Expert Opinion on Investigational Drugs

Transcription

1 Expert pinion on Investigational Drugs Forray & oble Subtype selective a1-adrenoceptor antagonists Monthly Focus: ncologic, Endocrine & Metabolic Subtype selective 1-adrenoceptor antagonists for the treatment of benign prostatic hyperplasia Review 1. Introduction 2. Pharmacology of α1-adrenoceptors 3. Launched compounds 4. ompounds in development 5. Expert opinion 6. onclusions Bibliography Patents arlos Forray & Stewart A oble Synaptic Pharmaceutical orporation, Paramus, ew Jersey 07652, USA Benign prostatic hyperplasia (BP) is highly prevalent in the male population beyond the age of 60. Impairment of urinary flow due to prostate enlargement gives rise to symptoms of prostatism that have a detrimental impact on the quality of life. The current trend in the management of symptomatic BP favours pharmacotherapy as a first line option, while the number of surgical procedures being performed has experienced a steady decline during the last ten years. Among the pharmacological treatments, the use of α1-adrenoceptor blockers has demonstrated to be an effective treatment option for BP. These agents reduce the adrenergic tone to the prostate and increase urinary flow, with a concomitant reduction of lower urinary tract symptoms. The α1-blockers currently approved include compounds such as alfuzosin, terazosin and doxazosin, originally developed for the treatment of hypertension, and more recently tamsulosin, an α1-subtype selective drug. The blockade of α1-adrenoceptors present in vascular smooth muscle is largely responsible for the most prominent side effects of current drugs, which can be severe and require patients dose titration. The limitation imposed by side effects naturally raises the possibility that complete blockade of prostatic α1 receptors is not attained at the maximum tolerated dose. The extensive efforts by the pharmaceutical industry towards the development of uroselective α1-blockers, is the subject of this review. Advances in the molecular cloning of genes encoding three α1-adrenoceptors led to the identification of the α1a-subtype as the predominant receptor responsible for the contraction of prostate smooth muscle. In preclinical animal models, selective α1a-antagonists have consistently been found to have minimal cardiovascular effects, thus providing a pharmacological rationale for uroselectivity. It has also become apparent, however, that uroselectivity can emerge in a poorly understood manner from the pharmacodynamic properties of compounds without α1a-subtype selectivity. linical experience with tamsulosin, an α1a/α1d selective drug, has failed to demonstrate a significant improvement in efficacy beyond that demonstrated for non-subtype selective α1-blockers, and gives support to the notion that α1a-selective antagonists might achieve greater efficacy for the treatment of BP. Given the demonstrated uroselectivity of α1a-selective antagonists in preclinical models, it is anticipated that third generation α1-blockers will exhibit improved urinary flow efficacy and be better tolerated than tamsulosin. The extent to which the improvement in urinary flow will translate to the relief of symptoms of prostatism, however, remains to be demonstrated in randomised placebocontrolled clinical trials of α1a-selective antagonists. Keywords: A , 1A-adrenoceptor, alfuzosin, benign prostatic hyperplasia, GW-1818, doxazosin, JT-601, KMD-3213, L-757,464, prazosin, Ashley Publications Ltd. ISS

2 2074 Subtype selective 1-adrenoceptor antagonists RE , RS , SAP6991, tamsulosin, terazosin, uroselectivity two risk factors have been identified: advancing age and the continuous presence of androgens [3]. Exp. pin. Invest. Drugs (1999) 8(12): Introduction 1.1 Epidemiology of BP The prostate is the organ most commonly afflicted by disease in males beyond the age of 60; benign prostatic hyperplasia (BP) is the predominant pathophysiology involved. The disease is characterised by a gradual increase in both the glandular and fibromuscular tissue in the periurethral and transition zones of the prostate. Thus, 70% of 70 year old males develop BP, of whom 40% or more experience severe lower urinary tract symptoms (LUTS), including bladder outflow obstruction with impact on the quality of life [1]. The disease becomes universal with advancing age, such that 90% of male population reaching age 80 exhibit histopathological evidence of BP. Given that that 50% of men with enlarged prostates will develop benign prostatic obstruction, this equates to a one in three probability that men above the age of 50 years will need treatment for BP. In 1997, more than 33 m men in the major industrialised countries suffered from moderate or severe BP. Given current estimates of life expectancy and demographic trends, the prevalence of BP will increase over the next few decades. There is a need to develop safe and effective medicines for the symptomatic relief of BP, thereby diminishing the inherent risks associated with surgical intervention in the elderly. 1.2 Pathophysiology of BP It is thought that the clinical manifestations of BP are related entirely to bladder outflow obstruction [2]. Prostatic enlargement will give rise to obstructive symptoms, such as hesitancy, intermittence, decreased force of the urinary stream, dribbling and retention. Furthermore, it is thought that as a result of the bladder outflow obstruction, bladder dysfunction develops and gives rise to irritative symptoms such as urgency, urinary frequency, a sense of incomplete voiding, nocturia and incontinence. Taken together, these symptoms are referred to as prostatism. If left untreated, symptoms will persist but will not further deteriorate, and on occasion may improve [3]. While the pathogenesis of BP remains largely unknown, 1.3 Treatment options for BP The ultimate objective of the surgical and medical treatment of BP has been to remove the lower urinary tract obstruction. Depending on the severity of LUTS, the treatment options for BP are watchful waiting, medical therapy or surgery. The most widely used surgical treatment for BP is transurethral resection of the prostate (TURP). owever, there are a growing number of minimally invasive therapies