The results of this study showed that the majority of the patients with lung cancer had a history of O&D use, and the use of O&D may be associated with an increased risk of LC. Also, in this study, there was a dose-response relation between O&D use and the risk of LC, and by increased O&D use, the risk of these types of cancers increased as well.
However, the relation between O&D use and cancer may be confounded by other risk factors such as age, gender, cigarette smoking, and alcohol use (30). In Iran, opioids consumers are mainly elderly people who are more likely to be cigarette smokers as well(26). However, after adjusting for confounding factors, including cigarette smoking, alcohol use, and diet, the results remained significant.
There are many reasons to suggest a causal relation between O&D use and LC. The first is the strong odds ratios observed in this study and the fact that O&D use increased the odds of LC over 7 times. Second, consistent with this study, other studies conducted in Iran have also shown a relation between O&D use and LC. A study conducted in Tehran, Iran, showed that O&D use increased the odds of LC over 7 times and consistent with this study, a dose-response relation between LC and opium was reported (14).
However, opioids may also be used to relieve cancer pain. Therefore, in order to prevent reverse causality, the history of O&D use from two years before the diagnosis of cancer was investigated.
Many mechanisms have been proposed about the carcinogenicity of O&D. Studies have shown that O&D and its alkaloids, including morphine have mutagenic effects (31). Empirical studies have shown that pyrolysed opium has mutagenic effects on Salmonella strains (32). Also, pyrolysates and morphine alkaloids have led to sister chromatid exchange in human lymphocytes and morphological changes in cultured Syrian hamster embryo cells (33). They have also caused carcinogenic changes after being injected subcutaneously, inside the trachea, or into the gastrointestinal system of rats (32). It has also been shown that aromatic hydrocarbons released from burning opioids, indirectly lead to DNA damage and, as a result, may stimulate mutagenic mechanisms (34). However, the carcinogenic mechanisms of opium has not been thoroughly identified yet and further studies are required.
It is worth to mention that many chemicals are added to opioids during their processing, which may have carcinogenic effects. One of these chemicals is lead, which is added in order to increase the weight of the product, and let drug dealers make more money. In the studies conducted on opioids and addicted people in Iran, the amount of lead in the opioids and blood samples of addicts was much higher than normal and at levels that could cause severe health effects(35). Studies have investigated the relation between occupational exposure to lead and the risk of lung cancer, it was shown that exposure to lead increased the risk of lung cancer by about three-fold and a dose-response relationship was observed(36).
There was no relationship between alcohol consumption and lung cancer in this study.In Iran, since the vast majority of people are Muslims, the prevalence of alcohol consumption is much lower than other parts of the world (37). Although, in a few previous studies, people who were heavy alcohol users had a relatively low risk of developing lung cancer(38), but here was no heavy alcohol user in this study.
Opioids use has a high prevalence in some parts of Iran and in Kerman province. There is a probability that a significant proportion of lung cancers in Iran may be related to opioids use. Most of the Bradford Hill criteria about the causal relation between LC and O&D use have been met; and temporality, a dose-response relation, consistency, and a strong association, can be seen in this relation.
This study had some limitations. There is a probability of recall, interviewer, or reporting bias. Under-reporting of opium use, especially in controls, was possible, although by standardizing the interview process and employing a trained interviewer to conduct all interviews, researchers tried to minimize this bias.
A strength of this study was using population-based neighbor controls rather than hospital controls, which helped partially control the confounding effects of socio-economic and environmental factors (39). The other strength of this study was the use of a well-designed questionnaire for measuring O&D use that its validity and reliability had been confirmed in previous studies (15).