|Home | About | Journals | Submit | Contact Us | Français|
The aim of this study was to determine whether the magnitude of the intraocular-pressure (IOP)-lowering response in monkeys to the nonselective prostaglandin (PG)F2a-isopropyl ester (ie) can be reproduced by combining other PG-subtype-selective compounds. IOP was lowered by approximately 25% after 4–5 days of topical administration with latanoprost (FP agonist, 1.5 μg, q.d.), bimatoprost (prostamide, whose metabolites have been shown to be FP agonists; 9 μg, q.d.), or travoprost (FP agonist, 1.2 μg, q.d) or the EP2 agonist, butaprost (25 μg, b.i.d.). The EP1 agonist, 17-phenyl trinor (PhT) PGE2 (b.i.d.), and EP3 agonist, sulprostone (b.i.d.), had no IOP-lowering effects. The addition of butaprost, sulprostone (10 μg), or 17PhTPGE2 (25 μg) to latanoprost did not lower IOP more than latanoprost alone. However, treatment with the combination of latanoprost, 17PhTPGE2, butaprost, and sulprostone produced a similar 50–55% reduction in IOP, as did PGF2α-ie (b.i.d.). In conclusion, latanoprost, travoprost, and bimatoprost produce similar IOP-lowering responses in normotensive monkeys and are most efficacious when administered q.d. pm, compared to b.i.d. The combination of the FP, EP1, EP2, and EP3 agonists used in this study was sufficient to lower IOP by the same magnitude as PGF2α-ie, suggesting that combining PG-subtype agonists may be a potent antiglaucoma strategy.
Prostaglandin (PG) analogs are the most potent, efficacious topical ocular hypotensive agents currently known for the treatment of human glaucoma. 1 The most effective PGs for lowering intraocular pressure (IOP) in humans are derivatives of PGF2α modified structurally to enhance ocular penetration and, after metabolism to the free acid, specifically activate the FP-prostanoid receptor. 2 The most undesirable side effects have also been eliminated as a result of these modifications (historic reviews3,4).
In normotensive monkeys, PGF2α- isopropyl ester (ie) dramatically lowers IOP following 3–5 days of twice-daily topical treatment before the onset of any adverse ocular effects. 5 However, direct comparison in monkeys of PGF2α-ie to three of the most commonly prescribed PG analogs for glaucoma therapy in the United States, latanoprost, bimatoprost, and travoprost,6,7 has not been done. Therefore, it is unknown if maximum IOP-lowering capability with PG derivatives has been achieved.
PGF2α has a substantial affinity for EP1, EP2, EP3, and EP4 PG receptor subtypes as well as for FP receptors.2,8 Therefore, it may be possible to add subtype-selective PG agonists to the current clinical FP-selective analogs 9 to produce additional IOP-lowering efficacy without increased side effects.
We first directly compared the IOP responses of latanoprost, bimatoprost, and travoprost in monkeys that responded strongly to PGF2α-ie and found that there was no difference in IOP-lowering efficacy. Therefore, subsequent studies were conducted by adding other EP-selective subtype compounds to latanoprost. Then, the most effective IOP-lowering combination of latanoprost and available EP agonists was compared to that of PGF2α-ie.
All animal experiments were conducted in accordance with the University of Wisconsin (Madison, WI) IACUC and National Institutes of Health (NIH; Bethesda, MD) guidelines.
Thirty-five (35) ocular normotensive cynomolgus monkeys (Macaca fascicularis) (both sexes, 3.1–7.6 kg) were studied, usually in groups of 4, 8, or 9. All monkeys were free of any ocular abnormalities. The monkeys were anesthetized with intramuscular (i.m.) ketamine (10 mg/kg, initial, supplement 5 mg/kg, as needed) for drug administration and IOP measurements. Lactated Ringer's fluids (10 mL/kg) were administered subcutaneously (s.c.) 2–3 times during the course of the 6- and 8-h IOP studies.
The following agents from Cayman Chemical (Ann Arbor, MI) were studied: PGF2α-ie (2 μg); latanoprost (1.5 or 4.5 μg); bimatoprost (9 or 27 μg); travoprost (1.2 or 3.6 μg;); 17-phenyl trinor (PhT) PGE2 (25 μg, 50 μg; EP1>EP3 selective10,11); butaprost (25 μg, 50 μg; EP2 selective11,12); and sulprostone (5 μg, 10 μg, 50 μg; EP3>EP1 selective11,13,14).
If solvent was present, it was first blown off with a stream of nitrogen passed through a 0.2-μm filter. Stock solutions of the study agents were prepared in 100% dimethyl sulfoxide (DMSO) and kept at –20°C. With some noted exceptions, treatment solutions and vehicles were formulated in 20% DMSO/phosphate-buffered saline (PBS) and were given as two 5-μL drops (unless otherwise noted) administered to the central cornea of the supine, anesthetized monkeys. This vehicle was chosen to minimize sources of confounding due to formulation variations in clinical preparations.
