Pre-clinical information

NSAIDs are essential drugs for the management of acute and chronic pain and inflammation. Serious adverse effects limit the usefulness of NSAIDs, especially gastrointestinal adverse effects and effects on renal function. Significant efforts went into developing NSAIDs that have less risk of gastrointestinal adverse effects.

• Two cyclo-oxygenase (COX) isoforms discovered: COX-1, COX-2 (COX-3 has since been discovered)

• Rofecoxib shown to selectively inhibit COX-2 in vitro—indeed it has greater selectivity for COX-2 than celecoxib and meloxicam

• What is known of its chemical, pharmacological and toxicological properties suggest it may be a safe, efficacious and feasible molecule to enter clinical drug development

Safety, efficacy and feasibility are equally important in clinical drug development—if the drug does not meet any one of them, it will not be brought to market. A couple of terminological notes: (i) it is worth distinguishing between “efficacy” and “effectiveness”. A treatment is efficacious if it has the expected beneficial effect on patients when tested under experimental conditions. A treatment is effective if it has the expected beneficial effect on patients when used in routine care. (ii) “Feasible” carries a lot of weight in this context. The bottom line is whether the drug is financially viable to bring to market. Clearly, this will depend on the safety and efficacy of the product, but it will also depend on whether the product can be given orally (pharmaceutics), how often the product will need to be taken (pharmacokinetics), what competitors are available, how much it will cost to make the drug, and much else.

Phase I

Key questions

• Does rofecoxib selectively inhibit COX-2 in healthy volunteers? (What is the \(IC_{50}\)?)

• Is selective inhibition of COX-2 likely to reduce gastrointestinal adverse effects?—how will we measure this?

• Is rofecoxib likely to be an efficacious anti-inflammatory?—will it cause less GI events compared to traditional NSAIDs at doses that are efficacious?

• What dose should be used?

Phase I trials for rofecoxib focussed on answering: how the drug works (pharmacology) and how the drug is treated by the body (pharmacokinetics).

These trials seek to define and measure the therapeutic target (selective COX-2 inhibition) and assess the safe dose range. This includes identifying the “maximum tolerated dose” (MTD). This this is the dose above which most participants enrolled in the study had to stop the drug due to toxicity.

The Phase I trials typically involve small numbers of healthy volunteers, between 20–80 participants

Key findings from Phase I (Scott and Lamb 1999)

• Dose-dependent COX-2 inhibition.

  • \(\mathit{IC_{50}}\) for COX-2 inhibition 0.77 \(\mu\)mol/L at doses 25–1000mg. No inhibition of COX-1 at the same doses.

  • Indomethacin: COX-1 \(\mathit{IC_{50}}\) 0.09 \(\mu\)mol/L; COX-2 \(\mathit{IC_{50}}\) 0.30 \(\mu\)mol/L

• No effect on COX-1 in healthy volunteers up to doses of 375mg/day

• Less endoscopically visible ulcers from healthy volunteers taking rofecoxib 250mg daily compared to ibuprofen 800mg tds or aspirin 650mg qid (49, 51 and 17 subjects respectively)

• Determined pharmacokinetics in a health population: \(C_{max}\), \(t_{\frac{1}{2}}\), \(V_{SS}\), …

• Identified: maximum dose at which rofecoxib will produce the desired effect without unacceptable toxicity.

Phase II

Key questions

• Is rofecoxib efficacious in (selected) patents?—does it demonstrate efficacy without significant adverse effects?
• What dose(s) should be used in the large studies designed to confirm the drug is efficacious enough to be marketed?
• What kind of study will be needed to confirm the drug is efficacious?—What sample size? What outcomes?

The question of which indications to pursue is addressed through out the drug development process. A large number of Phase II studies were conducted using rofecoxib for different indications, including: acute pain, osteoarthritis, rheumatoid arthritis, improvement of Alzheimer’s disease symptoms and the prevention of gastrointestinal polyps.

Phase II trials for rofecoxib

• Does the drug work in patients? How should the larger confirmatory trials be designed?

• Initial trials (Phase IIa) will contain 50–150 patients, are controlled (i.e. compared to active treatment or placebo), and conducted on a narrowly-defined population (such as “healthy patients”)

• Later trials in this phase (Phase IIb) are often larger (\(n \approx 500\)), include a broader range of patients, and look at a smaller range of doses

Key findings from Phase II (Scott and Lamb 1999)

• Study endpoints: patient and investigator global assessment of response to therapy, pain walking on a flat surface assessed according to Western Ontario and McMasters Universities Osteoarthritis Index (WOMAC).

• A multicenter dose ranging study in patients with osteoarthritis found: rofecoxib 5mg, 12.5mg, 25mg and 50mg were more effective than placebo.

• Rofecoxib demonstrated similar efficacy to diclofenac over 26 weeks, ibuprofen over 6 weeks, and nabumetone over 6 weeks (\(n = 784, 736, 341\) respectively)

• 12.5mg and 25mg well-tolerated; common adverse effects described

• Small studies suggest less endoscopically-observed ulcers with rofecoxib compared to older NSAIDs

Phase III

Key Questions

• Is the benefit:risk ratio sufficient to permit marketing the drug?

• Does rofecoxib work in a broader group of patients?

• Is rofecoxib well tolerated when given to a broader range of patients (i.e. different demographic groups and patients with different comorbidities)?

• Is rofecoxib as efficacious as alternatives and does it reduce the risk of adverse gastrointestinal events?

Phase III trials for rofecoxib

• A significant number of studies conducted, most comparing rofecoxib to an active control (different studies for different indications) with similar endpoints to the Phase II studies (there is significant overlap between the Phase IIb and Phase III studies)

• Studies are large (\(n \approx\) 1000–3000); often multicenter; include a patient population that simulates practice; and are conducted for longer

• Most trials tested 12.5mg/day and 25mg/day

• The primary purpose of some Phase III studies were to demonstrate reduced gastrointestinal risks

Findings from Phase III (Matheson and Figgitt 2001)

• Rofecoxib 12.5mg or 25mg/day as good as celecoxib 200mg/day, paracetamol 4g/day, ibuprofen 2400mg/day, diclofenac 150mg/day and naproxen 1000mg in patients with osteoarthritis

• General tolerability of rofecoxib was comparable to older NSAIDs

• Rofecoxib led to less endoscopically-observed ulcers compared to older NSAIDs

• Cardiovascular events were monitored due to the possible link between selective COX-2 inhibition and thrombotic events: no increase was observed

Rofecoxib was approved for marketing by the US Food and Drug Administration on May 1999.

Post-market problems

Five years following FDA approval, rofecoxib was withdrawn from the market after the APPROVe trial reported an increased risk of myocardial infarction in participants randomised to rofecoxib (Bresalier et al. 2005).

Figure 1: A selection of titles following withdrawal of rofecoxib

The APPROVe trial tested the hypothesis that rofecoxib would reduce the relapse risk of gastro-intestinal polyps in patients with a history of colorectal adenomas. Patients with a history of colorectal adenomas were randomized to receive rofecoxib 25mg daily or placebo for three years.

The relative risk for patients taking rofecoxib to suffer a confirmed thrombotic event during the study was 1.92 (95% CI 1.19–3.11; \(p\)=0.008). Absolute risk increase: 7 additional thrombotic events per 1000 patient years

It is possible to classify APPROVe in terms of the traditional phases of drug development in a number of ways. It was certainly post-marketing, so in that sense, it could be described as a Phase IV study. However, it was testing rofecoxib in a novel indication, and on this basis it would be classed Phase II or Phase III.