Drug Name: Bosentan (Tracleer^TM)

Medical Reviewer: Maryann Gordon, M.D.

Table of contents

Summary ……………………………………………………………………..	2
1.0 Overall clinical program………………………………………………...	3
2.0 Demographics……………………………………………………………	5
3.0 Deaths……………………………………………………………………..	11
4.0 Serious adverse events…………………………………………………...	18
5.0 Premature withdrawals because of adverse events…………………….	24
6.0 Liver toxicity……………………………………………………………...	30
7.0 Anemia…………….……………………………………………………...	45
8.0 All adverse events………………………………………………………...	56
9.0 Vital signs…………………………………………………………………	64
10.0 ECG changes……………………………………………………………	66
11.0 Clinical pharmacology………………………………………………….	68
Appendix ……………………………………………………………………..	74

Summary of safety

The safety of oral bosentan has been tested in just over 1000 patients[1]. The indications that have been studied include pulmonary arterial hypertension (for which this application has been submitted), congestive heart failure (still active with a large, double blind morbidity/mortality program scheduled to finish within the year), essential hypertension (discontinued) and subarachnoid hemorrhage (discontinued). Bosentan has been studied over a fairly wide dosing range (total daily doses 100 mg to 2000 mg). Twenty-nine patients with pulmonary hypertension have been treated for at least 1 year.

Adverse events leading to premature withdrawal occurred more frequently in the bosentan group (11.1%) compared to placebo (9.4%) and the most common reason for discontinuation in the bosentan group was abnormal hepatic function (3.4% placebo subtracted). Events leading to withdrawal but reported only in the bosentan group include anemia, hypotension, diarrhea, dyspnea, flushing, lower limb edema, pyrexia, vomiting, abdominal pain, face edema, and myocardial infarction.

Bosentan provokes elevation of LFTs, primarily ALT and AST (ALT for one patient[2] was increased 73 times over his baseline value), with examples of positive drug rechallenge. The increases in alkaline phosphatase tended to be less striking. Approximately 10 to 13% of bosentan patients had LFTs at least 3 times upper limit of normal and 4% had values greater than 8 times upper limit of normal. Overall, the higher the dose and the longer a patient takes any dose, the higher the risk of having an elevated LFT.

There are reports of patients with elevated bilirubin as well as two reports of jaundice[3] (plus an additional report of jaundice from the ongoing CHF trial (ENABLE)). There are examples of patients with elevated LFTs reporting abdominal pain, fever, flu-like syndrome.

Understanding that there is only a small population from which to draw conclusions, there is no indication at this time that the increase in liver enzymes will not be reversible, at least in the majority of patients, with discontinuation of bosentan, or an attempt to lower the dose. There is no indication that bosentan was tied to any death (with the possible exception of the REACH study that had an increase in deaths in the fast titration group). As of June 19, 2001, the sponsor has not received a report (s) of liver failure, liver transplant, or a death because of liver impairment in any patient who has received bosentan[4]. The protocols routinely excluded patients with AST/ALT > 3 times upper limit of normal and this should continue to be a contraindication if the drug receives approval.

In addition to the elevation of LFTs, there is a dose related mean decrease in hemoglobin of nearly 1 g/dl in patients taking bosentan for more than 12 weeks. About 3% of patients reported serious anemia requiring withdrawal and/or transfusion. (In comparison, there were 2 placebo patients (0.9%) who received blood transfusion and none was withdrawn for anemia.) The one case of bone marrow biopsy[5] reported as normal is reassuring. The cause of the anemia is unknown. The sponsor’s explanation that it is all the result of hemodilution seems unlikely.

Although anemia can be a serious side effect, there is no indication that there is irreversible harm done to patients who develop anemia when taking bosentan. Thus far, it appears that by stopping bosentan when patients show an unacceptable drop in hemoglobin/hematocrit results in patients returning to baseline levels. The protocols routinely excluded patients with hemoglobin and/or hematocrit <30% below normal limits.

Additional dose related adverse events include headache, flushing and edema (definitely peripheral and perhaps face as well). Although there was no evidence that bosentan is associated with hypotension, the protocols routinely exclude patients with systolic blood pressure < 85 mmHg.

Although there was one report of torsade de pointes in a patient receiving bosentan, there is no indication that bosentan affects electrical activity in the heart.

Medications whose concomitant use with bosentan is to be contraindicated include cyclosporin and ketoconazole. Great care should be used when bosentan is coadministered with any CYP3A4 inhibitors, especially with the first bosentan dose. This drug-drug interaction does present a predicament since there are so many drugs that inhibit CYP3A4. Bosentan also can be expected to decrease the concentrations (and may be the effect) of CYP3A4 and CYP2C9 substrates.

Other drugs that the sponsor prohibited study patients from taking include glibenclamide, encainide, flecainide, disopyramide, propafenone, moricizine, pinacidil, minoxidil and oral positive inotropic agents other than digitalis.

Patients receiving warfarin and bosentan must be monitored for decreases in prothrombin time. There are no data supporting the concomitant use of bosentan and Flolan therapy.

1.0 Overall Clinical Program

The bosentan program contains both completed and ongoing safety and efficacy studies involving several indications. There are also various clinical pharmacology studies.

The indications currently being pursued are pulmonary arterial hypertension (PAH) and congestive heart failure (CHF). Both the hypertension (HTN) and subarachnoid hemorrhage (SAH) indications were discontinued because of safety considerations (liver enzymes increase and decrease in hemoglobin) and lack of effect, respectively.

1.1 Completed studies

- 3 studies in PAH: 2 double blind, randomized efficacy trials and 1 open label safety study;

- 7 studies (> 2 weeks duration) in patients with CHF, HTN, or SAH;

- 23 clinical pharmacology studies and studies with special objectives (< 2 weeks duration).

1.1.1 PAH

The efficacy and safety of bosentan in the treatment of patients with PAH are based on the following studies:

- randomized, placebo-controlled, double-blind trial (AC-052-351) using doses of 62.5 mg bid up to 125 mg bid for 12 weeks. A total of 32 patients were enrolled.

- AC-052-352 (BREATHE-1): double blind, placebo controlled randomized 16-week efficacy study using 62.5 mg bid up to 250 mg bid that enrolled 214 patients. The study was submitted for review on May 18, 2001. There was a complete medical review of efficacy and safety of the study, but because it was submitted so late, its integration with the rest of the safety program is limited.

- small open label, single iv dose followed by multiple oral doses (1000mg bid) hemodynamic exploratory study (BD14884) used for safety only.

1.1.2 Other indications

Table 2 shows the 7 completed[6] studies of at least 2 weeks duration conducted in patients with diseases other than PAH.

a) CHF trials

-NC15462 (REACH) double blind, placebo controlled, 6 month study that enrolled 307 patients. Bosentan doses were 125 mg bid to 500 mg bid. Study was stopped because there was increased incidence of elevated liver enzymes and increased incidence of decreased hemoglobin. The primary efficacy objective was not met. Additional CHF trials (ENABLE) are using lower doses (62.5 mg bid to 125 mg bid).

b) HTN trial

NC15020 double blind, placebo controlled, 4 week efficacy trial that was stopped prematurely because of increased incidence of elevated liver enzymes. A total of 293 patients entered the study with doses of 100 mg qd to 1000 mg bid.

1.2 Ongoing studies

Interim reports containing data collected up to the clinical cut-off for the following ongoing, open-label or blinded studies:

PAH

- AC-052-353: small, ongoing open-label extension study of AC-052-351 (base study) using doses of 62.5 mg bid up to 125 mg bid in patients with PAH;

CHF

- NC15464B: ongoing open-label extension study of NC15462 in patients with severe CHF;

- AC-052-301 and AC-052-302 (The ENABLE Trials): ongoing, large, still blinded morbidity/mortality trials in patients with severe CHF.

1.3 Clinical pharmacology studies

There are 23 studies with healthy volunteers and various patient populations designed to characterize the pharmacokineties of bosentan and to assess potential interactions with other medications. These include:

- 3 single-dose i.v. studies in healthy volunteers;

- 5 single-dose i.v. studies in patient populations;

- 6 single-dose oral studies in healthy volunteers;

- 3 single-dose oral studies in patient populations;

- 3 multiple-dose oral studies in healthy volunteers;

- 7 drug-drug interaction studies.

1.4 Safety update

Data from the studies shown below have been added to the safety update.

2.0 Demographics

2.1 Numbers of patients

2.1.1 All bosentan patients

Total number of all subjects[7] in bosentan database (regardless of treatment group) at the time of the NDA submission is 759. The number of patients studied for the various indications are shown in the following chart.

There were 447 patients with CHF, 39 patients with PAH, 243 patients with HTN and 30 patients with SAH in a total of 10 studies (limited to those with a duration > 2 weeks).

The numbers of patients in the various types of trials are shown below.

A total of 571 patients received bosentan in 1 of 10 clinical trials.

2.1.2 All Placebo controlled trials

A total of 533 patients[8] received bosentan and 219 patients received placebo in the 7 placebo controlled clinical trials. There were 38 bosentan patients (who are not part of the 533) in open label extension trials. The 23 clinical pharmacology trials enrolled 571 subjects with 434 of these having received bosentan. This is a very small safety data base.

At the time of the NDA submission, the ongoing ENABLE trials (in CHF) had 1400 patients enrolled which increased to 1613 at the time of the submission of the safety update. The ENABLE trials remain blinded and are expected to be completed November 2001.

2.1.3 Safety update

There are 29 PAH patients in the open label extension study AC-052-353. These patients are carry overs from study AC-052-351.

A total of 427 new patients have been added to the bosentan safety database: 214 patients were recruited into the ongoing BREATHE-1 trial (AC-052-352, the second efficacy trial in PAH) and 213 patients were added to the 1,400 patients of the ongoing ENABLE trials (CHF) for a total of 1613.

2.2 Demographics and duration of exposure

2.2.1 All bosentan patients

Demographics and characteristics of the 571 patients who received bosentan in 1 of the 10 therapeutic trials are shown in the table below.

Most patients were male (71.5%) and white (91.6%), with a mean age of 59.6 years (range 25 to 90 years). Over half of the study patients were CHF patients, about one third were HTN patients and a minority (5.8%) were PAH patients. Total daily doses ranged from 100 mg to 2000 mg. Nearly two thirds of study patients received daily doses of > 1000 mg.

A figure showing the duration of exposure is displayed below.

Approximately 50 % of patients received bosentan for around 4 weeks or less. According to the above figure, at least 68 patients have received bosentan for 300 days.

2.2.2 PAH

a) Demographics

The table below displays information on the 31 patients who participated in the efficacy study and the 29 patients who entered into the extension study.

Most patients were white and female with a mean age around 50 years. Primary pulmonary hypertension was a more common etiology than primary pulmonary hypertension with systemic sclerosis. Mean amount of time from PAH diagnosis to randomization was less than 2 years.

b) Duration of exposure

All patients randomized to bosentan in study AC-052-351 completed the 12 week double blind phase and these patients were allowed to continue taking bosentan past the 12 weeks. Patients who had been randomized to placebo were allowed to receive bosentan after completing the double blind part of the study. The extension study AC-052-353 accepted patients who had been enrolled into and completed AC-052-351. All patients who entered the (ongoing) extension study received open label bosentan. The mean duration of treatment was 227+ 35 days for the 21 patients who had received bosentan in the base study and 98+ 8 days for the 9 patients who had received placebo in the base study.

2.3 Numbers of patients who stopped treatment for any reason

2.3.1 Placebo controlled trials

The table below shows the number and percent of patients who dropped out of a trial, by reason and treatment group (from fax sent 5-23-01).

Approximately equal incidence rates were reported for all study withdrawals: bosentan 45.0% and placebo 45.2%. The most common reason was administrative/other, which is puzzling. Adverse event was also common. It can usually be assumed that a patient’s decision to withdraw results from an (unreported) adverse event.

2.3.2 Open label trials

The incidence rates for patients in open label bosentan treatment are shown below by reason.

Death and adverse events were the most common reasons for withdrawing from bosentan treatment in the open label trials. The 7 patients who decided to stop probably did so because of an adverse event. The incidence rate for withdrawal for elevated LFTs was 3.3%.

3.0 Deaths

3.1 All bosentan patients

The deaths reported for the all bosentan population (n=571) are shown below by cause of death. The mean duration of exposure to bosentan for this group was 154 days (range from 1 to 1098 days).

A total of 51 (8.9%) deaths were reported: 50 deaths occurred in patients participating in CHF NYHA class III-IV trials (26 in placebo controlled trial NC15462B, REACH, and 24 in open label ongoing extension NC15464B) and 1 death occurred in the SAH trial (death attributed to cerebral infarction).

Most of the causes of death appear to be related to underlying cardiac disease: sudden death (9 patients, 17.6%), cardiac failure (7 patients, 13.7%), and myocardial infarction (6 patients, 11.8%). Two patients died of multi-organ failure and there were 3 deaths for which the cause was unknown/not reported.

There were 24 deaths in the bosentan 250 mg and 27 in the bosentan 1000-1500 mg daily dose groups.

Three deaths occurred in patients who were or had received iv bosentan (patient #806 died with bronchopneumonia, patient #03 died with hypotension, oliguria, thrombocytopenia, and patient #107 died with hydrocephalus and CVA). The deaths occurred either during or shortly after drug was administered. The iv formulation is no longer being developed.

3.2 All placebo controlled trials

There are 7 placebo controlled studies with a total of 219 placebo patients and 533 bosentan patients. Deaths reported for patients who participated in these trials are shown below.

The incidence rates for death are similar for the 2 treatment groups: 5.9% [CU1] for placebo and 5.1% for bosentan. The most common causes of death for placebo and bosentan groups were sudden death and cardiac failure. Most deaths were attributed to cardiovascular causes, which is to be expected in this predominantly CHF population.

There were 2 bosentan deaths attributed to renal failure:

-subject 18111/6022 was a 64 year old with ischemic heart disease, NYHA class IV, ejection fraction 15%, post MI and post CABG received bosentan 500-1000 mg daily for 12 days. The study medication was discontinued because of worsening heart failure. He made a temporary recovery but then developed acute renal failure secondary to low cardiac output. The patient had steady deterioration followed by sudden death.

-subject 18195/9068 was a 66 year old with ischemic heart disease, NHYA class IV, ejection fraction 29% and a history of COPD, MI, stroke and coronary artery bypass surgery received bosentan 500-1000 mg daily until day 44 when he was discontinued on day 44 because of worsening CHF and chronic renal failure. His death, on day 70, was attributed described as acute renal failure. No autopsy was performed.

Deaths with addition of AC-052-352

The numbers and percents of patients who died in the placebo controlled trials (including AC-0520352) are shown below.

from fax dated 6-04-01

There is a slightly higher death rate in the placebo group (5.2%) compared to the bosentan group (4.6%). Most deaths were cardiovascular in nature.

3.3 REACH--Protocol NC15462B

This was a large placebo controlled trial with CHF NYHA class III-IV patients. There were 370 patients randomized to placebo, bosentan fast titration to 500 mg bid or slow titration to 500 mg bid. Patients were followed for 26 weeks. The numbers of deaths per group are shown below:

Number and (percent) of patients

Placebo

N=126

Bosentan

slow titration

N=121

Bosentan fast titration

N=123

Total

bosentan

N=244

Death

11 (8.7)

9 (7.4)

17 (13.8)

26 (10.7)

Table 18 vol 40.

The death rates were similar for placebo (8.7%) and the bosentan slow titration group (7.4%) but higher for the bosentan fast titration group (13.8%). Overall, there is a small imbalance in the death rate favoring placebo.

The figure below is the Kaplan-Meier plot of time to death or heart failure morbidity for study NC15462B.

The break down by reason for the deaths is shown below.

The deaths do not seem unexpected in a heart failure population, but the imbalance is not reassuring.

There is no evidence that bosentan adversely affects mortality, at least in CHF patients, but there is no convincing evidence that it does not. The analysis of mortality from the large ENABLE trials may clarify this issue.

3.4 Safety update

3.4.1 PAH

a) ongoing, open label trial

There are 29 patients in the extension study AC-052-351. No deaths have been reported.

b) ongoing, blinded trial (AC-052-352)

A total of 214 patients have been randomized to bosentan 62.5mg-250 mg bid or placebo. At the time of the safety update, 2 deaths have been reported:

Patient 10014 was a 48-year-old white female with pulmonary hypertension due to systemic sclerosis. Medical history included cardiac failure, anemia, thrombocytopenia, lupus erythematosus, Raynaud's syndrome, rectal bleeding, urinary tract infection, cholelithiasis, hip arthroplasty and constipation. Concomitant medication included prednisone, acetylsalicylic acid, omeprazole, fluorazepam, paracetamol, docusate, and calcium. The patient reported nausea treated with promethazine on day 1, dyspepsia on day 6, tinnitus on day 31, and arthralgia on day 87. LFTs were found to be severely increased on day 99 (AST 659 U/L, ALT 554 U/L, alk phos 275 U/I). Total and direct bilirubin were high (22 and 36 mmol/1) Study drug was permanently discontinued on day 116. Hepatic function tests returned to normal by day 141. The patient reported to her local pneumologist on day 145 with symptoms of worsening pulmonary hypertension. After admission to the local hospital, epoprostenol therapy was initiated. Inotropic drugs were started and mechanical ventilation was needed. Three days later the patient died. The cause of death was reported as right heart failure secondary due to worsening pulmonary hypertension. No autopsy was performed.

Patient 10070 was a 50-year-old white female with PPH. Medical history included monoclonal gammopathy, irritable bowel syndrome, and anxiety. Concomitant medication included furosemide, spironolactone, digoxin, warfarin, fluticasone, salmeterol, promethazine, clonazepam, diazepam, paracetamol, codeine, dicycloverine, diphenydramine and estrogens. She collapsed at home on day 19 with cardiac arrest. CPR was unsuccessful. An autopsy was performed and the death was attributed to worsening PPH.

3.4.2 CHF

a) ongoing open label

There are 27 ongoing patients and 3 additional deaths. Overall, there have been 26 deaths in this long term extension study with severe heart failure patients.

Patient 20252/1283 had worsening heart failure and renal function. He continued to deteriorate, all medication was discontinued and he died 3 days later.

Patient 20266/1641 complained of abdominal pain with nausea and vomiting and was found unresponsive the next day. Resuscitation was unsuccessful. He died on study day 848. Patient was anemic (hb: 11.4 g/dl and hct: 36.5%). GTT was elevated (76 U/L). History included diabetes with worsening renal failure and excised melanoma.

Patient 20086/42 was a 79-year-old white male patient with ischemic heart disease, NYHA class IIIb-IV enrolled into the open-label study. Concomitant medication included enalapril, furosemide, digoxin, perhexiline, metoprolol, isosorbide mononitrate, acetylsalicylic acid, indomethacin, diclofenac, imipramine, temazepam, and ferrous supplements. He reported 2 episodes of gout, 2 vasovagal episodes that resolved spontaneously, and blurred vision resolved without treatment, upper respiratory tract infection that resolved with treatment. He had 2 episode of shortness of breath treated with readjustment of diuretic or rest. These events were reported months prior to his sudden collapse and death at home. He had been on study drug for about 2 years. No autopsy was performed.

b) ongoing, double blind (ENABLE trials)

There have been a total of 187 deaths in 1613 study patients. The reasons given for the deaths are shown below.

Until the study is completed and analyzed, it is difficult to draw conclusions about the deaths in patients with advanced heart failure.

3.5 Abrupt bosentan withdrawal

Reviewing the 4 deaths reported for PAH patients receiving bosentan (AC 052 352) discloses no evidence that abruptly stopping bosentan leads to death.

4.0 Serious adverse events

Clinical adverse events are included that were reported during or up to 28 days after treatment was stopped. Those events identified as part of the efficacy endpoint of a particular trial (e.g., worsening heart failure in CHF studies), however, were not consistently reported in safety so some numbers are unreliable.

4.1 All bosentan patients

Ten studies were conducted with bosentan in various indications with a total of 571 bosentan patients. Serious adverse events reported by at least 0.5% (3) of all bosentan patients are shown below, by indication as well as for the total bosentan population.

Overall, 16.8% of bosentan patients reported at least 1 serious event. For the indication with the largest number of patients (CHF with 323 patients), 27.2% reported at least one event.

The most commonly reported serious events were atrial fibrillation, pneumonia, cardiac failure, and cerebrovascular accident. Renal failure was reported by 1.1%, anemia 0.9%, abdominal pain by 0.7%, and abnormal hepatic function 0.5% of the total bosentan population.

4.2 All placebo controlled trials

There were 7 placebo controlled studies with a total of 219 placebo patients and 533 bosentan patients. Serious adverse events reported by at least 2 bosentan patients are shown below.

More serious events were reported for the placebo patients (16%) compared to the bosentan patients (12%). The incidence rate for reporting serious hepatic function, in this table, is 0.6% for the bosentan patients compared to 0 for the placebo patients. However, there were many more bosentan patients with abnormal hepatic function than the 3 reported here (see section 6). Also, anemia, sometimes requiring transfusions, were reported but not classified as a serious event (see section 7.0).

4.3 Serious events by dose

Most events were reported for patients on the higher doses, in part because most study patients were taking least 1000 mg daily.

There were 54 patients (17.4%) who received bosentan 1000-1500mg daily and reported at least 1 event compared to 31 placebo patients (16.8%). Events reported by at least 3 bosentan patients included renal failure (4 reports), pneumonia (4 reports), atrial fibrillation (3 reports), and abnormal hepatic function (3 reports).

There were 8 patients (7.8%) who received bosentan 2000 mg daily and reported at least 1 event compared to 3 patients (4.1%) who received placebo. Events reported by at least 2 patients included cardiac failure (3 reports) and dyspnea (2 reports). There were no reports of abnormal hepatic function with this bosentan dose.

4.3.1 REACH--Protocol NC15462B

This was a large placebo controlled trial with CHF patients. There were 370 patients randomized to placebo, bosentan fast titration to 500 mg bid or slow titration to 500 mg bid. Patients were followed for 26 weeks. The table below briefly discusses events reported by bosentan patients that are not expected in a CHF population.

Patient no.	Adverse event
18102/6162	Abdominal pain, elevated LFTs, diagnosis of cholelithiasis, torsades de pointe (discussed in section 6.0)
18106/6072	Abnormal hepatic function (discussed in section 6.0)
18108/6003	Postural changes on day 7, 3 episodes of syncope day 42 with SBP 68 mmHg. Received pacemaker day 48.
18120/2051	Vomiting, watery stools, syncope day 169
18122/2001	Pyrexia, elevated LFTs (discussed in section 6.0)
18127/8085	Duodenal ulcer and transfusion
18136/8143	Severe abdominal pain with diarrhea possible resulting from gastroenteritis
18141/6263	Severe abdominal pain with metastatic carcinoma
18149/6374	Fever, malaise, and elevated LFTs (discussed in section 6.0)
18166/6473	Severe abdominal pain, diaphragmatic hernia, worsening heart failure
18197/9006	Severe elevation of LFTs and mild dermatitis (discussed in section 6.0)
18197/9007	Decreased hemoglobin requiring transfusion (discussed in section 7.0)
18199/9071	Anemia requiring transfusion (discussed in section 7.0)
18207/4031	Flushing, lightheadedness, palpitations, tremor day 15
18219/3020	Asthenia, diarrhea, fever, elevated LFTs, hepatomegaly (discussed in section 6.0)
18229/3081	Syncope

Reported serious adverse events with addition of AC-052-352

The numbers and percents of patients with an adverse event that was a) reported by at least 2 bosentan patients and b) reported more often by bosentan patients than placebo patients are shown below.

No. and (percent) of patients

	Placebo N=288	Bosentan N=677	Placebo subtracted (%)
Any event	51 (17.7)	92 (13.6)	-4.1
Pneumonia	1 (0.3)	6 (0.9)	0.6
Pyrexia	0	3 (0.4)	0.4
Abdominal pain	1 (0.3)	4 (0.6)	0.3
Pneumothorax	0	2 (0.3)	0.3
Intestinal obstruction	0	2 (0.3)	0.3
Myocardial infarction	1 (0.3)	4 (0.6)	0.3
Gastroenteritis	1 (0.3)	3 (0.4)	0.1
Abnormal hepatic function	1 (0.3)	3 (0.4)	0.1

from fax dated 6-01-01

Overall, there were more patients in the placebo group who reported serious adverse events. The events with the highest placebo subtracted incidence rates included pneumonia, abdominal pain and pyrexia, all less than 1%.

4.4 Safety update

4.4.1 PAH

a) ongoing, open label trial

There are 29 patients in the extension study AC-052-351. Two serious adverse events have been reported:

Patient 10202 reported cramping abdominal pain and fever and was hospitalized on day 190. She also had new onset atrial fibrillation. Laboratory values were reported to be within normal limits except prothrombin time (elevated but INR was low). Patient was taking warfarin.

b) ongoing, blinded trial

A total of 214 patients have been randomized to bosentan 62.5mg to 250 mg bid or placebo. At the time of the safety update, there were 16 reports of serious adverse events: bronchitis (2), intestinal obstruction, renal failure, syndrome of malaise (with fever, chills, and elevated LFTs),worsening pulmonary hypertension leading to death, anemia needing transfusion, hyperkalemia, abdominal pain thought to be attributed to diverticulitis, hypokalemia, abdominal pain and diarrhea, atrial fibrillation/flutter, vomiting and abdominal pain with normal liver and renal function, gall bladder disease and abdominal pain, abdominal distension of unknown etiology, hypotension and bradycardia.

4.4.2 CHF

a) ongoing open label (NC 15464)

There are 27 ongoing patients out of 86 enrolled. The dose in this long term, open label study is 125 mg bid.

There were 41 patients with reports of serious adverse events at the time of the ISS cut off date (n=86). The events are listed below:

Acute bronchitis, CVA

CVA, death

Unstable angina

Atrial fibrillation, ventricular fibrillation, respiratory failure, death

Right ventricular failure

Pneumonia, cardiac failure (x2), atrial fibrillation

Atrial fibrillation, diabetes mellitus

Head trauma, subdural hematoma

Abdominal pain, jaundice (20060/00841, discussed in section 6.0)

Septicemia

Abdominal angiogram for aneurysm

Thyroidectomy

Chest pain (x5), anemia requiring transfusion, death (discussed in section 7.0)

Renal failure, death

CVA, death

Metastatic melanoma, anemia requiring transfusion

Cataract extraction, foot ulcer, amputation, wound infection,

COPD, death

Loin pain, angina, pneumonia, renal failure, cardiac failure, death

Hemorrhage, anemia requiring transfusion, hematuria after bladder surgery, abdominal pain, urinary retention, pnemonia, death (see section 7.0)

Skin carcinoma

Anemia, cardiac failure, ventricular tachycardia

Chest pain, dyspnea, pulmonary edema, atrial fibrillation, anemia requiring transfusion

Unstable angina

Cellulitis, muscle weakness, pneumonia

Pyrexia, dehydration, hypotension, chest pain, abdominal pain with laparoscopic cholecystectomy, fever, chills, pneumonia

Arthralgia, death resulting from worsening heart failure, metastatic carcinoma

Shortness of breath, weakness, chills, worsening heart failure, cellulitis

Epigastric abdominal pain, urinary retention, fecal impaction

Chest pain, shortness of breath, hypovolemia, acute MI, death

Fever, dyspnea, bronchitis, worsening heart failure, hypotension

Hematoma, anemia requiring transfusion, elbow pain (see section 7.0)

Cardiac arrest, death

Heat stroke, dizziness, syncope, hypoglycemia

Bilateral leg pain with lumbar laminectomy

Diabetes, gangrene requiring saphenous vein bypass

Pulmonary tumor, survived sudden death, bronchospasm, acute pulmonary edema, anemia, death

Epistaxis, ventricular tachycardia

There were 4 additional serious adverse events (and 2 were not reported previous). The 4 new ones include vomiting and palpitations, cardiac arrest and death, dental extraction, infection in pacemaker pouch and thyroid carcinoma.

The 2 additional reports include peripheral edema requiring hospitalization and cardiac failure followed by sudden death.

b) ongoing, double blind (ENABLE trials)

The bosentan doses used in these CHF trials are 62.5 mg with titration to 125 mg bid.

There have been a total of 336 patients reporting a serious adverse event at the time of the safety update. The events reported by at least 5 of the patients are shown below.

A total of 474 patients (29.4%) reported a serious adverse event thus far. There are 21 reports of abdominal pain (including 4 upper abdominal pain), 16 reports of serious anemia and 4 reports of serious LFT elevations in this table. Study drug assignment remains blinded.

5.0 Premature withdrawals because of adverse events

All adverse events including laboratory and ECG abnormalities as well as clinical events leading to study discontinuation were reviewed.

5.1 All bosentan patients

Events leading to discontinuation in the 10 therapeutic studies (571 bosentan patients) for at least 3 patients are shown below.

This table includes patients listed in appendix 88

A total of 81 (14.2%) of the 571 bosentan patients withdrew for a safety reason. Abnormal hepatic function was the leading cause of these discontinuations (35.8%, 29/81), followed by headache (11.1%, 9/81), anemia (7.4%, 6/81), hypotension (7.4%, 6/81). Lower limb edema led to drop out in 3 patients.

In addition to the above list, reasons for study withdrawal reported by 2 bosentan patients were abdominal pain, CVA, face edema, MI.

The number of patients and subjects who withdrew for unclear reasons include

-AC 052-101 #5 who did not return after 4.5 days receiving bosentan 125 mg bid. No adverse events were reported.

Descriptions of selected withdrawals for abdominal pain, fever, flushing, and edema are discussed below.

report 165230 Patient 16321/1305. 43 year old female withdrew for precordial sensations, not further discussed. Other events include flushing and gastritis. Narrative erroneously describes dropout because of flushing.

report 165230 Patient 16324/1424. 40 year old female reported periorbital edema reported on day 13 and then discontinued study. LFTs were increasing while hemoglobin was decreasing. She was treated with furosemide

report 165230 Patient 16325/1530. 59 year old male discontinued study drug on day 8 because of lower limb edema. A 3 pound weight gain was reported as was an increasing ALT.

report 165230 Patient 16339/1608. 70 year old male discontinued study drug because of “feeling strange,” face edema, leg edema, headache, insomnia. He had strongly positive hematuria, his hemoglobin was dropping and eosinophil count was elevated.

report 165230 Patient 16818/1802. 56 year old male reported headache, pain in limbs, dysuria, decreased appetite, and hot flushes. LFTs were increased, as were eosinophils. RBCs in urine.

Report b165230 patient 16309/0308 70 year old female was discontinued because of myocardial infarction. Also reported leg swelling. LFTs were increasing and hemoglobin was decreasing

Report b165231 patient 16945/1003 68 year old reported severe headache, nausea, fatigue about 4 hours after the first dose of study drug. He discontinued study drug.

Report b165231 patient 16945/1004. 70 year old male reported headache about 2 hours after the first dose. He was discontinued from study drug. In the follow up he reported nausea, fatigue, coughing up blood and body aches.

Report b166849 patient 18209/4082 76 year old male discontinued study drug on day 14 because of diarrhea, vomiting, fatigue, leg edema, cough, and insomnia.

Report b166849 patient 18106/6474 64 year old male discontinued study drug on day 53 because of diarrhea, nausea, vomiting that had started on day 15.

Report b166849 patient 18134/

Report 166849 patient 18223/3032. 76 year old female patient discontinued study medication because of a serious hypotensive episode at home. Her blood pressure was 65/40 mmHg at the clinic and she was treated with iv fluids.

Report 18223/3036 71 year old patient reported worsening of his gastritis with persistent itching. He had no post baseline laboratory values.

Study NC15464 patient 20045/00323. 65 year old female with CAD was hospitalized for ventricular fibrillation. “Patient was slow to respond to resuscitation efforts” and drug was discontinued. She died 4 months later from respiratory failure.

Study NC15464 patient 20251/01262. 69 year old male with CAD reported 2 episodes of light headedness, hypotension, leg muscle fatigue, and chest tightness. He withdrew from study for groin pain.

Report 166849 patient 18120/2051. 44 year old male experienced several episodes of vomiting and watery stool with syncope. He could not tolerate oral fluids and had a presyncopal episode. He recovered with iv fluids.

5.2 All placebo controlled trials

There are 7 placebo controlled studies with a total of 219 placebo patients and 533 bosentan patients. Serious adverse events reported by at least 2 bosentan patients are shown below.

The incidence of withdrawal for adverse events was higher in bosentan patients (12.4%) compared to placebo patients (10.0%). The leading reason for withdrawal, abnormal hepatic function, was reported by 4.7% of the bosentan patients compared to 0.9% of placebo patients. Other events reported by at least 4 bosentan patients and occurring more often in these patients than in the placebo group included headache (1.5% versus 0.9%), followed by anemia (0.9% versus 0), and hypotension (0.9% versus 0).

Premature withdrawals with addition of AC-052-352

The reasons for withdrawal (limited to reasons given by 2 or more bosentan patients) are shown below by treatment group.

from fax dated 6-01-01

There was a higher rate of withdrawal for the bosentan group (11.1%) compared to placebo (9.4%). The most common reason in the bosentan group (placebo subtracted) was abnormal hepatic function (3.4%). Events leading to withdrawal but reported only in bosentan group include anemia, hypotension, diarrhea, dyspnea, flushing, lower limb edema, pyrexia, vomiting, abdominal pain, face edema, and myocardial infarction.

5.2.1 REACH--Protocol NC15462B

Discontinuations because of increased LFTs or anemia reported for this large CHF trial are shown below:

Number and (percent) of patients

	Placebo N=126	Total bosentan 500 mg bid N=244
Any event	13^ (8.7)	52+ (21.3)
Increased LFTs	1 (0.8)	32 (13.1)
Anemia	0	4 (1.6)

^includes 2 who dropped out for liver enzyme increase

+includes 7 bosentan patients who dropped out for an adverse event but were recorded as refusal to continue, and 16 who dropped out for adverse event including cardiac transplantation (3) and liver enzyme increase (6).

Table 21 vol 40.

5.3 Safety update

5.3.1 PAH

a) ongoing, open label trial

There are 29 patients in the extension study AC-052-351. One patient (10503) was discontinued from study drug on day 105 because of worsening pulmonary hypertension.

b) ongoing, blinded trial (AC-052-352)

A total of 214 patients have been randomized to bosentan 62.5mg-250 mg bid or placebo. There have been 4 discontinuations, all attributed to worsening of the patient’s underlying condition.

5.3.2 CHF

a) ongoing open label

There are 27 ongoing patients with a total of 5 being discontinued prematurely from study drug. The 4 drop outs not previous discussed include 3 who refused to continue and 1 who withdrew for an unknown reason.

6.0 Liver toxicity

Bosentan provokes elevation of LFTs, primarily ALT and AST (ALT was reported up to 3000 U/L), including reports of positive rechallenge. There are few reports of patients with elevated bilirubin as well as two reports of jaundice[9]. Some of the patients with elevated LFTs are also reporting symptoms such as nausea, vomiting, abdominal pain, flu like syndrome.

6.1 All bosentan patients

The number and percent of patients in 3 categories of ALT and/or AST elevation (greater than 3 times but less than 5 times upper limit of normal[10], greater than 5 times but less than 8 times upper limit of normal, and greater than 8 times upper limit of normal) for all patients who took bosentan and had post baseline laboratory data (n=550) are shown below.

Number and (percent) of bosentan patients

	>x 3 but <x 5	> x 5 but <x 8	>x 8	total
ALT (n=550)	19 (3.5)	15 (2.7)	22 (4.0)	56 (10.2)
AST (549)	18 (3.3)	5 (0.9)	10 (1.8)	33 (6.0)
ALT and/or AST (n=550)	19 (3.5)	15 (2.7)	22 (4.0)	56 (10.2)

corrected fax dated 3-13-01

A total of 56 (10.2%) of all bosentan patients had LFT abnormalities; 22 (39%) of these patients had values greater than 8 times upper limit of normal. The most extreme ALT value reported was 2838 U/L (an increase of 73 times the baseline value of 39 U/l, patient 18164/6331).

Abnormal LFTs with addition of AC 052 352

The number and percent of patients with abnormal LFTs by category are shown below for patients with baseline and post baseline data including study AC 052 352.

Number and (percent) of patients

	>x 3 but <x 5	> x 5 but <x 8	>x 8	total
ALT (n=693)	28 (4.0)	18 (2.6)	29 (4.2)	75 (10.8)
AST (692)	27 (3.9)	6 (0.9)	15 (2.2)	48 (6.9)
ALT and/or AST (n=693)	29 (4.2)	18 (2.6)	29 (4.2)	76 (11.0)

Fax dated 6-26-01

The incidence rates for abnormal LFTs were roughly the same with the addition of the second PAH study even though the doses of bosentan were somewhat lower in this study than doses used in the entire clinical program.

The incidences of marked laboratory abnormalities (above 50 U/L for ALT, 60 U/L for AST, and 34.2 umol/l for bilirubin plus increase of 50% from baseline) in all bosentan patients are shown below.

ALT and GGT were markedly increased in 14% and 14.7 % of patients, respectively; AST and alk phos were increased in 9.7% and 2.4% of patients, respectively; bilirubin was high in 0.5%.

6.2 Placebo controlled trials

6.2.1 Mean changes

Mean changes from baseline for ALT in the placebo controlled trials are shown below by treatment group.

The group who took bosentan had a mean increase in ALT of 14 U/l from baseline (70%, 14/20) compared to placebo that had a 2 U/l increase (10.5%, 2/19). The increases for AST were less striking (37.5%, 6/16 for bosentan and 6.7%, 1/15 for placebo). This table shows mean changes in bilirubin and alk phos as being unremarkable.

6.2.2 Markedly abnormal values (see appendix)

Incidence rates for markedly abnormal clinical chemistry values are shown below, by treatment groups.

The placebo subtracted incidence rate of marked ALT, AST, GGT elevations (HH defined as a 50 % increase from the baseline value and above range) in the bosentan group were 10%, 6.1%, and 4.4%, respectively. Remarkable differences between drug and placebo were not seen for bilirubin or alkaline phosphatase.

Bosentan was similar to placebo for the other chemistry values.

Time of occurrence of liver abnormalities (ALT > 3xULN) is shown in the figure below. There were 412 bosentan patients (figure does not include 102 patients who received 2000 mg/day and 19 patients with no post baseline data) and 212 placebo patients included in the figure at time 0.

Abnormal ALT started being reported about 4 weeks after start of bosentan treatment. By week 12, roughly 15% of patients who were still on drug were reporting abnormalities and by week 28, nearly 25% of patients had abnormalities. The effect does not level off: the longer patients are treated with the bosentan, the greater is the chance that they will develop high LFTs. The crude incidence rate of ALT elevation was 11.4% when the high dose is excluded and 10.5% when the high dose is included (fax March 6, 2001).

6.2.3 Dose

Of the bosentan patients who withdrew for abnormal hepatic function or anemia, 24/25 were receiving 1000-1500 mg/d; all 5 withdrawals for anemia occurred in this dose group as well. This was the dose range that was received by the majority (58.2%, 310/533) of bosentan patients who participated in placebo controlled trials. Only 19.3% (103/533) received the higher doses (2000 mg). Elevated LFTs occurs at low as well as high bosentan doses.

The number of occurrences of high ALT values by time in study and dose is shown below.

Incidence (%) of patients with high ALT (3xULN)

	100 mg		250-500 mg		1000-1500 mg		2000 mg		ALL doses
WKS	B n=48	P n=47	B n=67	P n=58	B N=297	P n=179	B n=102	P n=69	B n=514	P n=212
2-12	2.1	0	4.3	0	9.5	1.8	6.9	0	6.2	1.3
>12	-	-	4.8	0	15.8	2.4	-	-	14.9	2.2
ALL wks	2.1	0	4.5	0	14.5	2.2	6.9	0	10.5	1.9

n=total number of patients who received the dose and had labs measured

Appendix 62

Overall, the higher the dose and the longer a patient takes any dose, the higher the risk of having an elevated ALT.

Percent of patients with ALT or AST > 3ULN with addition of AC-052-352

Percents of bosentan patients (placebo subtracted) who had elevated ALT or AST when the recently completed PAH study is included are shown below by dose and study duration.

Percent of patients (placebo subtracted)

Duration of Rx	100 mg^	250-500 mg+	1000-1500 mg^^	2000 mg++	All doses combined!
2-12 weeks	0	2.2	5.9	5.4	3.6
>12 weeks	-	12.7	13.4	-	13.1

^n=48 bosentan, n=47 placebo

+2-12 wks n=46 bosentan, n=47 placebo; >12 wks n=165 bosentan, n=79 placebo

^^2-12 wks n=63 bosentan, n=55 placebo; >12 wks n=234 bosentan, n=124 placebo

++2-12 wks n=102 bosentan, n=69 placebo;

from fax dated 6-04-01

! +2-12 wks n=259 bosentan, n=77 placebo; >12 wks n=399 bosentan, n=203 placebo

The longer the patients are on bosentan the greater the risk that they will have an abnormal ALT or AST.

Percent of patients with Abnormal LFT reported as an AE, with addition of AC-052-352

Percents of bosentan patients (placebo subtracted) who reported an abnormal LFT when the recently completed PAH study is included are shown below by dose and study duration.

Percent of patients (placebo subtracted)

Duration of Rx	100 mg^	250-500 mg+	1000-1500 mg^^	2000 mg++	All doses combined!
2-12 weeks	2.0	0	1.5	0	0.7
>12 weeks	-	6.0	6.7	-	6.4

^n=50 bosentan, n=49 placebo

+2-12 wks n=49 bosentan, n=49 placebo; >12 wks n=165 bosentan, n=80 placebo

^^2-12 wks n=66 bosentan, n=58 placebo; >12 wks n=244 bosentan, n=126 placebo

++2-12 wks n=103 bosentan, n=73 placebo;

! 2-12 wks n=268 bosentan, n=82 placebo; >12 wks n=409 bosentan, n=206 placebo

from fax dated 6-04-01

Approximately 6% of patients in a study with bosentan given for >12 weeks reported an abnormal LFT as an adverse event.

Associated symptoms

There were 29 bosentan patients (5.1%) who reported abdominal complaints including abdominal pain, abdominal cramps, abdominal distention, and/or abdominal discomfort. Of these, 14 had a ALT value that was at least 15% above baseline value and 13 had low hemoglobin value. There were 8 patients with abdominal complaints and both high ALT and low hemoglobin values (from database).

6.3 Individual cases of abnormal hepatic function

The patients listed below are some of the bosentan patients who had liver enzyme elevations and/or bilirubin elevations. Many, but not all, had at least 1 report of a adverse event, sometimes serious, and many, but not all, were discontinued for an adverse event, usually, but not necessarily, suggestive of liver abnormality and/or elevated LFTs. Upper limit of normal (ULN) was 60 U/l for ALT and 50 U/L for AST for the purposes of this review.

Jaundice, vomiting, edema. Died from end stage cardiac failure.	Patient 110 10037 (study 352) 43 year old white female with pulmonary hypertension (class IV) systemic sclerosis gastro-esophageal reflux disease, calcinosis, Raynaud's syndrome, sclerodactyly and telangiectasis and concomitant medication including digoxin, furosemide, omeprazole, metolazone, spironolactone, thyroid, hydrocodone, phenyltoloxamine, guafenesin, metoclopramide, triamcinolone, warfarin, and azithromycin. During the course of the study, the patient experienced vomiting, dehydration, edema, respiratory tract infection, alopecia, drying of mucus membranes, and jaundice. On day 127, the patient's pulmonary hypertension worsened (indicated by a decrease in walking performance) and the patient was treated with i.v. epoprostenol. The study medication was permanently discontinued on day 147 due to worsening of the condition. Highest bilirubin 64.3 umol/l; baseline was elevated (26.1 umol/l). ALT and AST remained within normal limits but alk phos was elevated (120 U/L). After bosentan was discontinued, bilirubin remained elevated. She was started on epoprostenol but continued to deteriorate, was hospitalized with multiorgan failure secondary to endstate cardiac failure, and died on day 36 of open label bosentan. (fax dated 6-21-2001)
Jaundice, vomiting and abdominal pain, anemia, increased LFTs, study drug withdrawal	Patient 20060/00841 bosentan 125 mg bid. A 81-year old male with ischemic heart disease and CHF NYHA class IV, previous MI, COPD, previous episode of epigastric pain and flatulence with raised liver enzymes and impaired renal function tests, and anemia and concomitant medications including furosemide, perindopril, digoxin, warfarin, acetylsalicylic acid, ISMN, salbutamol, ipratropium, gliclazide and ferrous sulphate was hospitalized on day 559 with one-week history of persistent vomiting, lower abdominal pain and weight loss. He was jaundiced, his abdomen was bloated and tender over the epigastrium with a palpable mass, and his urine was dark. Highest ALT 392 U/L and bilirubin 91 umol/l (normal bilirubin range 3-15 umol/l). Pleural effusion was observed. Biopsies of the gastric mucosa demonstrated chemical (reactive) gastritis. Questionable evidence of pancreatitis, inflammatory gall bladder or cholelithiasis were found. Bosentan was stopped 9 days after the start of the symptoms. Approximately 2 months later he was improved but with jaundice and an enlarged liver that was tender on palpation[11]. Laboratory values were normal except for total bilirubin (21 umol/l). Patient was reported to be doing well about 8 months later. The sponsor states that there was an episode of elevated ALT and alk phos with epigastric pain 1 year prior to study (fax dated 6-21-2001).
Elevated LFTs, anemia, lymphopenia.Hospitalized for a fib and developed ARF. Resolved. Study drug discontinued because of LFTs	Patient 18168/8001 bosentan 250-500 mg bid. 49 year old white male with coronary artery disease, NYHA Class IV, ejection fraction 26%, had a history of peripheral artery disease, coronary artery bypass surgery, chronic renal failure, diabetes mellitus and hyperlipidemia. Concomitant medications included captopril, furosemide, propranolol, isosorbide mononitrate, isophane insulin, insulin, simvastatin and acetylsalicylic acid. He was hospitalized on day 20 because of atrial fibrillation. He was cardioverted but 2 days later developed acute renal failure. He was discharged on day 30 with serum creatinine 2.1 mg/dl. The patient was found to have an elevation of GGT 172 U/l and anemia, (hb/hct11.4/36). Study medication was discontinued on day 107. GGT 72 U/l and normal hb/hct more than 100 days after drug was discontinued.
Elevated LFTs. Study drug discontinued	Patient 18140/6251 bosentan 125-250-500 mg bid. This 5l year old male patient had increased LFTs. Highest ALT: 457 U/L, AST: 255 U/L. Dose of study drug was decreased to 250 mg bid on day 115, and LFTs continued to rise. Study drug was discontinued on day 120 and within 72 days (sponsor states 35 days), LFTs were normal. Reported hypotension and cardiac failure.
Elevated LFTs with diarrhea, abdominal pain. Study drug discontinued	Patient 18134/8174 bosentan 125-250-500 mg bid. This 82 year old white female patient with coronary artery disease, NYHA class IV, ejection fraction 20%, with a history of angina pectoris, pacemaker insertion, polymyalgia rheumatica, and osteoporosis, treated with enalapril, furosemide, digoxin, carvedilol, omeprazole, cortisone, paracetamol, dextropropoxyphene, isosorbide dinitrate, and pravastatin, experienced diarrhea from days 20 to 29. The study drug was temporarily stopped on day 25. Study treatment was re-started on day 33. On day 43, AST and ALT were elevated (80 U/I and 97 U/l, respectively). The study medication was again temporarily stopped on day 50. After re-starting the study drug on day 84, the patient reported abdominal pain of approx. 1 h duration following the second and third doses. The patient stated he was afraid of the study drug and requested to be taken off the study. The study drug was permanently discontinued on day 85 and AST and ALT eventually become normal. Other reported adverse events included dizziness and postural hypotension.
Elevated LFTs, anemia. Fatigue, headache, flu-like sympomts. Study drug discontinued;death secondary to CHF.	Patient 18197/9002 bosentan 125-250-500 mg bid. 82-year-old white female with valvular heart disease and pulmonary rales, history of coronary artery bypass surgery, gout and diabetes, treated with furosemide, enalapril, amiodarone, ISMN, glyceryl trinitrate, digoxin and warfarin, allopurinol, colchicine, oxazepam, metochlopramide, paracetamol, panadeine, coloxyl with senna, was anemic at baseline (hb/hct 10.2/30). Hb/hct continued to drop (9.7/28) day 24 and ALT started to rise (77 U/L) day 29. Alk phos was elevated at baseline but other LFTs were normal. Patient was pale and extremely fatigued and she reported headache and flu-like symptoms. Study drug was discontinued on day 70; she was not rechallenged because of ongoing heart failure and because “she felt better off study medication”. Patient was transfused with 2 units of blood. Patient had 4 episodes of worsening CHF and died on day 157 after the last episode. LFTs were elevated 57 days (day 127) after bosentan had been discontinued. Other reported adverse events included headache, postural hypotension, back pain, and pruritic rash.
Elevated LFTs, abdominal pain and angina. Study drug was stopped for admini-strative reasons.	Patient 18147/6523 Bosentan 250-500 mg bid. This 61-year-old white male with coronary artery disease, NYHA class IIIB, ejection fraction 34%, history of MI, coronary artery bypass surgery, coronary angioplasty, hypertension and ongoing hypercholesterolemia and diabetes, treated with captopril, metoprolol, digoxin, furosemide, pravastatin, glibenclamide (glyburide), acetylsalicylic acid and potassium chloride. Patient was discontinued on day 43 because of elevated liver enzymes (AST 321 U/l, ALT 406 U/l, GGT 595 U/l, alk phos 131 U/l, total bilirubin 1.7 mg/dl). Baseline values were normal except for elevated bilirubin (1.6 mg/dl). LFTs started to decrease after bosentan had been stopped. Six weeks later, on day 87, patient presented with mild right upper quadrant pain, rated as of non-cardiac origin. HIDA scan, sonogram and CT scan were performed, results unknown. LFTs were again elevated 49 days (day 92) after drug had been discontinued for administrative reasons (AST 172 U/l, ALT 396 U/l, GGT 413 U/l, alk phos 134 U/l and total bilirubin 1.7 mg/dl), but again normalized (except mild ALT and GGT) after two weeks without further intervention. Patient reported recurrent chest pain diagnosed as angina on day 127. Metoprolol dose was increased. LFTs were at or below baseline levels.
Elevated LFTs, study drug withdrawn. Reported flushing, headache, pain on limbs, coughing, dysuria, and decrease of appetite	Patient 16818/1802 bosentan 1000 mg bid. This 56-year-old male patient with hypertension, treated with fish skin oils showed markedly elevated values for LFTs by day 15 (highest recorded ALT: 265 U/L, AST 144 U/L). Baseline values were normal. Drug was discontinued on day 18. Patient reported flushing, headache, pain on limbs, coughing, dysuria, and decrease of appetite beginning day 9. Follow-up evaluations after study medication discontinuation showed all LFTs except GGT returned to baseline levels by 34 days.
Elevated LFTs, flushing edema, headache, dizziness	Patient 16317/0907 bosentan 1000 mg bid. This 47-year-old female patient with hypertension and a history of dyspepsia, treated with ranitidine, showed markedly elevated LFTs by day 29 (highest ALT: 200 U/l, AST: 112 U/L). Baseline values were normal. LFTs with the exception of GGT returned to baseline by 16 days after study drug was stopped. Reported adverse events included dizziness, leg and face edema, flushing, headache. Completed study.
Elevated LFTs.	Patient 16317/0905 bosentan 1000 mg bid. This 43-year-old female patient with hypertension and a history of hypercholsterolemia showed markedly elevated LFTs by day 15 (highest ALT: 239 U/l, AST: 203 U/L). Baseline ALT was slightly above normal (37 U/L). Study drug was discontinued on day 29. Evaluations during active treatment showed LFTs returned to baseline levels by day 29 except. GGT which returned to baseline within 7 days after drug was stopped. Completed study. Reported adverse event was headache.
Elevated LFTs, study drug withdrawal for administrative reasons. Reported cystitis, headache, renal calculus, abdominal pain	Patient 16315/0523 bosentan 1000 mg qd. This 37-year-old female patient with hypertension treated with medroxyprogesterone, showed markedly elevated values for LFTs by day 8 (highest ALT: 161U/l, AST 82 U/L). Baseline ALT was slightly high (55 U/L). Follow-up evaluation showed return to baseline levels by 2 weeks. Reported adverse events included cystitis, headache, renal calculus, abdominal pain. Study drug was discontinued for administrative reasons.
Elevated LFTs Reported headache and influenza	Patient 16312/0707 bosentan 1000 mg qd. This 53-year-old female patient with hypertension showed markedly elevated values for LFTs by day34 (highest ALT: 256 U/l), AST 170 U/L. Pretreatment values were normal. Completed study. There was persistent abnormality for ALT levels and partial return to baseline levels for AST within six days of drug discontinuation. Reported adverse events included headache and influenza.
Elevated LFTs. Reported nosebleed, influenza, and rash.	Patient 16309/0311 bosentan 100 mg qd. This 70-year-old female patient with hypertension and history of asthma, treated with salbutamol, beclomethazone showed markedly elevated values for LFTs on starting on day 14 (highest ALT 566 U/l, AST 242 U/l). Pretreatment values for LFTs were above normal (ALT 163 U/l). Completed study. LFTs were returning to baseline levels within 20 days. Reported adverse events included nosebleed, influenza, and rash.
Elevated LFTs. Reported abdominal pain and feeling strange.	Patient 16306/0235 bosentan 500 mg qd. This 50-year-old male patient with hypertension and a history of lactose intolerance showed markedly elevated LFTs starting on day 29. Pretreatment ALT and GGT values were above normal (99 U/l and 274 U/L, respectively). Highest ALT (860 U/L) and AST (480 U/l) on day 29. Completed study. LFTs returned to baseline levels within 43 days. Reported adverse events included abdominal pain and feeling strange.
Elevated LFTs. Abdominal pain , edema	Patient 16306/0216 bosentan 1000 mg bid. This 52-year-old female patient with hypertension and a history of unspecified menopausal and postmenopausal disorder and treated with vitamin B complex, multivitamin combination, and estradiol, showed markedly elevated values for LFTs starting on day 16 (highest ALT: 120 U/l). Pretreatment values for ALT, AST, GGT, and bilirubin were above the normal range. Completed study. Eosinophils were markedly elevated (2.2x 10⁹/L). The patient complained of mild abdominal pain on day 33. Follow-up evaluations showed return to baseline levels within 32 days. Reported adverse events included chest pain, coughing, dyspnea, headache, edema, flushing, abdominal pain.
Elevated LFTs. Blurred vision, headache	Patient 16306/0207 bosentan 500 mg qd. This 47-year-old male patient with hypertension had markedly elevated LFTs by day 29 of treatment. (highest ALT: 600 U/l, AST 360 U/l) Baseline values for AST and ALT were normal. Completed study and follow-up evaluations showed a return to baseline levels within 28 days. Reported adverse events included blurred vision, headache, ocular disorder.
Elevated LFTs, study drug withdrawal	Patient 18106/6072 bosentan 125-250-500 mg bid. This 61-year-old white male with ischemic heart disease, NYHA class III, ejection fraction 10% and a history of peripheral and coronary artery disease, MI, coronary artery bypass surgery and diabetes, hyperlipidemia, esophageal reflux and pain in limb, treated with digoxin, metoprolol, metolazone, captopril, furosemide, omeprazole, lovastatin, insulin, acetylsalicylic acid and potassium chloride developed asymptomatic elevation of liver enzymes on day 41: ALT 155 U/l, AST 116 U/l, GGT 1039 U/l, alk phos 240 U/l. Study drug was permanently discontinued and LFTs returned to baseline values except AST (38 U/L) and GGT (168 U/L). Reported adverse events included lymphoma, constipation, anemia.
Elevated LFTs	18184/7074 This patient was reported as having an adverse event of elevated liver enzymes (reported as <3xULN). The dose of study drug was decreased by half. The enzymes had started to decrease while the patient was on drug. Study drug was stopped for administrative reasons and LFTs were normal less than a month later (fax dated 3-6-01)
Elevated LFTs, study drug withdrawal. Bronchitis, influenza	Patient 16945/1014 Bosentan 1000 mg bid. This 64-year-old female had elevated LFTs (ALT: 387 U/L, AST 159 U/L) starting day 29. Baseline values were normal.. She was withdrawn on day 76 and within 40 days values were normal except ALT (47 U/l). TSH was abnormally high on day 8. Reported adverse events included bronchitis and influenza.
Elevated LFTs, abdominal pain, surgery (cholelithiasis), positive rechallenge, study drug withdrawn. Report of tdp	Patient 18102/6162 Bosentan 125-250-500 mg bid. This 73-year-old white female with congestive (dilated) cardiomyopathy with heart failure NYHA class IIIB, ejection fraction 30%, and hypothyroidism, currently treated with lisinopril, furosemide, digoxin, and levothyroxine, was hospitalized on day 43 for severe upper quadrant abdominal pain, elevated liver enzymes AST 64 U/l, ALT 149 U/l, and alkaline phosphatase (166 U/l). Cholelithiasis was diagnosed by ultrasound and the study drug was stopped. The patient developed ventricular tachycardia (described as torsade de pointe). After successful defibrillation, she was intubated and treated with lidocaine intravenously. Prolonged QT interval was identified. The patient was extubated and a laparoscopic cholecystectomy was performed. She was discharged and resumed the intake of study drug. Drug was again stopped because of elevation of LFTs; 2 weeks later LFTs were reported as normal. Study drug was restarted and 2 weeks later LFTs increased (highest AST 161 U/I and ALT 283 U/l). Patient was permanently discontinued and LFTs returned to normal. Other reported adverse events included diarrhea, dizziness, cough
Elevated LFTs, study drug withdrawn	18119/2014 bosentan 125-250-500 mg bid. This 67-year-old male developed elevated LFTs on day 27. Highest ALT values 850 U/L, GGT 440 U/l, and AST 460 U/L and alk phos 871 U/l. Bilirubin was elevated but it was elevated at baseline as were alk phos and GGT.
Fever, chills, elevated LFTs, positive re- challenge, drug withdrawal	Patient 18122/2001 bosentan 125-250-500 mg bid. This was a 67-year-old white male with congestive (dilated) cardiomyopathy, NYHA class IIIB, ejection fraction 25% and a history of diabetes, stroke, coronary angioplasty, Parkinson's disease and healing gastric ulcer. Ongoing medications included quinapril, hydrochlorothiazide, digitoxin, furosemide, isosorbide dinitrate, molsidomine, phenprocoumon, memantin, glyburide, ranitidine, levodopa. The patient presented on day 63 with fever and chills and elevated ALT (118 U/l). He had been hospitalized previously (day 8 to day 39) because of pulmonary congestion, peripheral edema, and pneumonia and had been treated with ampicillin, ceftriaxone and imipenem. Highest ALT 820 U/l and AST 460 U/l. GGT, alk phos, total bilirubin, urea, and were elevated at baseline. Study medication was discontinued on day 27. ALT and AST normalized by day 56 and other values normalized in the following weeks. Study drug was restarted on day 63 (250 mg bosentan). Elevated ALT and fever were reported 8 hours after dosing. Study medication was permanently discontinued and LFTs became normal.
Elevated LFTs, drug withdrawal	Patient 18128/8122 (Bosentan 125-250-500 mg bid.). This 67-year-old male patient withdrew on day 65. Highest ALT: 174 U/L. Withdrew on day 65 and LFTs became normal. Reported adverse events included dizziness, cardiac failure, gout and eczema.
LFTs increased; study drug dis continued	Patient 18148/6363 bosentan 250--500 mg bid. This 48-year-old black female patient with CHF NYHA class IIIB, ejection fraction 15%, had a history of MI, diabetes, hepatitis C, chronic renal insufficiency, hypercholesterolemia, pancytopenia and secondary pulmonary hypertension. She received treatment with furosemide, losartan, digoxin, ISMN, suoralfate and alprazolam. At baseline liver function tests were above normal and were increased by study day 44. Study drug was reduced 250 mg bid and then discontinued on day 101 (AST 65 U/1, ALT 74 U/1, GGT 450 U/l. LFTs had decreased slightly 40 days later.
Dyspnea, urticaria, elevated LFTs, study drug dis continued	Patient 18149/6373 (bosentan 250-500 mg bid). This 64-year-old black female patient with congestive (dilated) cardiomyopathy with heart failure NYHA class IIIB, ejection fraction 20%, and a history of hyperlipidemia and mitral valve incompetence, treated with lisinopril, terazosin, digoxin, furosemide and metoprolol, was hospitalized because of dyspnea on day 114. Resolved in 1 day. She had received penicillin by her dentist at some unknown earlier time point Study drug was permanently discontinued on day 133 because of elevated LFTs (highest ALT 380 U/I, AST 111 U/l, GGT 97 U/l). Values returned to normal by 50 days. Reported adverse events included hearing impaired and tooth repair.
Fever, edema, elevated LFTs, study drug dis continued	Patient 18149/6374 bosentan 125-250-500 mg bid. This 58-year-old black male with congestive, dilated cardiomyopathy, NYHA class IIIB, ejection fraction 20% and a history of ongoing NIDDM, hypertension and lymphocytic leukemia, concomitant treatments of losartan, furosemide, digoxin, and glipizide, presented with fever and 'achy all over' on day 23. Ceftibuten was given on days 30 and 31. Hospitalized on Day 34 because of ongoing fever and concomitant elevation of liver enzymes: ALT 833 U/L and AST 385 U/L. Bilirubin rose from 0.5 mg/dl at baseline to 1.1 mg/dl day 61. Gall bladder sepsis was ruled out. Permanent discontinuation of study medication on day 34. LFTs normal within 110 days. Reported adverse events included bradycardia, blurred vision, dizziness, headache, diabetes, cardiac failure, ventricular extrasystoles, edema.
Extremely elevated ALT. Study drug discontinued	Patient 18164/6331 Bosentan 250-500 mg bid. This 44-year-old black male patient who was also taking glienclamide, was discontinued around day 50 because of elevated LFTs. ALT 2838 U/l, AST 846 U/L on day 40. Recovered within 50 days.
Elevated LFTs and anemia. Study drug discontinued	Patient 18168/8004 bosentan 250-500 mg bid. 76-year-old white female with congestive dilated cardiomyopathy, NYHA class IV, ejection fraction 32%, history of hypertension, knee arthroplasty and hysterectomy, treated with enalapril and furosemide, developed both anemia and elevation of liver enzymes by day 34 with peak values between day 62 and 69 (AST 153 U/l, ALT 185 U/l, GGT 550 U/l, alk phos 387 U/l). Bosentan dose was reduced to 250 mg bid on day 62 and permanently discontinued on day 65.. Last values on drug: ALT 167 U/L, AST 153 U/L, GGT 520 U/L, alk phos 361 U/l. Hb/hct: 11.5/38. LFTs were normal 55 days after drug was discontinued. Hb/hct remained decreased. Hb/hct and LFTs were normal at baseline.
Fever[12], elevated LFTs, study drug discontinued	Patient 18169/8025 bosentan 125-250-500 mg bid. This was a 73 year old male. On day 80, he reported pyrexia, his LFTs were elevated (ALT: 272 U/L, AST 118 U/l), and he was discontinued. LFTs were normal within 33 days.
Abnormal GGT and alk phos and weight gain. Study drug discontinued	Patient 18169/8027 bosentan 250-500 mg bid. This 45-year-old white female with coronary artery disease, NYHA class IV, ejection fraction 32%, with a history of NIDDM, stroke, and hypertension, treated with benazepril, furosemide, spironolactone, ISMN, atenolol, digoxin, propranolol, bezafibrate, warfarin, glibenclamide (glyburide), and metformin, permanently discontinued the study medication on day 23 because of elevated GGT: 274 U/I) alk phos (573 U/). ALT and AST were normal throughout the study. The patient gained 6 kg of weight in two weeks. On day 43, GGT: 101 U/l and normal alk phos and loss of the 6 kgs.
Elevated LFTs. Sustained ventricular tachycardia. Study drug discontinued	Patient 18170/8064 bosentan 250-500 mg bid. This 51-year-old white female with congestive (dilated) cardiomyopathy, NYHA class IIIB, ejection fraction 27%, was taking furosemide, enalapril and propranolol. Dose of study drug was reduced on day 43 because of elevated LFTs. Highest ALT: 135 U/L, AST 80 U/L Patient was hospitalized because of palpitations, dizziness and general weakness on day 72 and diagnosed with atrial fibrillation, and then 3 hours later with sustained ventricular tachycardia. Sinus rhythm was restored. Study drug was discontinued. The next day, ventricular tachycardia recurred and was successfully treated. LFTs normal within 84 days. Low blood pressure was also reported.
Elevated LFTs. UTI and brain tumor. Study drug discontinued	Patient 18171/8017 bosentan 125-250-500 mg bid. This 72-year-old white female with coronary artery disease, NYHA class IV, ejection fraction 22%, had a history of MI, arterial hypertension, anemia, NIDDM and chronic renal failure. Concomitant medications included captopril, furosemide, digoxin, acetylsalicylic acid, ISMN, ferrous sulfate, isophane insulin and glibenclamide (glyburide). The patient was hospitalized on day 8 for severe dehydration. She recovered with treatment by day 13. The patient was re-hospitalized on day 55 because of general deterioration, weakness, and dyspnea and evidence of worsening renal function. A urinary culture was positive for E. coli. She was discharged on day 65. The following day she was discontinued from study drug because of elevated hepatic enzymes: ALT 141 U/L, GGT 170 U/l. She was re-hospitalized on day 122 (56 days after study discontinuation) and CT scan of the brain revealed a meningioma. ALT and AST were within normal limits by 33 days.
Elevated LFTs. Study drug discontinued	Patient 18172/8042 bosentan 250-500 mg bid. This was a 57-year-old male with ALT 264 U/L and AST 219 U/L on day 53. Dose was reduced to 250 mg bid on day 60 and discontinued on day 64. LFTs normal within 66 days.
Lethargy, elevated LFTs, anemia. Study drug discontinued	Patient 18207/4032 bosentan 250-500 mg bid. This was a 64 year old male with ALT 203 U/L and AST 77 /l on day 65. He was discontinued that day and LFTs were within normal range by 62 days. Reported adverse events included lethargy, gout, cardiac failure, hematuria, hypokalemia
Elevated LFTs. Rash. Study drug discontinuation.	Patient 18197/9006 bosentan 250-500 mg bid. This 61 year old white male with coronary artery disease with heart failure NYHA class IIIB, ejection fraction 25%, and a history of angina, essential hypertension, atrial fibrillation, hypercholesterolemia and osteoarthritis, treated with atenolol, digoxin, furosemide, fosinopril, warfarin, piroxicam, roxithromycin, and simvastatin, was withdrawn from the study on day 29 for mild dermatitis and severe liver function abnormalities (ALT 713 U/l, AST 399 U/I, GGT 93 U/I). All values had returned to baseline within 56 days. Reported adverse events included sinusitis, dermatitis, rash.
Elevated LFTs. Study drug discontinuation.	Patient 18219/3019 bosentan 125-250-500 mg bid. This was a 57 year old male with ALT 350 U/L and AST 115 U/L on day 106. He was discontinued and LFTs were within normal limits by 42 days.
Fever, hepatomegaly, elevated LFTs, anemia. Hospitalized. Study drug discontinued	Patient 18219/3020 bosentan 125-250-500 mg bid. This was a 66-year-old white female with non-ischemic dilated cardiomyopathy, NYHA class IV, ejection fraction 23% and active herpes zoster, treated with torsemide, enalapril, captopril and phenprocoumon. She was hospitalized on day 6 because of asthenia and mild diarrhea and treated with cyanocobalamin and paracetamol. Patient was discharged without sequelae on Day 14. The patient developed fever on day 19 and was hospitalized on day 21. LFTs were elevated (ALT 118 U/l, AST 84 U/l, GGT 226 U/l, alk phos 719 U/l, total bilirubin 28 umol/l). LFTs at baseline were normal. Hepatomegaly was diagnosed on ECHO. Hepatitis, cancer, and immune disease were ruled out. Study drug was discontinued. Fever resolved and LFTs were within normal limits by about 42 days. Reported adverse events included asthenia, viral infection, cardiac failure.
Elevated LFTs. Study drug discontinued	Patient 18221/3025 bosentan 125-250-500 mg bid. This 58 year old female was down-titrated to 250 mg bid on day 49 because of elevated LFTs. The 500 mg bid was restarted, LFTs continued to rise (ALT: 130 U/L and AST: 62). She was discontinued from study drug and LFTs eventually normalized. Reported dizziness.
Elevated LFTs. Syncope. Study drug discontinued	Patient 18229/3081 (bosentan 125-250-500 mg bid). A 53-year-old with male with non-ischemic dilated cardiomyopathy, NYHA class IV., ejection fraction 20%, a history of TIA with syncope, active diabetes, hyperuricemia and hypertension, treated with digoxin, furosemide, enalapril, carvedilol, mexileline, potassium, L-camitine, ticlopidine, chlorpropamide, phenformin, allopurinol and ranitidine, experienced an episode of syncope on day 30 which resolved within 10 minutes. He was hospitalized and brain CT showed no change from when he had had a previous syncopal event (prior to start of study). He remained on same dose of study medication until day 70 when he was down titrated to 250 mg because of low blood pressure. Metformin and glibenclamide (glyburide) were started on day 77. Highest ALT 280 U/l and AST 168 U/l, bilirubin 24 umol/l. GGT had been elevated at baseline. He was discontinued on day 135. ALT was still elevated (163 U/l) and there was no follow up. Reported adverse events included UTI, paraesthesia, chest pain, cough, cardiac failure.
Elevated LFTs, anemia, and worsening heart failure. Study drug discontinued	Patient 18233/8053 bosentan 250-500 mg bid. 67 year-old white male with coronary artery disease, NYHA class IV, ejection fraction 16%, NIDDM, peripheral artery disease, MI, TIA, coronary angioplasty, and renal failure, treated with acetylsalicylic acid, benazepril hydrochloride, ISMN, spironolactone, furosemide, and digoxin, was permanently discontinued from study drug on day 54 because of worsening heart failure, elevated liver enzymes and anemia. Day 43: ALT 82 U/l, GGT 230 U/I, alk phos 329 U/l (AST remained within the normal range throughout the study), and hb/hct 11.6/36 (baseline 14/41), eosinophils were elevated. LFTs normalized. Hb/hct were not followed.
Elevated LFTs, study drug discontinued	Patient 20068/00862 bosentan 125 mg bid. This 65-year-old white male with ischemic heart disease, previous MI, and gout discontinued study drug because of elevated liver enzymes. ALT and AST were 600 and 972 U/l. Bosentan was stopped and values dropped to normal. In a previous study, patient had developed atrial fibrillation, was cardioverted, and had 3 episodes of ventricular tachycardia. Bosentan was temporarily discontinued and a defibrillator was inserted. Concomitant medications included digoxin, ISMN, captopril, furosemide, bendrofluazide, allopurinol, acetylsalicylic acid, warfarin and codydramol tablets. Reported adverse events included epistaxis, blurred vision, dyspnea, anorexia.
Elevated LFTs, positive rechallenge, nausea, study drug discontinued	Patient 20082/00063 bosentan 125 mg bid. This 73-year-old white female with non-ischemic dilated cardiomyopathy and previous MI experienced an increase in LFTs that resulted in study discontinuation. She had received bosentan 500 mg bid in a previous study without apparent abnormalities. Concomitant medications included lisinopril, digoxin, carvedilol, bumetanide, ISMN, simvastatin, naproxen, colchicine and metformin. Glibenclamide (glyburide) was started on day 149. ALT and AST were 763 and 381 U/l, respectively, by day 284. Alk phos and GGT were 218 and 432 U/l, respectively. Bosentan was stopped between days 285 and 369 and 4 weeks later, liver enzymes returned to baseline levels. Bosentan was restarted. Nausea and with elevated LFTs were reported on day 383 and bosentan was discontinued on day 389. Simvastatin was also discontinued. Reported adverse events included gout, hypotension, diabetes, diarrhea, dyspnea.
LFTs increased. Reported various adverse events. Positive rechallenge, discontinued study drug	Patient 20257/1502 bosentan 125 mg bid. This 61-year-old white male with heart failure, ischemic heart disease, angina, previous MI, CABG, and angioplasty discontinued study drug because of increased LFTs. ALT: 489 U/l and AST 263 U/l. Bosentan was stopped and enzymes normalized. Other reported adverse events included fatigue, bronchitis, irritability and vertigo. Patient had participated in a previous study during which he received bosentan 500 mg bid. He reported headaches, nightmares, palpitations, worsening heart failure, allergic drug rash, and increased LFTs during this first study. LFTs normalized when drug was stopped. Concomitant medications included furosemide, ISMN, acebutolol and acetylsalicylic acid.
Elevated LFTs, positive rechallenge, study drug discontinued	Patient 20273/1122 bosentan 125 mg bid. This 70-year-old white male patient with ischemic heart disease, NIDD discontinued study drug increased in LFTs. Concomitant medications included glibenclamide (glyburide) and metformin, lisinopril, furosemide, metoprolol, prinivil and simvastatin. ALT and AST increased to 238 and 119 U/l, respectively. LFTs normalized after bosentan was discontinued. Patient had received bosentan 500 mg bid in a previous study with ALT and AST peaking at 262 U/L and 98 U/l. Study drug was stopped and LFTs returned to baseline.
Elevated LFTs with positive rechallenge.	Patient 18127/8082 bosentan 250-500 bid. 71 year old female had ALT peak of 1420 U/L on day 47. AST was 840, alk phos 317. She was termporarily discontinued from drug with decrease in LFTs. A rechallenge caused re-elevation of LFTs (ALT to 161). Permanently discontinued.
	Patients added from safety update
Malaise, fever, chills, elevated LFTs, elevated bilirubin. Study drug discontinued	AC-052-352 Patient 106/10008. This 53-year-old black female with pulmonary hypertension secondary to systemic sclerosis was receiving prednisone, hydrochloroquine, lorazepam, zolpidem, and panadeine. Her medical history included systemic lupus erythematosus with previous hepatic involvement, systemic hypertension, and obesity. She presented to the emergency room with fever, chills, and malaise on day 143. The patient had reported a recent history of URI treated with multiple antibiotics. ALT 433 U/l, AST 298 U/l, AP 164 U/l, total bilirubin 17.9 umol/l). Study drug was stopped. The consulted local rheumatologist reported that the event occurring on study drug is consistent with patient’s previous pattern of flares in liver function abnormalities due to auto-immune hepatitis (lupus), cessation of hydrochloroquine, and a 2-day cessation of prednisone. The study medication was discontinued.
Elevated LFTs and bilirubin; nausea, fatigue, anorexia. Study drug discontinued. Died of worsening PAH.	Patient AC-052-352/105 10014. Bosentan 250 mg bid. 48-year-old white female with pulmonary arterial hypertension due to systemic lupus erythematosus. Reported adverse events included loss of appetite, nausea and fatigue with increases in LFTs (AST 659 U/l, ALT 554 U/l, alk phos 275 U/l, bilirubin 36 umol/l). Study drug was discontinued on day 116 followed by a return to baseline levels about 24 days later. Patient died of worsening of underlying condition 32 days after drug was stopped.
Elevated LFTs and bilirubin. Study drug discontinued	Patient AC-052-352/10050. 59-year-old white male with increase in LFTs (ALT 900 U/L, AST 596 U/L and bilirubin 1.9 mg/dl. Drug was discontinued and values were normal by about 4 weeks. Remained asymptomatic.

There were a few reports of very high elevations of ALT and most patients were discontinued when their LFTs started to increase. There were 2 reports of jaundice[13]. Adverse events of fever, nausea, vomiting, flu or flu-like symptoms were reported in some patients while other patients remained asymptomatic. The cases of rechallenge usually resulted in elevation of LFTs (noted as early as 8 hours after the rechallenge dose). At this time, there is no evidence of a liver related fatality or need for liver transplantation in this small sample of patients. It seems so far that stopping bosentan results in patients returning to baseline status.

There is a concern that patients who develop fever and abdominal pain could be having a drug reaction but be taken (mistakenly) to surgery for an acute abdominal event.

6.4 Ongoing ENABLE trials

This blinded, placebo controlled trial in patients with CHF is using bosentan doses 62.5mg bid titrated to 125 mg bid.

As of the clinical cut-off date, increases of ALT and/or AST to > 3 x ULN have been reported in 40 patients, with 11 reporting increases of 8 times upper limit of normal or higher. Of the 40 patients, 24 continued taking the study drug, at least for a while longer. The patients of note are discussed below. Although the blind remains intact, these are more likely to be bosentan than placebo patients:

-patient 50511: LFTs were starting to rise by week 8. By week 26, increases in ALT (262 U/L, AST 546 U/l), total bilirubin (6.75 mg/dl), and alkaline phosphatase (374 U/L) associated with nausea, jaundice, abdominal pain and weakness were reported. Treatment was stopped and within 10 days, LFTs had decreased markedly (ALT was not reported). Gastroscopy revealed severe esophagitis and duodenitis.

-patient 71114: After 4 weeks of treatment, ALT increased to 1485 U/l and total bilirubin to 40 mg/dl with reports of abdominal pain. Treatment was stopped 8 days later and LFTs fell to normal values by around 2 weeks. Patient is being re-challenged with the low dose. Results of this re-challenge are pending.

-patient 92806: LFTs at week 4 were elevated: ALT 442 U/L, AST 213 U/L, bilirubin (1.8 mg/dl, and alkaline phosphatase 622 with reports of nausea, vomiting and dizziness. Study drug was stopped and symptoms disappeared. All laboratory parameters returned to baseline values within 4 weeks.

There are 8 patients who stopped receiving the target dose and were re-challenged with the low dose after their transaminases returned to baseline values. Rechallenge with the low dose has resulted in 3 patients (40405, 71114, 96001) with no follow up information, 1 patient (93403) with no increase in transaminases, 2 patients (30302, 41103) with transient increases in ALT and/or AST, and 2 patients (50317, 95201) with recurrent increases in transaminases and permanently discontinued.

One patient (50427) had an increase in ALT and AST reported on day 59. Bilirubin was elevated (21.9 mg/dl) on day 243. Evidence for acute hepatitis A was reported. Study drug was bilirubin gradually decreased and was 5.5 mg/dl 2 weeks later (day 299). After 3 additional weeks, treatment with study medication (low dose) was restarted.

6.5 Safety Update

6.5.1 PAH

a) ongoing, blinded trial (AC-052-352)

A total of 214 patients have been enrolled and 7 of these had ALT/AST >3 times ULN. Of the 7, 3 had ALT/AST >8 times. Total bilirubin increased in 5 patients. Three of the 7 are discussed in the abnormal LFT list. The remaining 4 were not symptomatic and did not discontinue study drug.

The reporting of elevated LFTs, sometimes higher than 8xULN with and without elevations in bilirubin and with or without symptoms continues unabated.

b) ENABLE: 1613 patients have been randomized and treated with bosentan 125 mg or placebo for a minimum of 16 weeks. At the time of the Safety update, there have been 69 patients (4.3%) who have had increases in ALT and/or AST at least 3 times upper limit of normal. Three patients had study drug discontinued.

No. of patients

	Category of AST and/or ALT elevations
	>3 but <5 times ULN	> 5 but < times ULN	> 8 times ULN	> 3 times ULN
Any elevation	31	15	23	69

A total of 23 patients (1.4%) have had elevations of LFTs of greater than 8 times ULN. Although the study is still blinded, it is certain that most of the 69 patients are in the bosentan group taking doses < 125 mg bid.

7.0 Hematology

7.1 Anemia

7.1.1All bosentan patients

7.1.1.1 Adverse event

Of the 571 patients who received bosentan, 27 (4.7%) reported anemia as an adverse event.

7.1.1.2 Markedly abnormal laboratory value (see appendix)

The incidence rates of patients with markedly abnormal hematology values defined as a significant shift from baseline for and outside a reference range are shown below for all patients who received bosentan or placebo and had post baseline values.

Forty two (7.9%) bosentan patients had a markedly low hemoglobin[14] and 53 had markedly low hematocrit[15].

Other parameters including platelets and white blood cells were abnormal for less than 2% of bosentan patients. A markedly high eosinophil count[16] was reported in 0.6% of bosentan patients.

7.1.2 Placebo controlled trials

7.1.2.1 Adverse event

Of the patients who received bosentan in a placebo controlled trial, 3.7% (20/533) reported anemia (1 reported hemoglobin decreased) as an adverse effect compared to 0.9% (2/219) patients who received placebo.

7.1.2.2 Mean changes

Mean changes from baseline for hemoglobin in the placebo controlled trials are shown below by treatment group.

The bosentan group had a mean decrease from baseline in hemoglobin of 0.9 g/dl (6.1%) compared to 0.2 g/dl (1.4%) in the placebo group. Similar decreases were shown for the hematocrit (0.03, 6.8% decrease for bosentan and 0.01, 2.3% decrease for placebo).

Mean changes from baseline with addition of AC-052-352

Placebo subtracted mean changes for bosentan from baseline at last value, by dose and duration, is shown below.

Placebo subtracted change from baseline: hemoglobin (g/dl)

Duration of Rx

100 mg^

250-500 mg+

1000-1500 mg^^

2000 mg++

All doses combined!

2-12 weeks

-0.31

n=48

-0.29

n=45

-0.54

n=61

-0.93

n=95

-0.47

n=249

>12 weeks

-0.96

n=161

-0.93

n=208

-0.94

n=369

^n=48 bosentan, n=46 placebo

+2-12 wks n=45 bosentan, n=46 placebo; >12 wks n=161 bosentan, n=79 placebo

^^2-12 wks n=61 bosentan, n=53 placebo; >12 wks n=208 bosentan, n=116 placebo

++2-12 wks n=95 bosentan, n=67 placebo;

! 2-12 wks n=249 bosentan, n=74 placebo; >12 wks n=369 bosentan, n=195 placebo

From fax dated 6-01-01

There is a dose related mean decrease in hemoglobin that becomes more pronounced the longer the patient is on bosentan. For all doses combined and duration of treatment >12 weeks, the mean decrease from baseline is nearly 1 g/dl.

7.1.2.3 Markedly abnormal laboratory value (see appendix)

The incidence rates of patients reporting markedly abnormal hematology values (defined as a 15% change from baseline for hemoglobin and hematocrit) are shown below for patients in placebo controlled trials that had post baseline values.

The rates of markedly low hemoglobin and/or hematocrit are higher for bosentan group than placebo. Reports of abnormalities for the other hematology values, however, were similar for the 2 groups.

7.1.2.4 Dose and duration

Marked decreases in hemoglobin and hematocrit by dose and duration

The reports of decreased hemoglobin by time in study and dose is shown below.

Percent of patients with markedly low^ hemoglobin

	100 mg		250-500 mg		1000-1500 mg		2000 mg		ALL doses
WKS	B n=48	P n=46	B n=66	P n=57	B N=283	P n=172	B n=100	P n=69	B n=514	P n=212
2-12	0	0	0	0	11.1	1.9	4.0	1.5	4.3	2.7
>12	-	-	0	0	9.1	3.4	-	-	8.3	3.1
ALL wks	0	0	0	0	9.5	2.9	4.0	1.5	6.2	2.9

^15% decrease from baseline

n=total number of patients who received the dose and had labs measured

Appendix 61

In this limited database, doses of 1000 to 1500 mg were associated with a placebo subtracted percent incidence of 9.2% for markedly low hemoglobin within the first 12 weeks of treatment.

For all doses combined, the longer patients took bosentan, the higher the (placebo subtracted) incidence rate: 1.6% for patients taking bosentan 12 weeks or less and 5.2% for more than 12 weeks. This implies that the chances a patient will develop anemia increase the longer the patient is taking bosentan. This is supported by the figure shown below.

Percent of patients with low hemoglobin with addition of AC-052-352

Percents of bosentan patients (placebo subtracted) who had hemoglobin drop of 15% or more from baseline, by dose and duration, are shown below.

Percent of patients (placebo subtracted)

Duration of Rx	100 mg^	250-500 mg+	1000-1500 mg^^	2000 mg++	All doses combined!
2-12 weeks	2.0	2.3	18.4	18.0	4.4
>12 weeks	-	13.8	13.3	-	12.4

^n=48 bosentan, n=46 placebo

+2-12 wks n=45 bosentan, n=46 placebo; >12 wks n=163 bosentan, n=79 placebo

^^2-12 wks n=61 bosentan, n=53 placebo; >12 wks n=210 bosentan, n=117 placebo

++2-12 wks n=95 bosentan, n=67 placebo;

! +2-12 wks n=249 bosentan, n=74 placebo; >12 wks n=373 bosentan, n=196 placebo

from fax dated 6-01-01

The percent of patients (placebo subtracted) with a decrease in hemoglobin is shown below.

Percent of patients (placebo subtracted)

	Total dose of bosentan
Decrease of hb	100 mg n=48	250-500 mg n=206	1000-1500 mg n=269	2000 mg n=95	All doses N=618
> 1.0 g/dl	7.6	31.2	31.7	42.9	27.8
To lower limit of normal	2.0	12.4	17.2	18.0	7.5
> 15% from baseline and < 11 g/l	0	1.6	6.6	2.5	3.1
> 15% from baseline and < 10 g/l	0	1.1	0.6	0	0

Numbers of patients are roughly the same for all categories

Fax dated 6-26-01

Overall, there is a dose related increase in the percent of patients with abnormally low hemoglobin. The longer the patients are on bosentan the greater the risk that they will have an abnormality.

The Kaplan-Meier estimate of the time to the first occurrence of marked decrease in hemoglobin concentration (decrease of 15% from baseline value) is shown in the figure below.

As with the liver function test abnormalities, the longer patients take bosentan, the higher is the likelihood that they will have a marked decrease in hemoglobin.

7.1.3 REACH—NC15462

This 12 week placebo controlled trial in 370 CHF patients reported that hemoglobin, hematocrit, and RBCs were decreased by an average of 7-10% from baseline at week 3 in the bosentan group but unchanged in the placebo group. These changes in the bosentan group were present and of similar magnitude at weeks 12 and 26. There was a mean decrease in total leukocytes of 9%, 15% and 11% at weeks 3, 12, and 26, respectively. There were mean decreases in neutrophils and platelets as well.

The marked laboratory abnormalities in this study are shown below.

Percent of patients

		Bosentan (up to 500 mg bid)
Hematology variable	Placebo n=119	slow titration n=109	fast titration n=111
Low hematocrit	4.2	11.9	16.2
Low hemoglobin	3.4	8.2	9.8
Low WBC	1.7	0	0.9
Low lymphocytes	10.9	17.4	12.5

Table 22 vol 40

7.1.4 Anemia reported as serious adverse event

Of the 571 patients who received bosentan, 6 (1.1%) reported anemia (1 reported decreased hemoglobin) as a serious adverse event. In the placebo controlled trials, 2 (0.4%) bosentan patient and 2 (0.9%) placebo patient reported anemia (or decreased hemoglobin or secondary anemia) as a serious adverse event.

7.1.4.1 Reports of anemia

The bosentan patients listed below are those who reported anemia as a serious adverse event, needed a transfusion, and/or were discontinued from study drug for reasons that included anemia. N.B. hb=hemoglobin (units are g/dl); hct=hematocrit (units are %).

Hematoma, anemia requiring transfusion	Patient 20256/1481 bosentan 125 mg bid. This 90-year-old male patient with ischemic heart disease was hospitalized with hematoma and anemia on day 24. Medical history: atrial fibrillation , MI, and gout. Concomitant medications: warfarin, digoxin, furosemide and benazepril. Baseline hb/hct 12/33. Hct at hospitalization was 20 and he was transfused with two units of RBC. Hb/Hct 11.4/36 on day 31, 11.2/36 on day 88 and 12/35 last recorded value. Completed study.
Chest pain, atrial fibrillation, pulmonary edema, anemia requiring transfusion	Patient 20241/1061 bosentan 125 mg bid. This 78-year-old male patient with ischemic heart disease was hospitalized on day 649 with severe retrosternal chest pain with marked dyspnea and atrial fibrillation with rapid ventricular rate. Medical history: MI, CABG, CVA, Raynaud’s disorders and hypothyroidism. Concomitant medications: losartan, furosemide, spironolactone, digoxin, glyceryl trinitrate, warfarin, quinine, cholestyramine and levothyroxine. In hospital, patient was treated with amiodarone, metoprolol, fentanyl and streptokinase was started but stopped immediately. Pulmonary edema was diagnosed. Hb was 8.6 (10.3 at baseline). Stool was positive for occult blood. Patient received 1 unit of blood. Hb was above 11 for the rest of the study. Other adverse events included edema, dizziness, and severe diarrhea for which drug was interrupted.
Anemia, heart failure, ventricular tachycardia	Patient 20097/1821 bosentan 125 mg bid. 70-year-old female with ischemic heart disease, diabetes mellitus, and idiopathic tbrombocytopenia and concomitant medications: cilazapril, bumetanide, carvedilol, spironolactone, digoxin, actrapid and insulin retards. Baseline hb/hct 13.5/43. She was diagnosed on day 57 with anemia (hb/hct 9.9/35); treated with iron supplementation. She was hospitalized on day 92 for worsening heart failure requiring diuretics. Last hb/hct on drug 10.7/not recorded. Study drug was stopped on day 133 because of ventricular tachycardia. Other reported events included CVA.
Chest pain, abdominal pain, anemia, transfusion, dyspnea, death	Patient 20081/0122 bosentan 125 mg. 76-year-old patient with ischemic heart disease was hospitalized because of chest pain on day 14. Medical history: MI, angina pectoris, diabetes, gastric vascular ectasia, gastric ulcer, gastric polyp and mild depression. Concomitant medication: captopril, furosemide, digoxin, metoprolol, glyceryl trinitrate, ticlopidine, glipizide, metformin, omeprazole, teniazepam, fluoxetine, paracetamol and iron supplement. Anti-anginal treatment based on glyceryl trinitrate was initiated and the pain resolved within 2 hours. He was rehospitalized on day 32 because of chest pain with T-wave inversion He was treated and discharged. He was re-hospitalized on day 65 because of chest and severe abdominal pain with shortness of breath. Hb was 9.2 (baseline hb/hct 11.7/34). Endoscopy was negative. He was discharged and rehospitalized on day 90 because of abdominal pain and dyspnea (hb/hct 10.3/30 day 89). The patient received 4 units of packed RBC and was discharged. On day 148 the patient was re-hospitalized with angina. On day 184 the patient was re-hospitalized suffering from musculoskeletal chest pain. He experienced sudden death on day 376 thought to be secondary to MI.
Active bleeding, anemia, transfusion, followed by septicemia. Death	Patient 20086/0041 bosentan 125 mg bid. This was a 65 year old male patient with a medical history ischemic heart disease, congestive cardiac failure, bacterial endocarditis, pacemaker insertion, NIDDM, gout and prostatic complaints. Concomitant medications: furosemide, spironolactone, digoxin, prazosin, isosorbide mononitrate, warfarin, glipazide, colchicine, allopurinol, omeprazole, temazepam, terbutaline, scopolamine, paracetamol and iron supplement. There were no baseline lab values. On day 27, he was hospitalized for bladder neck incision and reported hematuria and acute urinary retention with hematuria. On day 380, he was hospitalized because of malaise and increasing dyspnea. Clinical examination revealed pneumonia, anemia, renal failure, gout and worsening of cardiac failure. Colonoscopy showed blood in the caecum. Anemia (Hb 7.8 mg/l), melena and hematuria were noted and he was treated with 5 units of packed red blood cells, 5 units of plasma. He was treated with furosemide, antibiotics. The next day, fever was noted during blood transfusion and blood culture detected positive microbial growth. Bacterial endocarditis was suspected. The patient's condition gradually deteriorated and he died about 2 weeks later (day 398). While receiving placebo in the prior study (REACH), he was noted to have mild elevation of liver enzymes, including bilirubin, and raised creatinine and BUN.
Elevated LFTs, anemia, and worsening heart failure. Study drug discontinued	Patient 18233/8053 bosentan 250-500 mg bid. 67 year-old white male with coronary artery disease, NYHA class IV, ejection fraction 16%, NIDDM, peripheral artery disease, MI, TIA, coronary angioplasty, and renal failure, treated with acetylsalicylic acid, benazepril hydrochloride, ISMN, spironolactone, furosemide, and digoxin, was permanently discontinued from study drug on day 54 because of worsening heart failure, elevated liver enzymes and anemia. Day 43: ALT 82 U/l, GGT 230 U/I, alk phos 329 U/l (AST remained within the normal range throughout the study), and hb/hct 11.6/36 (baseline 14/41), eosinophils were elevated (.8810e9/l).LFTs normalized. Hb/hct were not followed.
Elevated LFTs, anemia. Fatigue, headache, flu-like sympomts. Study drug discontinued;death secondary to CHF.	Patient 18197/9002 bosentan 125-250-500 mg bid. 82-year-old white female with valvular heart disease and pulmonary rales, history of coronary artery bypass surgery, gout and diabetes, treated with furosemide, enalapril, amiodarone, ISMN, glyceryl trinitrate, digoxin and warfarin, allopurinol, colchicine, oxazepam, metochlopramide, paracetamol, panadeine, coloxyl with senna, was anemic at baseline (hb/hct 10.2/30). Hb/hct continued to drop (9.7/28) day 24 and ALT started to rise (77 U/L) day 29. Alk phos was elevated at baseline but other LFTs were normal. Patient was pale and extremely fatigued and she reported headache and flu-like symptoms. Study drug was discontinued on day 70; she was not rechallenged because of ongoing heart failure and because “she felt better off study medication”. Patient was transfused with 2 units of blood. Patient had 4 episodes of worsening CHF and died on day 157 after the last episode. LFTs were elevated 57 days (day 127) after bosentan had been discontinued. Other reported adverse events included headache, postural hypotension, back pain, and pruritic rash.
Elevated LFTs, anemia, and worsening heart failure. Study drug discontinued	Patient 18233/8053 bosentan 250-500 mg bid. 67 year-old white male with coronary artery disease, NYHA class IV, ejection fraction 16%, with a history of NIDDM, peripheral artery disease, MI, TIA, coronary angioplasty, and renal failure, treated with acetylsalicylic acid, benazepril hydrochloride, ISMN, spironolactone, furosemide, and digoxin, was permanently discontinued from study drug on day 54 because of worsening heart failure, elevated liver enzymes and anemia. ALT 82 U/l, GGT 230 U/I, alk phos 329 U/l , hb/hct 11.6/36, eosinophils were elevated. Baseline hb.hct13/41. LFTs normalized. Hb/hct were not followed up.
Elevated LFTs, anemia, lymphopenia. Hospitalized for a fib and developed ARF. Resolved. Study drug discontinued because of LFTs	Patient 18168/8001 bosentan 250-500 mg bid. 49 year old white male with coronary artery disease, NYHA Class IV, ejection fraction 26%, had a history of peripheral artery disease, coronary artery bypass surgery, chronic renal failure, diabetes mellitus and hyperlipidemia. Concomitant medications included captopril, furosemide, propranolol, isosorbide mononitrate, isophane insulin, insulin, simvastatin and acetylsalicylic acid. He was hospitalized on day 20 because of atrial fibrillation. He was cardioverted but 2 days later developed acute renal failure. He was discharged on day 30 with serum creatinine 2.1 md/dl. The patient was found to have an elevation of GGT 172 U/l and anemia, (hb/hct11.4/36). Study medication was discontinued on day 107. GGT 72 U/l and normal hb/hct more than 100 days after drug was discontinued.
Elevated LFTs and anemia. Study drug discontinued	Patient 18168/8004 bosentan 250-500 mg bid. 76-year-old white female with congestive dilated cardiomyopathy, NYHA class IV, ejection fraction 32%, history of hypertension, knee arthroplasty and hysterectomy, treated with enalapril and furosemide, developed both anemia and elevation of liver enzymes by day 34 with peak values between day 62 and 69 (AST 153 U/l, ALT 185 U/l, GGT 550 U/l, alk phos 387 U/l). Bosentan dose was reduced to 250 mg bid on day 62 and permanently discontinued on day 65.. Last values on drug: ALT 167 U/L, AST 153 U/L, GGT 520 U/L, alk phos 361 U/l. Hb/hct: 11.5/38. LFTs were normal 55 days after drug was discontinued. Hb/hct remained decreased. Hb/hct and LFTs were normal at baseline.
Anemia, blood transfusion, study drug withdrawal	Patient 18197/9007 bosentan 125-250-500 mg bid. 65-year-old white male with coronary artery disease, heart failure NYHA class IIIB, ejection fraction 28% and a history of COPD, peripheral artery disease, MI, stroke/TIA, peripheral neuropathy, peptic ulcer, coronary artery bypass surgery and NIDDM, treated with furosemide, enalapril, glibenclamide (glyburide), metformin, ranitidine, glyceryl trinitrate and acetylsalicylic acid, was hospitalized on day 22 with worsening heart failure. Hb/hct 10.6/30 at baseline (secondary to vitamin B12 deficiency). The patient was treated with diuretics, blood transfusion, ferrous sulfate, and hydroxycobalamine. He was discharged 2 days later. The patient reported weakness, unsteadiness on feet, constipation, and diarrhea and withdrew from the study on day 55. Last hb/hct on drug was 10.7/32. Patient remained anemia through follow up. Other events included nausea and flushing.
Anemia, worsening CHF, transfusion, study drug withdrawn	Patient 18113/6061 Bosentan 125-250-500 mg bid. 64-year-old female with CHF NYHA class IIIB and a history of iron-deficiency anemia (> 6 months), hypothyroidism, diabetes, previous MI and peripheral artery disease. Concomitant treatments included ferrous sulfate, beta-carotene, glibenclamide, metformin, captopril, furosemide, hydrochlorothiazide, digoxin. Hb/hct 10.5/33 baseline. After six weeks of treatment with study drug, patient reported worsening of CHF symptoms leading to hospitalization. Study medication was discontinued. Hb/hct 8.9/27. Patient received 2 units of RBC. Gastroscopy was negative for bleeding. Patient remained anemic through 127 days of follow up. Other events included edema, dizziness, diarrhea, epistaxis.
Anemia, transfusion	Patient 18199/9071 bosentan 500 mg bid. 63 year old male with CHF NYHA class IIIB and a history of diabetes, MI, diverticulum, hyperlipidemia and obstructive sleep apnea was randomized with baseline hb/hct (14.2/41). He was hospitalized around week 11 for worsening of CHF and his hb was 10.8. Patient received blood transfusion. Iron studies, B12/folate showed no abnormality. History of melena was reported. The patient completed the study (173 days) with hb/hct 14.1/41. Other events reported included flushing and nausea.
Anemia, low monocyte count. Study drug discontinued	Patient 18168/8006 Bosentan 250-500 mg bid. 75 year old male with CHF NYHA class IIIB and ejection fraction of 22% at baseline. Medical history included diabetes, hypertension, and hyperlipidemia. Concomitant medication included captopril, furosemide, atenolol, simvastatin, aspirin and nitrates. Treatment with study drug was discontinued on day 70 because of a decrease in hb/hct:10.8 g/dL/33 (hb/hct at baseline: 13.1 g/dL/39).There was also a rise in alk phos 277 U/L), ALT 59 U/L and AST (76 U/L) while patient was taking study drug. Laboratory values were back to baseline 84 days after drug was stopped.
Anemia, transfusion, study drug discontinued	Patient 18169/8021 bosentan 500 mg bid. 59-year-old male with CHF NYHA class IV and a history of diabetes, angina pectoris, MI and coronary bypass surgery. He was anemic at baseline hb/hct: 11.3/53. GTT at baseline was elevated (248 U/L). He was hospitalized on day 8 because of worsening heart failure and was discharged 4 days later. Hb/hct continued dropping (9.6/29). The patient was hospitalized again on day 30 for worsening of CHF. He was transfused with 2 units of RBC for symptomatic anemia, his condition improved and he was discharged one week later. Day 135 GGT 490 U/L with elevated ALT (86 U/L). Hb/hct 10.1/31 at study discontinuation. He was anemic with elevated, but falling, ALT and AST (51/52 U/L) at last follow up (39 days after drug was discontinued).
Anemia, transfusion, normal bone marrow biopsy	Patient 18202/9032 bosentan 500 mg bid. 82-year old male with CHF NYHA class IIIB and a history of angina pectoris, MI and gout was randomized with decreased hb/hct 12.5/38 and elevated ALT and AST (64/47 U/L). Anemia continued to worsen (hb/hct 9.6/27 day 81) but LFTs normalized. Results of bone marrow biopsy showed a normocellular marrow with adequate hematopoietic reserves and no significant diagnostic features. Iron stores were normal and no pathological ring sideroblasts were detected. Three weeks later the patient received 2 units of packed cells. Patient completed the study. Other reported events include positional vertigo, heart failure, dehydration.
Anemia, Study drug discontinued	Patient 20083/0022 bosentan 125 mg bid. 64-year-old white male with valvular heart disease. He had received placebo in the previous study during which he experienced five episodes of worsening of heart failure. Medical history: coronary artery bypass surgery, coronary angioplasty, NIDDM, and COPD, atrial flutter, Concomitant medications included lisinopril, digoxin, furosemide, isosorbide dinitrate, hydralazine, warfarin, chlorpropamide, simvastatin, ranitidine, salbutamol, tolbutamide, beclomethasone, and ketoprofen. Patient had low hb/hct at baseline (12.7/37) which worsened during the study . He has an episode of mild CHF on day 74. Anemia of 9.7/29 was reported and the study drug was discontinued on day 284.
Anemia reported with Safety update
Anemia; transfusion	Patient No: AC-052-352 /10030. 67-year-old male with pulmonary hypertension secondary to systemic sclerosis receiving furosemide, spironolactone, aspirin, lisinopril, and omeprazole. Medical history included systemic hypertension, hiatal hernia, ischemic heart disease with CABG, prostate cancer, GERD, and CREST syndrome. He was diagnosed with anemia on day 28 (on day 36 hb/hct: 6.9/23). He received 4 units of blood over 2 days. He remained on study drug.

In comparison, there were 2 placebo patients (0.9%) in the clinical trials who received blood transfusions (fax 7-2-01). The one case of a normal bone marrow biopsy (#18202/9032) from a patient with anemia who received bosentan is reassuring.

7.1.5 Safety Update

7.1.5.1 ENABLE

This ongoing, still blinded, placebo controlled trial in patients with CHF is using bosentan doses 62.5mg bid titrated to 125 mg bid.

As of the clinical cut-off, 16 (1.0%) reported anemia as a serious adverse event. A marked

decrease in hemoglobin of 15% from baseline was reported for 7.4% of patients (120/1613). Eight patients had hemoglobin of <8 g/dl.

Number and (percent) of all study patients

15% decrease in hemoglobin from baseline and hemoglobin
<11 g/l but > 10 g/l	<10 g/l but > 9 g/l	<9 g/l but > 8 g/l	< 8g/l
55 (3.4)	39 (2.4)	18 (1.1)	8 (0.5)

Table 17 of safety update

7.1.5.2 PAH ongoing blinded (AC-052-352)

There were 6 reports of anemia. One patient (10030) required transfusion.

7.2 White blood count

7.2.1 Mean changes

Mean changes from baseline at endpoint for various parameters are shown below by treatment group.

Means at baseline, mean change from baseline, % change from baseline

	Placebo			Bosentan
	Baseline	Change	% change	Baseline	Change	% change
Leukocytes 10⁹/L	7.6	7.5	0	7.1	-0.5	-7.0
Neutrophils 10⁹/L	4.5	0.2	4.4	4.43	-0.4	-9
Lymphocytes 10⁹/L	1.96	-0.04	-2	1.88	-0.13	-6.9
Monocytes 10⁹/L	0.47	0	0	0.44	-0.01	-2.3
Eosinophils 10⁹/L	0.19	0.01	5.3	0.18	0.06	27.3
Platelets 10⁹/L	218	7	3.2	216	3	1.4

Fax sent 3-7-01

Overall, there tended to be a decrease in white blood cell parameters in the bosentan group compared to placebo except for eosinophils. However, the only consist finding was the increase in eosinophils. The placebo subtracted mean eosinophil count increased 22% over baseline suggesting a drug induced phenomenon. Changes in platelets were unremarkable.

8.0 All adverse events

8.1 All bosentan patients

Adverse events reported by at least 2% of the 571 patients who received bosentan are shown below.

A majority of bosentan patients (74.1%) reported at least one adverse event. Cardiac failure was reported most often (21.9%), followed by headache (14.2%), dizziness excluding vertigo (13.1%), and hypotension (13.1%).

8.1.1 Dose

The table below shows all events reported by at least 2% of the all bosentan group, by the dose of bosentan.

Since the sample sizes for the different dose groups are small and uneven, it is difficult to draw conclusions from this table. However, headache, flushing, lower limb edema were reported more often in the 2000 mg bosentan group and seem to be dose related.

8.1.1.1 Selected adverse events from NC15020 (hypertension)

This randomized, placebo controlled, parallel group, placebo and active controlled hypertension study used the widest dose range of all the bosentan studies. Duration of treatment was 4 weeks.

The percent of patients reporting at least of the selected adverse event, by treatment and dose group, is shown below.

Percent of patients

		bosentan				Enalapril
	Placebo n=49	100 mg n=50	500 mg n=49	1000 mg n=45	2000 mg n=50	20 mg n=50
Any event	57.1	42.0	49.0	51.1	56.0	50.0
Headache	18.4	14.0	14.3	20.0	24.0	12.0
Flushing	0	6.0	12.2	8.9	18.0	2.0
Edema+	0	6.0	10.2	4.4	24.0	0

+includes face and leg edema

Vol2.34 appendix 7

Although headache was reported at a high rate by the placebo group (18.4%), there was somewhat more headache reported in the 2000 mg bosentan group (24%). It is increasingly clear that flushing and various types of edema are related to bosentan use.

8.1.2 Duration in treatment

Reported adverse events, grouped by planned treatment duration (< 12 weeks and >12 weeks), are shown below.

As expected, the longer the patient is in a study, the higher the reporting of an adverse event (56.7% for patients in study 12 weeks or less compared to 90.2% for greater than 12 weeks). Therefore, without concurrent placebo groups, it is hard to draw conclusions about any findings. That said, while headache was reported at roughly the same rate regardless of length of planned treatment, dizziness and abnormal hepatic function were reported 10 times more often by patients who were on drug longer. Also, anemia was reported 7 times more often in patients with the longer planned treatment duration.

8.2 Placebo controlled trials

Adverse event reporting rates in the placebo controlled trials is shown below by treatment group.

Slightly more bosentan patients reported at least one adverse event (73.7%) compared to the placebo patients (71.2%).

Adverse events of reported by at least 7 bosentan patients (2.8%) and were reported by at least 1% more bosentan patients than placebo patients are shown below.

No. and (percent) of patients

	Placebo N=219	Bosentan N=533	% Placebo subtracted
Flushing	2 (0.9)	32 (6.0)	5.1
Lower limb edema	0	20 (3.8)	3.8
Headache	23 (10.5)	76 (14.3)	3.8
Abnormal hepatic function	4 (1.8)	26 (4.9)	3.1
Anemia^	2 (0.9)	20 (3.7)	2.8
Angina	2 (0.9)	15 (2.8)	1.9
Pruritus	0	7 (1.3)	1.3
Edema NOS	1 (0.5)	9 (1.7)	1.2
Hypokalemia	1 (0.5)	8 (1.5)	1.0

^includes hemoglobin decreased

Bosentan use is associated with more reports of flushing, edema, headache, abnormal hepatic function, anemia, angina, pruritus and hypokalemia compared to placebo use.

8.2.1 Dose

Selected adverse events are shown in the following table, by dose.

Number and (%) of patients reporting

	placebo	bosentan
Adverse event	n=219	<1000 mg n=120	1000-1500 mg n=310	2000 mg n=103	All bosentan n=533
Any event	156 (71.2)	65 (54.2)	265 (85.8)	63 (61.2)	393 (73.7)
Headache	23 (10.5)	20 (16.5)	35 (11.3)	21 (20.4)	76 (14.3)
Dizziness exc.vertigo	26 (11.9)	7 (5.8)	56 (18.1)	2 (1.9)	65 (12.2)
Flushing	2 (0.9)	11 (9.2)	10 (3.2)	11 (10.7)	32 (6.0)
Hepatic function abnormal	4 (1.8)	1 (0.8)	25 (8.1)	0	26 (4.9)
Lower limb edema	0	3 (2.5)	7 (2.3)	10 (9.7)	20 (3.8)
Cough	6 (2.7)	6 (5.0)	10 (3.2)	3 (2.9)	19 (3.6)
anemia	2 (0.9)	0	19 (6.1)	0	19 (3.6)
Angina	2 (0.9)	2 (1.7)	10 (3.2)	3 (2.9)	15 (2.8)

Although the numbers are small in some of the dosing groups, it looks like headache, flushing, and lower limb edema are dose related. Even though the pattern is not clear from this table, hepatic function abnormal and anemia are also dose related (see sections 6 and 7).

Reported adverse events with addition of AC-052-352

The numbers and percents of patients with an adverse event that was a) reported by at least 8 bosentan patients and b) at least 1% more bosentan patients reported the event compared to placebo patients are shown below.

No. and (percent) of patients

	Placebo N=288	Bosentan N=677	Placebo subtracted (%)
Any event	220 (76.4)	529 (78.1)	1.7
Flushing	5 (1.7)	45 (6.6)	4.9
Edema#	8 (2.8)	51 (7.5)	4.7
Abnormal hepatic function	6 (2.1)	40 (5.9)	3.8
Headache	37 (12.8)	107 (15.8)	3.0
Anemia^	3 (1.0)	25 (3.7)	2.7
Pruritus	0	12 (1.8)	1.8
Angina	3 (1.0)	15 (2.2)	1.2
Palpitations	5 (1.7)	18 (2.7)	1.0
Dry mouth	1 (0.3)	9 (1.3)	1.0

^includes hemoglobin decreased

#includes lower limb edema, edema NOS, peripheral edema

From fax dated 6-01-01

The reported events flushing, abnormal hepatic enzymes, anemia, edema, headache consistently show up in most of the lists of events.

9.0 Vital signs

Mean baseline and mean change from baseline at endpoint for blood pressure, heart rate, and weight, all doses combined and by treatment group, are shown below.

Bosentan, compared to placebo, produced a small mean decrease in systolic and diastolic blood pressure. Mean changes in heart rate and weight were similar in both groups.

Dose

The effects of bosentan on blood pressure by dose was examined in study NC15020, a 12 week hypertensive study. Results are shown for systolic and diastolic blood pressure and heart rate in the tables below.

Systolic blood pressure

Placebo subtracted changes from baseline at endpoint for bosentan total daily doses 100 mg, 250-500 mg, 1000-1500 mg, 2000 mg were -1.4 mmHg, -0.9 mmHg, -4.7 mmHg, and -3.2 mmHg, respectively.

Diastolic blood pressure

Placebo subtracted changes from baseline at endpoint for bosentan total daily doses 100 mg, 250-500 mg, 1000-1500 mg, 2000 mg were +1.4 mmHg, +0.2 mmHg, -1.8 mmHg, and +0.5 mmHg, respectively.

Overall, bosentan has only a minor effect on blood pressure.

Heart rate

Placebo subtracted changes from baseline at endpoint for bosentan total daily doses 100 mg, 250-500 mg, 1000-1500 mg, 2000 mg were +0.4 bpm, -0.3 bpm, +0.7 bpm, and –3.0 bpm, respectively. Bosentan probably has a minor effect on heart rate.

10.0 ECG changes

Treatment emergent ECG abnormalities reported in the placebo controlled trials are shown below.

There was a higher incidence rate of placebo patients with an ECG finding compared to bosentan patients. Nearly all findings were more frequent in placebo group with the exception of left anterior hemiblock and left atrial enlargement. Nothing in the above table suggests that bosentan has a adverse effect on the heart.

ECG intervals

There is no indication that bosentan has an effect on ECG intervals. There was 1 report of torsades: Patient 18102/6162 Bosentan 125-250-500 mg bid. a 73-year-old white female with heart failure NYHA class IIIB, treated with lisinopril, furosemide, digoxin, and levothyroxine, was hospitalized on day 43 for severe upper quadrant abdominal pain, elevated liver enzymes AST 64 U/l, ALT 149 U/l, and alkaline phosphatase (166 U/l). Cholelithiasis was diagnosed by ultrasound and the study drug was stopped. The patient developed ventricular tachycardia (described as torsades de pointe). After successful defibrillation, she was intubated and treated with lidocaine intravenously. Prolonged QT interval was identified. The patient was extubated and a laparoscopic cholecystectomy was performed. She was discharged and resumed the intake of study drug. Drug was again stopped because of elevation of LFTs and 2 weeks later LFTs were normal. Study drug was restarted and 2 weeks later LFTs increased (highest AST 161 U/I and ALT 283 U/l). Patient was permanently discontinued and LFTs became normal. The torsades event was probably unrelated to the use of bosentan.

11.0 Clinical Pharmacology

Summary

There were 23 clinical pharmacology studies. These are shown in the diagram below.

Healthy volunteers

Single dose

There were 5 studies with healthy volunteers. Total daily doses ranged from 3 mg to 2400 mg.

There were no deaths in the clinical pharmacology studies. One subjects (#14219/0011) withdrew because of T wave abnormalities on ECG.

Adverse events

There was a dose response for headache/head discomfort with 100% of subjects reporting headache with the 2400 mg dose.

Multiple dose

There were 54 healthy volunteers who received between 100 mg and 1400 mg doses of bosentan for a mean of 6.3 days. There were no deaths, serious adverse events, or withdrawals for adverse events. Three subjects had increases in ALT (<3xULN).

Patients

Single dose

There were 53 patients who received a single oral dose of bosentan with total daily doses ranging from 125 to 2000 mg. There were no deaths, serious adverse events, or study withdrawals because of an adverse event.

Interaction studies

Age

Adverse events by age group: there is no study that evaluated this potential interaction. The table below shows the placebo subtracted incidence rate of selected adverse events by age.

Placebo subtracted incidence rate (%)

	<50 yrs n=117	50-74 yrs n=368	>74 yrs n=48
Any event	-5.8	-1.1	13.7
Headache	9.0	1.5	4.0
Flushing	7	4.3	4.2
Abnormal LFTs	0.1	3.8	4.2
Lower limb edema	3.4	3.5	6.3
Anemia	0.9	2.7	8.2

Table 60

The overall incidence rates of reporting adverse events were higher for placebo in the younger age groups. The rate, however, for patients above the age of 74 was nearly 14%. Of the events listed above, the higher rate of anemia in the oldest age group (8.2%) is disturbing.

Gender

Adverse events by gender group: there is no study that evaluated this potential interaction. The table below shows the placebo subtracted incidence rate of selected adverse events by gender.

Placebo subtracted incidence rate (%)

	Male N=383	Female N=150
Any event	1.2	6.7
Headache	2.2	8.0
Flushing	6.0	3.2
Abnormal LFTs	2.0	5.3
Lower limb edema	3.1	5.3
Anemia	3.0	3.3

Table 62

Of the 533 patients who received bosentan in a placebo controlled trial, 28% were female. A higher incidence of females on bosentan reported at least one adverse event (6.7%), compared to males (1.2%). Headache was reported nearly 4 times more in females on bosentan than the males.

Race

Adverse events by race: there is no study that evaluated this potential interaction. There were very few black patients in the bosentan studies: 26 (4.9%) received bosentan and 9 (4.1%) placebo in the placebo controlled trials. Any attempt to draw conclusions about a drug-race interaction is of no value.

Drug-drug

There were 7 studies that evaluated the effects of concomitant use of bosentan and cyclosporin, warfarin, digoxin, nimodipine, ketoconazole, losartan, simvastatin, or glibenclamide.

Bosentan metabolism is shown below

Drugs that inhibit CYP3A4 and/or CYP2C9 are expected to cause an increase in the plasma concentration of bosentan. Conclusions about the pharmacokinetics of the drugs were obtained from Dr. Gabriel Robbie’s biopharmacology review.

Ketoconazole

Concomitant administration of ketoconazole significantly increases the Cmax and AUC of bosentan by 62% and 83%, respectively.

There were 12 study subjects. No adverse events were reported. One subject (#5) dropped out during the bosentan only phase; the sponsor gave no explanation. Five subjects had decreases from baseline in both hemoglobin and hematocrit.

Cyclosporin

Concomitant administration of bosentan and cyclosporin increased the trough concentration of bosentan by 30-fold. It is recommended that the concomitant use of bosentan and cyclosporin be contraindicated.

In the cyclosporin interaction study, 2 subjects receiving bosentan 2000 mg/d and cyclosporin reported severe headache and gastric disturbance resulting in dose reduction. Both completed the study. One subject receiving bosentan 1000 mg/d and cyclosporin reported severe nausea and vomiting. Subject continued to be symptomatic (myalgia, numbness of extremities, retrosternal burning and epigastric pain) even though dose was reduced. He was withdrawn from study drug.

Simvastatin

Coadministration of bosentan and simvastatin significantly decreased steady-state C_max and AUC of both simvastatin (31% and 49%, respectively) and its active metabolite, -hydroxy simvastatin, (by 33% and 60%, respectively). The concomitant use of bosentan and simvastatin could reduce the effectiveness of simvastatin.

There were 12 subjects: 4 male and 8 female. There were no serious events and no drop outs. Reported adverse events included flu (1) and headache (2). There were a few minor changes in LFTs. One subject had a decrease in hemoglobin/hematocrit from 12.6/36.9 at baseline to 10.6/31.5 at study end.

Digoxin

Bosentan 500-mg BID administered for 7 days slightly decreased the C_max and AUC of digoxin by 9% and 12%, respectively. Day 14 C_min of digoxin decreased by 30% in the presence of bosentan. Comparison of the pharmacokinetics of 500-mg BID bosentan from the present study with another study (B-159037) in healthy individuals indicated no effect of concomitant administration of digoxin on bosentan pharmacokinetics.

There were 19 subjects entered into the study. There were no deaths and no serious adverse events. One subject withdrew because of mild 2nd degree AV block after receiving digoxin alone and one subject withdrew because of bronchitis. The reporting of adverse events was similar for the 2 groups.

Warfarin

Steady-state bosentan increases the elimination of both R- and S-warfarin which, in turn, reduces the anticoagulation effect of warfarin as measured by prothrombin time and factor VII activity. Primary pharmacodynamic measures such as PT_max,cor decreased by 23% and AUC_PT,cor decreased by 38%. Consistent with the unaltered Cmax of R- and S-warfarin, the time to maximum effect and time course of warfarin effect on prothrombin time and factor VII activity were not altered by bosentan. The trough plasma concentration 12 hours after the warfarin dose were statistically significantly lower compared to the predose concentrations. The concomitant use of bosentan and warfarin might reduce the effectiveness of warfarin. Prothrombin times should be checked more often.

Single-dose warfarin decreased the mean steady-state trough concentration of bosentan by 63%.

Safety

There were 12 subjects. No subject died, reported a serious adverse event, or dropped out of the study because of an adverse event. Of the subjects receiving warfarin plus bosentan, 92% reported headaches compared to 26% of subjects receiving warfarin alone. There is considerable concern regarding the reduction of the anticoagulation effect of warfarin when used in conjunction with bosentan.

Losartan

Concomitant administration of losartan lowered both Cmax and AUC of bosentan by 15%.

Glibenclamide

Coadministration of bosentan and glibenclamide significantly decreased steady-state C_max and AUC of both bosentan and glibenclamide. Steady-state Cmax and AUC of bosentan decreased by 24% and 29%, respectively, while that of glibenclamide decreased by 22% and 40%, respectively. The concomitant use of bosentan and glibenclamide could reduce the effectiveness of glibenclamide. Blood glucose should be checked more often.

Safety

There were 12 subjects. There were no reported deaths, serious adverse, or study withdrawals because of an adverse event. Four subjects reported adverse events: headache, headache and fatigue, abdominal pain and increased LFT, increased LFTs. No hematology values were obtained beyond screening.

Drug-disease

Mean C_max and AUC of bosentan were 37% and 11% lower in severe renal impairment patients compared to healthy subjects. Median Tmax and T_1/2, however, were similar in both groups. The concentrations of the active metabolite Ro 48-5033 increased in severe renal impairment patients but were <10% of parent drug concentrations. Concentrations of the other metabolites, Ro 47-8634 and Ro 64-1056, increased by 100% in severe renal impairment patients compared to normal subjects indicating that the renal route is a major route of elimination of the metabolites of bosentan. The use of bosentan in severe renally impaired patients should be discouraged. Its use in less impaired patients is probably acceptable.

Safety

There were no reported deaths or serious adverse events. The only subject reporting an adverse event was a healthy subject with headache and weakness. Blood pressure was lowered to a greater extent in the renally impaired subjects. However, it is not possible to drawn conclusions about blood pressure effects since the study was open label. There were no reports of hypotension.

There was no study evaluating the effect of bosentan in patients with liver impairment. It is reasonable to contraindicate the drug in patients with elevation of LFT > 3 times ULN, as was done in the protocols.

[1] Total = 1066 (571 from therapeutic trials+144 from 352 trial+351 from clinical pharmacology trial)

[3] patient 110 10037 in study AC 052 352 (see summary of efficacy) and 20060 00841 in study RO15464

[6] ENABLE trials are included in the table although they are still ongoing

[8] this total does not include study AC 052 352

[9] patient 110 10037 in study AC 052 352 (see summary of efficacy) and 20060 0084 in study RO15464

[10] Upper limits of normal for ALT and AST were 30 U/l and 25 U/l, respectively.

[12] Sponsor states that fever was erroneously listed (fax dated 3-27-01).

[13] And a second patient reported jaundice from the recently concluded study AC 052 352 (see medical review of efficacy).

[14] defined as 15% decrease from baseline value and less than 11 g/dl

[15]defined as 15% decrease from baseline value and less than 36%

[16] defined as 100% increase from baseline value and less than 1.5 10⁹/L