The eyelid was held open for 30 sec following administration, and 1 min elapsed between drops. When a given eye received more than one drug solution during a given treatment session, the agents were separated by 5 min. Once- (pm) and twice-daily treatment regimens for 4–5 days were evaluated.
Before drug administration, slit-lamp examination (SLE) ensured that no inflammation was present and that the eyes were normal. Baseline IOP (with nondairy creamer as the interface) 15 was measured by Goldmann applanation tonometry. 16 For latanoprost, bimatoprost, and travoprost comparisons, animals were selected that demonstrated at least a 5-mmHg IOP reduction at 3–4 h on day 4 of b.i.d. treatment with 2 μg PGF 2α-ie. These animals were studied in groups of 3 as follows:
|Group 1||Group 2||Group 3|
Some of these animals were included in the remainder of the studies; no prescreening was done for any additional animals that were used.
IOP measurements for once-daily (q.d.) treatments were performed at 12, 14, 16, 18, and 20, and sometimes 36 h, and SLE at 14 and 20 h after the last treatment, usually given at approximately 7:30–8:00 pm on day 4. IOP measurements for twice-daily (b.i.d.) treatments were taken prior to and at 0.5, 1, 1.5, 2, 3, 4, 5, and 6 h (also at 7 and 8 h for studies including butaprost), and SLE prior to, and at 3 and 6 h after, either the seventh or ninth treatment. The monkeys were allowed to recover for at least 2 weeks between experiments.
The percentage change in IOP between treated and control eyes for pretreatment baseline, or for treated or control eyes compared to pretreatment baseline, was calculated as follows:
where T=experimental eye; C=vehicle or control eye; and BL=baseline prior to the first treatment. Data are given as the mean±standard error of the mean. Significance was determined by the two-tailed paired t-test for differences compared to 0.0.
The clinical doses of latanoprost (1.5 μg), bimatoprost (9 μg), and travoprost (1.2 μg) were equally effective in significantly lowering IOP by approximately 25% from hours 12–20 following the fourth treatment (P<0.001 at 12–20h; P<0.005 at 36h; Fig. 1A). Despite the apparent trend toward a lower IOP response with q.d. PGF2α-ie (1.6 μg in this case) there was no significant difference in percent change in IOP compared to that of latanoprost, bimatoprost, and travoprost, when the same 4 monkeys were compared from each group.
IOP response was also determined at 3× the clinical doses of each compound, latanoprost (4.5 μg), bimatoprost (27 μg), and travoprost (3.6 μg), administered q.d. in subsets of 3 animals (Fig. 1B). Agents were given in 6×5 μL drops. Comparisons of the results to those for the corresponding animals studied by using the clinical doses described above suggested no enhancement of the IOP-lowering response; therefore, additional animals were not studied. The combined data from the three compounds (3 different animals for each compound; therefore, n=9) also indicated that the 1× dose of each compound was as effective as the 3× dose at 16–20 h after the fourth treatment.
Twice-daily treatment with the clinical dose of latanoprost (1.5 μg) caused slight and variable decreases in IOP measured prior to, and for, 6 h following the seventh treatment. Bimatoprost (9 μg) had essentially no effect on IOP during this time. In contrast, previous screening of these animals for their IOP responsiveness to PGF2α-ie demonstrated a significant and dramatic decrease in IOP of nearly 50% (P<0.001) at 3–4 h after the seventh dose. For screening studies only (Fig. 1C), the PGF2α-ie dose (2 μg), was formulated in 5 μL containing 0.4% Tween 80/PBS. For subsequent studies, PGF2α-ie was formulated in 20% DMSO/PBS, as were other compounds, as stated in Methods. As can be seen in Table 1 (panels 1C vs. 3E and 3F), there was no diminution of the IOP response to PGF2α-ie (and, presumably, all the other compounds studied) by formulation in 20% DMSO/PBS instead of Tween 80/PBS.
Following the IOP measurements shown in Figure 1C, the corresponding monkeys received the eighth dose of latanoprost (1.5 μg) or bimatoprost (9 μg). The next day, 12–20 h after the eighth dose, IOP measurements were repeated. Neither agent consistently decreased IOP, compared to baseline, prior to the first treatment (not shown).
There was essentially no change in IOP prior to, and for, 6 h following the seventh doses of b.i.d. treatment with 17PhTPGE2 (25 or 50 μg) or sulprostone (5, 10, or 50 μg; Fig. 2A and 2C). The 50-μg dose of sulprostone was formulated in 2×5 μL drops of 30% DMSO/PBS.
Butaprost significantly (P<0.05, minimum) decreased IOP by approximately 15–30% from hours 4–7 following the seventh 25-μg dose of b.i.d. treatment (Fig. 2B, 2D, and 2E). No greater response was attained with the 50-μg dose. The 50-μg dose of butaprost was formulated in 2×5 μL drops of 30% DMSO/PBS.
The addition of 17PhTPGE2 (25 μg) or sulprostone (10 μg) to 25 μg of butaprost (Fig. 2D and 2E) did not significantly alter the IOP-lowering response to butaprost alone.
Latanoprost was used in subsequent subtype combination studies, since there was no difference in the IOP-lowering response to latanoprost, bimatoprost, or travoprost, as shown in Figure 1.
First, 17PhTPGE2 (25 μg; Fig. 3A), butaprost (25 μg; Fig. 3B), or sulprostone (10 μg; Fig. 3C) was administered b.i.d. to one eye; latanoprost (1.5 μg) was administered q.d. in the afternoon to both eyes. For the afternoon treatment, 17PhT-PGE2, butaprost, or sulprostone was administered first. Latanoprost was administered 5 min later. On day 5, 17PhT-PGE2, butaprost, or sulprostone was administered immediately after the 12-h post-latanoprost IOP measurement.
Compared to baseline IOP prior to the initial treatment, latanoprost alone decreased IOP on day 5 by approximately 15–22% from 14 to 18 h after the fourth latanoprost treatment (Fig. 3A, 3B, and 3C; Table 1). Latanoprost in combination with 17PhTPGE2 or butaprost may have slightly enhanced the IOP-lowering response to latanoprost alone, although differences between the eyes were not significant at any time points. Latanoprost in combination with sulprostone may have slightly attenuated the IOP-lowering response to latanoprost alone, but no differences were significant.
Sulprostone (10 μg) and butaprost (25 μg) formulated together in a single solution were administered b.i.d. to one eye; latanoprost (1.5 μg) was administered q.d. in the afternoon to both eyes. For the afternoon treatment, sulprostone+butaprost or vehicle were administered first. Latanoprost was administered 5 min later. On day 5, sulprostone+butaprost were administered immediately after the 12-h post-latanoprost IOP measurement. The combination of latanoprost with sulprostone+butaprost further decreased IOP by an additional 15% at the 16–18-h time points (P<0.05; n=11), compared to latanoprost alone (Fig. 3D).
In another series of experiments, treatment with the combination of sulprostone+butaprost+latanoprost, as described above, was then compared to b.i.d. PGF2α-ie (2 μg). Compared to initial baseline IOP prior to the first treatment, the combination of latanoprost+sulprostone+butaprost decreased IOP by approximately 30–40% from 14 to 16 h after the fourth latanoprost treatment (2–4 h after the ninth sulprostone+butaprost treatment). However, PGF2α-ie decreased IOP by an additional 20% or more following the ninth dose (P<0.05; n=8; Fig. 3E).
Last, all three EP subtype-selective agonists were formulated together in a single solution, that is, 17PhTPGE2 (25 μg)+sulprostone (10 μg)+butaprost (25 μg) and administered b.i.d. to one eye, and PGF2α-ie (2 μg) b.i.d. to the opposite eye. Latanoprost (1.5 μg) was administered q.d. in the afternoon to the EP agonist-treated eye. On day 5, 17PhTPGE2+sulprostone+butaprost or PGF2α-ie were administered immediately after the 12-h post–latanoprost IOP measurement to the corresponding eyes. The combination of latanoprost with 17PhTPGE2+sulprostone+butaprost decreased IOP on day 5 at 4 h post-treatment by 52%, whereas PGF2α-ie decreased IOP by 58% (Fig. 3F). There were no significant differences in IOP-lowering response between the eyes at any time point.
All eyes were free of biomicroscopic cells and flare, except in 4 different monkeys receiving butaprost alone or in combination with sulprostone or sulprostone and latanoprost. In both eyes 1+, trace or rare cells were present. In 2 cases, the measurements were discontinued and the data not included. In 2 other cases (rare or trace cells), the data were calculated with and without those monkeys. Since there was no difference in the results with or without those monkeys, the data were included.
Clinical doses of latanoprost, bimatoprost, and travoprost administered q.d. pm produced maximal IOP-lowering responses in normotensive monkeys, compared to b.i.d treatments or threefold higher doses given q.d. Latanoprost, bimatoprost, and travoprost were similar in their efficacies. Prescreening the monkeys for strong responsiveness to PGF2α-ie may have eliminated some variability in IOP efficacies, as reported in humans, where nonresponders to latanoprost may still respond to bimatoprost.17–19
The EP1 (EP1>EP3) agonist 17PhTPGE2 had little effect on IOP alone or in combination with butaprost or latanoprost. The EP1 receptor was shown to mediate the IOP-lowering response to PGF2α in cats. 10 EP1 and EP3 receptors were shown to be prominently expressed in human ciliary body and cornea 20 as well as in trabecular cells. 21 EP1 receptor mRNA transcripts were present in all muscle fibers of the human ciliary body. 22 It is also possible the b.i.d. treatment regimen utilized in the current study may not be optimal, as was the case with latanoprost and bimatoprost.
The EP2 agonist, butaprost, lowered IOP similarly to what has been previously reported. 23 The addition of butaprost to latanoprost did not significantly enhance the IOP-lowering response to latanoprost. Immunohistochemistry studies showed that EP2 receptors were the most abundantly expressed EP subtype in human ocular tissues. 20
The EP3 (EP3 ≥ EP111,13,14) agonist, sulprostone, had very little effect on IOP when used alone. This is in contrast to early studies in monkeys, where a 2.5–3-mmHg reduction was reported after a single 25-μg dose of sulprostone or an ~4-mmHg reduction after twice-daily treatment with 25 μg of another EP3 agonist, MB 28767. 24 The addition of sulprostone to latanoprost did not further enhance the IOP-lowering response. In EP and FP receptor knockout mice, part of the ocular hypotensive response to the PGF2α analog, tafluprost, were found to be mediated by EP3 receptors. 25 Similarly, in knockout mice, it was concluded that the EP3 receptor plays a role in the IOP-lowering response to latanoprost, travoprost, and bimatoprost. 26 This is in contrast to what would be expected from in vitro receptor binding2,14 and functional assays, 2 in which FP-receptor binding and functional responses (e.g., phosphoinositide turnover) were at least two orders of magnitude greater than for EP3 receptors. The EP3 receptor has also been detected in trabecular cells in human donor eyes. 21
It was hypothesized that the FP+EP2+EP3 combination would give the greatest effect, based on individual agonist IOP-lowering studies conducted in monkeys by other investigators.23,24 Combined therapy with latanoprost+sulprostone+ butaprost decreased IOP more that latanoprost alone, but the IOP-lowering response was not as great as with the nonselective agonist, PGF2α-ie, in normotensive cynomolgus monkeys (Table 1). However, the addition of the EP1 agonist, 17PhTPGE2, to the combination of latanoprost+sulprostone + butaprost did reproduce the magnitude and time course of the IOP-lowering response to PGF2α-ie. Perhaps PG subtypes, which alone have no effect on IOP, may act in synergistic ways to enhance IOP reduction when combined with current IOP-lowering PG compounds. The combination of latanoprost+17PhTPGE2+butaprost was not tested, so it is not known if sulprostone is necessary in the formulation to reproduce the magnitude of the PGF2α-ie IOP-lowering response. Once-daily treatment with the other compounds should be investigated as well in order to possibly further optimize the IOP-lowering responses and minimize side effects. Also, there was no analysis of the combined formulation to determine if any alteration of the individual components occurred when they were combined that could have contributed to the response.
The EP4 agonist, ONO-123A, has been reported to increase outflow facility by 43% in monkeys after a single dose. 27 Another EP4 agonist, 3,7-dithia PGE1, did not relax primate ciliary muscle precontracted with carbachol. 28 PGF2α is known to effectively bind to EP4 receptors. 2 No further enhancement would be expected by the addition of an EP4 agonist, since the magnitude of the PGF2α-ie IOP lowering was achieved with the combination of the FP+EP1+EP2+EP3 agonists. * However, this does not exclude the possibility that the addition of the EP4 agonist to PGF2α-ie, or to the combination that equaled it, might produce IOP lowering beyond either of those. It is also possible that an EP4 agonist might be substituted for one of the others to produce the same, or more consistent, IOP-lowering response and to eliminate adverse side effects, such as were encountered in some instances with butaprost.
PGF2α-ie lowers IOP in vivo by primarily increasing uveoscleral outflow29,30 with variable, if any, effects on total out-flow facility and no apparent effects on trabecular outflow facility. 31 Similar morphologic changes of the uveoscleral pathway in monkeys have been shown after long-term treatment with EP2, EP3, and FP agonists. 32 Increased scleral permeability also seems of similar magnitude with PGF2α and latanoprost.33,34 Therefore, combined therapy may be useful for short-term enhancement of IOP reduction, but long-term effectiveness may be similar, regardless of the receptor subtypes stimulated.
This work was presented, in part, at ARVO 2006, Fort Lauderdale, Florida, April 30–May 4, 2006.
* EP4 agonists were not studied due to a legal impass regarding material transfer agreements.
This study was supported by the following: NIH/NEI EY002698; NEI P30 EY016665 (Core Grant for Vision Research); Research to Prevent Blindness, Inc., (New York, NY); unrestricted departmental and Physician-Scientist awards; the Ocular Physiology Research and Education Foundation; and the Walter Helmerich Chair from the Retina Research Foundation.