Pharmac. & Ther. Vol. 55: 201-277, 1992


Department of Molecular and Cell Biology, 229 Stanley Hall, University of California at Berkeley, Berkeley, CA 94720, U.S.A.

4. The Drug-AIDS Hypothesis

After the global acceptance of the virus-AIDS hypothesis, several investigators have recently revived the original hypothesis that AIDS is not infectious (Section 2.2). In view of (1) the almost complete restriction (97%) of American AIDS to groups with severely compromised health, (2) the predetermination for certain AIDS diseases by prior health risks, and (3) the many links between AIDS and drug consumption (Sections 2.1.3 and 3.4, Table 2), it has been proposed that recreational drugs and AZT may cause AIDS (Lauritsen and Wilson, 1986; Haverkos, 1988a, 1990; Holub, 1988; Papadopulos-Eleopulos, 1988; Rappoport, 1988; Duesberg, 1990a, 1991a, 1992c,f; Lauritsen, 1990; Albonico, 1991a,b; Pillai et al., 1991; Cramer, 1992; Leonhard, 1992). Here the hypothesis is investigated that all American and European AIDS diseases, above the normal background of hemophilia and transfusion-related diseases, are the result of the long-term consumption of recreational and anti-HIV drugs.

4.1. Chronological Coincidence Between the Drug and AIDS Epidemics

The appearance of AIDS in America in 1981 followed a massive escalation in the consumption of psychoactive drugs that started after the Vietnam War (Newell et al., 1985b; Kozel and Adams, 1986; National Institute on Drug Abuse, 1987; Bureau of Justice Statistics, 1988; Haverkos, 1988b; Office of National Drug Control Policy, 1988; Flanagan and Maguire, 1989; Lerner, 1989; Shanon et al., 1990). The Bureau of Justice Statistics reports that the number of drug arrests in the U.S. has increased from about 450,000 in 1980 to 1.4 million in 1989 (Bureau of Justice Statistics, 1988; Shannon et al., 1990). About 500 kg of cocaine were confiscated by the Drug Enforcement Administration in 1980, about 9000 kg in 1983, 80,000 kg in 1989, and 100,000 kg in 1990 (Bureau of Justice Statistics, 1988, 1991; Flanagan and Maguire, 1989). In 1974, 5.4 million Americans had used cocaine at some point in their lives and in 1985 that number had gone up to 22.2 million (Kozel and Adams, 1986). Currently about 8 million Americans are estimated to use cocaine regularly (Weiss, S.H., 1989; Finnegan et al., 1992). The number of dosage units of domestic stimulants confiscated, such as amphetamines, increased from 2 million in 1981 to 97 million in 1989 (Flanagan and Maguire, 1989).

Several arguments indicate that these increases reflect increased drug consumption rather than just improved drug control, as has been suggested (Maddox, 1992a):

(1) The Bureau of Justice Statistics estimates that at most 20% of the cocaine smuggled into the U.S. was confiscated each year (Anderson, 1987).

(2) The National Institute on Drug Abuse reports that between 1981 and 1990 cocaine-related hospital emergencies increased 24-fold from 3296 to 80,355 and deaths from 195 to 2483 (Kozel and Adams, 1986; National Institute on Drug Abuse, 1990a,b). Thus cocaine-related hospital emergencies had increased 24-fold during 9 of the 10 years in which cocaine seizures had increased 100-fold.

(3) It is highly improbable that, before the jet-age, the U.S. would have imported annually as much cocaine as it did in 1990 plus the 100,000 kg that were confiscated in that year.

Further, the recreational use of psychoactive and aphrodisiac nitrite inhalants began in the 1960s and reached epidemic proportions in the mid-1970s, a few years before AIDS appeared (Newell et al., 1985b, 1988). The National Institute on Drug Abuse reports that in 1979-1980 over 5 million people used nitrite inhalants in the U.S. at least once a week (Newell et al., 1988), a total of 250 million doses per year (Wood, 1988). In 1976 the sales of nitrite inhalants in one American city alone amounted to $50 million annually (Newell et al., 1985b, 1988) at $5 per 12 mL dose (Schwartz, 1988).

Since 1987 the cytocidal DNA chain terminator AZT has been prescribed as an anti-HIV drug to AIDS patients (Kolata, 1987; Yarchoan and Broder, 1987b) and since 1990 to asymptomatic carriers of HIV (Editorial, 1990). Currently about 120,000 Americans and 180,000 HIV-positive persons worldwide, with and without AIDS, take AZT in efforts to inhibit HIV. This estimate is based on the annual AZT sales of $364 million and a wholesale price of $2000 per year for a daily dose of 500 mg AZT per person (Burroughs Wellcome Public Relations, 3 April 1992). In addition, an unknown number take other DNA chain terminators like ddI and ddC (Smothers, 1991; Yarchoan et al., 1991).

4.2. Overlap Between Drug-Use and AIDS Statistics

Drugs and AIDS appear to claim their victims from the same risk groups. For instance, the CDC reports that the annual mortality of 25- to 44-year-old American males increased from 0.21% in 1983 to 0.23% in 1987, corresponding to about 10,000 deaths among about 50 million in this group (Buehler et al., 1990). Since the annual AIDS deaths had also reached 10,000 by 1987, HIV was assumed to be the cause (Institute of Medicine, 1986; Centers for Disease Control, 1987, 1992b). Further, HIV infection was blamed for a new epidemic of immunological and neurological deficiencies, including mental retardation, in American children (Blattner et al., 1988; Institute of Medicine, 1988; Centers for Disease Control, 1992b).

However, mortality in 25- to 44-year-old males from septicemia, considered an indicator of intravenous drug use, rose almost 4-fold from 0.46 per 100,000 in 1980 to 1.65 in 1987, and direct mortality from drug use doubled (National Center for Health Statistics, 1989; Buehler et al., 1990), indicating that drugs played a significant role in the increased mortality of this group (Buehler et al., 1990). In addition, deaths from AIDS diseases and nonAIDS pneumonia and septicemia per 1000 intravenous drug users in New York increased at exactly the same rates, from 3.6 in 1984 to 14.7 and 13.6, respectively, in 1987 (Selwyn et al., 1989). Indeed, the cocaine-related hospital emergencies alone could more than account for the 32% of American AIDS patients that are intravenous drug users (Section 2.1.3). The emergencies had increased from "a negligible number of people" in 1973 to 9946 non-fatal and 580 fatal cases in 1985 (Kozel and Adams, 1986), when a total of 10,489 AIDS cases were recorded and to 80,355 nonfatal and 2483 fatal cases in 1990 (National Institute on Drug Abuse, 1990a,b), when a total of 41,416 AIDS cases were recorded by the CDC (Centers for Disease Control, 1992a). Moreover 82% of the cocaine-related and 75% of the morphine-related hospital emergencies were 20-39 years old (National Institute on Drug Abuse, 1990a), the age distribution typical of AIDS patients (Section 2.1.1).

Another striking coincidence is that over 72% of all American AIDS patients (Centers for Disease Control, 1992b) and about 75% of all Americans who consume "hard" psychoactive drugs such as cocaine, amphetamines and inhalants (National Institute on Drug Abuse, 1987, 1990a,b; Ginzburg, 1988) or get arrested for possession of such drugs (Bureau of Justice Statistics, 1988) or are treated for such drugs (National Institute on Drug Abuse, 1990a) are 20- to 44-year-old males. Thus there is substantial epidemiological overlap between the two epidemics (Lerner, 1989), reported as "The twin epidemics of substance use and HIV" by the National AIDS Commission (National Commission on AIDS, 1991).

Moreover, maternal drug consumption was blamed by some for the new epidemic of immunological and neurological deficiencies, including dementias, of American children (Toufexis, 1991). In view of this, the CDC acknowledges, "We cannot discern, however, to what extent the upward trend in death rates from drug abuse reflects trends in illicit drug use independent of the HIV epidemic" (Buehler et al., 1990).

4.3. Drug Use in AIDS Risk Groups

4.3.1. Intravenous Drug Users Generate a Third of All AIDS Patients

Currently 32% of the American (National Commission on AIDS, 1991; Centers for Disease Control, 1992b) and 33% of the European (Brenner et al., 1990; World Health Organization, 1992a) AIDS patients are intravenous or intrauterine users of heroin, cocaine, and other drugs (Section 2.1.3). These include:

(1) 75% of all heterosexual AIDS cases in America and about 70% of those in Europe,

(2) 71% of the American and 57% of the European females with AIDS,

(3) over 10% of the American and 5% of the European male homosexuals,

(4) 10% of the American hemophiliacs with AIDS,

(5) 70% of American children with AIDS including 50% born to mothers who are confirmed intravenous drug users and another 20% to mothers who had "sex with intravenous drug users" and are thus likely users themselves (Amaro et al., 1989),

(6) 80-85% of the European children with AIDS who were born to drug-addicted mothers (Mok et al., 1987; European Collaborative Study, 1991).

In an article entitled "AIDS and intravenous drug use: the real heterosexual epidemic" the AIDS researcher Moss points out that "90% of infected prostitutes reported in Florida, Seattle, New York and San Francisco have been intravenous drug users ... Drug use is also the source of most neonatal AIDS, with 70% of cases occurring in children of intravenous drug users ..." (Moss, 1987). Indeed, all studies of American and European prostitutes indicate that HIV infection is almost exclusively restricted to drug users (Rosenberg and Weiner, 1988), although all prostitutes should have the same risks of HIV infection, if HIV were sexually transmitted. Surprisingly, all of these studies only mention the incidence of HIV, rather than of AIDS, in prostitutes.

4.3.2. Homosexual Users of Aphrodisiac Drugs Generate about 60% of AIDS Patients

Approximately 60% of American AIDS patients are male homosexuals over the age of 20 (Table 1). They are generated by risk groups that have sex with large numbers of partners (Centers for Disease Control, 1982; Jaffe et al., 1983b; Darrow et al., 1987; Oppenheimer, 1992) that often average over 100 per year and have exceeded 1000 over a period of several years (Mathur-Wagh et al., 1984; Newell et al., 1985a; Turner et al., 1989; Callen, 1990). The following evidence indicates that these sexual activities and the corresponding conventional venereal diseases are directly proportional to the consumption of toxic sexual stimulants, which include nitrite- and ethylchloride inhalants, cocaine, amphetamines, methaqualone, lysergic acid, phenylcyclidine, and more (Blattner et al., 1985; Shilts, 1987; Lauritsen and Wilson, 1986; Darrow et al., 1987; Haverkos, 1988a; Rappoport, 1988; Raymond, 1988; Adams, 1989; Turner et al., 1989; Weiss, S.H., 1989; Ostrow et al., 1990; Lesbian and Gay Substance Abuse Planning Group, 1991a).

An early CDC study of 420 homosexual men attending clinics for sexually transmitted diseases in New York, Atlanta and San Francisco reported that 86.4% had frequently used amyl- and butylnitrites as sexual stimulants. The frequency of nitrite use was proportional to the number of sexual partners (Centers for Disease Control, 1982).

In 1983 Jaffe et al. investigated AIDS risk factors of 170 male homosexuals from sexual disease clinics, including 50 with Kaposi's sarcoma and pneumonia and 120 without AIDS. In this group, 96% were regular users of nitrite inhalants and 35-50% of ethylchloride inhalants. In addition, 50-60% had used cocaine, 50-70% amphetamines, 40% phenylcyclidine, 40-50% lysergic acid, 40-60% methaqualone, 25% barbiturates, 90% marijuana and 10% heroin (Jaffe et al., 1983b). Over 50% had also used prescription drugs. About 80% of these men had past or current gonorrhea, 40-70% had syphilis, 15% mononucleosis, 50% hepatitis and 30% parasitic diarrhea. Those with Kaposi's sarcoma had a median of 61 sex partners per year and those without AIDS about 26. The study points out that "lifetime exposure to nitrites ... (and) use of various 'street' drugs ... was greater for cases than controls." The lifetime drug dose of "cases" was reported to be two times higher than of asymptomatic HIV carriers (Jaffe et al., 1983b).

A study of a group of 359 homosexual men in San Francisco reported in 1987 that 84% had used cocaine, 82% alkylnitrites, 64% amphetamines, 51% methaqualone, 41% barbiturates, 20% injected drugs and 13% shared needles (Darrow et al., 1987). About 74% had past or current infection by gonococcus, 73% by hepatitis B virus, 67% by HIV, 30% by amoebae and 20% by treponema (Darrow et al., 1987). This group had been randomly selected from a list of homosexuals who had volunteered to be investigated for hepatitis B virus infection and to donate antisera to hepatitis B virus between 1978 and 1980. For the same group the 50% "progression rate" from HIV to AIDS was calculated to be 8-11 years (Table 2) (Moss et al., 1988; Lemp et al., 1990)-and reported to be relevant for "the (HIV-infected) population as a whole" (Moss et al., 1988)!

A study investigating AIDS risk factors among French homosexuals reported that 31% of those with AIDS, but only 12% of those without AIDS, had achieved "over 100 nitrite inhalations" (Messiah et al., 1988). The study included 53, or 45%, of all homosexual AIDS patients recorded in France by 1987.

The staggering oral drug use among male homosexuals at risk for AIDS was confirmed in 1990 by the largest survey of its kind. It reports that 83% of 3916 self-identified American homosexual men had used one, and about 60% two or more drugs with sexual activities during the previous six months (Ostrow et al., 1990). Similar drug use has been reported for European homosexuals at risk for AIDS (van Griensven et al., 1987).

A survey of homosexual men from Boston, conducted between 1985 and 1988, documented that among 206 HIV-positives 92% had used nitrite inhalants, 73% cocaine, 39% amphetamines, 29% lysergic acid in addition to six other psychoactive drugs as sexual stimulants; among 275 HIV-negative controls 71% had used nitrites, 57% cocaine, 21% amphetamines, 17% lysergic acid again in addition to six other psychoactive drugs (Seage et al., 1992). A similar survey of 364 HIV-positive homosexual men in Berlin conducted between 1983 and 1987 stated that 194 (53.3%) had used nitrite inhalants (Deininger et al., 1990).

According to Newell et al. (1985b), volatile nitrites had penetrated "every corner of gay life" by 1976. Surveys studying the use of nitrite inhalants found that in San Francisco 58% of homosexual men were users in 1984 and 27% in 1991, compared to less than 1% of heterosexuals and lesbians of the same age group (Lesbian and Gay Substance Abuse Planning Group, 1991b).

Several investigators have pointed out that nitrite inhalants, and possibly other drugs, are preferred by male homosexuals as aphrodisiacs because they facilitate anal intercourse by relaxing smooth muscles (Section 4.4.1) (Mirvish and Haverkos, 1987; Newell et al., 1985b; Ostrow et al., 1990; Lesbian and Gay Substance Abuse Planning Group, 1991a; Seage et al., 1992). "Nitrites were used primarily for heightened sexual stimulation during sexual activity by reducing social and sexual inhibitions, prolonging duration, heightening sexual arousal, relaxing the anal sphincter during anal intercourse, and prolonging orgasm" (Newell et al., 1985b).

4.3.3. Asymptomatic AZT Users Generate an Unknown Percentage of AIDS Patients

The DNA chain terminator AZT has been licensed in the U.S. since 1987 as a treatment for AIDS patients (Chernov, 1986; Kolata, 1987; Lauritsen, 1990; Yarchoan et al., 1991) based on a placebo controlled study sponsored by Burroughs Wellcome, the manufacturer of AZT (Section 4.4.2) (Fischl et al., 1987; Richman et al., 1987). In 1990 AZT was also licensed as AIDS prophylaxis for healthy HIV carriers (Section 4.4.2) (Volberding et al., 1990; Yarchoan et al., 1991).

The choice of this drug as anti-AIDS treatment is based entirely on the virus-AIDS hypothesis. According to Broder et al., "The rationale for anti-retroviral therapy for AIDS is ... that HIV is the etiologic agent of AIDS" and that HIV RNA-dependent DNA synthesis is inhibited by AZT (Yarchoan et al., 1991). In view of this and their faith in the virus-AIDS hypothesis, about 120,000 American HIV carriers, with and without AIDS, and 180,000 worldwide currently take AZT every day (Section 4.1). It follows that probably a high percentage of the 40,000 Americans and 15,000 Europeans that currently develop AIDS per year (Table 1 ) have used AZT and other DNA chain terminators prior to AIDS.

The drug is now recommended as AIDS prophylaxis for all AIDS-free persons with less than 500 T-cells per microliter by the director of AIDS research at the NIH (Kolata, 1992) and with some reservations also by the National Hemophilia Association of New York (personal communication), despite recent doubts about its usefulness (Kolata, 1992). For instance, AZT has been used indefinitely by over 1200 AIDS-free, but presumably HIV-infected, homosexual men from the Multicenter AIDS Cohort Study referenced above (Ostrow et al., 1990), including 7% of 3670 with over 500 T-cells per microliter, 16% of 1921 with 350-499 T-cells, 26% of 1374 with 200-349 T-cells and 51% of 685 with fewer than 200 T-cells (Graham et al., 1991). Yet the large study acknowledges finding "... no effects (of AZT) on rates of progression to lower CD4+ lymphocyte counts in any of the transition intervals" (Graham et al., 1991). In San Francisco 3.3% of 151 AIDS-free male homosexuals with over 500 T-cells, 11% of 128 with 200-500 T-cells and 36% of 42 with less than 200 T-cells were on AZT in 1989 (Lang et al., 1991). Another study reports that, in 1989, 26 out of 322 HIV-positive but AIDS-free homosexuals from San Francisco, Chicago and Denver had taken AZT for less than 6 months and 101 for over 6 months (Holmberg et al., 1992).

To distinguish between HIV and drugs as causes of AIDS, it is necessary to identify either HIV-carriers that develop AIDS only when they use drugs (Section 4.4) or to identify HIV-free drug users that develop AIDS indicator diseases (Section 4.5) and to demonstrate drug toxicity (Section 4.6).

4.4. Drug Use Necessary for AIDS in HIV-Positives

Studies demonstrating that drugs are necessary for AIDS among HIV-positives fall into two subgroups: (1) those demonstrating that AIDS among HIV-positives depends on the long-term use of recreational drugs and (2) those demonstrating that HIV-positive AIDS-free persons and AIDS patients on the antiviral drug AZT develop new AIDS diseases or AZT-specific diseases. Since the health of AIDS-free persons selected for AZT prophylaxis is compromised by prior AIDS risks, e.g. less than 500 T-cells, and since nearly all American and European AIDS patients have used recreational drugs or have been immunosuppressed by long-term transfusions, evaluating the role of AZT in the progression of AIDS is complicated by these confounding risk factors (Sections 3.4.4 and 4.3.3).

