Subject: AIDS Treatment News #127 Date: May 18 1991 (990 lines) &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& J O H N J A M E S writes on A I D S &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& Copyright 1991 by John S. James; permission granted for non-commercial use. AIDS TREATMENT NEWS Issue #127, May 17, 1991 phone 800/TREAT-12, or 415/255-0588 CONTENTS: [items are separated by "*****" for this display] The Home-Run Strategy: How to Revolutionize Drug Development and Approval Tat-Inhibitor AIDS Trials Cancelled; Business Reasons Cited Announcements: BI-RG-587 Trial Begins Chronic Diarrhea: Sandostatin Trial Recruiting Advertising Competition: Ads Against AIDS San Francisco: Fundraiser for AIDS TREATMENT NEWS June 16 at Eureka Theater ***** The Home-Run Strategy: How to Revolutionize Drug Development and Approval by John S. James Today's advances in basic science and biotechnology offer new opportunities for major progress in treating AIDS and other diseases. But current systems for developing and approving new drugs reflect an earlier era; they cannot respond rapidly to today's developments. Even the current debate about drug- approval reform (focused on issues like parallel track vs. conditional approval, and on drugs like ddI and ddC) has largely overlooked some critical opportunities which scientific advances are now making possible. The "home-run strategy" outlined below would collect treatment efficacy information starting with the first human trial, for the most important potential therapies (single drug or combination) -- those emerging from preclinical testing with the best laboratory results and scientific rationale. It would look for improvement in patients in the treatment group, which can take weeks -- instead of looking for death or deterioration in the control group, which can take years. A truly major treatment advance (a "home-run" drug) could easily be spotted in a small trial, perhaps with only a handful of patients, if the trial is designed for that purpose. Using such trials, the home-run strategy "skims" the most promising drugs and combinations as they begin human testing, to identify one or more major advances for emergency development and access. This approach could be used immediately, at almost no financial cost. And it could focus debate on the most critical improvements needed now to save lives. Part I: Overview Improved laboratory discovery of AIDS drugs has created opportunities which never existed before for wholesale bypassing of almost all of the current obstacles to speedy development and delivery of the most important new treatments. Advances in basic science and in understanding the life cycle of HIV are creating an unending stream of rationally designed, scientifically well supported potential drugs, each of which has a meaningful chance of being a major improvement over all existing therapies. But most remain secret in corporate laboratories, untested and undeveloped because the great costs and delays of today's drug- development system create powerful disincentives against entering new drugs into human trials. A "home-run" strategy is one designed and optimized for quickly identifying any decisively important drug or combination therapy. A major, dramatic advance -- such as a treatment which quickly restored almost everyone with AIDS or HIV to full health (although it might need to be taken indefinitely and would not be a cure) could be identified far more easily than a small, incremental improvement, which the trials now being run were designed to find. Once a major treatment advance is discovered and confirmed, what would happen then? Existing drug-development and approval practices could lead to years of additional studies before the new treatment became widely available, even for a drug which clearly was immensely beneficial. During the last decade, a continuing procession of reform efforts (such as compassionate use, treatment IND, and parallel track) have tried to correct this problem, with limited success. In addition to such efforts at structural reform, the AIDS community should also consider what might be called "approval by acclamation" -- by the irresistible professional and public momentum which would develop when the case for a major treatment advance is unambiguous and overwhelming. Formal approval could still be fit into some legal or regulatory framework: "expanded access," "conditional approval," "compassionate use," "subpart E," or whatever other option might be ready at the time. Which particular technicality is used to make the drug available will not be ultimately important, because the real dynamic which drives development and determines access will be elsewhere. Again and again we have seen that professional and community consensus about what clearly should and must be done is more powerful than formal procedures alone. Financially, a home-run strategy will cost almost nothing, since the necessary trials would need very few patients each -- until after success has clearly been achieved. It will not compete for scarce resources against other kinds of research. To implement the home-run strategy, we need to focus on the one key obstacle, in the existing drug-development and approval system, which has kept it from happening already. This obstacle is the extremely long delay between the discovery of a potentially major treatment advance, and the obtaining of any information about how it works in practice when used to treat AIDS or symptomatic HIV infection. The key to implementing the home-run strategy is to reduce this delay to the minimum required for ethics and safety -- for example, the delay required by acute toxicity studies in animals if the potential drug is a new chemical never given to humans before. (There would, of course, already have been successful laboratory studies and a coherent scientific rationale for the drug; otherwise it would never have been selected as a home-run candidate in the first place.) The critical path for saving lives is to reduce the time between the emergence of a