Long-term cannabis abuse may increase the risk of schizophrenia. Nerve growth factor (NGF) is a pleiotropic neurotrophic protein that is implicated in development, protection and regeneration of NFG sensitive neurones. We tested the hypothesis that damage to neuronal cells in schizophrenia is precipitated by the consumption of cannabis and other neurotoxic substances, resulting in raised NGF serum concentrations and a younger age for disease onset. The NGF serum levels of 109 consecutive drug-naive schizophrenic patients were measured and compared with those of healthy controls. The results were correlated with the long-term intake of cannabis and other illegal drugs. Mean (± SD) NGF serum levels of 61 control persons (33.1 ± 31.0 pg and 76 schizophrenics who did not consume illegal drugs (26.3 ± 19.5 pg/mi) did not differ significantly, Schizophrenic patients with regular cannabis intake (> 0.5 g on average per day for at least 2 years) had significantly raised NGF serum levels of 412.9 ± 288.4 pg/mI Cu 21) compared to controls and schizophrenic patients not consuming cannabis (p c 0001). In schizophrenic patients who abused not only cannabis, but also additional substances, NGF concentrations were as high as 2336.2 ± 1711.4 pg C = 12). On average, heavy cannabis consumers suffered their first episode of schizophrenia 3.5 years (n = 21) earlier than schizophrenic patients who abstained from cannabis. These results indicate that cannabis is a possible risk factor for the development of schizophrenia. This might be reflected in the raised NGF-serum concentrations when both schizophrenia and long-term cannabis abuse prevail.

Discussion

These results demonstrate that serum NGF concentrations in untreated schizophrenic patients differed greatly depending on their long-term intake of drugs of abuse. Whereas he drug-naive schizophrenic patients who had not consumed illegal substances n the past showed no significant difference in serum NGF concentrations, those abusing cannabis for longer than 2 years showed significantly elevated N compared to healthy controls. This has been shown unequivocally not only by a descriptional data analysis, but also by a confirmatory study design (Table 2). Schizophrenic patients with long-term abuse of multiple substances showed an even greater increase in their serum NGF concentrations up to 90-fold above non-abusing schizophrenic patients .

NGF-plasma levels in 26 male schizophrenic patients who had been kept free of neuroleptics for 14 days were reported to be significantly lower than those observed in controls (Bersani er al., 1999).  By contrast, in our much larger sample of patients, we found no significant differences with respect to serum NGF concentrations between schizophrenic patients and controls. However, our patients were comp!etely drug-naive whereas the patients of the formerly cited study previously had been treated with antipsychotics for various time spans. It is known that haloperidol can remain in the cerebral tissue for as long as year after application (Konihuber et al, 1999) thereby possibly modulating and influencing NGF values by its antidopaminergic properties. For this reason, a drug-free period of l day might be too short to rule out the pharmacological effects of footer antipsychotic medication on the NGF concentration. Haloperidol reduces the basal TGF plasma levels in eight formerly neuroleptic free schizophrenic patients (Aloe et al. 1997). One explanation could be cosecretion of NGF with prolactin. with both being controlled by activation of the dopamirie D receptor subtype (Missale et al. 1996) that, in turn, can be blocked by the antidopamine drug Haloperidol. Moreover brain-derived neurotrophic factor (BDNF). another NGF-related neurotrophin. was shown to be decreased in the serum of chronic schizophrenic patients who were already treated with neuroleptics (Tovooka ci at. 2002). However. we demonstrated highly signficant elevations of the NGF serum levels in schizophrenic patients who had consumed significant amounts of cannabis in the past,  more than 0.5 per day over at least 2 years). There’s strong circumstantial evidence for neuronal damage by toxic drugs, but only a few neurochemical studies to support this.    Schizophrenia. a disease of alleged neurodevelopmental origin begins in  adolescence at age 16—24 years and is thought to coincide with increased vulnerability to cannabinoids. sometimes even triggered by them (Andreasson et al. 1987: Linszen et al. 1994). In an analogy to the situation in neurodegenerative disease (for reviews, see Heliweg et al, 1998; Siegel and Chauhan, 2000). the high levels of NGF observed in our study might reflect the assumed adverse effects induced by cannabis consumption in schizophrenia development.

