PET Molecular Imaging in Drug Development: The Imaging and Chemistry Perspective

Positron emission tomography with selective radioligands advances the drug discovery and development process by revealing information about target engagement, proof of mechanism, pharmacokinetic and pharmacodynamic profiles. Positron emission tomography (PET) is an essential and highly significant tool to study therapeutic drug development, dose regimen, and the drug plasma concentrations of new drug candidates. Selective radioligands bring up target-specific information in several disease states including cancer, cardiovascular, and neurological conditions by quantifying various rates of biological processes with PET, which are associated with its physiological changes in living subjects, thus it reveals disease progression and also advances the clinical investigation. This study explores the major roles, applications, and advances of PET molecular imaging in drug discovery and development process with a wide range of radiochemistry as well as clinical outcomes of positron-emitting carbon-11 and fluorine-18 radiotracers.


INTRODUCTION
The selective radioligands used for positron emission tomography molecular imaging have vastly increased over the past decade because of many advances in biomedical research as well as clinical investigation (1). It has been estimated that radiopharmaceuticals with 39.1% and radiodiagnostics with 7% were used in the 5 years from 2020 to 2025 by the MEDrays Intell nuclear medicine world market report 2019 (2). Positron emission tomography (PET) molecular imaging with radioligand acts like precision pharmacology, provide accurate information related to therapeutic drug development in living subjects in vivo like determining dose and dosing interval, a biomarker for drug efficacy, identifying homogeneous groups whether a new drug reaches the specific region of interest in the body with an effective ED 50 concentration, specific/non-specific receptor binding, receptor occupancy, quantitative relationship, brain penetration, metabolism in the brain as well as peripheral, which could enhance clinical trials and biomedical research involving animal studies with more reliable data (3). The PET imaging modality visualizes and measures rates of various physiological processes at the molecular level in vivo such as glucose metabolism, neurotransmitters, uptakes, and blood flow, etc. in cancer, cardiovascular, and various neurological conditions (4). The basic principle of PET is the detection of two annihilated photons traveling in opposite directions at 180 • through the collision of an emitted positron from a radiotracer with a local electron in the living system, followed by giving medical images which map the location of radiotracer in the body (5). The PET is very sensitive because it detects photons from tracer quantity of radioligand in picomolar to nanomolar range, and which do not show any pharmacological responses due to lower concentration than ED 50 values, and the radiation exposure to a patient is also minimal (6). Its major limitation is to provide low anatomical information, whereas CT, and MRI provide good anatomical data, therefore PET is combined with other molecular imaging modalities and emerged as multimodality devices like PET-CT, and PET-MRI to provide both functional and anatomical data simultaneously. The spatial resolution of PET is 2-6 mm but the MRI has <<1 mm, whereas the sensitivity of PET is 10 −12 M, but the MRI has only 10 −4 M, hence it has higher significance in the field of medical imaging (7). To enable good performance in vivo, each radioligand should meet the criteria which include lipophilicity (log D) in the range of 1 ≤ log D ≤ 3, molecular weight < 500 daltons, number of hydrogen bond donors is <1, No active P-glycoprotein-mediated efflux pump, Affinity based on dissociation/inhibition constant (K d /K i ) is <10 nM, and the receptor density to affinity ratio should be >10 (8). Several fluorine-18 and carbon-11 radiotracers are being used currently in various biomedical research programs (preclinical studies) as well as in clinical PET centers (clinical studies), in which the fluorine-18 radiotracers include 18 Ffluorodeoxyglucose for glucose metabolism, 18 F-fluorothymidine for tumor cell proliferation, 18 F-fluoromizonidazole for tumor hypoxia, 18 F-fluoromisonidazole for myocardial ischemia, 18 Ffluorodopa for Dopamine neuronal densities, 18 F-flumazenil for benzodiazepine receptors densities, 18 F-estradiol for estrogen receptor expression, whereas the carbon-11 radiotracers include 11 C-methionine for amino acid transport and protein synthesis, 2-[ 11 C]thymidine for tumor cell proliferation, 11 C-acetate for the tricarboxylic acid cycle, [ 11 C]meta-hydroxyephedrine for presynaptic catecholamine reuptake, 11 C-raclopride for dopamine receptor densities, 11 C-methylspiperone for 5-HT2 receptor density, and 11 C-PK11195 for neuronal microglial activation (9,10). A generalized view of radiochemistry and PET molecular imaging studies has been shown in Figure 1, where the initial step focuses on generating positron-emitting radionuclides like carbon-11 and fluorine-18 using different targets in a cyclotron, the second step is critical and it involves the radiochemical synthesis by transferring radioactivity into either automated modules like GE Tracerlab MX, FX 2 C, FX 2 N or hot cell apparatus where the mixture of selective precursor for radiolabeling and solvents are present in a heating vial, and this step should adhere with cGMP compliances for human applications, and the third step focuses on the preparation of suitable radiopharmaceutical formulation for injection into a living subject after conformation with quality control studies like pH, endotoxin levels, purity, molar activity, solvent levels, etc. (11). The fourth step is the PET imaging data acquisition by setting up required parameters for different studies like static or dynamic PET studies, where simultaneous arterial blood sampling is required for dynamic PET study to quantify the available receptor concentration in living subjects. The majority of clinical PET centers use static PET studies just to monitor the physiological changes in different diseases conditions like oncology, neurology, and cardiology. The fifth step is the PET data pre-and post-processing for both static and dynamic studies, the parameters include motion correction, coregistration of MRI with PET, spatial normalization, generation of the region of interest (ROI), and partial volume correction (PVC). The static PET data analysis gives standardized uptake values (SUV), which differentiate patient conditions from healthy controls, whereas the dynamic PET data analysis provide pharmacokinetic and pharmacodynamic profiles of radiotracers like the volume of distribution, binding potential, metabolism, etc. using different methods like tissue compartmental analysis, spectral analysis, and mathematical graphical analysis. The last stage is the PET images of living subjects' brains or whole bodies with color-coding to observe the radiotracer uptake in the specific regions of the body, which provides information about the physiological changes in any disease condition compared to normal healthy controls (12).
The United States Food and Drugs Administration (USFDA) does not have regulations to monitor the production of PET drugs for research purposes, since they do not come under the drugs and cosmetic act, which outlines that they should exhibit pharmacological actions at therapeutic doses.  (13,14).
The traditional preclinical and clinical studies without any advancements have shown very low success rate in the drug development process since many factors contribute to the difficulty of developing novel therapeutic drugs, which include the low optimal level of drug-receptor binding at the target, non-specific binding, lack of expected therapeutic efficacy or biological effects of a drug candidate, severe side effects, and drug metabolism. Therefore, there have been several advancements in the drug discovery and development process to overcome the limitations as well as increase the success rate, in which the best and most popular technique commonly used in drug discovery is the PET tool, it can facilitate the development of therapeutic drugs at a faster rate based on various approaches like target FIGURE 1 | The schematic representation of receptor imaging with positron emission tomography (PET) and the flow from radionuclide generation to PET Imaging output, where the initial step is the generation of positron-emitting radionuclides, the second step is the radiochemical synthesis, the third step is the preparation of suitable radiopharmaceutical formulation, the fourth step is the PET imaging data acquisition, the fifth step is the PET data pre and post-processing using software technologies, the last stage is the output of PET images reveal physiological changes based on radiotracer uptake corresponding to the intensity and dynamics in the color of images. engagement, proof of mechanism, proof of principle, and proof of concept (Figure 2) (15,16). The PET has many roles in the early and late-stage drug discovery and development process like in different stages of preclinical and clinical trials by revealing information about pharmacokinetic profiles as well as pharmacodynamic actions of a new drug candidate in a living system (17).
The target engagement approach helps in finding key biological targets or biomarkers in various disease conditions, which can be identified by cross-sectional studies or longitudinal studies. The identified biological targets or biomarkers can also be used to monitor the therapeutic responses of a drug candidate. The proof of mechanism approach helps in finding specific binding of a drug candidate with biological targets or biomarkers through its volume of distribution and kinetic rates, metabolism in the brain, and penetration through the blood-brain barrier. The proof-of-concept approach helps in understanding the biological effects of a drug candidate and its interactions with receptors in various diseases conditions (18,19). PET plays a crucial role in the early and late-stage preclinical studies to increase the drug development process by providing information on the key biological targets or biomarkers, selection of a suitable radiotracer, biodistribution, target quantification, pharmacokinetics, efficacy, safety, and toxicology profiles (20,21), The PET imaging applications at various stages of the drug discovery and development process are shown in Figure 3.

QUANTITATIVE RECEPTOR BINDING AND BIODISTRIBUTION STUDIES USING PET
Quantitative receptor binding and biodistribution are the major applications of PET molecular imaging in the drug discovery and development process. PET provides a change in the concentration of receptors in various disease conditions based on the quantitative measurement of receptor density, and biodistribution using different compartment models. The in vitro radioligand binding assays are the basis to translate its techniques into PET molecular imaging like receptors imaging in vivo, which are mainly based on the equilibrium between the bound ligand-receptor complex FIGURE 2 | The roles of PET imaging studies in drug discovery and development, the PET advances the development of therapeutic drugs with various approaches like target engagement, proof of mechanism, proof of principle, and proof of concept at different stages of early and late-stage drug discovery research, and clinical development process.
(LR) and a free receptor (R) as well as a free ligand (F) with reaction rate constants k on and k off follow the law of mass action (22). (1) At equilibrium: association = dissociation the equilibrium dissociation constant (K D ) is derived from rate constants kon, and koff.
Based on in vitro radioligand binding as well as dissociation constant (K D ) for the first time, the term binding potential was introduced into in vivo PET molecular imaging. It is based on two parameters such as receptor density, and radioligand affinity toward the target to measure quantitatively the radioligand uptake in the brain. Therefore, it is considered that the binding potential (BP) is the ratio of receptor density (B max ) to radioligand equilibrium dissociation constant (K D ). Since the radioligand affinity is the inverse of the equilibrium dissociation constant (K D ), the BP is also considered as a product of B max and affinity (23). The in vivo binding potential quantifies a receptor population with the specific binding of radioligand to the receptor by omitting the free state of radioligand and non-specific binding, and the tracer quantity of radioligand occupies only <5-10% of the target but quantifies the entire receptor population through specific radioligand binding (24).
Equilibrium Michaelis − Menten : s B = B max * F/K D + F (4) In a PET study, a tracer quantity of radioligand is injected (KD >> F), and the Equation (5) is rearranged as below, Frontiers in Medicine | www.frontiersin.org Therefore, Radiotracer Binding Potential = B/ F = B max / K D . The term B max has been commonly used in radioligand binding assays in vitro due to the homogeneity of all available receptors, where the in vivo conditions show heterogenicity of receptors due to competition between radioligand and endogenous neurotransmitters for receptor binding. Hence, the term B max has been replaced with B avail (total available receptor concentration) for in vivo PET imaging studies. The in vivo binding potential (BP) quantifies the ratio of specific radioligand binding to other reference concentrations like free radioligand, bound with plasma proteins or non-displaceable binding at equilibrium, therefore, it was categorized into three individual binding potentials based on different reference concentrations of radioligand like BP F , BP P , and BP ND. The BP F refers to the ratio of the concentration of specific radioligand binding (Cs) to the concentration of free radioligand (C F ) at equilibrium. The BP P refers to the ratio of the concentration of specific radioligand binding (C S ) to the concentration of both free and plasma protein-bound radioligand (C P ) at equilibrium. The BP ND refers to the ratio of the concentration of specific radioligand binding (Cs) to the concentration of nondisplaceable uptake (C ND ). The below mathematical equations help us to understand different binding potential values with respect to reference concentrations of radioligand (25,26). The BP F and BP P require measurement of the arterial input function to quantify receptor occupancy, whereas the BP ND does not need blood sampling during PET analysis (27). All the parameters relating to BP F , BP P , and BP ND are summarized in Table 1.

VOLUMES OF DISTRIBUTION
Determining the volume of distribution (V T in mL.cm −3 ) is an essential step in the drug discovery and development process to establish the different binding potential values corresponding to receptor densities based on the Area Under Curves (AUC) as well as kinetic rate constants. V T is the ratio at the equilibrium of the concentration of radioligand in the brain and the plasma. In general, V T refers to an uptake of radioligand in the brain relative to how much radioligand is delivered via arterial plasma. V T is calculated by the area brain curve divided by the area plasma curve at the equilibrium. V T is B avail , Density of receptors available to bind radioligand but some receptors may be unavailable; bound by the transmitter or sequestered; f P , free fraction in plasma; f ND , free fraction in the nondisplaceable compartment.
always independent of injected activity (28,29). The total concentration of radioligand in the tissue (C T ) is based on the concentration of specifically bound with receptors (S), bound with plasma proteins or non-displaceable binding (ND), and free state in the tissue water (F). The volume of distribution measures the ratio of the concentration of radioligand in each component like specific bound, non-specific bound, or free radioligand in tissue water at equilibrium to the concentration of parent radioligand (C P ) in plasma, which is separated from radiometabolites by HPLC analysis. The V S refers to the ratio of the concentration of specific radioligand binding (Cs) to the concentration of parent radioligand (C P ) in plasma free from radiometabolites at equilibrium. The V ND refers to the ratio of the concentration of non-displaceable uptake (C ND ) to the concentration of parent radioligand (C P ) in plasma free from radiometabolites at equilibrium. The V T refers to the ratio of the concentration of radioligand in tissue (C T ) to the concentration of parent radioligand (C P ) in plasma free from radiometabolites at equilibrium. The below mathematical equations help us to understand different V T values with respect to reference concentrations of radioligand (30,31). All the parameters related to V S , V ND , and V T are summarized in Table 2.
The V S and BP P refer same as the ratio of the concentration of specific radioligand binding (C S ) to the concentration of both free and plasma protein-bound radioligand (C P ) without radiometabolites at equilibrium (32). The V T and V ND are calculated from specific region-of-interest using different reference tissue methods as an alternative to the arterial input function. In general, the ratio of V T /V ND is referred to as the distribution volume ratio (DVR), it is commonly used as a function of receptor availability and it requires arterial input function, but many methods have been developed to calculate DVR without the need for blood sampling with different reference tissue methods (33). The individual binding potential values are calculated using the volume of distribution, which in turn depends on rate constate values as shown in Table 3.
The rate constant values are based on two-tissue compartment models (Figure 4), here these rate constant values provide the volume of distribution, which in turn provide binding potential values based on receptor density and dissociation constant. In the one-tissue compartment (1TCM) the whole tissue is considered as one compartment including the nondisplaceable (free plus non-specific) along with the specifically bound, whereas in the two-tissue compartment (2TCM), two compartments are present in the tissue which include nondisplaceable (free plus non-specific) and specifically bound (34,35). The 2TCM is a commonly used model to obtain the volume of distribution and binding potential. Figure 5 illustrates a variety of examples on PET imaging of the human brain with abnormal conditions as well as normal healthy brain using different radioligands such [ 11 C]-PBR28 showed high binding in different cortical regions of Alzheimer's disease patient compared with a normal brain, [18F]-FDOPA showed a little uptake in regions of caudate and putamen corresponding to nigrostriatal pathway in Parkinson's disease patient from the healthy brain, (11) C-PK11195 showed uptake in the regions with high actation of microglia in a multiple sclerosis patients compared to a healthy normal brain, and the (11) C-PBR28 showed increased binding in frontal and temporal lobes in frontotemporal dementia compared with healthy controls (36)(37)(38)(39).

RECEPTOR OCCUPANCY STUDIES USING PET
The drug-receptor occupancy is one of the major applications of PET molecular imaging in the drug discovery and development process. Since PET is ready to provide receptor densities with selective radioligand, it can also easily test the efficacy of new drug candidates by measuring the plasma concentration with the receptor occupancy in a time course (40). This data provides a lot of pharmacological information of a new drug in the development process like the penetration capability across the blood-brain barrier, specific binding with the target, and rational dosage selection in different phases of drug development. The drug-receptor occupancy gives us a piece of information about the treatment-induced relative change in the concentration of available receptors, it can be easily calculated with the relative change in the binding potential before and after the drug treatment using PET baseline and blocking studies (41). The drug-receptor occupancy (RO%) is calculated using the below formula, and Figure 6 illustrates PET images of the human brain in Parkinson's Disease (PD) condition with baseline and after treatment with a blocking agent study, these studies were performed to measure the brain Monoamine Oxidase B (MAO-B) receptor occupancy by Rasagiline (selective MAO-B inhibitor) using the selective radiotracer 11 C-L-Deprenyl, and also showed treatment response in PD condition with selective drug Rasagiline (42).

THE CHEMISTRY OF RADIOLABELING WITH FLUORINE-18 AND CARBON-11 FOR PET IMAGING
Radiosynthesis Using Fluorine-18 The fluorine-18 is a commonly used radionuclide because of its maximum half-life of 110 min, high radiochemical yields, easy access to PET clinical centers, and best PET imaging properties (43). The [ 18 F]FDG is regarded as a potential PET radiotracer for imaging and quantifying various physiological changes in cancer, cardiovascular and neurological conditions. The number of research papers on the development of fluorine-18 based PET radiotracers have been increasing exponentially, and many cGMP compliance cyclotron facility centers were set up to produce 18 F-radiopharmaceuticals and distribution to nearby PET clinical centers (44). Many clinical and research centers produce fluorine-18 radionuclide with high molar activity through cyclotron facility using 18 (49). Recent advances have come to produce 18 F-radiopharmaceuticals through transitional metalmediated reactions and prosthetic groups involved reactions for 18F-labeled peptides, and proteins (50).
The electrophilic radiofluorination reactions with [ 18 F]F 2 fluorine are limited since the radiochemical yield is almost 50% due to dilution of 19F fluorine, leakage or contamination is very common as [ 18 F]F 2 fluorine comes as gas from the cyclotron, achieve minimum molar activity (0.05-0.5 GBq/µmol), and [ 18 F]F 2 fluorine is a non-selective electrophilic agent for regiospecific reactions (51). The generation of electrophilic [ 18 F]F 2 fluorine involves two steps, initially, fluorine-18 is generated in a metal target using irradiating 18 O water with high energy protons (10 MeV), later this fluorine-18 undergoes isotopic exchange with continuous irradiation of a mixture of molecular fluorine and noblegas (F: Ne = 2: 98) as a gas target, and finally the gaseous form of [ 18 F]F 2 fluorine releases for electrophilic radiofluorination reactions (52). Initially, the clinically used PET radiotracers like [ 18 F]FDG, 18 F-L-DOPA, 18 F-EF5, and 18 F-labeled 5-fluorouracil were synthesized using electrophilic radiofluorination reactions with 18 F-F 2 isotope gas, from which the synthesis of [ 18 F]FDG was shown in Figure 7A (53). Many synthetic challenges limited the use of electrophilic radiofluorination, and automated synthesis modules (54). The nucleophilic radiofluorination reactions with 18 F fluoride are most common since the radiochemical yield is >90% due to direct insertion of fluorine-18 into precursor, leakage or contamination is very limited as 18 F fluoride comes as a liquid from the cyclotron, achieve maximum molar activity (500-5,500 GBq/µmol), and 18 F fluoride is a selective nucleophilic agent for regiospecific reactions (55). The major limitation of 18 F fluoride is ready to form hydrogen bonds with the 18 O water and reduces the nucleophilic substitution reaction rate due to its nucleophilicity, but it can be achieved by FIGURE 4 | Biopharmaceutical compartment analysis models. (A) One-tissue compartment, in which the whole tissue is considered as one compartment including the non-displaceable (free plus non-specific) along with the specifically bound. (B) Two-tissue compartment. In which two compartments are present in the tissue which include non-displaceable (free plus non-specific) and specific bound. The rate constants are generally obtained via dissociation constant from the tissue to the plasma.  is widely used in radiochemistry to dry the fluoride-cryptand complex with acetonitrile or using a microwave heating method (58). The commonly used polar aprotic solvents include acetonitrile, N, N-dimethylformamide, dimethyl sulfoxide, and dimethylacetamide. Based on the nature of molecules the nucleophilic radiofluorination is possible either for aliphatic or aromatic substitution reactions. The clinically used PET radiotracers like [ 18 F]FDG, 18 F-FLT, 18 F-FES, and 18 F-choline were synthesized using nucleophilic radiofluorination reactions with 18 F-fluoride, from which the synthesis of [ 18 F]FDG was shown in Figure 7B (59). Many automated synthesis modules and methodologies are developed based on nucleophilic radiofluorination (60,61). Recently one study reported first transition metal assisted 18 F-labeling of heteroaromatic phenols through deoxyfluorination process via Ruthenium π-complexes. It is well-established that this transition metal π-complexes lower the electron density of the heteroaromatic phenols and activates these for nucleophilic aromatic substitution. They have translated this method into a fully automated synthesis of tyrosine and β-CFT derivatives in GE tracer lab FX 2 N modules and achieved high radiochemical yields (62). The same research group extended their research into 18F radiolabeling of biomolecules using the same approach of a chemoselective radio-deoxyfluorination of an aromatic amino acid residue, mainly a tyrosine residue for conjugation of biomolecules (63). Various aromatic and heteroaromatic iodonium salt precursors were developed for 18 F-labeling via Cu-mediated radiofluorination, and the utility of iodonium salts has been increasing vastly for the development of various 18 F-radiopharmaceuticals due to its high radiochemical yield, regioselectivity, high tolerance, and straightforward synthesis (64). Dr. Pike's group has done extensive research on different iodonium salt precursors and developed several useful PET radiotracers for oncology, and neurology using iodonium salts (65,66). One research group reported Cu-mediated 18Fradiofluorination of diaryliodonium salts using [ 18 F]KF and translated this method into clinically used radiotracers like 4-[ 18 F]fluorophenylalanine and 6-[ 18 F]F-DOPA, and achieved high radiochemical yields as shown in Figure 7C (67). However, nucleophilic radiofluorination has some limitations, mainly for 18 F-radiolabeling of protein conjugates since peptides, protein conjugates are unstable at higher temperature is being used in nucleophilic radiofluorination (68). Therefore, prosthetic groups have been used to introduce 18 F-radionuclide into peptides, protein conjugates ( Figure 7D). The 18 F-radiotracers are developed by conjugating 18 F-labeled precursors containing prosthetic groups with peptides, and proteins under normal conditions, but the limitations include lower yields and multistep process (69).

Radiosynthesis Using Carbon-11
Like fluorine-18, carbon-11 is also a widely used radionuclide for PET radiotracer development, and it is a short-lived positron emitter with a half-life of 20.4 min and β + emission of 99.8%. The carbon-11 radionuclide can be generated in two different sources from a cyclotron including [ 11 C]CO 2 or [ 11 C]CH 4 using different targets via the 14 N (p, α) 11 C nuclear reaction (70). These generated different forms of carbon-11 radionuclide are inserted into various organic biomolecules during radiochemical synthesis to develop target-based PET radiotracers. The [ 11 C]CO 2 is generated from a target having nitrogen 100 % and oxygen ∼0.1-1% with higher activities ∼118 GBq and 11-17 MeV energy, the [ 11 C]CH 4 is generated from a target having nitrogen 100% and hydrogen ∼ 10% with higher activities ∼67 GBq and 18 MeV energy (71). Several convenient methods were adopted to generate [ 11 C]CO 2 or [ 11 C]CH 4 in high molar activity using different targets (72). In general, the No-Carrier-Added (NCA) method was utilized to generate carbon-11 since the specific (GBq/mg)/molar activity (GBq/µmol) is achieved, whereas the fluorine-18 generation mainly the 18 F-Fluorine ( 18 F-F) is generated via Carrier-Added (CA) method, hence the specific (GBq/mg)/molar activity (GBq/µmol) is very low due to dilution with non-radioactive fluorine-19 atoms. Therefore, the molar activities (GBq/µmol) of carbon-11 are always higher than 18 F-Fluorine ( 18 F-F) radionuclide (73). The presence of non-radioactive atoms causes saturation of biological targets thereby lowering target-based PET signal and resolution of PET images, hence the NCA method is essential to increase the ratio of radioactive atoms to non-radioactive atoms (74). The molar activities depend on characteristic decay, therefore, is it essential to focus on the radiotracer injection time during a PET study, duration of radiosynthesis, radionuclide generation, and bombardment time (75). In general, a human PET study requires about 500-600 MBq of carbon-11 at the time of injection, whereas for an animal PET studies <200 MBq is sufficient based on body weight (76). The last stage radiochemical synthesis, rapid quality control studies, and fast dispensing for administration to a living subject are essential with carbon-11 since the half-life is about 20 min various carbon-11 radiotracers can be synthesized with different routes by generating primary and secondary precursors. The major advantage of carbon-11 is the direct insertion of carbon-11 synthons into a precursor to synthesize 11C-radiotracers without optimization trials (77).
The primary precursors [ 11 C]CO 2 or [ 11 C]CH 4 are directly synthesized from a cyclotron using specific targets with nitrogen gas, and a trace amount of hydrogen or oxygen via the 14 N(p,α) 11 C nuclear reaction. The radiochemical yields are very good due to the high concentration of nitrogen with low oxygen/hydrogen, whereas other primary precursors such as [ 11 C]CN or [ 11 C]CO can also be produced by changing target composition, but yields are moderate (78). Many studies reported the direct insertion of [ 11 C]CO 2 into the molecules via [ 11 C]carboxylation reactions with different organic compounds like organometallic reagents, amines, etc (79). For example, the direct insertion of [ 11 C]CO 2 can generate various carbon-11 functional groups like 11 C-carboxylic acid, 11 C-urea, 11 Camide, 11 C-carbamate, and 11 C-oxazolidinone, which can further be used for developing target-specific 11C -radiotracers like [ 11 C]AR-AO14418, is a selective GSK-3β inhibitor where it has 11 C-urea functional group, [ 11 C]Bexarotene, is a retinoid X receptor (RXR) agonist where it has an 11 C-carboxylic acid functional group, and ([ 11 C]GR103545, is a kappa opioid agonist where it has 11 C-carbamate functional group (80).
Due to limited access to developing various radioligands with only the primary carbon-11 precursors [ 11 C]CO 2 or [ 11 C]CH 4 , many secondary precursors have been developed and used as building blockers to synthesize target-specific radioligand with high accessibility. Many successful radiotracers are used in clinical and preclinical studies are developed using various secondary precursors like 11 CH 3 I, 11 CO, 11 COCl, 11 COCl 2, 11 HCN, 11 CH 2 O, 11 CS 2, and 11 CH 3 OTf (81). A few reported examples include [ 11 C]Raclopride is a cerebral D2 receptors antagonist for imaging Parkinson's diseases, which is synthesized by 11 CH 3 I precursor, 11 C(carbonyl)-estramustine phosphate is an estrogen receptors agonist for imaging solid tumors, which is synthesized by 11 CO precursor, [ 11 C]citalopram is a selective serotonin reuptake inhibitor for imaging depressive conditions, which is synthesized by 11 CH 3 I precursor (82). Various reaction pathways have been shown in Figure 8 to generate secondary precursors using the primary precursors [ 11 C]CO 2 or [ 11 C]CH 4 (83,84). Three important steps must consider for an effective strategy to synthesize carbon-11 radioligands since it has a shorter half-life, which includes designing a protocol in a way to keep radiolabeling at the final step, the reaction time should be minimal, and reduce isotopic dilution to get a high molar activity. Currently, many organic methodologies have been adapted for developing selective radioligands through C-C bond formation despite using simple methylation with [ 11 C]CH 3 I, which include alkylation of nucleophilic groups with carbon-11 inserted alkyl halides, cyanides, and nitroalkanes which can act as electrophilic groups (85), [ 11 C]carboxylation reactions with different organometallic copper and other metal-catalyzed reagents, transition metals such as palladium, rhodium, ruthenium, and selenium assisted reactions with different secondary precursors like [ 11 C]CH 3 I, [ 11 C]CO, and [ 11 C]CN (86)(87)(88). A few of reported carbon-11 radioligands are synthesized via C-C bond formation include [ 11 C]A85380 is a nicotinic Ach receptor agonist, which is synthesized from Pd2(dba) 3 / 11 CH 3 I, [ 11 C]MPEP, is a glutamate receptor subtype 5 receptors agonist, which is synthesized from Pd(PPh 3 ) 2 Cl 2 / 11 CH 3 I.

Gauthier et al. have reported an 18 F-Labeled Tropomyosin
Receptor Kinase (Trk) Inhibitor 18 F TRACK for PET Imaging. The overexpression of either Tropomyosin Receptor Kinase A, B, or C receptors induces tumor growth, metastasis, and some types of neurological diseases like Alzheimer's and Parkinson's diseases. The 18 F-TRACK was synthesized via Cu mediated radiofluorination reaction in DMF. 18 F-(R)-TRACK was found to cross the blood-brain barrier (BBB) and showed high brain uptake in studies with non-human primates (89). Zhang et al. have reported 18 F labeled sulfonamide analog as human carbonic anhydrase IX inhibitor for PET imaging to detect hypoxia tumors. It was synthesized via a copper-mediated radiofluorination reaction (90). The studies in the human colorectal cancer xenograft model suggested high uptake in tumors with 0.41 ± 0.06% ID/g. The transmembrane-bound human carbonic anhydrase IX and XII are highly expressed in the tumor cells, and these are considered tumor-associated enzymes, mainly in hypoxic tumors. Therefore, overexpression of human carbonic anhydrase IX and XII are associated with tumor growth, metastasis, and angiogenesis (91-93). Song et al. have reported 18 F-IRS as Epidermal Growth Factor (EGF) Receptor selective ligand for imaging mutant EGF receptors in non-smallcell lung carcinoma (NSCLC) patients. An overproduction of EGF receptor ligands in the tumor microenvironment leads to cause severe mutations in the EGFR receptors which in turn accelerate epithelial tumor cell growth, invasion, and metastasis (94,95). This radiotracer showed a good uptake in tumor cells via targeting mutant EGF receptors in vivo. Wilson et al. have reported 18 F-Olaparib for PET Imaging of PARP Expression.
In-vitro studies showed a good tumor uptake of 18 F -olaparib in PARP-1-expressing Capan-1 and MiaPaCa cells. In vivo studies showed specific uptake of 18 F -olaparib in PARP-1expressing tumors in mice xenografts model. Overexpression of PARP reduces cellular levels of NAD+, which leads to cause tumor growth, but the exact mechanism is not known (96). Nepal et al. have reported 18F-fluciclovine (Axumin) is a novel radiolabeled amino acid derivative for PET imaging in prostate cancer. This study suggested higher uptake of 18Ffluciclovine in prostate cancer cells. 18F-fluciclovine (Axumin) is the US Food and Drug Administration (FDA) approved radiotracer for PET imaging in patients with recurrent prostate cancer (97). Strebl et al. have reported a new radiotracer [ 18 F]MGS3 for PET imaging of histone deacetylase in living subjects, this study confirms high brain uptake, specific binding, and regional distribution of [ 18 F]MGS3 (98). Overexpression of histone deacetylases (HDACs) is associated with cancer, neurodegenerative diseases, and psychiatric conditions. Many studies reported various biological targets plays a vital role in cancer progression, one of the major biomarkers is human galectin-1 protein, which is highly expressed in multiple cancers (99,100), and therefore, which could be the potential targets for novel PET radiotracers to image various conditions in cancer, here a few examples of human galectin-1 selective inhibitors where they have been successfully labeled with fluorine-18 and proved to be PET imaging agents for the diagnosis of cancer (101,102). The importance of molecular imaging for diagnosis and therapy has been increased in recent days for various cancer types, hence, the current cancer research is spreading widely onto various receptor-based radiotracers.
Tyrosine kinases are highly upregulated in tumor development, and progression, and are considered cancer hallmarks. Several Tyrosine kinase inhibitors are developed to treat cancer (103). Recently, one study developed carbon-11 labeled 3-Piperidinylethoxyanilinoquinazoline (PAQ), it is a derivative of vandetanib (tyrosine kinase inhibitor), a transgenic mouse model of breast cancer study showed promising results with [ 11 C]PAQ to check the cancer treatment status and considered as a good radiotracer in cancer therapy response treatment (104). A thymidine derivative [ 11 C]4DST (4 ′ -[methyl-11 C]thiothymidine) is approved to use clinically for measurement of cancer cells development, metastasis, and invasion. It is one of the hallmarks of cancer cells. The first radiotracer developed was [ 11 C]thymidine to measure the proliferative rates in different tumors (105). The large neutral amino acid transporters (LAT) like LAT1, LAT2, LAT3, and LAT4 are highly upregulated in tumor cells due to the full utilization of these transporters for tumor growth and development (106). Various carbon-11 labeled amino acid radiotracers are developed. Recently one study reported an 11C-labeled amino acid radiotracer to quantify tumor cell proliferation (107). Several studies confirmed the upregulation of Translocator protein (TSPO) in gliomas, and a recent study by Su et al. reported the use of [ 11 C]PK11195 radiotracer can be easily differentiated between the low-grade astrocytomas and oligodendrogliomas through a PET dynamic study (108). The  PET molecular imaging studies with endogenous reporter genes have been translated into clinical research, and one recent study with trimethoprim labeled with carbon-11 ([ 11 C]TMP) showed high biodistribution and good sensitivity toward Escherichia coli dihydrofolate reductase expressing cells in a xenograft mouse model and suggested it could provide advancement in present PET reporter gene technologies (109). Natural compounds like coumarin and benzimidazole are the most popular ligands for the development of target-based selective anticancer agents, hence which trigger the development of target-based selective 11 Cradiotracer for molecular imaging in oncology and neurology (110,111) (Figure 9). Here, we have reported a list of recent patents reported for fluorine-18 and carbon-11 radiotracers with their specific biomarkers for various conditions and clinical use in oncology (  Tangles are a key pathological feature associated with cognitive decline in Alzheimer's disease. C]MC1 for imaging low-density cyclooxygenase 2 (COX-2) enzyme in the human brain. It is an essential target for neuroinflammation. The COX-1 expression is high in the brain, whereas the COX-2 expression is low in healthy brain tissue, but it is upregulated during the inflammatory process. This [ 11 C]MC1 is a selective radioligand for COX-2 and found that in rheumatoid arthritis conditions it detected high-density COX-2 in symptomatic joints. This study shows high uptake of [ 11 C]MC1 in the human brain and the specific binding to COX-2 was confirmed by blocking studies with celecoxib (129). Yan et al. have reported an improved PET radioligand [ 11 C]deschloroclozapine ([ 11 C]DCZ) to quantify the muscarinic DREADDs transfected in the brain of monkeys. In the monkey DREADDs model, the signal-to-background ratio of [ 11 C]DCZ was almost 2fold greater than that of [ 11 C]clozapine ([ 11 C]CLZ) due to a much lower background uptake. Hence, it is suggested that the [ 11 C]deschloroclozapine has high selectivity over DREADD hM4Di receptors than [ 11 C]CLZ (130). The Designer Receptors Exclusively Activated by Designer Drugs (DREADDs) are a novel chemogenetic technology used to activate or inhibit different neural populations in the brain regions leading to control various neurological conditions (131). Kim et al. have reported the first-in-human evaluation of a novel radioligand [ 11 C]PS13 to quantify cyclooxygenase-1 (COX-1) in the brain. It is also an essential target for neuroinflammation. The pharmacokinetic profiles were obtained through two-tissue compartment models, and Logan graphical analysis, the total distribution volume (V T ) was also determined. This study found that [ 11 C] PS13 shows high uptake in the occipital cortex and hippocampal regions of the human brain (132).  Wakabayashi et al. have reported a selective phosphodiesterase subtype 4D (PDE4D) radioligand, it is a potential target to improve cognition or antidepression category. This study has reported four lead compounds having an alkoxypyridinyl base with high in vivo PDE4D selectivity, potency, lipophilicity, and good brain uptake. The blocking studies with rolipram or BPN14770 (selective PDE4D inhibitor) confirmed specific binding of [ 11 C]T1650 in the cognitive functions associated regions like the prefrontal cortex, hippocampus, and temporal cortex, and the major limitation of this study is the high brain enzyme density (V T ) with scan duration because of accumulation of radiometabolites in the brain (133). The vesicular monoamine transporter 2 (VMAT 2 ) protein is considered a biomarker to evaluate the condition of Parkinson's disease. The synthesized dopamine at the striatal cells is transported and stored in the synaptic storage vesicles mediated by VMAT 2 receptors. Hence, the recent study developed a novel radiotracer 10-(+)-[ 11 C]DTBZ as in vivo PET imaging for VMAT 2 using micro-PET study to understand Parkinson's disease condition (134) (Figure 10). Here, we have reported a list of recent patents reported for fluorine-18 and carbon-11 radiotracers with their specific biomarkers for various conditions and clinical use in neurology ( Table 5).

FUTURE PERSPECTIVES
The PET is the best technology with great potential and limited challenges in the field of molecular imaging. It plays a tremendous role in the medical diagnosis of various conditions in oncology, cardiology, and neurology. The specialty of PET is to measure lower concentrations like 10 −6 -10 −9 gm of radiotracers in living subjects. The drawback of PET is to provide limited anatomic data which is overcome by hybrid technology such as PET-CT, and PET-MR for attenuation correction and anatomical orientation. The PET is more sensitive (10 −12 M) than other medical imaging modalities like MRI (10 −4 M), but it has a low special resolution (2-6 mm). Therefore, PET remains a remarkable molecular imaging modality in biomedical as well as clinical practice. Micro dosing is one of the best approaches in pharmaceutical development since the low concentration of radiopharmaceutical exhibits lower toxicologic risks. The totalbody PET/CT system, like a whole-body dynamic scan with high sensitivity, makes it useful for radiolabeled drug pharmacokinetic study. Therefore, the distribution of radiopharmaceuticals can be visualized, and the time of activity can be calculated in different tissue regions of interest simultaneously with dynamic images. Hence, pharmacokinetics, as well as pharmacodynamics, can be estimated more precisely. The total-body PET/CT has many advantages in the evaluation of various systemic diseases, which include higher sensitivity allows imaging at high signal to noise ratio (SNR) at later time points after radioligand injection, the total-body dynamic scan provides more kinetic information than normal conventional PET/CT, and the wholebody images can provide more comprehensive evidence of many of systemic diseases in oncology, cardiology, and neurology (145,146). The fluorine-18 and carbon-11 are promising radiotracers with many advantages but limited challenges. The fluorine-18 can withhold long-run radiochemical synthesis due to a longer half-life of 110 min, and multiple studies can be done by PET in a single day. Despite the shorter half-life of carbon-11 (t 1/2 = 20.4 min), it is widely used in various synthetic methodologies with different heterocycles because of the direct insertion of the carbon-11 radionuclide. Several automated methods are adapted for easy, rapid, and efficient synthesis of various PET radiopharmaceuticals. Therefore, advances in multimodality devices could help in getting accurate and reliable data on drug candidates at different stages of drug development.

CONCLUSIONS
In this article, we have focused on the importance of PET molecular imaging in the drug discovery and development process. The PET is mainly used to address essential factors to advance the drug development process, which include understanding the pathophysiology, measuring roles and applications of PET molecular imaging in drug discovery and development. In addition, it is also focused on a variety of carbon-11 and fluorine-18 radioligands, their modified and advanced procedures of production, and their PET imaging applications in early state diagnosis, prognosis, and treatment of disease conditions in oncology, and neurology. Various radiofluorination methods, easy synthesis, appropriate precursor identification, and automated synthesis measures have been explored to reach the goals of developing selective radiotracers. Clinical use of PET imaging agents with carbon-11 and fluorine-18 have vastly increased due to a wide range of advantages. It would be helpful to understand the significance of PET molecular imaging in drug development, and the importance of positrons emitting fluorine-18 and carbon-11 radiotracers.

AUTHOR CONTRIBUTIONS
SN and PS: data collection, writing, analyzing, and setting up paper. TS: data collection, analysis, and review. CD: editorial role to the paper. All authors contributed to the article and approved the submitted version.