Clinical trial: single- and multiple-dose pharmacokinetics of polyethylene glycol (PEG-3350) in healthy young and elderly subjects

Transcription

1 Alimentary Pharmacology & Therapeutics Clinical trial: single- and multiple-dose pharmacokinetics of polyethylene glycol (PEG-3350) in healthy young and elderly subjects R. W. PELHAM*, L. C. NIX, R.E.CHAVIRA, M.VB.CLEVELAND*&P.STETSONà *Braintree Laboratories Inc., Braintree, MA, USA; MDS Pharma Services, Lincoln, NE, USA; àwil Research Laboratories, LLC, Ashland, OH, USA Correspondence to: Dr R. W. Pelham, Braintree Laboratories Inc., 60 Columbian Street, West, Braintree, MA 02185, USA. Publication data Submitted 11 December 2007 First decision 6 January 2008 Resubmitted 25 March 2008 Resubmitted 25 April 2008 Accepted 25 April 2008 Epub OnlineAccepted 29 April 2008 SUMMARY Background The pharmacokinetics of polyethylene glycol 3350 (PEG-3350) have not been fully described because of lack of a sufficiently sensitive analytical method. Aim To describe the pharmacokinetics of PEG-3350 in humans. Methods A highly sensitive, high performance liquid chromatography with mass spectrometry (HPLC MS MS) method was developed for PEG-3350 in urine, plasma and faeces with quantification limits of 30 ng ml, 100 ng ml and 500 lg g respectively. Noncompartmental pharmacokinetics methods were used and the effects of gender, age, renal status and dosing frequency were examined after the oral administration of 17 g to healthy volunteers. Results Peak PEG-3350 plasma concentrations occurred at 2 4 h and declined to nonquantifiable levels usually within 18 h after single and multiple doses, with a half-life of about 4 6 h. Steady state was reached within 5 days of dosing. Mean urinary excretion of the administered dose ranged from 0.19% to 0.25%. Age, gender or mild kidney impairment did not alter the pharmacokinetics of PEG Mean faecal excretion of the administered dose was 93% in young subjects. Conclusions For the first time, a highly sensitive assay allowed comprehensive pharmacokinetics studies of PEG-3350 in humans. These studies confirmed that orally administered PEG-3350 is minimally absorbed, rapidly excreted and primarily eliminated via faeces. Aliment Pharmacol Ther 28, ª 2008 Braintree Laboratories, Inc. doi: /j x

2 CLINICAL TRIAL: PEG-3350 PHARMACOKINETICS IN HUMANS 257 INTRODUCTION Polyethylene glycols (PEGs) have extensive medical applications, many with significant relevance to gastroenterology. Although the absorption of the lower molecular weight PEGs provides insight into intestinal permeability, 1, 2 the larger PEGs have found enduring and expanding uses. In 1984, PEG with a molecular weight of 3350 (PEG-3350) plus electrolytes emerged in the US as a lavage solution for bowel cleansing prior to colonoscopy. 3 In 1999, PEG-3350 was cleared by the US Food and Drug Administration (FDA) for use as a stand-alone therapy for occasional adult constipation. PEG-3350 has also benefited patients with chronic constipation 4, 5 and in 2006 received OTC status in the US for the treatment of occasional constipation. In the intervening years, high molecular weight PEGs (>20 000) were covalently linked to interferons (IFNs), effectively extending the IFN half-lives. In comparison to IFN with ribavirin, pegylated IFN with ribavirin resulted in a greater proportion of subjects with sustained virological response and normalization of hepatic enzymes. 6 It is surprising that only a few published studies address PEG s absorption, distribution, metabolism and excretion. PEG is not known to be metabolized in humans and there are no known pharmacological actions or confirmed toxicity resulting from the limited absorption. There are no published articles on the blood levels of (nonradioactive) PEG-3350 and its kinetics after its administration to any species, by any route. Previous pharmacokinetic (PK) studies have only measured PEG in urine or faeces, leaving to speculation its bioavailability and PK in blood. 