Conducting studies on the possible association between CCBs and cancer using a pharmaco-epidemiological approach based on data collected in existing databases and cancer registries is challenging. The challenge lies in working within the limitations of the available data (in terms of quality and completeness) and simultaneously maximizing the value of the available data through thoughtful study design and statistical analysis.
CCBs represent a chemically and pharmacologically diverse group of agents that are widely used for the treatment of hypertension and angina. It has been proposed that CCBs may interfere with apoptosis, leading to an increased potential for abnormal cell proliferation and tumor growth. The underlying biological mechanism for this effect is thought to be linked to the role of transmembrane Ca2+. This hypothesis has been critically reviewed and results have shown that the action of CCBs on apoptosis are complex with both increases and decreases in intracellular Ca2+ linked to this form of programmed cell death (2). CCBs have also been shown to inhibit apoptosis in certain non-transformed cell lines but promote apoptosis in other non-transformed and transformed cell lines. The results from non-human genotoxicity studies have shown no link between CCB use and tumor development
Epidemiologic studies have also provided inconsistent results. While only a few follow-up analyses reported an increased risk for all cancer or breast cancer, further observational studies have so far provided no evidence to support the hypothesis that long-term use of CCBs might be carcinogenic. As a whole, these studies have been limited by lack of statistical power and/or inadequate methods for defining the exposure window of antihypertensive treatment in relation to the index date (cancer outcome), making the establishment of a causal relationship between CCBs use and risk of cancer problematical. Studies often assumed a relatively short period of CCB use (usually between 2 months and 1 year) before entering the study as users. In other cases, information on the use of CCBs was only available at study entry or during follow-up. Overall, most studies were limited by follow-up periods that could be considered too brief to measure a carcinogenic effect. The vast majority of studies collected information from electronic medical or administrative databases.
The current study aims to learn from the experience, strengths and limitations of previous research to present a best-practice approach to addressing the hypothesis in question. Combined with the objectives of the PROTECT program, we hope that this study will help to provide a framework for guiding methodological choices in future research and contribute to increasing the usefulness and reliability of pharmacoepidemiological studies for benefit-risk assessment and decision making.
