OBJECTIVE—Glucagon-like peptide-1 (GLP-1) has been proposed as a new treatment modality for type 2 diabetes. To circumvent the drawback of the short half-life of GLP-1, inhibitors of the GLP-1–degrading enzyme dipeptidyl peptidase IV (DPP IV) have been examined. Such inhibitors improve glucose tolerance in insulin-resistant rats and mice. In this study, we examined the 4-week effect of 1-[[[2-[(5-cyanopyridin-2-yl)amino]ethyl]amino]acetyl]-2-cyano-(S)-pyrrolidine (NVP DPP728), a selective, orally active inhibitor of DPP IV, in subjects with diet-controlled type 2 diabetes in a placebo-controlled double-blind multicenter study.

RESEARCH DESIGN AND METHODS—A total of 93 patients (61 men and 32 women), aged 64 ± 9 years (means ± SD) and with BMI 27.3 ± 2.7 kg/m2, entered the study. Fasting blood glucose was 8.5 ± 1.5 mmol/l, and HbA1c was 7.4 ± 0.7%. Before and after treatment with NVP DPP728 at 100 mg × 3 (n = 31) or 150 mg × 5 (n = 32) or placebo (n = 30), subjects underwent a 24-h study with standardized meals (total 2,000 kcal).

RESULTS—Compared with placebo, NVP DPP728 at 100 mg t.i.d. reduced fasting glucose by 1.0 mmol/l (mean), prandial glucose excursions by 1.2 mmol/l, and mean 24-h glucose levels by 1.0 mmol/l (all P < 0.001). Similar reductions were seen in the 150-mg b.i.d. treatment group. Mean 24-h insulin was reduced by 26 pmol/l in both groups (P = 0.017 and P = 0.023). Although not an efficacy parameter foreseen in the study protocol, HbA1c was reduced to 6.9 ± 0.7% in the combined active treatment groups (P < 0.001). Laboratory safety and tolerability was good in all groups.

CONCLUSIONS—We conclude that inhibition of DPP IV is a feasible approach to the treatment of type 2 diabetes in the early stage of the disease.

The gut hormones glucagon-like peptide-1 (GLP-1) and gastric inhibitory polypeptide (GIP) are both incretin hormones that are released postprandially and markedly augment glucose-stimulated insulin secretion through sensitizing the β-cell action of glucose (13). GLP-1 also exhibits other effects of importance for glucose homeostasis, viz., inhibiting glucagon secretion, delaying gastric emptying, and stimulating insulin biosynthesis (2,3). These effects, along with a potential increase in peripheral insulin action (4), will together be antidiabetogenic. GLP-1 has also been shown to reduce postprandial and fasting glycemia in subjects with type 1 and type 2 diabetes (3,49) and may, therefore, be a potentially useful new therapeutic agent in the treatment of diabetes. However, GLP-1 is rapidly degraded in plasma by the enzyme dipeptidyl peptidase IV (DPP IV), resulting in the short circulating half-life of intact GLP-1 being <1 min (3,10,11). Therefore, GLP-1 is unattractive as chronic therapy because multiple daily injections are required to maintain glycemic control.

The short half-life of GLP-1 has prompted development of alternate strategies to harness the potent antidiabetic activity of GLP-1. One approach is to inhibit DPP IV activity, thereby prolonging the circulating half-life of endogenous GLP-1 (11). DPP IV (or CD26) is an enzyme that is found throughout the body in both plasma and the endothelial lining of several organs, such as the kidney, liver, and intestine (12). It cleaves a number of biologically active peptides, including GLP-1, which is degraded from the active form of GLP-1, i.e., GLP-17–36amide, yielding GLP-19–36amide (10,13,14). DPP IV also degrades GIP through a similar mechanism, removing the two first NH2-terminal amino acids (12,13); therefore, inhibiting this enzyme will prolong the circulating half-life of the two most important incretins. Interestingly, the degradation product of GLP-1, GLP-19–36amide, has been shown to exhibit GLP-1 receptor antagonistic properties (11,14), suggesting that DPP IV inhibition increases the level of an antagonist, GLP-19–36amide. The potential of using this approach in the treatment of diabetes is illustrated in studies showing that DPP IV–deficient mice (15) and rats (16) exhibit increased insulin secretion and glucose tolerance. Animal studies have confirmed the potential of this strategy because inhibition of DPP IV with valine pyrrolidide potentiates the insulinotropic action of GLP-1 in pigs (17). Furthermore, in diabetic animal models, improved glucose tolerance and insulin response to oral glucose have been demonstrated by several different DPP IV inhibitors (1820). However, no studies have so far been reported on the potential use of this principle in the treatment of type 2 diabetes in humans.