1, 2, 7 9 Radiolabelled congeners of PEGs had been synthesized, in some cases adding non-peg ligands in the process and thereby complicating the interpretation of results. 10 When used for constipation or for bowel lavage, the dose of PEG delivered by mouth ranges from 17 to 420 g. About 0.1% of the dose appears in urine after 40 or 240 g of PEG , 7, 8 while most is (90 100%) excreted in faeces. 8, 9 This report focusses on the PK of PEG-3350 in three separate studies involving single-dose administration in non-elderly healthy subjects, multiple doses in non-elderly healthy subjects, and multiple doses in elderly and young healthy subjects. These studies were designed to characterize plasma PK and determine the role of urinary and faecal excretion, and were made possible by development and validation of an ultra sensitive high performance liquid chromatography with mass spectrometry (HPLC MS MS) method. MATERIALS AND METHODS Study designs The single-dose study was an open-label, PK evaluation of PEG-3350 in non-elderly (ages <65 years) healthy adult subjects following a single oral dose of 17 g PEG-3350 (MiraLAX polyethylene glycol 3350, NF powder for solution; Schering-Plough Healthcare Products, Kenilworth, NJ, USA). The multiple-dose study was an open-label, PK evaluation of PEG-3350 in non-elderly healthy adult subjects following an oral dose of 17 g of PEG-3350 once daily for 7 days. The effect of age on the PK of PEG-3350 was evaluated following multiple-dose administration of 17 g PEG once daily for 7 days in healthy subjects years of age (young population) and ages 65 years (elderly population). All doses were administered in 250 ml of water at room temperature to subjects who had fasted for 10 h. The subjects resided in a phase 1 unit for the duration of the studies, which were conducted according to Good Clinical Practice regulations. All study protocols were cleared by the MDS Pharma Services Institutional Review Board, which also approved the Informed Consent Form. All subjects provided written informed consent prior to study entry. Pharmacokinetic sampling In the single-dose study, blood samples were collected prior to dosing and at set times through 48-h postdose. Urine samples were collected prior to dosing and in fixed intervals over the next 48 h. In the multipledose study, blood samples were collected prior to dosing and at set times during the 24-h dosing interval on days 1, 5 and 7. Urine samples were collected prior to dosing on day 1 and in fixed intervals during the 24-h dosing interval on days 1, 5 and 7, with an additional collection for the 24- to 36-h period after dose 7. Predose samples were obtained prior to the dose on day 1 of the age-effect study, and sampling times for plasma and urine on day 7 were similar to the multiple-dose study. However, urine sample was collected for 60 h following the last dose. All faeces samples were collected from young subjects in the age-effect

3 258 R. W. PELHAM et al. study in 24-h intervals from day 1 up to the morning of day 11. Pharmacokinetic analysis HPLC MS MS methods in the positive electrospray ionization (ESI+) mode were developed and used for the determination of PEG-3350 concentration in human plasma, urine and faeces. The plasma, urine and faeces methods were validated over the following PEG-3350 concentration ranges: ng ml for plasma; ng ml for urine and lg g for faeces. Intra- and intersession assay accuracy and precision, as assessed with quality control samples spanning the calibration concentration ranges, met the acceptance criteria (relative standard deviation (RSD) 15%; % relative error (RE) within 15%). At the lower limit of quantitation (LLOQ) and for dilutional quality control (QC) samples of faeces 20% and within 20% were considered acceptable because of the low levels in some blood samples and high