4.4.1. AIDS From Recreational Drugs

(1) A study of 65 HIV-infected drug users from New York showed that their T-cell count dropped over nine months in proportion with drug injection, on average 35%, compared to controls who had stopped (Des Jarlais et al., 1987).

(2) The incidence of AIDS diseases and death among HIV-positive, asymptomatic intravenous drug users over 16 months were 19% (23/124) among those who persisted in injecting psychoactive drugs, 5% (5/93) among those who had stopped injecting drugs and 6% (5/80) among those on methadone treatment (Weber et al., 1990).

(3) Among male homosexuals, receptive anal intercourse carries a 2.75 times (Warren Winkelstein, personal communication) to 4.4 times (Haverkos, 1988b) higher AIDS risk than insertive intercourse, presumably reflecting a higher risk of infection by HIV (Moss et al., 1987; van Griensven et al., 1987; Winkelstein et al., 1987; Seage et al., 1992). However, if HIV were the cause of AIDS, the donors should have the same AIDS risk as the recipients, because recipients can only be infected by HIV donors. No microbe can survive that is only unidirectionally transmitted. All venereal microbes are therefore bitransitive. Indeed, Haverkos found no differences in sexually transmitted diseases between those practicing receptive and insertive intercourse (Haverkos, 1988b). The probable reason for the higher AIDS risk associated with receptive anal intercourse is that this sexual practice directly correlates with a 2-fold (van Griensven et al., 1987; Seage et al., 1992) to an 8-fold (Moss et al., 1987; Haverkos, 1988b) enhanced use of nitrite inhalants and other aphrodisiac drugs that facilitate anal intercourse (Sections 4.3.2 and 4.6).

(4) A Canadian study reports that every one of 87 HIV-positive male homosexual AIDS patients had used nitrite inhalants. Those who had used over 20 "hits" per month were more likely to have Kaposi's sarcoma and sarcoma plus pneumonia than those who had used less than 20 hits per month. HIV-free controls, described in a previous report of the same cohort (Section 4.5) (Marion et al., 1989), were not mentioned in this study (Archibald et al., 1992). The authors concluded that a "sexually transmitted agent," which is even more difficult to transmit than HIV(!) (Section 3.5.1), would explain the Kaposi's sarcomas among the AIDS patients. The nitrites were proposed to be a cofactor of this cofactor of HIV (Archibald et al., 1992). Thus nitrites were necessary for AIDS in HIV-positives.

To determine whether HIV was indeed necessary for these AIDS cases, the incidence of AIDS-defining diseases in HIV-positive and negative homosexuals who are matched for the duration and extent of drug consumption must be compared. This is what the Canadian team has recently attempted to do in a study termed "HIV causes AIDS: a controlled study" (Craib et al., 1992). The study asserts to meet the challenge of "Duesberg [who] wrote in 1988 (Science, 1988; 242: 997-998) and repeated in public addresses in 1991 that the necessary comparisons in controlled cohorts were not available...."

However the study failed to match the HIV-free control group with the HIV-positives for the extent and duration of drug consumption. It mentions that 49% of the HIV-negatives had used "psychoactive drugs," but fails to mention the percentage of drug users among the HIV-positives. In their previous study 100% of the HIV-positive AIDS patients had used such drugs (Archibald et al., 1992). In addition the authors failed to recognize that HIV-infection is a marker for the duration of drug consumption. Since an average of 1000 sexual contacts is required for sexual transmission of HIV (Section 3.5.2), HIV is a marker for the dosage of sexual stimulants that is used for 1000 contacts. Thus HIV-positives would have used more sexual stimulants, the equivalent for 1000 contacts, than HIV-negatives. Indeed the authors acknowledge problems with "claims that AIDS is caused by other exposures and not by HIV ... the problem may be semantics. No one has ever disputed that cofactors play a very important role ..." (Craib et al., 1992). Moreover the authors failed to mention whether AZT was prescribed to the HIV-positives.

(5) A survey of 99, including 92 "gay or bisexual," AIDS patients from an "HIV clinic" at St. Mary's Hospital in London reports that 78% used "poppers" (nitrite inhalants), 78% cannabis, 76% cigarettes, 68% alcohol, and 48% "ecstasy" (amphetamines). In addition, the patients received an average of three unspecified medications, probably including AZT (Valentine et al., 1992). HIV-tests were not reported, but are assumed to be positive because the patients were in an "HIV clinic." Clearly, the multiplicity of drugs consumed by these patients could be relevant to their pathogenesis.

(6) A European survey of HIV-positive infants with AIDS found that "nearly all children were born to intravenous-drug-abusing mothers" and that AIDS was 9.4 times more likely in children whose mothers had AIDS symptoms before delivery than in those who had no symptoms (Mok et al., 1987). "Children with drug withdrawal symptoms" were most likely to develop diseases, those with no withdrawal symptoms but "whose mothers had used recreational drugs in the final 6 months of pregnancy were intermediate on all indices, whereas children of former drug users did not significantly differ from those born to women who had no history of i.v. drug use" (European Collaborative Study, 1991). An American survey reported that 63 of 68 infants "with symptomatic HIV infections" had "at least one parent who had AIDS or was in an AIDS high-risk group" (Belman et al., 1988). Since the risk of infants to develop AIDS increased with maternal drug consumption and increased 10-fold with maternal AIDS symptoms, it would appear that disease or subclinical deficiencies during pregnancy rather than perinatal infection by HIV are responsible for pediatric AIDS.

4.4.2. AIDS from AZT and AZT Plus Confounding Recreational Drug Use

(1) A placebo-controlled study, sponsored by Burroughs Wellcome the manufacturer of AZT, investigated 289 patients with "unexplained" weight loss, fever, oral candidiasis, night sweats, herpes zoster and diarrhea for the licensing of the drug as AIDS therapy in the U.S. (Fischl et al., 1987; Richman et al., 1987). All but 13 of these patients were males. The study was planned for 6 months, but it was interrupted after 4 months, because by then the therapeutic benefits of AZT seemed too obvious to continue the placebo control:

(a) after 4 months on AZT 1 out of 145 in the AZT group but 19 out of 137 in the placebo group had died. Therefore the study claimed that AZT can "decrease mortality";

(b) T-cell counts first increased from 4-8 weeks and then declined to pretreatment levels within 4 months;

(c) the lymphocyte count decreased over 50% in 34% of the AZT recipients but in only 6% of the control group;

(d) 66 in the AZT group suffered from severe nausea, compared to only 25 in the control group;

(e) muscle atrophy was observed in 11 AZT recipients but in only 3 from the control group. Yet, the primary claim of the study, "decreased mortality" from AZT is not realistic if one considers that 30 out of the 145 in the AZT-group depended on multiple transfusions to survive anemia, compared to only 5 out of the 137 in the placebo group. Thus the number of subjects in the AZT-group who would have died from severe anemia if untreated was larger, i.e. 30, than the AIDS deaths and anemias of the control group combined, namely 19+ 5. The "decreased mortality"-claim is further compromised by numerous "concomitant medications" other than transfusions for AZT-specific diseases and failure to match the AZT and placebo groups for the cumulative effects of prior and parallel recreational drug use. In addition some of the AZT-specific AIDS diseases observed in the placebo group appear to be due to patient-initiated "drug sharing" between AZT and placebo recipients (Lauritsen, 1990; Duesberg, 1992d; Freestone, 1992) and falsification of the case report forms (Lauritsen, 1992).

Moreover the low mortality of 0.7% (1/145) claimed by the licensing study for the first 4 months on AZT could not be extended in a follow-up study which found the "survival benefits" of AZT rapidly declining after the original 4 month period. By 18 months 32% of the original AZT group had died and 35% of the former control group, which by then had also received AZT for 12 months (Fischl et al., 1989).

Since the original study considered AZT effective in decreasing AIDS mortality, subsequent placebo-controlled studies were deemed unethical. But the low mortality claimed by the licensing study has not been confirmed by later studies, which observed mortalities of 12-72% within 9-18 months (see items (3) to (6) below). In addition, a CDC study has recently reported a mortality of 82% in a cohort of 55 AIDS patients that had been on AZT for up to 4 years (Centers for Disease Control, 1991)-hardly recommending AZT as an AIDS therapy.

The brief transient gains of T-cells observed upon AZT treatment by the licensing study may reflect compensatory hemopoiesis, random killing of pathogenic parasites (Elwell et al., 1987) and the influence of concomitant medication, including multiple transfusions (Richman et al., 1989). Indeed the study concluded, based on the "hematological toxicity" described above, that "... the initial beneficial immunological effects of AZT may not be sustained" (Richman et al., 1987). A French study confirms "... the decrease of cell counts below the initial value after a few months of AZT suggests that this drug might be toxic to cells" (see item (3) below) (Dournon et al., 1988). And a recent American study also confirms "... no effects on rates of progression to lower CD4 + lymphocyte counts in (6 month) transition intervals" (Section 4.3.3) (Graham et a]., 1991). Moreover, the manufacturer states, "A modest increase in mean CD4 (T4) counts was seen in the zidovudine group but the significance of this finding is unclear as the CD4 (T4) counts declined again in some patients" (Medical Economics Data, 1992).

