.14 ± 0.67%ID/g). There was a significant correlation between tumor uptake of 111In-DTPA-PEG-Annexin A5 and apoptotic index determined by TUNEL assay (r = 0.87, p = 0.02). Imaging with 111In-DTPA-Annexin A5 failed to reveal tumors in either untreated or treated mice. Intratumoral distribution of 111In-DTPA-PEG-Annexin A5 showed a main distribution in the central zone of the tumors, whereas 111In-DTPA-Annexin A5 was largely confined to tumor periphery. The fraction of apoptotic cells in MDA-MB-468 tumors after poly(L-glutamic acid)-paclitaxel treatment was quite low (7.6% to 11.1%), suggesting 111In-DTPA-PEG-Annexin A5 is sensitive in the detection of this low number of apoptotic cells. In addition, by increasing the blood half-life of 111In-DTPA-PEG-Annexin A5, may have allowed the radiotracer to diffuse deeper in the tumor as 111In-DTPA-PEG-Annexin A5 showed strong radioactivity in the central zone in addition to the peripheral zone, whereas 111In-DTPA-Annexin A5 showed only weak radioactivity in the periphery of the tumor.56 In a recent study, Annexin A5 was conjugated to PEG-coated, core-cross–linked polymeric micelles (CCPMs) dually labeled with near-infrared (NIR) fluorophores and 111In. The potential application of Annexin A5-CCPM was investigated in a lymphoma xenograft mouse model treated with cyclophosphamide and etoposide. At 48 hours after injection, 111In-Annexin A5-CCPM showed significantly higher uptake in the treated tumors (8.01%ID/g) compared to the untreated tumors (3.2%ID/g). 111In-Annexin A5-CCPM also showed significantly higher uptake in the tumors of the treated mice than 99mTc-HYNIC-Annexin A5 (4.14%ID/g) and 111In-labeled CCPM (2.81%ID/g). Localization of radioactivity from 111In-Annexin A5-CCPM correlated with apoptotic cells stained with caspase-3. By conjugating Annexin A5 to the surface of PEG-coated CCPM, the mean half-life was highly increased (12.5 hours) compared to 99mTc-labeled Annexin A5 (<7 minutes). In this way, Annexin A5 may penetrate deeper into the tumor and it is possible to visualize apoptotic cells over a prolonged period, allowing improved detection of therapy-induced apoptosis. Indeed, 111In-Annexin A5-CCPM showed significantly higher uptake in the tumors compared to 99mTc-HYNIC-Annexin A5, indicating that prolonging the blood-life of Annexin A5 leads to increased uptake of the radiotracer in the tumor. Moreover, the dually labeled Annexin A5-CCPM (nuclear and optical) allows at the same time detection of apoptotic cells at microscopic level by optical imaging.57

ANNEXIN A5 MUTANTS FOR 99mTc LABELING

In an effort to develop improved Annexin A5-based imaging agents, an endogenous chelation site for 99mTc was added to the N-terminus of Annexin A5, allowing direct labeling with 99mTc. The three mutants were called Annexin V-116, V-117, and V-118. Seven amino acid sequences (containing either one or two cysteine residues) to the N-terminal side of Annexin A5 were introduced and the naturally occurring Cys-136 was mutated to Ser in all three proteins. All three proteins retained full binding affinity for erythrocyte membranes with exposed PS. Labeling occurred in the presence of SnCl2 and glucoheptonate as exchange agent. Labeling reaction was rapid, and high radiochemical yields and purities were achieved. Biodistribution studies in mice revealed predominant uptake in liver and kidneys (60 minutes p.i.) for the three Annexin A5 mutants (ranging from 5.9 to 11.2%ID and from 5.9 to 17.9%ID), although the extent of uptake in these organs, together with spleen and bone marrow uptake was significantly decreased in comparison with 99mTc-HYNIC-Annexin A5 (16.6%ID and 39.1% for liver and kidney, respectively). Of these three mutant Annexin A5 proteins, Annexin-V-117 showed the most beneficial biodistribution properties.58 99mTc-Annexin V-117 was used in rats bearing hepatoma and treated with cyclophosphamide 11 days after tumor inoculation, to detect if there is interference of repetitive doses of Annexin A5 (or V-117) on accumulation of 99mTc-Annexin V-117 in tumors. Cold Annexin A5 was administered 24 hours before or after the cyclophosphamide treatment. 99mTc-Annexin V-117 was injected intravenously and radioactivity in tissue was determined 6 hours later. This study demonstrated that accumulation of radiolabeled Annexin A5 in tumors was not significantly affected by previous treatment with cold Annexin A5 before and after chemotherapy, which is highly desirable in the clinical setting because repetitive detection of apoptosis with radiolabeled Annexin A5 may be necessary.59 Although the Annexin V-117 used in this study, showed similar biodistribution pattern compared to 99mTc-HYNIC-Annexin A5, with a more rapid clearance, the absolute tumor uptake in cyclophosphamide-treated hepatoma-bearing rats was three-fold lower compared to 99mTc-HYNIC-Annexin A5 uptake in the same rat model.60 Two mutant molecules of Annexin A5 were constructed, suitable for the 99mTc-tricarbonyl labeling method. Annexin V-122 and Annexin V-123 were constructed with N-terminal extensions of 3 or 6 His residues because His is known to form highly stable complexes with [99mTc(H2O)3(CO)3]+ resulting in high specific radioactivities.61The tricarbonyl labeling of His-tagged Annexin A5 still requires a two-step procedure in which first the [99mTc(H2O)3(CO)3,]+ chelate needs to be performed in a separate step at 95° to 100°C. However, labeling of the Annexin V-117, for example, can be performed directly at room temperature and is a much faster and simpler labeling method for routine production in the clinic. Another mutant Annexin A5 protein was developed with an endogenous 99mTc chelation site at the N-terminus of Annexin A5, called Annexin V-128. Additional mutations were made in other residues that altered molecular charge without altering membrane-binding affinity. Comparison of 99mTc-Annexin V-128 with 99mTc-HYNIC-Annexin A5 showed a similar biodistribution of the two proteins, however 99mTc-Annexin V-128 had an 88% lower renal uptake than 99mTc-HYNIC-Annexin A5 at 60 minutes after injection. Liver and spleen uptake was also somewhat lower for 99mTc-Annexin V-128. Small bowel showed a slightly higher uptake of 99mTc-Annexin V-128 compared to 99mTc-HYNIC-Annexin A5 (1.6%ID). Mice were treated with the protein synthesis inhibitor, cycloheximide, to induce apoptosis. A several-fold increase in uptake of 99mTc-Annexin V-128 in the liver and spleen that correlated well with histologic analysis of the degree of apoptosis determined by TUNEL assay was found.62

99mTc-LABELED SECOND-GENERATION ANNEXIN A5

Disadvantages of the native Annexin A5 include an unfavorable biodistribution (high renal and liver uptake) and low tumor-to-background ratio, so further improvement of radiolabeled Annexin A4 is required. More recently, a variant of the native Annexin A5, called “second-generation” Annexin A5 was developed. Cys-Annexin A5 was developed by incorporation of a single cysteine residue at its concave side using site-specific mutagenesis to allow conjugation through thiol chemistry, without affecting its apoptotic cell–binding characteristics. Cys-Annexin A5 was conjugated to 99mTc-HYNIC and evaluated as novel apoptosis imaging agents. Biodistribution studies of 99mTc-HYNIC-cys-Annexin A5 in normal mice showed a similar pattern to the first-generation 99mTc-HYNIC-Annexin A5 conjugates. A similar uptake in the kidneys was observed for both radiotracers, 60 minutes p.i.; however, uptake in the liver was significantly higher for the second-generation 99mTc-HYNIC-cys-Annexin A5, at 10 minutes and 60 minutes p.i. There was a 257% increase in hepatic uptake 4 hours p.i. of 99mTc-HYNIC-cys-Annexin A5 in a murine model of apoptosis (treated with anti-Fas) when compared to control mice, indicating that 99mTc-HYNIC-cys-Annexin A5 may yield similar apoptosis imaging results when compared to the first-generation 99mTc-HYNIC-Annexin A5.63 The second-generation 99mTc-HYNIC-cys-Annexin A5 was also studied in rats, treated with cycloheximide to induce cell death in the liver and spleen, whereas apoptosis in the prostate was induced by castration. Again, kidneys were the main organs of 99mTc-HYNIC-cys-Annexin A5 uptake, and much lower levels accumulated in the liver and spleen. Cycloheximide treatment resulted in increased 99mTc-HYNIC-cys-Annexin A5 accumulation in liver and spleen over controls (two- and four-fold respectively, 15 minutes after radiotracer injection), which correlated well with TUNEL staining in the corresponding tissues. However, following castration, no increased uptake of 99mTc-HYNIC-cys-Annexin A5 in the prostate was observed whereas there was a strong increase in TUNEL-positive prostate epithelial cells. One explanation may be that the radiotracer cannot reach easily its target because capillary endothelial cells in the prostate are tightly organized with the main apoptotic cell type situated behind this barrier. Moreover, an additional layer of smooth muscle cells may provide an additional barrier between the prostate cells and the circulation. The utility of the second-generation 99mTc-HYNIC-cys-Annexin A5 was demonstrated in a model of cyloheximide-induced liver and spleen apoptosis. It failed to detect, however, an increase in apoptotic prostate cells following castration.46 In a recent study, the second-generation Annexin A5 (cys-Annexin A5) was labeled with 99mTc in three different ways (HYNIC, (CO)3-DTPA, and (CO)3 His-tagged). Faster urinary excretion was shown for the 99mTc-(CO)3 His-cys-Annexin A5 compared to the other two and hepatobiliary excretion was found for all three. A disadvantage of the 99mTc-(CO)3-DTPA-cys-Annexin A5 was the lower stability in blood plasma: After 1.5 hours in plasma, 70% of 99mTc-(CO)3-DTPA-cys-Annexin A5 was still present. Competition reaction with an excess of His showed a shift of at least 30% of the 99mTc-label from the DTPA-cys-Annexin A5 to His. After 4 hours 99mTc-HYNIC-cys-Annexin A5 uptake in the liver was significantly lower than the other two. Blood clearance was fastest for 99mTc-HYNIC-cys-Annexin A5, whereas the other two remained longer in the blood. In the anti-Fas–treated mouse model, 99mTc-(CO)3 His-cys-Annexin A5 showed the highest liver uptake of the three 99mTc-labeled cys-Annexin A5 variants. However, this compound has also higher blood pool values, so that the tumor-to-background ratios were in the same order of magnitude compared to 99mTc-HYNIC-cys-Annexin A5.64 In a recent study, the 99mTc-(CO)3 His-cys-Annexin A5 was evaluated in a mouse tumor model. Daudi tumor (human lymphoma cell line) was implanted in the spleen of SCID mice and mice were treated with adriamycin and cyclophosphamide. 99mTc-(CO)3 His-cys-Annexin A5 SPECT was performed before and 1 day after therapy. No radiotracer uptake was observed in the baseline tumors whereas tracer uptake was visualized in the tumors after therapy. Ex vivo measurements demonstrate that the tumor uptake significantly increased 1 day after treatment (0.8 ± 0.1%ID/g) compared to control (0.3 ± 0.1%ID/g), which was confirmed histologically by TUNEL staining.65 An overview of all SPECT-labeled Annexin A5 tracers is given in Table 41.1.

