Lessons learned from the clinical development of oral peptides

Abstract

The oral delivery of peptides and proteins has been hampered by an array of obstacles. However, several promising novel oral delivery systems have been developed. This paper reviews the most advanced oral formulation technologies, and highlights key lessons and implications from studies undertaken to date with these oral formulations. Special interest is given to oral salmon calcitonin (CT), glucagon-like peptide-1 (GLP-1), insulin, PYY-(3-36), recombinant human parathyroid hormone (rhPTH(1-31)-NH₂) and PTH(1-34), by different technologies. The issues addressed include (i) interaction with water, (ii) interaction with food, (iii) diurnal variation, (iv) inter- and intra-subject variability, (v) correlation between efficacy and exposure and (vi) key deliverables of different technologies. These key lessons may aid research in the development of other oral formulations.

Introduction

The primary advantages of oral drug delivery compared with injection or nasal spray administration are greater acceptability and convenience for patients. However, oral delivery of peptide-based drugs is limited by a series of obstacles 1–3. Proteolytic cleavage in the digestive tract, and thus limited intestinal uptake of the intact molecule is just the start 1,2. Additionally, if the intact peptides are absorbed through the gastrointestinal tract, they have to cross the epithelial layer of cells lining the small intestine and then enter the hepatic portal vein 4,5. Importantly, the integrity and function of the gastrointestinal tract mucosa (such as prevention of permanent damage to the enterocytes or to the tight junctions) must be maintained while ensuring peptide uptake. In addition to challenges associated with the biology of the GI tract itself, absorption of peptides is hindered by their physical characteristics, including their polarity and size of the dosage form, and their susceptibility to aggregation or degradation by the local pH in the tract. Without a suitable ‘technical intervention’, such molecules cannot successfully be delivered via the oral route, as will be highlighted in the following ‘roads most travelled’ to obtain a functional oral formulation 6.

These difficulties are highlighted by the fact that, with the exception of desmopressin 7, no orally formulated peptide has been approved by the US Food and Drug Administration (FDA). Since an oral form of any therapeutic protein could improve patient compliance due to a more convenient delivery route, a number of investigations have been performed to develop an effective oral delivery system, allowing intestinal absorption of different peptides, such as calcitonin (CT), glucagon-like peptide-1 (GLP-1) and insulin 8–21. To overcome the barriers mentioned above, various strategies have been adopted to increase bioavailability of proteins and peptides 22,23, as will be discussed.

Numerous attempts at oral formulations of a variety of peptides and proteins have been made, some of which have reached late stage clinical testing. This list includes, but is not limited to, salmon CT, GLP-2, GLP-1, PYY-(3-36), insulin, parathyroid hormone (PTH) 24 and octreotide, all of which are at different stages of clinical testing. Several more, such as growth hormone, erythropoietin, granulocyte colony stimulating factor, interferon alpha, interleukins IL-4 and IL-12, luteinizing hormone releasing hormone, somatostatin, thrombin inhibitors, thyrotropin releasing hormone analogues, vasopressin and vaccines have undergone preclinical testing using numerous types of formulations 23,25,26. While the preclinical work is out of the scope for this manuscript, the list underscores the interest in oral formulation pharmacology. The list of peptides in clinical testing is expected to grow with recent key discoveries in the development of oral formulations of macromolecules.

Barriers to drug absorption – a brief introduction

There are three major barriers to the adequate uptake of orally delivered drugs: (i) physiological barriers, (ii) physicochemical barriers and (iii) enzymes 4,23,27, each of which are briefly discussed below.

1. he luminal side of the GI tract is covered with epithelial cells that are connected via tight junctions, which limit passage of molecules between cells. Accordingly, absorption of most proteins and peptides occurs only following digestion by enzymes. The absorption of small fragments or amino acids occurs by passive diffusion through the paracellular pathway or by active transport through a transcellular pathway in the intestinal epithelium 4, depending upon their physicochemical properties.

2. The physical barriers to oral delivery of a molecule are parameters such as size, charge and hydrophobicity, stability as well as the pH range for solubility, which can be far from the acidic environment of gastric fluid.

