White blood cell count and renin-angiotensin system inhibitors for the risk of cancer in type 2 diabetes

Bayesian estimation of the measurement of interactions in epidemiological studies

Background

Interaction identification is important in epidemiological studies and can be detected by including a product term in the model. However, as Rothman noted, a product term in exponential models may be regarded as multiplicative rather than additive to better reflect biological interactions. Currently, the additive interaction is largely measured by the relative excess risk due to interaction (RERI), the attributable proportion due to interaction (AP), and the synergy index (S), and confidence intervals are developed via frequentist approaches. However, few studies have focused on the same issue from a Bayesian perspective. The present study aims to provide a Bayesian view of the estimation and credible intervals of the additive interaction measures.

Methods

Bayesian logistic regression was employed, and estimates and credible intervals were calculated from posterior samples of the RERI, AP and S. Since Bayesian inference depends only on posterior samples, it is very easy to apply this method to preventive factors. The validity of the proposed method was verified by comparing the Bayesian method with the delta and bootstrap approaches in simulation studies with example data.

Results

In all the simulation studies, the Bayesian estimates were very close to the corresponding true values. Due to the skewness of the interaction measures, compared with the confidence intervals of the delta method, the credible intervals of the Bayesian approach were more balanced and matched the nominal 95% level. Compared with the bootstrap method, the Bayesian method appeared to be a competitive alternative and fared better when small sample sizes were used.

Conclusions

The proposed Bayesian method is a competitive alternative to other methods. This approach can assist epidemiologists in detecting additive-scale interactions.

Systematic review and meta-analysis of randomised controlled clinical trial evidence refutes relationship between pharmacotherapy with angiotensin-receptor blockers and an increased risk of cancer

AIMS

The potential influence of angiotensin-receptor blockers (ARBs) on carcinogenesis is a much-debated topic. Both observational, as well as preclinical studies in rodent carcinogenic assays, suggest a major role of the Renin-Angiotensin-Aldosterone-System (RAAS) in cancer development. Therefore, a systematic review and meta-analysis with available study data on ARBs and carcinogenicity in general as primary outcome were conducted. Secondary outcomes were defined as tumour-specific mortality rates and the frequency of new cases of specific tumour types with particular emphasis on lung, breast, and prostate cancer.

METHODS

A systematic literature research was performed in MEDLINE, EMBASE, Cochrane Library, and TOXLINE. We used a combination of MeSH terms, keywords and substance names of ARBs and searched between 1950 and 2016. At least 100 participants in each study arm and a minimum follow-up for one year were necessary for study inclusion. Odds ratios (OR) were calculated by a random-effects model.

RESULTS

A total of 8818 potentially eligible publications were identified of whom seven randomised controlled trials, four case-control studies and one cohort study met our inclusion criteria. As a key result, we found no effect on carcinogenesis in randomised controlled trials for ARB usage. (OR 1.02, 95% CI 0.87-1.19; p = .803). Conflicting results with observational studies could be explained by poor reporting- and study qualities.

CONCLUSIONS

The results of our meta-analysis focusing only on high evidence levels and study designs (RCTs) did not reveal any relationship between pharmacotherapy with an ARB and an increased risk for cancer in general.

Autoimmunity and dysmetabolism of human acquired immunodeficiency syndrome

Acquired immunodeficiency syndrome (AIDS) remains ill-defined by lists of symptoms, infections, tumors, and disorders in metabolism and immunity. Low CD4 cell count, severe loss of body weight, pneumocystis pneumonia, and Kaposi's sarcoma are the major disease indicators. Lines of evidence indicate that patients living with AIDS have both immunodeficiency and autoimmunity. Immunodeficiency is attributed to deficits in the skin- and mucosa-defined innate immunity, CD4 T cells and regulatory T cells, presumably relating human immunodeficiency virus (HIV) infection. The autoimmunity in AIDS is evident by: (1) overproduction of autoantibodies, (2) impaired response of CD4 cells and CD8 cells, (3) failure of clinical trials of HIV vaccines, and (4) therapeutic benefits of immunosuppression following solid organ transplantation and bone marrow transplantation in patients at risk of AIDS. Autoantibodies are generated in response to antigens such as debris and molecules de novo released from dead cells, infectious agents, and catabolic events. Disturbances in metabolic homeostasis occur at the interface of immunodeficiency and autoimmunity in the development of AIDS. Optimal treatments favor therapeutics targeting on the regulation of metabolism to restore immune homeostasis.

Drug-subphenotype interactions for cancer in type 2 diabetes mellitus

On the basis of data obtained from a prospective cohort of Chinese patients with type 2 diabetes mellitus (T2DM), we discuss cancer subphenotypes (risk factors) in patients with T2DM, which can lead to drug-cancer subphenotype interactions. These subphenotypes include HDL cholesterol levels <1.0 mmol/l, co-occurrence of LDL cholesterol levels <2.8 mmol/l and triglyceride levels <1.7 mmol/l, and co-occurrence of LDL cholesterol levels <2.8 mmol/l and albuminuria. The increased risk of cancer associated with low levels of HDL cholesterol, low LDL cholesterol levels plus low triglyceride levels, and low levels of LDL cholesterol plus albuminuria can be reduced by treatment with metformin, renin-angiotensin system (RAS) inhibitors and statins, respectively. Mechanistic studies support the hypothesis that dysregulation of the 5'-AMP-activated protein kinase pathway and crosstalk between the RAS and insulin-like growth factor 1-cholesterol pathways create a cancer-promoting milieu in patients with T2DM. These findings highlight that in Chinese individuals, multiple pathways are implicated in the link between T2DM and cancer, which can generate multiple subphenotypes as well as drug-subphenotype interactions.

Lower aldosterone-renin ratio is a risk factor for total and cancer death in Japanese individuals: the Takahata study

OBJECTIVE

A higher plasma aldosterone-renin ratio (ARR) is an established marker for screening for primary aldosteronism (PA). The association between higher ARR and mortality in a general population has not been fully explored. We here examined whether higher ARR is a risk factor for total and cause-specific mortality in a Japanese population.

SUBJECTS AND METHODS

A population-based, longitudinal study of 1,310 Japanese individuals (age: 63·9 ± 9·8 years) enrolled in the Takahata study between 2004 and 2006 and followed for up to 8 years. The incidence and causes of death were monitored annually until 10 January 2012 (median follow-up: 2691 days).

RESULTS

During the follow-up period, 64 subjects died. Kaplan-Meier analysis showed a significantly increased risk for total and cancer mortality in subjects with lower ARR (log-rank P < 0·001). Cox's proportional hazard model analyses with adjustment for age and gender showed that lower ARR was associated with increased total and cancer mortality in subjects with low (≦72) vs high (>72) ARR (hazard ratios and 95% confidential intervals: 2·56, 1·44-4·56 and 2·78, 1·16-6·65, respectively).

CONCLUSIONS

Lower ARR was a significant and independent risk factor for increased total and cancer mortality in this Japanese population. Subjects with higher ARR were not-at-risk for total death in general. These findings increase the necessity for identifying people with PA from those with higher ARR. People with higher ARR without PA may be at very low risk for total and cancer death.

Enhancers and attenuators of risk associations of chronic hepatitis B virus infection with hepatocellular carcinoma in type 2 diabetes

Chronic hepatitis B virus (HBV) infection promotes hepatocellular carcinoma (HCC) risk. In type 2 diabetes (T2D), use of insulin and statins was associated with reduced cancer risk while co-presence of low LDL cholesterol (LDLC <2.8 mmol/l) plus low triglyceride (TG; <1.7 mmol/l) increased cancer risk. There is experimental evidence showing that insulin insufficiency might promote HCC. In this study, we examined whether this lipid subphenotype and use of insulin or statins might modify the promoting effect of chronic HBV infection (indicated by the presence of hepatitis B surface antigen) on HCC. We analyzed data of 1319 T2D patients enrolled into the Hong Kong Diabetes Registry from December 1996 to January 2005 and followed up to 2005. Additive interaction was estimated using relative excess risk due to interaction and attributable proportion due to interaction. During 5782 person-years of follow-up, 1.74% (n=23) of patients developed HCC (incidence, 3.98; 95% confidence interval, 2.36-5.60/1000 person-years). HbA1c ≥7.0% and the lipid phenotype (LDLC <2.8 mmol/l plus TG <1.7 mmol/l) increased the hazard ratios (HRs) of chronic HBV infection for HCC from 3.74 to 74.96 and from 11.01 to 89.82 respectively with significant interactions. Use of insulin or statins decreased the HRs from 37.51 to 5.87 and from 64.94 to 16.99 respectively with significant interactions (all P values <0.05). These findings support our hypothesis that hyperglycemia and co-presence of low LDLC plus low TG might enhance, while insulin or statin usage might attenuate the promoting effect of chronic HBV infection on HCC in T2D.

