The Warfarin Pioneers

The discovery of warfarin was the result of a sequence of coincidences, which were probably regarded as anything but fortuitous at the time. The story begins in the US during the Great Depression in 1920s, where previously healthy cattle began dying as a result of internal gastrointestinal bleeding. This was a time of great financial

hardship, and livestock was one of the US' major industries, so identifying the cause of the condition became a priority. The cause was later attributed to spoiled sweet clover (Melilotus alba and M. officinalis) and the condition was referred to as 'sweet clover disease'¹. The financial pressures of the

time probably contributed to the problem, as cattle feed may well have been kept for longer than normal making spoilage more likely. Even the weather played a part - a series of wet summers giving rise to an epidemic of sweet clover disease during this period².

Despite the general interest, the discovery of warfarin was due to the efforts of a single farmer. In desperation, he decided to transport his dead cow 190 miles to the nearest state veterinarian, but since the office was closed, he ended up at the doorstep of the Biochemistry Building at the University of Wisconsin and the laboratory of one Professor Paul Link, whose pioneering work resulted in the identification and isolation of dicoumarol, a precursor to the first anticoagulants - figure 1. ►

The research was funded by the Wisconsin Alumni Research Fund (W.A.R.F.) after which warfarin was named. Initially, however, no medicinal purposes for warfarin had been identified and in 1948, warfarin was marketed as a rodenticide, replacing other more toxic chemicals³. Human safety was demonstrated shortly thereafter by the intentional ingestion of

warfarin by a despondent soldier in the early 1950s⁴. Consequently, without the benefit of studies that generally form the backbone of developing indications for treatment and monitoring the efficacy and safety of medicines, warfarin was introduced as a treatment following myocardial

infarction^5;6 - one of the more prominent early recipients of warfarin was President Eisenhower in 1955.

The earliest clinical trials of warfarin were performed in the late 1950s to evaluate warfarin administration following myocardial infarction⁷. Present day indications for warfarin include as a long-term oral anticoagulant for the prevention and treatment of venous thromboembolism, and for the prevention of cardiac and systemic emboli in patients with atrial fibrillation or prosthetic heart valves. These days, warfarin is used only very rarely following myocardial infarction^8;9.

Standardising Warfarin Treatment

One of the biggest problems limiting the early use of warfarin was not the production or availability of the drug, but the availability of a standardized

measure of its anticoagulant effect. The prothrombin time was an established measure, but the test was, and is, notoriously variable due to the differences between batches of tissue factor (thromboplastin) used as the reagent in this test. For this reason, the World Health Organisation (WHO) and the International Committee on Thrombosis and Hemostasis established the concept of the International Normalised Ratio (INR) which was devised to standardize reporting of results of blood coagulation as follows:

►

The prothrombin time (PT_test) is compared with a normal prothrombin time (PT_normal). This value is then raised to the power of the International Sensitivity Index (ISI) value. The ISI value (which usually varies between 1.0 and 2.0) is assigned to each batch of tissue factor produced by the manufacturer to indicate how a particular batch of tissue factor compares to an

internationally standardized sample¹⁰. Essentially, this means that there may be over 100% variation between batches of the manufactured tissue factor used as a reagent for the test, in addition to other sources of inaccuracy in relation to either the test or control sample (e.g.

sample collection, storage, machinery or testing procedure, medical error etc.)¹⁰, as for any laboratory test. So despite efforts to standardize coagulation monitoring, the INR still remains one of the important limitations of warfarin administration to this day, making it difficult to compare results between anticoagulant centres from the same or different regions, or interpret test results performed within the same patient at different times. Even if the INRs were considered reliable, a number of limitations and controversies complicate decision-making regarding warfarin administration and management, such as:

• How best to interpret consecutive measurements? Various approaches have been suggested including time in therapeutic range (TITR) and percentage time in range (PTIR) - figure 2 (next page)¹¹.
• What constitutes being outside of the therapeutic range? Is it the range quoted by the guidelines? Or the range with an additional buffer (0.2 units above and below the indicated range) to allow for the variation in testing?
• What to do with people that lie just above or below the therapeutic range? Warfarin dose holding and boosting might make doctors feel better, but this response may

• also disguise trends that should be corrected by changes to the regular dose, and might in fact lead to more time out of range¹².
• What is the optimal frequency for INR testing? Some guidelines state that stable patients should be monitored with a single INR test once every 4 weeks¹³. Recent evidence suggests that intervals up to 8 weeks might also be acceptable¹⁴. In the UK, however, 12 weeks is common practice, although anyone attending hospital might have their INR checked every few days, regardless of whether their medication is stable or unstable. ►

Warfarin Anticoagulation Monitoring

There has been a rapid rise in the number of anticoagulation centres and services. However, evidence from randomised controlled trials to support anticoagulation services is lacking. Many questions remain regarding the optimal set up in terms of staffing, follow-up, choice of treatment algorithm,

availability of trained staff, and quality of service.

Point-of-care self-testing and self-management systems have also been tested as methods for improving TITR; however, only self-management (which is not suitable for all patients)

consistently improved this measure of warfarin treatment. Walraven et al¹⁵ reported that TITR may vary from: (1) 50% in standard community follow-up; to (2) 66% with the utilization of anticoagulation clinics; and (3) up to 72% with additional patient self-management. Despite these benefits, the magnitude of the improvements in TITR required to have any impact on clinical endpoints is large, and the relative improvements in safety and efficacy are small; which implies that the overall clinical improvements by increasing the TITR alone is rather limited^12;16. In addition, anticoagulation self-management may be unsuitable for a large proportion of patients. ►

In addition to these major issues related to monitoring, warfarin has a narrow therapeutic window, and an unpredictable pharmacology - figure 3 ¹⁷. Warfarin has a long half-life and is metabolized by the cytochrome P450 enzyme. The wide genetic variation in this enzyme is reflected by the need for therapeutic maintenance doses that vary from 0.5 to over

20 mg per day among patients. In addition, a number of other drugs and dietary factors can influence warfarin metabolism via the cytochrome p450 enzyme, either inducing or inhibiting warfarin metabolism. Obviously, these factors have a great deal of impact on loading and maintenance dosing, necessitating frequent monitoring and dose adjustment, despite the previously mentioned clinical problems associated with this methodology. ►

New Age of Rational Drug Design

In contrast to the coincidental discovery of warfarin, the identification and development of the newer oral anticoagulants could not be more different. Rivaroxaban has been developed through a process of rational drug design: a systematic identification of potential drug candidates

through target molecule identification and substrate adaptations - figure 4. The candidates have subsequently undergone a rigorous selection process through a pre-clinical trial program to identify the most effective and safe candidates for human trials.

Rivaroxaban has been available since 2008 for the prophylaxis of venous thromboembolism following elective orthopaedic

surgery, and was recently approved for the treatment of acute deep venous thrombosis and the prevention of stroke in

patients with atrial fibrillatio^18;19. In a salute to warfarin's very first therapeutic indication, rivaroxaban is also undergoing trials for use in acute coronary syndrome²⁰.

With simple and convenient dosing regimens, rapid onset of action, predictable pharmacology, promising benefit:risk profile compared to traditional vitamin K antagonists such as warfarin, and no need for anticoagulation monitoring, rivaroxaban is set to replace warfarin over the broadest range of indications amongst any of the emerging oral anticoagulants.