(2) In view of the reported success of AZT as AIDS therapy, the drug was also tested for licensing as AIDS prophylaxis by much of the same team, including Fischl, Richman and Volberding, and again with support from the manufacturer Burroughs Wellcome (Volberding et al., 1990). The study treated AIDS-free, HIV-positive 25- to 45-year-old male homosexuals and intravenous drug users with "fewer than 500 T­cells" for one year either with AZT or with a placebo. The expected annual AIDS risk for intravenous drug users and male homosexual risk groups is about 4-6% per year without AZT (Section

The study reports AIDS diseases in: (1) 11 out of 453 on 500 mg AZT per day, (2) 14 out of 457 on 1500 mg AZT per day and (3) 33 out of 428 on a placebo (Volberding et a]., 1990). Thus the AZT-groups appeared to do better than expected and the placebo group did as expected. Therefore it was claimed that AZT prevents AIDS.

However, the price for the presumed savings of 22 (33-11) and 19 (33-14) AIDS cases with AZT, compared to the placebo group, was high because 19 AZT-specific cases of potentially fatal anemia, neutropenia and severe nausea appeared in the 500 mg AZT-group, and 72 such cases, including 29 anemias requiring life-saving blood transfusions, appeared in the 1500 mg AZT-group. This indicates cytocidal effects of AZT on hemopoiesis and on the intestines. Although the AZT-specific diseases were not diagnosed as AIDS, neutropenia generates immunodeficiency (Walton et al., 1986) and thus AIDS. If these AZT-specific cases were included in the calculation of benefits from AZT compared to the placebo group, the 500 mg-group no longer benefited and the 1500 mg-group tripled its disease risk.

The study was further compromised by its failure to match the treatment groups for their cumulative recreational drug use prior to and during the study and for the many compensatory treatments for the AZT-specific diseases of the subjects analyzed. The fact that 8 cases in the control group but only 3 and 1 in the 500 mg and 1500 mg-AZT groups developed AIDS cancers suggests that the control group could have been exposed to higher recreational drug doses.

Since the licensing study considered AZT effective in preventing AIDS, subsequent controlled trials were deemed unethical. However, several subsequent studies cast further doubt on the claim that AZT is a useful AIDS prophylactic. One study reported that persons with "early" AIDS, i.e. AIDS-free persons at risk for AIDS, died at the same rate of 12-14% as AIDS controls and that 82% developed leukopenia within less than a year (see item 6 below) (Hamilton et al., 1992). Another study described "no effects on rates of progression to lower CD4+ lymphocytes ..." recorded within 6 month periods in over 1200 AIDS-free men on AZT (Section 4.3.3) (Graham et al., 1991). A third study reported that 26 out of 127 HIV-positive, AIDS-free homosexuals had discontinued an unreported dose of AZT within less than 6 months, most because of severe toxicity (Section 4.3.3) (Holmberg et al., 1992). In view of these and other data, it is surprising that a loss of T-cells was not noted in the licensing study (Kolata, 1987).

(3) A French study investigated the effects of AZT on 365 AIDS patients. The patients included 72% male homosexuals and 11% intravenous drug users with a median age of 36 years and with opportunistic infections and Kaposi's sarcoma. The study, the largest of its kind, observed new AIDS diseases, including leukopenia, in over 40% and death in 20% within 9 months on AZT (Dournon et al., 1988). The AIDS diseases of 30% worsened during AZT treatment. The study reported no therapeutic benefits 6 months after initiating AZT therapy. The authors concluded: "... the rationale for adhering to high-dose regimens of AZT, which in many instances leads to toxicity and interruption of treatment, seems questionable."

(4) A Dutch study treating 91 male AIDS patients, averaging 39 years, after 67 weeks on AZT, observed mortality in 72% and AZT-specific myelotoxicity, requiring on average 5 blood transfusions, in 57%. About 34% of the myelotoxicity manifested in anemia and 20% in leukopenia. The authors concluded that "the majority of patients ... cannot be maintained on these (AZT) regimens, most commonly due to the development of hematological toxicity" (van Leeuwen et al., 1990).

(5) An Australian study involving 308 homosexual and bisexual men with Kaposi's sarcoma, lymphoma and opportunistic infections and a median age of 36 years, reported 30% mortality within 1-1.5 years on AZT. In addition one or more new AIDS diseases, including pneumonia, candidiasis, fever, night sweats and diarrhea were observed in 172 (56%) within one year (Swanson et al., 1990). Moreover, 50% needed at least one blood transfusion and 29% needed multiple blood transfusions to survive AZT treatment. Yet the authors concluded that the "risk:benefit ratio (is) advantageous to AIDS patients" (Swanson et al., 1990).

(6) A comparison of the effects of indefinite AZT treatment on 170 HIV-positive AIDS-free persons with "early" AIDS to 168 with "late" AIDS indicated that the mortality was the same in both groups, i.e. 12-14% per 1-1.5 years (Hamilton et al., 1992). The median age of the AZT recipients was 40 years; 63% were male homosexuals and 25% were intravenous drug users. AZT-specific diseases were observed in most "early AIDS" cases, i.e. leukopenia in 82%, severe leukopenia in 14%, anemia in 20%, severe anemia requiring transfusions in 5%, nausea in 40% and skin rashes in 47%. This indicates directly that AZT is toxic for AIDS-free HIV carriers, and that AZT toxicity is sufficiently dominant over other AIDS causes that it accelerates the progression to death of AIDS-free HIV carriers to the same rate that is observed in late AIDS patients (Duesberg, 1992d). The authors concluded that AZT, contrary to the Wellcome-sponsored study from 1987 conducted for licensing AZT, does not extend life.

(7) The annual lymphoma incidence of AZT-treated AIDS patients, with Kaposi's sarcoma, pneumonia and wasting disease, was reported to be 9% by the National Cancer Institute and was calculated to be 50% over three years (Pluda et al., 1990). The estimate of the 3-year incidence of lymphoma from this study was recently revised down to 31% (Yarchoan et al., 1991). An independent study observed in a group of 346 AIDS patients in London, most of whom were on AZT, "during the past three years a progressive increase in the number of patients dying from lymphoma, ..." to a current total of 16% in 1991 (Peters et al., 1991). And a CDC study reported a 15% lymphoma incidence during 24 months on AZT (Centers for Disease Control, 1991).

The lymphoma incidence of untreated, HIV-positive AIDS risk groups is 0.3% per year, derived from the putative average progression rate of 10 years from HIV to AIDS (Moss et al., 1988; Lemp et al., 1990; Duesberg, 1991a) and the 3% incidence of lymphoma in AIDS patients (Centers for Disease Control, 1992b). Therefore, the annual lymphoma risk of AZT recipients is about 30 times higher than that of untreated HIV-positive counterparts. It appears that the chronic levels of the mutagenic AZT, at 20-60 µM (500-1500 mg/person/day), were responsible for the lymphomas (Section 4.6.2).

An alternative interpretation suggests that AZT had prolonged life sufficiently to allow HIV to induce the lymphomas directly or via immunodeficiency (Pluda et al., 1990; Centers for Disease Control, 1991). However, this interpretation is flawed for several reasons: (1) Cancers, including malignant lymphomas, are not consequences of a defective immune system (Section 3.5.8). (2) There is as yet only a model for how HIV, the presumed killer of T-lymphocytes, could also cause cancer (Section 3.5.14) (Gallo, 1990). (3) AZT-induced lymphomas lack HIV-specific markers (McDunn et al., 1991). (4) Several studies indicate that AZT does not prolong life (see above) (Dournon et al., 1988; van Leeuwen et al., 1990; Hamilton et al., 1992; Kolata, 1992).

(8) Ten out of 11 HIV antibody-positive, AZT-treated AIDS patients recovered cellular immunity after discontinuing AZT in favor of an experimental HIV vaccine (Scolaro et al., 1991). The vaccine consisted of an HIV strain that was presumed to be harmless, because it had been isolated from a healthy carrier who had been infected by the virus for at least 10 years. Since there was no evidence that the hypothetical vaccine strain differed from that by which the patients were already naturally vaccinated, the only relevant difference between the patients before and during the vaccine trial was the termination of their AZT treatment. It follows that AZT treatment is at least a necessary, if not a sufficient, cause of immunodeficiency in HIV-positives.

(9) Four out of 5 AZT-treated AIDS patients recovered from myopathy two weeks after discontinuing AZT; two redeveloped myopathy on renewed AZT treatment (Till and MacDonnell, 1990), indicating that AZT is at least necessary for myopathy in HIV-positives.

(10) Four patients with pneumonia developed severe pancytopenia and bone marrow aplasia 12 weeks after the initiation of AZT therapy. Three out of 4 recovered within 4-5 weeks after AZT was discontinued (Gill et al., 1987), indicating that AZT is necessary for pancytopenia in HIV-positives.

4.5. Drug Use Sufficient for AIDS Indicator Diseases in the Absence of HIV

Studies demonstrating AIDS-defining diseases in drug users in the absence of HIV are chronologically and geographically censored by the virus-AIDS hypothesis. Before the general acceptance of this hypothesis in the U.S., there were numerous American studies blaming AIDS on recreational drugs, but afterwards there was but one American report describing HIV-free Kaposi's sarcomas in homosexuals who had used such drugs, and only a few American and some European studies describing AIDS-defining diseases in HIV-free intravenous drug users (see below).