3. The biochemical barriers include the secreted and non-secreted proteolytic enzymes, which are secreted in large amounts in the GI tract, to facilitate digestion and absorption of nutrients and exclusion of unwanted material. The most important are pepsin, lysozyme, chymotrypsin, trypsin, elastase, carboxypeptidase, aminopeptidase A and N, diaminopeptidase I and endopeptidase S.

Cleavage of peptide bonds starts in the stomach in the presence of pepsins, a class of aspartate proteases that are secreted by the gastric cells, and continues throughout the lumen of the small intestine by pancreatic proteases as well as proteases from the brush border membranes and the cytoplasm and lysosomes of the enterocytes 4,5. The process of proteolysis is an important and efficient mechanism that enables the digestion of proteins in food and also plays a role in the inactivation of some organisms. Approximately 94–98% of the total protein ingested is cleaved into small peptide fragments or amino acids, digested and absorbed. The most relevant pancreatic proteases involved in the digestion process are the serine endopeptidases trypsin, α-chymotrypsin and elastase, as well as the exopeptidases carboxypeptidase A and B. The rate and degree of degradation are dependent on the chemical structure, the type and concentration of the enzymes involved in the degradation reaction and the ambient pH 28.

In this regard, the frequently studied peptides, such as PTH, salmon CT, PYY-(3-36), GLP-1 and GLP-2 are of particular interest, as N-terminal truncation of these molecules may lead to the generation of an antagonist as opposed to the intended agonistic molecule 29. This emphasizes that it is important to determine whether the molecules present in the blood following oral dosing are intact.

Strategies for peptide absorption

Five different approaches have been tried to enhance systemic exposure of peptides following oral delivery by one of the following mechanisms: (i) permeation enhancers, (iia) enzyme inhibitors and (iib) modulation of protein to prevent degradation, (iii), particulate systems, (iv) multifunctional polymers, and finally (v) ligand-specific binding and uptake, which are summarized in Table 1, and described in detail elsewhere 6.

Table 1.

Approaches being tested for oral delivery of peptides

Approach	Modification	Intention	Target molecule	Level	References
i) Permeation enhancer	Glycosylation, PEGylation, lipidation	Increase uptake across epithelium	Salmon CT, insulin, PYY-(3-36), GLP-1, PTH, octreotide	Clinical	14,83–85
	Carnitine esters
iia) Enzyme inhibitors	Soybean trypsin inhibitor	Prevent degradation in the GI tract	Salmon CT, PTH	Clinical	86–93
	Deoxycholic acid
	Organic acids (e.g. citric acid)
iib) Modulate peptide towards a stable form		Reduce degradation in GI tract	Insulin	Clinical	72
iii) Particulate systems	Polymeric micro and nanoparticles, liposomes, micro-emulsions or polymeric micelles	Encapsulation to protect against degradation	Insulin	Pre-clinical	23,94,95
Multifunctional polymers	Poly(alkylcyanoacrylate) chitosan	Adhesion to mucous tissue, and entry via tight junction	Insulin	Pre-clinical	96,97
Ligand mediated uptake	Particles linked to:	Targeted delivery via food intake relevant route.	Insulin	Pre-clinical	23,98,99
	vitamin B12, biotin, folate and lectin

Introducing the absorption enhancers 5-CNAC, SNAC, 4-CNAB

The Eligen technology employs low molecular weight compounds (termed drug delivery agents or carriers) which interact weakly and non-covalently with the target proteins, increasing their lipophilicity and thereby their ability to cross the gastrointestinal epithelium 30. The delivery agents consist of N-acylated alpha-amino acids 31. These carriers, 8-(N-2-hydroxy-5-chloro-benzoyl)-amino-caprylate (5-CNAC), N-[8-(2-hydroxybenzoyl)amino] caprylate (SNAC) and monosodium N-(4-chlorosalicyloyl)-4-aminobutyrate (4-CNAB), interact with the target proteins, e.g. calcitonin, GLP-1 or insulin, without changing their biological properties. The combination of carrier and protein/peptide forms an insoluble entity at low pH values, thereby reducing the compound's susceptibility to degradation by peptidases in the upper digestive tract 30. In the upper part of the small intestine at higher pH, the complex dissolves, which facilitates intestinal uptake over the non-polar biological membrane 30. The proteins are absorbed systemically via the passive transcellular pathway, in which peptides pass through epithelial cells without chemical modification and without compromising the integrity of the intestinal epithelium 4,32,33. After crossing the membrane, the delivery agent disassociates from the peptide which re-establishes its natural form and retains its biologically active state 30.

