To the Editor:  Evidence of the therapeutic value and limited toxicity of marijuana in the treatment of various symptoms and syndromes is rapidly accumulating.  Patients with glaucoma, migraine, epilepsy, multiple sclerosis paraplegia, quadriplegia, the AIDS wasting syndrome, or nausea and vomiting due to chemotherapy for cancer have risked severe penalties to acquire cannabis, because they find it more useful than legally available medicines.1,2  The U.S. government has consistently refused to permit prescription sales of marijuana, insisting that the evidence of its safety and efficacy is inadequate.
        In response to a lawsuit brought by advocates of the medical use of marijuana, an administrative law judge with the Drug Enforcement Administration (DEA) recommended in 1988 that the federal legal classification of marijuana be changed so that physicians can prescribe it.
3  The DEA did not implement the recommendation.  In 1992, Richard Bonner, then DEA administrator, offered this suggestion: "Those who insist that marihuana has medical uses would serve society better by promoting or sponsoring more legitimate scientific research, rather than throwing their time, money, and rhetoric into lobbying, public relations campaigns, and perennial litigation."4
        Encouraged by this declaration, Donald Abrams, of the University of California, San Francisco, sought permission to conduct a privately financed pilot study comparing three potencies of inhaled marijuana (high, medium, and low) with oral delta-9-tetrahydrocannabinol (dronabinol) capsules, which are available by prescription in the United States, for the treatment of weight loss associated with the AIDS wasting syndrome.  Dr. Abrams's protocol (Investigational New Drug No. 43,542) was designed in consultation with the Food and Drug Administration (FDA) and approved by the FDA, the University of California, San Francisco, institutional review board, the California Research Advisory Panel, and the scientific advisory committee of the San Francisco Community Consortium.
        Unfortunately, the U.S. government will not allow Dr. Abrams to obtain a legal supply of marijuana.  The DEA refused to permit him to import marijuana from Hortapharm, a company licensed by the government of the Netherlands to cultivate cannabis for botanical and pharmaceutical research.  The National Institute on Drug Abuse (NIDA), which controls the domestic supply of marijuana for clinical research, rejected Dr. Abrams's request in April 1995.  The letter of rejection, from NIDA director Alan I. Leshner, was sent nine months after Dr. Abrams had submitted his request.
        Given the limited sources of support for research on marijuana (pharmaceutical companies, for example, are not interested in supporting such research), the U.S. government by now should have taken note of the increasingly persuasive therapeutic claims and sponsored its own clinical trials through the National Institutes of Health.  Instead, as Dr. Bonner's remark and the handling of Dr. Abrams's protocol indicate, the government has responded by urging others to investigate the medical potential of marijuana and then creating obstacles that make the research impossible to pursue.

Boston, MA 02115

Harvard Medical School

Cambridge, MA 02138

John F. Kennedy
School of Government 

  1. Grinspoon L, Bakalar JB.  Marihuana, the forbidden medicine.  New Haven: Yale University Press, 1993.
  2. Idem.  Marihuana as medicine: a plea for reconsideration.  JAMA 1995;273:1875-6.
  3. Young FL.  Opinion and recommended ruling, marijuana rescheduling petition.  Department of Justice, Drug Enforcement Administration.  Docket 86-22.  Washington, D.C.: Drug Enforcement Administration, September 1988.
  4. Marijuana scheduling petition: denial of petition: remand.  Fed Regist 1992;57:10503.

        Editor's note: Drs. Leshner and Abrams were offered the opportunity to respond.  Dr. Leshner declined to do so.  Dr. Abrams's reply follows.

        To the Editor: The San Francisco Community Consortium is an association of providers of care for patients infected with the human immunodeficiency virus (HIV) in the San Francisco Bay area.  The consortium conducts community-based clinical trials of therapies of potential benefit to our patients.  Inhaled marijuana is being used increasingly by people with HIV infection, especially for its purported benefit as an anti-emetic agent and an appetite stimulant in those with the AIDS wasting syndrome.  Up to 2000 people infected with HIV are reported to be obtaining marijuana at a cannabis buyer's club in our area.  The California State Assembly has recently endorsed legislation to make medicinal marijuana available to patients with life-threatening illnesses.
        In an effort to determine whether inhaled marijuana is truly of any potential benefit and, more important, to evaluate its safety in people with AIDS, the Community Consortium designed a pilot study.  We thought it would be appropriate to conduct a small study of the overall feasibility of investigating inhaled marijuana use by such patients, before embarking on a full-scale trial of its efficacy.  The pilot-drug-evaluation staff at the FDA provided valuable comments on the design of the protocol.  Pulmonary-function tests, lymphocyte-sub-group analyses, and measurements of viral burden by the branched-DNA technique were to be performed serially for safety.  Body composition was to be measured by state-of-the-art techniques to assess potential trends in the amelioration of wasting.
        The FDA and the institutional review board supported the study.  Unfortunately, the DEA and the NIDA opposed it.  Most disturbing was the absence of a response from either agency for an unacceptably long period, followed by the NIDA's outright rejection of the proposal without any opportunity for dialogue or compromise.  Such behavior is offensive not only to the investigators but to the patients for whom we seek to find safe and effective treatments.

San Francisco, CA 94143

University of California
San Francisco 

New England Journal of Medicine
Sept. 7, 1995
Vol. 333  No. 10  pp. 670-671