Medicinal Chemistry

13 downloads 0 Views 3MB Size Report
up and gastrointestinal effects (steatorrhoea and diarrhoea); the MTP inhibitor lomitapide has been recently approved for the treatment of familial homozygous.



For reprint orders, please contact [email protected]

Medicinal Chemistry

New perspectives on the development of antiobesity drugs

After many years of research, obesity is still a disease with an unmet medical need. Very few compounds have been approved, acting mainly on neuromediators; researches, in recent years, pointed toward compounds potentially safer than first-generation antiobesity drugs, able to interact with one or more (multitarget therapy) receptors for substances produced by the gut, adipose tissue and other targets outside CNS. Other holistic approaches, such as those involving gut microbiota and plant extracts, appeared recently in the literature, and undoubtedly will contribute to the discovery of a valuable therapy for this disease. This review deals with the positive results and the pitfalls obtained following these approaches, with a view on their clinical trial studies.

It is nowadays recognized that obesity is a predisposing factor for life-threatening disorders; owing to the increase in obese people, obesity represents one big medical challenge. Obesity is one of the major causes of the metabolic syndrome, which is characterized by a cluster of metabolic abnormalities including glucose intolerance, insulin resistance, central obesity, dyslipidemia and hypertension [1] . Accordingly, the term diabesity has been recently introduced to characterize the common clinical association between Type 2 diabetes and obesity. However, the precise mechanisms linking the two conditions remain unclear [2] . Current antiobesity medications include mainly agents acting on neuromediators; among peripherally acting compounds, there are intestinal lipase inhibitors (orlistat and cetilistat, an orlistat analog approved in Japan for the treatment of obesity with complications) [3] . However, the side effects shown by the antiobesity approved compounds and the market withdrawn of several others antiobesity drugs acting on CNS [4] (e.g., the cannabinoid CB1 antagonist rimonabant and sibutramine, a monoamine reuptake inhibitor) promoted the research of new, more effective and better tolerated drugs able to treat this metabolic

10.4155/FMC.14.167 © 2015 Future Science Ltd

Luca Costantino*,1 & Daniela Barlocco2 1 University of Modena & Reggio Emilia, Dipartimento di Scienze della Vita, Via Campi 183, 41100 Modena, Italy 2 University of Milano, Dipartimento di Scienze Farmaceutiche, Via Mangiagalli 25, 20133 Milano, Italy *Author for correspondence: Tel.: +39 059 205 5749 Fax: +39 059 205 5131 [email protected]

disease. Recently, on the basis of the recognized importance in signaling compounds producing gut and adipose tissue, a shift in the research was observed, from compounds acting on CNS, to compounds able to interact with the receptors of substances produced by tissues involved in metabolic homeostasis (gut, adipose tissue), in order to avoid CNS side effects seen with the compounds developed so far. Many compounds showed to be able to decrease bodyweight (BW) in animal models of obesity, but, often, these effects have not been reproduced in clinical studies. This review will deal with compounds that have been considered recently as new antiobesity agents, with a view on their progresses in clinical trial, in order to try to forecast which will be the most successful approach for the next generation of drugs designed for the treatment of this metabolic disease. Despite of the availability of many new molecules able to affect a selected target, active in in vitro and in vivo assays (preclinical), very often these compounds fail in Phase II studies; thus, it seems that polypharmacology (on the basis of bariatric surgery effects) or, perhaps, personalized medicine, could be the most appropriate solution for this medical need.

Future Med. Chem. (2015) 7(3), 315–336

part of

ISSN 1756-8919


Review  Costantino & Barlocco

Key terms Diabesity: Refers to the rapid increasing in obesity, Type 2 diabetes and associated long-term complications. It represents a major health problem. Antisense oligonucletides: Single strands of DNA or RNA that are complementary to a chosen sequence. Oligonucleotides have a wide range of applications.

Pathways connected to obesity Several pathways are involved in the regulation of BW homeostasis. While physiological mechanisms about hunger and the decision to start eating are not fully understood, other processes related to food intake are known, and involve CNS- and periphery interactions. In the periphery, hormones synthesized and released from the gastrointestinal tract and pancreas (glucagonlike peptide 1 [GLP-1], oxyntomodulin (OXM), peptide YY [PYY], cholecystokinin (CCK), ghrelin and glucose-dependent insulinotropic polypeptide [previously known as gastric inhibitory polypeptide, GIP]), together with adipokines secreted from adipose tissue (e.g., leptin) play a central role in communicating with the hypothalamus for the regulation of the energy homeostasis [5,6] . Integration of the diverse signals from the periphery is then integrated in the vagal afferent neurons [7] or communicate directly to the CNS. Inside the CNS, the hypothalamus plays a central role for energy homeostasis; this endocrine gland integrates signals from nutrients and hormones from the periphery (insulin, leptin and incretins) and stimulates anabolic or catabolic pathways. Circuits involved in energy homeostasis are then integrated with those involved in food reward; among the transmitters involved, dopamine, opioid peptides (giving the rationale for the proposed naltrexone–bupropione combination therapy, US FDA approved in 2014) and the endocannabinoid system, with its CB1 receptors, play a central role [5,6] . Compounds acting on CNS neuromediator receptors Historically, antiobesity therapy was based on compounds acting on CNS as sympathomimetic agents (phentermine, FDA approved in 1959, amphetamine), as monoamine reuptake inhibitor (sibutramine, approved by FDA in 1997 and withdrawn in 2010, bupropion) or as CB1 inverse agonist (rimonabant, available in Europe since 2006 for use as an adjunct to diet and exercise for obese or overweight patients with associated risk factors, failed to secure FDA approval in the USA; in October 2008, following a review of postmarketing data, the EMEA recommended suspension of the drug marketing authorization on safety grounds). A cocktail drug (phentermine/topiramate, an adrenergic drug and a GABA receptor activator/kainite and


