Bulletin of Experimental Treatments for AIDS, No. 3, June 1989

Oral versus Intravenous

Since the beginning of the development of dextran sulfate as a possible anti HIV drug, much confusion, skepticism and doubt has surrounded it. Part of this, should anyone need take blame, has been somewhat my fault because the original discussions I had with Ueno Fine Chemicals [a Japanese pharmaceutical firm] in 1986 indicated that they had a compound they thought worked against HIV. They wanted to bring it into clinical trials. The com- pound, they told me, was available in two forms in Japan: intravenous and oral.

At that time [July 1986] my decision was to develop and investigate oral dextran sulfate as opposed to intravenous dex- tran sulfate. It is not only easier for patients participating in a trial, but it is also less labor intensive and less expen- sive.

At that time, many people cautioned me that because dextran sulfate is such a large molecule, the body would be unable to absorb it through the gastrointestinal tract. If it were absorbed, it was felt a molecule of that size would not penetrate into a cell to do its work.

Measuring Dextran Sulfate in the Blood

As we got further into the planning, my Japanese collabora- tors brought me graphs demonstrating, in fact, that dextran sul- fate could be measured in the bloodstream a certain time after it is taken orally. It reaches a certain level, then disappears (as you would like to see with any orally absorbed drug). However, this assay, to detect the actual level of the dextran sulfate in the blood, was not reproducible in samples we sent from our patients. We cannot send HIV infected blood to Japan. It's part of their importation policies. Therefore, we had to heat inac- tivate blood specimens before we could send them. This process coagulated proteins, which made the assay they had developed for HIV negative blood not possible for the treated blood samples we sent them.

So, throughout the course of our phase I study here at San Francisco General, performed in 1987, there was no direct method of measuring the level of dextran sulfate in the bloodstream. There was no conclusive evidence that dextran sulfate taken by mouth got into the bloodstream. When I began to go to NIAID with this project, suggesting that we begin to investigate dextran sulfate on a larger scale, I was asked whether any of the drug was getting into the bloodstream. I replied that patients were experiencing side effects, they got "speedy," they had abnormali- ties of their liver tests, their white blood cell counts went down. It seemed to me that something was having an effect. In the test tube, the specific molecular weight of dextran sulfate and the amount of sulphur on the molecule is critical for whether or not it has antiviral activity. Skeptics cautioned that oral dextran sulfate could be broken down into smaller subfragments which are not then anti retroviral and that these subfragments could be causing some of the side effects and laboratory abnor- malities. In the absence of an assay, there was still no direct evidence that oral dextran sulfate was absorbed.

Effect on p24 Antigen and T helper Cell (T 4) Levels

The phase I study we did was performed to determine the max- imally tolerated dose and to learn more about dextran sulfate's side effects in people with HIV infection. In AIDS research, it is difficult to do a study which just looks at toxicity and side effects. There is a demand to evaluate potential agents quicker, and to also get some information on effectiveness simultaneously with phase I data. Therefore, in this first study, we also looked at T 4 cell counts and p24 antigen levels. We really did not see anything very positive. 4 cells did not change in a sus- tained fashion. HIV p24 antigen positivity was not a prere- quisite for entry into the study. Those patients who were posi- tive at study entry stayed positive and those who were negative stayed negative. Dextran sulfate, for an eight week course, did not appear to have any effect.

Finally, we did decide to do a larger trial of patients through the NIAID* AIDS Clinical Treatment Group mechanism. This second trial was to enroll patients who were HIV positive and p24 antigen positive so we could at least answer the question, "Is oral dextran sulfate effective in causing a decline in p24 antigen levels?" I also believe that one of the things we should now begin to question is whether p24 antigen is the only measure we have of a drug's effectiveness against the virus. AZT cer- tainly makes p24 levels decline. But can we expect a drug (such as dextran sulfate) that prevents virus attachment to cells and cell cell infection to really impact on p24 antigen levels?

The phase I study gave us only a small amount of information on people with AIDS taking dextran sulfate. Most people in this trial had lymphadenopathy*. In fact, the few AIDS patients who participated could not complete the trial because of adverse side effects. It was felt, like other treatments we evaluated, that the drug might be better tolerated in patients in earlier stages of infection and less well tolerated in PWA.

The current study being performed via the NIAID AIDS Clini- cal Trials Group program evaluated dextran sulfate in three patient groups: asymptomatics, ARC and AIDS. Ten patients in each group were enrolled at each dose. We started at 2700 mg and increased to 5400 mg; we were prepared to escalate to higher dosages if we hadn't reached the maximum tolerated dose by 5400 mg. If 5400 proved to be the maximum, we were planning to inves- tigate intermediate dosages (between 2700 mg and 5400 mg) as we had done in the initial phase I trial.

