Drug Truths

A site devoted to teaching about drug discovery and development.

Will “Reusable Drugs” Truly Be a Worthwhile Exercise?

with 12 comments

Researchers at the Stanford University School of Medicine recently announced that they have developed a computer program designed to find new uses for already approved drugs.  As reported by Amy Dockser Marcus in The Wall Street Journal, researchers Atul Butte, Joel Dudley and Marina Sirota believe that their technology allows them to screen rapidly genomic databases in such a way that they can identify examples where a drug creates a change in gene activity opposite to the gene activity caused by a disease.  Such an observation allows researchers to identify uses for drugs that they were not initially designed for.
At the outset, let me say that I firmly support all efforts to find new uses for drugs.  And hopefully, this new approach will yield a breakthrough.  But I am not convinced that this will be a bountiful source of new products.

It is interesting that whenever researchers look to repurpose old drugs for new uses, they always use sildenafil (tradename, Viagra) as their poster child, as it was serendipitously discovered as an agent to treat erectile dysfunction (ED), even though that wasn’t the specific use it was designed for.  However, this discovery was not so accidental.  Sildenafil was designed as a potent inhibitor of an enzyme known as PDE-5.  The interest in PDE-5 inhibitors stemmed from the fact that inhibition of this enzyme should result in elevation of nitrous oxide (NO) in vascular tissue beds.  NO is well known to be a vasodilator.  Pfizer scientists hoped that by blocking PDE-5 in the heart vasculature, arteries would dilate and the result would be enhanced blood flow in patients with cardiac disease like congestive heart failure.

Sildenafil did, in fact, cause vasodilation.  However, this vasodilation was first observed in the penis and not the heart.  Instead of being a breakthough medication for heart disease, sildenafil became a major treatment for ED.  So, yes, this was a biological consequence that was not initially envisioned.  The key in all of this is that Pfizer scientists designed and synthesized a safe and effective PDE-5 inhibitor that could be tested in clinical trials to determine what the utility of such an agent would be.  Sildenafil was, in fact, designed as a vasodilator.  Its effects, however, were manifest in an organ other than the heart.

When a new mechanism is found to be effective in patients, scientists often explore where else such an agent may be of use.  Pfizer researchers were also interested in learning whether sildenafil would cause vasodilation in other parts of the body.  One theory was that the small arteries in the lung might also be sensitive to sildenafil’s effects.  Patients with primary pulmonary hypertension (PHT) suffer from arterial constriction which is extremely debilitating, and people with this disorder have trouble breathing.  Clinical trials showed that sildenafil was very effective in treating PHT, and it is marketed for this condition as Revatio.

Another example is Pfizer’s tofacitinib, an inhibitor of the enzyme JAK-3.  This orally effective drug was initially designed to be used as an agent to prevent organ transplant rejection.  However, when the impressive early clinical data first came in, researchers began to envision other uses for a drug that acted by this mechanism, including rheumatoid arthritis and psoriasis.  Tofacitinib is now in late stage clinical trials for these and other indications.

These two examples illustrate two important points.  The first is that, while serendipity is always appreciated in any research program, for any pharmaceutical research program to be successful, you need to have a safe compound which targets a specific biological process.  Once in the clinic, you may find that the mechanism for which the drug was originally designed does not prove to be the optimum use for the new drug (there is also a downside to this – sometimes the new mechanism may have a mechanistically related side-effect that turns out to kill the drug).  But you don’t go blindly into clinical trials with the hope that a PDE-5 inhibitor might do something beneficial in people.  Rather, you must connect the mechanism to a biological effect.

The second point is that when a mechanistically exciting drug shows beneficial effects in a disease, the news spreads rapidly throughout a research organization.  Scientists will share these results and then hypothesize where else such a compound may be effective.  This leads to many other experiments to explore the new exciting finding and potentially, new uses for this drug in medicine.

The Stanford scientists have raised a lot of hopes that their new approach will uncover new uses for existing medications.  Hopefully, they will have success. But casting a broad net with the hope of finding something new will be very challenging.  As the understanding of the causes of diseases grows, and should a compound that would be expected to interact in this specific pathway already exist, chances are that the company that developed the drug is already onto this new potential use.


Written by johnlamattina

August 31, 2011 at 11:50 am

12 Responses

Subscribe to comments with RSS.

  1. Perhaps thalidomide is a better example of a repurposed drug, especially since we don’t know how it worked for the original indication (morning sickness; some literature claims its effectiveness there was ill-proven) and it’s pretty murky how it works in the new indications (hematologic cancers, leprosy).

    There are other candidates less developed, such as the work of Stegmaier and Golub and colleagues at the Broad Institute, which have used the same approach to identify possible anti-cancer compounds that were not obvious.