first-class medical and scientific rationale for a treatment, and the earliest flexible, practical test of that treatment under conditions of actual use. Then, if a drug or combination does prove to be a home run, it will be easy to mobilize on an emergency basis to do whatever is needed to complete the drug's development. If the drug is not clearly a home run, it can be left to other development tracks; the home- run strategy would then move on to select and test another candidate drug which could possibly be decisive. Today's debate on drug-development reform has focused on proposals (such as surrogate markers and conditional approval) which could be immensely beneficial. But the critical element still remains the delay between scientific rationale and first practical experience. We must not become so absorbed in the debate over general reforms that we lose sight of the critical path which will make the difference between life and death. How could a home-run strategy ever be accepted by a scientific and regulatory establishment which is accustomed to different ways of doing things? The first answer is that acceptance need not be all or nothing. By discussing, improving, and debating such approaches, the AIDS community can improve the existing system of drug development and approval, by focusing the spotlight of attention on precisely those delays which are most damaging and most in need of change. The second answer is that establishment acceptance is not strictly necessary. A home-run strategy would be best, of course, as an element of a national plan; for example, the opportunity for almost immediate approval of a drug which clearly worked far better than anything else available would bring forth the best of the compounds now languishing in corporate labs. But if there continues to be no national leadership or plan, and if in the absence of leadership the research and regulatory establishment is unwilling to change itself in time to save lives, then the AIDS community could implement much of the home- run strategy independently, with or without establishment cooperation. All that is needed is one credible, unambiguous, and easily repeatable demonstration of a truly dramatic treatment advance. The community's key advantage is that the better a drug works, the easier this is to do. Part II: Questions, Objections, and Explanations Question: How do you know that there are any home-run drugs for this system to find? Reply: No one can know until trials are done. But today there are many potential drugs which might be major advances -- either alone, or in combination treatments. Only a handful are reaching human trials, however, because of the cumbersome drug- development system which now prevails. Under this system, most of these possibilities will never be tested. Here are two indications which lead us to suspect that there are many home-run treatments which will be found -- sooner or later: (1) In an animal test, one experimental treatment (hypericin) did prove to be a home-run drug for treating an animal retrovirus: "In this mouse illness which is similar to AIDS, and which is widely used in screening new antiretroviral agents, mice live only 30 days after being infected. Appropriate hypericin blood levels restore the mice to a normal clinical state and to a normal life span of 240 days, with no apparent toxicity." (Bernard Bihari, M. D., "Ambulatory management of HIV related immune suppression and AIDS," unpublished manuscript.) Similar though less striking survival improvements in mice infected with retroviruses and treated with hypericin have been formally published; see D. Meruelo, G. Lavie, and D. Lavie, "Therapeutic agents with dramatic antiretroviral activity and little toxicity at effective doses: aromatic polycyclic diones hypericin and pseudohypericin," Proceedings of the National Academy of Sciences, USA, volume 85, pages 5230-5234, July 1988. Hypericin also shows strong activity against HIV in laboratory studies. In human blood freshly donated by HIV- positive patients -- the closest a laboratory study can come to a clinical trial -- hypericin does show antiviral activity in both lymphocytes and macrophages. It is believed to block the virus at two different points in its life cycle. But would it work as an AIDS/HIV treatment in people? Would it be safe? No one knows, because no one has ever tried a dose large enough to be effective (concentrated hypericin is not readily available). St. John's wort herbal preparations do contain hypericin, but too little to provide the blood levels which laboratory studies suggest are required. Hypericin has been known as a potential HIV treatment for about three years (although the mouse test described above was not done until more recently). The drug will soon begin human dose-escalation trials. If a home-run strategy had been in use three years ago, a small amount of hypericin would have been obtained for a human dose-escalation-plus-efficacy trial (after animal toxicity tests). If the drug worked as well in people as it did in the mice, the history of AIDS would have been different. We need a home-run strategy today, because without it, similar opportunities will continue to be lost. (2) In human trials of treatments for AIDS or HIV, there is no home run yet. But there is a "double" -- the unexpectedly good results of the combination of AZT and ddC, being tested in the U. S. -funded study ACTG 106. The study has not yet been formally published, but an early report was published by Project Inform, and then by AIDS TREATMENT NEWS (issue #115, November 23, 1990). What was impressive about the early data is that the median starting T-helper count was about 70, and these counts increased by a peak of about 160 (to well over 200) at the best doses. After the counts peaked, they gradually went down again; but a year later they were still higher than when they started. And at appropriate dose levels, no opportunistic infections occurred after the first few weeks (before the treatment had time to take effect) in this year-long trial. We have