To determine in a noninferiority study whether mesalamine foam is as effective as mesalamine liquid enema for inducing clinical remission in patients with active left-sided ulcerative colitis (UC).
In a multicenter investigator-blind trial, 375 patients with mild-to-moderate UC were randomized to receive mesalamine foam 1 g/80 mL/day or mesalamine liquid enema 1 g/100 mL/day for 4 wk (W). Inclusion criteria were: disease extension at least 5 cm from anorectal junction and not above splenic flexure and Clinical Activity Index (CAI) 1–4 ≥ 4. Primary end point was clinical remission at W4 defined as a CAI 1–4 ≤ 2. Noninferiority of the foam to liquid enema was declared if the lower limit of the 97.5% unilateral confidence interval (97.5% CI) of the difference in remission rates between foam and liquid enema groups was greater than −15% .
Remission rates at W4 in foam versus liquid were 68.3% versus 73.6% in per protocol (PP) population (lower limit of 97.5% CI −15.1%) and 66.7% versus 70.5% in intention-to-treat (ITT) population (97.5% CI −13.4%). Remission rates at W2 were 48.1 % versus 50.6% in ITT (97.5% CI −12.8%) and 49.1% versus 52.1% in PP (97.5% CI −13.8%) in foam versus liquid, respectively. Both treatments were well tolerated.
A 4-wk treatment of 1 g mesalamine foam induced a clinical remission in 68% patients versus 73% with 1 g mesalamine liquid enema. Although the noninferiority of mesalamine foam could not be strictly demonstrated at W4 in the PP analysis, it was achieved in the ITT population and at W2 in both populations. Mesalamine foam represents a therapeutic alternative to mesalamine liquid enema in patients with mild-to-moderate active proctitis and proctosigmoiditis.
Ulcerative colitis (UC) is a chronic inflammatory bowel disease of unknown etiology and pathogenesis. Several trials have demonstrated the efficacy and safety of rectal salicylates in active distal UC (1, 2, 3, 4, 5, 6, 7, 8). Mesalamine liquid enemas are the first-line treatment for active proctosigmoiditis and left-sided colitis because their volumes ensure that the drug is brought to the inflamed mucosa as far as the splenic flexure. However, many patients have difficulties to retain liquid enemas because of their fluid consistency. To overcome this problem, rectal foams were developed. Their higher viscosity favor the retention and patient's acceptance, reducing incontinence and enhancing adhesion to the mucosa. Furthermore, 5-ASA is more finely dispersed on the mucosa by foam so that a wider spreading can be obtained (9). The efficacy of several 5-ASA foams has been demonstrated in clinical studies and some have received marketing approval (7, 10, 11, 12, 13, 14).
A new highly concentrated micronized foam formulation of mesalamine was developed. As the mesalamine crystal's diameter was smaller than in all presently marketed nonmicronized, this formulation was able to provide a minimum of 14 doses of 1 g per vial. The extent of dispersion of this foam preparation was determined in a gamma scintigraphic study performed in nine healthy volunteers. Foam spread throughout the rectum and the sigmoid colon immediately after dosing in all subjects and in six volunteers the foam reached the lower descending colon. At the end of the 12-h imaging period, the foam was located only in the rectum and sigmoid colon in three subjects, in four the foam had spread as far as the lower descending colon, and in two the foam had reached the upper descending colon (15). A pharmacokinetic study performed after application of 1 g per rectum in healthy volunteers confirmed that with the micronized form, the absorption was comparable with that of a current marketed mesalamine formulation of rectal administration. In this study, no serious adverse event occurred; the 1 g micronized mesalamine foam administered rectally appeared well tolerated (15).
The objective of this study was to demonstrate that mesalamine foam enema 1 g/dose per day was as effective as liquid enema 1 g/100 mL per day for treatment of patients with active left-sided UC.
PATIENTS AND METHODS
The study was conducted at 67 sites: France (N = 55), Belgium (N = 7), and Netherlands (N = 5) from July 2003 to July 2005. All participants gave written informed consents. The protocol was approved by all relevant Independent Ethics Committee (first approval was obtained on June 3, 2003 from the Ethics Committee of Lille, France). This study was conducted in accordance with the Declaration of Helsinki, in compliance with the consolidated Good Clinical Practice (GCP) guideline set forth in the International Conference on Harmonization (ICH) and the applicable regulatory requirements.
