A large prospective open-label, multicentre SAMM study, comparing the safety of aceclofenac

with diclofenac in patients with rheumatic disease

Авторы:
Huskisson E.C.
Irani M.
Murray F.

• Summary
• Introduction
• Patients and methods
• Results
• Discussion

Издание:"European Journal of Rheumatology and Inflammation"; Vol 17, Issue 1, 2000; 14a Milford
House, 7 Queen Anne Street, London, England; Department of Rheumatology, Ashford Hospital,
Middlesex, England; Departments of Clinical Pharmacology/Gastroenterology, Hepatology Unit,
Beaumont Hospital and Royal College of Surgeons of Ireland, Dublin, Ireland

This large prospective community-based study indicates that aceclofenac is significantly better tolerated
than diclofenac. SAMM is a valuable method for detecting differences in drug profiles in clinical practice.

Patients (>=18 yrs) with rheumatoid arthritis, osteoarthritis or ankylosing spondylitis were recruited in an
open-label, multicentre observational study complying with the Safety Assessment of Marketed Medicines
(SAMM) guidelines. They received aceclofenac 100 mg bid or diclofenac (sustained release) 75 mg bid in
a 3:1 ratio. Safety was measured by the incidence of adverse events, discontinuation due to adverse
events, serious adverse events and death.

Overall, 1078 GPs enrolled 10142 analysable patients (aceclofenac n = 7890; diclofenac n = 2252).
Adverse events (P<0.001), gastrointestinal adverse events (P<0.001) and patients withdrawing from
treatment (P<0.001) were significantly less common with aceclofenac than diclofenac. Few serious
adverse events (1.6%) occurred with either medication.

KEY WORDS: Rheumatoid arthritis, osteoarthritis, ankylosing spondylitis, aceclofenac, diclofenac, safety,
adverse event, SAMM study.

Introduction

Non-steroidal anti-inflammatory drugs (NSAIDs) are widely used for a variety of musculoskeletal
disorders, and their efficacy is well established. However, adverse events, particularly upper
gastrointestinal, often lead to treatment withdrawal. More serious adverse drug reactions, including peptic
ulcer and its complications may occur, with the elderly and subjects with a history of gastrointestinal
orders being particularly at risk. Renal and hepatic side effects are also occasionally seen, There is an
obvious need for new treatments with better side-effect profiles.

Following oral administration, aceclofenac, a novel phenylacetic acid derivative ([2-(2',6'-

dichlorophenylamino) phenyl] acctoxyacetic acid) is rapidly and well absorbed, with a half-life of 4 hours.
Numerous clinical trials have established its efficacy as an effective NSAID in the treatment of
osteoarthritis, rheumatoid arthritis and ankylosing spondylitis compared with ketoprofen indomethacin
diclofenac naproxen piroxicam and tenoxicam. Preclinical animal studies suggested that the potential of
aceclofenac to cause gastrointestinal damage is less than that of diclofenac and other NSAIDs. In
rodents, the acute gastric ulcerogenic activity of aceclofenac was found less than naproxen, diclofenac or
indomethacin, respectively.

Randomised, double-blind trials in patients with osteoarthritis have suggested that aceclofenac causes
less gastrointestinal adverse effects and treatment withdrawal than comparator NSAIDs. A meta-analysis
of 13 phase III trials including 3574 patients compared aceclofenac with other NSAIDs. The incidence of
adverse events, gastrointestinal adverse events and withdrawals was significantly (P<0.001) lower with
aceclofenac. Short-term placebo-controlled trials have shown that aceclofenac has similar tolerability to
placebo in patients with rheumatoid arthritis.

This large observational SAMM (Safety Assessment of Marketed Medicines) study was carried out in
general practice to compare the incidence of adverse events of aceclofenac with diclofenac in patients
with rheumatic disease. The purpose of SAMM studies is to enhance the knowledge of a drug by
analysing its safety profile in conditions of normal clinical practice. Diclofenac was chosen as the
comparator, as it is the most widely used NSAID worldwide, with a well-documented side-effect profile.

Patients and methods

Study design

An open-label, non-blinded observational study was conducted comparing the safety of aceclofenac with
diclofenac. The study was conducted under the SAMM guidelines issued by a working party, which
included representation from the Medicines Control Agency (MCA), the Committee on the Safety of
Medicines, the British Medical Association, the Association of the British Pharmaceutical Industry and the
Royal College of General Practitioners.

Patient population

Over a 2-yr period, patients from the UK were enrolled by their GPs. Men and women >18 years of age
were eligible for study entry if they had clinical evidence of rheumatoid arthritis, osteoarthritis or
ankylosing spondylitis. Exclusion criteria were: active or suspected peptic ulcer, gastrointestinal bleeding,
moderate or severe renal impairment (serum creatinine >300 µmol/l), pregnancy or intended pregnancy
during the study, previous adverse events (eg asthma attacks, acute rhinitis or urticaria) or
hypersensitivity to NSAID or aspirin therapy, or previous NSAID therapy (except ibuprofen) during the 30
days before the baseline visit.

Study drugs, stratification and assessments

Patients were prescribed aceclofenac 100 mg bid or diclofenac (sustained release) 75 mg bid prior to
study entry at the discretion of their GP. The protocol recommended that patients were allocated
aceclofenac or diclofenac in a 3:1 ratio in order to allow a higher power to detect unusual adverse events
with aceclofenac. The treatment period lasted for 12 months, during which time concomitant medications
were permitted.

Baseline information (medical history, renal status, tobacco and alcohol use) was recorded for all
patients. During routine visits to the General Practitioners (GPs), adverse events were recorded at
months 1, 6 and 12, using case report forms. Adverse events were also followed up for 30 days after the
last dose of study medication.

Evaluation criteria and definitions

The primary study end point was medication safety, measured by the incidence of adverse events,
discontinuation due to adverse events and serious adverse events or death. In a subgroup analysis,
safety was also assessed in relation to age, gender, race and any other clinically relevant patient
characteristics (eg. disease type, disease severity, and tobacco and alcohol consumption).

Data from all patients who received a study medication were included in the safety analyses. Adverse
events were recorded by body system and categorised according to the Coding Symbols for a Thesaurus
of Adverse Reaction Terms (COSTART).

Adverse events were graded by the patient as mild, moderate or severe and were designated by the
patients' GP as drug-unrelated, possibly drug-related, probably drug-related or drug-related.

Serious adverse events were defined as any of the following: death, hospitalisation or prolongation of
hospitalisation, any other life-threatening condition, permanent disability, overdose, cancer or congenital
abnormality.
Discontinuation was attributed to any of the following: adverse event, insufficient therapeutic effect,
protocol deviation, non compliance, lost to follow up or other reason and unavailable information.

Statistical analyses

Statistical analyses were performed to compare the demographics, disposition, extent of study medication
exposure, incidence of all adverse events, serious adverse events and deaths in both treatment groups.

Categorical characteristics such as gender and race were compared using the chi-square test. The
treatment groups were compared with respect to continuous variables using analysis of variance
(ANOVA).

Comparisons of the incidence of adverse events, drug-related and serious adverse events in each group
were compared using Fishers exact test. Comparisons of the number of patients withdrawing from the
study prematurely were performed according to the reason using the chi-square test.

Results

Patient Characteristics

In total, 10142 patients were enrolled by 1078 GPs into the study. Among these, 7890 patients received
aceclofenac and 2252 patients received diclofenac, which represented a 3.5:1 ratio. In addition, 159
patients received an unspecified treatment.

Most patients were aged >= 50 years (73.9%), Caucasian (88.1%) and suffered from osteoarthritis
(91.4%). In addition, more patients were female (59.9%), except those with ankylosing spondylitis, where
males predominated (76.6%). The majority of patients had arthritis for >1 yr (76.1%), while 28.1% had the
disease for more than 5 yrs.

Overall, there were no differences in the demographic characteristics of the two treatment groups with
respect to age distribution, gender, race and disease type (table 1). However, the duration of the disease
was significantly longer for patients receiving aceclofenac than diclofenac (P = 0.019). Furthermore, a
significantly (P<0.001) greater number of patients (15.1%) receiving aceclofenac exhibited severe arthritic
symptoms compared with diclofenac (12.1%).

Table 1. Demographics of the patient population

Characteristic Aceclofenac (n = 7890) Dictofenac (n = 2252) P value

Age (mean +- SD) 58.5 +- 13.8 58.6 +- 13.9 NS
Age range 17-99 18-93 NS
Female gender (%) 59.7 56.8 NS
Caucasian (%) 87.7 89.7 NS
Osteoarthritis (%) 91.4 91.4 NS
Rheumatoid arthritis (%) 6.5 6.5 NS
Ankylosing spondylitis (%) 2.1 2.1 NS
Disease duration (%)
<1 yr 21.8 24.8
1-2 yrs 24.1 23.8
3-5 yrs 24.3 24.4
>5 yrs 28.9 26.3
Disease severity (%) <0.001
mild 11.4 13.9
moderate 73.5 74.0
severe 15.1 12.1
Gastrointestinal history (%)
peptic ulcer 2.0 1.5 NS
dyspepsia 12.0 10.2 0.025
epigastric pain 5.1 4.5 NS
gastrointestinal bleeding 0.8 0.7 NS

About 20% of the total patient population were suffering from or had previously suffered from
gastrointestinal disease. Overall there was little difference between the groups, but significantly more
aceclofenac patients had a history of dyspepsia (P = 0.025; table 1).

Safety analysis

In total, 27.9% receiving aceclofenac and 27.0% patients receiving diclofenac completed 12 months of
treatment. The mean duration of drug exposure was similar for both groups: 168.1 days for aceclofenac
and 170.0 days for diclofenac.

Adverse events

Overall, both study drugs were well tolerated with most adverse events being mild to moderate in nature.
Fewer patients receiving aceclofenac (22.4%) reported adverse events compared with patients receiving
diclofenac (27.1%; figure 1) (P<0.001). The most frequent adverse events recorded in both groups
involved the digestive system (11.5%), the body as a whole (6.8%) and the nervous system (2.7%).
Furthermore, the incidence of adverse events affecting the gastrointestinal system was also significantly
lower for aceclofenac than diclofenac (figure 1), with 10.6% of aceclofenac recipients reporting adverse
events compared with 15.2% for diclofenac (P<0.001).

The number of adverse events occurring in >1% of patients is presented in table 2. The reported
frequency of dyspepsia, nausea, abdominal pain, and diarrhoea were significantly more common among
the diclofenactreated patients. The incidence of these events was 1.3, 1.5, 1.8, and 2.5 fold higher,
respectively, for diclofenac compared with aceclofenac.

Table 2. Percentage of patients reporting adverse events with a frequency of >=1 %

Nature of adverse event Aceclofenac (n = 7890) Diclofenac (n = 2252) P value

Dyspepsia 5.4 6.7 0.017
Abdominal pain 2.5 4.4 < 0.001
Diarrhoea 1.5 3.6 < 0.001
Nausea 1.6 2.4 0.01
Dizziness 1.1 0.7 0.073
Infection 1.0 0.8 NS

NS= not significant

Most gastrointestinal adverse events were mild or moderate in nature for both groups. The incidence of
adverse events affecting the nervous system was low for both treatment groups and their severity was
mostly mild or moderate. Overall, the incidence was significantly higher for aceclofenac (3%) than
diclofenac (1.9%; P = 0.007). The difference was mainly due to the more frequent occurrence of
dizziness, depression, and headache with aceclofenac.

The incidence of adverse events affecting the other body systems was very low (<1%), and there was no
significant difference between the treatment groups, although, significantly more patients reported allergic
reaction, congestive heart failure, vasculitis, mouth ulceration, ecchymosis. apnoea, alopecia and
psoriasis with diclofenac (P<0.05). There was also no special concern regarding the occurrence of rare
adverse events in either treatment group.
The incidence of adverse events in the patient subgroups (eg gender, age, history of peptic ulcer, alcohol
and tobacco consumption) paralleled the trends observed in the overall study population with respect to
treatment group differences. Adverse events for both treatment groups were more frequent in the two
oldest age groups, and more females (25%) than males (21%) exhibited adverse events.

Discontinuation

During the study, 59.0% of patients receiving aceclofenac discontinued therapy compared with 63.7% of
patients receiving diclofenac (P<0.001). The most common reasons for discontinuation were insufficient
therapeutic effect (15.9%), adverse events (14.9%) and lost to follow-up (10.1%).

Significantly fewer patients receiving aceclofenac (14.1%) withdrew due to adverse events compared with
diclofenac (18.7%; figure 1) (P<0.001). The most common adverse events leading to withdrawal were
gastrointestinal in nature. In particular, the incidence of nausea, abdominal pain and diarrhoea leading to
discontinuation was 46%, 65% and 41% respectively, lower for aceclofenac compared with diclofenac
(P<0.001).

Serious adverse events

Serious adverse events were recorded by 1.5% of patients taking aceclofenac compared with 1.9% of
patients receiving diclofenac (NS). Among these, 22.2% were considered to be related to aceclofenac
and 34.1% were considered to be related to diclofenac. Serious adverse events resulting in
hospitalisation occurred in 1.2% patients receiving aceclofenac and 1.5% of patients receiving diclofenac
(NS).

The incidence of serious gastrointestinal bleeding, ulcers and perforations was low and similar in both
groups. Serious gastrointestinal haemorrhages only occurred in five patients. Among these, four (0.05%)
occurred in the aceclofenac group and one (0.04%) in the diciofenac group. Serious gastrointestinal
ulcers or perforations were only observed in one patient receiving aceclofenac and two patients receiving
diclofenac. Two patients (0.03%) receiving aceclofenac and one patient receiving diclofenac (0.04%)
developed haematemeses. There were no reports of serious hepatic adverse events.

Deaths

During the study, or within 30 days of taking the final dose, 26 (0.3%) patients receiving aceclofenac and
17 (0.8%) patients receiving diclofenac died (NS). The majority of these patients (74.4%) were elderly (
70 yrs of age).

Discussion

This report describes the findings of the largest long-term prospective observational study evaluating the
adverse drug reactions of NSAIDs in daily clinical practice. The study was designed to comply strictly with
SAMM guidelines and was approved by the MCA. The patient population was representative of normal
clinical practice.

The major findings were that aceclofenac was associated with a significantly better gastrointestinal
adverse event profile and tolerance than diclofenac, despite the fact that patients receiving aceclofenac
had a higher incidence of previous gastrointestinal symptoms than diclofenac recipients. The differences
were large and statistically highly significant. For example, the incidence of abdominal pain, diarrhoea
and nausea that led to discontinuation was approximately twice as high in patients receiving diclofenac as
in those receiving aceclofenac. This may lead to better treatment acceptability with aceclofenac.

The mechanism for the enhanced adverse event profile is currently unclear. The relevant mechanisms
are complex and thus are unlikely to be related solely to any one biological effect, such as differential
suppression of COX-1 and COX-2. However, a recent endoscopic study has shown that diciofenac
significantly impairs gastric mucosal hexosamine content and blood flow. In contrast, gastric mucosal
hexosamine was significantly increased and gastroduodenal blood flow remained unchanged with
aceclofenac. Although the exact mechanism of the hexosamine increase is unknown, the slight structural
modification of aceclofenac may exert a preventative effect on gastric mucosal damage by altering
mucosal secretion.
The improved gastrointestinal tolerability of aceclofenac compared with diclofenac reported in this large
community-based SAMM study confirms the findings of several clinical trials and a meta-analysis
comparing aceclofenac with a number of competitor NSAIDs. As SAMM studies enhance the knowledge
of a drug by analysing its safety in large numbers of patients representative of normal practice, we
speculate that this enhanced safety profile of aceclofenac may lead to fewer serious complications in the
everyday clinical situation.