At present, the causes and mechanisms of the observed rise in NGF serum concentrations in schizophrenia following long-term cannabis abuse remain speculative. Similarly, from the present data, it is not possible to establish whether the rise in NGF levels is due to disease development (enhanced by cannabis) as a state marker or whether NGF was even already high before disease onset. Theoretically, patients at risk for an unfavourable outcome of schizophrenia due to repeated cannabis consumption could be assumed to be a different patient population altogether and show premorbid rises in NGF concentrations as a risk-trait variable. a small sample (n = I of subjects without schizophrenia, but regular cannabis consumption of at least 0.5 g per day for longer than 2 years. We found no such elevation of NGF measurements in the serum. The same was true when serum NGF concentrations were measured in otherwise healthy controls acutely intoxicated with cannabis ( 5; Anders et al,  unpublished data), These findings indicate that the rise in NGF concentrations is not an effect of chronic cannabis consumption per se but rather reflects the combined damaging effects of cerebral vulnerability in schizophrenia and the chronic toxicity of long-term cannabis abuse. The earlier onset of schizophrenia in the cannabis consuming patients further substantiates this hypothesis.

Greatly raised NGF serum concentrations have been demonstrated in chronic diseases such as alcohol dependence (Aloe a at.. 1996) or Behcet’s disease (Lockers-Scherlibi C, 1996). They are not specific for a certain diagnosis. but rather are a marker for chronicity of disease, and possibly for poor prognosis as seen in Behcet’s disease. Our finding of even greater serum NGF concentration in schizophrenic patients with a long-term consumption of additional substances with neuroroxic effects, is consistent with the hypothesis of an NGF correlation with the cumulative dose and toxicity of drugs. The rise was up to 90-fold of the mean NGF serum concentrations of schizophrenic patients without drug consumption, which corresponds to the highest endogenous NGF concentrations reported for man to date. Accordingly, cumulative doses of ecstasy have been demonstrated to be neurotoxic and exert delayed and/or chronic cerebral alterations (Ricaurte and McCann. 1992). showing altered glucose metabolism in positron emission tomography studies in the hippocampus and amygdala of seven chronic ecstasy users (Obrocki et al. 1999). Previous magnetic resonance imaging studies with chronic drug abusers of various drugs including cocaine, amphetamines and psychedelics. also showed minor structural brain changes (Ansley et al. 1993). However, the investigators did note that the study probands had also been consuming alcohol, Therefore, the structural central nervous system changes did not clearly reflect those effects exclusively attributable to illegal drugs, but possibly also those due at least in part to alcohol. To avoid such confounding factors in our study, we excluded those patients who consumed alcohol on a regular basis, This explains the lack of a control group with polysubstance abuse but no schizophrenia because we were unable to find probands that were otherwise reasonably healthy and consumed no alcohol. Certainly, this remains a field for future research.

The origin of the NGF measured in serum is speculative: on the one hand, serum NGF could well stem from a central source and reflect the contral neurotrophin state, especially in pathological conditions such as pre-clinical Alzheimer’s disease (Schaub et al. 2002). On the other hand, it could reflect a peripheral immun ological reaction in terms of a cytokine released from peripheral immune cells (for reviews, see Levi-Montalcini et al. 1996;). Schizophrenia has been connected to autoimmune disease (Ganguli et al. 1994; Jones and Cannon. 1998) and to inflammatory disease (Lin et al. 1998). both possibly resulting in central or peripheral immune responses. An argument could be made about the principal evidence for a role of the neurotrophins NGF or BDNF in conjunction with schizophrenia. In the meantime, there are a number of experiments indicating a connection between BDNF and schizophrenia (Toyooka et at 2002) and some indicating NGF as not only having a role as a peripheral cytokine. but also as a factor relevant to schizophrenia (for a review, see Aloe et al.  2000).

In summary, we suggest that the raised NGF serum concentrations found in schizophrenics with long-term cannabis abuse, and more so in schizophrenics with long-term abuse of additional drugs, reflect the amount of cerebral damage by the combined effects of a primary cerebral vulnerability resulting from schizophrenia and the supposed additional drug-neurotoxicity. Apart from the biochemical evidence of an additive effect of schizophrenia and cannabis consumption on NGF serum concentrations in our confirmatory study design, we also demonstrated an earlier onset of disease in schizophrenic patients consuming cannabis chronically, thereby underlining a precipitating effect of the drug on disease onset. Those two findings are suggestive of a correlation but further studies are required to confirm this hypothesis. Thus, cannabis use may be a risk factor in schizophrenia development, but a predisposition to schizophrenia and cannabis use combined (but neither one independently) appears to be linked to increased NGR production.