under investigation [4]. The incidence of postoperative complications, treatment failure, and reoperation rates after TURP, together with the non-progressive nature of the disease [5,6], are among factors responsible for the continued decline of the number of prostatectomies being performed in urological practice. The opposite trend is observed with medical management of BP, an option that is becoming the first line of treatment for patients without an absolute indication for prostatectomy. At present, there are two pharmacological treatment modalities for BP: 5α-reductase inhibitors and α1-blockers. The rationale for the use of the former came from the observation that patients with congenital deficiency of, 5α-reductase, Type 2, have very small prostates [7], thus indicating the role of this enzyme in controlling prostatic mass. Finasteride, an inhibitor of 5α-reductase, has been demonstrated to be effective in reducing prostate size in BP patients after long-term treatment [8]. The rationale for the clinical use of α-blockers in BP came from the recognition that bladder outlet obstruction not only originates from a static component associated with prostatic enlargement, but also involves a second dynamic component which relates to prostatic smooth muscle tone. It should be appreciated that this prostatic smooth muscle tone is regulated by the adrenergic component of the sympathetic nervous system [9]. Thus, synaptic release of noradrenaline mediates the contraction of prostatic smooth muscle and is associated with constriction, which is inhibitory to urine flow. onversely it might be expected that blockade of noradrenaline action in the prostatic adenoma, capsule and bladder neck might counteract excessive adrenergic tone, mediate relaxation and thus facilitate urine flow. The fact that noradrenaline-induced Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

3 contraction of prostate strips in vitro is sensitive to prazosin rather than rauwolscine implies α1-, rather than α2-receptors as the target [10]. Phenoxybenzamine was the first α-blocker used for the treatment of BP. In pioneering clinical studies, aine and co-workers [11] demonstrated that 80% of BP patients treated with phenoxybenzamine showed symptom improvement and nearly 50% experienced a 2-fold or greater increase of urinary flow [12]. It is of note, however, that clinical efficacy was accompanied with side effects such as dizziness, weakness or tiredness and palpitations in 30% of the patients. These side effects were associated with a 10% dropout rate, indicating that phenoxybenzamine treatment was poorly tolerated. The studies of aine prompted significant interest in the field which has since resulted in the FDA approval and commercial launch of three α1-adrenoceptor antagonists for the treatment of BP. These are terazosin and doxazosin, which were originally developed for the treatment of hypertension, and tamsulosin, an α1-subtype selective antagonist. In addition, alfuzosin, a non-subtype selective α1-blocker is widely marketed in Europe for the treatment of BP. In the last six years there has been an intensive effort in the pharmaceutical industry to design and develop α1a-selective antagonists for the treatment of BP. The expectation is that these compounds would exhibit improved clinical efficacy and a reduced side effect profile relative to currently available agents. 2. Pharmacology of 1-adrenoceptors A key development to improving the pharmacological selectivity of α1-blockers came from the molecular cloning of three genes encoding α1-adrenoceptors, α1a, 1B and 1D [13-15]. Molecular cloning studies confirmed earlier pharmacological evidence that suggested the existence of heterogeneity among α1-adrenoceptors. The fact that three cloned α1-adrenoceptors were isolated was surprising since only two pharmacological subtypes were previously recognised (α1a, 1B) (see [16]), due in part to the limited selectivity of the pharmacological tools available at the time. Thus, WB4101, or 5-methyl urapidil, exhibits good selectivity between α1a- and α1b-adrenoceptors, but fails adequately to differentiate between the α1a- and α1d-adrenoceptors [17,18]. Forray & oble 2075 omparison of the antagonist binding affinity between the recombinant human α1-adrenoceptors with the potency of the same antagonist to block the agonist-induced contraction of human prostate smooth muscle, indicated that the α1a-subtype was responsible for the contraction of the prostatic smooth muscle [17]. The pharmacological profile, together with the demonstration of preferential expression the α1a-adrenoceptor subtype in the human prostate strongly supported the notion that the α1a-subtype was responsible for the contractile tone of the prostatic urethra [19]. Furthermore, four splice variants of the human α1a gene (α1a1 - α1a4) with identical pharmacological properties, have been identified, of which α1a1 is expressed preferentially (75%) by the prostate, followed by α1a2 (20%), α1a3 and α1a4 (3% each) [20,21]. To date, one single nucleotide polymorphism has been reported for the human α1a gene, whose frequency was not found to be associated with BP and which has no effects on the pharmacological properties of the α1a-receptor [22]. ollectively these findings create the opportunity for the development of uroselective α1-antagonists targeted to the α1a-subtype, with reduced cardiovascular side effects, provided it could be demonstrated that the antagonism of the α1a-subtype was not involved in vascular smooth muscle contraction. This notion was later confirmed by in vitro studies that demonstrated that the pharmacological profile of the α1-adrenoceptor responsible for the contraction of human compliance vessels was different from the receptor that mediated the contraction of the human prostate [23]. The in vitro pharmacological profile of the α1-adrenoceptor that mediates the contraction of dog and rat aorta was found to be consistent with that of the α1d-subtype [24,25]. Furthermore, in vivo studies in animal models have shown that selective α1a-antagonists do not induce orthostatic hypotension, and are more potent in blocking urethral pressure than arterial pressure when compared with non-subtype selective α1-antagonists [26,27]. The observation that prazosin and several highly selective human α1a antagonists inhibit the agonistinduced contraction of human prostate at low potency relative to their respective α1a binding affinities, led to the suggestion that α1l-adrenoceptors were responsible for the smooth muscle contraction of the human prostate [27-29]. This classification of α1-adrenoceptors was based on the observation that α-adrenoceptors in tissues can be subdivided into Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

