The curious stories of drugs with two lives: a new paradigm in drug development

Conventional drug development resulted in the synthesis of drug molecules with specific targets and responses in mind. However, in recent years there has been a rise in the number of drugs that have been redeveloped to be used in a range of disease states far removed from their original indications. Initially, these discoveries were mainly as a result of serendipitous case reports or occasionally by astute scientific observation. Drug developers are now actively investigating hundreds of ‘old’ drugs to ascertain if they have another life ahead of them. This is being done using complex scientific and computer-aided technologies, not relying on luck and uncertainty. The advantages and cost benefits that are attached to such processes may be significant.

Most drugs in use today have been developed as exogenous ligands that bind to specific receptor sites. The resulting effect either through agonism or antagonism generally results in a narrow set of responses. Therefore, most drugs tend to have a fairly selective set of actions following administration. This selectivity is, of course, exactly what Erlich was alluding to almost a century ago when he referred to the need to develop therapeutic agents with a specific target in mind – the so-called ‘magic bullets’.

It came as somewhat of a sensation therefore, to find some years ago, that thalidomide, a mild sedative and anti-emetic that had been withdrawn from use many years earlier for its now well-known teratogenic effects was showing promise in the treatment of multiple myeloma. ¹

Since then, a wide variety of agents, many of which have been in clinical use for prolonged periods, have been redeveloped to treat an entirely new and unsuspecting host of disease states. In many cases, this represents treatment of pathological states so far removed from their original use as to defy belief that somebody could have even suspected that they might have any effects.

The discovery that drugs can have uses removed from their original indications is not entirely new, with clonidine and amantadine being such examples from the 1960s. Clonidine was initially developed as a nasal decongestant by the pharmaceutical manufacturer Boehringer Ingelheim. Following its administration to a department secretary, it caused her to ‘fall asleep for 24 hours, develop low blood pressure, bradycardia, and a dry mouth’. ² As a result, focus changed, and it has since been used initially as an anti-hypertensive and anti-migraine product and more latterly in anaesthesia. In the case of amantadine, the discovery that this drug initially developed as an anti-viral agent possessed mild activity in the management of Parkinson’s disease was likewise a serendipitous discovery that occurred when a patient with advanced Parkinson’s disease noted a transient improvement in her symptoms after a few weeks of taking the drug for its intended use. ³ Although rarely used, nowadays it provided a useful adjunct in the management of the disease during the years when therapeutic options were limited.

The list of drugs that have been given a ‘second life’ has now grown significantly over the last 10 years, and now numbers are approaching a 100. ⁴ A selection of some of the more noteworthy examples is given below.

Drugs with new lives

Thalidomide: first life: sedative and anti-emetic; Current use: treatment of multiple myeloma

Few drugs have been associated with such a range of serious adverse effects, by way of severe teratogenesis than thalidomide, first marketed as an agent for pregnancy-induced vomiting and as a sedative. Its re-development 20 years later is perhaps the prime example of how drugs have been given second lives. ¹

In many cases, it is difficult to determine from the literature just who had the idea to use a drug for an alternative disease and where the idea arose. In the case of thalidomide, the picture is clearer than in most. The first hint that thalidomide might have other activity was when an Israeli dermatologist noted that patients with dermatological manifestations of leprosy often demonstrated a dramatic response when treated with the drug. Fast forward 20 years when a series of important studies demonstrated that thalidomide had significant inhibitory effect on the expression of tumour necrosis factor alpha. Patients with HIV/AIDS were known to have elevated levels of this substance, and its second ‘off label’ use was in the management of mouth ulcers in such patients. It was successful when most other treatments had failed. ⁵

Lastly by way of extrapolation, another group in the United States began using thalidomide in the management of multiple myeloma, a disease which at the time usually resulted in death within a time following diagnosis. There was success – and new hope. The company which held the patent for thalidomide went on to develop an analogue, lenalidamide, a derivative with improved anti-tumour activity and a more moderate adverse effect profile.