TABLE 41.1

OVERVIEW OF RADIOLABELED SPECT ANNEXIN A5 TRACERS

ANNEXIN A5 PET RADIOTRACERS

Positron emission tomography (PET) offers higher resolution (spatial resolution is about two times better), higher sensitivity, and is more quantitative than SPECT. Additionally, PET accurately corrects for attenuation of photons emitted from the organ of interest, making PET images more accurate in quantifying the tracer activity in the tissue. Disadvantages of PET include the higher cost and the necessity of a particle accelerator (cyclotron) for production of the radionuclides like 18F.

18F-Annexin A5

The short physical half-life of 18F (=1.83 hours) and lower energetic positron emission compared to Iodine-124(124I), results in a substantial lower radiation burden to the patient, making 18F-labeled Annexin A5 PET tracers very suitable in clinical settings. Labeling of proteins with 18F is performed through 18F-labeled prosthetic groups (also referred as bifunctional labeling agents) because direct labeling is very challenging because of harsh reaction conditions.

A four-step synthesis has been described by Zijlstra et al. for the synthesis of 4-18F-fluorobenzoyl-Annexin A5 (18F-FBA) by means of an N-succinimidyl-4-18F-fluorobenzoate (18F-SFB) synthetic group. The synthesis occurred by a microcomputer-controlled, automated module and 18F-FBA synthesis took 90 minutes. Radiochemical yields ranged from 15% to 20% with radiochemical purities of >95%. To determine whether Annexin A5 maintained its biologic activity after labeling with 18F, Jurkat T cells were treated with UV-irradiation to induce apoptosis. Apoptotic Jurkat T cells showed a 60% increase in 18F-Annexin A5 compared to uptake in nonirradiated Jurkat T cells. Radioligand binding appeared to be time- and concentration-dependent and could be saturated.66 The group of Toretsky has also labeled Annexin A5 with 18F via 18F-FSB by a manual process allowing researchers to prepare 18F-Annexin A5 without the need of the microcomputer-controlled, automated module. The overall radiochemical yield of 18F-Annexin A5 from 18F was 17.6 ± 5.6% in 3 hours. Stimulation of cancer cells with the chemotherapeutic agent etoposide for 6 hours showed an 88% increase of the binding of 18F-Annexin A5 when compared to control cells.67 A third method for production of 4-[18F]-FBA-Annexin A5 required 2 hours of synthesis whereas lowest radiochemical yields in the range of 10% and radiochemical purities of 99% were obtained. The radiotracer was further evaluated in a rat model of myocardial ischemia.6818F-Annexin A5 was also tested in rats treated with cycloheximide to induce liver apoptosis. The rats were imaged by PET over 2 hours. Biodistribution studies of normal rats showed the highest uptake of 18F-Annexin A5 in urinary bladder and kidney. After the bladder and kidney, organs with highest uptake were lungs, liver, and spleen. Radioactivity cleared rapidly from the heart, with less than 7% of maximum still present 1 hour after injection, and with a return to background levels before 2 hours. The peak of activity in the kidneys occurred at 10 minutes, at the same time that activity began to appear in the urine. After 1 hour, activity in the kidney had decreased to 16% of the maximum. After 2 hours of imaging, nearly 70% of the ID of 18F-Annexin A5 was found in the urine. 18F-Annexin A5 is cleared more rapidly from the liver and kidneys compared to 99mTc-HYNIC-Annexin A5. The rapid excretory clearance and shorter decay half-life of 18F should result in lower radiation dose to most internal organs compared to 99mTc-HYNIC-Annexin A5. Indeed, kidneys, liver, red marrow, and spleen would all receive a lower radiation dose with18F-Annexin A5 compared to 99mTc-HYNIC-Annexin A5. Only the radiation dose of the testis and bladder would be higher in the case of 18F-Annexin A5. Pretreatment of rats with cycloheximide resulted in a three- to nine-fold increase in 18F-Annexin A5 in the liver of treated rats at 2 hours compared to control animals. Morphologic analysis and TUNEL assay confirmed the apoptotic levels.69 In a recent study, longitudinal 18F-Annexin A5 PET imaging was performed in an H&NSCC tumor xenograft model, treated with two doses of doxorubicin (10 mg/kg each dose) with 1-day interval. Longitudinal PET was performed at 6 hours, 24 hours, 3 days, and 7 days after treatment started. After treatment with doxorubicin, the tumor uptake of 18F-Annexin A5 increased with time. At day 3, the treated tumors showed a significantly increased tracer uptake (1.56 ± 0.23%ID/g) compared to control tumors (0.89 ± 0.31%ID/g). At day 7 after treatment, the tumor uptake of 18F-Annexin A5 had returned to baseline level. TUNEL staining on the spleen tumor slides confirmed the presence of apoptotic cells, 3 days after start of the treatment.70 At this time, possibly because of a not yet standardized labeling method for 18F-Annexin A5, no clinical trials in human cancer patients have taken place.

Copper-64 (64Cu)-Annexin A5

64Cu with a longer physical half-life (12.7 hours) is ideal for in vivo monitoring of apoptosis in humans as imaging over a wider time window is possible. The lower energy of the radionuclide at the same time would cause a lower radiation burden to patients, in particular to the specific organs that normally retain the radiolabeled Annexin A5 (kidney, spleen, and liver). A drawback however, is the laborious synthesis of the 64Cu-DOTA-Annexin A5 complex.

McQuade et al. described the radiolabeling of 1,4,7,10-tetraazacyclododecane-N,N′,N′,N′″-tetraacetic acid (DOTA)-conjugated Annexin A5 with 64Cu as an alternative PET radionuclide. 64Cu-DOTA-Annexin A5 was prepared with high specific activities and radiochemical purities (90% to 94%). This radiotracer was evaluated in an animal model of anti-Fas–induced hepatic apoptosis. In mice pretreated with anti-Fas, a 240% increase in spleen uptake and a 238% decrease in kidney uptake was observed, however the liver showed no uptake. Blood metabolism study in mice pretreated with anti-Fas showed that after 1 hour only 3% of the total activity was bound to red blood cells whereas it remained 65% intact.71 Another research group used 64Cu-DOTA-biotin-Streptavidin (Sav) PET, following pretargeting of apoptotic cells with biotinylated Annexin A5 to visualize tumor apoptosis after photodynamic therapy (PDT) in tumor-bearing mice. An avidin chase after administration of the biotinylated Annexin A5 was performed to eliminate free biotinylated products thus increasing detection specificity and reducing blood background. PDT induces apoptosis by using phthalocyanine dyes as photosensitizers and red light. These dyes have been shown to act directly on the tumor cells causing rapid and extensive damage within the illuminated area. Three hours after the i.v. injection of 64Cu-DOTA-biotin-Sav, the plasma concentration of the tracer was 2.8% of the initial blood concentration, which was sufficiently low to assure image sufficient image contrast. PET scans were recorded up to 13 hours after PDT and delineated apoptosis in the treated tumor as early as 30 minutes after tracer injection. A maximum contrast between the treated tumor and the surrounding tissue was reached at 7 hours post PDT. The main excretory way was urinary (high uptake of the radiotracer in the bladder and kidney). The main disadvantages of this apoptosis radiotracer include the laborious synthesis (preparation of 64Cu-DOTA-biotin-Sav takes approximately 2 hours), the three-step protocol for administration and the fact that both biotin and Sav can provoke an immunologic response.72

94mTc-HYNIC-Annexin A5

McQuade et al. also described the radiolabeling of 94mTc-HYNIC-Annexin A5, to transform the SPECT 99mTc-HYNIC-Annexin A5 into a PET variant. With HYNIC as chelator, the radiolabeling could be easily performed. In mice pretreated with anti-Fas, a 340% increase in spleen uptake and a 388% decrease in kidney uptake were observed, whereas no uptake in the liver was detected.71

124I-Annexin A5

Annexin A5 has also been radiolabeled directly and indirectly with the positron emitter 124I (half-life = 4.18 days, Emax β+ = 1.53, 2.14 MeV). The longer half-life of 124I allows in vivo detection and quantification over a longer period and the radionuclide can be prepared well in advance so subsequent radiolabeling can be performed.

Glaser et al. investigated the use of 124I-Annexin A5 as potential probe for PET imaging of apoptosis. Annexin A5 was iodinated directly by the CAT method to produce 124I-Annexin A5, and indirectly by the prelabeled reagent N-succinimidyl-3-124I-iodobenzoate (124I-m-SIB). Under optimized condition, the radiochemical yield of directly labeled 124I-Annexin A5 was 22.3 ± 2.6% and radiochemical purity was >95%. Radiochemical yield of indirectly labeled 124I-Annexin A5 was 25% and radiochemical purity was also >95%. No deiodination was observed for both compounds upon storage in PBS at 4°C for 4 days. The biologic activities of both compounds were investigated in an in vitro study, using human leukemic HL60 cells stimulated with camptothecin by using the 125I counterparts of directly and indirectly iodinated Annexin A5. A nonsignificant increase of 17% of directly labeled 125I-Annexin A5 and a significant 21% increase of the indirectly labeled125I-Annexin A5 were demonstrated, suggesting a reduced biologic activity of the directly labeled 125I-Annexin A5.73 In another study, Annexin A5 was directly labeled with 124I and validated in an in vitro and an animal model of liver apoptosis. Jurkat cells treated with the anticancer drug camptothecin had a six-fold higher 124I-Annexin A5 uptake as compared to untreated cells. A good correlation was found between radioactivity per cell and FITC-Annexin A5 binding at different apoptosis levels (r = 0.93, p < .0001). Radioactivity was rapidly cleared from the blood, with a half-life of approximately 12 minutes for Annexin A5-bound radioactivity. A biodistribution study in normal mice (2 hours p.i.) revealed highest uptake in the blood, thyroid, kidneys, and stomach. The high uptake in thyroid and stomach suggests some dehalogenation in vivo. Administration of anti-Fas antibody to BDF-1 mice resulted in significantly higher 124I-Annexin A5 hepatic uptake compared with controls at 1 and 2 hours. PET images obtained at 2 hours after injection of 124I Annexin A5 in untreated and anti-Fas–treated mice revealed accumulation of the tracer in bladder and stomach, whereas treated animals had massive accumulation in the livers. Also, in vivo 124I- Annexin A5 uptake derived from PET images correlated with histologic derived apoptotic density (r = 0.67, p < .001).74 Collingridge et al. also synthesized directly labeled 124I-Annexin A5 (CAT method) and 124I-m-SIB-Annexin A5 (the 125I-Annexin A5 counterparts were synthesized in the same way for the in vitro and biodistribution studies). RIF-1 cells treated with 5-fluorouracil for 24 hours to induce apoptosis showed higher binding of 125I-Annexin A5 and 125I-SIB-Annexin A5 compared to control cells. The extent of binding was dependent on the concentration of 5-fluorouracil added, however the percentage of radiotracer that was bound to apoptotic cells was higher in the case of 125I-SIB-Annexin A5. The uptake of 125I-SIB-Annexin A5 by RIF-1 tumors growing in mice increased by 2.3-fold at 48 hours after a single injection of 5-fluorouracil, compared to a 4.4 fold increase in TUNEL-positive cells. Uptake of the directly labeled 125I-Annexin A5 did not significantly change in 5-fluorouracil–treated mice compared to control mice. 124I-SIB-Annexin A5 appeared to be superior for PET imaging in vivo compared to the directly labeled 124I-SIB-Annexin A5 radiotracer, however further preclinical and clinical research is warranted.75 The group of Dekker et al. also compared directly labeled 124I-Annexin A5 (by IodoGen method) and indirectly labeled 124I-Annexin A5. 124I-4IB Annexin A5 was made by using [124I]N-hydroxysuccinimidyl-4-iodobenzoate prepared by iododestannylation of N-hydroxysuccinimidyl-4-(tributylstannyl)benzoate. In comparison with 124I-Annexin A5, 124I-4IB-Annexin A5 has a higher rate of binding to PS in vitro. a higher kidney and urine uptake, a lower thyroid and stomach content uptake, greater plasma stability, and a lower rate of plasma clearance. In mice treated with anti-Fas antibody, 124I-4IB-Annexin A5 accumulated in the liver; however, the difference in uptake between untreated and treated animals was not as clear as that previously demonstrated with directly labeled 124I-Annexin A5.76

The same research group developed an 124I maltose-binding protein (MBP)-Annexin A5 and evaluated this radiotracer in vivo. The Annexin A5 was radiolabeled by direct iodination applying the IodoGen method and evaluated in a model of hepatic apoptosis. Iodinated MBP and albumin were used as control proteins. Treatment with anti-Fas antibody prior to radiotracer injection showed clearly apoptotic hepatocytes by PET imaging. Liver uptake of 124I-MBP-Annexin A5 was nine times greater for mice treated with anti-Fas antibody than for untreated control animals, which was confirmed when 124I-MBP-Annexin A5 was used as radiotracer. Immunohistochemistry on liver sections taken from animals treated with anti-Fas antibody confirmed the presence of apoptotic cells. However, biodistribution studies with125I-MBP-Annexin A5 revealed substantial uptake of the radiotracer in the bladder, stomach content, thyroid, and blood.77 Despite the promising results of 124I-Annexin A5 as alternative PET probe for imaging of apoptosis in vivo. some issues including deiodination of the radiolabeled Annexin A5 probe need to be overcome before 124I-labeled Annexin A5 may find its way into the clinic.

Carbon-11 (11C)-Annexin A5

Another PET radioligand for apoptosis imaging is the 11C-labeled Annexin A5. 11C has a short physical half-life (20.39 minutes) and emits positrons with Emax β+ = 0.961 MeV. A recent study performed PET imaging with 11C-labeled Selenium (Sel)-tagged Annexin A5 (11C-Annexin A5-ST). A nucleotide sequence encoding a bacterial-type SECIS element and the SEL-tag having the amino acid sequencer–Gly-Cys-Sec-Gly was previously introduced into the Annexin A5 gene by PCR.78 Radiochemical purity was >90%. In a mouse xenograft model (head and neck cancer), modest apoptosis was induced by doxorubicin. Apoptosis in tissue was verified ex vivo by caspase-3 immunostaining. In addition, serum biomarkers for apoptosis, caspase-cleaved keratin 18 (ccK18) and total keratin 18 (K18) were monitored. At 72 hours after a single dose of doxorubicin, 11C-Annexin A5-ST uptake in tumors was elevated compared to control tumors, which was confirmed by caspase-3 immunostaining. Also plasma levels of ccK18 at this time point were significantly increased; however, detection sensitivity was higher compared to PET imaging. These authors therefore conclude that combining apoptosis imaging with measuring circulating serum biomarkers of cell death increases the total sensitivity of detection.79 However, there are some disadvantages in using 11C-Annexin A5 as PET apoptosis ligand. The short physical half-life of 11C hampers in vivo monitoring of apoptosis over a long period, which is essential given the dynamic character of apoptosis. Detection will only occur at one time point, delivering only one snapshot of the apoptotic process, and no information on the apoptotic kinetics can be obtained (unless repeated injections are administered). In addition, so far no straightforward radiolabeling protocol for 11C-Annexin A5 is available, which is crucial for clinical applications.

Gallium-68 (68Ga)-Annexin A5

With the short half-life of this radionuclide (68 minutes) the radiation dose to the patient can be minimized and may also allow rapid sequential imaging. It emits positrons with Emax β+ = 1.9 MeV. As opposed to 64Cu and 18F, 68Ga can be eluted from an in-house Ga/Ge-generator and labeling procedures with this metal are straightforward. As 68Ga behaves like 111In, a great potential of labeling chelators is already available. However, not many 68Ga-labeled Annexin A5 tracers are described in the literature. This is mainly because of the fact that the chelator DOTA, which is predominantly used for radiometal nuclide complexation in radiochemistry, requires high temperatures or long reaction times for complexation with 68Ga. Because proteins are susceptible to heat-induced breakdown, chelators requiring a heating step for complexation are not suitable for such a radiolabeling approach. Especially for Annexin A5 this is the case as temperatures above 50°C start to destroy the biologic activity of Annexin A5. Wängler et al. described in 2011 a universally applicable 68Ga-labeling technique for proteins. They conclude that the 68Ga-labeling procedure for proteins should comprise only one simple, fast, and preferably quantitative labeling step at room temperature without the need for further purification. These authors introduced the 68Ga chelating agent NODA-GA-T (2,2′-(7-(1-carboxy-4-(2-mercaptoethylamino)-4-oxobutyl)-1,4,7-triazonane-1,4-diyl)diacetic acid) by reaction with proteins chemically processed with sulfo-SMCC (4-(N-maleimidomethyl)cyclohexane-1-carboxylic acid 3-sulfo-N-hydroxy-succinimide ester sodium salt) which resulted in labeling precursors, enabling a simple and rapid kit-labeling procedure. The 68Ga-labeling technique provided the 68Ga-Annexin A5 in a straightforward synthesis within a short preparation time (15 minutes). High radiochemical yields of 95% were obtained. The applicability and attained biologic activity of 68Ga-Annexin A5 was demonstrated in mice with left ventricular myocardial infarctions.80

Two Annexin A5 variants were labeled with 68Ga-DOTA-maleimide, namely Cys2-Annexin A5 and Cys165-Annexin A5 containing a single available cysteine residue at respectively the 2-position and the 165-position. Biodistribution and pharmacokinetics were studied by micro-PET in healthy mice and in a mouse model of hepatic apoptosis. Dynamic PET with both radiotracers was also performed in lymphoma-bearing mice before and after treatment (cyclophosphamide [20 hours before apoptosis imaging] plus radiotherapy [10 Gy/tumor, 4 to 6 hours before apoptosis imaging]). Radiochemical yields of 70 ± 10% and radiochemical purities of >98% were obtained for both radiotracers and total synthesis time was about 55 minutes. The blood clearance half-life of 68Ga-Cys2-Annexin A5 and 68Ga-Cys165-Annexin A5 were 6.5 and 5.5 minutes, respectively. Clearance was mainly via the urinary tract (high uptake in the kidneys) with only minor uptake in the liver and intestines. In anti-Fas–treated mice, a major shift of radioactivity occurred from the kidneys to the apoptotic liver. Significant differences in the tumor standardized uptake values (SUVs) (10 and 30 minutes) before and after start of the therapy were observed for both radiotracers, however tumor delineation based solely on PET images proved impossible. MRI-PET fusion images did allow a clear delineation of each tumor.81 The low absolute tumor uptake of 68Ga-Annexin A5 may be problematic when only PET is used, however in the standard clinical situation combined PET/CT usually occurs and CT can be used to delineate the tumors accurately. An overview of all PET-labeled Annexin A5 tracers is given in Table 41.2.

