Re: 나이아신(비타민 B3) 약물역동학 - cMax 3시간, 반감기 5시간 -

[문형철]

beyond reason

나이아신을 효과적으로 복용하는 방법

50mg 나이아신을 10분단위로 3시간 연속복용

100mg 20분단위로 3시간 연속복용하기...

The metabolic profile of niacin is influenced by the rate of niacin administration. This study characterizes the effect of administration rate on the pharmacokinetics of niacin and its metabolites. Twelve healthy males were enrolled in an open-label, dose-rate escalation study and received 2000 mg niacin at 3 different dosing rates. Plasma was analyzed for niacin, nicotinuric acid, nicotinamide, and nicotinamide-N-oxide. Urine was analyzed for niacin and the metabolites nicotinuric acid, nicotinamide, nicotinamideN-oxide, N-methylnicotinamide, and N-methyl-2-pyridone5-carboxamide. Cmax and AUC0-t for niacin and nicotinuric acid increased with an increase in dosing rate. The changes observed in plasma nicotinamide and nicotinamide-N-oxide parameters, however, did not correlate to dosing rate. The total amount of niacin and metabolites excreted in urine was comparable for all 3 treatments. However, with the increase in dosing rate, urine recovery of niacin and nicotinuric acid showed a significant increase, whereas N-methyl-2-pyridone5-carboxamide and N-methylnicotinamide showed a significant decrease.

Niacin is used both in immediate-release (IR) and extended-release (ER) formulations for the treatment of dyslipidemia.1-5 Although efficacy appears to be related to the extent of niacin absorption, the rate of niacin administration is believed to affect the adverse event profile of niacin, probably by influencing its metabolic profile.2-5 Extended-release niacin produces lower flushing as compared to IR niacin while avoiding hepatotoxicity that the sustainedrelease niacin formulations can cause.6-8 Niacin undergoes extensive, saturable metabolism via 2 major pathways (Figure 1).2,4,9-12

One pathway (pathway 1) is via glycine conjugation with niacin to form nicotinuric acid (NUA). The other pathway (pathway 2) contributes to the formation of nicotinamide adenine dinucleotide (NAD). Pathway 1 is responsible for the flushing seen with niacin, whereas pathway 2 is responsible for hepatotoxicity.3,4 Although the specific metabolic steps in pathway 2 have not been completely elucidated, niacin appears to be first converted to nicotinamide (NAM), which is, in turn, rapidly methylated to N-methylnicotinamide (MNA) or conjugated to form NAD. The further metabolism of MNA has been defined more exactly: MNA is oxidized to N-methyl-2-pyridone-5-carboxamide (2PY) and to the minor metabolite N-methyl-4-pyridone3-carboxamide (4PY). Hydroxy nicotinamide (6HN) and nicotinamide-N-oxide (NNO) also appear to be minor niacin metabolites; however, it is unclear whether NNO is formed directly from niacin or indirectly from nicotinamide. Only very small quantities of NNO and 4PY have been found in humans, whereas niacin, MNA, 2PY, and NUA were the predominant compounds excreted following oral dosing with niacin.13 The objective of the study was to characterize the effect of niacin absorption rate on the plasma and urine pharmacokinetics of niacin and its metabolites.To achieve this, niacin solutions were administered orally at 3 different rates representing slow, intermediate, and fast absorption rates, and the plasma and urine pharmacokinetics of niacin and its metabolites were determined.

METHODS

Subjects

Twelve healthy, nonsmoking male subjects were enrolled in the study. The clinical study was conducted at Peninsular Testing Corporation (Miami, Fla) under the direction of the late Albert Cohen, MD. The protocol and the informed consent form were approved by an institutional review board from Peninsular Testing Corporation. Subjects included in the study had to be within 40 to 55 years of age, had to be 154 to 212 pounds, and had not donated blood 6 weeks prior to dosing. Subjects were selected based on the absence of any clinically significant findings from the medical history, physical examination, electrocardiogram (ECG), and clinical laboratory eval‎uations. All subjects confirmed abstinence from other medications, alcohol, tobacco, and caffeinated products throughout the study.