Synergistic effects of low LDL cholesterol with other factors for the risk of cancer in type 2 diabetes: the Hong Kong Diabetes Registry

We have reported associations of cancer with low triglyceride and high high-density lipoprotein cholesterol (HDL-C) as well as co-presence of low low-density lipoprotein cholesterol (LDL-C) and albuminuria in type 2 diabetes (T2D). This analysis aims to test (1) whether low LDL-C and low triglyceride have synergistic effects to increase cancer risk in T2D and (2) whether high HDL-C enhances the effect of co-presence of low LDL-C and albuminuria on cancer risk. A prospective cohort of patients with T2D, established within the Prince of Wales Hospital, was used in the analysis. A total of 3,476 T2D patients in Hong Kong enrolled between 1996 and 2005, free of cancer at enrolment and not using statins or fibrates within 2.5 years before enrolment and during follow-up, were followed until 2005. The study measured additive interactions of low LDL-C with other factors for cancer using relative excess risk due to interaction (RERI) and attributable proportion due to interaction (AP). A statistically significant RERI > 0 or AP > 0 indicates additive interaction. During 5.11 years of follow-up, 199 patients developed cancer. Co-presence of triglyceride <1.70 mmol/L and LDL-C < 2.80 mmol/L was associated with increased cancer risk (multivariable hazard ratio [HR]:2.13, P = 0.0008) with significant interaction. Co-presence of HDL-C ≥ 1.30 mmol/L and LDL-C < 2.80 mmol/L plus albuminuria was also associated with increased cancer risk (HR: 3.84, P < 0.0001) with significant interaction. In T2D, low triglyceride may potentiate cancer risk associated with low LDL-C while high HDL-C enhances the synergistic effect of low LDL-C with albuminuria towards increased cancer risk.

Diabetes and cancer: the mechanistic implications of epidemiological analyses from the Hong Kong Diabetes Registry

Diabetes is a disorder of energy metabolism associated with increased cancer risk, but the underlying mechanism is poorly understood. In a prospective cohort of patients enrolled in the Hong Kong Diabetes Registry, we explored risk factors for cancer including drug usage in type 2 diabetes. In a series of published papers, we reported a linear risk association of cancer with glycated haemoglobin with a threshold at 6.0%-6.5% and non-linear risk associations of body mass index, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, triglyceride and white blood cell (WBC) count in V-shaped or A-shaped manners. Detailed pharmacoepidemiological analysis revealed markedly attenuated cancer risk in patients treated with insulin and oral anti-diabetic drugs compared with non-users of these drugs. We further observed significant drug-subphenotype interactions with attenuated cancer risk in metformin users with low high-density lipoprotein cholesterol, renin-angiotensin system (RAS) inhibitor users with high WBC count and statin users with co-presence of low low-density lipoprotein cholesterol plus albuminuria or low triglyceride. These novel observations corroborate with experimental findings of possible consequences of hyperglycaemia on dysregulation of cholesterol metabolism, renin-angiotensin system and adenosine 5'-monophosphate-activated protein kinase pathways, all of which may be implicated in carcinogenesis. On the basis of these epidemiological and experimental findings, we argue for the strong need to strengthen the health care system to ensure that type 2 diabetes subjects receive appropriate drugs to optimize internal milieu to reduce all events including cancer. Apart from mechanistic studies, large-scale, randomized clinical trials using medications such as statin, renin-angiotensin system inhibitors and metformin in patients with risk-conferring subphenotypes are needed to confirm their anti-cancer effects.

Higher plasma renin activity is a risk factor for total mortality in older Japanese individuals: the Takahata study

Plasma renin activity (PRA) is accepted as a marker for increased risk of cardiovascular diseases. However, the association between PRA and total mortality has not been fully explored in a general population. We here examined whether PRA is associated with increased total mortality in a general Japanese population. The participants of the Takahata study (3502 subjects; age, 62.5 ± 10.4 years), a population-based, longitudinal study of Japanese held from 2004 to 2006, were enrolled and followed up for up to 7 years. The incidence of death and causes of death were monitored annually to the end of 2010 (median follow-up, 2280 days). During the follow-up period, 143 subjects died. Kaplan-Meier analysis showed a significantly increased risk for total mortality in subjects with higher PRA (log-rank P < .001). Cox proportional hazard model analyses with adjustment for factors correlated with PRA (age, sex, weight, diastolic blood pressure, high-density lipoprotein cholesterol, uric acid, B-type natriuretic peptide, serum total protein, antihypertensive treatment, and diabetes) showed that higher PRA was associated with increased total mortality in linear regression models (per 1 increase in log 10 × PRA [nanograms per milliliter per hour]: hazard ratio, 2.12; 95% confidence interval, 1.47-3.06), between groups of patients stratified by quartiles of PRA (highest vs lowest quartile: 2.63, 1.57-4.41) and in subjects with high (≥ 2.0 ng/[mL h]) vs low (<2.0 ng/[mL h]) PRA (1.97, 1.37-2.83). Higher PRA was a significant and independent risk factor for increased total mortality in this Japanese population and may be a marker for subjects at an increased risk of total mortality.