Future Med. Chem. (2015) 7(3)

AMPA receptor antagonist, respectively) and lorcaserin (a centrally acting serotonin 5HT2c receptor agonist) have been recently approved, but concerns about their side effects rise some doubts about their usefulness [8,9] . Another cocktail drug, naltrexone/bupropion sustained release (opioid antagonist/dopamine and noradrenaline reuptake inhibitor, respectively), has been FDA approved in 2014; Phase III clinical trial showed that this combination is well tolerated [10] but long-term data about its safety about cardiovascular risks are not available yet. Accordingly, safety aspects in compounds acting in this therapeutic area became a very big concern. Some compounds acting on CNS neuromediator receptors are, however, in development, as reported below. Compounds acting as CB1 receptor antagonists/inverse agonists Researches showed that the endocannabinoid system modulates feeding by means of a central- and a peripheral mechanism. The importance of the central mechanism was supported by the effect of the centrally acting antiobesity agent CB1 receptor (CB1R) inverse agonist rimonabant; unfortunately psychiatric side effects led to its withdrawn. However, evidence was given that cannabinoids may promote feeding by acting on peripheral sites [11] . There is evidence that CB1R are expressed in vagal afferent neurons, where it mediates the transmission of orexigenic signals to brain [12] , and in the duodenal epithelium [13] . Even if evidence from CNS-specific CB1R knockout (KO) mice supported the relevance of central CB1R for the antiobesity and metabolic benefits of rimonabant, and CNS-specific CB1 receptor KO mice showed BW and metabolic benefits associated with global CB1R KO mice [14] , nevertheless CB1R antisense oligonucleotide treatment in diet-induced obese (DIO) AKR/J mice, unable to cross the blood–brain barrier, led to decrease in BW (fat mass), insulin and leptin levels, improved insulin sensitivity and glucose homeostasis, strongly supporting the notion that selective inhibition of the peripheral CB1R, without blockade of centrally located CB1R, may serve as an effective approach for treating Type 2 diabetes, obesity and metabolic syndrome [15] . Thus, in order to avoid side effects linked to the CB1R antagonism in the CNS, research was addressed to the discovery of compounds able to target CB1R in the periphery. Several compounds have been recently designed to act on the peripherally located CB1 receptor, adding polar groups onto known CB1 receptor inhibitors to increase polar surface area or conferring to the compounds the ability to interact with P-gp transporters [16] ; these compounds were shown to be active in

future science group

New perspectives on the development of antiobesity drugs 

animal models of obesity [17] . Table 1 reports the structures and data of the most advanced compounds: (1) entered clinical trial, and an IND was recently filled for compound (2). Successful results have been reported in Phase I clinical trial for (1) [18] , but no news about the clinical development of this compound has been reported since 2010. Other strategies have been suggested in the attempt to save the efforts that led to the discovery of rimonabant, namely the combination of low-dose rimonabant with another anorectic agent (an opioid agonist [12] , 5-HT2c receptor agonist mCPP [23] , the gut peptide CCK-8 [24] or others [25–28]). Moreover, the reduction in endocannabinoid biosynthesis could represent an alternative strategy to counteract peripheral endocannabinoid overactivity through dietary n-3 polyunsatured fatty acids or the development of diacylglycerol lipase inhibitors [29] . It has also been suggested that polymorphism of some genes could be involved in rimonabant psychiatric side effects, and this could help in finding individuals who may benefit from treatment with centrally acting CB1 receptor antagonists without psychiatric side effects (personalized therapy) [30] . Histamine H3 receptor antagonists Several studies showed that H3 receptor (H3R) antagonists increase histamine release from the hypothalamus


and reduce energy intake in normal and leptin-resistant DIO mice [31] . Several compounds were active in rodent model of obesity [32–34] and diabetes [35] ; NNC 38–1202 (compound (3), Table 2) showed for the first time an anorectic effect in monkeys [36] . Another H3R antagonist, HPP404 (compound (4), Table 2), a compound that belongs to a series of compounds developed by Novo Nordisk [37] , is effective in producing sustained reductions of food intake and causes in animals a significant reduction in food intake, cumulative BW gain and in total body fat comparable to the CB1 antagonist, rimonabant. HPP 404 has completed a 30-day trial in healthy obese subjects where the drug was well tolerated with no CNS related side effects. Despite lack of dietary restrictions, a meaningful weight reduction was seen. A 6-month Phase II dose-range study in overweight/obese individuals with HPP404 in conjunction with diet and behavioral modification has been recently terminated [38] (data not available). Melanin concentrating hormone receptor 1 antagonists Melanin concentrating hormone (MHC) is an orexigenic peptide predominantly expressed in the lateral hypothalamus and zona incerta, acting on its receptors melanin concentrating hormone receptor 1 (MHCR1)

Table 1. Representative peripherally acting CB1R antagonists. Compound


(1) TM 38837 N O2S HN




CB1R Ki (nM)

B/P Ratio



0.03 0.02

Antagonist. BW reduction in DIO mouse, no effects at CNS level