Clinical experience with 5-CNAC

Particularly, the combination of 5-CNAC and salmon CT has been intensively studied in both healthy and diseased populations in relation to shedding light on (1) the optimal dose of oral salmon CT, (2) the bioavailability of orally vs. nasally administered calcitonin, (3) a comparison of the exposure and efficacy at day 1 and at day 14 of treatment, (4) the interaction with food intake 34, (5) the interaction with water intake 15,18,35–38 and (6) prolonged efficacy studies in osteoporotic patients.

1. The studies showed that dosing with oral salmon CT at 0.8 mg was superior to 0.6 mg in terms of exposure, both at the first day of dosing and after 14 days, and that dosing manifested itself in significant reductions in the biochemical markers of bone and cartilage turnover applied as efficacy output (Table 2 and 37). Furthermore, a direct relationship between exposure and efficacy was observed, although the efficacy profile was protracted, a phenomenon which recently has been explained to be linked directly to the interaction with the calcitonin receptor in the target cells 35–37,39–42.

2. When comparing both exposure and efficacy of the orally formulated salmon CT with the nasal formulation (Miacalcin), it was apparent that the oral formulation led to both a high systemic exposure and a higher efficacy as assessed by the biomarker of bone turnover C-terminal telopeptide of type I collagen (CTX-I), and hence continued treatment is expected to result in an overall improved efficacy of the orally formulated drug 35.

3. One study investigated the effect of repeat dosing on both exposure and efficacy (see Figure 1). The study firmly demonstrated that, based on CTX-I, increased exposure leads to increased efficacy at both time points. However, it also showed that the correlation between exposure and efficacy at day 1 and at day 14 is poor 38, due to reasons which are not fully understood.

4. When studying the effect of food intake on exposure of orally delivered salmon CT, a significant interaction with food intake was observed. This study demonstrated that ingestion of the tablet within 4 h after a meal reduced plasma exposure by up to 74% when compared with ingestion 10 min before the meal, and again exposure was linked directly to efficacy assessed by CTX-I 34. These data corroborate the need for careful consideration of when to dose in relation to oral peptide therapies, as seen for other types of oral therapy 43–46.

5. One study compared the effects of 50 ml or 200 ml of water taken with a single tablet of calcitonin, and it demonstrated a significant sensitivity to water intake, resulting in a two- to three-fold higher exposure at 50 ml of water than at 200 ml 34. These data are the first to demonstrate that water intake has an important effect on oral peptide uptake with this technology, with smaller quantities of liquid improving bioavailability by as much as 400%. This finding may be important for the development of other formulations.

6. Long term clinical studies using 5-CNAC as the carrier.

Table 2.

Summary of findings from the PK−PD studies of 5-CNAC in combination with salmon CT

Carrier and protein	Parameter	Conclusion	Reference
5-CNAC and salmon CT	Water	50 ml is superior to 200 ml	34–36,39,40
5-CNAC and salmon CT	Food	Dosing before eating is superior to after eating
5-CNAC and salmon CT	Dose	0.8 mg is better than 0.6 mg
5-CNAC and salmon CT	Repeat dose	No statistically significant correlation between exposure and efficacy at day 1 and day 14.
5-CNAC and salmon CT	Administration route	Oral is better than nasal

Figure 1.

Linear relationship between plasma sCT concentrations and reduction in CTX-I. Reproduced from 6 with permission