Use of angiotensin-converting-enzyme inhibitors or angiotensin-receptor blockers and cancer risk: a meta-analysis of observational studies

BACKGROUND

Epidemiologic studies have reported inconsistent findings regarding the association between the use of angiotensin-converting-enzyme (ACE) inhibitors or angiotensin-receptor blockers and the risk of cancer. We performed a meta-analysis of observational studies to assess the association.

METHODS

We searched MEDLINE, EMBASE and the Cochrane Library to identify studies through January 2011. Two evaluators independently reviewed and selected articles of cohort and case-control studies on the basis of predetermined selection criteria.

RESULTS

Of 3970 screened articles, 12 cohort studies and 16 case-control studies were selected for analysis. We found no significant association between the use of ACE inhibitors or angiotensin-receptor blockers and the overall risk of cancer (relative risk [RR] 0.96, 95% confidence interval [CI] 0.90-1.03). We found a decreased risk of cancer associated with use of either medication when we restricted the analyses to cohort and nested case-control studies (RR 0.90, 95% CI 0.83-0.97) or to studies with long-term follow-up of more than five years (RR 0.89, 95% CI 0.83-0.96). In the subgroup meta-analyses by cancer site, a decreased risk was identified for esophageal cancer, whereas an increased risk was found for melanoma and kidney cancer.

INTERPRETATION

No significant association was found between the use of ACE inhibitors or angiotensin-receptor blockers and overall risk of cancer. A possible beneficial effect associated with use of either medication was suggested in sensitivity analyses, including those of studies with long-term follow-up. Large randomized controlled trials with long-term follow-up are needed to specifically test the effect of each of these medications on the risk of cancer.

The Complexity of Vascular and Non-Vascular Complications of Diabetes: The Hong Kong Diabetes Registry

Diabetes is a complex disease characterized by chronic hyperglycemia and multiple phenotypes. In 1995, we used a doctor-nurse-clerk team and structured protocol to establish the Hong Kong Diabetes Registry in a quality improvement program. By 2009, we had accrued 2616 clinical events in 9588 Chinese type 2 diabetic patients with a follow-up duration of 6 years. The detailed phenotypes at enrollment and follow-up medications have allowed us to develop a series of risk equations to predict multiple endpoints with high sensitivity and specificity. In this prospective database, we were able to validate findings from clinical trials in real practice, confirm close links between cardiovascular and renal disease, and demonstrate the emerging importance of cancer as a leading cause of death. In addition to serving as a tool for risk stratification and quality assurance, ongoing data analysis of the registry also reveals secular changes in disease patterns and identifies unmet needs.

Estimating measures of interaction on an additive scale for preventive exposures

Measures of interaction on an additive scale (relative excess risk due to interaction [RERI], attributable proportion [AP], synergy index [S]), were developed for risk factors rather than preventive factors. It has been suggested that preventive factors should be recoded to risk factors before calculating these measures. We aimed to show that these measures are problematic with preventive factors prior to recoding, and to clarify the recoding method to be used to circumvent these problems. Recoding of preventive factors should be done such that the stratum with the lowest risk becomes the reference category when both factors are considered jointly (rather than one at a time). We used data from a case-control study on the interaction between ACE inhibitors and the ACE gene on incident diabetes. Use of ACE inhibitors was a preventive factor and DD ACE genotype was a risk factor. Before recoding, the RERI, AP and S showed inconsistent results (RERI = 0.26 [95%CI: -0.30; 0.82], AP = 0.30 [95%CI: -0.28; 0.88], S = 0.35 [95%CI: 0.02; 7.38]), with the first two measures suggesting positive interaction and the third negative interaction. After recoding the use of ACE inhibitors, they showed consistent results (RERI = -0.37 [95%CI: -1.23; 0.49], AP = -0.29 [95%CI: -0.98; 0.40], S = 0.43 [95%CI: 0.07; 2.60]), all indicating negative interaction. Preventive factors should not be used to calculate measures of interaction on an additive scale without recoding.