dilutions of faeces (1:50 or more) respectively. Sample preparation. Meta-phosphoric acid was added to 1-mL human plasma to precipitate plasma proteins. PEG-3350 was extracted from the deproteinized plasma with chloroform. Human urine (1 ml) was also extracted with chloroform. The chloroform extracts were evaporated to dryness, reconstituted with mobile phase and injected for HPLC MS MS analysis. For the determination of PEG-3350 concentration in human faeces, deionized water was added to the faecal samples prior to homogenization. Aliquots of the homogenized samples were further diluted with water with the final dilution conducted with acetonitrile:water (1:1, v v). The diluted samples were extracted with chloroform. The chloroform extracts were evaporated to dryness, reconstituted and injected for HPLC MS MS analysis. HPLC MS MS. The HPLC MS MS system used for the PEG-3350 assays included an Agilent 1100 liquid chromatograph and a Micromass Quattro Ultima tandem quadrupole mass spectrometer (Waters Medical Instruments, Rochester, MN, USA). The analytical column for the plasma and urine assays was a SUPELCO TSK gel (30 cm 7.8 mm, 6-lm particle size; Sigma-Aldrich Corp., St Louis, MO, USA) column with a TSK gel PWXL guard column (4.0 cm 6.0 mm). The mobile phase consisted of 1.0-mM ammonium acetate in 1:1 (v v) acetonitrile:deionized water. The retention time for PEG-3350 was approximately 10 min. For the determination of PEG-3350 in human faeces, the analytical column used was a Zorbax SB-C18 ( mm, 5-lm particle size; Agilent Medical Technologies, Santa Clara, CA, USA) column. A gradient elution (30 95% B) over 7 min with mobile phase A (1.0-mM ammonium acetate) and B (1.0-mM ammonium acetate in 90:10 (v v) acetonitrile:water) resulted in a PEG-3350 retention time of approximately 5 min. The mass transitions monitored for quantification of PEG-3350 in plasma and faeces were: m z 700 (77-mer + 5NH4) + 5 fi 45; m z 726 (80-mer + 5NH4) + 5 fi 45; m z 735 (81- mer + 5NH4) + 5 fi 45; m z 744 (82-mer + 5NH4) + 5 fi 45; m z 753 (83-mer + 5NH4) + 5 fi 45 and m z 761 (84-mer + 5NH4) + 5 fi 45. The mass transitions monitored for quantification of PEG-3350 in urine were: m z 691 (76-mer + 5NH4) + 5 fi 45; m z 700 (77-mer + 5NH4) + 5 fi 45 and m z 735 (81-mer + 5NH4) + 5 fi 45. The PEG-3350 peak area and the theoretical concentrations of the calibration samples were fit to an lnquadratic function, excluding the origin. Pharmacokinetic parameter calculations. The area under the curve to the last quantifiable concentration (AUC 0 t ), the maximum observed concentration (C max ) and the time to maximum concentration (T max ) were determined from plasma PEG-3350 concentration vs. time profiles. AUC 0 t was estimated using the linear trapezoidal method. When possible, the terminal elimination rate constant (k z ) and terminal half-life (t ½ ) were determined. Following a single or first dose, AUC 0 was determined as AUC 0 t + C last k z, where C last is the last measured concentration. After multiple dosing, the area under the curve for the 24-h dosing interval, AUC 0 s was obtained as AUC 0 t +(C last C 24 ) k z when T last was <24 h. In cases where k z was not calculable, the parameters t ½, AUC 0 and AUC 0 s were reported as missing. Urine concentrations and volumes were used to determine amount of PEG-3350 excreted in the urine (X u ), average excretion rate per interval (DX u Dt) and the per cent of dose excreted in the urine (%D u ). In addition, the k z and t ½ were determined from the urinary excretion rate vs. midpoint time data, as appropriate. Renal clearance (CL r ) was determined as the X u divided by AUC in plasma for the same time intervals. Faecal concentrations and weights were used to calculate the amount of PEG-3350 recovered in the faeces (X f ) and per cent of total dose recovered in the faeces (%D f ) over the 7 days of dosing plus 3 days of follow-up.