Two studies using 5-CNAC as a carrier for an extended period of time have been conducted. One study was in post-menopausal women who were treated with different doses of oral salmon CT for 3 months. In these women, clear and dose-dependent reductions in the efficacy parameters CTX-I and C-terminal telopeptide of type II collagen (CTX-II) were observed both acutely and chronically (Table 3) 15,18. Furthermore, this study also demonstrated that the reductions in CTX-I at day 1, month 1 and month 3 are related, which indicates low intra-subject variability, i.e. that a ‘good’ responder is always a good responder. Intriguingly, no major reduction in bone formation markers was observed, indicating an ability to reduce bone resorption without reducing formation 47. Interestingly, a post hoc analysis of this study showed that oral salmon CT possesses the ability to reduce the cartilage degradation marker CTX-II, especially in those with high levels of cartilage degradation 18. Along a similar line of thinking, Manicourt et al. found that oral calcitonin treatment led to improvements in both biomarkers of cartilage turnover and a functional index in osteoarthritis (OA) patients, further supporting the efficacy of salmon CT in this oral formulation for this indication 19. Overall, the oral formulation was well tolerated, with mild to moderate gastrointestinal and skin manifestations apparent mainly in the high dose groups. This formulation has been in phase III studies for (osteoporosis) and OA, but no data have been published yet.

Table 3.

Overview of the absorption enhancers and the peptides used in combination and the indication for which they are being tested along with the adequate references

Carrier	Molecule	Indication(s)	References
5-CNAC	Salmon CT, PTH	Osteoarthritis, osteoporosis	15,18,25,35–37,40,100
SNAC	Insulin, GLP-1, PYY-(3-36)	Diabetes	48,49
4-CNAB	Insulin	Diabetes	50

5-CNAC has also been tested in combination with human PTH(1-34) 25. This was a single dose study using different combinations of 5-CNAC and PTH(1-34) and a comparison with teriparatide. All doses of PTH(1-34) were rapidly absorbed, and the PK profile was dose-dependent. Interestingly, PTH(1-34) disappeared from blood faster after oral than after s.c. administration, and exposure superior to that with teriparatide could be obtained. Furthermore, a dose-dependent increase in urine cAMP was observed indicating biological efficacy. Importantly, the high exposure was not associated with clinically significant reports of hypercalcaemia 25, and in general the AEs were related to PTH(1-34) rather than 5-CNAC.

Clinical experience with SNAC

Two studies have been conducted using the carrier SNAC in combination with either GLP-1 or PYY-(3-36) 48,49. The initial study was a single ascending dose type of study, investigating both GLP-1 and PYY-(3-36) in healthy adults. For oral delivery of GLP-1 t_max was reached quickly (15 min) after which the molecule was cleared from the circulation within 1 h, but the bioavailability was less than 4%. The dose−response was not linear. Furthermore, the PK profile was different from i.v. dosing, where a t_max plateau was observed, and the variation was markedly larger. On the other hand, exposure of PYY-(3-36) was completely dose-dependent and yielded substantial exposure, with a t_max at 15 min and a linear elimination from the circulation resulting in complete clearance even of the high dose after 2.5 h. The exposure levels reached manifested themselves in efficacy at the level of insulin secretion for GLP-1 and a reduction in ghrelin secretion by PYY-(3-36) 48. In the second study, a combination of orally formulated PYY-(3-36) and GLP-1 was added, and again short-lived but significant exposure was observed for both. Additionally, increased efficacy was observed at the level of satiety, as well as post-prandial glucose regulation, when compared with either peptide alone 49.

Clinical experience with 4-CNAB

A single study has been conducted with the carrier 4-CNAB, and in this case it was a combination of 400 mg of the carrier combined with 300IU of insulin, which was compared with a subcutaneous (s.c.) dose of 15 units of regular human insulin in a 1 day glucose clamp study in type 2 diabetics 50. This study highlighted a markedly different PK profile, namely that the oral version had a t_max at t = 20 min, and a very sharp exposure peak with complete clearance within less than 2 h. By contrast, s.c. delivery led to a prolonged exposure profile, and the plasma concentration did not begin to fall until 5 h after dosing. However, this could possibly be due to the selection of insulin analogue. Importantly, the efficacy parameter (glucose infusion rate) clearly matched the exposure for both dosing regimens, illustrating a direct PK–PD relationship. The variance in exposure was larger with the oral formulation than with the s.c. formulation 50.

Safety of these carriers

The safety profile for the carriers described in the above sections, in general, is rather good. While AEs were reported, most of them appeared to be linked to the active molecule, as opposed to the carrier and they rarely led to dropouts. Furthermore, the AEs were related to dose of the active molecule, i.e. salmon CT, and corresponded to those reported for other forms of the molecule, clearly indicating a good safety of the carrier 15.

Introducing the Enteris technologies

Enteris oral delivery technology was first developed by exploring the parameters required for the oral delivery of salmon CT 51. Several peptides, differing in size, charge and stability, such as salmon CT and a recombinant human PTH analogue [rhPTH(1-31)-NH₂], have been tested in phase 1 and phase 2 studies, and a pivotal phase 3 study, using this technology 52–54.

Mechanism of the oral delivery technology

A tablet core containing the peptide, an organic acid, a permeation enhancer, and other excipients is covered with an acid stable enteric coat that allows it to remain intact in the stomach (Figure 2), and prevent its degradation by gastric acid and pepsin. The enteric coat also makes absorption of the peptide less susceptible to variability due to administration with meals or large volumes of liquid. When the intact tablet exits the stomach into the duodenum, the local pH increases to 6 and the enteric coat begins to dissolve. For this technology to work optimally, the peptide, as well as the tablet excipients, need to be released simultaneously in a small, localized area in the small intestine. This bolus release is facilitated by a water soluble subcoat underneath the enteric coat. Subcoat performance is a critical pharmaceutics design feature and acts to prevent the acid core from leaching into the enteric coat and interfering with complete dissolution of the pH sensitive enteric coat. One of the main excipients released from the tablet is the organic acid, generally citric acid, which is present in the form of maltodextrin-coated beads. The maltodextrin coating prevents the acid and the peptide from coming into contact with each other until the point of release in the intestine when the water soluble coating dissolves, thus avoiding the potential problem of peptide degradation under acidic conditions during storage of the tablets. In the localized area where the tablet contents are released, the organic acid creates an acidic environment. Since all of the proteolytic enzymes in the intestine, whether from the pancreas or the brush border membranes, have a neutral to alkaline pH optimum, their activity is inhibited, which allows the peptide to remain intact. The citric acid is also helpful in increasing peptide absorption across the lumen of the intestine since it chelates intracellular calcium within the enterocytes, and also because the acidic environment increases intestinal flux. The second main excipient is lauroyl-L-carnitine (LLC), a 12 carbon chain ester of carnitine, which acts as a permeation enhancer 55. The simultaneous release of the acylcarnitine enhances peptide absorption by increasing passive paracellular transport. LLC acts by transiently loosening tight junctional complexes, resulting in a net increase in the radius of tight junctions between adjacent gastrointestinal epithelial cells. Since LLC is also a surfactant, it increases the solubility of peptides, and enhances the ability of peptides to traverse the mucus layer that lines the intestine. For oral delivery of peptides by paracellular transport, several parameters predict the extent to which the technology will enhance oral bioavailability. The size of the peptide is clearly important, but so is the charge, hydrophobicity, stability, propensity to aggregate, etc. 56,57.

Figure 2.

Barriers to oral delivery of peptides

1. Acidic milieu in the stomach induces need for protection of peptide, i.e. protection against acid-mediated degradation is essential through a coating or a chemical modification
2. Removal of acid protective coat needed to ensure exposure and allow uptake
3. Uptake across the tight junctions in the epithelium, made difficult by the presence of tight junctions sealing off the lumen from the surrounding regions to ensure that only ‘the right material’ gets taken up. Requires permeability enhancers

Clinical studies with the Enteris technology

This formulation has been successfully tested in phase I, II and III studies, and hence a substantial amount of clinical data are available on the performance of the technology.

CR845

The analgesic peptide, CR845, which is a κ opioid agonist developed by Cara Therapeutics, was studied in a rising dose phase 1 study. The data demonstrated linearity with dose, and resulted in an overall absolute bioavailability of 16%. However, while i.v. dosing results from phase II have been reported 58, no news on the ongoing further oral development has been published 59.

Oral PTH

As part of a 6 month phase 2 study with oral rhPTH(1-31)-NH_2, PK measurements were made at the beginning and the end of the study 53. As shown in Figure 3, there was a consistent exposure of the intact PTH analogue at the beginning and at the end of the study at a daily dose of 5 mg. The mean C_max values achieved with this oral PTH analogue were significantly higher (295 pg ml⁻¹ and 207 pg ml⁻¹ at week 0 and week 24, respectively) than those of the subcutaneously administered Forsteo® (120 pg ml⁻¹ at both week 0 and week 24) that was used as the active comparator. However, the AUC values of the oral PTH analogue and teriparatide were comparable (Figure 3). As seen for other oral formulations the variation in exposure was larger in the oral delivery group than in the s.c. group 53. Importantly, the oral PTH analogue was delivered with a pulsatile PK profile, which resulted in a 2.2% increase in lumbar spine bone mineral density (LS-BMD) after 6 months of treatment. Interestingly, no increase in bone resorption markers was observed, while a moderate and sustained increase in bone formation parameters were reported. Compared with teriparatide, the effect was numerically lower, which appeared to be caused primarily by a slower onset of action, although future studies are needed to clarify this 52.

Figure 3.

Pharmacokinetic parameters from a phase 2 proof of concept study with oral tablet formulation of PTH(1-31)NH₂. A 6 month phase 2 study was carried out to compare the bone anabolic response of oral PTH(1-31)NH₂ with that of s.c. PTH1-34 (Forsteo®) by measurement of lumbar spine bone mineral density (BMD). Blood samples were collected after the first dose (week 0) and at the end of the study (week 24). Samples were analysed with a sandwich ELISA assay that specifically measures only the intact PTH molecules. Figure reprinted from 49 with permission. , PTH(1-31)NH₂ (week 0); , PTH(1-31)NH₂ (week 24); , Forsteo (week 0); , Forsteo (week 24). *time relative to t_max for oral PTH

Oral salmon CT

An oral formulation of salmon CT was evaluated by Tarsa Therapeutics in two late stage clinical studies. A 1 year phase 2 study for prevention of post-menopausal osteoporosis in women with low bone mass demonstrated that oral CT was superior to placebo with respect to change in LS-BMD at week 28 and week 54 and resulted in prompt suppression of bone resorption 60. In this study the patients in the active arm were divided into two cohorts, one dosed daily at dinner and the other dosed at bedtime. The increase in LS-BMD in both cohorts was equivalent, demonstrating that the efficacy was not adversely affected by food intake. A pivotal phase 3 study of oral salmon CT post-menopausal osteoporotic women was also conducted, and it was demonstrated that subjects treated with oral salmon CT tablets for 48 weeks had an improvement in LS-BMD which was superior to that of subjects receiving the comparator nasal spray calcitonin (Figure4). Interestingly, in addition to the increase in efficacy compared with nasal, the oral formulation also resulted in a five-fold decrease in patients with circulating anti-salmon CT antibodies 54.

Figure 4.

One year double-blind placebo controlled phase 3 study conducted by Tarsa Therapeutics comparing orally delivered salmon calcitonin with a nasal spray calcitonin product (Miacalcin®). Change in lumbar spine bone mineral density (BMD) from baseline to week 48. Figure reprinted from 50 with permission. , rsCT tablet; ,nasal spray; , placebo

Safety of the Enteris formulation

These formulations are the ones with the most substantial safety database, and the data from the phase II and phase III studies reported gastrointestinal AEs as being prominent, and in these studies these effects appeared somewhat related to the carrier, as they were present in the treatment groups receiving the carrier and not the ones without 52,54. However, the reason for this is still somewhat elusive, and a puzzling phenomenon was observed with the orally formulated PTH analogue, namely that the combination of the analogue and the formulation led to higher numbers of gastrointestinal events than the formulation alone. On the other hand these issues were mostly mild or moderate, were often resolved, and only in very few cases led to drop outs 52. 

Introducing the Oramed technologies

As described in the introduction, the major challenges for oral delivery of peptide and protein (p/p) drugs is their susceptibility to acid hydrolysis in the stomach, proteolytic degradation in the intestine, limited permeability to cross membranes, and their tendency for complexation and adsorption in the gut 61. Of these barriers the proteolytic degradation by luminal pepsins, pancreatic enzymes and brush border enzymes has been studied extensively as a venue to improve enteral absorption, as described in the introduction. Protease inhibitors afford protection from degradation. For example, a formulation containing insulin and duck ovomucoid offers 100% protection against trypsin- or α-chymotrypsin-mediated insulin degradation. Polymer inhibitor conjugates, such as carboxymethylcellulose Bowman Birk inhibitor and carboxymethylcellulose elastinal (CMC–Ela), have offered in vitro protection against trypsin, chymotrypsin and elastase 62.

In the context of drug delivery, it has been observed that most of the methods that reduce the proteolytic degradation of a peptide, increase its fraction absorbed and consequently its bioavailability. As mentioned above, there are several methods being investigated as to how best to protect p/p from degradation in the intestine. The ideal protease inhibitor will consist of a substrate specific inhibitor to match to the specific bonds in the p/p. However because of the multitude of degrading enzymes and the susceptibility of many different p/p to these degrading enzymes, ‘broad acting’ protease inhibitors (PI) are more practical. Moreover, some proteins are predominantly susceptible to a particular protease inhibitor. For example insulin has been found to be primarily susceptible to α-chymotrypsin which is responsible for the initial cleavage and unfolding of insulin globular structure, exposing the molecule to subsequent proteolytic degradation 63. Thus by using an α-chymotrypsin inhibitor in the formulation, a significant fraction of the molecule can be spared degradation. The common PIs considered in drug development are trypsin or α-chymotrypsin inhibitors, such as pancreatic inhibitor 64, soybean trypsin inhibitor 64, FK-448 65, camostat mesylate 66, aprotinin 67 and chicken and duck ovomucoids 68. Interestingly, some of the more potent PIs that inhibit serine endopeptidases trypsin and α-chymotrypsin are found in plants which are consumed in great quantities worldwide such as wheat grain, soy beans, quinoa, potatoes and eggs and should allay to some degree, safety concerns 69.

Oramed employ PIs, in combination with encapsulation in a polymer with pH-dependent solubility that utilizes pH gradients in the intestine for site-specific dissolution, as part of their core technology of drug delivery.

Clinical studies

Two studies using this approach have been carried out. In the first study healthy volunteers were exposed to five different oral formulations of insulin, which was used to select formulations leading to the most robust reduction of blood glucose and C-peptide, both of which are standard efficacy parameters in insulin studies 70. The second study was performed with eight type I diabetes patients using continuous blood glucose measurements (Figure 5). In this study, clinically relevant lowering of blood glucose was observed in the patients exposed to oral insulin therapy, clearly indicating the usefulness of this formulation 71. However, the number of patients was too low to provide data about variation in exposure as well as adverse events, two issues observed with the oral formulations that have gone through substantial clinical testing.

Figure 5.

Continuous blood glucose measurements (full traces) and standard errors (dotted traces) before and after treatment with ORMD-0801. Reproduced from 66. , pretreatment; , ORMD-0801

Oral insulin using pegylation by Biocon

IN-105 is a human insulin analogue, in which a single short chain amphiphilic oligomer is covalently linked by a non-hydrolysable amide bond to the free amino acid group on the Lys-β29 residue of recombinant human insulin. The modification has been shown to improve the solubility over human insulin, as well as the stability against enzymatic degradation in the digestive tract, both parameters which help in the systemic absorption of the intact peptide. In vitro studies showed increased stability, with unaltered insulin receptor activation profile 72. A dose−response of this analogue was studied in type 2 diabetes subjects poorly controlled on metformin, with the primary objective of examining the effect of sequential single ascending doses of IN-105 on the plasma glucose concentration under fed conditions. In this study, the PK profile showed a clear dose–response, and a corresponding inverse effect on blood glucose and C-peptide following a meal was observed, clearly demonstrating efficacy of the drug. Although trends towards hypoglycaemia were observed, these were not clinically significant 72. For this formulation, the number of study subjects was too low to draw conclusions on safety. However, as seen with the other formulations the inter-subject variability was substantial.

Gastrointestinal permeation enhancement technology by Merrion pharmaceuticals

This platform is intended to deliver macromolecules and polypeptides, including insulin. The system is based on promoting drug absorption through the use of matrices consisting of medium chain fatty acids (C8 and C10 medium chain fatty acids) and formulated as solid dosage forms 73,74. The matrices enjoy food additive, or Generally Recognized as Safe (GRAS) status, and are normal dietary components with long records of safe use. The target molecule and the other ingredients are prepared as a physical mix and are formulated into a tablet designed to be released in the duodenum 73,74. Phase I studies of acyline, which is a decapeptide, have been performed, and shown to provide robust exposure as well as clinically relevant reductions in serum testosterone, which was used as efficacy end point. However, high inter-individual variation was observed 74. Recently studies of insulin with this technology are ongoing. However, data from these have not been released, although a press release stated that it was successful 75.

Transient permeability enhancers (TPE) by chiasma pharma

This system facilitates intestinal absorbance of drug molecules with limited intestinal bioavailability. It is compatible with peptides, small proteins (up to 20 kDa in size), saccharides and poorly absorbed small molecules. The TPE protects the drug molecule from inactivation by the hostile gastrointestinal environment and at the same time acts on the gastrointestinal wall to induce permeation of its cargo drug molecules. These two attributes ensure that when delivered in the TPE formulation, the drug reaches the bloodstream effectively in its native active form. All excipients used in manufacturing of the drug product are of pharmaceutical/food grade and are on the FDA GRAS list 76. Using this formulation of octreotide for treatment of acromegaly, four single dose studies were conducted in 75 healthy volunteers 77 and a phase III is presently ongoing 78.

Oral or nasal delivery

An important consideration in the development of oral formulations of peptides is whether the nasal alternative is more attractive, considering that nasal delivery of peptides already has succeeded for calcitonin and desmopressin 35,79. Importantly, there are still significant limitations to nasal delivery in terms of volume delivered, which is quite small. Importantly, uptake over the nasal mucosa is also quite limited and the bioavailability of larger peptides, such as CT, is very low 35. Additionally, two other aspects limit nasal delivery, namely irritation of the nasal mucosa upon chronic treatment and large variations in uptake 79–81. Thus, in summary while nasal formulations are somewhat useful there is still room for oral delivery; however, during the development of oral delivery of a peptide it should always focus on improving bioavailability and efficacy compared with a nasal delivery system.

Conclusions

In conclusion, a range of challenges need to be overcome when developing oral formulations of peptides. Most importantly, technological inventions as well as co-administration of solvents (water), while exploiting the natural physiology of the disease (such as by dosing in the evening in patients with osteoporosis when bone resorption is at its peak and before food), may all aid in the identification of the optimal dosing regimen.

The technical and clinical development of calcitonin in a tablet has shown that an oral formulation of a peptide is at last achievable, and this current formulation produces increased levels of drug exposure compared with nasal delivery.

Ongoing phase III drug registration trials may prove it to be an effective, convenient option, with a high safety profile, for patients with osteoporosis and potentially osteoarthritis 38. However, a key point to be followed in the larger trials is whether increasing permeability across the gastrointestinal tract could be detrimental long term. At present studies have reported abdominal pain, but this was not related to chronic exposure and hence more data are needed 52,54. A key point is also the potential interaction with oral drugs altering the gastric/intestinal environment, such as food and fluid intake, proton-pump inhibitor therapy etc. However, there are limited or no data to shed light on these aspects yet.

Another important aspect is the costs associated with the development of these approaches, as large amounts of peptide and the different excipients are needed, and these aspects are likely to be expensive. Efforts towards reducing the costs of producing the peptides in sufficient amounts for oral delivery have led to the use of recombinant peptides, despite the knowledge that synthetically produced peptides are easier to handle from a regulatory point of view. Additionally, the list of excipients needed also contributes to increasing the overall costs of the orally formulated peptides. Hence, the costs of development and manufacturing are likely to be substantial, and thus the subsequent drug could end up being expensive, which may limit the interest although this remains to be seen.

Oral formulation of many peptides remains a challenge, and the potential of this approach is still somewhat unclear 82, but hopefully the lessons learned from the development and clinical trials described here may aid others in the development of oral formulations of therapeutic peptides.

Competing Interests

All authors have completed the Unified Competing Interest form at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare no support from any organization for the submitted work. MAK, KH, CC and BJR are employed by Nordic Bioscience, NM and WS are employed by Enteris Biopharma and EA has been employed by Oramed within the previous 3 years. MK, BJR, and CC own stock in Nordic Bioscience. NM and WS own stock in Enteris Biopharma and EA owns stock in Oramed. All authors have patents related to oral formulations of peptides.