Study Design

This was an open-label, dose-rate escalation, crossover study. Each subject received the following treatments (described by administration rate), in order, each consisting of 2000 mg niacin in solution form:

Slow: 25 mg niacin aqueous solution administered every 10 minutes for 80 doses (13 hours and 10 minutes)

Intermediate: 50 mg niacin aqueous solution administered every 10 minutes for 40 doses (6 hours and 30 minutes)

Fast: 100 mg niacin aqueous solution administered every 10 minutes for 20 doses (3 hours and 10 minutes)

After receiving 2000 mg niacin in each treatment, each subject received 10 mL of water every 10 minutes up to 13 hours and 10 minutes after the initiation of dosing to maintain consistency between all treatments in the total amount of water administered. The study had a fourth treatment arm in which niacin ER tablets were administered; however, this treatment is not included here as the results were not relevant to the objective of this analysis. Dosing for each treatment was initiated at approximately 10 PM following a low-fat snack. Subjects were allowed to sleep between procedures but were required to remain seated upright in a chair or bed during all phases of dosing. Subjects could ambulate only to use the bathroom. Meals controlled for niacin content were administered at approximately 7 AM, noon, 5 PM, and 9:45 PM daily. No other medications were permitted during the study. Subjects were confined to the clinic during the 6-day study period of each treatment.

Pharmacokinetic Sampling

Blood samples were obtained within 1 hour prior to the start of dose administration and for approximately 17 hours after, with sampling times customized for each treatment. In period 1, 19 samples were obtained at predose and from 2 hours 5 minutes to 16 hours 35 minutes after starting to dose the slow treatment. In period 2, for the intermediate treatment, 18 samples were obtained at predose and from 35 minutes to 9 hours 55 minutes after the start of dosing. In period 3, 13 samples for the fast treatment were obtained at predose and from 35 minutes to 8 hours 5 minutes after the start of dosing. Blood samples were collected into 10 mL vacutainers containing sodium heparin, then centrifugated at 3000 rpm for 15 minutes in a refrigerated centrifuge to separate the plasma. Plasma samples were stored frozen at approximately –70°C until analysis. Urine was collected for 24 hours prior to and for 96 hours following each treatment. Urine was collected at the following time intervals: 24 to 18, 18 to 12, 12 to 6, and 6 to 0 hours before dosing and 0 to 6, 6 to 12, 12 to 18, 18 to 24, 24 to 48, 48 to 72, and 72 to 96 hours after starting administration of each dose. Aliquots for urine from each interval were stored frozen at approximately –20°C until analysis.

Bioanalysis

Plasma was analyzed for niacin, NUA, NAM, and NNO. Urine was analyzed for niacin, NUA, NAM, NNO, MNA, and 2PY. All assays were conducted using validated high-performance liquid chromatography (HPLC) methods with ultraviolet (UV) detection following liquid-liquid extraction using methods adapted from the literature.14-17 The plasma assay for niacin and NUA used a Waters C-18 column, whereas for the urine assay, the column used was a Waters PicoTag column; UV detection for plasma and urine was at 254 nm. The internal standard used in the assays was isonicotinic acid. The assay range for niacin and NUA was 0.20 to 25 µg/mL in plasma and 20 to 500 µg/mL in urine. Nicotinamide and NNO concentrations in plasma were determined using the same method, whereas NAM, NNO, and 2PY in urine were determined using the same method. The analytes were detected after separation using UV detector at 260 nm. The column used for separation was Phenomenex Inertsil 5 ODS, and the internal standard used in the assay was isonicotinamide. The assay range for NAM and NNO in plasma was 0.25 to 15 µg/mL, whereas that for NAM, NNO, and 2PY in urine was 5 to 500 µg/mL. The urine assay for MNA used a Beckman Ultrasphere ODS column and UV detection at 264 nm. The internal standard used was hydroxyethyltheophylline. The assay range for MNA was 2.5 to 150 µg/mL. The precision and accuracy of the assay, as measured by percent relative standard deviation of the various runs and percent difference from nominal values, were within 10% for all calibration standards and quality control samples. Quality control samples were run with each analysis to monitor assay performance.

Pharmacokinetic Analysis

Pharmacokinetic parameters were determined using WinNonlin Version 4.1 (Pharsight Corporation, Mountain View, Calif) and Microsoft Excel 2000 (Microsoft Corporation, Redmond, Wash). For niacin and metabolites (NUA, NAM, and NNO) in plasma, the following parameters were calculated: maximum observed plasma concentration (Cmax), time to Cmax (tmax), and area under the plasma concentration-time profile from time 0 to the last measurable concentration (AUC0-t). Concentrations below the limit of quantitation (LOQ) were treated as zero.