4 CLINICAL TRIAL: PEG-3350 PHARMACOKINETICS IN HUMANS 259 Statistical analysis Results from the single-dose study were summarized using descriptive statistics. The Wilcoxon signed-rank test (a = 0.05) was used to compare the C max, T max, AUC, and 24-h X u between the study days (days 5 vs. 1, days 7 vs. 1 and days 7 vs. 5) of the multiple-dose study in non-elderly subjects. The nonparametric test was chosen because the distribution of the data appeared skewed because of a few subjects with exceptionally high values. A general linear model with the fixed-effect factor of age was used for the comparison of steady-state PK parameters C max, AUC 0 t, AUC 0 s, 24-h X u and CL r from the age-effect study. The C max, AUC and CL r parameters were normalized by body weight and all parameters were log-transformed prior to analysis. The percentage mean ratios (elderly young) were reported for C max, AUC, X u and CL r, and 90% CI were reported for exposure parameters C max, AUC, X u (calculated from the ANOVA model). The differences between the parameters at the a = 0.05 level were also reported. The Wilcoxon rank-sum test was used to compare the t ½ and T max between the age groups (a = 0.05). RESULTS Single-dose Pharmacokinetics Six subjects (three females, five Caucasians and one African-American) with a mean age of 54 years (range 37 61) and a mean weight of 82.3 kg (range ) participated. The arithmetic means and standard deviations of PEG-3350 plasma and urine PK parameters following single 17-g dose of PEG-3350 administration are summarized in Table 1 and plasma PEG-3350 concentrations vs. time curves are presented in Figure 1. PEG-3350 was quantifiable in plasma for h. Peak concentrations (C max ) ranged from 412 to 801 ng ml and were observed between 2- and 4-h postdose. AUC 0 t ranged from 3749 to 6311 ng h ml. The terminal elimination rate constant (k z ) could not be determined from the plasma data because of a poorly defined terminal elimination phase. The mean t ½ and k z for PEG-3350, as determined from the urinary excretion data for four of the six subjects, were 5.83 h and per hour respectively. For these same subjects, the mean %D u over 48 h was 0.19%. Therefore, more than 99% of the urinary excretion should be accounted for by 48 h based on the t ½. Table 1. Single-dose PEG-3350 pharmacokinetics in non-elderly healthy subjects Parameters Multiple-dose Pharmacokinetics n Arithmetic mean s.d. Range Plasma C max (ng ml) T max (h) AUC 0 t (ng*h ml) Urine t ½ (h) k z (1 h) X u (mg 48 h) %D u (per 48 h) %D u (over 0 ) AUC 0 t, area under the curve to the last quantifiable concentration; C max, maximum observed concentration; T max, time to maximum concentration; k z, terminal elimination rate constant; t ½, terminal half-life; X u, amount of PEG-3350 excreted in the urine; %D u, per cent of dose excreted in the urine. Fourteen subjects (seven females, eight Caucasians, four African-Americans, one Hispanic and one Asian) with a mean age of 44 years (range 24 61) and a mean weight of 76.8 kg (range ) participated in this study. The plasma and urine PEG-3350 data were quite variable and median values for PK parameters, C max, AUC and 24-h X u for day 7 were lower than the means because a few subjects had exceptionally high values. As the distribution of the data appeared skewed, median and range values were provided in Table 2. The plasma PEG-3350 concentrations vs. time curves are presented in Figure 2. The average half-life ranged from 4 to 7 h. Based on this half-life, very little accumulation would be expected with single daily doses and steady state would be expected after 2 days. The PEG-3350 exposure between days 5 and 1 was similar, with <15% difference in median AUC 0 s, <20% difference in median amount excreted over 24 h and nearly identical median C max values (P > 0.1 for all parameters). Considerable variability (>48% CV) was observed in these PK parameters. The AUC 0 s on day 7 was compared to the AUC 0 s on day 5 to assess whether steady state was reached. There was no significant difference in AUC 0 s on these days. The median X u over 24 h increased 36% between days 5 and 7 (P = 0.01). However, there was no

