Five recommendations
to proactively expedite medical marijuana research.

[Submitted in a Report to NIDA's Expert Panel on Medical Marijuana Research submitted February
25, 1997 following the February 19-20, 1997 NIH Workshop on the Medical Utility of Marijuana
held at the NIH Campus in Bethesda, MD.]

1. Make available a list of the inventory of NIDA marijuana currently in stock.

NIDA currently has a monopoly on the supply of marijuana available to researchers who have
obtained FDA approval for their proposed protocols. In light of this, research protocols could be
designed to use marijuana which is currently in stock rather than being blindly designed for
marijuana that may be need to be planted and grown before the studies can begin. This simple
request is my litmus test to determine whether NIDA is serious about proactively supporting medical
marijuana research or whether expressions of support for the need for research are purely cosmetic.
An October 30, 1996 Los Angeles Times article by Eric Bailey reported that NIDA had about
one ton of marijuana already in stock. Is this true? What are the amounts and potencies of NIDA
marijuana currently in stock? How much of this marijuana is needed for purposes other than medical
marijuana research?

2. Estimate the amount of NIDA-supplied marijuana which would be needed to supply all medical marijuana research projects likely to be approved by the FDA over the next three years.

Developing such an estimate should guide NIDA in determining how much additional marijuana,
if any, needs to be grown for medical marijuana research. This estimate should be linked to a Clinical
Plan for the study of the medical uses of marijuana. I developed such a Clinical Plan at the request of
the FDA and with its consultation. The final draft of this Clinical Plan(see enclosure)was submitted
to the FDA in March 1994 as an adjunct to Dr. Donald Abrams's IND (#43,542) to study the use of
marijuana in the treatment of AIDS wasting.
By my estimate, NIDA would need to supply roughly 25 kilograms of marijuana per year for
each of the following three years to all the studies investigating the medical use of marijuana.

3. Provide estimates of the cost of NIDA-supplied marijuana.

The cost to NIDA of its marijuana has been cited as a main justification for the need for extensive
NIH peer-review of all new medical marijuana protocols. Just how expensive is NIDA marijuana?
Dr. ElSohly, director of NIDA's marijuana farm at the University of Mississippi, estimates that the
production cost is $1,120 per kilogram. This is a relatively minor cost compared to NIDA's

estimated $487 million budget. Dr. Leshner says that Dr. ElSohly's estimates are low and that the
entire marijuana production program costs about $600,000 per year in total.
The Panel should obtain and provide to the public the marginal costof supplying 25 kilograms
per year for the next three years to studies investigating the medical use of marijuana in several
patient populations. The true cost of supplying marijuana for a new use is the additional cost over
and above the costs already assumed by NIDA as part of its current and regular production program.
The Panel should also obtain and provide to the public NIDA's average production costs.

4. Propose an NIH- endorsed peer-review procedure that considers the unique nature of marijuana research and takes less time than the standard NIH review process.

The NIH peer-review process is very lengthy. There are only three application deadlines per year
and it takes about eight months or so from the application deadline to the availability of funding.
NIDA's requirement that it will only provide marijuana to NIH-approved protocols, regardless of
funding needs, has so far prevented the initiation of FDA-approved medical marijuana research and
will definitely slow the pace of any research that does manage to secure NIH approval.
The Panel should propose other peer review processes that can respond to state and privately
funded projects, that will act more rapidly than the NIH process, and that will still satisfy NIDA's
concerns that the protocols be well designed.The specification of alternative routes to NIDA
marijuana is the most difficult litmus test to determine if the Panel, and NIDA, are really serious
about expediting medical marijuana research.

5. Propose that NIDA fund studies into the effectiveness of vaporizers--as distinct from aerosol delivery systems--in filtering marijuana smoke, or formally encourage Research Triangle Institute (RTI) to accept private contracts for such research.

Marijuana smoke contains compounds that are neither therapeutic nor without health risks. From
1994-1995, research into the effectiveness of water pipes and vaporizers in filtering marijuana smoke
was conducted at the American Health Foundation (AHF), a leading center for research into the
constituents of tobacco and marijuana smoke. The study was funded by MAPS and California
NORML and used marijuana donated by NIDA outside of any NIH peer-review process. This
research determined that water pipes were counterproductive and selectively filtered out more
cannabinoids than particulate matter but that vaporizers did improve the ratio of cannabinoids to
particulate matter.
Vaporizers are existing technology that can be rapidly tested and improved. Their political
downside is that they are designed to improve the safety profile of smoked marijuana. Aerosol
technology is also important to support but will take more time and cost more money to develop.
Its political advantage, and medical advantage as well, is that it can utilize only the therapeutic

cannabinoids and need not involve the delivery of any extraneous substances.
Unfortunately, AHF has decided not to continue with the water pipe/vaporizer research project,
citing the controversial nature of researching marijuana smoke. MAPS has tried to continue this
research, but a research proposal that I submitted to RTI has not been accepted. I think that a
recommendation to RTI from the Panel that encourages it to conduct research specifically on
vaporizer technology would be a sign to RTI that such research is acceptable to NIDA.