no access to the data, and are waiting for publication to obtain the most accurate picture. We have heard that the results still look as good today as they did last November. These results support our belief that there probably are home-run drugs because: * Before this study, it was widely believed that no antiviral alone could raise T-helper cells much if they were already well below 200. This trial shows that the immune system can recover, even from quite low T-helper counts, if the virus can be adequately stopped. * AZT plus ddC is probably NOT one of the better combinations. Each drug is mediocre by itself. And both are in the same class of drugs (nucleoside analogs). The best drug combinations, however, are likely to be those which work differently, which block different steps in viral reproduction. The unexpectedly good results of AZT plus ddC suggests the power of combination antiviral therapy -- especially since this combination is far from ideal, and much better ones are likely to be found. If home-run therapies do exist (which, for the reasons outlined above, seems likely), then it is important to discover them as quickly as possible. Current drug-development procedures cannot do this. We need a strategy which will. Then the chances will be excellent that one or more major treatment advances can be identified quickly. Question: Explain how this strategy would work. How would drug development differ from what prevails today? Reply: The home-run strategy would focus on immediately conducting very small trials of drugs which might be major advances for a certain group of patients. (An example of a treatment which might be tested this way would be one of the new non-nucleoside-analog RT inhibitors, such as BI-RG 587, in combination with AZT.) By "home run," we mean a drug or combination which, if it works, would quickly reverse almost all clinical and laboratory signs of illness for almost all patients, at least within a selected group of patients which can be defined in advance. If any drug works this well, it will be easy to determine that it has important activity, through a tiny, and very rapid trial -- one which is easy to design, finance, recruit for, conduct, and analyze. Before a new drug is given to humans for the first time, the laboratory and animal work would be completed, of course, as it is today. [The preclinical development system, such as animal toxicity testing, also needs major reforms, but that is a different issue to be discussed separately; here we are considering only what happens after a drug is ready for its first human test.] In the system prevailing today, the new drug's sponsor first goes to the FDA (or to another country's regulatory authority) for permission to conduct a single-dose pharmacokinetic test in a small number of HIV-negative persons (for example, students in Europe) to see how well the drug is absorbed, how long it stays in the bloodstream, etc. If this test is successful, then the sponsor designs a dose-ranging trial, and goes back to the FDA for approval (often by new reviewers, because of low pay and resulting turnover at the agency) for this trial. After approval (usually with some changes) the sponsor recruits for the dose- ranging study, conducts the study, analyzes the results, develops a new protocol, and then goes back to the FDA for approval (likely by new reviewers again) for a phase II/III study which, under the "fast track" system of developing critical drugs, might lead to approval. Under the system in place today, these final trials usually take years to organize, recruit for, conduct, and analyze. The biggest cause of this delay is the time needed to accumulate enough deaths or clear disease progressions in the control group -- those not receiving the drug of interest -- to achieve statistical significance. The second biggest delay is the need to recruit, and administer, huge numbers of eligible volunteers in order to reduce the first delay. Not only are results unavailable when patients and physicians need them, but also the great human and financial resource requirements of large trials assure that few new drugs can be tested. Hopefully the FDA will accept T-helper count as a "surrogate marker" to substitute for death or major disease progression as trial endpoints. Then the years required for this final trial before drug approval might be reduced to months. But the other delays will still be in place. We are proposing a much faster system. For potential home- run drugs or treatment combinations, the three steps in the process -- pharmacokinetics, dose-ranging, and efficacy studies -- will still be done, but all at the same time, with the same patients, and with one visit instead of three to the regulatory agency to approve the trial. And because it is so much easier to detect and prove efficacy of a home-run drug than of one which is equivalent to or only slightly better than existing therapy, the entire course of human trials of the drug, from pharmacokinetics to dose ranging to definitive proof of efficacy (to customary levels of statistical significance at least, and usually much better) could be reduced, in the best possible case, to a few weeks or less. Long-term safety and efficacy data will, of course, still be needed before a drug is approved for widespread use; the point is that the delays in the process which are not medically or scientifically necessary can be eliminated. The very drugs which are most important, because they have a major, dramatic impact, are for the same reason the ones which can be tested most rapidly. A typical trial would proceed as follows. First, there must be reason to believe that a particular drug or combination might make a dramatic difference for a certain group of patients. The necessary laboratory and animal studies must be completed so that the drug is ready for human testing. The procedure so far is the same as is done today. But the first human test would be in patients judged likely to benefit dramatically from the new drug, if it works as hoped. HIV-negative volunteers