If HIV were necessary for AIDS among drug users, only HIV-positive drug users should develop AIDS. However, there is not even one controlled study showing that among matched drug users only HIV-positives get AIDS. On the contrary, such studies all indicate that drugs are sufficient to cause AIDS.

4.5.1. Drugs Used for Sexual Activities Sufficient for AIDS Diseases

(1) The first five AIDS cases, diagnosed in 1981 before HIV was known, were male homosexuals who had all consumed nitrite inhalants and presented with Pneumocystis pneumonia and cytomegalovirus infection (Gottlieb et al., 1981).

(2) In 1985 and again in 1988 Haverkos analyzed the AIDS risks of 87 male homosexual AIDS patients with Kaposi's sarcoma (47), Kaposi's sarcoma plus pneumonia (20) and pneumonia only (20) (Haverkos et al., 1985; Haverkos, 1988b). All men had used several sexual stimulants, 98% had used nitrites. Those with Kaposi's sarcomas reported double the amount of sexual partners and 4.4-times more receptive anal intercourse than those with only pneumonia. The median number of sexual partners in the year prior to the illness was 120 for those with Kaposi's and 22 for those with pneumonia only. The Kaposi's cases reported 6-times more amylnitrite and ethylchloride use, 4-times more barbiturate use, and twice the methaqualone, lysergic acid and cocaine use than those with pneumonia only. Since no statistically significant differences were found for sexually transmitted diseases among the patients, the authors concluded that the drugs had caused Kaposi's sarcoma.

Although the data for Haverkos' analysis had been collected before HIV was declared the cause of AIDS, Haverkos' conclusion is valid. This is because (1) all patients had AIDS but only the heavy drug users had Kaposi's sarcoma in addition to immunodeficiency and because (2) not all can be assumed to be infected by HIV because transmission depends on an average of 1000 contacts (Section 3.5.2). Indeed, HIV was found in only 24% (Deininger et al., 1990), 31% (van Griensven et al., 1990), 43% (Graham et al., 1991; Seage et al., 1992), 48% (Winkelstein et al., 1987), 49% (Lemp et al., 1990), 56% (Marion et al., 1989) and 67% (Darrow et al., 1987) of cohorts of homosexuals at risk for AIDS in Berlin, Amsterdam, Chicago-Washington DC-Los Angeles-Pittsburgh, Boston, San Francisco and Canada that were similar to those described by Haverkos.

(3) A 4.5 year tracking study of 42 homosexual men with lymphadenopathy but not AIDS reported that 8 had developed AIDS within 2.5 years (Mathur-Wagh et al., 1984) and 12 within 4.5 years of observation (Mathur-Wagh et al., 1985). All of these men had used nitrite inhalants and other recreational drugs including amphetamines and cocaine, but they were not tested for HIV. The authors concluded that "a history of heavy or moderate use of nitrite inhalant before study entry was predictive of ultimate progression to AIDS" (Mathur-Wagh et al., 1984).

(4) Before HIV was known, three controlled studies compared 20 homosexual AIDS patients to 40 AIDS-free controls (Marmor et al., 1982), 50 patients to 120 controls (Jaffe et al., 1983b) and 31 patients to 29 controls (Newell et al., 1985a) to determine AIDS risk factors. Each study reported that multiple "street drugs" were used as sexual stimulants. And each study concluded that the "lifetime use of nitrites" (Jaffe et al., 1983b) were 94% to 100%-consistent risk factors for AIDS (Newell et al., 1985a).

(5) Early CDC data indicate that 86% of male homosexuals with AIDS had used oral drugs at least once a week and 97% occasionally (Centers for Disease Control, 1982; Haverkos, 1988b). The National Institute on Drug Abuse reports correlations from 69% (Lange et al., 1988) to virtually 100% (Haverkos, 1988a; Newell et al., 1988) between nitrite inhalants and other drugs and subsequent Kaposi's sarcoma and pneumonia.

(6) A 27- to 58-fold higher consumption of nitrites by male homosexuals compared to heterosexuals and lesbians (Lesbian and Gay Substance Abuse Planning Group, 1991 a,b) correlates with a 20-fold higher incidence of Kaposi's sarcoma (Selik et al., 1987; Beral et al., 1990) and a higher incidence of all other AIDS diseases in male homosexuals compared to most other risk groups (Tables 1 and 2).

(7) During the last 6-8 years the use of nitrite inhalants among male homosexuals decreased, e.g. from 58% in 1984 to 27% in 1991 in San Francisco (Lesbian and Gay Substance Abuse Planning Group, 1991b). In parallel, the incidence of Kaposi's sarcoma among American AIDS patients decreased from a high of 50% in 1981 (Haverkos, 1988b), to 37% in 1983 (Jaffe et al., 1983a), to a low of 10% in 1991 (Centers for Disease Control, 1992b). It follows that the incidence of Kaposi's sarcoma is proportional to the number of nitrite users.

(8) After the discovery of HIV, 5 out of 6 HIV-free male homosexuals from New York with Kaposi's sarcoma have reported the use of nitrite inhalants (Friedman-Kien et al., 1990). Some of these men had no immunodeficiency. Soon after another six cases of HIV-free Kaposi's sarcoma were reported in a "high risk population" from New York (Safai et al., 1991). This indicates directly that HIV is not necessary and suggests that drugs are sufficient for AIDS.

(9) A 44-year-old, HIV-free homosexual man from Germany developed Kaposi's sarcoma and had a T4 to T8-cell ratio of only 1.2. The man "had used nitrite inhalants for about 10 years," but had no apparent immunodeficiency (Marquart et al., 1991). Likewise, Kaposi's sarcoma was diagnosed in a 40-year-old, promiscuous HIV-free homosexual from England who admitted "frequent use of amyl nitrite." The patient was otherwise symptom-free with a normal T4/T8 cell ratio (Archer et al., 1989). In 1981 an English male homosexual with a "history of amylnitrite inhalation," hepatitis B, gonorrhea and syphilis was also diagnosed with Kaposi's sarcoma. In 1984 he was found to be free of HIV, but in 1986 he became antibody-positive (Lowdell and Glaser, 1989).

(10) A prospective study from Canada identified immunodeficiency in 33 out of 166 HIV-free homosexual men (Marion et al., 1989). The study did not mention drug consumption, but a later report on homosexual men with AIDS from the same cohort documented that all had been using either more or less than 20 "hits" of nitrites per month (Section 4.4) (Archibald et al., 1992). Thus nitrites and possibly other drugs were sufficient for immunodeficiency.

Likewise, Lang, et al. (1989) described a steady decline of T4-cells in 37 homosexual men in San Francisco from 1200 per µL prior to HIV infection to 600 or less at the time of infection. Although recreational drug use and AZT were not mentioned, other studies of the same cohort of homosexual men from San Francisco described extensive use of recreational drugs (Section 4.3.2) (Darrow et al., 1987; Moss, 1987) and AZT (Lang et al., 1991).

4.5.2. Long-term Intravenous Drug Use Sufficient for AIDS-defining Diseases

(1) Among intravenous drug users in New York representing a "spectrum of HIV-related diseases," HIV was observed in only 22 out of 50 pneumonia deaths, 7 out of 22 endocarditis deaths, and 11 out of 16 tuberculosis deaths (Stoneburner et al., 1988).

(2) Pneumonia was diagnosed in 6 out of 289 HIV-free and in 14 out of 144 HIV-positive intravenous drug users in New York (Selwyn et al., 1988).

(3) Among 54 prisoners with tuberculosis in New York state, 47 were street-drug users, but only 24 were infected with HIV (Braun et al., 1989).

(4) In a group of 21 long-term heroin addicts, the ratio of helper to suppressor T-cells declined during 13 years from a normal of 2 to less than 1, which is typical of AIDS (Centers for Disease Control, 1987; Institute of Medicine, 1988), but only 2 of the 21 were infected by HIV (Donahoe et al., 1987).

(5) Thrombocytopenia and immunodeficiency were diagnosed in 15 intravenous drug users on average 10 years after they became addicted, but 2 were not infected with HIV (Savona et al., 1985).

(6) The annual mortality of 108 HIV-free Swedish heroin addicts was similar to that of 39 HIV-positive addicts, i.e. 3-5%, over several years (Annell et al., 1991).

(7) A survey of over a thousand intravenous drug addicts from Germany reported that the percentage of HIV-positives among drug deaths (10) was exactly the same as that of HIV-positives among living intravenous drug users (Puschel and Mohsenian, 1991). Another study from Berlin also reported that the percentage of HIV-positives among intravenous drug deaths was essentially the same as that among living intravenous drug users, i.e. 20-30% (Bschor et al., 1991). This indicates that drugs are sufficient for and that HIV does not contribute to AIDS-defining diseases and deaths of drug addicts.

(8) In 1989, the annual mortality of 197 HIV-positive, parenteral drug users from Amsterdam with an average age of 29 years was 4% and that of 193 age-matched HIV-negatives was 3% (Mientjes et al., 1992). The annual incidence of pneumonia was 29% in the HIV-positives and 9% in the negatives. Clearly, a 3-fold higher morbidity is intrinsically inconsistent with a near identical mortality. However, the slightly higher mortality of HIV-positives is compatible with the fact that the positives had injected more drugs for a longer time, e.g. 84% of the positives vs 64% of the negatives had injected over the last 5 years, 85% vs 72% over the last 6 months and 59% vs 50% had injected heroin and cocaine.

(9) Lymphocyte reactivity and abundance were depressed by the absolute number of injections of drugs not only in 111 HIV-positive, but also in 210 HIV-free drug users from Holland (Mientjes et al., 1991).

(10) The same lymphadenopathy, weight loss, fever, night sweats, diarrhea and mouth infections were observed in 49 out of 82 HIV-free, and in 89 out of 136 HIV-positive, long-term intravenous drug users in New York (Des Jarlais et al., 1988).

(11) Among intravenous drug users in France, lymphadenopathy was observed in 41 and an over 10% weight loss in 15 out of 69 HIV-positives, and in 12 and 8, respectively, out of 44 HIV-negatives (Espinoza et al., 1987). The French group had used drugs for an average of 5 years, but the HIV-positives had injected drugs about 50% longer than the negatives.

(12) In a group of 510 HIV-positive intravenous drug users in Baltimore, 29% reported one and 19% reported two or more AIDS-defining diseases. In a control group of 160 HIV-negative intravenous drug users matched with the HIV-positives for "current drug use," again 29% reported one and 13% reported two or more AIDS-defining diseases (Munoz et al., 1992).

Nevertheless, the average T-cell count of HIV-negatives was about 2-times higher than that of HIV-positives (Munoz et al., 1992). As in the above French study (Espinoza et al., 1987), this appears to reflect a higher lifetime dose of drugs, because HIV is a marker for the duration and extent of drug consumption (Sections 3.4.3, 4.4 and 5).

(13) Among 97 intravenous drug users in New York with active tuberculosis, 88 were HIV-positive and 9 were HIV-negative; and among 6 "crack" (cocaine) smokers with tuberculosis, 3 were HIV-negative and 3 were positive (Brudney and Dobkin, 1991).

(14) The mental development and psychomotor indices of 8 HIV-infected and 6 uninfected infants were observed from 6-21 months of age. The mothers of each group were HIV-positive and had used intravenous drugs and alcohol during pregnancy (Koch, 1990; Koch et al., 1990; T. Koch, R. Jeremy, E. Lewis, P. Weintrub, C. Rumsey and M. Cowan, unpublished data). The median indices of both groups were significantly below average, e.g. 80/100 mental development and 85/100 psychomotor units. The uninfected infants remained on average about 5/100 units higher. A control group of 5 infants, born to HIV-negative mothers who had also used intravenous drugs and alcohol during pregnancy, also had subnormal indices averaging about 95/100 for both criteria.

The degree of neurological retardation of the infants correlated directly with maternal drug consumption: 80% of the mothers of infected infants were "heavy" and 10% occasional parenteral cocaine users and 33% were "heavy" and 33% occasional alcohol users during pregnancy; 45% of the mothers of uninfected infants were "heavy" and 30% occasional parenteral cocaine users and 35% were "heavy" and 30% occasional alcohol users; and 21% of the HIV-free mothers were "heavy' and 58% occasional parenteral cocaine users and 12% were "heavy" and 44% occasional alcohol users. In addition 66% of the HIV-positive and 63% of the negative mothers reported the use of opiates during pregnancy (T. Koch, R. Jeremy, E. Lewis, P. Weintrub, C. Rumsey and M. Cowan, unpublished data).

(15) The psychomotor indices of infants "exposed to substance abuse in utero" were "significantly" lower than those of controls, "independent of HIV status." Their mothers were all drug users but differed with regard to drug use during pregnancy. The mean indices of 70 children exposed during pregnancy were 99 and those of 25 controls were 109. Thus maternal drug use during pregnancy impairs children independent of HIV (Aylward et al., 1992).

The same study also reports a "significant difference" based on the HIV status of these children. The mean scores of 12 HIV-positives was 88 and that of 75 negatives was 102. But the study did not break down the scores of the HIV-positive infants based on "exposure to substance abuse in utero." Indeed, the scores of 4 of the 12 HIV-infected infants were "above average," i.e. 100-114, and 4 of the 12 mothers did not inject drugs during pregnancy.

(16) Ten HIV-free infants born to intravenous drug-addicted mothers had the following AIDS-defining diseases, "failure to thrive, persistent generalized lymphadenopathy, persistent oral candidiasis, and developmental delay ..." (Rogers et al., 1989).

(17) One HIV-positive and 18 HIV-free infants born to intravenous drug-addicted mothers had only half as many leukocytes at birth than normal controls. At 12 months after birth, the capacity of their lymphocytes to proliferate was 50-70% lower than that of lymphocytes from normal controls (Culver et al., 1987).

(18) Two studies to test the role of HIV on neurological function confirm the drug-AIDS hypothesis indirectly and directly. The first of these, which excluded users of psychoactive drugs, found that neuropsychometric functions of 50 HIV-negative homosexuals were the same as those of 33 HIV-positives (Clifford et al., 1990). Another study of intravenous drug users on methadone found that neither the drug-impaired neuropsychological functions of 137 HIV-negatives nor those of 83 HIV-positives were deteriorating over 7.4 months (McKegney et al., 1990). However, the study notes that the functions of HIV-positives were lower than those of HIV-negatives because "a greater number of injections per month, more frequent use of cocaine ... were strongly associated with HIV seropositivity."

Thus, a critical lifetime dosage of drugs appears necessary in HIV-positives and sufficient in HIV-negatives to induce AIDS-indicator and other diseases.

4.6. Toxic Effects of Drugs Used By AIDS Patients

4.6.1. Toxicity of Recreational Drugs

From as early as 1909 (Achard et al., 1909) evidence has accumulated that long-term consumption of psychoactive drugs leads to immune suppression and clinical abnormalities similar to AIDS, including lymphopenia, lymphadenopathy, fever, weight loss, septicemia, increased susceptibility to infections and profound neurological disorders (Terry and Pellens, 1928; Briggs et al., 1967; Dismukes et al., 1968; Sapira, 1968; Harris and Garret, 1972; Geller and Stimmel, 1973; Brown et al., 1974; Louria, 1974; McDonough et al., 1980; Cox et al., 1983; Kozel and Adams, 1986; Selwyn et al., 1989; Turner et al., 1989; Kreek, 1991; Pillai et al., 1991; Bryant et al., 1992). Since the early 1980s, when T-cell ratios became measureable, low T4 to T8-cell ratios averaging 1 or less were reported in addicts who had injected drugs for an average of 10 years (Layon et al., 1984).

Intravenous drugs can be toxic directly and indirectly. Indirect toxicity can be due to malnutrition, because of the enormous expense of illicit drugs, or to septicemia because most illicit drugs are not sterile (Cox et al., 1983; Stoneburner et al., 1988; Lerner, 1989; Buehler et al., 1990; Pillai et al., 1991; Luca-Moretti, 1992). Typically, intravenous drug users develop pneumonia, tuberculosis, endocarditis and wasting disease (Layon et al., 1984; Stoneburner et al., 1988; Braun et al., 1989; Brudney and Dobkin, 1991). Oral consumption of cocaine and other psychoactive drugs has been reported to cause pneumonitis, bronchitis, edema (Ettinger and Albin, 1989) and tuberculosis (Brudney and Dobkin, 1991). Physiological and neurological deficiencies, including mental retardation, are observed in children born to mothers addicted to cocaine and other narcotic drugs (Fricker and Segal, 1978; Lifschitz et al., 1983; Alroomi et al., 1988; Blanche et al., 1989; Root-Bernstein, 1990a; Toufexis, 1991; Finnegan et al., 1992; Luca-Moretti, 1992). According to the National Institute on Drug Abuse, "Cocaine is currently the drug of greatest national concern, from a public health point of view ..." (Schuster, 1984).

Because inhalation of alkylnitrites relaxes smooth muscles, it has been prescribed since 1867 against angina pectoris and heart pain at doses of 0.2 mL (Cox et al., 1983; Newell et al., 1985b; Shorter, 1987; Seage et al., 1992). No AIDS defining diseases have been reported at these doses in patients with those relatively severe, terminal cardiovascular diseases (Cox et al., 1983; Shorter, 1987), possibly because they did not live long enough to develop them. However, immediate and late toxicities have been observed in recreational users who have inhaled millilitres of nitrite inhalants (Newell et al., 1985b; Schwartz, 1988). Alkylnitrites are directly toxic as they are rapidly hydrolyzed in vivo to yield nitrite ions, which react with all biological macromolecules (Osterloh and Olson, 1986; Maikel, 1988). Addicts with 0.5 mM nitrite derivatives and 70% methemoglobin in blood have been recorded (Osterloh and Olson, 1986). Toxicity for the immune system, the central nervous system, the hematologic system and pulmonary organs has been observed after short exposure to nitrites in humans and in animals (Newell et al., 1985b, 1988; Wood, 1988). In 1982, Goedert et al. found that the helper to suppressor T-cell ratio was lower in homosexual men who had used volatile nitrite inhalants than among nonusers. Further, alkylnitrites were shown to be both mutagenic and carcinogenic in animals (Jorgensen and Lawesson, 1982; Hersh et al., 1983; Mirvish et al., 1988; Newell et al., 1985b, 1988).

By comparing the AIDS risk factors of 31 homosexual men with AIDS to 29 without, Newell et al. and others determined a direct "dose-response gradient": the higher the nitrite usage the greater the risk for AIDS (Marmor et al., 1982; Newell et al., 1985a; Haverkos and Dougherty, 1988) and deduced a 7-10 year lag time between chronic consumption and Kaposi's sarcoma (Newell et al., 1985b). Likewise, a French study of homosexual men with and without AIDS who had inhaled nitrites documents that "cases were significantly older (approximately 10 years) than controls" (Section 4.3.2) (Messiah et al., 1988). Also, a German study observed Kaposi's sarcoma in an HIV-free man after he had inhaled nitrites for 10 years (Section 4.5.1) (Marquart et al., 1991). These studies indicate that about 10 years of nitrite inhalation are necessary to convert "controls" to "cases."

In view of this several investigators have proposed that nitrite inhalants cause pulmonary and skin Kaposi's sarcoma and pneumonia by direct toxicity on the skin and oral mucosa (Centers for Disease Control, 1982; Marmor et al., 1982; Haverkos et al., 1985; Mathur-Wagh et al., 1985; Newell et al., 1985a; Lauritsen and Wilson, 1986; Haverkos, 1990). Because of their toxicity a prescription requirement was instated for the sale of nitrite inhalants by the Food and Drug Administration in 1969 (Newell et al., 1985b) and because of an "AIDS link" (Cox, 1986) the sale of nitrites was banned by the U.S. Congress in 1988 (Public Law 100-690) (Haverkos, 1990) and by the "Crime Control Act of 1990" (January 23, 1990).

Although a necessary role of HIV in HIV-positive AIDS patients cannot be excluded, this role would be stoichiometrically insignificant compared to that of the drugs. This is because drug molecules exceed HIV molecules by over 13 orders of magnitude. Given about 1010 leukocytes per human, of which at most 1 in 104 are actively infected (Section 3.5.1), and that each actively infected cell makes about 100 viral RNAs per day, there are only 106 T-cells with 102 HIV RNAs in an HIV-positive person. By contrast, 1 mL (or 0.01 mol) of amyl nitrite with a molecular weight of 120 contains 6 x 1021 molecules, or 6 x 107 nitrite molecules, for every one of the 1014 cells in the human body. Thus, based on molecular representation, HIV's role in AIDS, if it existed, would have to be catalytic in comparison with that of drugs.

Pillai, Nair and Watson conclude from a recent review on the role of recreational drugs in AIDS: "Circumstantial and direct evidence suggesting a possible role for drug ... induced immunosuppression appears overwhelming. What is required now is better and more accurate detection of substance abuse, a direct elucidation of the immune and related mechanisms involved, and appropriate techniques to analyze it" (Pillai et al., 1991).

4.6.2. Toxicity of AZT

Since 1987 AZT has been used as an anti-HIV agent (Section 4.3.3) based on two placebo-controlled studies reporting therapeutic and prophylactic benefits (Section 4.4.2). However, AZT was originally developed in the 1960s for cancer chemotherapy to kill human cells via termination of DNA synthesis (Cohen, 1987; Yarchoan and Broder, 1987a; Yarchoan et al., 1991). The primary AZT metabolites are 3'-termini of DNA which are cell-killing, 3'-amino-dT which is more toxic than AZT, and 5'-O-glucuronide which is excreted (Cretton et al., 1991). As a chain terminator of DNA synthesis, AZT is toxic to all cells engaged in DNA synthesis. AZT toxicity varies a great deal with the subject treated due to differences in its uptake and in its cellular metabolism (Chernov, 1986; Elwell et al., 1987; Yarchoan and Broder, 1987b; Smothers, 1991; Yarchoan et al., 1991).

AZT is prescribed as AIDS prophylaxis or therapy at 500-1500 mg per day, corresponding to a concentration of 20-60 µmol/L in the patient. Prior to the licensing of AZT, Burroughs Wellcome, the manufacturer of the drug, and the NIH have jointly claimed selective inhibition of HIV by AZT in vitro because human lymphoblasts and fibroblasts appeared over 1000-fold more resistant to AZT (inhibited only at 1-3 mM) than was replication of HIV (inhibited at 50-500 nM) (Furman et al., 1986). On this basis they calculated an in vitro antiviral therapeutic index of 104. This "selective" sensitivity of HIV to AZT was explained in terms of a "selective interaction of AZT with HIV reverse transcriptase" (Furman et al., 1986). Accordingly the manufacturer informs AZT recipients: "The cytotoxicity of zidovudine [AZT] for various cell lines was determined using a cell growth assay ... ID50 values for several human cell lines showed little growth inhibition by zidovudine except at concentrations > 50 µg/mL (³200 µM) or less." (Medical Economics Data, 1992). Further, it informs them that enterobacteria including E. coli are inhibited "by low concentrations of zidovudine [AZT]," between 0.02 and 2 µM AZT, just like HIV (Medical Economics Data, 1992).

However, an independent study showed in 1989 that AZT is about 1000-times (!) more toxic for human T-cells in culture, i.e. at about 1 µM than the study conducted by its manufacturer and the NIH (Avramis et al., 1989). Other studies have also found that AZT inhibits T-cells and other hemopoietic cells in vitro at 1-8 µM (Balzarini et al., 1989; Mansuri et al., 1990; Hitchcock, 1991). Since normal deoxynucleotide triphosphates are present in the cell at micromolar concentrations, toxicity of AZT should be expected in the micromolar range. Indeed, when AZT is added at a micromolar concentration to the culture medium, it and its phosphorylated derivatives quickly reach an equivalent or higher concentration in the cell, and thus effectively compete with their natural thymidine counterparts (Avramis et al., 1989; Balzarini et al., 1989; Ho and Hitchcock, 1989; Hitchcock, 1991).

Thus the low cellular toxicity reported by the manufacturer and the NIH for human cells appears erroneous-possibly because "the clinical development of AZT was exceedingly rapid; it was approved for clinical use in the U.S. about 2 years after the first in vitro observation of its activity against HIV" (Yarchoan et al., 1991). It follows that AZT does not selectively inhibit viral DNA synthesis and is prescribed at concentrations that exceed 20- to 60-fold the lethal dose for human cells in culture.

In view of its inevitable toxicity, the rationale of using AZT as an anti-HIV drug must be reconsidered and its potential antiviral effect must be weighed against its toxicity. AZT not a rational anti-HIV drug. A rational antiviral therapy depends on proof that the targeted virus is the cause of the disease to be treated and that toxicity for the virus outweighs that for the host cell. Such proof cannot be supplied for AZT for the following reasons:

(1) There is no proof that HIV causes AIDS (Section 3.3).

(2) Even if the hypothesis that HIV causes AIDS by killing T-cells were correct, it would be irrational to kill the same infected cells twice, once presumably with HIV and once more with AZT.

(3) Since many healthy persons with antibodies against HIV have equal or even higher percentages of infected T-cells than AIDS patients (Section 3.3), there is no reverse transcription of HIV during progression to AIDS that could be targeted with AZT. Even if some reverse transcription occurred in antibody-positive persons, AZT could not differentially inhibit viral DNA, because HIV DNA comprises only 9 kb but cell DNA comprises 106 kb. Thus cell DNA is a 100,000-fold bigger target for AZT than HIV. And even if AZT showed a 100-fold preference for reverse transcriptase of HIV over cellular DNA polymerase, as has been claimed by the study conducted by Burroughs Wellcome and the NIH (Furman et al., 1986), cell DNA would still be a 1000-fold bigger target for AZT than viral DNA. It follows that cell DNA is the only realistic target of AZT in antibody-positive persons.

(4) Since AZT cannot distinguish infected from uninfected leukocytes and on average less than 1 in 1000 is infected (Section 3.3), AZT must kill at least 1000 leukocytes in AIDS patients and in asymptomatic HIV-carriers to kill just 1 infected cell-a very high toxicity index, even if HIV were the cause of AIDS.

It follows that there is no rational basis for AZT therapy or prophylaxis for AIDS (Duesberg, 1992d). Toxicity of AZT in AIDS Patients and AIDS-free Persons. The following AZT-specific diseases have been recorded in AIDS patients, in AIDS-free persons and animals treated with AZT, based on studies listed here (Section 4.4.2) and reviewed elsewhere (Smothers, 1991; Medical Economics Data, 1992):

(1) anemia, neutropenia and leukopenia in 20-80%, with about 30-57% requiring transfusions within several weeks (Gill et al., 1987; Kolata, 1987; Richman et al., 1987; Dournon et al., 1988; Walker et al., 1988; Swanson et al., 1990; van Leeuwen et al., 1990; Smothers, 1991; Hamilton et al., 1992),

(2) severe nausea from intestinal intoxication in up to 45% (Richman et al., 1987; Volberding et al., 1990; Smothers, 1991),

(3) muscle atrophy and polymyositis, due to inhibition of mitochondrial DNA synthesis in 6-8% (Richman et al., 1987; Bessen et al., 1988; Gorard and Guilodd, 1988; Helbert et al., 1988; Dalakas et al., 1990; Till and MacDonnell, 1990; Yarchoan et al., 1991; Hitchcock, 1991),

(4) lymphomas in about 9% within 1 year on AZT (Section 4.4.2),

(5) acute (nonviral) hepatitis (Dubin and Braffman, 1989; Smothers, 1991),

(6) nail dyschromia (Don et al., 1990; Smothers, 1991),

(7) neurological diseases including insomnia, headaches, dementia, mania, Wernicke's encephalopathy, ataxia and seizures (Smothers, 1991), probably due to inhibition of mitochondrial DNA (Hitchcock, 1991),

(8) 12 out of 12 men reported impotence after 1 year on AZT (Callen, 1990),

(9) in addition AZT is carcinogenic in mice, causing vaginal squamous carcinomas (Cohen, 1987; Yarchoan and Broder, 1987a), and it transforms mouse cells in vitro as effectively as methylcholanthrene (Chernov, 1986).