5 260 R. W. PELHAM et al. 600 Plasma PEG-3350 Concentration (ng/ml) Hours from Dosing Figure 1. Mean plasma PEG concentrations vs. time in six non-elderly (ages <65 years) healthy adult subjects following a single 17-g dose of PEG significant difference in median X u between days 1 and 7, and there was a 24% decrease in median X u from days 1 to 5. Although exposure did not increase from day 1 to 5, an increasing trend in exposure was observed from day 5 to 7, where the increase in C max and AUC 0 s was 31% and 35%. Age effect Twenty-four subjects (12 females, 15 Caucasians, 8 Hispanics and 1 African-American) participated in this study. Twelve were elderly, with a mean age of 72 years (range 65 79) and a mean weight of 71.4 kg (range ) while 12 were young with a mean age of 28 years (range 20 39) and a mean weight of 69.1 kg (range ). The PEG-3350 PK parameters and statistical results for young and elderly subjects are presented in Table 3 and the plasma PEG-3350 concentrations vs. time curves are presented in Figure 3. There were no statistically significant differences in the PEG-3350 plasma and urine PK parameters between the elderly and young subjects (P > 0.05 for all parameters). Mean total PEG-3350 exposure, adjusted for body weight and based on geometric mean AUC 0 s from the ANOVA, was 21% higher in the elderly. Mean AUC 0 t was approximately twofold higher in the elderly. The greater difference in AUC 0 t was because of one young subject with no detectable plasma concentrations of PEG Mean peak exposure (C max ) was 35% higher in the elderly. PEG-3350 exposure was highly variable, with considerable overlap in individual parameter values between the two groups. The mean plasma half-life of PEG-3350 was approximately 5 h in both groups (N.S.). Median T max (2 h) was also not different between the two groups. Mean X u over 24 h and mean CL r of PEG-3350 based on geometric means were similar (difference of 20%) in the two groups and no statistically significant differences were found. The cumulative %D u did not differ between young (0.23%) and elderly (0.25%) subjects. In eight of 12 elderly subjects, the estimated creatinine clearance at baseline ranged from 50 to 80 ml min, which is classified as mild renal impairment by FDA criteria (FDA Group 2 Mild Kidney Impairment). 11 All subjects had a serum creatinine within the lab normal range of mg dl at screening. No significant relationship between PEG-3350 exposure (AUC 0 s ) and creatinine clearance was observed. Gender The PEG-3350 PK parameters and statistical results for females and males from the multiple-dose and ageeffect studies are presented in Table 4. Mean PEG exposure, based on C max and AUC, appeared higher in females. With the exception of C max in the young age group, no statistically significant differences were found in the comparison between females and males (a = 0.05). Variability was high and the sample sizes were small, both of which contributed to very wide 90% confidence intervals. Faecal excretion Faecal excretion of PEG-3350 was determined for the young subjects in the age-effect study (Table 5). The subjects received a total of 119-g PEG-3350 over 7 days. An average of 108-g PEG-3350 (93.3% of the dose) was excreted in the faeces over 240 h (first dose through 3 days after the last dose). The cumulative X f

6 CLINICAL TRIAL: PEG-3350 PHARMACOKINETICS IN HUMANS 261 Table 2. Multiple-dose PEG-3350 pharmacokinetics in non-elderly healthy subjects Day 1 Day 5 Day 7 Parameters n Median s.d. Range n Median s.d. Range n Median s.d. Range Plasma C max (ng ml) Tmax (h) AUC0 t (ng*h ml) * * AUC 0 s (ng*h ml) * t½ (h) kz (1 h) Urine t½ (h) kz (1 h) X u (mg 24 h) %Du (per 24 h) AUC0 t, area under the curve to the last quantifiable concentration; Cmax, maximum observed concentration; Tmax, time to maximum concentration; kz, terminal elimination rate constant; t ½, terminal half-life; AUC 0 s, area under the curve for the 24-h dosing interval; X u, amount of PEG-3350 excreted in the urine; %D u, per cent of dose excreted in the urine. * Significantly different compared to day 1 (P < 0.05). Significantly different compared to day 5 (P < 0.05).

