Is There an Association Between Benzodiazepine Use and Cancer?
Introduction
Since 1959, when chlordiazepoxide was introduced, benzodiazepines (BZDs) have become widely known and prescribed to hundreds of millions of people. Because of their high therapeutic index, the availability of flumazenil as an antagonist in cases of BZD overdose, and their rapid onset of action, this class of drugs has become difficult to replace in everyday medical practice. Depending on the country, the prevalence of BZD use ranges from 10% to 40% among older people. BZDs are routinely prescribed for sleep and anxiety disorders and in combination with other psychotropic medications (1).
However, soon after the introduction of BZDs, reports emerged about their potential to cause dependence, as well as isolated reports suggesting a possible association between BZD use and cancer. Associations between specific BZDs and the development of cancer have been analyzed using medical databases. In presenting their conclusions, researchers divided BZDs into two groups: “safe” and “unsafe.” Medazepam, diazepam, chlordiazepoxide, nitrazepam, and oxazepam were classified as “safe.” Clonazepam, bromazepam, alprazolam, lorazepam, zopiclone, and zolpidem were classified as “unsafe,” meaning that they were associated with a higher risk of cancer. This article briefly reviews the conflicting results of scientific studies on BZD-induced carcinogenesis.
Studies That Found an Association Between BZD Use and Cancer Development
Using data from Taiwan’s National Health Insurance database from 1996–2000, a study was conducted to investigate possible associations between BZD use and cancer risk. Two groups matched for age and sex were formed: a study group consisting of 59,647 patients with a mean age of 47.9 years who used BZDs, and a control group consisting of 59,592 individuals with a mean age of 46.4 years who did not use BZDs. Each participant was followed until a cancer diagnosis was established or the participant was lost to follow-up because of insufficient data or death. Statistical analysis showed that the overall risk of cancer was 19% higher in the BZD group than in the control group (3).
Another study analyzed associations between individual BZDs and the development of cancer using a medical database. In presenting their conclusions, the researchers divided BZDs into two groups: “safe” and “unsafe.” Medazepam, diazepam, chlordiazepoxide, nitrazepam, and oxazepam were classified as “safe.” Clonazepam, bromazepam, alprazolam, lorazepam, zopiclone, and zolpidem were classified as “unsafe,” meaning that they were associated with a higher risk of cancer. More information about “safe” and “unsafe” BZDs is presented in Table 1.
In addition, an association was observed between BZD use and cancers at certain sites compared with non-use of BZDs: brain tumors occurred 98% more frequently, colon cancer 25% more frequently, and lung cancer 10% more frequently. It was also noted that long-term BZD use may increase the risk of cancer (2).
Table 1. “Safe” and “Unsafe” BZDs (2)
| Number of Cases (42,500 Deaths) / Healthy Controls | Control Group (255,000 Deaths) / Healthy Controls | Odds Ratio (95% CI) | P Value | *Adjusted Odds Ratio (95% CI) | P Value | |
|---|---|---|---|---|---|---|
| “Safe” BZDs | ||||||
| Chlordiazepoxide | 165/42,335 | 761/254,239 | 1.32 (1.11–1.56) | 0.001 | 1.16 (0.97–1.37) | 0.097 |
| Diazepam | 1,273/41,227 | 6,883/248,117 | 1.11 (1.04–1.18) | 0.001 | 1.00 (0.94–1.06) | 0.991 |
| Medazepam | 13/42,487 | 40/254,960 | 2.00 (1.07–3.75) | 0.031 | 1.76 (0.94–3.32) | 0.080 |
| Nitrazepam | 109/42,391 | 620/254,380 | 1.06 (0.87–1.30) | 0.559 | 0.98 (0.80–1.21) | 0.867 |
| Oxazepam | 65/42,435 | 353/254,647 | 1.12 (0.86–1.46) | 0.414 | 0.99 (0.76–1.29) | 0.944 |
| “Unsafe” BZDs | ||||||
| Alprazolam | 2,566/39,934 | 11,860/243,140 | 1.32 (1.26–1.38) | <0.0001 | 1.18 (1.13–1.24) | <0.0001 |
| Bromazepam | 876/41,624 | 4,054/250,946 | 1.30 (1.20–1.40) | <0.0001 | 1.16 (1.08–1.25) | <0.0001 |
| Clonazepam | 966/41,534 | 4,408/250,592 | 1.32 (1.23–1.42) | <0.0001 | 1.22 (1.13–1.31) | <0.0001 |
| Lorazepam | 2,426/40,074 | 11,189/243,811 | 1.32 (1.26–1.38) | <0.0001 | 1.18 (1.13–1.24) | <0.0001 |
| Zopiclone | 558/41,942 | 2,640/252,360 | 1.27 (1.16–1.39) | <0.0001 | 1.14 (1.04–1.25) | 0.006 |
| Zolpidem | 2,053/40,447 | 10,074/244,926 | 1.23 (1.18–1.30) | <0.0001 | 1.13 (1.07–1.18) | <0.0001 |
Adjusted odds ratio – adjusted for concomitant diseases, medications used, and socioeconomic status.
Another scientific study conducted in Taiwan aimed to evaluate the association between BZD use and the risk of brain cancer. The study used information collected in the National Medical Database between 2000 and 2009. Brain cancer was found to occur 3.33 times more frequently among BZD users than among non-users. More cancer cases were detected among people aged 50–59 years. When the effect of dose was examined, an annual dose greater than 36 mg was found to substantially increase tumor risk among patients with anxiety (4).
Several studies investigated the association between the use of non-benzodiazepine hypnotics and cancer. Some authors reported that the probability of cancer was 35% higher among people using hypnotics (5). This view was also supported by other authors, who reported that high doses of zolpidem, more than 300 mg per year, increased cancer risk 2.38-fold. The association between zolpidem and the development of cancer at specific sites is presented in Table 2 (6).
Danish Study: BZDs Do Not Affect the Risk of Cancer
Following publication of the previously mentioned study reporting that BZDs increased the risk of cancer development by as much as 35%, another study using a similar methodology was conducted in Denmark. The results led to a very different conclusion: long-term BZD use or use at high doses was not associated with the development of cancer. The exception involved cases in which other factors with carcinogenic effects, such as smoking, were present (7). Comparisons between the study and control groups are presented in Table 3.
Table 2. Increased Risk of Cancer Associated With Zolpidem Use (Fold Increase) (6)
| Cancer Site | Overall Increase in Cancer Risk | Increase in Cancer Risk in Men | Increase in Cancer Risk in Women |
|---|---|---|---|
| Oral cavity | 2.36 | 1.89 | 1.70 |
| Stomach | 1.95 | 2.11 | 1.91 |
| Pancreas | 1.28 | 1.22 | 1.32 |
| Liver | 1.81 | 1.42 | 2.12 |
| Colon | 1.04 | 1.13 | 0.96 |
| Lung | 1.64 | 1.31 | 1.85 |
| Breast (women) | 1.84 | — | — |
| Uterus | 1.62 | 1.62 | — |
| Prostate | 1.39 | — | 1.39 |
| Endometrium | 1.20 | 1.20 | — |
| Urinary bladder | 1.60 | 1.66 | 1.54 |
| Other | 2.18 | 2.54 | 1.68 |
Significance of Peripheral-Type BZD Receptors
The peripheral-type benzodiazepine receptor (PBR) is an 18 kDa protein that is abundant in tissues synthesizing steroid hormones, including the gonads, adrenal glands, placenta, and brain. PBR was identified because of its ability to bind diazepam in peripheral tissues without affecting central BZD receptors (8). PBR is also known as the translocator protein because it is located in the outer mitochondrial membrane and is one of the major transporters of cholesterol into mitochondria. Thus, PBR plays an important role in cholesterol synthesis (9).
PBR dysregulation is associated with increased proliferation of cancer cells. For example, one study investigated the influence of PBR on human breast cancer cells. Under laboratory conditions, two cell cultures were isolated: the non-aggressive MCF-7 cell line, characterized by very low PBR expression, and the highly aggressive MDA-MB-231 cell line, which demonstrated substantially higher PBR expression. Studies showed that PBR is directly involved in the survival and growth cycle of human breast cancer cells (8).
Table 3. Comparison of the Study and Control Groups (7)
| Parameter | Cancer Patients (Study Group) | Healthy Participants (Control Group) |
|---|---|---|
| Men | 52.8% | 52.8% |
| Women | 47.2% | 47.2% |
| Mean age | 65 (56–74) | 65 (56–74) |
| All BZDs combined | ||
| Never used | 79% | 80.2% |
| Ever used | 21.0% | 19.8% |
| Long-term use (interval 1–5 years) | 1.7% | 1.5% |
| BZDs (ATC groups N05BA and N05CD) | ||
| Never used | 85.9% | 86.8% |
| Ever used | 14.1% | 13.2% |
| Long-term use (interval 1–5 years) | 0.7% | 0.5% |
| Z-drugs (ATC group N05CF) | ||
| Never used | 88.9% | 89.4% |
| Ever used | 11.1% | 10.6% |
| Long-term use (interval 1–5 years) | 1.0% | 0.9% |
Hypotheses Attempting to Explain the Influence of BZDs on Carcinogenesis
Several hypotheses may explain how BZDs could be associated with the risk of cancer:
- People reporting anxiety and sleep disorders may be more likely to undergo neuroimaging, leading to the incidental detection of an already existing tumor.
- Slowly growing benign tumors often present with episodes of dizziness, sleep disturbances, and other conditions for which BZDs may be prescribed. In such cases, BZDs may simply “mask” the symptoms of an undiagnosed tumor.
- BZDs increase gamma-aminobutyric acid neurotransmission through GABA chloride ion channels. GABA is not only a neurotransmitter but also influences cell growth and differentiation. This may therefore be associated with tumor growth through receptor-mediated mechanisms (4).
Summary
Possible associations between BZD use and an increased risk of cancer have recently been the subject of discussion. At present, the influence of BZDs and hypnotics on the development of cancer can neither be unequivocally confirmed nor ruled out. When describing the limitations of the studies, the authors note that other risk factors, such as smoking, heredity, and previous diseases, which may influence the development of certain cancers, were not comprehensively evaluated. In addition, BZDs are among the most widely used medications, and suddenly prohibiting their use because of a presumed carcinogenic effect would create numerous medical and ethical problems.
It is logical that the first symptoms of cancer are often general, nonspecific complaints such as anxiety, insomnia, irritability, and similar symptoms. In such cases, one or another BZD may be used as first-line medication. Therefore, when a tumor is diagnosed in a patient taking BZDs, it cannot be firmly concluded that BZD use contributed to this process.
PBR represents a target for future research because its role is not entirely clear. It is evident that these receptors are involved in one way or another in the cycles of cell division and growth. Identifying these molecular mechanisms may provide a new direction for the treatment of cancer.
Thus, the association between BZDs and cancer is an area requiring thorough and long-term research and may have implications for everyday clinical practice. Physicians should be aware that some BZDs may increase the risk of cancer, particularly in the presence of comorbidities or risk factors with established carcinogenic effects.
Source: Internistas, 2016