Overall AZT is not a rational prophylaxis or a therapy for AIDS and is capable of causing potentially fatal diseases, such as anemia, leukopenia and muscle atrophy. Yet, despite its predictable toxicity, AZT is thought to have serendipitous therapeutic and prophylactic benefits according to those investigators who have studied its effects together with the manufacturer for licensing of the drug (Section 4.4.2) (Fischl et al., 1987; Richman et al., 1987; Volberding et al., 1990). Confronted with the difficulties in rationalizing anti-HIV prophylaxis and therapy with AZT, the Wellcome researcher Freestone cites the Burroughs Wellcome study analyzed above (Section 4.4.2, item 1): "the primary end-point for the study was death (1 in 145 zidovudine recipients, 19 in 137 placebo recipients ...)-an end-point little subject to observer error or bias" (Freestone, 1992).

The popularity of AZT as an anti-HIV drug can only be explained by the widespread acceptance of the virus-AIDS hypothesis, the failure to consider the enormous difference between the viral and cellular DNA targets and a general disregard for the long-term toxicity of drugs (Section 6). In the words of the retrovirologist Temin "but the drug generally becomes less effective after six months to a year ..." (Nelson et al., 1991)-a euphemism for its fatal toxicity by that time. This is a probable reason that AZT was licensed without long-term studies in animals compatible with human applications and that the need for such studies is neither mentioned nor called for in reviews of its toxic effects in humans (Chernov, 1986; Yarchoan and Broder, 1987b; Smothers, 1991; Yarchoan et al., 1991), although AZT must be the most toxic drug ever approved for indefinite therapy in America. Even the manufacturer acknowledges that "... the drug has been studied for limited periods of time and long term safety and efficacy are not known" (Shenton, 1992) and recommends that "patients should be informed ... that the long-term effects of zidovudine are unknown at this time" (Medical Economics Data, 1992). And after prescribing it for five years, even AIDS "experts" have recently expressed doubts about the "survival benefit" of AZT (Kolata, 1992).

4.7. Drug-AIDS Hypothesis Correctly Predicts the Epidemiology and Heterogeneous Pathology of AIDS

(1) The long-term consumption of drugs, but not the hosting of a latent virus, predicts drug-specific pathogenicity after "long latent periods." These long latent periods of HIV are in reality the lag periods that recreational drugs (Schuster, 1984; Newell et al., 1985b) and frequent transfusions of foreign proteins take to cause AIDS-defining diseases (Section Drugs are molecularly abundant (Section 4.6.1) and biochemically active as long as they are administered and thus cumulatively toxic over time. It is for this reason that it typically takes 5-10 years for recreational drugs, and months for AZT, to cause AIDS-defining and other diseases (Sections 3.1 and 5). But HIV, after a brief period of immunogenicity (Clark et al., 1991; Daar et al., 1991), is chronically dormant and thus molecularly and biochemically irrelevant for the rest of the host's life.

(2) Drugs and other noninfectious agents also exactly predict the epidemiology of AIDS. About 32% of American AIDS patients are confirmed intravenous drug users, 60% appear to use recreational drugs orally, and an unknown but large percentage of people in both behavioral and clinical AIDS risk groups use AZT. Moreover, the consumption of recreational drugs by AIDS patients is probably under-reported because the drugs are illicit, and because medical scientists and support for research are currently heavily biased in favor of viral-AIDS (Section 6) (Ettinger and Albin, 1989; Lerner, 1989; Duesberg, 1991b). In sum, more than 90% of American AIDS is correlated with drugs. The remainder would reflect the natural background of AIDS-defining diseases in the U.S. (Duesberg, 1992). Indeed, only drug users do not benefit from the ever improving health parameters and increasing life spans of the Western World (Hoffman, 1992; The Software Toolworks World Atlas™, 1992). The widespread use of AZT in hemophiliacs (Section 4.3.3) unfortunately predicts a new increase in their mortality.

The dramatic increase in America in the consumption of all sorts of recreational drugs since the Vietnam War also explains the simultaneous increase of AIDS in intravenous drug users and male homosexuals (Centers for Disease Control, 1992b). AIDS of both risk groups followed closely the above listed drug-use statistics during the last 15 years, with increases in 1987 that corresponded to the expanded AIDS definition (Centers for Disease Control, 1987) and the introduction of AZT treatment. By contrast a sexually transmitted AIDS would have spread much faster among homosexuals than among intravenous drug users (Weyer and Eggers, 1990; Eggers and Weyer, 1991). The apparent exponential spread of AIDS during the period from 1984 to 1987 (Heyward and Curran, 1988; Mann et al., 1988; Weyer and Eggers, 1990) probably reflected an exponential spread of "AIDS testing," which resulted in an exponential spread of AIDS diagnoses for drug diseases (Section 4.2). AIDS testing had increased from 0 in 1984 to 20 million tests per year in 1986 in the U.S. alone (Section 3.6).

(3) The drug hypothesis further predicts that the 50-70% of American and 50-80% of European intravenous drug users who are HIV-free (Stoneburner et al., 1988; Turner et al., 1989; Brenner et al., 1990; U.S. Department of Health and Human Services, 1990; National Commission on AIDS, 1991), and the HIV-free male homosexuals who use sexual stimulants will develop the same diseases as their HIV-positive counterparts-except that their diseases will be diagnosed by their old names. This has been amply confirmed for intravenous drug users (Section 4.5). But since AIDS research became dominated by the virus-hypothesis, only a few studies have published HIV-free homosexual immunodeficiencies and "AIDS cases" (Section 4.5, Note added in proof). Yet more such cases must exist because the CDC allows "presumptive diagnosis" of HIV disease and only about 50% of all American AIDS cases are confirmed positives (Sections 2.2 and 3.4.1) and because only about 50% of homosexuals from many different cohorts at risk for AIDS are confirmed HIV-positive (Section 4.5.1).

(4) The drug hypothesis also correctly predicts drug-specific AIDS diseases in distinct risk groups due to distinct drugs (Sections 2.1.3, 3.4.5 and 5, Table 2).

4.8. Consequences of the Drug-AIDS Hypothesis: Risk-Specific Preventions and Therapies, but Resentment by the Virus-AIDS Establishment

The drug-AIDS hypothesis predicts that the AIDS diseases of the behavioral AIDS risk groups in the U.S. and Europe can be prevented by stopping the consumption of recreational and anti-HIV drugs, but not by "safe sex" (Institute of Medicine, 1988; Weiss and Jaffe, 1990; Maddox, 1991b) and "clean needles," i.e. sterile injection equipment (National Commission on AIDS, 1991) for toxic and unsterile street drugs. Indeed AIDS has continued to increase in all countries that have promoted safe sex to prevent AIDS for over 5 years now (Centers for Disease Control, 1992b; World Health Organization, 1992a; Anderson and May, 1992). Further, the hypothesis raises the hopes for risk-specific therapies.

According to the drug-AIDS hypothesis, AZT is AIDS by prescription. Screening of blood for antibodies to HIV is superfluous, if not harmful, in view of the anxiety that a positive test generates among the many believers in the virus-AIDS hypothesis (Grimshaw, 1987) and the toxic AZT prophylaxis prescribed to many who test "positive." Eliminating the test would also reduce the cost of the approximately 12 million annual blood donations in the U.S. (Williams et al., 1990) and of examining annually 200,000 recruits and 2 million servicemen for the U.S. Army (Burke et al., 1990) by $12 to $70 each (Irwin Memorial Blood Bank, San Francisco, personal communication).

Further, it would lift travel restrictions for antibody-positives to many countries including the U.S. and China, and would lift quarantine for HIV-positive Cubans, and would acquit all those antibody-positive Americans who are currently imprisoned for having had sex with antibody-negatives, and would grant to HIV "antibody-positives" the same chances to be admitted to a health insurance program as to those who have only antibodies to other viruses.

Despite its many potential blessings, the drug hypothesis is currently highly unpopular-not because it would be difficult to verify, but because of its consequences for the virus-AIDS establishment (Section 6). The drug hypothesis is very testable epidemiologically and experimentally by studying the effects of the drugs consumed by AIDS patients in animals. Indeed most tests have already been done (Section 4). To disprove this hypothesis it would be necessary to document that an infectious agent exists which-in the absence of AZT (!) causes AIDS diseases above their normal background in the nondrug using population. The medical, ethical and legal consequences of the drug-AIDS hypothesis, should it prevail, have recently been summarized under the title "Duesberg: An enemy of the people?" (Ratner, 1992). Ratner points out that, "The loss of confidence of Americans in their scientists and perhaps, by extension, their physicians, could rival their current disillusionment with politicians" and wonders, "What would happen to the reservoir of good will painstakingly built up for the victims of AIDS?"