7 262 R. W. PELHAM et al. Plasma PEG-3350 Concentration (ng/ml) Time (hr) Figure 2. Mean plasma PEG concentrations vs. time in 14 non-elderly (ages <65 years) healthy adult subjects on days 1, 3 and 7 following once daily 17 g of PEG-3350 for 7 days. Table 3. Steady-state PEG-3350 pharmacokinetics for young and elderly healthy subjects Parameters Elderly subjects Young subjects n Mean s.d.* n Mean s.d.* % Mean ratio 90% CI P-value Plasma C max (ng ml kg) T max (h) AUC 0 t (ng*h ml kg) AUC 0 s (ng*h ml kg) t ½ (h) k z (1 h) Urine t ½ (h) k z (1 h) CL r (ml min kg) X u (mg 24 h) %D u (per 24 h) AUC 0 t, area under the curve to the last quantifiable concentration; C max, maximum observed concentration; T max, time to maximum concentration; k z, terminal elimination rate constant; t ½, terminal half-life; AUC 0 s, area under the curve for the 24-h dosing interval; X u, amount of PEG-3350 excreted in the urine; %D u, per cent of dose excreted in the urine; CL r, renal clearance. * Median and range are presented for T max and plasma t ½. % Mean ratio, 90% CI and P-value are from the ANOVA. P-values for T max and t ½ are from the Wilcoxon rank-sum test. ranged from 65 to 143 g and the %D f ranged from 56% to 123%. Within the first 24 h after the first dose, 6.35 g (38.4%) were excreted in the faeces. DISCUSSION Three studies described herein report the first comprehensive PK evaluation of PEG-3350 in any species. The effects of single and multiple oral doses of 17 g, and the effects of gender and age were examined. The sensitivity of the analytical method allowed the detection of PEG-3350 concentrations in serum and urine even after a single oral dose of only 17 g. Overall, the results attest to the accuracy of prior inferences, based on urinary and faecal excretion, 2, 7 9, 12, 13 that PEG is poorly absorbed. The percentage of administered PEG-3350 dose excreted in the urine by our subjects was comparable to that reported when g of PEG was administered. These new results allow quantification of the systemic exposure, estimation of

8 CLINICAL TRIAL: PEG-3350 PHARMACOKINETICS IN HUMANS 263 Figure 3. Mean plasma PEG concentrations vs. time in 12 young (squares) and 12 elderly (circles) healthy adults following once daily 17-g doses of PEG-3350 for 7 days. Plasma PEG-3350 Concentration (ng/ml) Time (Hours) Table 4. Steady-state PEG-3350 pharmacokinetics for males and females Males Females Parameters n Mean s.d. n Mean s.d. % Mean ratio 90% CI P-value Non-elderly C max (ng ml) AUC 0 t (ng*h ml) AUC 0 s (ng*h ml) X u (mg 24 h) Young C max (ng ml kg) AUC 0 t (ng h ml kg) AUC 0 s (ng h ml kg) X u (mg 24 h) Elderly C max (ng ml kg) AUC 0 t (ng h ml kg) AUC 0 s (ng h ml kg) X u (mg 24 h) AUC 0 t, area under the curve to the last quantifiable concentration; C max, maximum observed concentration; AUC 0 s, area under the curve for the 24-h dosing interval; X u, amount of PEG-3350 excreted in the urine. renal clearance and comparison of the PK parameters according to selected demographic categories. They also permit a calculation of the mass balance of the administered dose. The three PK studies included a single-dose study, a 7-day multiple-dose study in non-elderly subjects, and a 7-day multiple-dose study comparing young and elderly subjects. It can be seen from Tables 1 4 that there was little difference among the means medians of the key PK parameters from study to study. Overall, there were no differences in PK parameters with respect to gender, age or mild kidney impairment. Most of the subjects were Caucasian; hence, conclusions about the effect of race cannot be made. Based on comparisons of PK between days 1 and 5 of multiple daily dosing of PEG-3350 in non-elderly subjects, steady state was reached by day 5. A trend of increasing PEG-3350 exposure with multiple dosing was observed from days 5 to 7. According to PK theory, once steady state is reached, no further accumulation should occur. From days 5 to 7 of PEG-3350 administration, AUC 0 s decreased in five subjects and increased