For niacin and metabolites (NUA, MNA, 2PY, NAM, and NNO) in urine, urinary recovery for each analyte over 96 hours was expressed as percent of niacin dose administered. In each case, the urine concentration data were corrected for baseline recovery and molecular weight of the analyte and then expressed as the percent of niacin dose administered. Statistical Analysis Demographic and statistical analyses were performed using SAS System for Windows, Version 6.11 (SAS Institute, Cary, NC). Comparisons between treatments for Cmax and AUC0-t of each analyte in plasma, as well as percent recovery of each analyte in urine and total percent recovery, were made using the Wilcoxon signed-rank test. Nonparametric comparisons between treatments were used instead of parametric comparisons due to the small sample size and the large amount of variability in the data. No statistical comparison was performed on tmax because the time over which the dose was administered was not constant between treatments, and tmax was a function of the length of dosing.

Safety Eval‎uation

During screening and at study termination, a physical exam, ECG, and laboratory eval‎uations (clinical chemistry, hematology, urinalysis) were conducted on all subjects. Subjects were monitored throughout the study for adverse events through questions from the clinic staff and were encouraged to volunteer any untoward effects. The severity of each adverse event was determined by the clinic staff based on direct observation and interview with the subject. The investigator judged the relationship of the adverse event to the study treatments.

RESULTS

Subject Disposition and Demographics

Twelve male subjects were enrolled and completed the study. The mean (range) age was 46 (41-51) years, the mean (range) weight was 168 (144-190) pounds, and the mean (range) height was 70 (67-72) inches. Seven subjects were Caucasian, 3 were Hispanic, and 2 were African American. Ten of the 12 subjects were of medium frame size, whereas 1 subject each was of small and large frame size.

Pharmacokinetic Eval‎uation

Plasma Data

Mean concentration-time profiles for plasma niacin, NUA, NAM, and NNO are shown in Figures 2 through 5. Plasma pharmacokinetic parameters and statistical results are provided in Table I. For niacin and NUA, Cmax and AUC0-t were significantly different for all treatment comparisons. The niacin data were highly variable, with intersubject percent coefficient of variation (%CV) for Cmax and AUC0-t at least 90% for the slow treatment. The variability of the data decreased as the dosing rate increased. The NUA data were considerably less variable than the niacin data, with %CVs for Cmax and AUC0-t in the range of 24% to 34%. Mean plasma NAM Cmax was significantly different for the slow treatment as compared to the intermediate and fast treatments. There were no differences in Cmax between the intermediate and fast treatments. The AUC0-t value for the slow treatment was not significantly different than the AUC0-t for the intermediate or fast treatment. However, the AUC0-t for the intermediate treatment was significantly different from the AUC0-t for the fast treatment. The NAM data were more variable than the NUA data and were similar in variability to the niacin data, with %CVs in the range of 49% to 84% for Cmax and AUC0-t. Similar to niacin, the variability in NAM data decreased as the dosing rate increased. The plasma NNO data were extremely sparse, with most subjects having concentrations below the limit of quantification regardless of the niacin input rate. There were no significant differences between any treatments for Cmax or AUC0-t, and the NNO data were extremely variable, with %CVs of 165% or greater in all cases.

Urine Data

Mean urine recovery data are listed in Table II and depicted in Figure 6.

