Homocysteine-lowering clinical trials in Norway. Cardiovascular and cancer outcomes in the Western Norway B Vitamin Intervention Trial and the Norwegian Vitamin Trial

Abstract

Introduction: Observational studies have reported associations between levels of the amino acid homocysteine in the circulation and risk of cardiovascular disease. Oral administration of the synthetic B vitamins folic acid and cyanocobalamin (vitamin B12) can lower plasma total homocysteine levels. In the Western Norway B Vitamin Intervention Trial (WENBIT) and the Norwegian Vitamin Trial (NORVIT), patients with ischemic heart disease were randomized to groups receiving folic acid plus vitamin B12, or no such treatment, to assess whether they would benefit from lowered homocysteine levels with respect to major adverse clinical events, such as myocardial infarction, stroke, cardiovascular death or all-cause death. Using a 2 x 2 factorial design, participants were also randomized to groups receiving vitamin B6 or no vitamin B6. Aims: The overall aim of the present dissertation was to investigate the clinical effects of B vitamin treatment in patients with established ischemic heart disease. Materials and methods: We used clinical and laboratory data on 6837 patients with ischemic heart disease, recruited from 36 hospitals in Norway (1998 to 2004) collected during in-trial followup, and data on cancer incidence and cause-specific and all-cause mortality on these patients collected during extended follow-up throughout the year 2007. Clinical outcomes were analyzed for groups assigned to folic acid plus vitamin B12 treatment vs no folic acid/vitamin B12, and for groups assigned to vitamin B6 treatment vs no vitamin B6. Survival curves were constructed using the Kaplan-Meier method, and estimates of hazard ratios with confidence intervals were obtained using Cox proportional hazards regression. Results: Folic acid plus vitamin B12 treatment lowered plasma total homocysteine substantially in both trial populations. In the WENBIT study population, this treatment was not associated with the incidence of major adverse cardiovascular events or all-cause mortality during in-trial follow-up of median 38 months. In the combined NORVIT-WENBIT study population, it was not associated with the incidence of major adverse cardiovascular events or any of its constituents (myocardial infarction, stroke or cardiovascular death) during in-trial follow-up of median 39 months, or associated with long-term cardiovascular mortality during extended follow-up of median 78 months. However, among NORVIT-WENBIT participants with hyperhomocysteinemia at baseline, treatment with folic acid plus vitamin B12 was associated with increased risk of in-trial major cardiovascular events, and of long-term cardiovascular mortality. Exploratory analyses in NORVITWENBIT showed that baseline plasma total homocysteine was not independently associated with cardiovascular outcomes, whereas homocysteine measured after 1-2 months of folic acid plus vitamin B12 treatment was a strong predictor of in-trial major cardiovascular events. In the combined NORVIT-WENBIT study population, folic acid plus vitamin B12 treatment was associated with increased cancer incidence, cancer mortality and allcause mortality during extended follow-up of median 78 months. These findings were consistent in both trial populations, among patients with age below or above the median, in both genders, among never and ever smokers and among patients with baseline serum folate level below or above the median. However, hazard ratios for folic acid plus vitamin B12 treatment vs no such treatment were higher among individuals with the TT genotype than among those with the CC or CT genotypes of the methylenetetrahydrofolate reductase 677C→T polymorphism. Vitamin B6 treatment led to a 10-fold increase in plasma levels of pyridoxal 5’ phosphate in both trial populations, but was not associated with outcomes during intrial follow-up in the WENBIT study population, or with any outcomes during in-trial or extended follow-up of the combined NORVIT-WENBIT study population. Discussion and conclusions: Our findings with respect to cardiovascular outcomes are consistent with the null effects of homocysteine-lowering B vitamin treatment demonstrated in large randomized controlled trials to date. The increased risk of cardiovascular outcomes by folic acid plus vitamin B12 among patients with baseline hyperhomocysteinemia was contrary to what would be expected if homocysteine has a causal role in cardiovascular disease progression. Thus, B vitamins to lower homocysteine should not be recommended for patients with cardiovascular disease. The increased cancer incidence and cancer mortality during extended follow-up observed in the groups who received folic acid plus vitamin B12 for median 39 months may be explained by the so-called acceleration phenomenon; that this treatment influenced growth in cancers that were silent at baseline or during trials, leading to excess subsequent clinical surfacing and diagnosis during extended followup. However, reports on cancer outcomes from other completed homocysteinelowering B vitamin treatment trials to date do not support our findings, and our results need confirmation in other populations.