9 264 R. W. PELHAM et al. Table 5. Faecal PEG-3350 parameters in young healthy subjects Parameters n Mean s.d. X f (g 240 h) %D f (per 240 h) X f, amount of PEG-3350 recovered in the faeces; %D f, per cent of total dose recovered in the faeces. in the remaining nine subjects. Median C max, AUC 0 s and X u over 24 h increased from 31% to 36%. This trend is not explained by the drug disposition. PEG was not detected in a majority of subjects predose samples on days 5 and 7. The remaining subjects had values below the quantifiable limit of the assay, with the exception of one subject with a predose value of 158 ng ml on day 5. As the serum PEG assay has a relative error of 15%, the variability in PEG absorption could have been primarily responsible for the apparent increase in median PEG-3350 exposure. This large variability in PK deserves further comment. For example, even though the dose was administered to fasting subjects at the same time of the day, the T max value varied by 36-fold (from 0.5 to 18 h) and the C max varied by 20-fold (from 106 to 2152 ng ml). C max was more reproducible within a subject, with CV percentage ranging from 10 to 101 (median 46%) for three repeated measures. Variability in exposure is not unusual in a drug that is very poorly absorbed. The amount of PEG-3350 excreted in the urine after a single 17-g dose was approximately 29 mg (0.17%) over 48 h. Given that this and other published studies 8, 9 have estimated nearly 100% excretion of PEG-3350 in the faeces, the 29 mg is a good estimate of the amount absorbed. The mechanism by which PEG-3350 is absorbed is not known. Because of its size, it is probably not passive transport, but rather an active process such as pinocytosis. With only 29 mg being absorbed, individual differences in gastrointestinal absorption would have a greater impact, resulting in high variability. Our subjects did not have apparent gastrointestinal diseases that could have affected absorption, and the differential variability of renal clearance of PEG-3350 did not appear to be responsible for the individual variability. Despite the wide variations, plasma PEG-3350 levels were unquantifiable in most subjects by h, thereby limiting the extent of exposure. Others have also reported that variability of PEG urinary excretion in healthy volunteers exceeded 50-fold and that patients with Crohn s disease excrete almost ten times greater amounts of high molecular weight PEGs than do controls. 12, 13 These observations underscore the potential usefulness of the measurement of blood levels of PEG-3350 in the identification of disease-related differences in motility and or permeability, as others have sought in studies of urinary excretion of various molecular weight PEGs in different disease states. 2, 9, 12, 13 A further observation that was rather surprising is the rapid appearance of PEG-3350 in faeces. Approximately 38% of the single dose of 17 g appeared in the faeces in the first 24 h and this may be in part related to the action of PEG-3350 to improve bowel motility. The rapid faecal elimination corroborates the clinical observation that a substantial proportion of patients obtain relief from their constipation within 1 3 days of using PEG The recovery of PEG from faeces, even up to 6 days after dosing, was variable. In three subjects, the amount of PEG recovered was 56 65%, which is lower than can be expected from the assay precision. This low recovery may indicate a relatively slower transit of PEG through the GI tract of some subjects. A higher recovery of the administered dose from those subjects with low PEG excretion could have been achieved, if the collection period had been longer or, if whole gut lavage had been used to collect all faeces. 9 The cumulative faecal excretion ranged from 65 to 143 g and the percentage of dose excreted ranged from 56% to 123%. In seven of the 11 subjects, 90% of the dose or greater was excreted in the faeces. Five (5) subjects had values >100% ( %) for percentage of dose excreted in the faeces, possibly because of some imprecision in the extraction dilution procedures. The standard operating procedure required that two-thirds of the high QC samples ( lg g), which required a 50-fold dilution prior to assay, to be within 20% of the nominal concentration. Only one subject s apparent faecal PEG recovery exceeded the 20% relative error that was allowed for the assay. As the subjects did not consume additional sources of PEG during their confinement, the apparent recovery of more than 100% of the dose administered in this and four of the other subjects most likely represents dilution and assay variability. Four (4) subjects had a lower faecal cumulative amount excreted (65 86 g) compared with the

