Hypertension and the risk of cancer: is there new evidence?

The idea that hypertension is in some way linked to cancer is under controversial discussion. A recent review of the issue revealed that, indeed, raised blood pressure per se or its treatment may confer a low-grade risk for several malignancies [¹]. In 1960, de Waard et al. were the first to suggest that there might be an association between hypertension and breast cancer, speculating that certain hormonal mechanisms were responsible for both conditions [²]. Subsequently, three retrospective studies indicated a possible association between the use of reserpine, the antihypertensive drug of first choice in those days, and breast cancer [^3–5]. However, strong clinical and scientific interest in a possible association between raised blood pressure and cancer was not aroused until after the first prospective study carried out by Dyer et al. [⁶] some 25 years ago. Among a cohort of white males followed for 14 years, these authors found that both systolic and diastolic blood pressures were significantly associated with subsequent mortality from cancer of the lung, colon and all other sites combined, even after adjustment for age, cholesterol and smoking. A careful examination of the possibility that antihypertensive medication (in particular reserpine) was responsible for the excess mortality indicated no evidence that this was the case. Subsequently, several prospective studies, most of which have been confined to men, have addressed the issue of hypertension and total cancer mortality (rather than incidence), and have provided further evidence indicating the possibility that hypertension in itself is a significant risk factor for the development of cancer [^7–17]. However, many uncertainties regarding a possible connection between hypertension and cancer still exist, mainly concerning cancer site specificity, sex, age and duration of the disease, and also complex interactions with other factors, including smoking, obesity, diabetes, alcohol and diet [¹]. With respect to cancer site specificity, the organ most frequently related to pre-existing hypertension, even in long-term prospective studies, is the kidney [^12,18,19], but positive relationships with blood pressure were also reported for cancer of the colon [^6,9], lung [^6,15] and even the endometrium [^20,21]. It has been suggested that some classes of antihypertensive agents may be carcinogenic and be associated with increased cancer risk, particularly renal cancer [^13,19,22]. A recent review by Grossmann et al. [²³] revealed that, with the exception of diuretics and renal cell carcinoma (risk ratio 1.54), the association between antihypertensive drugs and malignancy was either low-grade (rauwolfia), uncertain (atenolol), absent (angiotensin-converting enzyme inhibitors), or absent with a yet to be investigated inverse association (calcium antagonists). The reported association between diuretic use and renal cell carcinoma remains problematic. Therefore, ongoing and future studies assessing the outcome of antihypertensive therapy should pay attention to the effects of these agents on extra-cardiovascular morbidity, and mortality such as malignancies [²³]. As to a possible carcinogenic effect of calcium antagonists [^24–26], the results obtained from large and methodologically robust studies failed to provide support for the hypothesis that the use of calcium antagonists is associated with an increased susceptibility for cancer incidence, and the existing data make it unlikely that calcium antagonists increase the risk of any malignancy; for a review, see [^23,27,28].

In this issue of the journal, Lee et al. [²⁹] make an important observation that provides support for a link between hypertension and the risk of lung cancer. The authors present the results of an analysis of a huge database obtained from a prospective observational study of more than 450 000 Korean men of whom no less than 58.3% were identified as smokers and 20.8% were ex-smokers. During the 5-year follow-up (1995–99), 883 (0.19%) individuals (mean age 45.4 years) died from lung cancer. Hypertension increased the mortality risk of lung cancer by 30%. However, after stratification of smoking status, the risk ratio of hypertension on lung cancer was only increased for smokers (risk ratio 1.4). Interestingly, on introducing interaction terms in the multivariate model used, the authors discovered a significant interactive effect of hypertension with current smoking on lung cancer risk (risk ratio 1.8) but were unable to qualify hypertension per se as an independent risk factor. Lower body mass index, lower serum-cholesterol, no exercise and an increased alcohol consumption (risk ratio up to 1.8) were all identified as risk factors related to cancer mortality in their own right. No significant effect of hypertension treatment status on lung cancer risk was observed, a finding consistent with data obtained in other recent prospective studies [^8–17,28]. Unfortunately, nothing is known about the classes of antihypertensive drugs used in the hypertensive study population. Therefore, the possibility that the increased risk of mortality from lung cancer was attributable in part to some kind of antihypertensive medication cannot be excluded. Because of the small percentage of subjects (1.8%) who had a history of taking drugs for raised blood pressure, it was probably impossible for the authors to trace subjects to obtain incidence rates of lung cancer among those assigned to the various classes of antihypertensive agents compared with those assigned no treatment. A limitation of the study might be the relatively short follow-up period of 5 years. The relationship between blood pressure and cancer may differ with respect to follow-up time. A recent prospective study [¹⁵] revealed an inverse relationship between cancer and blood pressure in the first 5 years of follow-up. However, in the subsequent follow-up of up to 13 years, a significant positive association was seen between systolic blood pressure (but not diastolic blood pressure) and risk of cancer mortality. This positive relationship was only seen in current smokers, a finding in agreement with the conclusion of Lee et al. [²⁹] that hypertension increases susceptibility to lung cancer only in the presence of smoking. It might be possible that, with a longer follow-up period, the relationship between hypertension and lung cancer observed in the study by Lee et al. [²⁹] would have been even stronger and enabled the authors to disentangle raised blood pressure as an independent risk factor. Overall, the huge database and the impressive follow-up record support the conclusions of the authors that smoking interacts with hypertension to influence the risk of mortality from lung cancer. Although the risk was comparatively modest, the findings may be of public health importance for both hypertension and lung cancer are common conditions.

At present, the nature of the relationship between hypertension and cancer is uncertain, and we do not know of any mechanism of hypertension as a risk factor for any malignancy. Genetic predisposition both to hypertension and to cancer might be one possibility [³⁰]. The results of a recent study by Norden et al. [³¹] investigating DNA repair synthesis in lymphocytes exposed to carcinogens indicate that lymphocytes from individuals with elevated diastolic blood pressure show a higher frequency of chromosome aberrations compared to those from individuals with normal or low diastolic pressures. Similarly, studies in spontaneously hypertensive rats have shown that these animals are more sensitive than normotensive rats to chromosome aberrations induced by chemical carcinogens [³²]. It has been suggested that cell-death by apoptosis [^33,34], the process used to ‘eliminate’ cells (e.g. for example, preneoplastic, initiated, or damaged cells [³⁵]) can control the growth of vascular smooth muscle cells, and that significant abnormalities are observed in hypertension and other cardiovascular disorders [^1,35–37]. Calcium plays an important role in apoptosis and, in many cell types, a sustained rise in cytosolic ionized calcium may initiate the apoptotic process [³⁸]. One suggested explanation for the reported association between the use of calcium-blockers and cancer is the putative effect of these agents on apoptosis; for example, their action as a tumour promoter by interfering with the programmed death of DNA damaged cells [^33,35]. However, a comprehensive analysis of the basic and clinical evidence does not support a causal relationship between the therapeutic use of calcium antagonists and an increased incidence of cancer development as a result of interfering with apoptosis [³⁹]. Another mechanism previously discussed is the possibility that hypertension in itself might cause cancer through pathways that include the promotion of growth or dissemination of malignant cells [^11,30], or through the mitogenic effect of several hormones known to play a role in the development of high blood pressure. This is particularly true for vasoactive hormones, including angiotensin II, catecholamines and vasopressin, and for some metabolic hormones, such as insulin, mineralocorticoids and glucocorticoids. Rees [⁴⁰], in a recent review of the biosynthesis of hormones by tumours, revealed that adrenocorticotropic hormone production leading to blood pressure elevation can be found in cancer of the lung, pancreas, thyroid and ovary, and in neuroblastomas. Similarly, an elevation of adrenocortical hormones has been observed in the blood of patients who had cancer at a variety of sites [⁴¹]. These findings raise the interesting question of the direction of causality in any association between hypertension and cancer, and favour the hypothesis that the causal direction is from cancer to hypertension rather than the reverse [¹¹]. Incidence data for both cancer and hypertension are needed to examine the possibility of a causal association in either direction. Another explanation for the reported connection between cancer and hypertension is that the small to modest overall association between raised blood pressure and cancer may have resulted from confounding but often unmeasured factors, such as diet (low-dietary fibre intake and high fat, meat, high salt, low potassium and low magnesium) or obesity, which are all components that are related in some way to both hypertension and specific types of cancer [³⁰]. At present, the possibility that both hypertension and cancer are the result of a common aetiological factor may be plausible but is unproven. Prospective studies are needed to search for such putative factors and to determine their contribution. With respect to previous reports suggesting that some antihypertensive agents are associated with a greater than normal risk of cancer, it is important to recognize that long-term treatment of hypertension is associated with an overall lower risk of mortality for the population because of a lower risk of cardiovascular death [⁴²]. It is tempting to speculate that approaches aimed at preventing the development of hypertension might be important in lowering the total burden of both cardiovascular disease and cancer [³⁰].

In conclusion, many uncertainties remain concerning a possible relationship between hypertension and the risk of malignancies, and unresolved differences mainly persist concerning the interaction with other factors. Lee et al. [²⁹] make a valuable contribution by suggesting that hypertension increases the modest risk of death from lung cancer among smokers. Although the data cannot be considered ‘new evidence', and the dominant aetiological role of smoking for lung cancer is well recognized, they provide additional information to the accumulating database indicating that hypertension is linked in some way to cancer and warrant the search for factors which affect blood pressure, interact with smoking and are potentially carcinogenic.