    Keith Robison

    August 31, 2011 at 3:52 pm

    • Keith,
      I would agree that thalidomide is one viable example. However, there aren’t many over decades of work. Pfizer had started an “Indications Discovery Unit” in their St. Louis labs in 2007 with the expressed purpose of doing this. It is my understanding that they abandoned it recently. As I said, repurposing is a nice idea – I just don’t think it will provided a bounty of new drugs. I hope that I’m wrong!
      – John


      August 31, 2011 at 5:37 pm

  2. Great points raised in your article… and you are right that expertise within a pharmaceutical company can be used to envision additional uses for compounds in development. However, not every pharmaceutical company has experts in every possible therapeutic area (TA), nor do I think all pharmaceutical companies work that synergistically across TAs. Repositioning efforts aren’t going to replace conventional drug discovery, maybe just enhance certain aspects of it.

    I do want to point out that these two compounds for which we found new uses are now off patent… once off patent, I think the incentive essentially collapses for the original patent holders to find additional uses. If these “out of the box” aren’t specifically searched for, I don’t think they will be found, whether using our methods or those of Kim Stegmaier, Justin Lamb, Todd Golub, Pankaj Agrawal, Monica Campillos, Peer Bork, and several others.

    Best regards, — Atul Butte

    Atul Butte

    September 2, 2011 at 4:59 am

    • Atul,
      I think your point about off-patent drugs is a good one. The downside is that unless patents can be filed for new uses, it might be difficult to generate the necessary funds for development. Good luck in this endeavor. – John


      September 2, 2011 at 10:54 am

  3. Even new use patents won’t be very valuable; both NitroMed and CombinatoRX went down in flames trying that strategy. If these drugs are to be developed for these indications, it will have to be public funding.

    Keith Robison

    September 2, 2011 at 1:03 pm

    • Keith,
      Thnat’s a valid concern. – John


      September 2, 2011 at 4:44 pm

  4. I find it interesting that the approach to pay out; of mapping drug change in gene activity vs. gene activity caused by a disease, is shadowed by another mapping: follow the science vs. “follow the money”. Some of the above comments acknowledge the low percentage of unanticipated “hits” from the past. However is it not clear if the activity mapping is more a matter of getting the information into an algorithm (relatively cheap) than creating new data (expensive.) Given the complexity of disease mechanisms, there are a lot of permutations that could be tested.

    But – if the data is already in hand from the normal drug development process, by leveraging existing disease models those “hits” may pay out for in-development drugs as well as those that are in-market. Finding novel uses could extend the exclusivity life of a drug that is already throwing off enough cash to fund clinical trial proofs (unlike those off-patent.) “Following those monies” is easy if the science proves out. Finally, if the algorithmic tests are cheap enough, there are many, many earlier pipeline molecules that could be evaluated for these “bonuses”, guiding portfolio decisions.

    Similar mappings via novel information algorithms have unearthed risk factors from past clinical results, and related them to chemosimilar “domains” of candidates. That helped ID potential sources of “bad actor” sidechains or structures that would otherwise have taken years and many millions of clinical dollars to unearth. Using information innovatively in ways like these is the one main hope for R&D to improve its return on investment – whether by society or by Big Pharma.

    Terry McCormick

    September 4, 2011 at 1:13 pm

  5. Plenty of good points here. But there are lots of new indications for old drugs (I know over 1600 such associations) that are not being developed because of commercial barriers. The Stanford researchers are looking to increase the size of this haystack. Moreover, some of their ‘new’ findings are actually re-discovered examples that have been in the literature for a long time. Cimetidine, which they posit for lung cancer, was first suggested to be useful in cancer since 1979! My preference would be for careful analysis of the existing data and optimisation of the active moiety or formulation for the new indication. This is a modification of the drug repurposing strategy, using clinical efficacy pointers to de-risk new product development.

    …and I agree that this strategy is not a cure-all. But can be useful and highly productive.

    • David,
      Thanks for your comments. I was not aware of the cimetidine potential for treating cancer. But it raises a lot of questions for me. Presumably, GSK (or SmithKline) scientists knew of this activity. Surely, they would have pursued it with cimetidine. Presumably, they could have gotten a use patent on this and then, if successful, been able to obtain premium pricing for such an important indication. If they didn’t do this, it undoubtedly wasn’t due to laziness or lack of interest. They must have had a strong rationale not to conduct such an R&D program.
      I support this general concept. I just believe that it is receiving way too much publicity that isn’t proportional to the lielihood of success.
      – John


      May 15, 2012 at 5:01 pm

  6. […] repurposing,” “drug repositioning,” “reusable drugs” – have been previously discussed in my blog. I do hope that people can have success in this type of work. But I believe successes are going to […]

  7. […] more on the trickiness of drug repurposing, see John LaMattina here and here. And the points he raises get to the “as long as it doesn’t cost too much” line […]

    Roche Repurposes | Healthcare

    November 30, 2012 at 4:27 am

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: