Tuberculosis: New Drug Class Investigated through Public-Private Partnership

A June 21 press release from the U.S. National Institute of Allergy and Infectious Diseases (NIAID) described a collaboration between government and industry which produced an experimental drug candidate which may be the first new kind of tuberculosis drug in more than 30 years. The new compound, still in laboratory testing, was active against all MDR (multi-drug resistant) TB strains tested(1).

The compound, called PA-824, was developed by PathoGenesis Corporation in Seattle; NIAID studied how it worked.

To develop the drug, PathoGenesis scientists searched the literature for compounds with anti-TB activity, and found one which had been abandoned as a medicine because it also caused potentially harmful mutations. But they made over 300 chemical variants of that molecule, and found some that had even better anti-TB activity and did not cause the mutations. The one they selected is PA-824.


Tuberculosis kills more than two million people a year, but little has been done to find new drugs because most people with the disease could not pay high prices. It is not clear that PA-824 will be affordable, because it was developed under the "CRADA" system (Cooperative Research and Development Agreement), by which the U.S. government and private companies can share resources -- but the government does not provide funds or control prices. Instead, "CRADAs provide one way for companies to pursue promising but 'low profit' drugs -- such as those for TB, AIDS and malaria -- without spending huge amounts of money starting up new laboratories" (quoted from the NIAID press release).

But this example does show that critically important drug development can be started at relatively low cost (most of the cost in developing new drugs is in the large-scale clinical trials later in the process). There is plenty of money (from many potential sources, including some in developing countries) to pay for the initial investigation and discovery, as in the project reported here. Qualified scientists and laboratories can be found. Synthesizing and testing hundreds of variations of a chemical is not always rocket science, but sometimes a boring job which may be given to a chemistry graduate student at a university. U.S. government laboratories could participate through a CRADA (which does not require low prices for an eventual product, but does not require high prices either).

What this project would need at the start is money, and the commitment and sophistication to do the job right (without the funder committed in advance to some particular "cure," for example). The goal would be a convincing proof of principle -- a candidate drug which clearly works in animals and has no known major drawbacks. Then the project would need to do the organizing, business, and politics necessary to get the potential drug into larger, more expensive human studies -- or better yet, to set up credible but less expensive trial networks, not necessarily geared toward U.S. FDA approval, to avoid the huge inefficiencies in clinical trials in this country.


1. Stover CK, Warrener P, VanDevanter DR and others. A nitroimidazopyran drug candidate for the treatment of active and latent tuberculosis. Nature 2000; volume 405, pages 926-966.

ISSN # 1052-4207

Copyright 2000 by John S. James. Permission granted for noncommercial reproduction, provided that our address and phone number are included if more than short quotations are used.

Back to the AIDS Treatment News July 7, 2000 contents page.