Background

On November 10, at the American Heart Association (AHA) Scientific Sessions, Merck announced the Phase III clinical results of the world’s first oral PCSK9 inhibitor, MK-0616 (Enlicitide). Enlicitide decanoate demonstrated significant lipid-lowering efficacy with a favorable safety profile.

The precursor of MK-0616 was identified by Merck through an mRNA display-based cyclic peptide library developed in collaboration with Ra Pharma. The synthesis of this structurally complex molecule was achieved via enzymatic catalysis combined with a “Northern Fragment and Southern Fragment” convergent strategy. By optimizing crystallization processes to avoid chromatographic purification, Merck ultimately realized large-scale production at the hundred-kilogram level.

Significant Phase III Clinical Efficacy of MK-0616

Enlicitide is a small-molecule macrocyclic peptide compound that specifically binds to PCSK9 (proprotein convertase subtilisin/kexin type 9) expressed in the liver. By blocking the interaction between PCSK9 and the LDL receptor, Enlicitide reduces LDL receptor degradation, enhances hepatic clearance of LDL-C, and thereby significantly lowers plasma low-density lipoprotein cholesterol levels.

Two Phase III studies in the lipid-lowering field—CORALreef Lipids and CORALreef HeFH—demonstrated the following results:

1. CORALreef Lipids 

evaluated the efficacy and safety of Enlicitide in adults with atherosclerotic cardiovascular disease (ASCVD) risk or a history of ASCVD events. Among adults receiving stable lipid-lowering therapy or intolerant to statins, once-daily oral Enlicitide achieved a primary LDL-C reduction of 55.8% at Week 24.

Additionally, 67.5% of patients achieved at least a 50% reduction from baseline with an absolute LDL-C level below 55 mg/dL (1.42 mmol/L). The LDL-C reduction was sustained at 47.6% at Week 52.

2. CORALreef HeFH 

evaluated Enlicitide in adults with heterozygous familial hypercholesterolemia (HeFH) receiving stable lipid-lowering therapy. LDL-C reductions of 59.4% at Week 24 and 61.5% at Week 52 were achieved.

mRNA Display Technology

Peptide screening technologies include DNA-encoded libraries (DEL), phage display, and mRNA display.

DEL libraries can reach sizes up to 10¹¹ and readily incorporate non-proteinogenic amino acids, enabling diverse modifications of cyclic peptide backbones and side chains, while the attached DNA facilitates hit identification. Phage display has lower diversity and intrinsic limitations in in vivo surface expression.

mRNA display technology, first reported in 1997, enables library sizes of 10¹²–10¹⁴ due to high sequence diversity. It is performed entirely in vitro, avoiding biological constraints and allowing the incorporation of non-natural amino acids.

Experimental Workflow of mRNA Display Technology

1. DNA Library Construction and In Vitro Transcription

A large-capacity DNA library encoding mRNA sequences is constructed and transcribed in vitro using T7 RNA polymerase.

2. Puromycin Conjugation

The mRNA library is linked to puromycin via a short linker.

3. In Vitro Translation

Ribosomes translate the mRNA–puromycin templates in vitro. Upon reaching the 3′ end, puromycin enters the ribosomal A site and forms an amide bond with the C-terminus of the nascent peptide, terminating translation and generating an mRNA–puromycin–peptide conjugate.

4. Peptide Cyclization

Cyclization is achieved using reagents such as DSG, DBX, TCEP, NaBH₃CN, or CuAAC to connect peptide segments.

5. cDNA Reverse Transcription

Reverse transcription via RT-PCR generates a complementary DNA strand, yielding a cDNA/mRNA–puromycin–peptide complex.

6. Selection

The target protein is immobilized on magnetic beads. After 5–10 rounds of affinity selection, high-affinity target-binding peptide complexes are enriched.

7. Amplification and Enrichment

Proteins are released by heat or high pH, followed by PCR amplification and sequencing to identify peptide structures with high specificity for the target protein.

Merck and Ra Pharma’s mRNA Display Screening Strategy

Merck collaborated with Ra Pharma to construct a linear peptide library containing two reactive thiol groups using mRNA display technology. Cyclization was achieved via 1,3-bis (bromomethyl) benzene (DBX) crosslinking, leading to the identification of two candidate peptides.

Compound 13 exhibited an LDL receptor binding activity of 134 nM and showed significant inhibition in the LDLR-FRET assay. Compound 12 showed weaker activity (956 nM) but possessed a unique spatial structure favorable for further optimization.

Kilogram-Scale New Synthesis Process for the MK-0616 Northern Fragment

MK-0616 originates from an mRNA display-derived macrocyclic peptide and features a highly complex structure:

an amino acid sequence of eight residues (six of which are non-proteinogenic and chemically modified), a 37-membered macrocycle formed via triazole, alkene, and thioether linkages that restrict molecular conformation and enhance stability, a non-peptidic moiety increasing molecular complexity and binding specificity, a terminal ammonium side chain influencing target binding and solubility, and a decanoate permeation enhancer to improve oral bioavailability.

This structural complexity renders solid-phase synthesis highly challenging.

A new synthetic route for Intermediate 1 (Northern Fragment) was developed. Starting from 5-fluoro-tryptophan, the process involves five steps, achieving >150 kg per batch with an overall yield of 50%.

Step 1: Allylation to Form Compound 6

Using NMP as solvent and sodium tert-butoxide as base, the allylation with allyl chloride was optimized. Increasing allyl chloride to 1.8 equivalents, controlling sodium tert-butoxide at 2.01–2.05 equivalents, and maintaining NMP water content below 300 ppm reduced byproduct 14 to below 1%.

During workup, the reaction mixture was acidified to pH 1–2 with HCl to hydrolyze byproduct 14 back to starting material 5, then neutralized with 5% aqueous Na₂CO₃ to induce crystallization. Filtration yielded target compound 6.

Step 2: Amide Coupling to Form Compound 8

Using 1.0 equivalent HATU and 1.2 equivalents DIPEA initially generated 17.2% of byproduct 15. Mechanistically, residual HATU/DIPEA activated the carboxyl group of compound 8 to form intermediate 8-INT, which reacted with compound 6, while ester exchange under DIPEA also contributed.

Optimization to 1.0 equivalent each of HATU and DIPEA, combined with reversing the addition order—slowly adding activated intermediate 7-INT into the solution of compound 6—reduced byproduct 15 to 1.2%.

Step 3: One-Pot Synthesis of Compound 10

When purified compound 8 was used, additional HOAt was required to control isomer content below 1.5%. However, as HOAt was already generated in situ during the first coupling, no extra HOAt was needed, further reducing costs.

Compound 9 (1.15 eq), HATU (1.3 eq), and DIPEA (2.4 eq) were added directly to the reaction stream of compound 8. Following standard workup, compound 10 was obtained as an MTBE solution.

At scales exceeding 200 kg, the two-step process achieved 87% yield with 90.8% purity, improving yield by 10% and shortening cycle time by 36%.

Synthesis of Compound 9 Based on Cbz-trans-3-hydroxy-L-proline

One-Pot Synthesis of Compound 10

Step 4: Deprotection

Conventional deprotection using mineral acids (HCl, H₂SO₄) or TFA resulted in low yield and incomplete deprotection.

TMSI-mediated deprotection of compound 10 in MeCN (a more environmentally friendly solvent than DCM) proved optimal, with 5.0 equivalents of TMSI achieving complete conversion.

This method enabled crystallization of compound 11, allowing isolation of high-purity material through a single separation step. Quenching with 5.0 equivalents of water followed by solvent exchange to 2-methyltetrahydrofuran afforded crystalline compound 11. At 147 kg scale, the isolated yield reached 89% with purity exceeding 98%.

Step 5: Cyclization

Compound 11 contains two unprotected amines, leading to regioselectivity challenges. Under initial conditions (T3P 6.0 eq, NMI 1.4 eq, DMF 16 mL/g, –5 °C, 40 h), 6–8% of byproduct 16 was formed.

High-throughput condition screening identified EDCI/HOPO as an optimal coupling system. Using EDCI (3.4 eq), HOPO (2.0 eq), MgCl₂ (3.0 eq), DMAc (10 mL/g), and –5 °C for 25 h achieved 97.4% purity with 75% yield.

Hzymes Supplies Non-Proteinogenic Amino Acid Raw Materials

Leveraging its enzymatic synthesis platform for intermediates, protein-directed evolution capabilities, and the Hevo AI+ assisted design platform, Hzymes has developed enzymatic production processes for key MK-0616 raw materials: L-5-fluoro-tryptophan and trans-3-hydroxy-L-proline.

Both products achieve >99% purity with single impurity <0.5%, fully meeting the large-scale raw material requirements for MK-0616 production.

References

[1] Peacock H, Suga H. Discovery of De Novo macrocyclic peptides by messenger RNA display. Trends in Pharmacological Sciences, 2021, 42(5), 385-397

[2] Kamalinia G, Grindel B J, Takahashi T T, Millward S W. Directing evolution of novel ligands by mRNA display. Chem. Soc. Rev. 50 9055–103

[3] Alleyne C, et al. Series of Novel and Highly Potent Cyclic Peptide PCSK9 Inhibitors Derived from an mRNA Display Screen and Optimized via Structure-Based Design. J. Med. Chem. 2020, 63 (22), 13796–13824

[4] Kai-Jiong Xiao，Yong-gang Chen. Process Development toward a Key Fragment of the PCSK9 Inhibitor Enlicitide Decanoate. Organic Process Research & Development Article ASAP