Chemical Space of Fluorinated Nucleosides/Nucleotides in Biomedical Research and Anticancer Drug Discovery

Abstract

Fluorinated nucleos(t)ide drugs have proven to be successful chemotherapeutic agents in treating various cancers. The Food and Drug Administration (FDA) has approved several drugs that fit within the fluorinated nucleoside pharmacophore, and many more are either in preclinical development or clinical trials. The addition of fluorine atoms to nucleos(t)ides improves the metabolic stability of the glycosidic bond and, in certain instances, facilitates additional interactions of nucleons(t)ides with receptors. The insertion of fluorine either on sugar or the base of nucleos(t)ides proved to enhance the lipophilicity, pharmacokinetic, and pharmacodynamic properties. Overall, the fluorine atom feeds diverse advantages to the biological profile of nucleos(t)ide analogs by improving their drug-like properties and therapeutic potential. This review article covers the often-used fluorinating reagents in nucleoside chemistry, the clinical significance of [¹⁸F]-labeled nucleosides, the synthesis and anticancer activity of FDA-approved fluoro-nucleos(t)ide drugs, as well as clinical candidates, which are at various stages of clinical development as anticancer agents.

1. Introduction

Nucleos(t)ides are the fundamental building blocks of life. They encode the genetic information of all living organisms and play a significant role in biological processes. Thus, these molecules have been extensively studied in the pursuit of modern medicine and have shown considerable potential as antiviral and anticancer agents. Nucleos(t)ides are crucial parts of the cellular process and offer potential drug scaffolds for treating a diverse range of diseases. Nucleos(t)ide drugs can treat cancer by inhibiting the DNA/RNA synthesis of highly dividing cancerous cells [1]. In the fight against cancer, 6-mercaptopurine was among the first nucleoside-base-driven drugs developed for cancer treatment [2]. Later, a structural analog of guanine, 6-thioguanine (6-TG), was also discovered to treat various cancers [3]. To date, FDA-approved nucleos(t)ide-based therapies are very effective at slowing cancer progression and extending patients’ lifespans.

The development of 5-fluorouracil (5-FU) marked a significant breakthrough in cancer therapy; it has been established as one of the most widely used drugs for cancer treatment [4]. 5-FU also works by inhibiting the DNA elongation of cancer cells [5]. The discovery of 5-FU gained attention for the importance of the 5-fluoro atom in uracil moiety and its noted benefits in the invention of nucleos(t)ide drugs. Various strategies for drug design and the development of fluorinated nucleos(t)ides proved the importance of the fluorine atom in this class of molecules [6]. The incorporation of fluorine into nucleoside molecules alters their electronic properties, conformation, and stability, which drastically impacts the biological profile of the parental molecule [7]. Over the past five decades, numerous fluorinated nucleos(t)ide drugs have been developed and received FDA approval for treating cancer and emerging viral infections [8]. Among these, gemcitabine, clofarabine, and tegafur have become widely prescribed medications for cancer treatment [9,10]. It revealed that 20–25% of drug candidates contain a fluorine atom within their therapeutic profile. The presence of fluorine atom(s) in drug molecules often facilitates improvements in lipophilicity, pharmacokinetic, and pharmacodynamic (PK/PD) profiles [11]. Additionally, fluorine is an excellent hydrogen bond acceptor and univalent isostere of the hydroxy group. Replacing the hydroxy with the fluoro group reduces the molecule’s polarity and increases the hydrophobicity, bioavailability, and cellular uptake [12]. These advantages make fluorine an excellent addition to the altered nucleos(t)ides and pave the way for many approved and future drug molecules [13]. Thus, several fluorinating agents have been developed to construct various analogs of fluorinated nucleos(t)ides of biological interest, and much effort is being made to invent more fluorinating agents [14]. The present article covers the practice of fluorinating agents in the synthesis of fluorinated nucleos(t)ide, the utility of the [¹⁸F] radio labeled in nucleos(t)ides in the medical field, and the synthetic methodology and anticancer activity of FDA-approved fluoro-containing nucleos(t)id drugs as well as ongoing drug candidates in clinical trials.

2. Fluorinating Reagents

The selective synthesis of organofluoride molecules is important and often challenging. In traditional fluorinating methods, molecular fluorine and hydrogen fluoride (HF) are used as fluoride sources. However, these reagents are extremely reactive, toxic, and corrosive, and due to radical mechanistic pathways, they present a non-selective method of fluorination [15]. Additionally, their handling requires a highly skilled chemist and special equipment. Therefore, alternative fluorinating reagents have been developed with stable, safe, and selective ways to install fluorine atoms at the desired position of the molecule. Several highly efficient reagents, such as diethylaminosulfur trifluoride (DAST), Selectfluor, etc., have been developed for fluorinating both C-(sp²) and C-(sp³) bonds. These reagents have mainly been divided into two classes: nucleophilic and electrophilic fluorinating agents. However, based on the nature of the F-species generated and the plausible reaction mechanism, C-F bond formation reactions may be categorized into three types: (i) electrophilic, (ii) nucleophilic, and (iii) thermal or photo-induced radical fluorination.

2.1. Nucleophilic Fluorinating Agents

The nucleophilic character of a fluoride ion depends on the reaction environment; in a protic solvent, it acts as a poor nucleophile; however, in an aprotic solvent, it demonstrates a strong nucleophilic nature. In the case of sugar fluorinated nucleosides, at 2′-OH or 3′-OH, a S_N2, substitution of the hydroxy group with fluorine via a leaving group strategy has been well explored with various fluorinating agents and conveys a standard replacement method of an OH group with a fluoro group [16].

(i)

HF-based nucleophilic reagents

Hydrogen fluoride (HF) is a traditional and the simplest nucleophilic fluorinating reagent, and most nucleophilic or electrophilic reagents are produced from it [17].

However, HF is a hazardous gas at room temperature and is tedious to handle. Therefore, stabilized versions of HF–amine complexes with organic bases, such as the 1:1 pyridine/HF complex (pyridine/HF, 1, Olah reagents) and triethylamine. 3HF complex (NEt₃°3HF, 2), (Figure 1) [18] were invented, which are easy to handle. The disadvantage of HF/pyridine (1) is that it can make a strong complex with many transition metals and reduces metal catalytic activity. Okoromoba et al. reported two ideal HF-based alternate reagents, 1,3-dimethylhexahydropyrimidin-2-one hydrofluoride (DMPU/HF) complex (3) and KHSO₄°(HF)x (4), which are less basic, express weak coordination with metals, and do not interfere in transition metal-catalyzed reactions [18,19]. These reagents are particularly useful in nucleoside chemistry for converting primary and secondary alcohols to their corresponding fluoro moiety [20].

In a representative example (Scheme 1), Pulido et al. reported that the reaction of compound 5 with HF/pyridine (Olah’s reagent) provided a regioisomeric mixture of 10-fluoro (6), 9-fluoro (7), and 8-fluoro (8) derivatives with an isomeric ratio in a 95% yield [21].

Another example published by Izawa et al. is the industrial-scale synthesis of Lodenosine (FddA), an acid-stable purine nucleoside active against HIV [22]. In the reported method, the noncorrosive nature of triethylamine NEt₃^.HF is more beneficial and has better yields in the conversion of intermediate 9 to 10 compared to DAST reagent (Scheme 2).

(ii)

Metal Fluorides

Many popular nucleophilic metal fluoride salts [23], like M¹F (M¹ = K, Cs, Ag), are widely used for fluorination [24]. Of these, CsF is the most reactive and well utilized in nucleoside chemistry. However, metal fluorides’ poor solubility in organic solvents limits their application as fluorinating reagents. Consequently, using a phase-transfer catalyst increases both the solubility and reactivity of metal fluorides. Another drawback of this category of reagents is that the fluoride anion acquires high basicity, which propagates side reactions during fluorination and results in low yields.

Verdine et al. reported the efficient synthesis of 4′-F nucleoside analogs using radical bromination followed by silver fluoride (AgF) fluorination (Scheme 3). This unique procedure eliminates the previously reported multi-step synthesis of fluoro insertion at the 4′ position [23].

(iii)

Quaternary Ammonium Fluoride

Tetra alkyl ammonium fluorides such as tetra-n-butylammonium fluoride (TBAF) are often employed as nucleophilic sources of fluoride with enhanced solubility in organic solvents. TBAF is usually used to convert alcohol to fluoro by activating OH groups and making them a good leaving group, such as tosylate, mesylate, or triflate [25]. Due to the basicity of the fluoride anion, fluorination, usually in organic substrates bearing β-hydrogen(s) in anhydrous condition, may lead to Hofmann elimination. To overcome this Hofmann elimination, the use of tetramethyl ammonium fluoride (TMAF) is in practice for safe fluorination. TMAF is commercially available and thermally stable, and via this reagent, fluorination can be performed under milder reaction conditions [26].

Recently, Singh et al. described the stereoselective synthesis of unique 4′-α-fluoro-methyl carbocyclic nucleosides. In this communication, selective fluorination at 4-OH was carried out using a 1 M solution of TBAF in THF, which was advantageous over the commonly used DAST fluorination (Scheme 4) [25].

• (iv) Deoxy fluorinating reagents

Diethylaminosulfur trifluoride (DAST, 19) is a versatile and admired deoxyfluorination reagent for synthesizing fluorinated nucleosides. The deoxyfluorination reaction of alcohols via DAST occurs in situ by forming an activated leaving group followed by the nucleophilic fluoride anion attack [27]. In the case of secondary alcohol, the replacement of alcohol to fluoro occurs via an S_N2 displacement with an inversion of configuration. DAST has also been extremely popular in converting ketones and aldehydes to geminal difluorides [28]. Singh et al. reported the conversion of the 2-β-fluorination of 2′-α-OH using DAST conditions (Scheme 5) [29].

Several other dialkylamino sulfur trifluoride derivatives of DAST, such as bis(2-methoxy)aminosulfur trifluoride (DeoxoFluor 20, Figure 2) and morpholine sulfur trifluoride (Morpho-DAST or MOST, 21), are also frequently utilized for specific fluorination in nucleoside synthesis [30]. Yarovenko’s reagent (22) and Ishikawa’s reagent (23) have been introduced to overcome the limitations and restrictions of DAST reagents [31]. Aminodifluorosulfinium salts, such as XtalFluor-E (24) and XtalFluor-M (25), are also used to fluorinate nucleoside moieties [32]. Furthermore, DAST reagents have limited functional group tolerance and cause a substantial formation of eliminated side products. Nielsen et al. reported 2-pyridinesulfonylfluoride (PyFluor, 26), which readily fluorinates primary and secondary alcohols in the presence of tertiary alcohols. With this reagent, substrates containing carbonyl/aldehydes do not undergo gem defluorination [33]. Another deoxyfluorination reagent, PhenoFluor (27), fluorinates a wide range of alcohols without a substantial formation of elimination side products, resulting in better yields of the desired product [34].

2.2. Electrophilic Fluorinating Agents

The fluorination of electron-rich double bonds and aromatic rings can be performed via electrophilic fluorinating agents. In these reagents, fluorine binds to an electronegative atom, such as nitrogen, and is activated either by strong electron-withdrawing groups such as sulfonyl or by positive charges in the same molecule. In this class, Selectfluor (28, Figure 3), N-fluoropyridinium salts (29), and N-fluorobenzenesulfonimide (NFSI, 30) are commercially available fluorinating reagents, which are widely used in the fluorination of nucleosides [35,36]. These N-F fluorinating agents are non-hygroscopic and easier to handle than nucleophilic fluorinating agents such as potassium fluoride (KF) and hydrogen fluoride (HF) [37].

H. Kumamoto et al. [38] reported the synthesis of fluoroneplanocin analogs, a class of S-adenosylhomocysteine hydrolase (SAH) inhibitors, via the construction of key intermediate 32 with the installation of fluorine on intermediate 31 (Scheme 6). The electrophilic addition of fluoro to carbocyclic ketone 31 was performed with Selectfluor to produce a crucial intermediate 32, which progressed for the synthesis of targeted fluoroneplanocin analogs.

In nucleoside chemistry, Selectfluor is used to fluorinate the heterocyclic ring (Scheme 7) or electron-rich double bond positions [39]. Selectfluor is equivalent to F_2, which is solid and much safer to handle. It exhibits selective fluorination in most organic solvents. Selectfluor may also selectively fluorinate certain electron-rich sugar moieties containing double bonds via electrophilic addition [40,41].

Notably, N-fluoropyridinium salts (29) are particularly useful for the fluorination of aromatic rings, carbanions, and enol ethers [42].

Recently, many nucleosides containing trifluoromethyl groups have demonstrated potential biological activities. The -CF₃ group is an excellent isostere of the methyl group to overcome possible metabolic oxidation of the latter [43]. Togni-I (37, Figure 4), Umemoto reagent-II (38), and dimethyl 2-(phenyl(trifluoromethyl)-l4-sulfanylidene)malonate (ETF-1, 39) are well used for the insertion of trifluoromethyl either in the nucleobase or sugar moiety of nucleosides. Through the application of these reagents, direct trifluoromethylation of nucleos(t)ides may be used as a synthetic method for the construction of trifluoromethylated nucleosides [44,45].

Baran et al. reported zinc bis(alkanesulphonate) salt reagent as a transfer alkyl radical that installs various alkyl fluorides (-CF₃, -CF₂H, -CH₂CF₃, and -CH₂F) under C-H functionalization on a highly electron-rich heterocycle analog. This methodology is highly efficient and eliminates earlier reported multi-step approaches of trifluromethylated nucleosides [45,46,47] to install alkyl fluoro functionality under operationally simple and mild conditions [48,49,50]. Recently, by utilizing zinc sulphinate salt (CF₃SO₂)₂Zn reagent, the installation of a trifluoromethyl group at the C8 position of purine nucleos(t)ides has been carried out (Scheme 8) [51].

3. [¹⁸F]-Radiolabeled Nucleosides in Biomedical Research and Their Applications

¹⁸F-labeled molecules are used as imaging probes in positron emission tomography (PET)/computed tomography (CT) for oncology [52], neurology [53], and cardiology diagnosis applications [54]. PET is a non-invasive method that is extremely useful in understanding disease pathology and assists in drug discovery [55]. It enables early detection of various diseases, increasing the likelihood of successful treatment by facilitating timely intervention [56]. ¹⁸F radionuclide reveals exceptional chemical and nuclear properties and has become a highly desirable element in the PET field. ¹⁸F has relatively low positron energy (0.635 MeV) due to its decay to ¹⁸O by positron emission (β⁺97%). It also has a favorable half-life (t_½ = 109.7 min) over other radionuclides ((¹¹C, t _½ = 20.3 min); (¹³N, t _½ = 10 min); (⁶⁸Ga, t _½ = 68 min)) that permits the synthesis, quality control, and transportation of ¹⁸F radiolabeled probes before administering it to the patient [57,58,59]. As of November 2024, out of 19 FDA-approved radionuclides, 12 are ¹⁸F-containing PET tracers [60].

In PET techniques, 2-[¹⁸F]-fluoro-2-deoxy-D-glucose ([¹⁸F]-FDG, 42, Figure 5) is widely used in oncology for the detection of various cancers [61,62] and for imaging disorders of cardiology [63,64] and neurology [65,66]. Though [¹⁸F]-FDG is the breakthrough in transformative success in PET applications, over the years, clinical data have proven that it expresses non-specific uptake [67] and demonstrates metabolic limitations [68]. Considering the drawbacks of [¹⁸F]-FDG, several other nucleoside-based radiotracers such as 3′-deoxy-3′-[¹⁸F]fluorothymidine ([¹⁸F]FLT, 43) and 1-(2′-deoxy-2′-¹⁸F-fluoroarabinofuranosyl) cytosine ([¹⁸F]-FAC, 44) have also been developed (Figure 5) for imaging tumor cell proliferation [69,70].

Several research groups extensively studied radioactive ¹⁸F installation on nucleobases such as pyrimidine (thymine, uracil, and cytosine), purine (adenine and guanine), and C-2′ and C-3′ of a sugar moiety. Because they are DNA-specific, thymidine-based nucleosides are the most prominent, whereas uracil counterparts are used in the replacement of RNA. To date, the thymidine-based tracers [¹⁸F]-FLT (43) and 2′-deoxy-2′-[¹⁸F]fluoro-5-methyl-1-β-D-arabinofuranosyluracil ([¹⁸F]-D-FMAU (46) are the most studied [70]. Pharmacokinetic studies revealed that [¹⁸F]-D-FMAU is more beneficial due to its ability to incorporate into DNA, whereas [¹⁸F]-FLT (43) acts as the terminator of the growing DNA chain [58,71,72]. Currently, [¹⁸F]-D-FMAU (46) is undergoing preclinical and clinical studies to investigate the diagnosis and characterization of prostate, breast, and non-small cell lung carcinomas [73].