Non-steroidal anti-inflammatory drugs (NSAIDs): first life: analgesics; now: possible anti-neoplastic agents

As one reviewer has noted, ⁶ there are now compelling epidemiological and clinical studies suggesting that NSAIDs have anti-neoplastic properties, and that population-based studies suggest that the incidence of colorectal cancers is reduced by as much as 50% in patients who are also chronic users of NSAIDs. What is perhaps more interesting is that recent research has demonstrated that this unexpected effect is not related to the well-known cyclooxygenase inhibitory effects of this drug class. Studies have suggested that a COX (Cyclo-oxygenase)-independent or off-target effect such as phosphodiesterase inhibition and cGMP elevation could be the mechanism for the anti-neoplastic effect. This would enable the synthesis of NSAIDs like drugs that have minimal COX-inhibitory effects, thus eliminating the raft of adverse effects usually associated with their use. ⁷

Metformin: first life: anti-diabetic agent; now: anti-neoplastic agent

The initial possibility that metformin might have a role in treating cancer was first noted when examining survival rates in patients with type 2 diabetes who were taking metformin.^8,9 These retrospective analyses suggested that even at therapeutic doses, metformin administration had a positive benefit in suppressing neoplastic activity.

The complex and multiple process by which metformin might produce these responses has recently been extensively reviewed, ⁹ and although the precise mechanism remains unclear, there is a theory that suggests that increased insulin levels are associated with enhanced tumour growth and progression, resulting in enhanced expression of insulin receptors in some tumour cells.

Thioridazine: first life: anti-psychotic; now: possible anti-malarial

The history of the phenothiazine derivative thioridazine in its first life is almost as interesting as its second! Its precursor was promethazine, a sedating anti-histamine still in use today.

Phenothiazines, in addition to their psychotomimetic actions, have significant anti-microbial activity, a feature identified as early as 1891 when Erlich demonstrated such activity in another member of that group, methylene blue. ¹⁰

Thioridazine in particular has been shown to possess significant activity against Methicillin resistant Staphloccus aureus, Enetrococcus and of even greater importance Mycobacterium tuberculosis and Plasmonium falciparum. The current challenge now is to develop an analogue of thioridazine with enhanced anti-microbial characteristics, but with minimal dopaminergic activity, so reducing troublesome adverse effects.

Anti-epileptic agents: first life management of epilepsy; now: management of bipolar disorder and neuropathic pain

Perhaps the jump from epilepsy to management of psychiatric disorders is not that great. As noted by Bailer, ¹¹ bipolar disorder and epilepsy have a number of features in common – notably their episodic nature and kindling phenomena (phenomena where repeated withdrawals lead to increasingly severe symptoms).

With respect to neuropathic pain, it has been suggested that it bears some similarities with epilepsy, as both are associated with spontaneous generation of electrical impulses and an increase in sodium channel excitability. Alterations in the expression of sodium channel expression can thus be related to both epileptic hyperexcitability and that seen in peripheral nerves in neuropathic pain. While carbamazepine and phenytoin share this mechanism, their use in neuropathic pain has been superseded by anti-epileptic agents with calcium channel blocking properties, notably gabapentin and pregabalin.

Beta adrenergic blocking drugs: first life anti-hypertensives; now: anti-neoplastic agents

One of the more recent developments in the field of drug repositioning is some interesting studies suggesting that non-selective beta adrenergic blocking drugs such as propranolol can slow or halt the progression of certain tumours, notably breast cancers, in mice. This has been supported from epidemiological work with cancer recurrence rates in patients taking beta blockers. This would be in line with some theories that the growth and spread of adrenergic receptors found in some tumours are enhanced by sympathetic activation. ¹² Not all studies have shown convincing results, but this bears watching.

Success not always assured

Although there are many cases where an unexpected outcome has shown promise, such is not always the case. There were early reports from animal studies, for example that ceftriaxone might offer some neuroprotective effects in the management of that most untreatable of afflictions, motor neurone disease. Unfortunately, a large-scale human study was commenced in 2010 but ceased in 2012 by the investigators who determined that the treatment was unlikely to be of any clinical benefit, and further studies were abandoned. ¹³

Clear advantages in the process

Closely examining ‘older’ drugs for a second life has some real promise, and it has been estimated that there are in the order of 10,000 such drugs available for re-evaluation or ‘repositioning’ as the new term goes. ¹⁴

For example, Chong and Sullivan ¹⁵ have pointed out there is strong logic behind this approach. For one thing, it has been suggested that it takes an average of 15 years and over $US500 million to bring a single drug onto the market. While re-launching drugs into a second life will still necessitate considerable costs in Phase II–IV trials, at least some time and money can be potentially saved by not having to repeat Phase I studies.

There are other clear advantages in this process. The first is that most of these agents are no longer covered by patent making them freely available for testing and clinical use. Second, many of these agents have been used extensively in their first life and have therefore undergone significant safety assessment, been used in a range of patient groups and have undergone extensive post-marketing surveillance. Lastly, the fact that there is often already a significant body of data covering safety and pharmacokinetics means that investigators can accelerate research quite quickly in Phase II studies.

In many of the cases referred to above, the link between original and new use seems a huge jump. In some cases, there was a thread that might connect the two uses.

There have been a number of recent papers in which the authors have pointed out that such new drug development need not be a ‘hit and miss’ affair, but rather there should be a concerted effort to re-examine older drugs to ascertain whether they might indeed have a second life.

In fact, many pharmaceutical companies are already following this lead. It has been suggested that from 2007 to 2009, about 40% of drugs or related products that were approved or launched for the first time in the United States were either drugs repositioned for new indications, reformulations or new combinations of existing drugs. ¹⁶ While this can be seen in one respect as useful reuse of current medications, others have suggested that this practice of so-called ‘evergreening’ simply represents a method that pharmaceutical companies can use to prolong the period of patent protection. ¹⁷

More recent papers dealing with drug repositioning have set out to investigate a different question, namely – how can companies best identify potential targets for re-positioning? Clearly, although serendipitous observation have proved useful in past examples, passively waiting for such events to occur is not only likely to be a very low yield process but also immediately rules out discovering possible repositioning prospects for drugs that are not in current use. Lastly, another group of patients likely to benefit from this new approach will be those with relatively rare or neglected disease. For such people, mainstream drug development is unlikely ever to be pursued because of the failure of any such discoveries to return any significant financial reward to the developers. ¹⁵

How will new targets be identified?

A detailed discussion of the proposed methods for detecting possible drugs suitable for repositioning is beyond the scope of this article, but one review has suggested it could be approached in a number of ways. ¹⁸ First, although serendipitous observations may on first inspection appear to be of little benefit, if there were to be set up some sort of centralised agency that could collate such data, it may well yield results. Patients, of course, would need to be encouraged to report such findings. Other approaches are set out to use complex mathematical computational models, where drug attributes and possible receptor affinities can be matched with target proteins or receptors known to exist in a particular disease state.^18,19 Such models have made detailed studies of a drugs molecular structure and side chains and then matched them with drugs with similar structural properties that are known to be effective in the target illness. ¹⁹

Lastly, there are already in existence vast data banks such as OpenPhacts, DrugBank and the delightfully named ‘Promiscuous’ that provide a wealth of pharmacological and physicochemical data, as well as information on drugs’ protein and active-site structures and associations with related disease states, ²⁰ that lie ready to be mined. These existence of data banks, together with the fact that a new journal entirely devoted to the subject of drug repositioning is to be launched shortly, ²¹ suggest that drug repositioning may indeed herald a new dawn in the complex world of drug discovery and development.