This was a multicenter, controlled, investigator-blinded design, randomized, parallel, and stratified group study.
Patients were randomly assigned in a 1:1 ratio to rectal foam or reference liquid enema, based on a central computer-generated randomization scheme. The numbers were allocated sequentially in the order in which the patients were enrolled. Under no circumstances patients enrolled in the study could be permitted to reenrol for a second time in this study.
After obtained informed consent, investigators used an interactive voice response system (IVRS) that assigned the eligible patient to the next randomization number for the concerned centre.
The central randomization was stratified according to the disease's extent: the first stratum included patients with proctitis (≥5 cm and <15–18 cm from the anal margin) or proctosigmoiditis (≥18 cm and <60 cm) and the second stratum included patients with left-sided UC with disease extension from 60 cm to splenic flexure.
According randomization (visit 1), patients received either mesalamine rectal foam or reference liquid enema for 4 wks (W) following Visit 1. Appropriate medication was dispensed by the investigators in a blinded manner. Patients attended the study centers three times: at visit 1 (baseline, day 1), at visit 2 (W2, day 15 ± 3), and visit 3 (final visit, W4, day 29 ± 3). Physical examination including vital signs and weight, blood and pregnancy tests, endoscopy (or sigmoidoscopy), use of concomitant medication, and acceptability of treatment were assessed during the study.
Included patients began treatment at once and recorded in diaries stool frequency, presence of blood in stools, and abdominal pain/cramps.
Men or nonpregnant women older than 18 yr were eligible if they had newly diagnosed or relapsing active mild-to-moderate left-sided UC with a disease extension of at least 5 cm from the anal margin and not above the splenic flexure. At least one total colonoscopy in the disease history was required. For relapses, an endoscopy having visualized the superior limit of the disease within the previous 6 months was necessary. Activity of UC was retrospectively assessed on the 7 days before inclusion according to modified Rachmilewitz score Clinical Activity Index (CAI) 1–4 (16) that had to be at least of 4 for inclusion. Activity of UC was thereafter assessed according to CAI 1–4 calculated at each visit using patient's diary card completed on a weekly basis.
Oral maintenance treatment was permitted if given at a stable dose for at least 1 month before inclusion for 5-ASA and for at least 6 months for azathioprine/methotrexate and if maintained at the same dose during the study. Antibiotics, nonsteroidal anti-inflammatory drugs (NSAIDs), and rectal steroids were not permitted within 1 wk prior to baseline, oral steroids within 1 month, and other immunomodulators within 3 months prior to baseline. Patients with known significant hepatic or renal function abnormalities and/or clearance creatinine ≤ 80 mL/min were not included. All patients gave their written informed consent before inclusion.
The test medication was a foam enema in a pressurized vial suitable for rectal application and containing sufficient medication for 2 wk. The propellant was a safe mixture of butane/isobutane/propane/nitrogen. A single actuation of the vial released 1 g of mesalamine in a volume of about 80 mL foam. The reference medication was the commercially available liquid enema Pentasa (Ferring Pharmaceuticals, Gentilly, France) containing 1 g mesalamine in a 100 mL volume. Patients received a single dose of 1 g mesalamine either from foam or liquid enema every evening for 4 wk.
Because the difference between the rectal forms precluded a double-blind design, an investigator-blinded design was chosen. Although the enemas were not identical in appearance, the treatment boxes were equal in size and weight. The patient was informed to avoid giving any information about the form of the treatment.
At W4, patients had to return the treatment box containing used and unused medication, and answer questions about compliance. The accountability of the brought-back treatments (number of enemas or weight of the vials containing the foam) was checked by an external observer.
Efficacy and Safety Variables
The primary efficacy variable was clinical remission at 4 wk defined on the basis of the modified CAI 1–4 according to Rachmilewitz (16) that covers clinical symptoms for 1 wk and includes four variables: number of stools/wk, presence of blood in stools, abdominal pain/cramps, and investigator's global assessment. The first three variables were noted daily in the patient's diary and were summed by the physician to give a weekly score. Active UC was defined by a CAI 1–4 ≥ 4 and clinical remission by a CAI 1–4 ≤ 2. Further secondary efficacy variables were improvement of clinical symptoms assessed by weekly CAI 1–4 and endoscopic scoring at W4. Endoscopic findings were graded according to the endoscopic index of Rachmilewitz that considers granulation, vascular pattern, vulnerability of the mucosa, and mucosal damage.
The global acceptability at W4 or at discontinuation was assessed by a 100 mm Visual Analogue Scale (VAS) delimited from 0 (not at all) to 100 (perfectly).
Treatment acceptability was assessed by summing: (a) an enema retention score summing a retention volume score (from 0 if completely retained to 3 × episodes if not retained) and a retention duration score (from 0 if all night long duration to 2 × episodes for duration < 1 h), (b) an anal pain score (from 0 if none to 3 × episodes in severe intensity), (c) an abdominal pain score (from 0 if none to 3 × episodes in severe intensity) and (d) a flatulence score (from 0 if none to 3 × episodes in severe intensity).
Primary study end point was clinical remission defined by a CAI 1–4≤ 2 at the final/withdrawal visit in the per protocol (PP) population. Secondary end points were: clinical remission at W2, endoscopic remission at W4 defined as an endoscopic score <4, weekly CAI 1–4, treatment acceptability per visit, and global acceptability at W4. Safety variables were adverse events, relevant changes in laboratory values (central analysis), and the overall tolerability.
Analysis (Statistical Methods)
PRIMARY END POINT.
The main goal of this study was to evaluate the noninferiority of the mesalamine foam to the liquid enema in the treatment of active left-sided UC, with clinical remission at W4 as primary variable. A per protocol analysis, which is the most conservative in noninferiority trials, was chosen as primary population, in order to follow ICH of technical requirement for registration of pharmaceuticals for human use recommendations (17, 18). In consequence, the confirmatory statistical analysis was based on the PP population analysis and the intention-to-treat (ITT) analysis was supportive. The estimated rate of remission with the liquid enema was 65%. An acceptable difference between foam and liquid enema for remission rate was less than 5%. The noninferiority of the foam was declared if the lower limit of the 97.5% unilateral confidence interval (97.5% CI) of the difference in remission rates between the foam and the liquid enema groups was greater than −15%.
SECONDARY END POINTS.
Secondary efficacy analysis was conducted in the ITT population as the main analysis and in the PP populations as a supportive analysis.
Clinical remission at W2 was submitted to the same analysis as the primary efficacy. Endoscopic remission was assessed using a Cochran-Mantel-Haenszel (CMH) χ2 test adjusted on disease location. Weekly CAI 1–4 score, items of CAI 1–4 score, global acceptability, and treatment acceptability score were submitted to an analysis of covariance (ANCOVA).
Safety analyses were conducted according to the treatment actually received rather than according to the treatment assigned.
DETERMINATION OF SAMPLE SIZE.
Assuming 65% clinical remission for liquid enema, with a first kind error of 5% and a power of 80%, and to insure a lower limit of the 95% CI of the difference in remission between foam and liquid enemas greater than −15%, 160 evaluable patients have to be analyzed in each group in the PP population. With a 12% withdrawal, a total of 378 patients had to be included.
Participant Flow and Follow-Up
A total of 375 patients were randomized (191 in the foam group and 184 in the liquid enema group), of which 373 could be evaluated for safety (foam: N = 191, liquid enema: N = 182), 368 for the ITT analysis (foam: N = 189, liquid enema: N = 179), and 330 for the PP analysis (foam: N = 167, liquid enema: N = 163) (Fig. 1).
Major protocol deviations, leading patients to be excluded from the PP analysis, were observed in 12.6% in the foam group and 11.4% in the liquid enema group. The most frequent deviation in both groups was bad compliance (6.8% and 6.0% in the foam and liquid enema groups, respectively), followed by no efficacy evaluation and inclusion criteria not fulfilled.
Forty-five patients were prematurely withdrawn from the study medication. The main reasons were nonserious adverse event (foam: N = 7, liquid enema: N = 6), worsening or, insufficient response (foam: N = 5, liquid enema: N = 6), patient lost to follow-up (liquid enema: N = 3), pregnancy (foam: N = 1), serious adverse event (foam: N = 1, liquid enema: N = 1), and other reason (foam: N = 7, liquid enema: N = 8). Among these 45 premature withdrawals, 26 patients had at least one major protocol deviation (foam: N = 12, liquid enema: N = 14) that excluded them from the PP analysis.
The two groups were well balanced for basic demographic data and disease history (Table 1).