In this study, we examined the antidiabetic effect of daily administration of1-[[[2-[(5-cyanopyridin-2-yl)amino]eth-yl]amino]acetyl]-2-cyano-(S)-pyrrolidine(NVP DPP728) for 4 weeks in drug-naive subjects with type 2 diabetes previously treated with a diet and exercise regimen only. As previously reported in detail (21), NVP DPP728 is an orally active and highly selective DPP IV inhibitor containing a 2-cyanopyrrolidide moiety that is responsible for a slow-binding mechanism characterized by high potency, competitive behavior, and rapid reversibility inhibiting both human and rodent DPP IV activity. The compound has been shown to reduce DPP IV activity in plasma and to improve glucose tolerance in diabetic Zucker rats (19). Based on pharmacokinetic and pharmacodynamic data in healthy subjects, a total daily dose of 300 mg was selected for NVP DPP728, and our study compared the therapeutic usefulness of this dose administered twice daily or three times per day with placebo.

Study design

The study was a randomized, double-blind, placebo-controlled, multicenter study comprising a 4-week run-in period followed by a 4-week study period in which patients were randomized to receive blinded treatment with either placebo or NVP DPP728 100 mg t.i.d. or 150 mg b.i.d.. Five centers in Sweden participated (the University Hospitals in Lund, Malmö, Göteborg, and Umeå and the Karolinska Hospital in Stockholm). The study was approved by the Ethics Committee at these institutions and by the Swedish Medical Products Agency (Uppsala, Sweden). Written informed consent was obtained from each participant. Patients attended at weeks –4, −2, 0, 2, and 4, at which time fasting blood samples were collected and details of adverse events were obtained. At weeks –4, 0, and 4, a clinical examination was also performed. At weeks 0 and 4, the subjects underwent a 24-h study with standardized meals and frequent sampling for glucose and insulin levels. During these days, patients received a standard meal schedule comprising a total of 2,000 kcal, consisting of 55% carbohydrate, 25% fat, and 20% protein (23% consumed at breakfast, 36% consumed at lunch, 36% consumed at dinner, and 5% consumed at snack time). Breakfast consisted of 450 kcal with 50% as carbohydrate, 23% as fat, and 27% as protein. After the 24-h study period at week 0, patients were randomly assigned to placebo, NVP DPP728 100 mg t.i.d., or NVP DPP728 150 mg b.i.d.. Participants were provided with glucometers and instructed to monitor their blood glucose levels if they experienced symptoms of hypoglycemia. NVP DPP728 was administered as either 100- or 150-mg tablets or identical placebo tablets. Tablets were taken 10 min before breakfast and dinner in the twice-daily arm and 10 min before breakfast, lunch, and dinner in the three-times-daily arm. In each arm, a 2:1 randomization was used to achieve an overall equal number of patients on each active treatment regimen and on placebo.

Subjects

Patients eligible for the study were men or women aged >30 years with a history of type 2 diabetes diagnosed at least 12 weeks before entry into the study. Women were required to be nonfertile or using a medically approved method of birth control. Patients were excluded from randomization if they had mean fasting blood glucose concentration <7.2 or >10.0 mmol/l, mean HbA1c level <6.3% or >10.0% (reference value <5.3%), or BMI <20 or >32 kg/m2. Also excluded were patients with evidence of diabetic end-organ disease (renal, cardiac, neurological, retinal), other clinically relevant conditions (for example, previous gastric surgery), clinically significant laboratory abnormalities, or any conditions that might alter drug absorption or metabolism or gastric emptying.

Analyses

Glucose was measured by the glucose oxidase method (Boehringer Mannheim, Mannheim, Germany), insulin levels were measured by radioimmunoassay (Pharmacia, Uppsala, Sweden), HbA1c levels were determined by ion-exchange high-performance liquid chromatography, and triglycerides and cholesterol were measured with an enzymatic method (Johnson & Johnson, Rochester, NY).