Total recovery of niacin in urine as niacin and 5 metabolites combined was 73.2%, 74.6%, and 75.2% of niacin dose for slow, intermediate, and fast treatments, respectively. There was no significant difference in the percent recovery among the treatments. Niacin urine concentrations were below LOQ in the predose collection interval and at each collection interval for all subjects following administration of the slow treatment, except 1 subject who had a detectable quantity of niacin in the 6- to 12-hour postdose collection interval. However, almost all subjects excreted unchanged niacin in the urine in the treatments where niacin absorption rates were faster. The majority of niacin was excreted during the 0- to 6-hour postdose collection interval. The percentage of the niacin dose excreted as unchanged drug increased with increased dosing rates. Mean niacin recovery was 0.102%, 4.03%, and 12.1% of the dose for slow, intermediate, and fast treatments, respectively. Niacin recovery in all treatments was significantly different from each other. Nicotinuric acid, NAM, and NNO urine concentrations were below LOQ in the predose collection interval for all subjects, whereas MNA and 2PY urine concentrations were quantifiable in most of the urine samples collected during the predose collection intervals. All subjects excreted NUA, MNA, and 2PY following administration of all treatments, but NAM and NNO concentrations were not quantifiable in all subjects. The majority of NUA was excreted during the 0- to 6-hour postdose collection interval following administration of slow, intermediate, and fast treatments. The percentage of the niacin dose excreted as NUA increased in the treatments with increased dosing rates. Mean NUA recovery was 8.19%, 13.5%, and 19.9% of the dose for slow, intermediate, and fast treatments, respectively. All treatments were significantly different from each other in NUA recovery. In the majority of those subjects (4 of 5) who did have quantifiable NAM in the urine, it was not detected until at least the 6- to 12-hour postdose collection interval. The percentage of the niacin dose excreted as NAM appeared to be independent of dosing rate. Mean NAM recovery was 0.134%, 0.220%, and 0.0914% of the dose for slow, intermediate, and fast treatments, respectively. The percent recovery of NAM was not significantly different between any treatments. Nicotinamide-N-oxide urine concentrations were low and often only quantified in 1 collection interval. In general, NNO was not quantified until at least the 6- to 12-hour postdose collection interval. The percentage of the niacin dose excreted as NNO appeared to be independent of dosing rate. Mean NNO recovery was 0.421%, 0.712%, and 0.426% of the dose for slow, intermediate, and fast treatments, respectively. The percent recovery of NNO was not significantly different between treatments. The excretion of MNA was highest in either the 6- to 12-hour or the 12- to 18-hour postdose collection intervals. The percentage of the niacin dose excreted as MNA decreased in the treatments with increased dosing rate. Mean MNA recovery was 20.0%, 16.5%, and 11.9% of the dose for slow, intermediate, and fast treatments, respectively. All treatments were significantly different in MNA recovery from each other. The excretion rate of 2PY was highest in either the 12- to 18-hour or the 18- to 24-hour postdose collection intervals. Similar to MNA, the percentage of the niacin dose excreted as 2PY decreased as the dosing rate increased. Mean 2PY recovery was 44.3%, 39.6%, and 30.8% of the dose for slow, intermediate, and fast treatments, respectively. Recovery of 2PY in all treatments was significantly different from each other. 

Safety and Tolerability

Vital signs (sitting blood pressure, pulse rate, respiratory rate, and oral temperature) were measured before dosing and throughout the study and were normal for all subjects. Prior to each subject’s release from the study, a physical exam, ECG, and laboratory studies (clinical chemistry, hematology, urinalysis) were performed. No clinically significant changes from screening were noted in the end-of-study physical examinations or ECGs. Most subjects experienced an adverse event during the study. Of the 12 subjects receiving study medication, only 1 subject reported no adverse events. Overall, the most frequently occurring treatmentemergent adverse events were vasodilation (cutaneous flushing) and pruritus. Overall, the number of adverse events appeared to increase as the dosing rate increased because only 4 subjects reported adverse events with the slow treatment, whereas 10 and 11 subjects reported adverse events with the intermediate and fast treatments, respectively. Vasodilation appeared to be correlated to dosing rate as none of the subjects had vasodilation at the niacin 2.5-mg/min dosing rate (slow treatment), but 5 and 10 subjects reported vasodilation when the dosing rate was increased to 5 mg/min (intermediate treatment) and 10 mg/min (fast treatment), respectively. Pruritus was reported for 2, 5, and 4 subjects on the slow, intermediate, and fast treatments, respectively. However, no statistical comparison of adverse events between treatment groups was performed due to the small sample size. All adverse events were reported as mild in severity. No serious adverse events were reported.

DISCUSSION

Immediate-release niacin is associated with cutaneous flushing, whereas sustained-release formulations have been associated with hepatotoxicity.6-8 The rate and extent of the metabolite formation via the glycine conjugation pathway (pathway 1, NUA) following niacin administration have been shown to depend on the rate of niacin release from the niacin formulations following single-dose administrations of 500 mg niacin.18,19 However, the effect of rate on the NAM metabolic pathway (pathway 2) or at therapeutic doses of 2000 mg has not been characterized. The objective of this study was to characterize the single-dose pharmacokinetics of 2000 mg niacin in solution delivered orally at different input rates. The input rate of the intermediate treatment (5 mg/min) was twice that of the slow treatment (2.5 mg/min), and the input rate of the fast treatment (10 mg/min) was twice that of the intermediate treatment. Plasma Cmax and AUC0-t values were determined across treatments to compare the rate and extent of niacin absorption and metabolism. The amount of niacin and metabolites excreted in urine also measures the extent of niacin absorption. The data were not sufficient for any of the subjects to compute terminal half-life for NNO (all treatments) or niacin (slow treatment). Hence, AUC0-t was used to compare treatments instead of AUC0-∞. The plasma concentrations for niacin and NUA were at or close to the LOQ at the last sampling time for all 3 treatments. Also, for all analytes, the amount excreted in urine showed the same results as plasma AUC0-t, indicating that despite the different sampling times, AUC0-t results was a reasonable parameter for comparison across the treatments. The increase in dosing rate led to an increase in niacin Cmax, as expected, and also caused an increase in niacin AUC0-t, suggesting that either the amount of niacin absorbed increased or that the clearance of niacin decreased as dosing rate increased. Niacin is metabolized first to either NUA or NAM (Figure 1), and NAM is further metabolized to NNO, MNA, 2PY, and 4PY. The Cmax and AUC0-t values for NUA increased with an increase in the absorption rate. The Cmax values increased 2.1 times when the rate was doubled from 2.5 to 5 mg/min and increased 2.5 times when the rate was doubled from 5 to 10 mg/min. Plasma NAM and NNO, however, did not show ratedependent changes in Cmax and AUC. This suggests that metabolic clearance of niacin to NAM and NNO decreased, indicating that niacin metabolism via the NAM pathway (pathway 2) is saturated at higher dosing rates, whereas the NUA pathway is not. Niacin is available in the diet, and hence subjects were put on a niacin-restrictive diet during the study. Predose samples collected for 24 hours prior to dosing showed niacin, NUA, NAM, and NNO levels below the LOQ, whereas MNA and 2PY had detectable predose levels. The urine recovery data after dosing indicated that niacin, NUA, NAM, and NNO showed maximum excretion within 12 hours of dosing, whereas MNA and 2PY showed maximum excretion after 12 hours of dosing. The positive predose levels for MNA and 2PY can be explained by niacin from the diet before check-in being excreted as MNA and 2PY during the predose collection period. On average, at least 73% of the dose administered was recovered in urine as niacin, NUA, MNA, 2PY, NAM, and NNO, indicating that the change of the niacin rate of administration did not affect extent of absorption. As the drug was administered in solution, it can be assumed that the extent of absorption was complete as unmeasured analytes probably accounted for the remaining 27% of the administered dose. Although the amounts of NAM and NNO excreted in urine were not affected by the dosing rate, the amounts of MNA and 2PY excreted were found to decrease with the increase in dosing rate, demonstrating that the clearance of niacin via the NAM pathway was saturable. In contrast, the amount of NUA excreted increased as the dosing rate increased. This corroborates the plasma results that the metabolism of niacin to NUA is not saturated, but the NAM pathway is saturated at the concentrations reached in this study. The metabolism of niacin to NUA has a higher capacity than does the metabolism of niacin via the NAM pathway. The 3 solution treatments demonstrated that the AUC0-t of niacin, NUA, NAM, and NNO are poor metrics of niacin extent of absorption. The AUC0-t of niacin and NUA increased as the dosing rate increased, whereas the AUC0-t of NAM and NNO decreased as the dosing rate increased, even though the total administered dose of niacin was the same. Only the total urinary recovery of niacin and metabolites demonstrated consistency across the 3 treatments. This study, therefore, suggests that the total urinary recovery of niacin and its metabolites is a better metric for determining relative extent of niacin absorption. The lower variability of NUA Cmax (relative to that of niacin Cmax) and its close relationship to niacin administration rate suggest that NUA is a good surrogate to compare relative rates of niacin absorption. Vasodilation appeared to be correlated to dosing rate because increasing the dosing rate from 2.5 to 5 mg/min and 10 mg/min showed a proportional increase in the number of subjects who experienced cutaneous flushing. Because the increase in dosing rate resulted in both increased NUA exposure and cutaneous flushing, it can be argued that increasing NUA exposure increases the incidence of flushing. In conclusion, an increase in the niacin dosing rate led to an increase in total exposure to niacin and NUA, even though the total dose administered was the same. The overall amount of niacin and its metabolites excreted in the urine was not affected by the dosing rate of the solutions, although the percentage of each metabolite excreted in the urine was affected. As the dosing rate increased, the percentage of the dose excreted as unchanged niacin and NUA increased, and the percentage excreted as MNA and 2PY decreased, whereas the percentage excreted as NAM and NNO did not change.

REFERENCES

1. Niaspan [prescribing information]. Cranbury, NJ: Kos Pharmaceuticals, Inc; 2005. 2. Pieper JA. Overview of niacin formulations: differences in pharmacokinetics, efficacy, and safety. Am J Health Syst Pharm. 2003;60:S9-S14. 3. Pieper JA. Understanding niacin formulations. Am J Manag Care. 2002;8:S308-S314. 4. Piepho RW. The pharmacokinetics and pharmacodynamics of agents proven to raise high-density lipoprotein cholesterol. Am J Cardiol. 2000;86:35L-40L. 5. Stern RH, Freeman D, Spence JD. Differences in metabolism of time-release and unmodified nicotinic acid: explanation of the differences in hypolipidemic action. Metabolism. 1992;8:879-881. 6. Morgan JM, Capuzzi DM, Guyton JR. A new extended-release niacin (Niaspan): efficacy, tolerability, and safety in hypercholesterolemic patients. Am J Cardiol. 1998;82:29U-41U. 7. Carlson LA. Niaspan, the prolonged release preparation of nicotinic acid (niacin), the broad-spectrum lipid drug. Int J Clin Pract. 2004;58:706-713. 8. McCormack PL, Keating GM. Prolonged-release nicotinic acid: a review of its use in the treatment of dyslipidaemia. Drugs. 2005; 65:2719-2740. 9. Fumagalli R. Pharmacokinetics of nicotinic acid and some of its derivatives. In: Gey KF, Carlson LA, eds. Metabolic Effects of Nicotinic Acid and Its Derivatives. Bern, Switzerland: Hans Huber Publishers; 1971:33-49. 10. Carlson LA, Hanngren A. Initial distribution in mice of H3- labeled nicotinic acid studied with autoradiography. Life Sci. 1964;3: 867-871. 11. Mrochek JE. Metabolic responses of humans to ingestion of nicotinic acid and nicotinamide. Clin Chem. 1976;22:1821-1827. 12. Ding RW, Kolbe K, Merz B, Vries JD, Weber E, Benet ZB. Pharmacokinetics of nicotinic acid-salicylic acid interaction. Clin Pharmacol Ther. 1989;46:642-647. 13. Felsted R, Chaykin S. N-Methylnicotinamide oxidation in a number of mammals. J Biol Chem. 1967;242:1274-1279. 14. Shibata K. Simultaneous measurement of nicotinic acid and its major metabolite, nicotinuric acid, in blood and urine by reversedphase high performance liquid chromatography. Agric Biol Chem. 1988;52:2973-2976. 15. Hengen N, Seiberth V, Hengen M. High-performance liquidchromatographic determination of free nicotinic acid and its metabolite, nicotinuric acid, in plasma and urine. Clin Chem. 1978;24:1740-1743. 16. Carter EG. Quantitation of urinary niacin metabolites by reversedphase liquid chromatography. Am J Clin Nutr. 1982;36:926-930. 17. Shibata K. Simultaneous micro-determination of nicotinamide and its major metabolites, N-methyl-2-pyridone-5-carboxamide and N-methyl-4-pyridone-3-carboxamide, by high-performance liquid chromatography. J Chromatogr. 1988;424:23-28. 18. Neuvonen PJ, Roivas L, Laine K, Sundholm O. The bioavailability of sustained release nicotinic acid formulations. Br J Clin Pharmacol. 1991;32:473-476. 19. Figge HL, Figge J, Souney PF, Sacks FM, Shargel L, Janosik JE. Comparison of excretion of nicotinuric acid after ingestion of two controlled release nicotinic acid preparations in man. J Clin Pharmacol. 1988;28:1136-1140.