or healthy seropositives would be avoided, because they do not have symptoms which could show if there is clinical benefit from the drug. The first doses administered would be studied for pharmacokinetics; if the drug failed here, it would be discarded, and the trial would be over. Otherwise, the volunteers would continue into a dose escalation phase, with careful safety monitoring as in any phase I trial. But efficacy would be monitored just as carefully, and if the drug worked as hoped -- meaning that it led to almost total remission of symptoms and normalization of laboratory values in almost all patients -- then the drug would be known to be at least a short-term success, and the trial would continue (with individualized dose adjustments as necessary to obtain optimum benefit) in order to begin collecting long-term information. On the other hand, if dose-limiting toxicity were found before efficacy, then the trial would be ended as having failed to show benefit. Note that this approach to trial design is fully consistent with obtaining long-term safety and efficacy information. In fact it optimizes such data collection, because the first patients treated can go all the way through, from pharmacokinetics to dose ranging to efficacy to long-term studies (in contrast to traditional drug development, which starts over with new patients for each new stage of trial). Drugs could be approved before the long-term studies end; as in current practice, post-marketing studies may continue after approval. Objection: How can you claim that a trial could obtain statistical proof of efficacy of a home-run drug with very few patients -- perhaps as few as only two or three -- within weeks? Even the smallest trial is likely to need 20 patients or more -- and some need hundreds to get a statistically significant result, and can last for years. And without statistical significance, you cannot know whether the results you see are due to coincidence. Reply: Large trials are necessary to detect small differences between treatments. The home-run strategy is not looking for anything except very large treatment effects. The clinical-trial designs which have prevailed so far in AIDS are inherited from those which were successful in developing cancer chemotherapy. These trials are suited for finding small incremental differences between treatment regimens, such as between different combinations of chemotherapy agents. Also, they are optimized for testing highly toxic drugs, which are expected to be used near the maximum tolerated dose. But these designs are poorly matched to the needs of AIDS drug development today. There are much faster and less expensive ways to scan potential treatments in order to find any major advances among them. Obtaining statistical significance means finding a result which would have been unlikely by chance alone (if the treatment had no effect). Conventional trials show significance by having many more deaths (or other endpoints) in the control group than in the treatment group; a major problem, of course, is that it takes a long time to obtain this proof. A much faster way to credibly show that a treatment is effective is to look for benefit. For example, if a certain remission (or other specified benefit) is very unlikely for certain patients without treatment, or with only standard or conventional treatments, and yet a great majority of those patients experience this benefit when given a new treatment, then it can be possible to obtain credible, and statistically significant, evidence of benefit, with only a few people in a treatment group. The odds against most or all of them benefitting by chance could be astronomical. Objection: How can you draw any conclusion from such a trial, without a control group? Reply: Well-known methods exist for using each patient as his or her own control, and also for historical controls, comparing those who receive the experimental treatment with past experience of similar patients without it. It is not necessary for every study to have a simultaneous control arm receiving a different treatment. The randomized clinical trial is admittedly the scientific "gold standard" for clinical trials -- the kind of design that, at least in theory, can produce the most certain results eventually. But it it not the only scientifically acceptable design. Nor is it always the best public policy, when the time required to get a useful, practical answer matters greatly. In a well known article in the November 8, 1990, New England Journal of Medicine, experts in AIDS clinical trials considered uncontrolled efficacy trials: "There are special situations in which the use of an uncontrolled phase III clinical trial to evaluate the efficacy of a new drug may be justified. We believe that such special situations must meet all the following requirements: (1) there must be no other treatment appropriate to use as a control; (2) there must be sufficient experience to ensure that the patients not receiving therapy will have a uniformly poor prognosis; (3) the therapy must not be expected to have substantial side effects that would compromise the potential benefit to the patient; (4) there must be a justifiable expectation that the potential benefit to the patient will be sufficiently large to make interpretation of the results of a nonrandomized trial unambiguous; and (5) the scientific rationale for the treatment must be sufficiently strong that a positive result would be widely accepted. (D. P. Byar, D. A. Schoenfeld, S. B. Green and others, "Design considerations for AIDS trials," New England Journal of Medicine, November 8, 1990, volume 323, number 19, pages 1343-1348.) The home-run trials we are proposing are not phase III, but they can meet all five of the requirements. Note that trials which use death or disease progression as primary endpoints are inherently slow, because they must wait for statistically significant numbers of these endpoints in the control group (which, in trials of an AIDS antiviral, is usually receiving AZT), no matter how well the new drug being tested may work. Even if the new drug were perfect and cured everyone immediately, it could still take a year, two years, or more to accumulate enough endpoints in the control group to obtain statistical significance. Also note that in most of the discussion about using surrogate markers (especially T-helper count) instead of death or disease progression to measure drug effects, it is assumed that using the markers will lead to faster trials because changes in markers occur faster than death or other traditional "major" endpoints. This assumption is true, of course, but it overlooks what may be an even greater contribution of the use of surrogate markers -- allowing the trial to look for improvement in the treated group, instead of only looking for deterioration in patients not receiving the treatment of interest. The alternative design we outlined above does rely on physicians' judgment that, for certain patients, a rapid, complete remission is improbable. Physicians do make such judgments, and often they are correct (although we all know of highly publicized cases where they have been wrong). One of the goals of science is to eliminate subjectivity as an element in experiments, so that results can be publicly verifiable and reproducible. But because our task now is not elegant science but rather a practical response to a public-health emergency, a potential improvement should not be rejected merely because it violates a general principle or a rule. Rather, the skeptic must show some plausible way that breaking the rule could, in the real world, lead to a wrong result -- to accepting a drug as a "home run" when in fact it is not, or vice versa. Since the trials we propose could easily be designed to produce statistical significance levels many times better than those accepted in common practice in clinical trials, the chance of error seems small, unless some systematic bias is at work. Question: Would "home-run" trials rely on surrogate markers, such as T-helper counts? Reply: A truly home-run drug would be expected to cause T- helper and other laboratory tests to return to normal or near- normal values. It would also be expected to relieve clinical symptoms, such as chronic infections, night sweats, fatigue, weight loss, etc. Both kinds of improvements would be required. Objection: Your trial design is assuming that the drug will work for everyone. But even a "home-run" drug could not promise this. Reply: For the sake of discussion, we provided simple examples in which the drug would be rejected if it failed for any patient in the trial. In practice, statisticians could work out a more reasonable arrangement. For example, if we set the standard that the "home-run" drug must produce a specified remission in, say, 90 percent of a defined group of patients, and if the remission were extremely unlikely without the new treatment, then statisticians could compute a sample size to obtain significance while also minimizing the probability of falsely rejecting a good drug. Instead of two or three, perhaps four or five patients would be required for the trial. The point is that this kind of trial can be approached with standard statistical methodology, despite the radically smaller number of patients involved. The difference is that this trial is designed for efficiently finding home-run drugs, whereas most of the trials used in AIDS have been designed for finding small improvements. Those trials were appropriate in the past, when few if any home-run candidates were available. Different kinds of trials are appropriate today. Objection: What if a drug fails the home-run test? It might then be discarded, even if it still has value. Won't pharmaceutical companies be afraid to participate in such trials, for fear that a good drug which is less than a home run will be prematurely killed? Reply: Even drugs which fail as home runs will still benefit by being tested. The practical efficacy information, whatever it is, will tell researchers where they stand with the drug, and help them design more targeted trials for it in the future. If the early trial suggests that the drug is harmful or ineffective, that is something the sponsors should want to know, to avoid wasting money on a hopeless project. Will sponsors cancel drugs just because they are believed not to be home runs? If so, if they must remain ignorant in order to safeguard their enthusiasm, then they are not practicing either good science or good business. Decisions should be based on all relevant information, without key data needing to be overlooked to maintain commitment. Objection: You propose the first human test of a new drug in persons with symptomatic HIV infection. This could be dangerous, because early signs of toxicity might be masked by HIV symptoms. Reply: Running these tests on persons who are HIV- negative, or HIV-positive but asymptomatic, does make it easier to see early signs of toxicity when it develops. But today this advantage is being reduced, because it is now possible to develop many drugs which are likely to have a large therapeutic range between effective and toxic doses. This raises the possibility of trials which will never reach dose-limiting toxicity, because they will reach effective therapeutic doses first, and never need to go higher. If physicians must know what toxicity would occur from too high a dose, in order to know what side effects to watch for when the treatment is used, then tests which escalate doses until toxicity is found should be conducted after the new drug is already known to work. Why subject volunteers to the risk of deliberately inducing toxicity to test a drug which may never be used? And such human toxicity testing, if needed at all, should not delay treatments which thousands of people urgently require. ***** Tat Inhibitor Trials Cancelled; Business Reasons Cited by John S. James In a serious setback to AIDS treatment development, Hoffmann-La Roche Inc. indefinitely postponed trials of its tat inhibitor (code named RO 24-7429), the only such drug in development. Researchers at Johns Hopkins University in Baltimore, where the trial was about to begin, were described as shocked by the news, which they received by phone on April 29. Hoffmann-La Roche has issued no written statement about the decision, which has received almost no media coverage. The Johns Hopkins team, probably the world's experts on the drug since they conducted preclinical studies and a small single-dose human trial there, are reluctant to speak publicly but have insisted that there was no medical or scientific reason to stop the trial. Paul Oestreicher, Ph.D., spokesperson for Hoffmann-La Roche, also acknowledged that there was no scientific reason to stop the trial. He said the company decided that, after a review of its entire antiviral program, it will pursue the development of ddC and a protease inhibitor while it "aggressively" seeks a partner (another pharmaceutical company) to continue tat antagonist development. Dr. Oestreicher said that Hoffmann-La Roche is open to a wide range of potential business arrangements (to continue development of the potential tat drug), including co-development or licensing. Importance of the Drug The Hoffmann-La Roche tat inhibitor is the only drug developed with its mode of action -- interfering with the protein produced by the tat gene of HIV, which is essential for viral replication. Since this drug could potentially offer a new kind of therapeutic alternative, it may increase the treatment options available, and also provide an excellent candidate for use in combination therapy. The drug is also important because researchers believe that a successful tat inhibitor would have specific action against KS (Kaposi's sarcoma), besides its anti-HIV effect. In addition, Johns Hopkins researchers had previously discovered that the tat protein causes immune suppression in laboratory cell cultures, similar to the immune suppression seen in AIDS; they published this finding in Science, December 22, 1989. An effective tat inhibitor would also have a unique business advantage -- one reason researchers thought Hoffmann-La Roche was especially interested in this drug. If such a drug is effective against KS, then it could bypass the unworkable HIV-drug pipeline full of other antivirals waiting to be compared to AZT. It could be approved as a KS treatment without such comparison, and then physicians could also use it "off label" for treating HIV in patients without KS. Such a drug might become a standard AIDS therapy before its rivals get approved, giving it a momentum hard for other drugs to overcome. (There are efforts to clear the drugjam, through proposals such as surrogate markers, or conditional approval. But the acceptance of these reforms is not assured.) Background "Tat" (an abbreviation for "transactivator") is a gene in HIV. This gene produces a protein which turns on other HIV genes, greatly increasing the activity of the virus. Without tat, HIV is largely or totally inactive. The Hoffmann-La Roche drug is believed to stop this protein from working. Some researchers have questioned whether the Hoffmann-La Roche drug is really a tat inhibitor. It is a benzodiazepine derivative, chemically similar to other potential AIDS drugs (such as Merck's L-661, or Boehringer Ingelheim's BI-RG-587) which inhibit reverse transcriptase, not tat; therefore, it is possible that this drug could be effective against HIV by other than the tat mechanism. If so, it might still be an important AIDS drug, but without the unique advantages of a tat inhibitor. Some people suspect that Hoffmann-La Roche may have stopped the trial because of doubts about whether the drug is in fact a tat inhibitor. (We consider it doubtful that Hoffmann-La Roche would have made a mistake about the mechanism of action, since methods for testing tat activity have been published, and only persons working with the company, not the skeptics on this issue, have full access to the laboratory test results.) The tat protein is also believed to have a major role in the development of KS; it could not be the sole cause, however, as KS can develop without HIV. A human trial of the drug as a KS treatment had been discussed for next autumn in Los Angeles. The reason the KS trial had been scheduled to start later than the HIV trial was to allow an animal test first -- an attempt to treat human KS in mice. But the animal study could not be run because the mice had not developed KS yet. Only the laboratory of Robert Gallo, M. D., at the U. S. National Cancer Institute, has been known to have the KS mouse model, which it first reported some years ago. Apparently Hoffmann-La Roche decided to redevelop this model elsewhere. Comment For months there has been concern among some AIDS treatment activists that Hoffmann-La Roche might pull out of its new- generation AIDS drug development if ddC proved unexpectedly difficult or expensive. The company now has over 3,000 people on its expanded-access ddC program -- which is costly because of the labor and record-keeping involved, even though the drug itself costs almost nothing. (In addition, about two thousand people may be using "underground" ddC obtained from chemical-industry sources, where ddC has been available for years.) The drug is not especially attractive as a single therapy, but it is an important option for thousands of persons who cannot successfully use AZT. The expanded-access program originally required paperwork burdensome enough that many physicians could not enroll their patients, because they did not have the time required to fill out the forms. The paperwork burden led to a boycott against Hoffmann-La Roche, first approved by ACT UP/New York on February 11, 1991. The boycott had largely been settled before the tat drug trials were stopped, as the paperwork problems had eased; and Hoffmann-La Roche claims that it had no effect on the company's business. But such boycotts, if supported by the medical community, can be a serious threat to pharmaceutical companies, because hospital, laboratory, and other medical personnel can divert millions of dollars of business to competitors, in cases where equivalent products are available. ACT UP/New York had distributed detailed boycott packets, and claimed that hundreds of physicians were refusing to prescribe Hoffmann-La Roche pharmaceuticals or laboratory services. (One unpublished advertisement, professionally prepared for medical journals and sent to the company by ACT UP/New York, shows a morgue with two sheet-covered bodies, under the headline, "Thanks to Roche, a lot of people with AIDS don't have it anymore.") We cannot read the minds of Hoffmann-La Roche executives, but it is hard to see how stopping the Johns Hopkins trial now would make business sense if the company wanted to find a partner to develop the product. The drug, if it works, would have much greater value after the study than before, because this trial would have produced the first human efficacy information ever for this drug. (We asked Dr. Oestreicher about this, and he replied that the trial was stopped now to allow the partner a role in designing new studies.) Another reason in favor of continuing was the ethical issue of stopping a trial for business reasons after the sponsor had already given the drug for the first time to volunteers, who had therefore undergone the risk of taking a new chemical never given to humans before. (It is our understanding that no efficacy information was collected from this earlier single-dose study.) Yet stopping the trial now would make business sense if the purpose was to avoid being pressured into another expanded-access program, with its resulting expense and possible threat to non- AIDS-related products. Without a trial, the data to support the public demand for such a program will never come into existence. Unfortunately, if this scenario is correct, then the prognosis for continuing the trial by finding a future partner is not good -- unless responsibility could be transferred to an entity willing to finance such a program, or to one clearly unable to do so. Since such business negotiations are almost always secret, the public will not know what is happening with a drug on which, in the future, thousands of lives may depend. This case illustrates the long-range problem of disincentives to pharmaceutical companies built into the current system of clinical trials. Once companies begin trials, they may incur millions of dollars in costs, which they might never recover. The tat inhibitor is unusual in that development was stopped so late, for a unique and potentially valuable drug. Other drugs which would have been developed, if the disincentives were not so severe, never get out of the laboratory, and we do not hear about them. At this time there seems to be little chance to convince Hoffmann-La Roche to continue the trial, not even to fill the gap until an acceptable partner is found. It might be possible for the AIDS community to help in locating a potential partner. Meanwhile, what AIDS TREATMENT NEWS can do is to make available all information we can find about this drug. Persons associated with the project have become even more secretive since the trial was cancelled than they had been before. We would like to hear from anyone with information about this drug, the research so far, the cancellation of the trials, or medical or scientific reaction to the cancellation. Anyone willing to talk on or off the record, or who knows someone who might, could call John S. James at 415/255-0588, or write to AIDS TREATMENT NEWS, Attn.: tat, P. O. Box 411256, San Francisco, CA 94141. Acknowledgement: Garey Lambert, reporter for the Baltimore Alternative, provided major assistance with this article. ***** Announcements ** BI-RG-587 Trial Begins On April 19 AIDS TREATMENT NEWS reported that the planned dose-ranging trial of BI-RG-587 had not yet begun, and that the drug had been taken by no more than a dozen people, none of whom received more than a single dose. The new trial (ACTG 0164) opened later in April, but at this time only 30 volunteers are being enrolled, at two sites: San Diego, California, and Worcester, Massachusetts. The San Diego site already has enough volunteers to fill its immediate openings; we do not know about Worcester. This trial requires weekly visits to the study site, so it is not feasible for persons living in other areas. The first three doses tested will be 12.5, 50, and 250 mg per day. BI-RG-587 is taken orally, once per day. Doses will escalate until there is either toxicity, or good efficacy. Volunteers for this trial must have T-helper counts under 400 and meet other medical criteria. For more information about possible enrollment, call 800/TRIALS-A. If this small trial proves successful, then a larger trial (ACTG 0168) could begin soon, possibly as early as July. Plans call for ACTG 0168 to include a very important test of BI-RG-587 in combination with AZT. In addition, a pediatric trial (ACTG 0165) of BI-RG-587 could also begin in the next couple months. ** Chronic Diarrhea: Sandostatin Trial Recruiting Sandostatin (also called octreotide acetate), a drug which provides symptomatic relief from certain kinds of diarrhea, is being tested at trial sites in at least 15 locations across the U. S. This trial is sponsored by Sandoz Pharmaceuticals Corporation. Sandostatin is already available as a prescription drug, approved by the FDA for treating diarrhea caused by certain cancers. This trial was designed to test whether the drug is also effective for AIDS-related diarrhea which cannot be treated by other means. Volunteers may receive a placebo, but only during the first four weeks. After that time, plans are to treat everyone with the drug, for a second phase of the study lasting at least six months. Sandostatin, a peptide with eight amino acids, mimics the action of somatostatin, a hormone which occurs naturally in the body. Sandostatin lasts much longer in the blood, however; it has a half-life of two to three hours, compared to one to three minutes for somatostatin. The drug is given by subcutaneous injection. Comment The first choice for diarrhea therapy is, of course, to find and treat the cause -- often an infection by protozoa, bacteria, or viruses. But if the cause cannot be diagnosed and treated successfully, it is still important to control the diarrhea, to prevent malnutrition and to improve quality of life. Sandostatin is available by prescription, but the drug is expensive. Also, physicians may be reluctant to prescribe it, since at this time there is no proof that it works for AIDS- related diarrhea. In the trial, the drug is free. For More Information The Sandostatin trial is recruiting in at least 15 locations in the U. S.: Boston, Chicago, Detroit, Galveston, Houston, Philadelphia, Portland, New York City, Long Island, Miami, Madison, Los Angeles, Washington, D. C., San Francisco, and San Diego; other sites may be added. For information about trial locations and enrollment, call the Sandoz hotline, 800/732-8096, Monday through Friday, 9 a.m. to 5 p.m. Eastern time. ** Advertising Competition: Ads Against AIDS Advertising professionals and others are invited to submit their ideas to a nationwide competition organized by Ads Against AIDS, which will then fully produce the winners and air them nationally and locally as public-service TV and radio spots. The goal is to raise peoples' awareness that AIDS is their concern, and to give them an option for a positive response. Awards will be presented at a benefit dinner in New York on November 6, co-produced by People Taking Action Against AIDS (PTAAA), a fundraising organization which since 1987 has raised over $800,000 for AIDS care and research. Advertising leaders supporting this effort include Phil Dusenberry, chairman-CEO of BBDO, New York; Jack Mariucci, executive VP-executive creative director of DDB Needham, New York; and Helayne Spivack, executive VP-executive creative director of Young and Rubicam, New York. Mr. Mariucci is co- chairman of Ads Against AIDS. The winning public-service announcements will also be made available to grassroots organizations throughout the country, for placement in their local media. TV spots will include a five- second tag for the local group's identification. For more information, contact Ads Against AIDS, 31 West 26th Street, Ground Floor, New York, NY 10010, phone 212/481-1374, fax 212/689-5291. ** San Francisco: Fundraiser for AIDS TREATMENT NEWS June 16 at Eureka Theater by Tim Wilson AIDS TREATMENT NEWS and Project Inform will be the beneficiaries of a special performance of Tony Kushner's Angels in America: Part One, a community fundraising event hosted by San Francisco's Eureka Theatre, 2730 16th Street (at Harrison), on Sunday, June 16, at 6:30 p.m. One hundred tickets at $35 are now available through ATN Publications on a first come basis; an additional 100 tickets will be available through Project Inform. Angels in America is a world-premiere commission by the Eureka Theatre comprised of two complete plays directed by David Esbjornson: a fully conceived production of Millennium Approaches and a minimalistic presentation of the just completed Perestroika. The June 16 benefit will feature a reception with playwright Tony Kushner at 6:30 p.m., followed by a special performance of Millennium Approaches at 7:30 p.m. The benefit date was chosen to coincide with the opening of the VII International Conference on AIDS in Florence. Angels in America: A Gay Fantasia on National Themes is a compelling yet funny epic set during the eight years of the Reagan administration. Millennium Approaches (Part One) begins the stories of two couples, one gay and one Mormon, as they engage in parallel struggles to reconcile love and responsibility. As the new millennium draws closer, relationships and old orders begin to crumble, the hole in the ozone widens, AIDS knocks at the door of the Justice Department, and supernatural intervention becomes unavoidable. Perestroika (Part Two, a work in progress -- not part of the benefit performance) moves from the death bed of notorious right wing lawyer Roy Cohn to the Soviet Union, and from Salt Lake City to Heaven itself. Having broken old bonds, the characters of Millennium seek hope, as well as spiritual and political transformation, within a cataclysmic world. The largest project ever mounted by the Eureka Theatre, Angels in America was originally commissioned by the Eureka in 1987. The project has been in development ever since. During a 1989 Mark Taper Forum workshop production, Millennium Approaches was hailed as "an epic theatrical fever dream" (Variety) which "transcends conventional sexual, ethnic, literary and political boundaries" (Los Angeles Times). Millennium has received a major grant from the Fund for New American Plays, a project of the John F. Kennedy Center for the Performing Arts with support from American Express Company in cooperation with the President's Committee on the Arts and the Humanities. Interested persons can send a check for tickets to the special June 16 benefit performance ($35 each) directly to ATN Publications, Attn.: Eureka Benefit, P. O. Box 411256, San Francisco, CA 94141. Tickets and/or a confirmation letter will be sent by return mail. After Friday, May 31, call 415/255-0588 to make sure tickets are still available. This special performance has been supported by American Express Company. ***** Statement of Purpose AIDS TREATMENT NEWS reports on experimental and complementary treatments, especially those available now. It collects information from medical journals, and from interviews with scientists, physicians, and other health practitioners, and persons with AIDS or HIV. Long-term survivors have usually tried many different treatments, and found combinations which work for them. AIDS TREATMENT NEWS does not recommend particular therapies, but seeks to increase the options available. We also examine the ethical and public-policy issues around AIDS treatment research and treatment access. [Obsolete subscription information has been removed. See the latest issues for up-to-date information. -- sysop] &&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& End of display