Safety and tolerability

Safety parameters included physical examination, vital signs, electrocardiographic evaluations, laboratory evaluations (hematology, chemistry, and urinalysis), adverse events, and self-monitoring of blood glucose levels for suspected hypoglycemia. Symptomatic suspected hypoglycemic episodes were recorded even if not confirmed by a low blood glucose level. Asymptomatic low blood glucose levels (<2.8 mmol/l) obtained on fingerstick were also recorded.

Statistics

The primary efficacy parameter was the change in mean 24-h glucose, calculated as the area under the 24-h glucose curve divided by 24 compared with placebo. Secondary parameters included mean 24-h insulin, fasting plasma glucose (FPG) and the prandial glucose excursion (defined as the difference between the maximum glucose value observed in the 4-h postbreakfast period minus the mean of the prebreakfast measurements). All values were expressed as means ± SEM unless otherwise stated. Statistical comparisons were made between each of the treatment groups and the placebo group for the primary and the secondary efficacy parameters by analysis of variance. Although not included in the study protocol as an efficacy parameter because of the short treatment duration, the change in HbA1c between baseline and study end was also evaluated.

Patients

Of the screened 141 Caucasian patients, 93 (61 men) were eligible for the study. Median or mean values for age (median 64; mean ± SD 64 ± 9; range 35–80 years), duration of diabetes (3.6; 4.6 ± 5.6; 0.2–27 years), BMI (27.2; 27.0 ± 2.9; 21.3–31.9 kg/m2), HbA1c (7.2; 7.4 ± 0.7; 6.3–9.9%), FPG (8.1; 8.6 ± 1.5; 7.2–10.1 mmol/l), or fasting insulin (80; 79 ± 6; 36–101 pmol/l) did not differ significantly between the three study groups.

24-h Glucose and insulin levels

The 24-h plasma glucose and insulin levels before and after 4 weeks’ treatment in the three study groups are shown in Fig. 1. After treatment, both groups receiving NVP DPP728 displayed a significantly reduced mean 24-h glucose compared with those receiving placebo. The reduction in the 24-h mean glucose (difference from placebo) was −1.0 mmol/l (95% CI −1.4 to −0.7 mmol/l; P < 0.001) in the patients who had received 100 mg t.i.d. and –1.0 mmol/l (−1.4 to −0.6 mmol/l; P < 0.001) in those who had received 150 mg b.i.d.. Also, mean 24-h insulin levels were significantly reduced in both active treatment groups. In the patients receiving 100 mg t.i.d., the reduction compared with the placebo group was –26 pmol/l (−46 to −5 pmol/l; P = 0.017); in those receiving 150 mg b.i.d., the reduction was –26 pmol/l (−47 to −4 pmol/l; P = 0.023).

FPG and prandial glucose and insulin levels

FPG was significantly reduced over the 4-week period in both active treatment groups; the reduction (difference from placebo) was –1.0 mmol/l in the 100 mg t.i.d. group (95% CI −1.4 to −0.6 mmol/l; P < 0.001) and −0.7 mmol/l in the 150 mg b.i.d. group (−1.2 to −0.3 mmol/l; P < 0.001) (Table 1 and Fig. 2). There was also a significant difference between the groups in the number of patients achieving an FPG <7 mmol/l after treatment. Therefore, 35% of the subjects in the group treated with 100 mg t.i.d. and 37% in the group treated with 150 mg b.i.d. achieved an FPG <7 mmol/l after active treatment, versus only 3% in the placebo group. Also, the prandial glucose excursion was reduced by NVP DPP728, the difference from placebo being −1.2 mmol/l in the 100 mg t.i.d. group (−1.7 to −0.7 mmol/l; P < 0.001) and −1.3 mmol/l in the 150 mg b.i.d. group (−1.8 to −0.7 mmol/l; P < 0.001) (Fig. 2). In contrast, fasting and postprandial insulin levels were not significantly altered (Table 1, Fig. 1). However, peak postprandial insulin divided by peak postprandial glucose (i.e., insulin-to-glucose excursion ratio) increased in the 150 mg b.i.d. group versus the placebo group (P = 0.038) (Table 1).

HbA1c

Although it was not one of the efficacy parameters defined in the study protocol, because of the short treatment period, HbA1c was also reduced compared with placebo in both active treatment groups; the change from baseline (mean ± SD) in the treatment groups was −0.1 ± 0.3% with placebo, −0.5 ± 0.4% with 100 mg t.i.d., and –0.5 ± 0.3% with 150 mg b.i.d. (Fig. 2). The difference from placebo (Student’s t test) was significant (P < 0.001) for both active treatments. Because this analysis was not preplanned, these P values should, however, be regarded as descriptive.

Lipids

Except for a small but still significant reduction compared with placebo in total cholesterol in the two active treatment groups and a similar reduction in VLDL cholesterol and triglycerides in the 100 mg t.i.d. treatment group, lipid levels did not change during the treatment period (Table 1).

Safety and tolerability

Body weight was not altered during the study period (Table 1). Treatment was well tolerated. The adverse events in the active treatment groups are shown in Table 2. Four patients taking NVP DPP728 and one patient in the placebo group reported symptoms that could have indicated hypoglycemia, but only one of the patients given NVP DPP728 had a blood glucose level <3.3 mmol/l. Of patients taking NVP DPP728, four experienced symptoms of nasopharyngitis and five experienced pruritus. These symptoms were short-lived and transient, despite continuation of therapy. Finally, one patient in the NVP DPP728 b.i.d. treatment group developed a transient nephrotic syndrome during the first week of active treatment. This patient, however, already had albuminuria during the run-in period. Treatment was discontinued in this patient.

This study provides the first evidence that pharmacological DPP IV inhibition is feasible for the treatment of type 2 diabetes in humans. The patients studied all had mild type 2 diabetes (HbA1c 7.4%) during treatment with diet and exercise only. HbA1c or FPG did not change during the 4-week run-in period or during the 4-week treatment in the placebo group, confirming that the diabetes in the patients studied was stable. Therefore, the marked improvement in glycemic control, as determined by the 24-h glucose profile, the fasting glucose levels, and the 4-h prandial glucose excursion during the 4-week treatment with NVP DPP728, is a clear effect of the treatment. We also found that there was no difference between a three-times-daily treatment schedule versus a twice-daily treatment schedule with NVP DPP728, indicating that either dosing regimen could probably be used with equal efficacy. Despite the short duration of the study period (4 weeks), a reduction in HbA1c levels by 0.6% was observed after treatment with NVP DPP728. Combined, these results encourage further development of DPP IV inhibition in the treatment of type 2 diabetes.

This study was not designed to establish the mechanisms underlying the beneficial effects of NVP DPP728 on the glycemic control in humans, which therefore remains to be established. Nevertheless, an increased insulin response to glucose might contribute, because the prandial insulin levels were not significantly altered in combination with reduced prandial glucose excursion after 4 weeks of treatment, which was also illustrated by the increased insulin-to-glucose excursion ratio by NVP DPP728 versus placebo. This could be exerted by GIP and GLP-1, which increase the β-cell sensitivity to glucose, because inhibition of DPP IV results in increased levels of these incretins (11). Other mechanisms, however, may also be contributing, because GLP-1 has been shown to delay gastric emptying and inhibit glucagon secretion (24), both of which would reduce plasma glucose levels. In addition, the significant reduction in fasting glucose suggests that mechanisms beyond the prandial period may also be playing a role, but further studies will be needed to define the exact mechanism or mechanisms by which NVP DPP728 and similar drugs achieve their clinical effects.

The dyslipidemia associated with type 2 diabetes has been shown to be improved in conjunction with improved diabetic condition after treatment with GLP-1 (22). We found, however, only minimal changes in the lipid profile after the 4-week treatment with NVP DPP728. Further studies will be needed to elucidate the possible long-term effect of DPP-IV inhibition on lipid levels in type 2 diabetic patients.

Overall, NVP DPP728 was well tolerated in the patients. Four patients receiving active treatment exhibited symptoms suggestive of hypoglycemia, but these were generally not associated with low blood glucose values and therefore cannot be classified as true hypoglycemia. Hypoglycemia was confirmed by a blood glucose value <3.3 mmol/l in only one subject. This patient, however, had reported heavy use of alcohol the evening before the event, and a causal relation to NVP DPP728 is therefore doubtful. However, the occurrence of hypoglycemia after treatment with DPP IV inhibitors should be monitored carefully in future studies in view of a previous report that hypoglycemia can occur after administration of GLP-1 to fasted individuals (23), even though it is likely to be very rare due to the glucose dependency of the action of GLP-1 (1). A more important safety concern is that DPP IV inhibition will slow or completely prevent the degradation of >20 other bioactive peptides in addition to GLP-1 and GIP, of which insulin-like growth factor-1, neuropeptide Y, GLP-2, and substance P may be of interest for the treatment of type 2 diabetes (12). The occurrence of pruritus in five patients treated with NVP DPP728 might be explained by increased levels of other biologically active peptides apart from GLP-1 and GIP, such as substance P, which has been suggested to be associated with pruritus (24). However, in all of these five cases, the pruritus, localized mainly to the palms, was transient and disappeared within 48 h; treatment was not discontinued in any of these patients. To what extent peptides other than GLP-1 and GIP contribute to the improved metabolic control or side effects during treatment with NVP DPP728 remains to be established.