Initially, Alauddin et al. reported the synthesis of [¹⁸F]-D-FMAU under milder conditions for a PET imaging agent for cellular proliferation. In the reported synthesis, triflate intermediate 47 was treated with tetrabutylammonium [¹⁸F]fluoride (n-Bu₄N¹⁸F) to produce protected radiofluorinated sugar moiety 48. For the conversion of 5–6 mg (~10 µmol) of triflate 47 to 48, the in situ generation of n-Bu₄N¹⁸F was carried out from n-Bu₄NHCO₃ (~7 µmol) with aqueous H¹⁸F. The crude reaction mixture of 48 was passed through a Sep-Pak silica cartridge to remove unreacted radioactive fluoride, followed by preparative HPLC to obtain >90% radiochemically pure (by HPLC) intermediate 48. Furthermore, 48 was treated with 30% HBr/AcOH to give crude 48A. Intermediate 48A was coupled with silylated 2,4-bis-O-(trimethylsilyl)thymine in 1,2-dichloroethane (DCE) under heating conditions to produce coupled product 49 as an anomeric mixture. Finally, the debenzoylation of 49 was carried out by using sodium methoxide (NaOMe) in MeOH to furnish targeted compound 46. At the end of synthesis (EOS), [¹⁸F]-D-FMAU was constructed with >99% average specific activity 2300 mCi/µmol. In this four-step synthesis process (Scheme 9), the 20–30% decay-corrected (d.c.) radiochemical yield (RCY) was reported as 18 mCi of labeled product, which was obtained starting from 246 mCi of [¹⁸F]-fluoride with a synthesis time of 3.5–4.0 h from the end of bombardment (EOB) [74].

Concurrently, Mangner et al. reported more highly efficient and reliable synthesis of [¹⁸F]-D-FMAU [72]. In the reported synthesis, cyclotron-produced [¹⁸F]fluoride-containing [¹⁸O]water was directly trapped on an anion exchange (bicarbonate) column. The [¹⁸F]fluoride was released using a solution of kryptofix 2.2.2 (KRP), K₂CO₃, water, and anhydrous acetonitrile into v-vail. The solvent was azeotropically dried under the flow of argon. A solution of intermediate 47 in DMF was added to the resultant [¹⁸F]fluoride/KRP/K₂CO₃ and heated to 150 °C for 5 min. After that, the solvent was removed, and the residue was purified by the silica gel Sep-Pak light to produce radio-fluorinated sugar 48. Intermediate 48 reacted with 30% HBr/AcOH at 125 °C for 12–15 min to give bromide intermediate 48A. Furthermore, 48A was coupled with silylated 2,4-bis-O-(trimethylsilyl)thymine in CHCl₃ at 150 °C for 30 min to give coupled product 49 as an β/α anomeric mixture. Finally, the deprotection of 49 was performed in basic conditions, and targeted [¹⁸F]-D-FMAU was purified by semi-prep HPLC. In this synthesis process, more controlled cyclotron-produced nucleophilic radio fluorination was accomplished, and all stages of reactions were conducted under high temperatures, with a total synthetic time of 3 h in a better yield. The RCY of the overall synthetic scheme was reported as 35–45% of the desired β-isomer with >98% radiochemical purity and specific activity of >3 Ci/µmol. Starting from ~220 mCi of [¹⁸F]fluoride resulted in ~25–30 mCi of [¹⁸F]-D-FMAU, which is usually sufficient to image two patients.

4. Synthesis and Anticancer Activity of Fluorinated Nucleos(t)ides

4.1. 5-Fluorouracil (5-FU, 50)

5-Fluorouracil (5-FU) is a uracil analog with a fluorine atom at the C-5 position. It has been widely used in treating a variety of solid tumors for more than fifty years and has been included in the WHO’s list of essential medicines [75]. 5-FU was approved by the FDA in 1962. Since then, it has been considered a first-line treatment for colorectal cancer (CRC). According to the American Cancer Society (ACS), in 2024, 152,810 new patients were diagnosed with CRC. It is estimated that 53,010 people will die from this disease, including 3750 younger patients [76,77]. 5-FU, as a chemotherapeutic, has an excellent impact, and it is also prescribed as a first-line drug to treat several gastrointestinal cancers [78]. Additionally, either as a single regime or in combination with other drugs, 5-FU is being used to treat several cancerous solid tumors of the breast, head and neck, pancreas, and stomach [79,80,81]. 5-FU is administered intravenously and has good water solubility. 5-FU rapidly enters into the cell using the same facilitated transport mechanism as the uracil base [82]. To exert a cytotoxic effect, 5-FU interacts intracellularly with phosphorylated sugar, which is further enzymatically catalyzed into several active metabolites. One of the 5-FU metabolites, 5-fluorodeoxy uridine monophosphate (FdUMP), forms a stable ternary complex with thymidylate synthase (TS) enzyme and 5,10-methylene tetrahydrofolate (CH₂THF) and inhibits the synthesis of DNA that causes cell death [83]. Another 5-FU metabolite, 5-fluorouridine triphosphate (FUTP), is extensively incorporated into RNA and disrupts standard RNA processing and function, which leads to abnormalities in RNA occupations at several stages, including messenger RNA (mRNA), transfer RNA (tRNA), and small nuclear RNA (sn RNA) [84,85,86]. A significant drawback associated with 5-FU is that it substantially catabolizes in the liver by dihydropyrimidine dehydrogenase (DPD), and only 3% of the original dose of 5-FU mediates the cytotoxic effect in tumor cells [78]. 5-FU also poses several disadvantages, including high cell toxicity, poor bioavailability, a short half-life, and a rapid metabolism, which limits the effectiveness of 5-FU in cancer chemotherapy [87]. Metabolically, 5 FU is unstable, and most of the injected dose converts to α-fluoro-β-alanine by the intracellular pyrimidine degrading enzyme [88]. To lower the 5-FU toxic effect on healthy cells and enhance the chemotherapeutic results in CRC management, several new methods of 5-FU delivery are under investigation. To enhance the effectiveness of 5-FU, the use of novel combination therapies, encapsulated drugs, and prodrug strategies have been actively pursued. Particularly, 5-FU prodrugs such as capecitabine (55) and tegafur (56, Figure 6) are orally available and activated in the human body via different mechanisms [89]. Capecitabine 55 is a cytosine-based carbamate prodrug of 5-FU that activates in the liver and cancer cells via sequential reactions catalyzed by carboxylase ester and cytidine deaminase [90]. Tegafur 56 activation involves either a 5′-hydroxylation catalyzed by the cytochrome P450 enzyme CYP2A6 [91] or an enzymatic cleavage of the N1-C2′ of the molecule [92]. Currently, tegafur is under clinical trials to treat various cancers [93,94].

Figure 6. (a) FDA-approved fluorinated nucleoside/nucleotide anticancer drugs; (b) structure of nucleos(t)ide undergoing clinical trials as of 31 October 2024.

Figure 6. (a) FDA-approved fluorinated nucleoside/nucleotide anticancer drugs; (b) structure of nucleos(t)ide undergoing clinical trials as of 31 October 2024.

4.2. Fludarabine Phosphate (F-ara-AMP, 51)

Fludarabine phosphate (F-ara-AMP, 51) is the monophosphate prodrug of fludarabine (F-ara-A, 59, Figure 7). It is commonly marketed under the brand name Fludara^®. Montgomery and Hewson first reported fludarabine (59); later, it was used as a 5′-monophosphate prodrug (51, F-ara-AMP) [95]. F-ara-AMP is a chemotherapeutic agent used to treat chronic lymphocytic leukemia (CLL). It is administered via an intravenous infusion (IV) and rapidly dephosphorylated at a physiological pH to F-ara-A 59. Furthermore, in the form of F-ara-A, it transports into cells by nucleoside transporters, where it re-phosphorylates intracellularly by cellular kinases to the active triphosphate, 2-fluoro-ara-ATP (60). The active fludarabine triphosphate (F-ara-ATP, 60) inhibits ribonucleotide reductase and DNA polymerase, which ultimately causes the death of cancer cells [96,97].

Synthesis of F-ara-A (59): The synthesis of fludarabine (F-ara-A, 59) was initiated with well-known ribose intermediate 2,3,5-tri-O-benzyl-1-O-p-nitrobenzoyl-D-arabinofuranose 61, which was treated with dry HCl gas to give chloro sugar intermediate 62. Chloro intermediate 62 was coupled with 2,6-di acetamidepurine 67 to afford coupled nucleoside which, upon treatment with methanolic sodium methoxide, produces compound 63. The conversion of 2-amino to 2-fluoro was carried out by treating 63 with fluoroboric acid and sodium nitrite to obtain 64. Finally, the O-benzyl deprotection of 64 was achieved by using boron trichloride to render F-ara-A (59) in a qualitative yield (Scheme 10) [98].

The original synthesis process depicted in Scheme 10 has scalable limitations because this procedure requires boron trichloride (BCl₃) for final debenzylation, which is unsafe for large-scale synthesis. To avoid this issue, Blumbergs et al. [99] developed a new method for the preparation of fludarabine phosphate (51, F-ara-AMP). Condensation of 2,3,5-tri-O-benzyl-α-D-arabinofuranosyl chloride 62 with protected nucleobase 68 affords coupled compound 69, which, upon treatment with NaOMe in methanol, yields 70.

Compound 70 was treated with fluoroboric acid and sodium nitrite to afford 2-fluoro-nucleoside 71. Final debenzylation of compound 71 was carried out with palladium (II) chloride on activated carbon (PdCl₂/C) under hydrogenation conditions to yield F-ara-A (59), which, upon further 5′-phosphorylation with phosphorus oxychloride (POCl₃) and trimethylphosphate, rendered targeted fludarabine monophosphate 51 (Scheme 11).

