Mendelian randomization: using genes as instruments for making causal inferences in epidemiology

Observational epidemiological studies suffer from many potential biases, from confounding and from reverse causation, and this limits their ability to robustly identify causal associations. Several high-profile situations exist in which randomized controlled trials of precisely the same intervention that has been examined in observational studies have produced markedly different findings. In other observational sciences, the use of instrumental variable (IV) approaches has been one approach to strengthening causal inferences in non-experimental situations. The use of germline genetic variants that proxy for environmentally modifiable exposures as instruments for these exposures is one form of IV analysis that can be implemented within observational epidemiological studies. The method has been referred to as 'Mendelian randomization', and can be considered as analogous to randomized controlled trials. This paper outlines Mendelian randomization, draws parallels with IV methods, provides examples of implementation of the approach and discusses limitations of the approach and some methods for dealing with these.     

Severity of bias of a simple estimator of the causal odds ratio in Mendelian randomization studies

Mendelian randomization studies estimate causal effects using genetic variants as instruments. Instrumental variable methods are straightforward for linear models, but epidemiologists often use odds ratios to quantify effects. Also, odds ratios are often the quantities reported in meta‐analyses. Many applications of Mendelian randomization dichotomize genotype and estimate the population causal log odds ratio for unit increase in exposure by dividing the genotype‐disease log odds ratio by the difference in mean exposure between genotypes. This ‘Wald‐type’ estimator is biased even in large samples, but whether the magnitude of bias is of practical importance is unclear. We study the large‐sample bias of this estimator in a simple model with a continuous normally distributed exposure, a single unobserved confounder that is not an effect modifier, and interpretable parameters. We focus on parameter values that reflect scenarios in which we apply Mendelian randomization, including realistic values for the degree of confounding and strength of the causal effect. We evaluate this estimator and the causal odds ratio using numerical integration and obtain approximate analytic expressions to check results and gain insight. A small simulation study examines finite sample bias and mild violations of the normality assumption. For our simple data‐generating model, we find that the Wald estimator is asymptotically biased with a bias of around 10% in fairly typical Mendelian randomization scenarios but which can be larger in more extreme situations. Recently developed methods such as structural mean models require fewer untestable assumptions and we recommend their use when the individual‐level data they require are available. The Wald‐type estimator may retain a role as an approximate method for meta‐analysis based on summary data. Copyright © 2012 John Wiley & Sons, Ltd.     

Aetiological factors for oral manifestations of HIV

OBJECTIVES: Describe the oral diseases in HIV-infected individuals in London, UK and identify social and medical factors related to the presence of specific oral diseases.
DESIGN: Cross-sectional analytic study.
SETTING: Dental clinics.
SUBJECTS: Consecutive sample of 456 patients with HIV infection.
METHODS: Social and medical history and clinical examinations. Univariate and logistic regression analysis.
OUTCOMES: Presence of HIV-associated oral disease.
RESULTS: 80% of patients with AIDS and 50% of patients with HIV had a specific oral disease. The most common diseases were hairy leukoplakia (30%), erythematous candidiasis (24%), pseudomembranous candidiasis (14%), angular chielitis (6%), necrotising periodontal disease (8%) and non-recurrent ulceration (6%).
CONCLUSIONS: The presence of erythematous candidiasis was not related to advanced HIV disease. Pseudomembranous candidiasis, hairy leukoplakia and mucosal ulceration were significantly associated with advanced HIV disease. Smoking was also identified as a strong aetiological factor in oral diseases. Longitudinal studies are required to further explore the prognostic significance of oral diseases in HIV infection.     

Prostate Tumor Growth Can Be Modulated by Dietarily Targeting the 15-Lipoxygenase-1 and Cyclooxygenase-2 Enzymes

The main objectives of our study were to determine the bioavailability of omega-3 (ω-3) to the tumor, to understand its mechanisms, and to determine the feasibility of targeting the ω-6 polyunsaturated fatty acids (PUFAs) metabolizing 15-lipoxygenase-1 (15-LO-1) and cyclooxygenase-2 (COX-2) pathways. Nude mice injected subcutaneously with LAPC-4 prostate cancer cells were randomly divided into three different isocaloric (and same percent [%] of total fat) diet groups: high ω-6 linoleic acid (LA), high ω-3 stearidonic acid (SDA) PUFAs, and normal (control) diets. Tumor growth and apoptosis were examined as end points after administration of short-term (5 weeks) ω-3 and ω-6 fatty acid diets. Tumor tissue membranes were examined for growth, lipids, enzyme activities, apoptosis, and proliferation. Tumors from the LA diet-fed mice exhibited the most rapid growth compared with tumors from the control and SDA diet-fed mice. Moreover, a diet switch from LA to SDA caused a dramatic decrease in the growth of tumors in 5 weeks, whereas tumors grew more aggressively when mice were switched from an SDA to an LA diet. Evaluating tumor proliferation (Ki-67) and apoptosis (caspase-3) in mice fed the LA and SDA diets suggested increased percentage proliferation index from the ω-6 diet-fed mice compared with the tumors from the ω-3 SDA-fed mice. Further, increased apoptosis was observed in tumors from ω-3 SDA diet-fed mice versus tumors from ω-6 diet-fed mice. Levels of membrane phospholipids of red blood cells reflected dietary changes and correlated with the levels observed in tumors. Linoleic or arachidonic acid and metabolites (eicosanoid/prostaglandins) were analyzed for 15-LO-1 and COX-2 activities by high-performance liquid chromatography. We also examined the percent unsaturated or saturated fatty acids in the total phospholipids, PUFA ω-6/ω-3 ratios, and other major enzymes (elongase, Delta [Δ]-5-desaturase, and Δ-6-desaturase) of ω-6 catabolic pathways from the tumors. We observed a 2.7-fold increase in the ω-6/ω-3 ratio in tumors from LA diet-fed mice and a 4.2-fold decrease in the ratio in tumors from the SDA diet-fed mice. There was an increased Δ-6-desaturase and Δ-9 desaturase enzyme activities and reduced estimated Δ-5-desaturase activity in tumors from mice fed the SDA diet. Opposite effects were observed in tumors from mice fed the LA diet. Together, these observations provide mechanistic roles of ω-3 fatty acids in slowing prostate cancer growth by altering ω-6/ω-3 ratios through diet and by promoting apoptosis and inhibiting proliferation in tumors by directly competing with ω-6 fatty acids for 15-LO-1 and COX-2 activities.     

Kinetics of peripheral blood mononuclear cell mobilization with chemotherapy and/or granulocyte-colony-stimulating factor: implications for yield of hematopoietic progenitor cell collections

BACKGROUND: Peripheral blood progenitor cells (PBPCs) are commonly collected and used to reconstitute hematopoiesis after high-dose chemotherapy. However, strategies for optimal collection and assessment of leukapheresis components are not standardized. STUDY DESIGN and METHODS: Hematopoietic progenitor cell assays were performed on 369 leukapheresis components collected from 95 patients who had received doxorubicin-based chemotherapy and/or granulocyte-colony-stimulating factor (G-CSF). Precollection patient hematologic values, leukapheresis collection values, component hematopoietic progenitor cell assays, and patient outcome measures were summarized. The kinetics of mononuclear cell (MNC) and PBPC mobilization were assessed among four patient groups. RESULTS: Patient group was a significant predictor of the peripheral blood MNC count on the day of collection (p<0.0001), and that value was a significant predictor of granulocyte-macrophage– colony-forming unit (CFU-GM) yield (p<0.0001). This relationship between the peripheral blood MNC count on the day of collection and CFU- GM yield differed according to patient group (p<0.0001). CFU-GM made up a larger fraction of peripheral blood MNCs collected from patients who received chemotherapy plus G-CSF than collected from those who received G-CSF alone. Moreover, the peripheral blood MNC count and the corresponding CFU-GM yield increased significantly on consecutive days of collection in patient groups receiving chemotherapy and G-CSF but were unchanged or decreased in patients receiving G-CSF alone. CONCLUSION: The relationship between peripheral blood MNC count and leukapheresis component CFU-GM yield differed significantly between patients who received chemotherapy and G-CSF and those who received G- CSF alone for the mobilization of PBPCs. Patient peripheral blood MNC count and component CFU-GM yield are useful for both assessing and suggesting revisions to PBPC mobilization and collection strategies.