The present study in human subjects with type 2 diabetes shows that DPP IV inhibition as monotherapy is feasible as a treatment for type 2 diabetes in patients at a relatively early stage in the disease. Because the study had a duration of only 4 weeks, further long-term studies will be needed to examine the long-term effects of DPP IV inhibition as well as to fully understand the mechanism of the effects and to define the use of this approach in patients with more advanced diabetes and in combination with other antidiabetic drugs.

The study was supported by the Swedish Research Council (Grant 6834); the Faculty of Medicine, Lund University (Lund, Sweden); and Novartis (Basel, Switzerland).

We thank research nurses Margaretha Persson, Kerstin Nilsson, and Lena Bryngelsson (Malmö); Kajsa Sundqvist and Annica Clark (Stockholm); Pia Roslin (Umeå); Marie Wilson, Margit Bergström, and Bertil Nilsson (Lund); Thordis Andrén (Göteborg); and Dr. Maria Svensson (Umeå) for their expert assistance. We also thank the general practitioners who participated in the recruiting of patients for the study and all the patients who participated in the study.

1
Fehmann HC, Göke R, Göke B: Cell and molecular biology of the incretin hormones glucagon-like peptide-I and glucose-dependent insulin releasing polypeptide.
Endocr Rev
16
:
390
–410,
1995
2
Drucker DJ: Glucagon-like peptides.
Diabetes
47
:
159
–169,
1998
3
Ahrén B: Glucagon-like peptide 1 (GLP-1) - a gut hormone of potential interest in the treatment of diabetes.
BioEssays
20
:
642
–651,
1998
4
Nauck MA, Holst JJ, Willms B, Schmiegel W: Glucagon-like peptide 1 (GLP-1) as a new therapeutic approach for type 2-diabetes.
Exp Clin Endocrinol Diabet
105
:
187
–195,
1997
5
Gutniak M, Ørskov C, Holst J, Ahrén B, Efendic S: Antidiabetogenic effect of glucagon-like peptide-1 (7–36)amide in normal subjects and patients with diabetes mellitus.
N Engl J Med
326
:
1316
–1322,
1992
6
Gutniak MK, Linde B, Holst JJ, Efendic S: Subcutaneous injection of the incretin hormone glucagon-like peptide 1 abolishes postprandial glycemia in NIDDM.
Diabetes Care
17
:
1039
–1044,
1994
7
Nauck MA, Wollschlaeger D, Werner J, Holst JJ, Ørskov C, Creutzfeldt W, Willms B: Effects of subcutaneous glucagon-like peptide 1 (GLP-[7–36amide]) in patients with NIDDM.
Diabetologia
39
:
1546
–1553,
1996
8
Gutniak MK, Larsson H, Sanders SW, Juneskans O, Holst JJ, Ahrén B: GLP-1 tablet in type 2 diabetes in fasting and postprandial conditions.
Diabetes Care
20
:
1874
–1879,
1997
9
Rachman J, Barrow BA, Levy JC, Turner RC: Near-normalization of diurnal glucose concentrations by continuous administration of glucagon-like peptide-1 (GLP-1) in subjects with NIDDM.
Diabetologia
40
:
205
–211,
1997
10
Deacon CF, Johnsen AH, Holst JJ: Degradation of glucagon-like peptide-1 by human plasma in vitro yields an N-terminally truncated peptide that is a major endogenous metabolite in vivo.
J Clin Endocrinol Metab
80
:
952
–957,
1995
11
Holst JJ, Deacon CF: Inhibition of the activity of dipeptidyl-peptidase IV as a treatment for type II diabetes mellitus.
Diabetes
47
:
1663
–1670,
1998
12
Mentlein R: Dipeptidyl-peptidase IV (CD26) - role in the inactivation of regulatory peptides.
Regul Pept
85
:
9
–24,
1999
13
