Rectal Tacrolimus in the Treatment of Resistant Ulcerative Proctitis

The Inflammatory Bowel Diseases (IBDs) are life-long conditions often diagnosed between 15 and 35 years of age that are increasing in frequency. In developed countries, IBD prevalence was about 100-220/100,000 in 2002. More recently the estimated prevalence of CD and UC in the USA are 201 and 238 per 100,000 respectively. This indicates more than 620,000 CD and 730,000 UC patients in the USA alone. It has also been noted that the incidence of IBD has been increasing in the ethnic minorities in American over the last two decades and it is predicted that IBD, which was previously thought of as a 'Caucasian disease', will reach a comparable levels in Asia as in the Western Countries within 10-15 years. One of the major costs in IBD treatment is the need for a hospital admission. These have been falling for CD patients secondary to the increasing use of the biological therapies, however, admissions for UC have doubled over the last 15 years. IBD now costs the Australian economy of 22,000,000 people over $2.7 billion annually and this is comparable throughout the western World.

Tacrolimus and cyclosporin are classical calcineurin inhibitors and are widely used as immunosuppressive medications with promising results in UC. Calcineurin, or protein phosphatase 2B (PP2B), is a ubiquitously expressed cytosolic Ser/Thr protein phosphatase, that is highly conserved in eukaryotes. It has the ability to dephosphorylate a broad range of proteins and can regulate interleukin (IL)-2, IL-4 and interferon (IFN)γ expression, as well as modulating the activity of transcription factors like NF-κB. Enhanced NF-kB activity is well described in CD and UC and induces the proinflammatory cytokine IL-1β, IL-6 and TNFα expression. It is primarily through the reduction in the levels of these cytokines that clinical remission may be achieved.

The efficacy of oral tacrolimus has been examined in the management of medication resistant CD and UC. Unfortunately, the majority of these studies have been open labelled with only one randomised controlled trial reported in UC. This demonstrated a short-term clinical improvement but without a significant increase in the remission rate, potentially due to low patient numbers. Despite this, there are numerous open labelled studies in both UC and CD that suggest efficacy in the short term and with promising longer term data. The evidence would suggest, however, that the blood trough level should be at least 10ug/L in order to achieve the best efficacy (therapeutic range 5-20ug/L), but the higher the trough level the more likely a patient will suffer an adverse effect. These, unfortunately, can be numerous and include hypertension, nausea and diarrhoea, haematological abnormalities, renal impairment. Increase in the rate of skin cancers is also a concern, a concern supported by animal studies.

Use of topical tacrolimus has been effective in the treatment of perioral and perineal inflammation in paediatric CD patients with resolution of symptoms in 75%. Work examining topical perianal tacrolimus therapy in adult CD patients also demonstrated clinical efficacy, and although tacrolimus is absorbed well transdermally, only low trough levels of tacrolimus were detected in the blood. In these preliminary studies, the use of topical tacrolimus was associated with very few side effects. Long-term topical use, as with oral formulations, may be associated with an increased risk of skin cancer formation. Epidemiological evidence, however, would suggest that the risk is low and localised to the tacrolimus-treated sun-exposed skin.

The recent pilot study by our group demonstrated that 75% (6/8) of patients with resistant distal colitis responded and achieved a remission of their disease following 4 and 8 weeks of tacrolimus rectal ointment8. A dose of 0.3 to 0.5mg/ml 3ml twice a day was identified in the majority of patients to induce remission. In these patients tacrolimus trough levels were taken regularly and were either undetectable (<1.5 μg/L) sub-therapeutic (<5 μg/L) or, at their highest, in the low therapeutic range (therapeutic range 5-20 μg/L). The ointment was well tolerated without any systemic adverse effects. Use of the same rectal tacrolimus preparation in further 8 patients was also associated with a clinical response in 6 patients after 4 weeks of therapy.

The efficacy of topical tacrolimus in UC is further supported by a separate pilot study that examined topical tacrolimus in patients with resistant distal colitis. In this study a clinical and histological improvement in 10 of 12 patients by 4 weeks without any major side effects reported and the preparation was well tolerated.

METHODOLOGY

Study medication and supply:The preparation of the tacrolimus rectal ointment will be undertaken in respective hospital pharmacy at each centre. Gloves and mask shall be worn when making the preparation. The concentration of tacrolimus in the ointment shall be 0.5 mg⁄mL. Five millilitres of propylene glycol will be slowly mixed into the desired amount of tacrolimus powder on a clean glass slab. To this, 70 mL of paraffin liquid BP (LP) will be gradually added by serial dilution and triturate until evenly mixed. This process will be repeated with 125 mL of paraffin white soft BP (WSP). The resulting cream will be packed into tubes and labelled. The preparation will be formulated using the LP ⁄WSP base for ease of rectal use.

Medication Stability: The cream expires after 30 days at room temperature or 90 days at 4-8 0C

Administration of the study medication: A total of 3ml of the study medication or placebo will be applied rectally, via an applicator, twice a day by the patient over the 8-week period. The application will take place prior to retiring to bed and following breakfast in the morning. For shift workers application will occur just prior to retiring to bed and following the first meal of the day.

Dosage schedule:

1. Randomisation: Patients will be randomly assigned in a 1:1 ratio to receive the tacrolimus or placebo.
2. Blinding: The trial patients and the clinical investigators will be blinded to the randomisation.

Concomitant therapy: Rectal preparation will be ceased a day prior to commencing the trial. Patients will be continued on the same oral medications they are on at commencement of the study.

Screening visit: All patients will sign informed consent prior to any data collection. Patient demographics will be recorded. Stool culture and microscopy, blood pressure (BP), full blood count (FBC), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), urea,electrolytes (U&E) and GFR will be taken and measured locally at each site. Results of these tests will be faxed through to the central site at Fremantle hospital within 3 days of collection for review. Faecal calprotectin will be taken and stored at -200C. These specimens will be sent in batches on dry ice every 3 months to the central site at Fremantle Hospital for analysis.

Within 7 days prior to week 0 a flexible sigmoidoscopy will be performed. Two sets of photos will be taken at endoscopy one set distal (within 15cm of the anus) and one set proximal (15-25cm from the anus). The region depicting the worse level of inflammation will 2 biopsies will be taken from each region for histological analysis. Biopsies will be fixed and blocked at each hospital site. All the paraffin blocks will be sent to Fremantle Hospital be noted. Following the photos for histological analysis. Four additional biopsies will be taken from each region for RNA isolation. The biopsies from each region will be placed immediately into 2 tubes labelled 'proximal' and 'distal' containing RNAlater and stored at -80oC. These will then be sent in batches on dry ice every 3 months to the central site at Fremantle Hospital for analysis. A full Mayo score will calculated to confirm the presence of active UC (Mayo score between 6 and 12 required for inclusion).

Week 0: A Mayo score will be calculated using the screening endoscopic findings and treatment initiated. Baseline IBDQ calculated and BP measured

Weeks 2, and 4: A modified Mayo score will be calculated, BP taken, and FBC, ESR, CRP, U&E as described above and the IBDQ assessed. The faecal calprotectin will be taken and stored at -200C. Blood for the tacrolimus trough will be taken, the serum collected and this will be stored at -20oC. The serum will then be sent on dry ice to the central site at Fremantle Hospital for assessment within 2 days of collection. The medical monitor (CIA) at Fremantle Hospital will review all the tacrolimus trough levels to ensure patient safety.

Week 8 or withdrawal: A flexible sigmoidoscopy will be undertaken ± 3 days of the week 8 visit (or at early withdrawal) and a full Mayo score calculated. Photos and biopsies will be taken and processed as detailed at the screening sigmoidoscopy. Routine FBC, U&E, ESR, CRP, BP measured, and faecal calprotectin as described above

Week 12: Follow-up phone call to assess any adverse events. Each histological specimen will be assessed centrally at Fremantle Hospital by a blinded specialist gastroenterological histopathologist. These findings will be correlated with the endoscopic scoring and photos taken at endoscopy.

Early withdrawal from the study: Patients are free to withdraw from the study at any time. If the patient is withdrawing due to lack of efficacy of the study preparation 4 weeks or more following the week 0 visit, they will be eligible for open label use topical tacrolimus ointment for 4 weeks. A withdrawal flexible sigmoidoscopy will be required prior to the supply of the topical tacrolimus ointment. Photos will be taken prior to the biopsies as described for the week 0 and 8 procedures. Routine FBC, U&E, ESR, CRP, BP measured, tacrolimus trough levels and faecal calprotectin will be taken after 2 weeks of therapy as described above. These patients will be eligible for analysis for the primary and secondary end points

Interim Analysis: An interim analysis by an independent statistician and clinician after 20 patients have completed the 8-week visit will be done to confirm the drug safety and to assess the efficacy. This shall be required prior to allowing the study to continue.

mRNA expression levels: Biopsies (x4) taken from two regions for the distal colon (within 15cm of the anus and 15-25cm from the anus). The region containing the worse inflammation seen on endoscopic examination at weeks 0 and 8 (or at early withdrawal) will have been noted. All samples will have been immediately placed in RNA later and stored at -800C prior to transfer on dry ice to the central site at Fremantle Hospital. Levels of TNFα, IFN-γ, IL-1β, IL-2, IL4, IL-10 and IL-6 mRNA expression will be determined on each group of samples. Total RNA will be isolated and DNase-treated using the Qiagen 96 RNeasy® kit (Qiagen, Clifton Hill, Australia) according to the manufacturer's instructions. The quality of the RNA samples will be verified, reverse transcription performed and cDNA quantitated by real-time PCR using QuantiTect SYBR Green Master Mix (Qiagen) on the ABI PRISM 7900HT (Applied Biosystems). Specific primers for the mRNAs will be used. Dissociation curve analysis will confirm specificity of amplification and copy numbers will be determined from plasmid or PCR product standard curves. Target mRNA copy numbers will be normalised against reference gene copy number for each cDNA sample. The above methods are already available in CIA's laboratory at Fremantle Hospital. The limit of mRNA detection is 3pg/ml for all cytokines

Statistical analysis: Comparisons between the mRNA expression in endoscopic biopsies taken at weeks 0 and 8 (or at withdrawal) in the tacrolimus treatment arm will undertaken and compared with each other and against those from the placebo control arm at the same time points. The mRNA expression levels will be correlated with the endoscopic and histological inflammatory scores. Differences between the groups will be assessed using one-way ANOVA on log-transformed data. Statistical analysis will be conducted using the Statistical Package for the Social Sciences (SPSS) version 17. Statistical significance is considered as p<0.05 for all analyses.