The angiodysplasias may be responsible for recurrent gastrointestinal bleeding and in some cases bleeding remaining inaccessible to endoscopic treatment. Several observational studies suggest that treatment with somatostatin analogue would reduce transfusion requirements in patients with recurrent bleeding due to angiodysplasia.

No randomized studies are available. The main objective of this study multicenter, prospective, randomized, was to assess the transfusion requirements in patients with recurrent bleeding due to angiodysplasia treated by a new analogue of somatostatin, Pasireotide, versus placebo.

Patients with recurrent gastrointestinal bleeding related to angiodysplasias, endoscopic treatment failure, with a need transfusion at least 6 red blood cells during the 6 months prior to inclusion could be randomized to receive monthly intramuscular injection of Pasireotide 60 mg or placebo for a period of 6 months. Patients were then followed for an additional 6 months after stopping treatment. A test monthly clinical and laboratory was performed during the six months of treatment then quarterly during the six months of surveillance.

The angiodysplasias are abnormal communications between dilated capillaries and veins. Their presence in the digestive tract is probably due to venous obstruction associated with smooth muscle contractions. These vascular malformations are not a particularly significant cause of bleeding in elderly patients. The angiodysplasias are responsible for about 1% of upper gastrointestinal bleeding and up to 6% of the lower bleeding. However in case of gastrointestinal bleeding obscures the presence of angiodysplasia was found in 23% of cases through an exploration of the small intestine endoscopic video capsule enteroscopy or double balloon. Some conditions promote the development of angiodysplasias such as kidney failure with a prevalence of 13% in case of bleeding or cirrhosis. The responsibility of aortic stenosis is debated and could be a chance association. On the other hand, are more symptomatic bleeding in case of bleeding disorders, in particular von Willebrand disease.

The first line treatment is based on the destruction of endoscopic lesions by laser, or argon plasma coagulation electrocoagulation. However, these lesions may be inaccessible to endoscopic treatment, when they are too numerous or if patients with severe comorbidities, indicating against invasive endoscopic treatments. Iterative transfusions remain the only possible treatment for a significant proportion of patient which induces frequent hospitalizations, complications related to transfusions and a significant cost.

Hormonal treatments have been tried without success to prevent rebleeding. Anti-angiogenic as thalidomide have been tested in phase II trials in low number of the patients.

Somatostatin is a cyclic peptide secreted by D cells in the gastric and intestinal mucosa, by the cells of the islets of Langerrhans. There are five known somatostatin receptors which when activated by the ligand caused an inhibition of gastric acid secretion, pancreatic and biliary secretions. The somatostatin analogues have different vascular effects documented as downregulation of VEGF expression or decreasing the splanchnic flow.

Thus, the somatostatin analogs can have an effect on the prevention of rebleeding secondary to angiodysplasia by several mechanisms: inhibition of angiogenesis, reduction of splanchnic blood flow, increased vascular resistance and increased platelet aggregation.

Several cohort studies suggest that treatment with somatostatin analogues, octreotide can reduce the need for transfusion in patients with recurrent bleeding due to angiodysplasia. A meta-analysis of these studies, which included 62 patients in total found a relative risk of rebleeding of 0.76 (95% CI: 0.64 to 0.85) and a decrease of 2.2 transfusions (95% : -3.9 to -0.5) between the period before the treatment and the period of treatment with octreotide.

To date, there are no randomized phase III study to suggest the efficacy of somatostatin analogues in the prevention of rebleeding due to angiodysplasia.

Pasireotide is a new analogue of somatostatin which has a higher affinity of Octreotide to like receptors 1, 3 and 5. A preliminary study showed that pasireotide could be effective in the symptomatic treatment of endocrine tumors beyond the octreotide therapy. This superior efficacy of pasireotide compared to Octreotide justifies the evaluation pasireotide in the prevention of rebleeding due to angiodysplasia. Side effects described with the LP pasireotide are the possibility of post prandial hyperglycemia transient dose-dependent. Episodes of intestinal disorders have been reported (diarrhea, nausea, vomiting), but usually require no medicinal treatment and disappearing spontaneously during treatment.

All these data suggests for a randomized study against placebo to establish the efficacy of treatment with pasireotide LP in the prevention of rebleeding due to angiodysplasia. A randomized phase II feasibility study will be conducted to verify the the investigators hypotheis and followed if necessary by a phase III study.

Somatostatin analogs have vascular action that has been shown in other pathologies (gastrointestinal bleeding in patients with portal hypertension). Cohort studies suggest efficacy in reduction of transfusion requirements in patients with gastrointestinal angiodysplasia. In case of failure of endoscopic treatment there is no currently licensed treatment for the prevention of rebleeding in these patients.

The investigators objective is to evaluate the reduction in the number of red blood cells transfused to M6 in patients treated with pasireotide LP or placebo.

Inclusion Criteria:

1. Angiodysplasia of the stomach, small intestine or colon confirmed by endoscopy.
2. 6 or more of packed red cells unit transfusion during the 6 months prior inclusion.
3. Failed endoscopic therapy or cons-indication for endoscopic treatment.
4. Patient affiliated to a social security insurance.
5. Age > 18 years.
6. Consent signed by the patient.

Exclusion Criteria:

1. Treatment with somatostatin analogue in the 6 months prior to inclusion
2. Symptomatic cholelithiasis
3. Rendu-Osler disease
4. Uncontrolled diabetes (HbA1c > 8%)
5. Breaking of the esophageal varicose veins bleeding older than six months.
6. Patients treated with anti vitamin K at baseline and during the study.
7. Patients with (AST, ALT> 2 ULN) and / or total bilirubin > 1.5 ULN.
8. TP < 50%, platelets <75 000/mm3, aPTT> 1.5 times the control
9. Uncontrolled heart disease: myocardial infarction within 6 months, status epilepticus angina, congestive heart failure grade III and NYHA, ventricular tachycardia, ventricular fibrillation, heart block, severe
10. Family medical history of the idiopathic sudden death
11. Syncope like medical history
12. QTcF> 450 ms
13. Metastatic malignancy
14. Pregnant or nursing women, women of childbearing age who have not achieved pregnancy test, women and men of reproductive age without effective contraception
15. Impossible follow for psychological and/or geographical reasons.