4 2076 Subtype selective 1-adrenoceptor antagonists Figure 1: Structures of launched α1-blockers. l 2 Phenoxybenzamine Terazosin 2 2 Doxazosin Alfuzosin S Prazosin (R)-(-)-Tamsulosin subtypes with high (α1) and low (α1l) affinity for prazosin [30]. Although prostate α1l-adrenoceptors were initially postulated to be distinct from the α1a-subtype, it has since been demonstrated that prazosin exhibits lower functional potency in intact cell assays than might be expected on the basis of it binding affinity for the recombinant human α1a-adrenoceptor [29]. Thus, it was concluded that the pharmacological properties of α1l-adrenoceptors arise, in a poorly understood manner, from the receptor protein encoded by the α1a-adrenoceptor gene. In agreement with this notion is the fact that to date there are no known α1l-selective compounds with low affinity for α1a-adrenoceptors. Moreover, given the fact that α1a-selective antagonists have been found to have little cardiovascular effects in vivo (see below), it is apparent that there would be no advantage in trying to optimise the α1l-activity of the so-called α1a/α1l antagonists. The available data therefore indicate that the α1a-adrenoceptor is the primary site of noradrenaline action in controlling prostatic smooth muscle tone; agents directed against this receptor should prove uroselective and efficacious in the treatment of BP. 3. Launched compounds 3.1 Alfuzosin, terazosin and doxazosin linical studies with prazosin, a selective α1-adrenoceptor antagonist, demonstrated a robust increase in urinary flow with a concomitant reduction of the obstructive symptoms, little or no reduction of the irritative symptoms, and a much lower incidence of side effects when compared to phenoxybenzamine [31-33]. Since then, three quinazoline derivatives similar to prazosin, which were originally developed for the treatment of hypertension, have been approved for the treatment of BP: alfuzosin, terazosin and doxazosin (Figure 1). The last two differ from prazosin and alfuzosin in that they have longer plasma half-lives, thus allowing for once-a-day dosing. Alfuzosin, introduced largely in Europe, has a short plasma half-life (t 1/2 ) and has been found to produce discrete improvements in symptoms as well as urinary flow in patients with BP [34-36]. Terazosin and doxazosin have been approved for the treatment of BP in the United States. Both drugs have been extensively studied and in placebo-controlled studies improve urinary flow 20-30%, versus 7-10% for placebo, with total symptom score reductions Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

5 between 30-80% versus 11-25% for placebo (for a review see [37]). At the recommended clinical doses, terazosin and doxazosin are relatively well-tolerated, with adverse events ranging between 5-17%. Since most of these drugs were initially developed for the treatment of hypertension, it is not surprising that adverse events in BP patients are related to blockade of the cardiovascular α1-adrenoceptors. These receptors play an important physiological role in the maintenance of blood pressure through constricting smooth muscle cells in the vascular wall of capacitance and resistance vessels. In normotensive patients, the blockade of α-adrenoceptors removes the adrenergic tone and impairs the responsiveness of compensatory mechanisms that maintain blood pressure, resulting in symptoms such as tiredness, asthenia, reduced exercise capacity, lightheadedness, postural dizziness, faintness and syncope. Thus, the blockade of the lower urinary tract α1-adrenoceptors by drugs such as alfuzosin, doxazosin, prazosin and terazosin is inextricably linked to the blockade of cardiovascular α1-adrenoceptors. In fact, in placebo-controlled studies with terazosin the incidence of cardiovascular adverse events is more frequent than with placebo [38]. Doxazosin has also consistently been found to have a greater incidence of hypotension than placebo, but has been found to be useful in BP patients with hypertension [39-41]. The collective clinical evidence suggests that the doses at which α1-selective antagonists, such as prazosin, alfuzosin, terazosin and doxazosin, are used for BP better reflect their tolerability than their maximal efficacy. 3.2 Tamsulosin Tamsulosin is the only drug approved for BP with some degree of α1-adrenoceptor subtype selectivity. Tamsulosin has been labelled as an α1a-selective antagonist since it has high affinity for the human α1a-subtype, with 2- and 25-fold lower affinity for the α1d- and α1b-subtypes, respectively [42,43]. In agreement with its poor α1a-subtype selectivity, tamsulosin was found to have, at best, modest uroselectivity as determined by the ratio of its potency to reduce arterial pressure over its potency to reduce urethral pressure in preclinical animal models [27,42]. In Phase I clinical trials, single doses of tamsulosin (up to 0.2 mg) in conventional formulation resulted in a reduction of blood pressure and orthostatic hypotension in 50% of subjects [44]. This result is not unexpected in light of the in vitro, and preclinical in Forray & oble 2077 vivo pharmacodynamic profile, which suggest that tamsulosin exhibits insufficient selectivity between α1a- and α1d-subtypes to pre-empt cardiovascular side effects. In another Phase I study, multiple doses of tamsulosin (0.4 mg) as a controlled release formulation were found to be well-tolerated with no effects on blood pressure or orthostatic hypotension [45]. The difference in tamsulosin tolerability relates to the lower plasma levels of tamsulosin achieved at max with the sustained release formulation. It is noteworthy that all clinical trials for BP have been conducted with the controlled release formulation. Double-blind placebo-controlled Phase III studies to evaluate the safety and efficacy of tamsulosin in Europe have yielded results similar to those conducted in the United States [45-47]. Thus, urinary flow was increased by % in patients receiving 0.4 mg once daily against 5.3-6% increase in the placebo group. The symptom score in treated patients was reduced between 35-48% compared to % in the placebo groups. verall, tamsulosin was well-tolerated, although it should be noted that abnormal ejaculation (8.4; 18.1%) and rhinitis (13; 17.9%) were among the side effects that were significantly more frequent in patients receiving 0.4 and 0.8 mg tamsulosin, respectively. Abnormal ejaculation which includes, ejaculation failure, retrograde ejaculation and ejaculation decrease, is thought to be related directly to the relaxation of the bladder neck, and probably reflects the potency of tamsulosin to block urethral α1a-adrenoceptors. Although dizziness was dose related and higher than placebo [48], it was not statistically significantly different than in the placebo group (10.1%). In the United States trials there was extensive testing for orthostatic hypotension. Thus, 16% of the patients receiving 0.4 mg, and 19% of the patients receiving 0.8 mg of tamsulosin showed at least one positive test, compared to 11% of the patients in the placebo group. Although the incidence of syncope was extremely rare (0.2% at 0.4 mg), the symptom was not apparent in the placebo group. Taken together, these data indicate that tamsulosin in a timed release formulation has a lower propensity to elicit cardiovascular side effects than terazosin or doxazosin. This property of tamsulosin allowed its prescription without the need for titration at the 0.4 mg dose. Extensive clinical experience supports the notion that tamsulosin does not reduce blood pressure in hypertensive patients, and does not Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

6 2078 Subtype selective 1-adrenoceptor antagonists interact unfavourably with other antihypertensive agents [46]. Data from two open-label studies with tamsulosin for BP showed that 84% of the patients rated the treatment efficacy good, or very good, and 90% rated the treatment with good, or very good tolerability [49]. The clinical experience with tamsulosin in BP seems to suggest that it provides better tolerability, albeit with comparable efficacy to terazosin and doxazosin in relieving LUTS and increasing urinary flow. Furthermore, it is apparent that the enhanced tolerability of tamsulosin is achieved by optimisation of its formulation and pharmacokinetic properties, rather than via its inherent pharmacological selectivity. 4. ompounds in development 4.1 Uroselective 1-adrenoceptor antagonists Although non-selective α1-adrenoceptor antagonists such as terazosin and doxazosin are proven to relieve the symptoms associated with BP and are in widespread clinical use, the requirement to titrate the dose raises the possibility that complete blockade of prostatic α1 receptors is not attained at the maximum tolerated dose. The need to titrate arises from the relatively high incidence of side effects upon initiation of dosing with these agents. These side effects include dizziness, asthenia, orthostatic hypotension and occasionally syncope, as previously discussed and are traceable, at least in part, to the blockade of vascular α1-adrenoceptors. The notion that these dose-limiting side effects restrain the urodynamic efficacy of current agents has provoked an intensive search for compounds that exhibit increased selectivity for the prostate. The requirement is thus for uroselective compounds that maintain uroflow efficacy via blockade of α1-adrenoceptors in the lower urinary tract while maintaining vascular tone. It should be appreciated that the meaning of uroselective is context-dependent and involves consideration of pharmacological, physiological and clinical perspectives [50,51]. Two independent approaches to the design of uroselective α1-adrenergic antagonists have emerged: pharmacological and physiological. The pharmacological approach involves the initial identification of agents that possess a high degree of selectivity for the α1a receptor in in vitro assays of binding and functional antagonism. The physiological approach requires only that compounds exhibit functional uroselectivity in animal models where α1-adrenoceptor mediated urethral and vascular effects are compared. With regard to animal models, in vivo efficacy evaluation is most commonly performed in anaesthetised dogs instrumented to measure prostatic urethral pressure in response to either electrical stimulation of the hypogastric nerve (GS) or to systemic administration of phenylephrine (PE) [52-55]. ardiovascular side effects are preferably assessed concomitantly in the same animal [48,56] and/or in a second species, such as the spontaneously hypertensive rat [57]. It is worth noting that an excellent correlation exists between selectivity data derived from radioligand binding and uroselectivity in the dog [43]. rthostatic blood pressure effects are usually assessed in tilt or lift models usually in the rat or dog. Quantification of such data is not only subject to variability, but the uncertainty implicit in any in vivo paradigm is amplified in the context of the dose-ratio calculation that forms the basis of uroselectivity. This caveat notwithstanding, the available data suggests that the dog is a good predictor of clinical urodynamic and haemodynamic profiles in humans. n a cautionary note, it is important to note that claims of selectivity are species-dependent, as evidenced by the finding that RE is approximately 10-fold less uroselective in the rabbit than the dog [58], reflecting a species difference in the distribution of vascular α1-adrenoceptor subtypes [59]. In recognition of these limitations, many investigators have elected to incorporate elements of each approach and have pursued the identification of clinical candidates by filtering novel chemical entities through a succession of in vitro and in vivo screens. Regardless of the process by which preferred molecules are selected, it should be appreciated that a clinically relevant definition of uroselectivity can only be determined in man and that even here, definition is dependent upon the patient population under evaluation. Thus, given that 41.3% of BP patients present with cardiovascular symptoms [60], drug associated reductions in blood pressure may be considered beneficial in a hypertensive patient population but are regarded as a liability in normotensive subjects. Progress over recent years has resulted in the identification of a plethora of novel α1-adrenceptor antagonists, which are either the subject of ongoing or recently completed clinical trials. The following discussion summarises the key attributes of these molecules and reports the results of published clinical Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

7 Forray & oble 2079 Figure 2: Structures of some novel α1-adrenceptor antagonists. l 2 l 2 SL SL S A Rec 15/2739 / SB Table 1: Pharmacological profile of α1-blockers: Binding affinities at human α1-adrenoceptors and antagonism of agonist-induced contractions in human prostate in vitro. ompound Binding affinity (pki) at recombinant 1-adrenoceptors uman prostate α1a α1b α1d Ref pa 2 Ref Launched compounds Prazosin [17] 8.7 [42] Terazosin [17] 7.4 [42] Doxazosin [17] 8.2 [42] Alfuzosin [17] 9.8 [43] Tamsulosin [42] 9.8 [42] ompounds under development SL 89, [42] 8.6 [42] A [66] - RE [42] 8.8 [42] JT [83] 8.8 [84] SAP [90] 9.4 [91] SAP [91] 9.2 [92] L-757, [208] 9.8 [208] GW [210] - KMD [100] 9.5 [101] RS [108] 8.6 [129] trials. The chemical structures of the compounds in question are depicted in Figure 2. A synopsis of the relevant in vitro and in vivo data for these agents is presented in Tables 1 and 2. For the purposes of the following discussion it is convenient to consider these compounds as belonging to one of two categories. The first group, comprising SL , SL , A , RE and JT-601 are characterised by high affinity, but only moderate selectivity for the α1a receptor subtype. The second group, consisting of SAP6201, SAP6991, L-757,464, GW 1818, KMD-3213 and RS (Ro ), all exhibit high affinity and selectivity for the α1a receptor subtype as defined in radioligand binding assays. 4.2 SL /SL SL [201], is a moderately uroselective α1-adrenoceptor antagonist from Synthelabo, that has Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

8 2080 Subtype selective 1-adrenoceptor antagonists Table 2: Uroselectivity of α1-blockers in the anaesthetised dog model. ompound Intraurethral pressure Blood pressure BP/IUP References Launched compounds Prazosin ID 50 in µg/kg; data from Kenny et al. [43] Terazosin K b in µg/kg; data from Broten et al. [93] Doxazosin pa 2 ; data from Kenny et al. [42] Alfuzosin ID 50 in µg/kg; data from Kenny et al. [43] Tamsulosin pa 2 ; data from Kenny et al. [42] ompounds under development SL 89, pa 2 ; data from Kenny et al. [42] A Pseudo pa 2 ; data from ancock et al. [72] RE pa 2 ; data from Kenny et al. [42] SAP K b in µg/kg; data from Broten et al. [93] L-757, K b in µg/kg; data from eremberg et al. [95] GW ED 50 in µg/kg; data from Drewry [96] RS ID 50 in µg/kg; data from Blue et al. [56] completed Phase II clinical trials. Data that support SL efficacy in BP patients has yet to be reported and the development programme has been terminated [61]. The compound was initially identified [62,63] on the basis of its relative ability to antagonise the phenylephrine-induced contraction of isolated rabbit urethra in vitro, to antagonise the increase in intraurethral pressure (IUP) induced by electrical GS in the anaesthetised cat and the evaluation of depressor effects on mean arterial pressure (MAP) in spontaneous hypertensive rats (SR). Thus, SL is 46-fold uroselective, as defined by the ratio of the dose required to produce a 20% reduction in MAP in the SR (1.2 mg/kg i.a.) to that required to effect a 50% inhibition of the increase in IUP on GS in the cat (0.026 mg/kg iv.). It is of note that control experiments with prazosin indicate a ratio of 0.8 under these conditions, consistent with the compound s clinical profile. Experiments performed by others [64] in anaesthetised dogs, however, are less compelling (Table 2: SL BP/IUP = 5) and exemplify the need to exercise caution in the interpretation of uroselectivity data, particularly with cross-species comparisons. Synthelabo have recently reported [60] the discovery of SL , a new balanced α1-adrenoceptor antagonist, that exhibits equal affinity for each of the cloned α1 receptor subtypes (Figure 2). The compound is reported to show dose-dependent effects on IUP in the absence of effects on blood pressure, which is proposed to occur via the preferential distribution of the compound to the prostate. The prostatic concentration of SL is reported to be 10X that found in plasma at 1 h following oral administration and rises to 24X at 6 h post dose. Although such preferential distribution was originally claimed for SL and been seen in other contexts by others, it is unclear to what extent such effects contribute to the observed uroselectivity of SL The clinical status of SL is presently unknown. Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

9 4.3 A A is a novel uroselective α1-adrenceptor antagonist that is under development at Abbott Laboratories [65]. The compound exhibits subnanomolar affinity for the α1a-receptor (K i = 0.22 nm) and an α1 subtype selectivity profile similar to that of tamsulosin, in that A retains appreciable affinity for α1d (K i = 0.97 nm) and is somewhat selective over the α1b-receptor (K i = 6.95 nm) [66,67] (Figure 2). In isolated canine prostate and rat vas deferens [68], A functions as a potent α1a antagonist under conditions of PE challenge (pa and 8.93, respectively). Similar experiments in rat aorta [69] indicate that the compound also functions as a potent α1d antagonist (pa 2 = 9.11). Studies with A in rat spleen [68] also reveal α1b antagonism (pa 2 = 7.89), with diminished potency being in line with the α1b receptor binding affinity of the compound. Thus, A exhibits 11- to 13-fold selectivity for α1a over α1b, an attribute that distinguishes the compound from tamsulosin, which marginally favours the latter. In isoflurane anaesthetised dogs, A antagonised the adrenaline-induced increases in IUP with a pseudo pa 2 = In SRs, the compound caused a transient dose related decrease in MAP as assessed by area under the curve analysis over 60 min, with a log index value of Taken together, these data are suggestive of > 600-fold selectivity between A s IUP and MAP effects [66]. Experiments in which the effect of A on IUP and MAP were concurrently evaluated in anaesthetised dogs support these findings (Table 2) [70,71]. Thus, A antagonised PE-induced increase of IUP with pseudo pa2 = 8.0, whereas doses in excess of 1000 nmol/kg iv. were required to effect a 10 mm g reduction in blood pressure, ped 10 = In conscious dogs, A dose-dependently inhibits PE-induced increases in IUP with long duration of action [55] at doses above 0.3 mg/kg po. [55,71,72]. These observations are striking since pharmacokinetic analysis of A following oral administration in the dog reveals a plasma t 1/2 less than 1 h ( h). It is of note, therefore, that fast plasma kinetics may well contribute to the transitory nature of the effects of A on basal MAP or the blockade of PE-induced pressor responses. Experiments designed to assess the time dependence of the prostatic concentration of A following oral administration to dogs have yet to be reported. The rate of Forray & oble 2081 dissociation of A from prostatic compared to vascular α1-adrenoceptors is also unknown. With regard to the potential liability for S-mediated side effects, studies with radiolabelled terazosin and A indicate that the brain to plasma ratio for A is a small fraction of the ratio observed for terazosin after administration of equivalent doses. These data suggest that A exhibits a level of S penetration that is considerably reduced relative to terazosin. To what extent this difference will translate to a reduced incidence of S-mediated adverse events remains to be established. The development status of A or other candidates is presently unknown. 4.4 RE /SB RE is an α-1 antagonist from Recordati that was under development at SmithKline Beecham (as SB216469) and subject to evaluation in Phase II clinical trials in the US and UK [73]. The results from these clinical trials were disappointing and development was suspended in 1996 [74]. Further trials were conducted by Recordati in Europe and RE was reported to have re-entered Phase II clinical trials in 1998 [75]. Recordati has since discontinued development of the compound stating that the results of these trials did not confirm the competitive profile of the compound [76]. Although a detailed analysis of the failure of RE is not presently available, it is understood that the initial clinical trials failed to provide data in support of compound efficacy. RE [202] is a 4-1benzopyran-8- carboxamide derivative that is structurally related to 5-methyl urapidil and SL 89,0591 (Figure 2). Binding selectivity for α1a over α1b was initially demonstrated in E-pretreated rat hippocampus and rat liver tissue preparations respectively [77,78]. The compound was subsequently assayed against both rat [71] and human [79,80] recombinant α1-receptor subtypes. RE has potent affinity for the human α1a-receptor, is selective over α1b and has intermediate affinity for α1d (Table 1). RE has been evaluated for its ability to antagonise noradrenaline-induced contraction of isolated rabbit [79] and human prostate [81] tissue. The compound exhibits Kb = 2.7 nm in both species, with 39- and 6-fold selectivity over effects in rabbit and human vascular arterial tissue, respectively. Studies in the anaesthetised dogs indicate that RE exhibits modest uroselectivity under a variety Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

10 2082 Subtype selective 1-adrenoceptor antagonists Figure 3: Structure of JT-601 and its metabolite 3 3 JT JT-601-G1 of experimental conditions. Under conditions of phenylephrine stimulation, RE antagonises the increase in IUP with pa 2 = 8.74 compared to blood pressure with pa 2 = 7.51, indicative of 17-fold uroselectivity (Table 2) [203]. Independent studies in which blockade of GS increases in IUP was compared to PE-induced increase in MAP indicate that RE is 30-fold uroselective [56]. In the conscious dog RE (0.1-3 mg/kg po.) preferentially attenuates the PE-induced increase in IUP, with potency equal to terazosin, relative to the increase in MAP. This finding is in contrast to studies in isolated canine prostate tissue in which RE is approximately 100-fold more potent than terazosin and may reflect the 5-fold differential bioavailability of the compounds. It is of note that the poor intraduodenal availability of RE has precluded the use of the compound in comparative studies by others [56]. The duration of RE action in conscious dogs was shorter than that found with terazosin, doxazosin and tamsulosin, consistent with the observation that the compound is rapidly cleared from plasma [72]. To what extent these pharmacokinetic parameters conspired to adversely impact the clinical efficacy profile of RE is presently unclear. n a positive note, preliminary data in models of orthostatic hypertension in rats markedly differentiate RE , which had no effect at doses up to 10 mg/kg iv. from prazosin, terazosin and tamsulosin, which are all associated with orthostaticinduced falls in blood pressure in rats at much lower doses ( µg/kg iv.). 4.5 JT-601 JT-601 [204] is a high affinity α1a/1l-adrenoceptor antagonist reported to be in early clinical development at Japan Tobacco Inc [82]. haracterised in tissue binding assays, JT-601 exhibits 10-fold higher affinity for α1l, similar binding affinity at α1a and 10-fold lower affinity at α1b and α1d than prazosin in parallel experiments (Table 1) [83]. In functional studies, JT-601 antagonised noradrenaline-induced contraction of human prostate with pa 2 = 8.84 (with respect to tamsulosin: 9.78; WB4101: 8.39; prazosin: 8.23) being the only compound to exhibit selectivity (10-20X) over effects on mesenteric artery (α1d) under these conditions. Interestingly, JT-601-G1, the principal JT-601 human metabolite also behaved as a potent antagonist in human prostate (pa2 = 8.12) and exhibited selectivity similar to the parent compound [84] (Figure 3). In vivo binding experiments substantiate claims that JT-601 exhibits tissue selective behaviour. Thus, JT-601 dose-dependently inhibits specific [ 3 ]tamsulosin binding following iv. injection in rats and exhibits greater selectivity of action than either tamsulosin or prazosin, as assessed by the magnitude of the rat aorta ID 50 /rat prostate ID 50, or spleen ID 50 /submaxillary gland ID 50 ratios [85]. JT-601 is claimed to lower intraurethral pressure and exhibited less pronounced effects on blood pressure than either tamsulosin or prazosin in a phenylephrine-challenge model in anaesthetised rabbits, or following intraduodenal administration in anaesthetised dogs [60]. 4.6 SAP 6201/SAP 6991/L-757,464 Scientists at Synaptic were the first to report the discovery of compounds with appreciable affinity and selectivity for the α1a receptor [203]. The initial leads, typified by SAP5089 [86] and SAP5399 [87] were derived by structural manipulation of the 1,4-dihydropyridine calcium channel antagonist (S)-(+)-niguldipine [88]. Thus, SAP5089 binds the human α1a-receptor with subnanomolar affinity (Ki = 0.4nM) and is 630 and 1500-fold selective for this receptor over α1b and α1d, respectively. SAP5399 has a receptor binding profile similar to SAP5089 but superior physicochemical properties such that the compound exhibits dose-dependent inhibition of PE-induced contraction of the dog prostate with Kb = nm, in close agreement with the compound s affinity profile at α1a (pk i = 9.1). Issues relating to bioavailability were initially addressed by replacement of the 1, 4-dihydropyridine Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

11 nucleus of SAP5399, which is potentially prone to metabolic oxidation, by a dihydropyrimidinone template [89,205], which after modification resulted in the discovery of SAP6201 [90]. The in vitro and in vivo profile of SAP 6201 is presented in Tables 1 and 2, respectively. Pharmacokinetic studies with SAP 6201 in Sprague Dawley rats and Beagle dogs indicated 19% and 26% bioavailability with t 1/2 of 2.0 and 2.5 h, respectively. SAP6201 metabolite analysis revealed the major degradative pathway involves -dealkylation at the propyl piperidine centre. Replacement of the piperidine by an aminocyclohexyl moiety led to the discovery of SAP6991 [91]. SAP6991 has an in vitro [92] and in vivo [93] profile that suggest the molecule has significant potential. The data presented in Table 1 indicate that SAP6991 is a potent α1a-subtype selective agent. The compound inhibited all, or the majority of [ 3 ]prazosin/[ 125 I]EAT binding in human rat and dog prostate (α1a) with high affinity (K i = nm) and was 100 times less potent in tissues known to contain non-α1a subtypes, such as rat liver and dog aorta (K i = nm). SAP6991 antagonised PE-induced contraction of isolated rat (Kb = 8.1 nm) and dog prostate (Kb = 2.7 nm). It also inhibited A induced contraction in rat prostate (Kb = 1.2 nm), human prostate (Kb = 0.57 nm) and rat tail artery (Kb = 1.1 nm). In contrast, the E-induced contraction of rat aorta was relatively resistant to SAP6991 (Kb > 1 mm). SAP6991 potently inhibits PE-induced contraction in the in situ rat prostate (AD 50 =35µg/kg iv.) with duration of action>4h,buthadnoeffect on PE-induced increases in BP in pithed rats at doses of up to 1 mg/kg iv. Furthermore SAP6991 [94] did not alter basal BP in conscious SR rats or conscious dogs, or potentiate the orthostatic BP response to head-up tilt in anaesthetised rats. These results sharply contrast with those obtained in parallel studies with terazosin for which significant decreases in basal BP in conscious SR and dogs (100 µg/kg, iv.) and potentiation of the orthostatic BP response to tilt in rats (30 µg/kg) were apparent. It is of particular note that in anaesthetised dogs SAP6991 potently inhibits PE-induced increases in prostatic IUP with AD 50 =48µg/kg, with no effect on basal BP at doses up to 3 mg/kg iv., indicative of 62-fold uroselectivity. Taken together, these data indicate that SAP6991 may well offer clinical advantage over current treatment options for dysuria associated with BP. The finding that SAP6991 exhibits 43% Forray & oble 2083 bioavailability and t 1/2 = 6.7 h in dogs is encouraging in this regard. This knowledge has since been exploited, resulting in the discovery of compounds in additional chemical series. In this regard, it is of note that scientists at Synaptic and Merck have recently disclosed compounds in an oxazolidinone series [ ]. Scientists at Merck have also described a series of -alkylated saccharin derivatives [209] related to L-757,464 [95] (Figure 4). L-757,464 is a potent uroselective α1a-antagonist that is structurally unrelated to SAP6201 or SAP6991. L-757,464 binds the human α1a-receptor with high affinity (K i = 0.08 nm) and manifests 300 and 1000-fold selectivity over α1b and α1d, respectively (Table 1). This selectivity is also apparent in tissue binding studies which demonstrate the compound has high affinity for dog (K i = 0.24 nm) and human prostate (K i = 0.17 nm), with 310- and 530-fold selectivity over their respective aortic counterparts. In vitro functional assays in tissues, performed under the conditions described for SAP6201 and SAP6991 above, demonstrate similar results. Thus, L-757,464 antagonised PE-induced contraction of isolated rat (Kb = 2.0 nm) and dog prostate (Kb = 1.7 nm), but had limited effects on rat (Kb = 110 nm) and dog aorta (Kb=>100 nm). In anaesthetised dogs L-757,464 potently inhibits PE-induced increases in IUP with AD 50 = 5.5 µg/kg, whereas inhibition of increases in DBP occurred with Kb = 156 µg/kg, which equates to 28-fold uroselectivity. The potential for L-757,464 selectivity of action was further demonstrated by studies that indicate the compound has no effect on BP in the SR (3 mg/kg, iv.), conscious dogs (1 mg/kg, iv.) and conscious rhesus monkeys (1 mg/kg, iv.). From a developmental perspective, a compound has been evaluated in initial clinical trials, but results are not yet available. 4.7 GW-1818 GW-1818 is a potent, selective α1a-antagonist that has been evaluated in early clinical trials by GlaxoWellcome. GW-1818 is a 4-trifluoroethoxymethylene substituted oxazole derivative, structurally related to tamsulosin [96,210] (Figure 5). It is of note that trifluoroethoxyl functionality is also resident in RS-100,975 and KMD-3213, suggesting that this group may be a key structural element that confers selectivity for the α1a receptor. The discovery of GW-1818 is the subject of a recent review [97]. As Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

12 2084 Subtype selective 1-adrenoceptor antagonists Figure 4: Structures of Synaptic and Merck compounds (S)-iguldipine SAP F F SAP 5399 SAP 6201 F F S l SAP 6991 L-757,464 detailed in Table 1, GW-1818 binds the human α1areceptor with sub-nanomolar affinity, and is selective for this receptor over α1b and α1d by 81 and 112-fold respectively. GW-1818 potently inhibits increases in IUP in the dog GS model (ED 50 = 13 µg/kg), weakly inhibits diastolic blood pressure responses to PE administered iv. in the dog and exhibits 550 selectivity for the prostate as defined by the BP/IUP ED 50 ratio. GW-1818 has a low volume of distribution (L = 0.9 ml/min/kg), a mediocre half-life (4 h) and excellent bioavailability (F = 93%) in male mongrel dogs. Allometric scaling of clearance and volume from rats and dogs predicted a half-life of approximately 5 h in humans which was initially perceived to be inadequate for once-daily dosing. Duration of action studies indicated that GW-1818 elicits a sustained pharmacological response in dogs after single dose administration (50 µg/kg iv.). Thus, it is of note that whereas plasma levels of parent drug declined nearly 50-fold over the 26 h period, the efficacy response, as determined by inhibition of the increase in IUP following PE challenge was only reduced 2-fold. The discrepancy between the duration of plasma drug levels and pharmacological response was explored. It is of note relative to comments made earlier with respect to A , that experiments with titrated GW-1818 demonstrated reversible binding to the α1a receptor with a fast on/off rate. ther studies revealed the presence of several potent metabolites in dog urine and plasma, however, the pharmacokinetic time profiles of these were not studied and their contribution to the pharmacodynamic properties of GW-1818 remains unknown. GW-1818 was reported to have commenced Phase I clinical trials in ovember o details regarding Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)

13 Forray & oble 2085 Figure 5: Structure of GW Figure 6: Structure of KMD F F 3 GW 1818 S F 3 KMD-3213 the results of these studies have been published, although it is understood that the compound is no longer being progressed. The extent to which the discovery that GW-1818 inhibits cardiac IKR with high specificity [98] was influential in decision making is unknown. 4.8 KMD-3213 KMD-3213 is a potent selective α1a-antagonist from Kissei Pharmaceutical that is presently the subject of Phase II development in Japan and the USA and under Phase I evaluation in Europe for the potential treatment of dysuria associated with BP [99]. The compound shares structural features with tamsulosin and exhibits a pharmacological profile that is similarly dependent upon the stereochemical configuration at the 2-aminopropyl centre (Figure 6). As shown in Table 1, KMD-3213 binds to cells expressing the human α1a-receptor with sub-nanomolar affinity and is selective for this receptor over α1b and α1d by 583- and 56-fold, respectively [100]. The compound inhibits noradrenaline-induced increase in intracellular a 2+ concentrations arising from functional activation of the human α1a-receptor with I 50 = 0.32 nm, but exhibited a diminished inhibitory effect at α1b and α1d. In human isolated prostate, KMD-3213 potently inhibited noradrenaline-induced contraction with pk B = 9.45 [101]. Tissue selectivity was initially evaluated in rabbit prostate, rabbit and rat aorta where KMD-3213 was found to inhibit the PE-induced contractions with pa 2 = 10.05, 9.36 and 8.13, respectively [102]. In ex vivo experiments in the rat, oral administration of KMD-3213 at mg/kg caused a significant dose-dependent-reduction of the binding potential (B max /K d ) of [ 3 ] prazosin in the prostate and submaxillary gland, but not in spleen, heart or cerebral cortex [103]. Finally, iv. and intraduodenal administration of KMD-3213, prazosin and tamsulosin in rats under conditions of PE challenge dose-dependently inhibited IUP and decreased blood pressure. In these studies KMD-3213 uroselectivity, as defined by the BP ED 15 /IUP ID 50 ratio was 29- to 32- and 5- to 8.2-fold higher than that of prazosin and tamsulosin, respectively. It is also of note that KMD-3213, like A has been found to exert longer lasting effects on lower urinary tract function than in vascular tissues [104]. As previously mentioned, KMD-3213 is the subject of a clinical development program at Kissei Pharmaceutical o. Ltd. As yet, no information relating to the efficacy of the compound in BP patients is available. 4.9 RS offmann-laroche have recently disclosed [211] a series of compounds structurally related to 5-methyl urapidil from which RS [105] and RS (Ro ) [56] were selected for further evaluation. RS is known to have commenced clinical studies, although the present status of the compound is unclear. RS is reported to have reached and completed Phase II clinical trials, although data from these studies are not available [61]. A report that a Roche compound reached Phase II but failed to show sufficient efficacy has surfaced [60], but this remains yet to be confirmed. The development of RS-17053, an early clinical candidate is known to have been discontinued [106]. RS is an -aryl piperazinyl--propylamino derivative structurally related to 5-methyl urapidil and SL 89,0591 (Figure 7). The compound was initially selected on the basis of primary functional assay in isolated rabbit bladder neck (RB) strips, under conditions of noradrenaline stimulation, rather than in isolated perfused rat kidney, since data indicated that the former system was more predictive of α1a/1l-adrenoceptor pharmacology in the human lower urinary tract. Selectivity over α1d was established via functional assay in rat thoracic aortic rings (RTA). As assessed in these assays, RS exhibits RB pa 2 = 8.8 and RTA pa 2 = 7.7, being Ashley Publications Ltd. All rights reserved. Exp. pin. Invest. Drugs (1999) 8(12)