Kieffer TJ, McIntosh CH, Pederson RA: Degradation of glucose-dependent insulinotropic polypeptide and truncated glucagon-like peptide 1 in vitro and in vivo by dipeptidyl peptidase IV.
Endocrinology
136
:
3585
–3596,
1995
14
Knudsen LB, Pridal L: Glucagon-like peptide-1-(9–36)amide is a major metabolite of glucagon-like peptide-1-(7–36)amide after in vivo administration to dogs, and it acts as an antagonist on the pancreatic receptor.
Eur J Pharmacol
318
:
429
–435,
1996
15
Marguet D, Baggio L, Kobayashi T, Bernard AM, Pierres M, Nielsen PF, Ribel U, Waranabe T, Drucker DJ, Wagtmann N: Enhanced insulin secretion and improved glucose tolerance in mice lacking CD26.
Proc Natl Acad Sci U S A
97
:
6874
–6879,
2000
16
Nagakura T, Yasua N, Yamazaki K, Ikuta H, Yoshikawa S, Asano O, Tanaka I: Improved glucose tolerance via enhanced glucose-dependent insulin secretion in dipeptidyl peptidase IV-deficient Fischer rats.
Biochem Biophys Res Commun
284
:
501
–506,
2001
17
Deacon CF, Hughes TE, Holst JJ: Dipeptidyl peptidase IV inhibition potentiates the insulinotropic effect of glucagon-like peptide 1 in the anesthetized pig.
Diabetes
47
:
764
–769,
1998
18
Pederson RA, White HA, Schlenzig D, Pauly RP, McIntosh CHS, Demuth HU: Improved glucose tolerance in Zucker fatty rats by oral administration of the dipeptidylpeptidase IV inhibitor isoleucine thiazolidide.
Diabetes
47
:
1253
–1258,
1998
19
Balkan B, Kwasnik L, Miserendino R, Holst JJ, Li X: Inhibition of dipeptidyl peptidase IV with NVP DPP728 increases plasma GLP-1 (7–36 amide) concentrations and improves oral glucose tolerance in obese Zucker rats.
Diabetologia
42
:
1324
–1331,
1999
20
Ahrén B, Holst JJ, Mårtensson H, Balkan B: Improved glucose tolerance and insulin secretion by inhibition of dipeptidyl peptidase IV in mice.
Eur J Pharmacol
404
:
239
–245,
2000
21
Hughes TE, Mone MD, Russell ME, Weldon SC, Villhauer EB: NVP DPP728 (1-[[[2-[(5-cyanopyridin-2-yl)amino]ethyl]amino]acetyl]-2-cyano-(S)-pyrrolidine), a slow-binding inhibitor of dipeptidyl peptidase IV.
Biochemistry
38
:
11597
–11603,
1999
22
Juntti-Berggren L, Pigon J, Karpe F, Hamsten A, Gutniak M, Vignati L, Efendic S: The antidiabetogenic effect of GLP-1 is maintained during a 7-day treatment period and improves diabetic dyslipoproteinemia in NIDDM patients.
Diabetes Care
19
:
1200
–1206,
1996
23
Edwards CM, Todd JF, Ghatei MA, Bloom SR: Subcutaneous glucagon-like peptide-1 (7–36) amide is insulinotropic and can cause hypoglycemia in fasted healthy subjects.
Clin Sci
95
:
719
–724,
1998
24
Singh LK, Pang X, Alexacos N, Letourneau R, Theoharides TC: Acute immobilization stress triggers skin mast cell degranulation via corticotropin releasing hormone, neurotensin, and substance P: a link to neurogenic skin disorders.
Brain Behav Immunol
13
:
225
–239,
1999

Address correspondence and reprint requests to Dr. Bo Ahrén, Department of Medicine, Lund University, B11 BMC, S-221 84 Lund, Sweden. E-mail: bo.ahren@med.lu.se.

Received for publication 15 August 2001 and accepted in revised form 5 February 2002.

Bo Ahrén, Peter Båvenholm, Suad Efendic, Per-Anders Jansson, Madeléne Sandqvist, and Erik Simonsson have received grant support and Jan W. Eriksson has received consulting fees from Novartis, Basel, Switzerland.

A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances.