Pharmacokinetics and *dynamics of Dabigatran Etexilate and Rivaroxaban in patients requiring PArenteral Nutrition (the PDER PAN study)

Chronic intestinal failure (IF) is caused by either surgically induced
anatomical short bowel, severe motility, or absorption disorders. These
patients require partial or total parenteral nutrition (PN and TPN,
respectively) and are thereby critically dependent on maintaining venous access
with a central venous catheter (CVC). Recurrent CVC-related thrombosis is a
serious complication and will ultimately lead to loss of central venous access
with intestinal transplantation being the only treatment option left.
Therefore, patients at risk for recurrent thrombosis and requiring (T)PN are
treated with long-term anticoagulants to prevent CVC-thrombosis, although no
prospective studies have assessed the efficacy and the safety of any long-term
anticoagulant treatment in these patients.
In patients with a normal intestinal absorption, INR-adjusted vitamin K
antagonists (VKAs) are the most used drugs for long-term anticoagulant
treatment. Most patients with intestinal failure requiring (T)PN cannot be
treated with these orally absorbed anticoagulants. This is due to unstable INR
levels mainly caused by variable drug absorption. In addition, most patients do
not use (T)PN daily, resulting in fluctuating vitamin K levels. Therefore,
subcutaneous low molecular weight heparins (LMWHs) or intravenous warfarin are
usually used for long-term anticoagulation in these patients. LMWH is
administered by daily subcutaneous injections, which is burdensome for patients
and can reduce compliance. Furthermore, LMWHs can lead to adverse reactions,
including allergic skin reactions, heparin-induced thrombocytopenia, and
osteoporosis.
A new generation of oral anticoagulants (NOACs), including dabigatran etexilate
and rivaroxaban, has recently been approved for prevention and the treatment of
venous thromboembolism (VTE) and ischemic stroke prevention in patients with
atrial fibrillation (AF). In contrast with VKAs, these drugs do not interact
with vitamin K metabolism, have a more predictable pharmacokinetic profile,
have less interaction with other drugs, and are prescribed at a fixed dose
without routine laboratory monitoring. Importantly, these drugs are absorbed
proximally in the gastrointestinal tract (stomach, duodenum, and proximal small
bowel). Therefore, NOACs may be an attractive oral alternative to both LMWH
injections and VKAs in these patients.

The aim of this phase I study is to assess the extent of intestinal absorption
of rivaroxaban and dabigatran etexilate in adult patients with short bowel
syndrome and treated with long-term TPN.

The study will be performed as an investigator initiated, single center,
randomized, phase I, cross-over study (Figure 1).
Patients will be screened before the randomization and instructed after
collecting the informed consent. The treating physician will screen patients
for eligibility.
On Day 0 patients will visit the Metabolic Research Unit and randomized in two
groups: subjects in Group 1 (n = 3) will take rivaroxaban 20 mg once daily (h
8.00) from Day 0 to Day 4 (five total doses); patients in Group 2 (n = 3) will
take dabigatran etexilate 150 mg twice daily (h 8.00-20.00) from Day 0 to Day 4
(nine total doses). The dose of either anticoagulant will be taken without the
consumption of food. After a period of at least 4 days of wash-out, patients in
both groups will switch to the other NOAC (dabigatran etexilate/rivaroxaban or
rivaroxaban/dabigatran etexilate) and the procedure will be repeated. The
patients will be randomized in two groups only for the order of NOACs.
Blood samples will be collected at the research unit at the following times: 1)
after rivaroxaban administration: - Day 0: T = 0 (= blank), T = 3 hours
following the first dose; - Day 4 at steady state: T = 0 (= trough), T = 1, 2,
3, 4, 5, 6, 8, 10, 24 hour(s) following the fifth dose. 2) after dabigatran
etexilate administration: - Day 0: T = 0 (= blank), T = 3 hours following the
first dose;
- Day 4 at steady state: T = 0 (= trough), T = 1, 2, 3, 4, 5, 6, 8, 12 hour(s)
following the ninth dose. At each timepoint 12 mL will be collected: one EDTA
blood sample will be collected for determination of rivaroxaban and dabigatran
concentrations, and two citrated blood samples will be collected for PD
parameters. A total of 276 mL of blood will be collected during the whole study
period. Plasma will be separated and stored at -20°C until analysis.
In exceptional situations, according to patient*s preferences and/or in case of
major personal limitations, blood samples will be collected at patient*s home
(Day 0 sample and T 24 sample only).
NOACs will be administered for 5 days to reach an appropriate steady state
level. On the basis of the calculated PK parameters in healthy volunteers,
rivaroxaban needs about 35-45 hours and dabigatran needs about 60-70 hours to
reach the steady state. In consideration of a theoretical reduction of the
absorption in these patients due to the intestinal resection, and in
consideration of a theoretical reduced bioavailability of the drugs, a 5-days
NOACs administration is reasonable.
The duration of the study for enrolled subjects can vary from 14 to 30 days
according to the length of the wash-out period decided by the patient.

Rivaroxaban and dabigatran etexilate will be used in the commercially available
formulation and will be obtained through the pharmacy of the Academic Medical
Center. Batchnumbers of medication will be recorded. Adherence to study
medication will be evaluated at the end of treatment checking the empty box.
For the treatment of acute VTE rivaroxaban and dabigatran etexilate have been
administrated in clinical trials at the doses used in our protocol (20 mg
once-daily and 150 mg twice-daily, respectively).

The burden for participants consists of 1) daily drug intake (5+5 consecutive
days), 2) one baseline visit before the randomization, and one day visit at the
hospital or at the patient*s home with placement of one peripheral venous
catheter for the blood samples (twice). An estimated total amount of 160 mL of
blood will be drawn per subject during the whole study.
The risks for all subjects participating in this study are the risks of a
normal venapuncture and the risk of bleeding due to the use of an
anticoagulant. A peripheral venous catheter for the sampling of blood will be
temporarily placed. CVC and/or catheter used for PN will not be used for
sampling of blood.
In the phase III trials in patients with atrial fibrillation, the risk of major
bleeding in patients treated with dabigatran etexilate and rivaroxaban was 2-4%
per year. In the phase-III Einstein PE clinical trial rivaroxaban at the same
dose of the present study showed an incidence of major bleeding events of 1.1%
per patient and estimated in 0.005% per day per patient (mean study treatment
duration was 216 days). In the phase-III RE-COVER VTE clinical trial
dabigatran etexilate at the same dose of the present study showed an incidence
of major bleeding events of 1.6% per patient and estimated in 0.009% per day
per patient (mean study treatment duration was 163 days).
The risk of bleeding of 5 days of treatment (twice) is therefore very small. If
the absorption in patients requiring TPN would be lower than in non-TPN
patients, these bleeding risks will likely be even lower.