Shen et al. [100] disclosed a novel synthetic approach for the synthesis of F-ara-A 59, which involves a bulkier ortho-alkyne benzoyl ester protecting group at 2-OH of 72 to give 73 that demonstrates neighboring group participation in Vorbrüggen glycosylation to afford the corresponding β-nucleoside (74) exclusively. This approach utilizes commercially available compound 72 as a starting material. First, the 2-OH of compound 72 was protected with 2-iodobenzoyl, followed by subsequent Sonogashira coupling with hex-1-yne to give 73 in a 74% yield.

Vorbrüggen glycosylation of compound 73 with in situ silylated nucleobase exclusively produced the desired β nucleoside 74 (Scheme 12). Deprotection of the 2′-ortho-alkyne-benzoyl nucleoside 74 was carried out using a gold (I) catalyzed reagent Ph₃PAuOTFA to render compound 75 in high yields [101]. Furthermore, the 2′-OH of 75 was transformed into trifluoromethyl sulfonate compound 76, which, upon treatment with nucleophilic KNO₂ in the presence of a phase-transfer catalyst, 18-crown-6, followed by in situ hydrolysis, gives intermediate 77. The deprotection of benzoyl of 77 provides targeted F-ara-A, 59.

Fludarabine phosphate (51) is an FDA-approved drug for the treatment of chronic lymphocytic leukemia (CLL). Its mechanism of action involves inhibiting DNA synthesis [102]. Fluoro-ara-ATP (61) incorporation into DNA is a process that is essential for the induction of apoptosis in leukemic cells. It also demonstrates lymphocytotoxic activity with preferential activity toward T-lymphocytes. In initial preclinical studies, fludarabine (59) has shown antitumor activity against L1210 murine leukemia. During phase I studies of this drug, myelosuppression was identified as a dose-limiting toxicity in patients with solid tumors, and fatal neurotoxicity was identified as the dose-limiting toxicity in adult patients with acute hematologic malignancies [103].

In in vitro studies, fludarabine has shown anticancer activity against multiple myeloma (MM) in RPMI 8226 cells with an IC₅₀ value of 1.54 μg/mL (

Table 1. Anticancer and antiviral activities of fludarabine (59).

Tumor or Virus	Cell Lines	IC50	CC50/tox	Ref
Multiple Myeloma (MM)	RPMI 8226	1.54 µg/mL	-	[104]
	MM.1S	13.48 µg/mL	-
	MM.1R	33.79 µg/mL	-
ZIKV	Vero Cells	0.13 +/− 0.04 µM	3.10 µM	[105]
	BHK-21 Cells	0.41 +/− 0.04 µM	3.61 µM
SFTSV	Vero Cells	0.83 +/− 0.03 µM	3.10 µM	[105]
	BHK-21 Cells	0.27 +/− 0.001 µM	3.61 µM
EV-A71	Vero Cells	0.04 +/− 0.001 µM	3.10 µM	[105]

ZIKV: zika virus; SFTSV: severe fever with thrombocytopenia syndrome virus; EV-A71: enterovirus; IC₅₀: 50% inhibitory concentration.

). However, it expressed IC₅₀ values of 13.48 µg/mL and 33.79 µg⁄mL in MM.1S (dexamethasone-sensitive) and MM.1R (resistant human MM cell lines) cells [104]. The antiviral profile of fludarabine (59) has also been evaluated. In antiviral assays in Vero cells, fludarabine (59) expressed activity against zika virus (ZIKV), severe fever with thrombocytopenia syndrome virus (SFTSV), and enterovirus EV-A71 with IC₅₀ values of 0.13 μM, 0.83 μM, and 0.04 μM, respectively (Table 1) [105].

In BHK-21 cells infected with ZIKV and SFTSV, fludarabine (59) showed similar activity, with IC₅₀ values of 0.41 µM and 0.27 µM, respectively [105]. Fludarabine showed CC₅₀ values of 3.10 µM and 3.61 µM in Vero and BHK-21 cells.

4.3. Clofarabine, 52

Clofarabine is chemically known as 2-chloro-2′-fluoro-2′-deoxyarabinofuranosyladenine. It is a second-generation hybrid nucleoside analog identical to fludarabine (59) with a fluorine atom at the crucial 2′-position of the ribose sugar and 2-chloro at the adenine base. The substitution of fluorine at the C-2′ of the sugar increases the stability of the glycosidic bond of clofarabine [106]. 2′-Fluorine at clofarabine provides enzymatic high resistance to purine nucleoside phosphorylase and metabolic stability to acid hydrolysis. Clofarabine is an antimetabolite used to treat patients between 1 and 21 years of age with relapsed or refractory acute lymphoblastic leukemia (ALL) [107,108]. It is marketed in the United States (U.S.) and Canada under the trade name Clolar and in Europe and Australia/New Zealand under the name Evoltra.

Montgomery and co-workers first synthesized clofarabine in six steps with an overall 6% yield [109]. Later, ILEX Products Inc. reported [110] the improved synthesis of clofarabine (Scheme 13). The reported procedure utilizes protected sugar 78 as a starting material, treated with HBr in DCM, to produce 2-hydroxy-1,3,5-tri-O-benzoyl sugar 72. Furthermore, 2-hydroxy of sugar 72 was converted to 2-O-imidazolyl sulfonyl intermediate 79. An S_N2 nucleophilic fluorination was executed on intermediate 79 with potassium bifluoride (KHF₂) to afford 2-fluorinated sugar 80. Bromination at the anomeric center of 80 was carried out with HBr/AcOH to produce key intermediate 1-bromo-3-O-benzyloxy-2-deoxy-2-fluoro-5-O-benzoylarabinose 81. The coupling of critical bromo-intermediate 81 with 2-Cl-adenine 82 in the presence of potassium tert. butoxide furnished nucleoside 83, which, upon further debenzoylation with 30% NaOMe solution in MeOH, afforded clofarabine 52 in a 14% overall yield [109,111,112].

Later, a process development and the concise synthesis of clofarabine were investigated with commercially available 2-deoxy-2-β-fluoro-1,3,5-tri-O-benzoyl-1-α-D-arabinofuranose 80. Compound 80 was treated with HBr/AcOH to obtain the glycosylated 1-α anomer bromide 81, which, upon coupling with anionic 2-chloroadenine using potassium tert. Butoxide, afforded the heavily desired β isomer 83 (Scheme 14). The coupled nucleoside 83 was triturated with n-butyl acetate, using heptane and hot MeOH to enrich the desired β-isomer purity. The final debenzylation of 83 was carried out with catalytic NaOMe in MeOH to afford semi-pure 84. Recrystallization from MeOH afforded clofarabine 52 in an overall high yield [113].

The FDA has approved clofarabine (52) to treat pediatric patients with acute lymphoblastic leukemia at a dosage of 52 mg/m² over two hours daily for five consecutive days [114]. In vitro, in colon tumor cell lines, namely HCT116, HT-29, DLD-1, and WiDr, clofarabine has shown activity with an IC₅₀ value range of 0.12 to 0.67 µM (

Table 2. In vitro anti-colon cancer activity of clofarabine, 52.

Compound	Cell Lines	Antiproliferative Activity IC50 (µM)
Clofarabine (52)	HCT116	0.12
	HT-29	0.77
	DLD-1	0.07
	WiDr	0.67

) [115]. This drug expresses anticancer activity by the induction of apoptosis. Clofarabine, in the form of active triphosphate metabolite, inhibits DNA synthesis in cells. It also impedes the activity of ribonucleotide reductase and demonstrates DNA chain termination that results in cell death [116]. Due to its broad cytotoxicity profile, it is effective against various leukemia subtypes. It is administrated intravenously in combination with other anticancer drugs to treat both leukemias and solid tumors.

4.4. Gemcitabine, 53

Gemcitabine (dFdC) is a cytidine analog of 2′-deoxy-2′,2′-difluororibose sugar [117]. Initially, this molecule was investigated for antiviral activity, but the results of preclinical studies turned it into an anticancer agent. It was first approved by the FDA in 1996 and marketed under the trade name Gemzar (Gemcitabine HCl salt) and administered by intravenous infusion. Gemcitabine (dFdc) has been prescribed as a first-line chemotherapeutic agent alone for pancreatic ductal adenocarcinoma (PADC) and a variety of other solid tumors, including non-small cell lung cancer, bladder cancer, ovarian cancer, and breast cancer [118]. Recent studies have shown that gemcitabine, in combination with platinum or fluorouracil-based drugs, may significantly improve patients’ overall survival (OS) and progression-free survival (PFS). Combining gemcitabine with cisplatin (Cis), one of the main platinum-based drugs, and fluorouracil-based drugs is effective in treating advanced pancreatic cancer [119,120,121,122]. Gemcitabine cellular uptake is mediated by human equilibrate nucleoside transporter-1 (hENT-1) and human concentrative nucleoside transporters (hCNTs) [123]. After cellular uptake into malignant cells, it undergoes initial phosphorylation by deoxycytidine kinase (dCK), and nucleoside monophosphate kinase (NMPK) converts it into diphosphate (dFdCDP). Furthermore, cellular kinases transform dFdCDP into active gemcitabine triphosphate (dFdCTP) metabolites. Gemcitabine triphosphate (dFdCTP) enters growing DNA chains and leads to chain termination (Figure 8) [124,125]. The incorporation of dFdCTP into the cell’s DNA, thus replacing deoxycytidine during DNA replication, creates an irreversible error that inhibits DNA synthesis, results in apoptosis, and impedes tumor growth [126]. Additionally, gemcitabine diphosphate (dFdCDP) proves activity by inhibiting ribonucleotide reductase (RNR), an essential enzyme for creating new DNA nucleotides [121].

Figure 8. Gemcitabine design strategy inspired by natural nucleoside deoxycytidine; metabolism of gemcitabine and its mechanism of action. CDA—cytidine deaminase; dck—deoxycytidine kinase; NMPK—nucleotide monophosphate kinase, NDPK—nucleotide diphosphate kinase.

Figure 8. Gemcitabine design strategy inspired by natural nucleoside deoxycytidine; metabolism of gemcitabine and its mechanism of action. CDA—cytidine deaminase; dck—deoxycytidine kinase; NMPK—nucleotide monophosphate kinase, NDPK—nucleotide diphosphate kinase.

Originally, gemcitabine synthesis was accomplished by Hertel and co-workers (Scheme 15) [117] at Eli Lilly; the reported synthesis process initiated with the reaction of (R)-2,3-O-isopropylidene glyceraldehyde 85 with ethyl bromodifluoroacetate in the presence of activated Zn in ether/THF under Reformatsky conditions to produce a 3:1 diastereomeric mixture of compound 86. The major β-isomer of 86 was selected for further synthesis, and its isopropylidene group was removed by Dowex-H⁺, which, upon in situ cyclization, afforded the key difluoro lactone 87. The 2,5-hydroxy of lactone 87 was protected with tert-butyldimethyl silyl (TBDMS) to provide the silyl-protected derivative 88. The reduction of protected lactone 88 was carried out by diisobutylaluminum hydride (DIBAL-H) to render lactol intermediate 89, which was converted to reactive mesylate intermediate 90 by reaction with methanesulfonyl chloride (MsCl) in the presence of NEt₃ in DCM. Condensation of disilyl difluoro mesylate sugar 90 with trimethylsilylated cytosine 91 was performed under Vorbrüggen conditions in the presence of trimethylsilyl triflate in 1,2-dichloroethane to produce a protected nucleoside in a 4:1 mixture of the α/β anomeric ratio. Eventually, gemcitabine 53 was obtained by the removal of protecting groups.

The original synthesis process depicted in Scheme 15 lacks the desired higher β-selectivity of coupled nucleoside at the condensation step with difluororibanofuranose 90 and silylated cytosine 91, making it unsuitable for the large-scale synthesis of gemcitabine.

To address this drawback, in 2010, Chang et al. [127] revisited the synthesis and reported a highly efficient large-scale synthesis of gemcitabine from key scaffold 2,2-fluoro-α-ribofuranosyl bromide 96 (Scheme 16). In the revised procedure, the coupling of α-bromo difluoro sugar 96 with trimethylsilylated cytosine 91 enriched the β-anomeric selectivity of coupled product 97. This synthesis process was also initiated with D-glyceraldehyde 85, and via Reformatsky reaction, difluoro ester 86 was constructed in a 3:1 diastereomeric mixture. The hydroxyl group of 86 was protected with a p-phenyl benzoyl group to produce protected difluoro ester 92. The hydrolysis of ester 92 was carried out with K₂CO₃ in THF/MeOH/H₂O to afford the pure potassium salt of erythro-pentanoate compound 93 with high diastereomeric purity. Deprotection of the isopropylidene group of 93 with 12 N HCl in acetonitrile afforded 5-hydroxy-3-PhBz-protected difluororibanolactone, which was further treated with benzoyl chloride in the presence of pyridine in EtOAc to furnish fully protected difluororibanolactone 94. The reduction of lactone 94 with lithium tri-tert-butoxyaluminumhydride afforded a lactol intermediate. Furthermore, this lactol was treated with diphenyl chlorophosphite to yield pure β-phosphate 95. β-phosphate 95 was treated with HBr in AcOH to furnish the key α-ribofuranosyl bromide intermediate 96. Furthermore, S_N2 coupling of 96 with trimethylsilyl-protected cytosine 91 gives the major desired β-nucleoside 97 (β/α = 5.5/1). The deprotection of protecting groups of 97 with ammonia solution in methanol yielded pure gemcitabine 53 in a 20% overall yield.

Gemcitabine (53) is an FDA-approved drug for treating breast, ovarian, non-small cell lung, and pancreatic cancers [128]. Gemcitabine has also shown good activity in cervical cancer cell lines and against solid tumors [129]. This drug was initially developed for its antiviral effect but later advanced as an anticancer agent. Gemcitabine monotherapy generates an objective tumor response in 18 to 26% of patients with advanced non-small cell lung cancer (NSCLC), and it demonstrates similar efficacy with cisplatin plus etoposide [130]. Gemcitabine is also a beneficial novel chemotherapy alternative for patients with advanced pancreatic cancer [131]. Its apparent survival and palliative benefits to patients with pancreatic cancer have been proven. The administration of gemcitabine improves both the duration and quality of the survival of patients with cancer [130]. In vitro gemcitabine against cervical cancer cell lines has demonstrated IC₅₀ values of 0.89 μM, 0.32 μM, and 0.11 μM in CaLo, HeLa, and CasKi cells, respectively (

Table 3. In vitro anticancer activity of gemcitabine (53) against cervical cell lines.

Compound	Cervical Cancer Cell Lines	IC50 (µM)	IC30 (µM)	Ref.
Gemcitabine	SiHa Cells	203 ± 12.3 µM	0.1 ± 0.05 µM	[129]
	CaLo Cells	0.89 ± 0.10 µM	0.2 ± 0.06 µM
	InBl Cells	0.63 ± 0.07 µM	0.07 ± 0.01 µM
	HeLa Cells	0.32 ± 0.02 µM	0.05 ± 0.01 µM
	C33A Cells	0.27 ± 0.08 µM	0.04 ± 0.01 µM
	CasKi Cells	0.11 ± 0.04 µM	0.02 ± 0.09 µM

IC₅₀: concentration of drug to inhibit cell growth by 50%; IC₃₀: concentration of drug to inhibit cell growth by 30%.

) [129].

Gemcitabine also exhibited good antiviral activity against coronaviruses. It displayed an EC₅₀ value of 1.2 μM against MERS-CoV, 4.9 μM against SARS-CoV-1, and 1.2 μM against SARS-CoV-2 in Vero cells (

Table 4. In vitro antiviral profile of gemcitabine against various viruses in different cell lines.

Compound	Virus	Cell Line	EC50	CC50	Ref
Gemcitabine (53)	MERS-CoV	Vero E6 Cells	1.2 µM	>10 µM	[132]
	SARS-CoV	Vero E6 Cells	4.9 µM	>10 µM	[132]
	ZIKA	RPE Cells	0.01 µM	>10 µM	[134]
	HCV	Huh-7 Cells	12 nM	>44,444 nM	[135]
	HIV	U373-MAGI-CXCR4CEM	16.3 nM	ND	[136]
	IAV (Influenza A)	RPE Cells	0.068 ± 0.005 µM	ND	[137]
	HSV-1	RPE Cells	>4 logs at 3 µM	ND	[137]
	SARS-CoV-2	Vero Cells	1.2 ± 1.1 µM	>300 µM	[133]

ND: not determined.

) [132,133].

In the SARS-CoV-2 assay, gemcitabine’s cytotoxicity was assessed to be greater than 300 μM (CC₅₀) with a selective index greater than 250 [133]. It has expressed an EC₅₀ value of 0.068 μM against the influenza virus and a degree of inhibition > 4 logs at a concentration of 3 μM against herpes simplex virus 1 (HSV1) in RPE cells [137]. In U373-MAGI-CXCR4_CEM cell lines against human immune deficiency virus (HIV), gemcitabine exhibited a sub-micromolar EC₅₀ of 16.3 nM (Table 4) [136]. Against zika virus, in RPE cells, it expressed an EC₅₀ value of 0.01 µM with a cytotoxicity (CC₅₀) of >10.0 µM [134]. Gemcitabine has also been tested against HCV in the Huh-7 cell line, where it demonstrated an EC₅₀ value of 15 nM with a cytotoxicity (CC₅₀) > 44,444 nM [135].

4.4.1. Drug Resistance of Gemcitabine (53)

The long-term use of gemcitabine is associated with the development of drug resistance, which significantly limits its efficacy against cancer cells. Several factors, such as alterations in nucleoside transporter proteins (hENT-1), limit the uptake of gemcitabine into cancer cells, and cytidine deaminase at the N4 position of the cytidine base generates the deaminated toxic metabolite 2′,2′-difluoro-2-2′deoxyuridine (dFdU, Figure 8) [138]. Other challenges include a high expression of drug efflux pumps, a poor conversion of initial phosphorylation by dCK (the rate-limiting step), poor oral bioavailability, and innate or acquired drug resistance, limiting its monotherapy response rates to <10% in pancreatic cancer treatment [139,140].

4.4.2. Implementation of Prodrug Strategies to Overcome Gemcitabine (53) Resistance and Cellular Uptake

To deal with the drug resistance, biostability, and bioavailability of gemcitabine, several chemical modifications via prodrug strategies have been practiced. Various prodrug modifications have been adopted, including the insertion of ester and amide at the cytidine N⁴ site and at C-5′-OH. Additionally, the phosphoramidate approach at C-5′-OH of gemcitabine was implemented. In the case of nucleoside molecules, prodrug tactics have proven to be an effective technique for oral delivery [141,142,143]. The introduction of amide and ester substitutions at the N⁴ site of gemcitabine may inhibit cytidine deamination and improve in vivo stability. Phosphoramidate prodrugs bypass the first-rate liming step of mono phosphorylation and directly metabolize to monophosphate gemcitabine (dFdCMP, Figure 8) [144].

Amidon et al. first reported amino acid ester prodrug strategies targeting the hPEPT1 transporter. The group synthesized a series of L and D aliphatic and aromatic amino acid ester prodrugs and evaluated their cellular affinity. The in vitro studies of 5′-L-valyl and 5′-L-isoleucyl aliphatic amino esters displayed enhanced transport into the cells and increased cytidine deaminase resistance [145].

LY2334737 (100): Bender et al. at Eli Lilly synthesized orally active gemcitabine prodrug LY2334737 (Figure 9). It is an amide prodrug of valproic acid at the N⁴ position of cytosine and has entered phase 1 clinical trials [146]. In preclinical in vitro and in vivo studies, the anticancer profile and pharmacokinetics of LY2334737 were evaluated. The data revealed that it was significantly absorbed through the intestinal epithelium with enhanced bioavailability. Carboxylesterase 2 (CES2) performs the hydrolysis of LY2334737. However, in vivo, the hydrolysis of LY2334737 is relatively slow, which results in a slow release of gemcitabine [147]. Furthermore, interestingly, mice treated with LY2334737 showed a lower ratio of metabolized dFdU than gemcitabine. These results demonstrate that an N⁴-modification strategy of amide linkage blocks the site of deamination by cytidine deaminase [147,148]. Prodrug LY2334737 with an amide linkage is proven to be both chemically and enzymatically stable and expresses a high release of parental nucleoside in systemic circulation.

CP-4126 (101): CP-4126 is a 5′-elaidic acid ester lipid conjugate prodrug of gemcitabine. This prodrug was designed to bypass hENT1-related resistance to gemcitabine [149]. In vitro studies revealed that the lipid conjugate prodrug CP-4126 (Figure 9) exhibited enhanced cellular uptake compared to gemcitabine (53) independent of hENT1 cell transporters [150]. Furthermore, in various mouse models, intraperitoneally and orally, CP-4126 was as effective as gemcitabine [151]. The encouraging in vitro and in vivo results of CP-4126 led to clinical trials of this prodrug under orphan drug designation in the United States (US) and Europe. However, in phase I/II clinical trials, the safety and efficacy of CP-4126 were evaluated with oral dosage administration, and no patients achieved a partial or complete response to solid tumors. As a result, CP-4126 was halted for further development [149].

NUC-1031 (101): NUC-1031 is a ProTide transformation of gemcitabine that was designed to overcome key gemcitabine resistance mechanisms. McGuan et al. reported the first anticancer drug with phosphoramidate prodrug NUC-1031 (Figure 9) to bypass dCK-mediated activation [152]. Gemcitabine has three known resistance mechanisms: (i) The suppressed expression of hENT1 results in impaired cellular uptake [153]. (ii) After gemcitabine enters into cells, phosphorylation by dCK is considered to be a major rate-limiting step that can result in reduced drug activation [154]. (iii) The overexpression of CDA in cancerous cells mediates gemcitabine inactivation to produce major dFdU, diminishing the overall drug activation [155]. Interestingly, NUC-1031 expressed cellular uptake independent of nucleoside transporters and mono phosphorylation without dCK. Furthermore, it was also not a substate of CDK for deamination [156]. Additionally, it expressed greater plasma stability and enhanced intracellular levels of dFdCTP with fewer toxic metabolites. Building on NUC-1031’s promising preclinical data, several clinical phase studies were conducted to evaluate its efficacy against various malignancies. However, despite the encouraging preclinical results, the clinical trial outcomes revealed no significant advantage of NUC-1031 over its parent drug, gemcitabine [157,158].

Figure 9. Prodrug structures of gemcitabine (53).

Figure 9. Prodrug structures of gemcitabine (53).

Cyclic Phosphoramidate (102): Recently, Zhang et al. reported a cyclic phosphate prodrug (103, Figure 9) named gemcitabine. Cyclic prodrug 102 exhibited superior in vitro antiproliferative activity with an IC₅₀ value in the range of 3.6–19.2 nM against multiple cancer cell lines compared to the linear phosphoramidate prodrug NUC 1031 (101) [159]. Several modified prodrug strategies are currently being explored to address drug resistance and enhance the PK/PD properties of gemcitabine.

4.5. Floxuridine (FUDR, 54)

Floxuridine is chemically known as 5-fluoro-2′-deoxyuridine (FUDR). It is a chemotherapeutic 2′-deoxyribose nucleoside analog that rapidly catabolizes to 5-FU after intravenous administration [160]. FUDR was approved by the FDA in 1970 and is used to treat colon, rectum, breast, stomach, and pancreas cancers. It is the drug most often used to treat colorectal cancer. Due to its short plasma half-life, FUDR is an ideal drug for continuous hepatic artery infusion (HAI) [161]. So far, a total of 67 active clinical trials have been reported [162]. Mechanistically, it inhibits DNA synthesis by the inhibition of thymidylate synthase (TS) [163]. Additionally, FUDR also inhibits RNA synthesis via incorporation into RNA, leading to the formation of damaged RNA [164].

Primarily, Hoffer et al. reported the synthesis of FUDR, which was valuable for large-scale synthesis over enzymatic procedures. In the reported method, the synthesis of FUDR was achieved by the coupling of protected 1-chloro-2-deoxyribose with monomercurypyrimidines [165]. However, several approaches have been reported for FUDR, but the synthesis reported by Robins et al. has revealed an advantage of selective starting material 103 with fixed β-conformation (Scheme 17). This synthesis process was started with the desired stereochemistry and readily available 2′-deoxynucleoside 103. The acetylation of compound 103 was carried out with acetic anhydride (Ac₂O) in the presence of 4-dimethylaminopyridine (DMAP) to give the acetyl-protected uracil intermediate 104. Fluorination on uracil 104 with CF₃OF/CCl₃F, followed by acetyl deprotection of 105, afforded floxuridine (54) [166].

Later, Hajime et al. reported the efficient stereoselective synthesis of floxuridine (54) [167]. The synthesis process commenced with 2-deoxy ribose 106. The protection of 3 and 5 hydroxyls of 106 with p-choro benzoyl, followed by 1-chlorination of the sugar, afforded critical intermediate 107 [168]. The coupling of 107 with 5-fluoro silyl uracil base in the presence of p-nitro phenol constructed majorly β-anomer nucleoside 108, whereas in the same coupling conditions, the addition of pyridine formed undesired α-nucleoside 109. Final deprotection of β-nucleoside 108 provided floxuridine 54 (Scheme 18).

FUDR is predominantly administered by continuous infusion into the hepatic artery to treat colon cancer. It is also used for the treatment of solid tumors, including liver tumors, gastrointestinal tumors, and adenocarcinoma, with a dosage range of 0.1–0.6 mg/kg/day [169].

In an acute lymphoblastic leukemia assay, floxuridine showed an IC₅₀ value of 0.02 against mouse leukemia L1210 cells (ATCCCCL 219) and an IC₅₀ value of 0.05 µM against the human T lymphoblastoid CCRF-CEM cell line (

Table 5. Cytostatic activity data for floxuridine.

Compound	Cell Lines	IC50 (µM)
Floxuridine (FUDR, 54)	L1210	<0.02 ± 0.002
	L929	>25
	HeLa S3	>25
	CCRF-CEM	0.05 ± 0.004

IC₅₀: inhibitory concentration of drug to inhibit 50% growth of cells; L929: murine L929 cells (ATCC CCL 1); HeLaS3: human cervix carcinoma HeLa S3 cell.

) [170]. In toxicological studies, floxuridine demonstrated toxicity in mice, dogs, and rabbits with LD₅₀ values of 880, 157, and 94 mg/kg, respectively. Notably, these doses are more than 100 times higher than the typical human dosage [169].

4.6. Capecitabine, 55

Capecitabine is an FDA-approved drug used to treat metastatic breast, colorectal, different gastrointestinal, and pancreatic cancers both as a single regimen or in combination [171]. It is chemically known as N⁴-pentyloxycarbonyl-5′-deoxy-5-fluoro cytidine and is being sold under the brand name Xeloda. Capecitabine is an orally administered prodrug of 5-FU, which expresses selectivity to the tumor by utilizing a higher intratumor concentration of thymidine phosphorylase. It was invented to overcome the disadvantages and side effects associated with 5-FU. 5-FU therapy causes bone marrow, central nervous system, gastrointestinal tract, and skin toxicity. Metabolically, capecitabine converts to 5-FU via a three-step sequential metabolic conversion. After intestinal absorption in the liver, it converts to 5′-deoxy-5-fluorocytidine (5′-DFCR) by carboxylesterase. Furthermore, 5′-DFCR transfers to 5′-deoxy-5-fluorouridine (5′-DFUR) by cytidine deaminase (CDA) in the liver and tumor tissues. Finally, the thymidine phosphorylase enzyme, predominantly expressed in tumor tissues, converts 5′-deoxy-5-fluorouridine (5′-DFUR) to 5-FU [116,172]. Several clinical trials involving capecitabine, in combination with other approved anticancer drugs, are underway to determine its ability to treat various cancers [173].

The first design and synthesis of capecitabine was reported by Shimma et al. (Scheme 19) [141]. The synthesis process started with the glycosylation of 1,2,3-tri-O-acetyl-5-deoxyribose 113 with TMS-protected 5-fluorocytosine 112 in the presence of stannic chloride (SnCl₄) to produce 114. Acylation of N⁴ amino derivative 114 was carried out with n-pentyl chloroformate to afford 115. Finally, the deprotection of acetyl groups of 115 yielded capecitabine 55.

As depicted in Scheme 20, Mekala et al. reported the scalable synthesis of capecitabine [174]. The synthesis process started with commercially available D-ribose, by which key intermediate 120 was constructed. Methylation followed by protection of the 2,3 dihydroxy ribose with an isopropylidene group rendered 116. After that, 5-hydroxy mesylation of 116 furnished 117, which, upon reduction with sodium borohydride/lithium chloride (NaBH₄/LiCl), afforded intermediate 118. The in situ ketal deprotection of 118 was carried out by 0.1% aq. H₂SO₄ to obtain intermediate 119. O-acetylation of 119 by acetic anhydride produced key intermediate 120 in a good yield. Furthermore, the coupling of 120 with silylated 5-fluorocytosine 112 afforded coupled nucleoside 114. The treatment of 114 with n-pentyl chloroformate rendered 115, which, upon deacetylation, gave capecitabine (55) in an overall good yield.

In cases where single-agent fluoropyrimidine is preferred, the FDA has approved capecitabine as a first-line treatment for metastatic CRC. It is also approved as a single agent to treat anthracycline- and paclitaxel-resistant metastatic breast cancer or when anthracycline treatment is contraindicated [175]. Capecitabine is also approved in combination with docetaxel for the cure of anthracycline-failed cancer therapies [176]. Studies revealed that capecitabine yields a higher concentration of 5-FU in tumors than in healthy cells/tissues [177]. In studies of a human cancer xenograft model, capecitabine was more efficient at a more comprehensive dose range than 5-FU, with fewer adverse effects. Furthermore, capecitabine has additional characteristics that are antimetastatic and anticachectic [178]. Capecitabine has also demonstrated potency in tumors known to be resistant to 5-FU [179]. Studies also proved that capecitabine may demonstrate benefits to cure prostate, renal, and ovarian carcinomas as a single agent or in combination with other anticancer drugs [179].

In in vitro models, capecitabine was found to be more sensitive in both colorectal LS174T WT and thymidine phosphorylase-transfected LS174T-c2 cells cultivated without and with Hepatoma HepG2 cells [180]. LS174T WT showed an IC₅₀ value of 890 µM without HepG2 and 630 µM with HepG2 (

Table 6. In vitro anticancer activity of capecitabine (55) in human colorectal and human colorectal + hepatoma cell lines.

Compound	Human Colorectal and Hepatoma Cells	IC50 (µM)	Ref.
Capecitabine (55)	LS174T WT	890 ± 48 µM	[180]
	LS174T-c2	330 ± 4 µM
	LS174T WT + HepG2	630 ± 14 µM
	LS174T-c2 + HepG2	89 ± 6 µM

) [180]. In LS174T-c2 cell lines, it showed an IC₅₀ value of 330 µM without HepG2 and 89 µM with HepG2 [180].

4.7. Tegafur, 56

Tegafur (1-(2-tetrahydrofuryl)-5-fluorouracil) is also a chemotherapeutic prodrug of 5-fluorouracil (5-FU). It is used in a 1:4 molar ratio as a combination of tegafur–uracil (UFT) to enhance oral bioavailability [181]. Combination with uracil protects the gastrointestinal tract from 5-FU toxicity and reduces the adverse effects of 5-FU [182]. Tegafur–uracil is being used as a first-line treatment for metastatic colorectal cancer and is approved in many countries. However, the US-FDA has not yet approved it, and it is currently in a phase II clinical trial in the US and Canada. The next generation of oral drug formulation in combination with tegafur, gimeraci (122), and oteracil (123, Figure 10) is called S-1, with a fixed molar ratio of 1:0.4:1, and is undergoing extensive preclinical studies [183]. Currently, S-1 is approved in China, Japan, Taiwan, Korea, and Singapore for treating gastric cancers [184].

Figure 10. Structures of tegafur combination drugs.

Figure 10. Structures of tegafur combination drugs.

Tegafur converts into 5-FU by liver microsomal cytochrome P450 enzymes, mainly by CYP 2A6 [185]. Intracellularly, 5-FU converts into its active metabolites FdUMP and FUTP [186]. FdUMP inhibits the activity of the thymidylate synthase enzyme, resulting in decreased thymidine synthesis, and ultimately inhibits DNA synthesis. The other active metabolite, FUTP, is integrated into cellular RNA, disrupts RNA function, and causes toxicity to tumor cells [187,188].

Scheme 21 illustrates the synthesis of tegafur (56). To afford the coupled product (56), 5-fluorouracil was coupled with 2-acetoxytetrahydrofuran 124 in the presence of 1,8-diazabicyclo [5.4.0] undec-7-ene (DBU) in DMF. The product was further treated with aqueous ethanol and finally crystallized by ethanol to obtain pure tegafur (56) [189]. Since no significant difference in the anti-tumor activity of enantiomers was observed, tegafur is used as a racemic mixture.

Tegafur offers a safe option for colorectal cancer in individuals with DPD deficiency [190]. The direct use of 5-FU in patients with DPD deficit expressed life-threatening adverse effects. In in vitro studies, tegafur expressed activity in multiple human cancers, including colon carcinoma (HT-29 cells; IC₅₀ value of 201 µM), pancreatic carcinoma (BxPC-3 cells; IC₅₀ value of 172 µM), and Ewing sarcoma (SK-ES cells; IC₅₀ value of 174 µM,

Table 7. In vitro evaluation of tegafur in various cancers.

Compound	Cells	IC50	Ref
Tegafur (56)	HT-29	201 µM	[191]
	BxPC-3	172 µM	[191]
	SK-ES	174 µM	[191]

) [191].

4.8. RX-3117, 57

RX-3117 is a novel fluorocyclopentenyl cytosine nucleoside that belongs to a particular class of molecules known as carbocyclic nucleosides [192]. RX-3117 contains significant modifications in the sugar part that include the replacement of the oxygen atom of the ribose ring with a carbon consisting of an endo alkene between the 4′ and 6′ positions of the cyclopentyl ring. It is an orally active nucleoside that has shown excellent anti-tumor activity against various cancers, including pancreatic, colon, renal, and lung cancers [193]. In xenograft models, RX-3117 has demonstrated strong activity against various cancer cell lines, including gemcitabine-resistant variants. In addition, RX-3117 has a better pharmacological profile and oral bioavailability relative to gemcitabine and is not a substrate for the CDK enzyme [192,194]. Mechanistically, RX-3117 is converted to its active triphosphate form by uridine–cytidine kinase 2 (UCK2), a tumor-specific enzyme predominantly expressed in cancer cells. The active triphosphate form of RX-3117 is incorporated into cancer cells’ DNA and RNA and induces apoptotic cell death. Some studies suggested that RX-3117 acts as a DNA methyl transferase 1 (DNMT1) inhibitor [193]. DNMT1 is an enzyme responsible for maintaining the methylation pattern of DNA during cell division, which is an essential epigenetic mechanism. The overexpression of DNMT1 and abnormal DNA methylation patterns are responsible for various types of cancer. However, the complete mechanism of action of RX-3117 still needs to be established [193]. RX-3117 is currently undergoing clinical trials in combination with Abraxane^® to treat metastatic pancreatic cancer [195] (clinicalTrails.gov identifier: NCT03189914), metastatic bladder cancer, and solid tumors (clinicalTrails.gov identifier: NCT02030067) [196].

Initially, Choi et al. reported the synthesis of RX-3111, as depicted in Scheme 22 [197]. The acetonide protection of D-ribose followed by TBDPS protection constructed 125. The sequential treatment of compound 125 with methyl triphenylphosphonium bromide (Ph₃PCH₃Br), potassium tert-butoxide (^tBuOK), and Swern oxidation, followed by Grignard reaction with vinyl magnesium bromide, furnished diene 126 as a single stereoisomer. TBDPS deprotection followed by benzyl re-protection afforded 127. The ring-closing metathesis of 125 was carried out with a Grubbs second-generation catalyst to give β-hydroxycyclopentene 128. The oxidation of intermediate 128 with pyridinium dichromate (PDC) in DMF undergoes oxidative rearrangement to yield cyclopentenone 129. Iodination of 129 with iodine in pyridine/THF constructed the iodo intermediate, which, upon a further stereoselective reduction of ketone with NaBH₄/CeCl₃, followed by TBDPS protection, afforded intermediate 130. The electrophilic vinyl fluorination of 130 was achieved by treating it with n-BuLi and N-flurobenzenesulfonimide (NFSI), and the deprotection of TBDPS with TBAF gave key intermediate fluoro-cyclopentenol 131. The coupling of 131 with N³-benzoyl uracil under Mitsunobu conditions yielded the coupled nucleoside, which, upon the sequential deprotection of benzoyl and isopropylidine, produced the final uracil analog 132. Finally, uracil analog 132 was converted to targeted compound RX-3117 (57), as depicted in Scheme 22.

RX-3117 has expressed broad-spectrum anti-tumor activity against numerous tumors, with IC₅₀ values in the sub-micromolar range. In in vitro screening, it has demonstrated an IC₅₀ of 0.25 μM against lung cancer NCIH226 cell lines, an IC₅₀ value of 0.19 μM against colon HCT116 cell lines, an IC₅₀ value of 0.83 μM against brain U251 cell lines, an IC₅₀ value of 0.82 μM against leukemia K562 cell lines, an IC₅₀ value of 0.63 μM against prostate PC-3 cell lines, an IC₅₀ value of 0.79 μM, against liver HepG2 cell lines, and an IC₅₀ value of 0.83 μM against kidney UMRC2 cell lines (

Table 8. In vitro broad-spectrum anticancer activity of RX-3117 in several cancer cell lines.

Compound	Cancer Cell Line	IC50	Ref
RX-3117 (57)	NCIH226 (lung)	0.25 μM	[197]
	HCT116 (colon)	0.19 μM
	U251 (brain)	0.83 μM
	K562 (Leukemia)	0.82 μM
	PC-3 (prostate)	0.63 μM
	HepG2 (liver)	0.79 μM
	UMRC2 (kidney)	0.83 μM

) [197]. Additionally, in vivo, RX-3117 significantly reduced tumor growth by 58.1% at 3 and 10 mg/kg doses in a nude mouse tumor xenograft model with lung cancer [197].

The U.S. FDA and the European Commission (EC) recently designated RX-3117 as an orphan drug for treating pancreatic cancer.

4.9. NUC-3373, 58

Fosifloxuridine nafalbenamide is an innovative pyrimidine phosphoramidate nucleotide anticancer prodrug of FUDR (54). It was developed by Nucana to overcome the limitations and pharmacological challenges of 5-FU treatment [198,199]. NUC-3373 is a potent targeted inhibitor of thymidylate synthase (TS). It is currently being investigated in phase 1b/2 clinical trials to treat various solid tumors [200]. The anticancer agents 5-FU (50), fluorodeoxyuridine (FUDR, 54), and capecitabine (55) exert their main anticancer activity by generating 5-fluorodeoxyuridine triphosphate (5-FdUTP) via 5-fluorodeoxyuridine monophosphate (5-FdUDR-MP), which is a thymidine kinase (TK) enzyme-dependent activation [201]. Therefore, TK mutations mainly contribute to resistance to these drugs. Additionally, DPD’s metabolism diminishes its therapeutic efficacy [202,203].

The large-scale synthesis of NUC-3373 is depicted in Scheme 23. Silylated 5-fluorouracil was coupled with 1-chloro-3,5-di-(4-chlorobenzoyl)-2-deoxy-D-ribose 133 in the presence of p-toluenesulfonic acid (p-TSA) in DCM to give the desired coupled β-isomer intermediate 134. The deprotection of the 4-Cl benzoyl group was carried out by methanolic ammonia to obtain FUDR 54. Furthermore, the coupling of phosphoramidate reagent 137 with 54 was performed in the presence of N-methylimidazole (NMI) in THF to produce a 1:1 diastereomer mixture of NUC-3373 (58) [204].

The phosphoramidate moiety of NUC-3373 protects it from DPD degradation, which results in the reduced exposure of catabolites and associated toxicities compared to 5-FU. NUC-3373 also demonstrates a significantly prolonged half-life in plasma (6–10 h) compared to 5-FU (8–14 min) [205]. Owing to the better PK properties and direct delivery of FdUMP, NUC-3373 is administered over a much shorter period than 5-FU. A phase Ib/II study of NUC-3373 in combination with other anticancer agents used to treat advanced metastatic CRC is ongoing (clinicalTrails.gov identifier: NCT03428958). In these phase II studies, the promising anticancer activity of NUC-3373 has been observed mainly in pre-treated patients who have relapsed on prior fluoropyrimidine therapy [206]. The data from the phase II study also demonstrate that NUC-3373 exerts toxicities of neutropenia, mucositis, diarrhea, and hand–foot syndrome [203]. Early in vitro data of NUC-3373 show that it is much more active than 5-FU. Its cytostatic activity was evaluated against several established tumor cell lines. The anticancer studies of NUC-3373 were performed in wild-type L-1210, CEM, and Hela cells and in the thymidine kinase-deficient (TK⁻) mutant of parental cell lines to investigate the effect of TK deficiency on the cytostatic activity of NUC-3373.

NUC-3373 has demonstrated an IC₅₀ value of 0.011 µM in the L1210 cell line. Thus, it demonstrated 30-fold greater activity than 5-FU in the same cell lines. However, NU-3373 also maintains its anticancer potency in the case of TK-deficient L1210 cell lines (IC₅₀ of 0.045 µM,

Table 9. In vitro cytostatic activity of NUC-3373 in various cell lines.

Compound	Tumor Cells Lines	IC50 (µM)
NUC-3373 (58)	L1210/W	0.011 ± 0.007
	L1210/TK−	0.045 ± 0.027
	CEM/0	0.068 ± 0.035
	CEM/TK−	0.31 ± 0.06
	HeLa	0.065 ± 0.013
	HeLa/TK−	2.5 ± 1.3

L1210/TK⁻, CEM/TK⁻, and HeLa/TK⁻: thymidine kinase (TK)-deficient cell lines.

) [204]. Similarly, it also expressed potent anticancer activity, with IC₅₀ values of 0.068 and 0.065 µM against CEM/0 and HeLa, respectively. This compound also revealed excellent anticancer activity against TK-deficient cell lines CEM/TK⁻ and HeLa/TK⁻.

5. Conclusions

Nucleos(t)ides (NAs) are well-practiced molecules in the field of anticancer and antiviral drug discovery and development; among them, fluoro-containing nucleos(t)ides demonstrate potent therapeutic efficacy. Numerous fluoro-containing nucleosides/nucleotides have received FDA approval as drugs to treat various cancers. It is crucial to understand how the judicial utilization of chemical space in the active nucleoside molecule to install fluorine in the sugar or base determines whether it stabilizes or destabilizes the parental molecule. The scientific findings demonstrate that incorporating a fluorine atom into the nucleos(t)ides provides distinct advantages, including enhanced stability, improved receptor binding interactions, and favorable pharmacokinetic properties. Nucleos(t)ides as therapeutics interact with the DNA/RNA polymerase and inhibit the elongation of the DNA/RNA of highly dividing cancerous cells. [¹⁸F]-radiolabeled nucleosides are well utilized in the diagnosis and progression of several diseases. This review focuses on the fluorinating reagents used in the synthesis of fluorinated nucleos(t)ide analogs, the clinical utilization of [¹⁸F]-radiolabeled nucleosides, and the synthesis and anticancer activity of FDA-approved drugs as well as ongoing fluoro-containing nucleos(t)ide clinical candidates.