The success of the GLP-1 agonists and related follow-on drugs has renewed interest in peptide-based drugs. Industry experts share their perspectives on the latest in peptide synthesis.  

By Patricia Van Arnum, Editorial Director, DCAT, pvanarnum@dcat.org

Inside the technology 
Solid-phase peptide synthesis (SPPS) is the dominant method in peptide synthesis, but other approaches, such as liquid-phase peptide synthesis (LPPS) and hybrid approaches are being used. Which methods are gaining the technology edge? Industry experts share how these approaches are being deployed and their benefits and limitations.   

“Solid phase peptide synthesis (SPPS) is a manufacturing technology whereby a peptide is grown on a solid support, typically this is a resin-type support, so it can be polystyrene or PEG [polyethylene glycol] or another polymer,” explains Alaric Desmarchelier, PhD, Business Development Manager, Peptides, Almac Sciences. “The peptide’s starting fragments or the starting amino acid are grafted onto that solid support. And then through an iterative process, we will add the amino acids one by one or fragment by fragment to build the peptide from one end to the other.”  

SPPS then consists of two cycles: a coupling reaction or loading of the amino acid on the solid-state support and then a cleavage step allowing for peptide chain elongation to the desired length, explains Mimoun Ayoub, PhD, Senior Vice President and Global Commercial Head of Peptides at CordenPharma. “So one cycle consists of two reactions or two steps. There is one coupling and then one cleavage for each of the cycles. For example, if you have a 40 amino-acid peptide, then you run 40 cycles, but 80 steps. You attach the amino acid, and then you deprotect that attached amino acid to prepare it for the next reaction, which is the next coupling. That’s why we have these two steps per cycle. The conversion for each reaction is nearly quantitative; it’s much higher than 99%. Now imagine for 40 amino-acid peptides, you will have to run 80 steps. Even if the conversion is nearly quantitative, you will still accumulate a very small fraction of impurities after each step. And then after 80 steps, of course, we are no longer talking about a small amount of impurities.” 

SPPS is considered the “workhorse” in peptide synthesis and is a well-established method, advanced by Robert Bruce Merrifield, PhD, winner of the 1984 Nobel Prize in Chemistry. It is favored for its linear process with high yields but requires large amounts of solvents as part of the washes and the purification process, and in scale-up, excess reagents and solvents quickly add up to the environmental and economic footprint. In addition, because the peptide is built on a solid-state support, it limits the amount of peptide strands per resin bead, which is a function of the amino-acid chain length. In general, SPPS can accommodate chain lengths of approximately 40–50 amino acids by linear synthesis, which as a point of reference, and would encompass glucagon-like peptide-1 (GLP-1) drugs although much larger peptides are feasible. For example, semaglutide, the active ingredient in Novo Nordisk’s GLP-1 agonist, Ozempic/Wegovy, is composed of 31 amino acids and tirzepatide, the active ingredient in Lilly’s Mounjaro/Zepbound, a dual-activating GIP (glucose-dependent insulinotropic polypeptide) and GLP-1 medication, is composed of 39 amino acids. 

At smaller scales, longer amino-acid chain length can be achieved through SPPS; at larger scale, long peptides benefit from LPPS or a hybrid approach. “At smaller scales, this technology [SPPS] has been applied to make very long peptides,” says Almac’s Desmarchelier. “In our R&D team, we make up to about 150 amino acids, small proteins, histones typically, by solid phase peptide synthesis, linearly or by a convergent approach. In larger scales, typically shorter peptides have been manufactured through this method. We at Almac have made up to 100 amino acids in solid phase synthesis by linear manufacturing, but typically the peptides are on the shorter end of the scale, so below 50 amino acids for most candidates that we see in development or on the market.” 

Scale, phase of development, desired chain length, and yield are all factors that go into the decision-making into what type of approach should be used. “At the end, you need to find a balance between the cost for the process development, the time you spend to run the manufacturing, but also the yield,” explains CordenPharma’s Ayoub. “If the objective is to generate material very quickly, to perform toxicology studies and generate data for IND [investigational new drug] fillings, then I think the straight SPPS in your process without too much process development, or at least what we call phase appropriate, is the best approach. But if the objective is really long-term supply, and especially if you have or expect large volumes for the market later on, then it’s most probably better to develop an appropriate process that fits the peptide sequence, but also that can address cost efficiently, the market demand or the volume of the API [active pharmaceutical ingredient].” 

One of the challenges in making longer peptides using straight SPPS is the loading of the resin. “The resin you start with has a certain loading, and you choose the loading based on the peptide length,” says CordenPharma’s Ayoub. “If the peptide is very long, of course, you cannot start with a highly loaded resin, which means you will be limited in terms of the solid-phase peptide synthesis batch size. So I would say for a 40-mer peptide, [a peptide with 40 amino acids], you would typically be starting with 0.6 millimole per gram, which means you cannot load beyond this. Otherwise, you will be facing some challenges with the reaction kinetics, but also the swelling of the resin, which impacts the overall impurity generation.” 

Liquid-phase peptide synthesis and hybrid approaches  
To address the issue of increasing chain length for longer chain peptides and larger volumes, a hybrid approach, where the full length peptide is assembled by connecting smaller lengths of peptide. The fragments can be made either using traditional SPPS followed by LPPS for coupling the fragments, or with LPPS to both make the fragments and couple the fragments.  

Liquid-phase peptide synthesis (LPPS) is where the process of growing the peptide chain is carried out as a homogeneous solution; typically in organic solvent, much like traditional small-molecule manufacture takes place. LPPS processes typically require significantly lower amounts of reagents and solvents and operate at higher concentration. All of which allows for the large-scale production of peptides to be much more efficient with significantly less waste and using equipment that is widely available.  

“The chief advantage of liquid phase is that it operates much as any small-molecule manufacturing does,” explains Matt Bio, PhD, Chief Scientific Officer, Cambrex. “The peptide is  protected at the amino acid end, at the nitrogen. You deprotect that, activate an acid, then couple and repeat that process over and over again. In between, you use aqueous washes to remove the impurities from the organic phase while your polymer or your peptide is growing in the organic phase. The challenge is that you have to keep that growing peptide in solution and effectively remove all the impurities that come from the activation of the acid from the deprotection of the amino acid. So, you need to develop a process, much like you would a small-molecule synthetic process, that both maintains the product in solution in the organic phase, but then removes the organic impurities that you don’t want into the aqueous phase after each wash.” 

A common strategy is to combine SPPS and LPPS in a hybrid approach. “But what the industry’s done is they’ve moved to a hybrid approach,” says Cambrex’s Bio. “Regardless of whether they’re using solid phase for the fragments or liquid phase, where they build up fragments of the molecule from tetramers up to 10-12 mers, and then combine them to make the longer chains. You can control quality at each one of the fragments and then link them together as part of the endgame to make the full-length molecule. And even if you were going to use solid phase, you would still want to take this approach. This is exactly how tirzepatide was made or is manufactured in the first instance.” 

Innovation in peptide synthesis 
Some innovation in peptide synthesis involves continuous manufacturing, particularly in the purification of the peptide, says Cambrex’s Bio, pointing to an innovative approach of using nanofiltration for purification during the hybrid assembly of molecules to be able to remove the small-molecule components and leave behind the desired peptide. For example, scientists recently reported on the use of a fragment convergent hybrid SPPS/LPPS approach that used flow chemistry for fragment condensation and nanofiltration for intermediate purification for application in large-scale peptide production (1).  

Continuous chromatography is another advanced technology in peptide purification, notes CordenPharma’s Ayoub. In general, continuous chromatography can lower solvent consumption compared to batch high-performance liquid chromatography (HPLC) and reduce solvent usage, resulting in higher yields at higher purity, such as through multicolumn countercurrent solvent gradient purification. 

Enzymatic ligation is another advance in peptide synthesis. “Historically, there have been a lot of chiral building blocks that can be made through enzymatic transformations, including natural amino acids,” says Almac’s Desmarchelier. “We’re looking more and more into enzymes for peptide-specific transformations, so ligation technologies, for example, to stitch together smaller fragments of peptides. We do that very successfully with oligonucleotides at Almac, but we’re looking at applying that to peptide manufacturing as well, or use enzymes to cyclize peptides, either head to tail or to form disulfide bridges, for example. Those are transformations that usually need a lot of control and are traditionally done in very large reactors in dilute conditions to avoid oligomerization. Enzymatic technologies would be very useful there to make the processes much more efficient. And part of that is reducing their footprints in a factory and making them in much smaller equipment. So, I think using biological tools for transformations rather than the whole fermentation synthesis, but for specific steps is quite an interesting field that we’re heavily involved in now.” 

Process analytical technology (PAT) and automation are also being used. “Nowadays we try to reduce the raw material amounts of the solvent through process analytical technologies or continuous processes at some of the steps in the peptide unit operations, but also the online monitoring of the reaction,” says CordenPharma’s Ayoub. “Without having to pull a sample each time and test the sample in the lab, there is an online monitoring of the coupling reaction, but also the cleavage, for completion of those reactions. 

Digital tools are also being applied to optimize the synthesis of peptides. “In terms of predictive modeling or digital tools used, people are using algorithms based on the data that we feed into the system to predict some potential challenges such as racemization, aggregation, or gel formation of the peptide sequence, difficult couplings, or formation of aspartamide impurities, for example,” says CordenPharma’s Ayoub. 

Almac’s Desmarchelier also points to the value of predicting the many reactions that are performed on a resin in SPPS. “We have a tool, for example, that is not [yet] an AI-driven tool, that draws from our historical know-how in the company, he says. “It will look at the peptides that we’re targeting, and it will look at regions in the peptide that might pose challenges based on what challenges we have encountered in the past. And it will suggest a first-time manufacturing approach to overcome those challenges. Obviously, we also have a lot of chemists with a wealth of know-how in the company. And so they will then have a critical view on those predictions and also apply their own knowledge to help guide the manufacturing. I know that there are also a lot of AI-driven tools that are being developed, both in academia and in the private sector, on how peptides behave, on predicting how a peptide will behave in terms of physical chemical properties, in terms of aggregation potential. “ 

Reference 

1. PJ Jansen et al., “A Convergent Hybrid Gram-Scale Synthesis of Tirzepatide: Tangential Flow Filtration Assisted Native Chemical Ligation-Desulfurization Approach,” Angew Chem Int Ed Engl. 2026 Feb 2;65(6), doi: 10.1002/anie.202520060.