TABLE 41.2

OVERVIEW OF RADIOLABELED PET ANNEXIN A5 TRACERS

ADDITIONAL ANNEXIN RADIOTRACERS

99mTc-HYNIC-Annexin B1

Annexin B1 is a PS-binding protein with a slightly longer N-terminal domain compared to Annexin A5. In contrast to Annexin A5, Annexin B1 can be produced using nonfusion expression systems making it a convenient and cheap alternative for apoptosis imaging in the clinical setting. 99mTc-HYNIC-Annexin B1 was prepared and detection of dexamethasone-induced thymic apoptosis and Fas-mediated liver apoptosis in mice were investigated. Biodistribution studies revealed similar pharmacokinetics of 99mTc-HYNIC-Annexin B1 in mice when compared to 99mTc-HYNIC-Annexin A5: Rapid blood clearance, high renal uptake, and no bowel excretion. Thymus uptake of 99mTc-HYNIC-Annexin B1 increased significantly 15 hours after dexamethasone injection (0.45%ID/g in control and 1.47%ID/g in treated mice, p < 0.001). A marked increase in 99mTc-HYNIC-Annexin B1 uptake in the liver was seen 2 hours after anti-Fas treatment.82

99mTc-Labeled Annexin 13 Peptides

The PS-specific sequence of Annexin A5 is attributed to a chain consisting of 13 amino acids at the N-terminal. Small peptides can be easily synthesized and modified for subsequent radiolabeling. In addition, these peptides (owing to the lower molecular weight) may have a more favorable biodistribution (rapid blood clearance and quick renal clearance) and a fast uptake in the tumor. 99mTc-labeled Annexin 13 fragments may therefore prove to be a better tool for apoptosis imaging compared to the native Annexin A5. Peptides CC-Annexin 13, C-Annexin 13, and H-Annexin 13 were synthesized and labeling of the first two peptides with 99mTc was carried out via 99mTc-nitrido intermediate whereas labeling of H-Annexin 13 with 99mTc was carried out via the 99mTc-(CO)3 core. Radiochemical purities for all three Annexin 13 fragments were >95%, although HPLC analysis of 99mTc-C-Annexin 13 revealed the formation of two species with different retention times. Biodistribution studies of 99mTc-CC-Annexin 13 in tumor-bearing mice revealed rapid clearance out of the blood. Kidneys were the main organs of uptake, followed by the lungs, intestine (+ gall bladder), and lungs. Tumor uptake was 0.52 ± 0.017%ID/g at 1 hour p.i. The retention of 99mTc-CC-Annexin 13 in soft tissue was relatively low in comparison to 99mTc-HYNIC Annexin A5, and this Annexin 13 fragments could be an alternative agent for apoptosis imaging.83

ADDITIONAL PS-BINDING AGENTS

Radiolabeled Synaptotagmin-I

Another PS-binding protein is synaptotagmin-I, a synaptic vesicle–associated membrane protein that contains C2A and C2B domains. The C2A domain is an approximately 120-amino acid segment and binds anionic phospholipids with relatively high affinity. The C2A domain is typically expressed and used as a recombinant gluthatione-S-transferase (GST) fusion protein. Radiolabeling of C2A with 99mTc and 18F leads to imaging probes for visualizing cell death that are also investigated in animal models for anticancer treatment.84 C2A variants containing His tags and/or free cysteine have also been developed for efficient and site-specific radiolabeling, as has been done for Annexin A5.85 The feasibility of 99mTc-labeled C2A-GST in imaging NSCLC apoptosis induced by paclitaxel treatment was investigated in a mouse model of human NSCLC. The tumor-to-normal tissue ratio (T/NT) significantly increased after paclitaxel treatment, whereas it was low in untreated tumors (T/NT = 1.24 ± 0.23). Tracer activity in tumors gradually increased with time, and in terms of biodistribution the kidneys and liver showed highest tracer accumulation, whereas uptake in the gastrointestinal tract was relatively low. The radiotracer uptake in the tumors was also positively correlated with the apoptotic index and caspase-3 activity determined ex vivo on the tumors.86The same research group developed an 18F-labeled C2A-GST probe and assessed the response of paclitaxel treatment in VX2 rabbit lung cancer model. Biodistribution of 18F-C2A-GST was determined in mice ex vivo and by small-animal PET. FDG-PET/CT was performed before and 72 hours after a single dose of paclitaxel. 18F-C2A-GST was obtained with more than 95% radiochemical purity, with radiochemical yield of 35% to 40% and bound in vitro to apoptotic Jurkat cells stimulated with camptothecin. Biodistribution studies revealed that 18F-C2A-GST was mainly excreted from the kidneys and rapidly cleared from the blood and nonspecific organs. After paclitaxel treatment, intense uptake of 18F-C2A-GST was found in the tumors. The SUVmax after therapy was 0.47 ± 0.28, significantly higher than in control mice (0.009 ± 0.0001), which correlated with caspase-3 activity in the tumors. However, a nonspecific accumulation of the radiotracer in the intestines after therapy was observed, which is presumably because of inflammation and injury caused by paclitaxel treatment.87 Preclinical studies with this promising radiotracer are however rare so further optimization and validation in different oncologic models are recommended. In addition, if the GST could be eliminated from the radiotracer, the molecular weight could be reduced (from 37 to 14 kDa), which would lead to more favorable biodistribution (faster renal clearance).

Radiolabeled PS-Targeting Antibodies

As apoptosis is a dynamic and transient process, with specific peaks after treatment that may vary based on therapy, tumor type, dose of therapy, and many other factors, optimal timing of imaging for detecting maximal apoptotic response will be extremely difficult with short-lived tracers. Because of the short plasma half-lives, limited accumulation of these probes may occur in the tumor, which may provide inadequate imaging signals. A solution for this problem is the use of specific PS-targeting antibodies with a long plasma-life (7 to 21 days), which may allow imaging of apoptosis over longer periods of time without the need for reinjection of the radiotracer. Many PS-targeting antibodies have been radiolabeled and tested in chemotherapy-treated tumor models and even in clinical trials.88 The chimeric PS-targeting monoclonal antibody bavituximab is currently in Phase II clinical trial. A fully human PS-targeting antibody has also been developed (PGN635) and radiolabeled with the longer living (78.4 hours) PET isotope Zirconium-89 (89Zr). It was tested as probe for detection of apoptosis in four mouse xenografts of human carcinoma (breast, colorectal, and NSCLC) treated with apoptosis-inducing agents: Paclitaxel for breast cancer and the other two models with PRO598 (an agonistic death-receptor monoclonal antibody). Radiochemical yield of 89Zr-PGN635 was 74 ± 4% and >95% radiochemical purity was obtained. At 2 days post paclitaxel treatment, the uptake of tumoral 89Zr-PGN635 was significantly higher than in the control tumors and the average tumor uptake of 89Zr-PGN635 reached 15.3 ± 2.9%ID/g at day 5. In the two PRO598 sensitive xenograft models (one colorectal and the other NSCLC model), tumor uptake of 89Zr-PGN635 at day 2 and at day 5 was significantly higher in the treated groups, than the control groups. The tumor-to-blood ratio reached 11.5 for the colorectal and 13.6 for the NSCLC xenograft model. In the PRO598 resistant colorectal xenograft model, 89Zr-PGN635 tumoral uptake was not significantly different from the control animals. For sure, these high–molecular-weight PS-targeting antibodies expand the time window in which apoptosis imaging can be performed, however they may also cause higher background levels compared to the low–molecular-weight probes.89

Small-Molecule Mimics of Annexin A5: Zinc Dipicolylamine (Zn2+-DPA)

Although radiolabeled Annexin A5 and its derivatives have clearly demonstrated their ability to image apoptosis, the in vivo applications remain limited because of low target-to-background ratios, long biologic half-life with high uptake in liver and kidney, and questionable in vivo stability.90 This initiated the development of conjugates with better biodistribution profiles. Although derivatives of Annexin A5 continue to evolve, the second-generation design mentioned earlier is limited for protein-based probes like Annexin A5. Low–molecular-weight probes are more amenable to structural optimization and display more favorable pharmacokinetic properties compared to protein-based probes like Annexin A5. Based on the x-ray crystal structure of Annexin A5, small-molecule mimics of Annexin A5 were developed. Swairjo et al.91 reported a crystal structure of Annexin A5 complexed with glycerophosphoserine in which two bridging Ca2+ ions in one of the four binding domains of the protein are coordinated to the PS headgroup, one bound to the phosphoryl oxygen and the other to the carboxylate oxygen. This Ca2+-bridging binding mechanism raised the idea that rationally designed metal coordination complexes with appropriate charge, geometry, and spatial orientation could potentially act as functional mimics of AnnexinA5.92 The group of Smith et al. discovered that synthetic Zn2+-DPA coordination complexes can mimic the apoptosis sensing function of AnnexinA5 and can thus serve as an alternative approach to PS-targeting. Similar to Ca2+ in Annexin A5 binding sites, Zn2+ ions mediate the interaction of the dipicolylamine ligand and the anionic PS headgroup.93

FIGURE 41.8. Chemical structure of PSS-794.

To enable in vivo imaging of cell death, synthetic fluorescent Zn2+-DPA probes were developed. PSS-794 (Fig. 41.8) (PSVue794, Molecular Targeting Technologies Inc., West Chester, PA) with a Zn2+-DPA affinity group conjugated to an NIRcarbocyanine fluorophore (ex. 794 nm, em. 810 nm) was shown to selectively stain the same cells as fluorescently labeled Annexin A5 in cell culture.94

FIGURE 41.9. Representative NIR fluorescence images of a mouse treated with a synthetic ionophore in the hind leg and injected with either PSS-794 (probe 2, top row) or Annexin-Vivo 750 (bottom row)via the tail vein. Both cohorts of mice were injected with ionophore in the right hind leg muscle and saline in the left hind leg muscle. The mice were dosed with either PSS-794 or Annexin-Vivo 750 2 hours post treatment. Images were acquired at the indicated time points after probe injection. The calibration bar applies to all images. (Reprinted with permission from Smith BA, Akers WJ, Leevy WM, et al. Optical imaging of mammary and prostate tumors in living animals using a synthetic near infrared zinc(II)-dipicolylamine probe for anionic cell surfaces. J Am Chem Soc.2009;132:67–69. Copyright 2011 American Chemical Society).

In an in vivo comparative study with NIR fluorescent Annexin A5 conjugate Annexin-Vivo 750 (VisEn Medical, Bedford, MA), using a mouse model of acute cell death, PSS-794 displayed a better pharmacokinetic profile with rapid clearance through the liver and higher T/NT ratio than the Annexin A5 probe (Fig. 41.9).95 Lower T/NT for the Annexin A5 probe was attributed to the high background signal because of accumulation of the probe in the bladder. This urinary clearance profile correlates with the high water solubility of Annexin A5 and its peptidic metabolites and has likewise been demonstrated for radiolabeled derivatives of Annexin A5. PSS-794 on the other hand, associates with serum proteins that promotes uptake by the reticuloendothelial system. The authors conclude that the different clearance pathways for PSS-794 (liver/intestines) and Annexin A5 (bladder/kidney) might give them value as complementary probes for imaging cell death in different anatomical locations. PSS-794 was evaluated for tumor cell death–targeting capacity in vivo in xenograft tumor models. Smith et al.94 demonstrated selective tumor accumulation of PSS-794 24 hours following i.v. injection of the probe (3 mg/kg) in immunocompetent Lobund Wistar rats with PAIII prostate tumors and athymic nude mice containing EMT-6 mammary tumors. Additional ex vivo imaging of the excised tumors and in vitro microscopy suggest preferential targeting of the necrotic region of the tumors. In another study they used PSS-794 to assess the level of therapy-induced tumor cell death.96 For this, they treated a xenograft PAIII prostate tumor in rat rear flank with 20-Gy focal beam irradiation whereas the tumor on the other flank remained untreated. Seventeen hours after irradiation, rats were injected intravenously with PSS-794 (3 mg/kg) or with 794 control (3 mg/kg) which lacks the Zn2+-DPA unit. Planar epifluorescence imaging 24 hours post probe injection demonstrated PSS-794 accumulation in the irradiated tumor that was almost double than in the nontreated tumor (Fig. 41.10). Uptake in the control tumor could be ascribed to moderate–high levels of necrotic tissue in this tumor model. No tumor uptake could be seen in either tumor of rats treated with the control dye, demonstrating the importance of the Zn2+-DPA group for targeting of exposed PS. Besides for tumor cell death targeting, several other studies demonstrate that PSS-794 can be used for in vivo imaging of cell death in animal models of tissue damage, thymus atrophy, traumatic brain injury, brain infarct, and infection (the latter through binding to apoptotic and necrotic neutrophils).9699 Although PSS-794 clearly demonstrated its PS-targeting capacities in several animal models of cell death, the usefulness of an NIR fluorescent probe remains limited to superficial lesions caused by tissue attenuation of the signal and can only supply semiquantitative information. PET and SPECT can circumvent the challenges imposed by fluorescence imaging and could give the Zn2+-DPA–based probes broad clinical applications. Therefore, radiolabeled Zn2+-DPA derivatives were recently developed and evaluated. Wyffels et al.100 reported on the development of 99mTc-labeled Zn2+-DPA probes using the HYNIC and 99mTc-(CO)3-labeling method for SPECT imaging of cell death. The labeling of Zn2+-DPA via 99mTc-tricarbonyl resulted in higher labeling yields, higher purity, and a higher stability compared to the 99mTc-HYNIC-Zn2+-DPA probe. A biodistribution study in normal mice revealed fast blood clearance and lower kidney uptake as compared to 99mTc-labeled forms of Annexin A5 with high uptake in the liver for both probes. This clearance profile is comparable to what has been described for the fluorescent Zn2+-DPA probe. Both probes were evaluated in vivo in an anti-Fas liver apoptosis mouse model and a rat model of myocardial ischemia-reperfusion injury. Although both models revealed uptake in areas of cell death, specific uptake in the rat model remained low. Liu et al.101 labeled Zn2+-DPA with 111In through a DOTA chelator and evaluated the PS-targeting properties in vitro in cultured human breast cancer BT-474 cells expressing different levels of apoptotic PS They demonstrated increased cell binding of 111In-DOTA-Zn2+-DPA to cells with higher PS expression. No in vivo evaluation was published so far. For PET imaging purposes, 18F-labeled Zn2+-DPA derivates were developed. Li et al.102 described the labeling of DPA using different 18F-prosthetic groups (18F-SFB, 18F-NFP, 18F-FET), without the Zn complexation step. Although the DPA unit could be labeled successfully using either prosthetic group, in vitro or in vivo evaluation of the 18F-labeled DPA complexed with Zn was not yet reported. In a meeting abstract, Chen et al.103 reported on 64Cu labeling of DPA using a direct labeling approach in which the DPA unit was coordinated with 64Cu instead of Zn. The probe was evaluated in vivo in a U87MG xenograft tumor mouse model and displayed good tumor uptake and high tumor-to-muscle ratio (14.99 at 24 hours p.i.). High liver accumulation of the tracer was observed.

FIGURE 41.10. Representative planar epifluorescence images of rats bearing two subcutaneous PAIII prostate tumors and dosed with PSS-794 (A in vivo, B ex vivo) or 794-control (C in vivo, D ex vivo) after radiation therapy. The right flank tumor (arrow) received focal beam radiation therapy, and the left flank tumor (arrow head) was not treated. At 17 hours after radiation, each rat was injected intravenously with probe (3 mg/kg), and 24 hours later the in vivo image was acquired. The 794-control does not target the tumors but accumulates in the kidneys (n = 5). (Reprinted from Smith BA, Xiao S, Wolter W, et al. In vivo targeting of cell death using a synthetic fluorescent molecular probe. Apoptosis. 2011;16:722–731 with kind permission from Springer Science+Business Media).

Although increasing effort has been applied into the development of small-molecule mimics of Annexin A5 for apoptosis imaging with promising cell death–targeting properties, no clinically translatable tracer has been discovered yet. Additional in vivo evaluation including blocking studies to demonstrate specificity of the signal of the Zn2+-DPA analogs developed so far is warranted as well as optimization of the pharmacokinetics.

PS-BINDING PEPTIDES: PHAGE PANNING TECHNOLOGY

Proteins and antibodies have a relatively large molecular weight, which may cause unfavorable biodistribution, slow diffusion rates, and slow blood clearance rates. For these reasons, it would be advantageous to develop small molecules (peptides) with the same target specificity as proteins but with a lower molecular weight. In addition, the radiolabeling and pharmacokinetics of these small molecules are somewhat easier to fine tune. To discover PS-binding peptides that can be used to image apoptosis, M13 phage display peptide library was screened onto PS-coated ELISA plates. The phage with the highest affinity-bound PS in ELISA carried the peptide TLVSSL.104 Another peptide, with sequence CLSYYPSYC was discovered and tumor-bearing mice treated with a single dose of an anticancer drug (camptothecin) and fluorescein-labeled CLSYYPSYC peptide showed peptide homing to the tumor. This peptide represents a novel PS-recognizing agent and may be used in the future as radiolabeled or fluorescent probe to detect apoptosis in vivo.105

CLINICAL VIEWPOINTS ON PCD IMAGING

From a clinical point of view, noninvasive imaging of PCD creates promising opportunities in oncology. A biomarker-targeting PCD may help in the early detection of cancers, monitoring of disease course, or the assessment of the effectiveness of cancer therapy. So far, no probe for imaging of PCD has made it into routine clinical practice. A valuable probe for clinical molecular imaging of PCD must successfully incorporate several distinct qualities. First, the radiotracer should have high selectivity and specificity for cells undergoing PCD in the early stage of the death process. Second, the probe should be suitable for in vivo use. This includes a favorable biodistribution on intravenous administration, a rapid distribution throughout the body and a rapid clearance from nontarget tissues with a favorable signal-to-background ratio. In addition, the tracer should be able to detect PCD in vivo. Third, the probe should be safe (nontoxic) and chemically and immunologically inert. Fourth, the probe should have high stability in vivo, with minimal metabolism. Ultimately, the probe should be compatible with routine clinical practice. Radiolabeling should be easy to perform, fast, reproducible, and not require an additional purification step. Although radiolabeling Annexin A5 with positron emitters (like 18F) is preferred to single positron emitters in an image point of view (better image quality and quantification), the radiolabeling presents more difficulties regarding rapid and efficient radiolabeling.106 A radiotracer imaging PCD meeting all this needs for routine clinical application has not yet been discovered, however the search is still ongoing.

IMAGING PCD: CHALLENGES AND CONSIDERATIONS

Although there has been considerable progress over the last 5 years, in vivo imaging of apoptosis has still not reached clinical practice: Some challenges and considerations must be taken into account. Apoptosis is a dynamic process in which newly generated apoptotic bodies are rapidly engulfed by phagocytes. Therefore, there is a short window for the detection of apoptotic cells. As apoptosis occurs over an interval of time after administration of apoptosis-inducing therapy in tumors, the best timing for detection of apoptosis in vivo must be determined. In general, the increase in apoptosis in tumors appears to be heavily dependent on the exact time after start of the chemotherapy. The time window in which peak(s) of activity occurs, is only a matter of hours and thus timing of apoptosis imaging after a given therapy is of paramount importance.21,48,49 Moreover, many factors affect the timeline of apoptotic events, including the cell line used for the xenograft model, or tissue examined apoptosis-inducing agent, drug concentration, and exposure time. The final goal of apoptosis imaging in clinical setting is to evaluate the efficacy of anticancer therapy shortly after the start of the therapeutic regimen. Contradicting results previously reported may be because of an inadequate interval chosen between the administration of the therapy and imaging, so the peak of activity is missed. Consequently, in the clinical setting it is crucial to initially characterize the time course of the apoptosis radiotracer uptake so the peak of apoptosis is detected and early tumor response to therapy can be evaluated. In addition, some limitations of the promising clinical studies herein described need to be addressed. First, the group of patients is always relatively small (<40) and quite heterogeneous. Caution must to be taken when extrapolating the results to a general oncologic population. Second, quantification of the radiolabeled Annexin A5 uptake was not standardized between the different studies and in some cases remained rather subjective. Finally, radiolabeled Annexin A5 tracers do not exclusively detect apoptotic cell death, but also many other forms of PCD. So, rather than claiming that radiolabeled Annexin A5 uptake in the tumors represent apoptotic cell death, it is more accurate to state that it detects and quantifies an overall cytotoxic effect in the tumor (including necrosis, autophagy, mitotic catastrophe). Larger clinical trials of more homogeneous cancer patients, with a standardized quantification method are therefore indispensable.

CONCLUSIONS

Molecular imaging of cell death in oncology has gained much interest to assess tumor response to cancer therapy. Inducing apoptosis (or other forms of PCD) of tumor cells is the main goal of treatment in oncology. Regrettably, treatment is not always effective and many of the therapies may cause (severe) side effects. Ineffective and toxic therapy should be avoided. The most effective treatment should be determined for each individual patient. The final goal of apoptosis imaging is therefore to achieve “personalized medicine.” However, this remains a big challenge because this strategy requires rapid information on the effectiveness of a drug in individual patients, and demands a rapid and reliable readout of treatment efficacy. Apoptosis is a very attractive cellular process for several reasons: (1) Induction of apoptosis is among the earliest events in the tumor-killing effect of anticancer therapy, (2) induction of apoptosis is a universal mode of action of anticancer therapies, (3) apoptosis generates a positive signal, and (4) it is directly related to the therapeutic intervention. To date, PS is the most promising biomarker to measure the extent of cell death as an early readout of efficacy of anticancer therapy. Annexin A5 is the ligand with the highest affinity for PS. It has been used in molecular imaging protocols to measure cell death in vitro and in vivo in preclinical and clinical studies. Experience with Annexin A5 indicates that aspects such as biodistribution and target-to-background ratio require further improvement. An approach is to stick to Annexin A5 and improve biodistribution and target-to-background ratio by changing Annexin A5’s primary structure and the label that is coupled to Annexin A5. The pretargeting strategy also shows promise to achieve better target-to-background ratio. The nuclear imaging methods, PET and SPECT, represent probably the most sensitive method for molecular imaging of apoptosis. Both imaging techniques have a very high sensitivity (in the picomolar range), unlimited depth penetration, an excellent signal-to-noise background ratio, and a broad range of clinical apoptosis-targeting probes are currently available. Both preclinical and clinical studies support the notion that 99mTc-HYNIC-Annexin A5 and SPECT imaging allows for noninvasive, repetitive, quantitative apoptosis imaging and in assessing tumor response as early as 24 hours following treatment instigation. As PET offers higher resolution, higher sensitivity and more accurate quantification compared to SPECT, Annexin A5 has also been labeled with 18F, 64Cu, and 124I. However, these labeling procedures proved laborious and time-consuming and the use of directly iodinated Annexin A5 in a clinical setting has proven to be not suitable. More recently, attempts have been made to label Annexin A5 with 68Ga for PET as the short half-life of this radionuclide minimizes the radiation dose to the patient and allows for rapid sequential imaging. Preclinical and clinical studies using these agents are needed.

FUTURE CONSIDERATIONS

Clinical imaging of apoptosis has undergone a dynamic evolution in the past decade, from SPECT imaging with 99mTc-Annexin A5, to PET imaging of apoptosis and finally the development of novel small-molecule structures detecting new targets of the apoptotic cascade. Radiolabeled caspase inhibitors/substrates have been developed for PET and SPECT apoptosis imaging because of their central role in the early phase of the apoptotic process. Available in vitro and preclinical studies suggest that these radiopharmaceuticals bear great potential for apoptosis imaging by means of PET and SPECT. However, thorough toxicologic studies are required before they can be applied in clinical studies. Finally, a novel unclassified tracer, called the ApoSense family (e.g., 18F-NST-732 and 18F-ML-10) was shown to selectively accumulate in apoptotic cells by an unknown mechanism. Although agents of the ApoSense family have shown very promising features, further investigation is warranted to further unravel their mechanism of action and applicability in human studies for predicting therapy response. A novel promising SPECT/PET apoptosis tracer that has recently been described is 99mTc/18F-duramycin. Duramycin is the smallest known polypeptide of 19 amino acids and binds the membrane phospholipid phosphatidylethanolamine that becomes externalized to the surface of apoptotic cells in a way similar to PS. The favorable in vivo imaging profiles of this new agent warrant further development and characterization. To conclude, during the past decade, a lot of progress has been made in the field of apoptosis research and novel apoptosis-targeted anticancer drugs have been developed. However, the entire mechanism behind the apoptotic cascade is still unknown and undoubtedly new apoptosis markers will soon be revealed. In the future, rapid progress in this important and exciting field of clinical research can be expected.

REFERENCES

1. Gerl R, Vaux DL. Apoptosis in the development and treatment of cancer. Carcinogenesis. 2005;26:263–270.

2. Reed JC. Dysregulation of apoptosis in cancer. J Clin Oncol. 1999;17:2941–2953.

3. de Bruin EC, Medema JP. Apoptosis and non-apoptotic deaths in cancer development and treatment response. Cancer Treat Rev. 2008;34:737–749.

4. Vermes I, Haanen C, Steffens-Nakken H, et al. A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V. J Immunol Methods. 1995;184:39–51.

5. Zwaal RF, Comfurius P, Bevers EM. Surface exposure of phosphatidylserine in pathological cells. Cell Mol Life Sci. 2005;62:971–988.

6. Corsten MF, Hofstra L, Narula J, et al. Counting heads in the war against cancer: Defining the role of annexin A5 imaging in cancer treatment and surveillance. Cancer Res. 2006;66:1255–1260.

7. Hengartner MO. The biochemistry of apoptosis. Nature. 2000;407:770–776.

8. Bevers EM, Comfurius PC, Dekkers DWC, et al. Lipid translocation across the plasma membrane of mammalian cells. Biochim Biophys Acta. 1999;1439:317–330.

9. Blankenberg FG. Imaging the molecular signature of apoptosis and injury with radiolabeled Annexin V. Proc Am Thorac Soc. 2009;6:469–476.

10. Gerke V, Moss SE. Annexins: From structure to function. Physiol Rev. 2002;82: 331–371.

11. Kemerink GJ, Boersma HH, Thimister PWL, et al. Biodistribution and dosimetry of 99mTc-BTAP-Annexin A5 in humans. Eur J Nucl Med. 2001;28:1373–1378.

12. Boersma HH, Liem IH, Kemerink GJ, et al. Comparison between human pharmacokinetics and imaging properties of two conjugation methods for 99mTc-Annexin A5. Br J Radiol. 2003;76:553–560.

13. Belhocine T, Steinmetz N, Hustinx R, et al. Increased uptake of the apoptosis-imaging agent 99mTc recombinant human Annexin A5 in human tumours after one course of chemotherapy as a predictor of tumour response and patient prognosis. Clin Cancer Res. 2002;8:2766–2774.

14. Belhocine T, Steinmetz N, Green A, et al. In vivo imaging of chemotherapy-induced apoptosis in human cancers. Ann NY Acad Sci. 2003;1010:525–529.

15. Yang DJ, Azdarinia A, Wu P, et al. In vivo and in vitro measurement of apoptosis in breast cancer cells using 99mTc-EC-Annexin A5. Cancer Biother Radiopharm. 2001;16:73–83.

16. Kurihara H, Yang DJ, Cristofanilli M, et al. Imaging and dosimetry of 99mTc-EC-Annexin A5: Preliminary clinical study targeting apoptosis in breast tumours. Appl Radiat Isot. 2008;66:1175–1182.

17. Vanderheyden JL, Liu G, He J, et al. Evaluation of 99mTc-MAG3-Annexin A5: Influence of the chelate on in vitro and in vivo properties in mice. Nucl Med Biol. 2006;33:135–144.

18. Blankenberg FG, Katsikis PD, Tait JF, et al. In vivo detection and imaging of phosphatidylserine expression during programmed cell death. Proc Natl Acad Sci USA. 1998;95:6349–6354.

19. Blankenberg FG, Katsikis PD, Tait JF, et al. Imaging of apoptosis (programmed cell death) with 99mTc annexin V. J Nucl Med. 1999;40:184–191.

20. Ohtsuki K, Akashi K, Aoka Y, et al. Technetium-99m HYNIC-annexin V: A potential radiopharmaceutical for the in-vivo detection of apoptosis. Eur J Nucl Med. 1999; 26:1251–1258.

21. Blankenberg F. To scan or not to scan, it is a question of timing: Technetium-99m-Annexin A5 radionuclide imaging assessement of treatment efficacy after one course of chemotherapy. Clin Cancer Res. 2002;8:2757–2758.

22. Takei T, Kuge Y, Zhao S, et al. Time course of apoptotic tumor response after a single dose of chemotherapy: Comparison with 99mTc-annexin V uptake and histologic findings in an experimental model. J Nucl Med. 2004;45:2083–2087.

23. Erba PA, Manfredi C, Lazzeri E, et al. Time course of Paclitaxel-induced apoptosis in an experimental model of virus-induced breast cancer. J Nucl Med. 2010;51:775–781.

24. Wong E, Kumar V, Howman-Giles RB, et al. Imaging of therapy-induced apoptosis using 99mTc-HYNIC-annexin V in thymoma tumor-bearing mice. Cancer Biother Radiopharm. 2008;23:715–725.

25. Guo M-f, Zhao Y, Tian R, et al. In vivo 99mTc-HYNIC-annexin V imaging of early tumor apoptosis in mice after single dose irradiation. J Exp Clin Cancer Res. 2009; 28:136.

26. Kemerink GJ, Liu X, Kieffer D, et al. Safety, biodistribution, and dosimetry of 99mTc-HYNIC-annexin V, a novel human recombinant annexin V for human application. J Nucl Med. 2003;44:947–952.

27. Vermeersch H, Loose D, Lahorte C, et al. 99mTc-HYNIC-Annexin A5 imaging of primary head and neck carcinoma. Nucl Med Commun. 2004;25:259–263.

28. Van de Wiele C, Lahorte C, Vermeersch H, et al. Quantitative tumour apoptosis imaging using technetium-99m-HYNIC annexin A5 single photon emission computed tomography. J Clin Oncol. 2003;21:3483–3487.

29. Vermeersch H, Ham H, Rottey S, et al. Intraobserver, and day-to-day reproducibility of quantitative 99mTc-HYNIC-Annexin A5 imaging in head and neck carcinoma. Cancer Biother Radiopharm. 2004;19:205–210.

30. Rottey S, Van de Bossche B, Slegers G, et al. Influence of chemotherapy on the biodistribution of 99mTc-Hydrazinonicotinamide-Annexin A5 in patients. Q J Nucl Med Mol Imaging. 2009;53:127–132.

31. Kartachova M, Haas RLM, Valdés Olmos RA. In vivo imaging of apoptosis by 99mTc-Annexin A5 scintigraphy: Visual analysis in relation to treatment response. Radiol Oncol. 2004;72:33–39.

32. Kartachova M, van Zandwijk N, Burgers S, et al. Prognostic significance of 99mTc-HYNIC-rh-annexin A5 scintigraphy during platinum-based chemotherapy in advanced lung cancer. J Clin Oncol. 2007;25:2534–2539.

33. Kartachova M, Valdés Olmos RA, Haas RLM, et al. 99mTc-HYNIC-rh-Annexin A5 scintigraphy: Visual and quantitative evaluation of early treatment-induced apoptosis to predict treatment outcome. Nucl Med Commun. 2008;29:39–44.

34. Rottey S, Slegers G, Van Belle S, et al. Sequential 99mTc-hydrazinonicotinamide-annexin A5 imaging for predicting response to chemotherapy. J Nucl Med. 2006; 47:1813–1818.

35. Haas RLM, de Jong D, Valdés Olmos RA, et al. In vivo imaging of radiation-induced apoptosis in follicular lymphoma patients. Int J Radiat Oncol Biol Phys. 2004;59:782–787.

36. Hoebers FJP, Kartachova M, de Bois J, et al. 99mTc HYNIC-rh-Annexin A5 scintigraphy for in vivo imaging of apoptosis in patients with head and neck cancer treated with chemoradiotherapy. Eur J Nucl Med Mol Imaging. 2008;35:509–518.

37. Ono M, Arana Y, Uehara T, et al. Intracellular metabolic fate of radioactivity after injection of technetium-99m-labeled hydrazine nicotinamide derivatized proteins. Bioconjug Chem. 1999;10:386–394.

38. Banerjee SR, Maresca KP, Francesconi L, et al. New directions in the coordination chemistry of 99mTc: A reflection on technetium core structures and a stragey for new chelate design. Nucl Med Biol. 2005;32:1–20.

39. Biechlin M-L, Bonmartin A, Gilly F-N, et al. A. Radiolabeling of annexin A5 with 99mTc: Comparison of Hynic-Tc vs. iminothialane-Tc-tricarbonyl conjugates. Nucl Med Biol. 2008;35:679–687.

40. Waibel R, Alberto R, Willuda J, et al. Stable one-step technetium-99m labeling of His-tagged recombinant proteins with a novel Tc(I)- carbonyl complex. Nat Biotechnol. 1999;17:897–901.

41. Sterin-Speziale N, Kahane VL, Setton CP, et al. Compartmental study of the rat renal phospholipids metabolism. Lipids. 1992;27:10–14.

42. Lang L, Jagoda E, Wu CH. Factors influencing the in vivo pharmacokinetics of peptides and antibody fragments: The pharmacokinetics of two PET-labeled low molecular weights proteins. Q J Nucl Med. 1997;41:53–61.

43. Subbarayan M, Häfeli UO, Feyes DK, et al. A simplified method for preparation of 99mTc-annexin V and its biological evaluation for in vivo imaging of apoptosis after photodynamic therapy. J Nucl Med. 2003;44:650–656.

44. Tait F, Cerqueira MD, Dewurst TA, et al. Evaluation of annexin V as a platelet-directed thrombus targeting agent. Thromb Res. 1994;75:491–501.

45. Kemerink GJ, Liem IH, Hofstra L, et al. Patient dosimetry of intravenously administered 99mTc-annexin V. J Nucl Med. 2001;42:382–387.

46. Greupink R, Sio CF, Ederveen A, et al. Evaluation of a 99mTc-labeled annexin A5 variant for non-invasive SPECT imaging of cell death in liver, spleen and prostate. Pharm Res. 2009;26:2647–2656.

47. Vangestel C, Peeters M, Oltenfreiter R, et al. In vitro and in vivo evaluation of [99mTc]-labeled tricarbonyl His-annexin A5 as an imaging agent for the detection of phosphatidylserine – expressing cells. Nucl Med Biol. 2010;37:965–975.

48. Vangestel C, Van de Wiele C, Mees G, et al. SPECT imaging of the early time course of therapy-induced cell death using 99mTc-(CO)3 His-Annexin A5 in a colorectal cancer xenograft model. Mol Imaging. 2012;11:135–147.

49. Vangestel C, Van de Wiele C, Van Damme N, et al. (99)mTc-(CO)(3) His-annexin A5 micro-SPECT demonstrates increased cell death by irinotecan during the vascular normalization window caused by bevacizumab. J Nucl Med. 2011;52:1786–1794.

50. Lin K-J, Wu C-C, Pan Y-H, et al. In vivo imaging of radiation-induced tissue apoptosis by 99mTc-(I)–His6-annexin A5. Ann Nucl Med. 2012;26:272–280.

51. Lahorte CMW, Slegers G, Philippé J, et al. Synthesis and in vitro evaluation of 123I-labelled human Annexin A5. Biomol Eng. 2001;17:51–53.

52. Lahorte CMW, Van de Wiele C, Bacher K, et al. Biodistribution and dosimetry study of I-123-rh-Annexin A5 in mice and humans. Nucl Med Commun. 2003;24: 871–880.

53. Cornelissen B, Lahorte CMW, Kersemans V, et al. In vivo apoptosis detection with radioiodinated Annexin A5 in LoVo tumour-bearing mice following Tipifarnib (Zarnestra, R115777) farnesyltransferase inhibitor therapy. Nucl Med Biol. 2005;32:233–239.

54. Zavitsanou K, Nguyen V, Greguric I, et al. Detection of apoptotic cell death in the thymus of dexamethasone treated rats using [I123]Annexin A5 and in situ oligonucleotide ligation. J Mol Hist. 2007;38:313–319.

55. Russell J, O’Donoghue JA, Finn R, et al. Iodination of Annexin A5 for imaging apoptosis. J Nucl Med. 2002;43:671–677.

56. Ke S, Wen X, Wu Q-P, et al. Imaging taxane-induced tumour apoptosis using PEGylated, 111In-labelled Annexin A5. J Nucl Med. 2004;45:108–115.

57. Zhang R, Lu W, Wen X, et al. Annexin A5-conjugated polymeric micelles for dual SPECT and optical detection of apoptosis. J Nucl Med. 2011;52:958–964.

58. Tait JF, Brown DS, Gibson DF, et al. Development and characterization of annexin V mutants with endogenous chelation sites for (99m)Tc. Bioconjug Chem. 2000;11: 918–925.

59. Kuge Y, Sato M, Zhao S, et al. Feasibility of 99mTc-annexin V for repetitive detection of apoptotic tumor response to chemotherapy: An experimental study using a rat tumor model. J Nucl Med. 2004;45:309–312.

60. Mochizuki T, Kuge Y, Zhao S, et al. Detection of apoptotic tumor response in vivo after a single dose of chemotherapy with 99mTc-annexin V. J Nucl Med. 2003;44:92–97.

61. Tait JF, Smith C, Gibson DF. Development of annexin V mutants suitable for labeling with Tc(I)-carbonyl complex. Bioconjug Chem. 2002;13:1119–1123.

62. Tait JF, Smith C, Blankenberg FG. Structural requirements for in vivo detection of cell death with 99mTc-Annexin A5. J Nucl Med. 2005;46:807–815.

63. Fonge H, de Saint Hubert M, Vunckx K, et al. Preliminary in vivo evaluation of a novel 99mTc-labelled HYNIC-cys-Annexin A5 as an apoptosis imaging agent. Bioorg Med Chem Lett. 2008;18:3794–3798.

64. De Saint-Hubert M, Mottaghy FM, Vunckx K, et al. Site-specific labeling of ‘second generation’ annexin V with 99mTc(CO)3 for improved imaging of apoptosis in vivo. Bioorg Med Chem. 2010;18:1356–1363.

65. De Saint-Hubert M, Wang H, Devos E, et al. Preclinical imaging of therapy response using metabolic and apoptosis molecular imaging. Mol Imaging Biol. 2011;13:995–1002.

66. Zijlstra S, Gunawan J, Burchert W. Synthesis and evaluation of a 18F-labelled recombinant Annexin A5 derivative, for identification and quantification of apoptotic cells with PET. Appl Radiat Isot. 2003;58:201–207.

67. Toretsky T, Levenson A, Weinberg IN, et al. Preparation of F-18 labeled annexin V: A potential PET radiopharmaceutical for imaging cell death. Nucl Med Biol. 2004;31:747–752.

68. Murakami Y, Takamatsu H, Taki J, et al. 18F-labelled annexin V: A PET tracer for apoptosis imaging. Eur J Nucl Med Mol Imaging. 2004;31:469–473.

69. Yagle KJ, Eary JF, Tait JF, et al. Evaluation of 18F-Annexin A5 as a PET imaging agent in an animal model of apoptosis. J Nucl Med. 2005;46:658–666.

70. Hu S, Kiesewetter DO, Zhu L, et al. Longitudinal PET imaging of doxorubicin-induced cell death with 18F-Annexin V. Mol Imaging Biol. 2012;14:762–770.

71. McQuade P, Jones LA, Vanderheyden JL, et al. 94mTc and 64Cu labelled annexin-V positron emitting radiopharmaceuticals to study apoptosis. J Labelled Comp Radiopharm. 2003;46(suppl 1):S335.

72. Cauchon N, Langlois R, Rousseau JA, et al. PET imaging of apoptosis with 64Cu-labelled streptavidin following pretargeting of phosphatidylserine with biotinylated Annexin A5. Eur J Nucl Med Mol Imaging. 2007;34:247–258.

73. Glaser M, Collingridge DR, Aboagye EO, et al. Iodine-124 labelled Annexin A5 as a potential radiotracer to study apoptosis using positron emission tomography. Appl Radiat Isot. 2003;58:55–62.

74. Keen HG, Dekker BA, Disley L, et al. Imaging apoptosis in vivo using 124I Annexin A5 and PET. Nucl Med Biol. 2005;32:395–402.

75. Collingridge DR, Glaser M, Osman S, et al. In vitro selectivity, in vivo distribution and tumour uptake of Annexin A5 radiolabelled with positron emitting radioisotopes. Br J Cancer. 2003;89:1327–1333.

76. Dekker B, Keen H, Shaw D, et al. Functional comparison of Annexin A5 analogues labelled indirectly and directly with iodine-124. Nucl Med Biol. 2005;32:403–413.

77. Dekker B, Keen H, Lyons S, et al. MBP-Annexin A5 radiolabelled directly with iodine-124 can be used to image apoptosis in vivo using PET. Nucl Med Biol. 2005;32:241–252.

78. Chen Q, Stone-Elander S, Arnér ESJ. Tagging recombinant proteins with a Sel-Tag for purification, labelling, with electropholic compounds or radiolabeling with carbon-11. Nat Protocols. 2006;1:604–613.

79. Cheng Q, Li L, Grafström J, et al. Combining [11C]-AnxA5 PET imaging with serum biomarkers for improved detection in live mice of modest cell death in human solid tumors xenografts. PLoS One. 2012;8:e42151.

80. Wängler C, Wängler B, Lehner S, et al. A universally applicable 68Ga-labeling technique for proteins. J Nucl Med. 2011;52:586–591.

81. Bauwens M, De Saint-Hubert M, Devos E, et al. Site-specific 68Ga-labeled Annexin A5 as a PET imaging agent for apoptosis. Nucl Med Biol. 2011;38:381–392.

82. Luo QY, Wang F, Zhang ZY, et al. Preparation and bioevaluation of 99mTc-HYNIC-B1 as a novel radioligand for apoptosis imaging. Apoptosis. 2008;13:600–608.

83. Mukherjee A, Kothari K, Toth G, et al. 99mTc-labeled annexin V fragments: A potential SPECT radiopharmaceuticals for imaging cell death. Nucl Med Biol. 2006; 33:635–643.

84. Smith BA, Smith BD. Biomarkers and molecular probes for cell death imaging and targeted therapeutics. Bioconjug Chem. 2012;23:1989–2006.

85. Tavare R, De Rosales RTM, Blower PJ, et al. Efficient site-specific radiolabeling of a modified C2A comain of synaptotagmin I with [99mTc-(CO)(3)](+): A new radiopharmaceutical for imaging cell death. Bioconjug Chem. 2009;11:2071–2081.

86. Wang F, Fang W, Zhao M, et al. Imaging paclitaxel (chemotherapy)-induced tumor apoptosis with 99mTc C2A, a domain of synaptotagmin I: A preliminary study. Nucl Med Biol. 2008;35:359–364.

87. Wang F, Fang W, Zhang M-R, et al. Evaluation of chemotherapy response in VX2 rabbit lung cancer with 18F-labeled C2A domain of synaptotagmin I. J Nucl Med. 2011;52:592–599.

88. Jennewein M, Lewis MA, Zhao D, et al. Vascular imaging of solid tumors in rats with a radioactive arsenic-labeled antibody that binds exposed phosphatidylserine. Clin Cancer Res. 2008;14:1377–1385.

89. Ogasawara A, Tinianow JN, Vanderbilt AN, et al. ImmunoPET imaging of phosphatidylserine in pro-apoptotic therapy treated tumor models. Nucl Med Biol. 2013;40(1):15–22.

90. De Saint-Hubert M, Prinsen K, Mortelmans L, et al. Molecular imaging of cell death. Methods. 2009;48:178–187.

91. Swairjo MA, Concha NO, Kaitzel MA, et al. Ca2+-bridging mechanism and phospholipid head group recognition in the membrane-binding protein annexin V. Nat Struct Biol. 1995;2:968–974.

92. Koulov AV, Stucker KA, Lakshmi C, et al. Detection of apoptotic cells using a synthetic fluorescent sensor for membrane surfaces that contain phosphatidylserine. Cell Death Differ. 2003;10:1357–1359.

93. Hanshaw RG, Smith BD. New reagents for phosphatidylserine recognition and detection of apoptosis. Bioorg Med Chem. 2005;13:5035–5042.

94. Smith BA, Akers WJ, Leevy WM, et al. Optical imaging of mammary and prostate tumors in living animals using a synthetic near infrared zinc(II)-dipicolylamine probe for anionic cell surfaces. J Am Chem Soc. 2009;132:67–69.

95. Smith BA, Gammon ST, Xiao S, et al. In vivo optical imaging of acute cell death using a near-infrared fluorescent zinc-dipicolylamine probe. Mol Pharm. 2011;8:583–590.

96. Smith BA, Xiao S, Wolter W, et al. In vivo targeting of cell death using a synthetic fluorescent molecular probe. Apoptosis. 2011;16:722–731.

97. Smith BA, Xie BW, van Beek ER, et al. Multicolor fluorescence imaging of traumatic brain injury in a cryolesion mouse model. ACS Chem Neurosci. 2012;7:530–537.

98. Chu C, Huang X, Chen C, et al. In vivo imaging of brain infarct with the novel fluorescent probe PSVue 794 in a rat middle cerebral artery occlusion-reperfusion model. Mol Imaging. 2012; published online.

99. Thakur ML, Zhang K, Paudyal B, et al. Targeting apoptosis for optical imaging of infection. Mol Imaging Biol. 2012;14:163–171.

100. Wyffels L, Gray BD, Barber C, et al. Synthesis and preliminary evaluation of radiolabeled bis(zinc(II)-dipicolylamine) coordination complexes as cell death imaging agents. Bioorg Med Chem. 2011;19:3425–3433.

101. Liu X, Wang Y, Gray B, et al. 111In-DOTA-DPA-Zn(II) binding with increasing of phosphatidylserine expression in cells. J Nucl Med. 2011;52(S1):1604.

102. Li J, Gray BD, Pak KY, et al. Optimization of labeling dipicolylamine derivative, N,N′-(5-(4-aminobutoxy)-1,3-phenylene)bis(methylene)bis(1-(pyridin-2-yl)-N-(pyridin-2-ylmethyl)methanamine), with three 18F-prosthetic groups as potential imaging agents for metastatic infectious disease. J Labelled Comp Radiopharm. 2012;55:149–154.

103. Chen K, Li-Peng Y, Park R, et al. Evaluation of 64Cu-labeled dipicolylamine (DPA) as a small-molecule PET probe for in vivo imaging of phosphatidylserine exposure. J Nucl Med. 2011;52(S1):1502.

104. Laumonier C, Segers J, Laurent S, et al. A new peptide vector for molecular imaging of apoptosis identified by phage display technology. J Biomol Screen. 2006;11:537–545.

105. Thapa N, Kim S, So IS, et al. Discovery of a phosphatidylserine recognizing peptide and its utility in molecular imaging of tumour apoptosis. J Cell Mol Med. 2008;12:1649–1660.

106. Reshef A, Shirvan A, Akselrod-Ballin A, et al. Small-molecule biomarkers for clinical PET imaging of apoptosis. J Nucl Med. 2010;51:837–840.