There was a significant difference (P < 0.009) between groups with regard to gender distribution with more men than women in the foam group. Recruitment resulted in more than 90% of proctitis or proctosigmoiditis. In around 80% of the patients, the current episode lasted for more than 3 months.
The most frequently taken concomitant medication was oral aminosalicylates (27.5% and 26.3%, respectively, in foam and liquid enema groups). Only 2% of patients took azathioprine in each group.
Primary Efficacy Evaluation
The clinical remission rates in the PP analysis were 68.3% for the foam and 73.6% for the liquid enema (Fig. 2). The observed difference was −5.4% and the lower limit of the unilateral 97.5% CI was −15.1%. In the ITT analysis, the remission rates were 66.7% for the foam and 70.5% for the liquid enema. The observed difference was −3.8% and the lower limit of the 97.5% CI was −13.4%.
Secondary Efficacy Evaluation
The clinical remission at W2 in the ITT analysis was 48.1% and 50.6%, respectively, in the foam and liquid enema groups. The observed difference was −2.5% and the lower limit of the 97.5% CI was −12.8%. In the PP analysis, the remission rates were 49.1% for the foam and 52.1% for the liquid enema. The observed difference was −3.0% and the lower limit of the unilateral 97.5% CI was −13.8%. The CAI 1–4 score decreased, that is, improved in each group with further visits. Similar results were observed in ITT and PP analysis (Fig. 3). The rate for endoscopic remission, defined as an endoscopic score <4, was 64.2% in the foam group and 72.7% in the liquid enema group in ITT (P = 0.09) and 64.4% and 75.5% in PP analysis (P = 0.03).
The weekly changes of CAI 1–4 are listed in Table 2.
Mean scores of global acceptability assessed on a VAS at W4 were not different in both groups: 76.1 (26.20) in the foam, 78.6.0 (22.77) in the liquid enema group (ITT). In both foam and liquid enema groups the total score of treatment acceptability by visit decreased, that is, improved between visit 2: 20.4 (25.4) and 24.2 (24.8) and visit 3: 14.6 (21.3) and 17.3 (23.5) (mean; SD).
The scores for anal pain, abdominal pain, and flatulence improved in both groups between visit 2 and 3, in the ITT and the PP populations. Although nonsignificant, the treatment acceptability score improved more in the foam group than in the liquid enema group. The enema retention total score was significantly lower (i.e., better retention) in the foam group than in the liquid enema group: 2.4 ± 6.4 versus 6.1 ± 11.9 (P < 0.001) at visit 2 and 2.0 ± 5.9 versus 5.0 ± 11.6 (P = 0.004) at visit 3. Similar results were observed in the PP population (Fig. 4).
Influence of Covariates on Clinical Remission
The influence of baseline covariates on the clinical remission rates was analyzed in the ITT and PP populations. Localization of the disease was assessed on a post hoc analysis comparing remission rate in patients with proctitis (15–18 cm from anal verge) to those with proctosigmoiditis (until 60 cm from anal verge). There was no difference in remission rate either in ITT or PP analysis indicating that both foam and liquid enema were effective in proctitis and in proctosigmoiditis. The sample size of the subgroup of left-sided colitis was too small for relevant statistical analysis. The following covariates did not influence the clinical outcome: gender, intake or not of oral aminosalicylates during the study, CAI 1–4 score at baseline, and storage of the foam in vertical or horizontal position.
Safety and Tolerability
In total, 373 patients received at least one dose of study medication. One hundred ninety-one patients actually received the foam and 182 patients actually received the enema.
Both foam and liquid enema were well tolerated. In total, 85 adverse events (AEs) occurred in 52 patients (27.2%) in the foam group and 96 AEs in 59 patients (32.4%) in the liquid enema group. In both groups, the most frequently reported AEs were gastrointestinal disorders (Table 3).
AEs leading to permanent study medication discontinuation were reported in 7.3% and 6.6% of patients in the foam and liquid enema groups, respectively, without statistically significant difference between the groups. Gastrointestinal disorders were the main reason of discontinuation because of AE in both groups.
Besides, the mean changes in laboratory criteria (creatinine clearance levels) and in physical examination were small and devoid of clinical relevance.
A number of studies have shown that mesalamine liquid enemas were highly effective in acute UC provided the disease does not extend beyond the splenic flexure (2, 3, 4, 5, 19). Rectal formulations of mesalamine represent the first-choice treatment for distal colitis being significantly superior both to placebo and topical corticosteroids (4, 20). Mesalamine suppositories are regarded as the most suitable treatment for patients with proctitis while liquid enemas and foams, thanks to a better retrograde spread, are suitable for more extensive UC (21).
In this randomized controlled trial comparing a 80-mL 1 g mesalamine foam to a liquid enema in patients with active left-sided UC, both formulations could be regarded as similarly effective, achieving clinical remission after 4 wks of treatment in around 70% of patients. Although the noninferiority of the foam was not fully demonstrated at 4 wk in the PP analysis (−15.1% for a limit of −15%), it was achieved in the ITT analysis at 4 wks and in both PP and ITT analysis at 2 wk. Furthermore, the shifts of CAI 1–4 indicated similar rates of clinical improvement and recorded symptoms (stools frequency, blood in stools, abdominal pain/cramps) improved similarly in both groups as well as the symptoms assessed by the investigators. The definition retained in this trial for left-sided colitis (from 60 cm of anal verge until left angle) may explain the small number of patients with left-sided colitis. In consequence, the analysis of proctitis plus proctosigmoiditis versus left-sided colitis was not possible as initially planned.
Rectal self-administration of mesalamine foam proved to be safe and well tolerated. Data regarding the patient's global acceptability showed that both enemas were well tolerated but that, within the treatment acceptability score, the new mesalamine foam was significantly better retained. In another study comparing mesalamine foam with a standard liquid enema in active distal UC (12), patients in the foam group had more adverse events than in the liquid enema group. Flatulence was the most prominent symptom, and the authors attributed this effect to an incorrect application of the foam. The different rating of different rectal foams might indicate that the individual characteristics of each foam application device as well as volume and pressure of the foam might influence tolerability results.
Here, both treatments were well tolerated with similar adverse event profiles and no significant change in renal clearance.
Because mesalamine is safe and effective for inducing clinical remission in patients with mildly to moderately active UC, true placebo controlled trials are not possible for ethical reasons. Thus, in order to evaluate new mesalamine formulation as potential first-line treatments, noninferiority studies are usually required although few of them have been performed in patients with inflammatory bowel disease (IBD) (22, 23, 24). Indeed a recent systematic review reported that a majority of noninferiority studies in digestive trials were actually failed superiority studies (25). In our study, remission rate in patients treated with liquid enema was estimated at 65% and the difference of clinical remission rate accepted between foam and liquid enema was 5%. This was a stringent delta because we have chosen at the same time a one-sided 97.5% confidence interval for the difference between both groups according to the EMEA recommendations (26). As far as the −15%, lower limit of the confidence interval of the difference, it was similar to those retained in comparable noninferiority studies in active UC (assessing local or general treatments), which vary from −15 to −20% (12, 13, 14, 21, 27, 28, 29).
The findings reported here are in good agreement with data in the literature: in most trials, clinical and endoscopic remission with mesalamine foam enema are close to that achieved with liquid enema, endoscopic remission being 10% somewhat lower with both formulations. Lee et al. compared a 2 g/day mesalamine foam to a prednisolone foam in 149 patients with distal active UC. At 4 wk, the clinical and endoscopic remission rates were 52% and 40%, respectively, in the mesalamine foam group (7). In another trial, a 1 g mesalamine foam in a volume of 30 mL was compared with a placebo foam of a similar volume in patients with active proctosigmoiditis. After 6 wk of treatment, the clinical and endoscopic remission rates were 65% and 57% in mesalamine foam-treated patients (13). Gionchetti et al. compared a 120 ml 2 g mesalamine foam to a 2 g gel enema. At 4 wks clinical remission was 69% and endoscopic remission 52% (10). Another 2 g mesalamine foam in a volume of 60 ml was compared with a 60 mL 4 g mesalamine liquid enema at 4 wk. Clinical remission rates were 64.9% and 69.5%, respectively, and the endoscopic remission about 38% in both groups (12). Two 1 g mesalamine foams (30 mL and 60 mL) administered twice daily were compared in active proctosigmoiditis. At 4 wk, clinical remissions were 77.4% and 75.5% and endoscopic remissions were 59.6% and 65.5%, respectively (14).
Although the spread of mesalamine foam was not assessed in our study, good distribution of mesalamine was apparent from the activity in all subgroups of patients classified according to the UC location. Campieri et al. compared the spread and distribution of mesalamine foam with those of liquid enema in 10 patients with UC and found a more uniform distribution and prolonged persistence in the descending and sigmoid colon for the foam formulation (9). An equivalent or a better spread and distribution of Salofalk foam versus mesalamine liquid enema were also found in healthy volunteers and UC patients (13).
The study has a limitation that we want to acknowledge. As a consequence of recruitment being mainly proctitis and proctosigmoiditis, the results of the study concerns these two forms and it is not possible to conclude for left-sided UC. The low number of enrolled left-sided UC patients could be explained by the definition of disease's extent used in this study for proctosigmoiditis: up to 60 cm. It is noteworthy that 60 cm from the anal verge is, for some authors, the definition of left-sided ulcerative UC (8).
Finally, we can conclude that 1 g mesalamine foam enema represents a clinically efficient and well-tolerated therapeutic alternative to mesalamine liquid enemas in patients with mild-to-moderate active proctitis and proctosigmoiditis. It may be especially appropriate in case of poor tolerance of enemas because of acute rectal inflammation.
What Is Current Knowledge
Good agreement with data in the literature: in most trials, clinical and endoscopic remission with mesalamine foam enema are close to that achieved with liquid enema, endoscopic remission being 10% somewhat lower with both formulations.
What Is New Here
Real noninferiority study with a good methodology.
Both formulations were similarly effective with high clinical remission rates after 4 wk of treatment in around 70% of patients.
Both enemas were well tolerated but with better retention for the new mesalamine foam (acceptability score).
CONFLICT OF INTEREST
Name of guarantor: Antoine Cortot, M.D.
Specific author contributions: Conception and writing of study design protocol: A. Cortot, S. Forestier, I. Idier, and M. Lémann; coordination of the study: A. Cortot, S. Forestier, I. Idier, and M. Lémann; interpretation of rough data and of statistical analysis: A. Cortot, S. Forestier, I. Idier, and M. Lémann; oral presentations of the data: A. Cortot; writing and corrections of the paper: A. Cortot, S. Forestier, I. Idier, and M. Lémann; other authors: inclusion of patients into the study.
Potential competing interests: None.
Financial support: This study was sponsored by Ferring France Company.
Conflict of interest: None.
This study was sponsored by Ferring France Company.
Investigators having participated in the study were:
Belgium: F. Baert, P. Caenepeel, G. D'Haens, M. De Reuck, O. Dewit, F. Fontaine, M. Stevens, P. Rutgeerts, F. Van De Mierop.
France: N. Abdelli, J.-L. Alexandre, P. Baron, C. Bastid, P. Benfredj, S. Beorchia, J.-M. Bidart, P. Bonvarlet, R-M. Bory, M. Bougnol, J. Bruhiere, P. Brulet, F. Castex, J.-P. Caucanas, J.-B. Cazals, G. Cerf, M. Chaillou, C. Chamiot-Prieur, J.-F. Claerbout, P. Colonna, M. Colson, P. Coulom, P. Dalbies, E. Degoutte, O. Delette, G. Deregnaucourt, P. Emery, P. Estable, M. Evreux, B. Flourié, M. Gompel, D. Gorce, L. Grozel, B. Grunberg, X. Hébuterne, F. Hedelius, T. Helbert, A. Ivanovic, R. Kornhauser, A. Landivier, J.-L. Larpent, P. Lebourgeois, N. Lemarchand, S. Lemiere, A. Leplat, E. Leprince, B. Leurent, D. Lescut, P. Levy, J.-J. Lugand, D. Maetz, M. Mahé, P. Meunier, B. Mory, B. Napoléon, M. Peyrot, R. Pierrugues, P. Pommelet, P. Potier, B. Pujol, M. Queralto, C. Revol, B. Richard-Molard, J.-M. Suduca, J. Silvie, Y. Teste, P. Vandevenne, L. Vandromme, G. Vauban, A. Verneau, B. Vernisse, G. Vigreux, J.-Y. Wallez, B. Watrin.
Netherlands: L.C. Baak, D.J. Bac, P.-J. Bus, R.K; Linskens, M. Oudkerk, A.C. Poen, M. Russel, T.G. Tan, L.A. Van Der Waaij.