Johns Hopkins Experiment

In the middle of our trial, the pharmacology group at Johns Hopkins University reported on an experiment involving 6 healthy people. They gave them intravenous dextran sulfate followed by a single dose of oral dextran sulfate. They followed what I would have to call 'surrogate markers' for dextran sulfate activity. In Japan, the intravenous preparation of dextran sulfate is used as an anticoagulant, or blood thinner. The oral preparation is used to lower blood fats. This group looked at the activated partial thromboplastin time [PTT] which measures how thin the blood is getting, and found that after intravenous dextran sulfate, the PTT prolonged significantly. However, after a single oral dose of 1800 mg (which is what our patients on 5400 mg take three times/day) there was no prolongation, which suggested to them that there was no absorption.

I objected to that reasoning because a single oral dose may not be enough to create the effect. However, I was told a single IV dose does, so if you're checking to see how much of that oral dose becomes available in the bloodstream (the so called "bio availability"), if the intravenous dose produces such an effect, then the oral dose should as well. They also measured an enzyme called lipoprotein lipase [an enzyme that promotes breakdown of blood fats]. Patients receiving the IV compound had a marked increase in the activity of this enzyme; patients after the oral single dose also had an increase but it was not as marked. This again suggested to me that there was some absorption of the oral form, but that it was very low.

New Information on Absorption

At about the same time, Ueno in Japan seemed again to come up with what they felt was a perfected assay to measure actual dextran sulfate directly, not using these surrogate markers for its activity [i.e., the PTT and the lipoprotein lipase]. Ueno also concurred that the oral bioavailability of dextran sulfate is probably less than 1%. That means that less than 1% of what was given orally actually gets into the bloodstream. This made people question whether or not oral dextran sulfate could ever be effective against HIV.

The pharmacokinetic* data became available in January. At that time a conference call discussion among the participating centers and the central office staff at the NIAID concluded that we should explain this information to patients participating in our protocols. They could then make an informed decision as to whether they wanted to continue in the trial, or withdraw and try to participate in other studies. We held a meeting of patients participating in our dextran sulfate trial on January 23rd to tell them about this paper, which was being published on February 1st, and to inform them about the preliminary data from Johns Hopkins lab.

Confusion in the Press

Then the L.A. Times article appeared February 19th, alleging that the dextran sulfate investigators knew the drug was effec- tive , and were not sharing that information. We did not know the drug was not effective; we suspected it was not being absorbed. Our tack was now to collect all the available efficacy data and to run as many of the p24 antigens levels and T 4 counts as possible. We would analyze the data to see if dextran sulfate may still be effective even if it is not absorbed to any great extent. We did not want to throw the baby out with the bathwa- ter! That was our plan.

What particularly upset me about the L.A. Times article was that it constantly referred to the drug as "dextran." Dextran is a completely different compound than dextran sulfate. To me, that sort of inaccuracy and sloppiness served only to perpetuate the confusion.

Currently, 16 patients nationwide out of the 60 who began the study are continuing on the trial. We are hoping that by the end of April more patients will have completed the 6 months of treatment. Then we can begin to assess whether oral dextran sul- fate is worth pursuing or not.

In the meantime, we know many people continue to use dextran sulfate that they obtain in many creative fashions. I received a recent letter which highlights our dilemma. This is an anonymous letter, signed only "name withheld." To me, it highlights both the need for well controlled clinical trials and some of the problems we have as clinical investigators in attempting to per- form them.

Dear Dr. Abrams,

I buy my dextran sulfate from one of the participants in your study. He does not take the dextran, and his lab values show it. He stays in your study to get the dextran to sell and to have free lab tests regularly. On the other hand, I do take the dextran. I take 1500 mg in three doses of 600 mg, 600 mg and 300 mg at bedtime. I have been since April of '88. My T cells have gone from 289 to 884, gradually, as a result. I understand from my friend in your study that several participants are not taking the doses you think they are. I am writing because I know personally that dextran is a valuable antiviral medication. I am hoping your test results will in some way reflect this erroneous input. Would a lie detector test be helpful?
890601
BETA0303

Copyright © 1989 - Bulletin of Experimental Treatments for AIDS (BETA). Reproduced with permission. BETA is published four times a year by the San Francisco AIDS Foundation. All rights reserved. Noncommercial reproduction is encouraged. Subscription lists are kept confidential. Call 415.487.8060; FAX: 415.487.8069. Mailing Address: P.O. Box 426182, San Francisco, CA 94142-6182.  beta@sfaf.org  http://www.sfaf.org/beta.html