10 CLINICAL TRIAL: PEG-3350 PHARMACOKINETICS IN HUMANS 265 remaining subjects ( g). Accurate and complete recovery of faeces from human subjects is difficult if not impossible even under the conditions of our study. There were no observed or reported protocol deviations, such as missed samples, that could account for the lower excretion. The subjects with lower excretion did not have fewer bowel movements than the other subjects. However, the three subjects with the lowest excretion had lower total faecal weights compared with the total faecal weights of other subjects. The total weights for these three subjects were 644, 849 and 840 g respectively; whereas the weights for the remaining subjects ranged from 1034 to 1817 g. The systemic exposure of PEG-3350 was not lower in these three subjects compared to the remaining subjects. The plasma AUC 0 t for these three subjects ranged from 1665 to 5555 ng h ml, whereas the values for the remaining subjects ranged from 50.5 to 6001 ng h ml. The amount excreted in the faeces on the fourth day after the last dose (216- to 240-h collection interval) ranged from to 3.47 g compared to a total nominal dose of 119 g, indicating that faecal excretion was nearly complete. Even with this consideration, our data are consistent with previous observations that the faeces represent the major route of PEG-3350 excretion. 8, 9 These studies provide much insight into the gastrointestinal absorption and PK of PEG-3350 following administration as a laxative. The studies confirm low blood levels, minimal absorption (<0.28%) and lack of substantial accumulation of PEG-3350 on multiple dosing regardless of age or gender. ACKNOWLEDGEMENTS We wish to thank Nichole R. Myers, BA, WIL Research Laboratories, LLC and her staff for the PEG-3350 analyses. These studies would not have been completed without the excellent Phase 1 operations staff and the cooperation of subjects at MDS Pharma Services. MiraLAX is a registered trademark of SCHERING-PLOUGH HEALTHCARE PRODUCTS, INC. Declaration of personal interests: R. W. Pelham and M. vb Cleveland are employees of Braintree Laboratories. L. Nix and R. Chavira are employees of MDS Pharma Services. P. Stetson is an employee of WIL Research, LLC. Declaration of funding interests: This study and the preparation of this paper were funded by Braintree Laboratories, Inc. Initial data analyses were undertaken by employees of MDS Pharma Services and received funding from Braintree Laboratories, Inc. WIL Research LLC developed and applied the HPLC MS MS method. REFERENCES 1 Chadwick VS, Phillips SF, Hofmann AF. Measurements of intestinal permeability using low molecular weight polyethylene glycols (PEG 400). I. Chemical analysis and biological properties of PEG 400. Gastroenterol 1977; 73: Ryan CM, Yarmush ML, Tompkins RG. Separation and quantitation of polyethylene glycols 400 and 3350 from human urine by high-performance liquid chromatography. J Pharm Sci 1992; 81: Davis GR, Santa Ana CA, Morawski SG, Fordtran JS. Development of a lavage solution associated with minimal water and electrolyte absorption or secretion. Gastroenterology 1980; 78: DiPalma JA, Cleveland MV, McGowan J, Herrera JL. A randomized, multicenter, placebo-controlled trial of polyethylene glycol laxative for chronic treatment of chronic constipation. Am J Gastroenterol 2007; 102: Di Palma JA, Cleveland MV, McGowan J, Herrera JL. An open-label study of chronic polyethylene glycol laxative use in chronic constipation. Aliment Pharmacol Ther 2007; 25: Simin M, Brok J, Stimac D, Gluud C, Gluud LL. Cochrane systematic review: pegylated interferon plus ribavirin vs. interferon plus ribavirin for chronic hepatitis C. Aliment Pharmacol Ther, 2007; 25: DiPiro JT, Michael KA, Clark BA, et al. Absorption of polyethylene glycol after administration of a PEG-electrolyte lavage solution. Clin Pharm 1986; 5: Brady CE, DiPalma JA, Morowski SG, Santa Ana CA, Fordtran JS. Urinary excretion of polyethylene glycol 3350 and sulfate after gut lavage with a polyethylene glycol electrolyte lavage solution. Gastroenterol 1986; 90: Schiller LR, Santa Ana CA, Porter J, Fordtran JS. Validation of polyethylene glycol 3350 as poorly absorbable marker for intestinal perfusion studies. Dig Dis Sci 1997; 42: Yamaoka T, Tabata Y, Ikada Y. Distribution and tissue uptake of poly(ethylene glycol) with different molecular weights after intravenous administration to mice. J Pharm Sci 1994; 83: Guidance for industry. Pharmacokinetics in Patients with Impaired Renal Function Study Design, Data Analysis, and Impact on Dosing and Labeling. US FDA Washington, DC, USA, CDER CBER, Rothfuss KS, Bode JC, Stange EF, Parlesak A. Urinary excretion of polyethylene glycol 3350 during colonoscopy preparation. Z Gastroenterol 2006; 44: Parlesak A, Bode JC, Bode C. Parallel determination of gut permeability in man with M(r) 400, M(r) 1500, M(r) 4000 and M(r) 10,000 polyethylene glycol. Eur J Clin Chem Clin Biochem 1994; 32: