The Association Between Antihypertensive Drug Use and Risk of Cancer Death in Finland

Antihypertensive drugs (anti-HT drugs) are very commonly used drugs. They include several different drug molecules with different mechanisms of action. Based on their mechanism of action, they can be categorized into five main groups: angiotensin converting enzyme (ACE) inhibitors, angiotensin-receptor (ATR) blockers, beta-blockers, calcium-channel blockers, and diuretics. These drugs are used in also other conditions than hypertension such as in the treatment of coronary artery disease and heart failure; the loop diuretic compound, furosemide, is mainly used in the treatment of oedema.

Since antihypertensive drugs are so widely prescribed, it would be important to know whether they would affect also other diseases; one can speculate that the mechanism of action of some antihypertensive drug groups may have relevance in non-cardiovascular diseases e.g., cancer. Although there are some indications that the drugs affecting the renin-angiotensin aldosterone (RAA) system (ACE-inhibitors and ATR- blockers) might affect cancer cell growth and improve prognosis of some cancer types, like prostate cancer, the published results have been conflicting.

The purpose of this thesis is to evaluate the association between antihypertensive drug use in general as well as subdivided according to the drugs’ mechanism of action and the risk of cancer death from prostate (PCa), urothelial (UC), breast (BCa) and ovarian (OC) cancer. The study was conducted by obtaining the information on cancer diagnoses from The Finnish Cancer Registry during 1995-2013 and combining this with information on drug purchases from the national prescription database. Drug use was evaluated separately before and after cancer diagnosis as well as the amount of use by calculating drug-specific Defined Daily Dose (DDD) for each participant. Cox regression was used to calculate hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the risk of cancer death after the diagnosis of each cancer type.

Anti-HT drug use before diagnosis was analysed as a time-fixed variable taking into account the duration and cumulative dose of use occurring between 1995 and the year of cancer diagnosis. Post-diagnostic use was analyzed as a time-dependent variable to minimize immortal time bias. Time-dependent variables were formed by updating medication user status as well as estimating the cumulative amount, duration, and intensity of use separately for each follow-up year after the cancer diagnosis. Dose dependence was evaluated by stratifying medication users by tertiles of DDD amount, duration and intensity of use based on the level reached on each follow-up year.

The simultaneous use of multiple anti-HT drug groups was modelled by forming separate time-dependent variables for use of each drug group with these variables included in a Cox- regression model. The long- term association between anti-HT drug use and the risk of cancer death was investigated in lag time analyses where the exposure was lagged forward in the follow-up time and analyzing medication use that occurred before that event. Analyses were adjusted for several confounders such as primary treatment, tumor extent, comorbidities, and simultaneous use of other drugs like statins and antidiabetic medication.

In prostate cancer, the use of ATR-blockers was associated with a decreased risk of cancer death. The risk decrease was dose-dependent and concerned usage both before and after diagnosis. The risk decrease remained for five years after usage. In contrast, use of beta-blockers was associated with an increased PCa death risk with both pre-and post-diagnostic use.

Similarly, the post-diagnostic use of ATR-blockers was dose-dependently associated with a reduced risk of cancer death also in bladder (BC) cancer. In BCa, ATR-blockers but also beta-blockers and calcium-channel blockers displayed a dose-dependent association with a reduced BCa death risk in post-diagnostic use. In ovarian cancer, ACE-inhibitors were associated with a decreased cancer death risk but only with very long follow-up periods exceeding 10 years.

ATR-blockers, unlike any other anti-HT drug group, were associated with decreased risk of cancer death in PCa, BC and BCa and furthermore in these three cancer types, the risk decrease was dose-dependent. This indicates that the association was causal. RAA- inhibition may confer benefits in preventing the progression of cancer. However, similar results were not observed for ACE-inhibitors even though they also inhibit RAA-system but in a different way. The role of the RAA- system in cancer progression warrants further study.

In BCa, decreased risk associations were observed for multiple anti-HT drug groups after cancer diagnosis. This suggests that the control of hypertension may be more important in this cancer type than any given mechanism of action. However, ATR-blockers were the only drug group also in this cancer type in which the pre-diagnostic use was associated with a decreased BCa death risk and furthermore in a dose-dependent manner.

In conclusion, the use of ATR-blockers is associated with improved cancer-specific survival in multiple cancer types. ATR- blockers differ in this regard from all other anti-HT drug groups, even ACE-inhibitors. Our studies suggest a possible prognostic role of ATR- inhibition in cancer. It would be enlightening to clarify the underlying biological mechanisms to explain why the blockade of the ATR- receptor can exert this potential anti-cancer effect.