Orestis A. Panagiotou

Comparison of Effect Sizes Associated With Biomarkers Reported in Highly Cited Individual Articles and in Subsequent Meta-analyses

CONTEXT Many biomarkers are proposed in highly cited studies as determinants of disease risk, prognosis, or response to treatment, but few eventually transform clinical practice. OBJECTIVE To examine whether the magnitude of the effect sizes of biomarkers proposed in highly cited studies is accurate or overestimated. DATA SOURCES We searched ISI Web of Science and MEDLINE until December 2010. STUDY SELECTION We included biomarker studies that had a relative risk presented in their abstract. Eligible articles were those that had received more than 400 citations in the ISI Web of Science and that had been published in any of 24 highly cited biomedical journals. We also searched MEDLINE for subsequent meta-analyses on the same associations (same biomarker and same outcome). DATA EXTRACTION In the highly cited studies, data extraction was focused on the disease/outcome, biomarker under study, and first reported relative risk in the abstract. From each meta-analysis, we extracted the overall relative risk and the relative risk in the largest study. Data extraction was performed independently by 2 investigators. RESULTS We evaluated 35 highly cited associations. For 30 of the 35 (86%), the highly cited studies had a stronger effect estimate than the largest study; for 3 the largest study was also the highly cited study; and only twice was the effect size estimate stronger in the largest than in the highly cited study. For 29 of the 35 (83%) highly cited studies, the corresponding meta-analysis found a smaller effect estimate. Only 15 of the associations were nominally statistically significant based on the largest studies, and of those only 7 had a relative risk point estimate greater than 1.37. CONCLUSION Highly cited biomarker studies often report larger effect estimates for postulated associations than are reported in subsequent meta-analyses evaluating the same associations.

Evaluation of Excess Significance Bias in Animal Studies of Neurological Diseases

The evaluation of 160 meta-analyses of animal studies on potential treatments for neurological disorders reveals that the number of statistically significant results was too large to be true, suggesting biases.

The Power of Meta-Analysis in Genome-Wide Association Studies

Meta-analysis of multiple genome-wide association (GWA) studies has become common practice over the past few years. The main advantage of this technique is the maximization of power to detect subtle genetic effects for common traits. Moreover, one can use meta-analysis to probe and identify heterogeneity in the effect sizes across the combined studies. In this review, we systematically appraise and evaluate the characteristics of GWA meta-analyses with 10,000 or more subjects published up to June 2012. We provide an overview of the current landscape of variants discovered by GWA meta-analyses, and we discuss and assess with extrapolations from empirical data the value of larger meta-analyses for the discovery of additional genetic associations and new biology in the future. Finally, we discuss some emerging logistical and practical issues related to the conduct of meta-analysis of GWA studies.

Comparative effect sizes in randomised trials from less developed and more developed countries: meta-epidemiological assessment

Objective To compare treatment effects from randomised trials conducted in more developed versus less developed countries. Design Meta-epidemiological study. Data sources Cochrane Database of Systematic Reviews (August 2012). Data extraction Meta-analyses with mortality outcomes including data from at least one randomised trial conducted in a less developed country and one in a more developed country. Relative risk estimates of more versus less developed countries were compared by calculating the relative relative risks for each topic and the summary relative relative risks across all topics. Similar analyses were performed for the primary binary outcome of each topic. Results 139 meta-analyses with mortality outcomes were eligible. No nominally significant differences between more developed and less developed countries were found for 128 (92%) meta-analyses. However, differences were beyond chance in 11 (8%) cases, always showing more favourable treatment effects in trials from less developed countries. The summary relative relative risk was 1.12 (95% confidence interval 1.06 to 1.18; P<0.001; I2=0%), suggesting significantly more favourable mortality effects in trials from less developed countries. Results were similar for meta-analyses with nominally significant treatment effects for mortality (1.15), meta-analyses with recent trials (1.14), and when excluding trials from less developed countries that subsequently became more developed (1.12). For the primary binary outcomes (127 meta-analyses), 20 topics had differences in treatment effects beyond chance (more favourable in less developed countries in 15/20 cases). Conclusions Trials from less developed countries in a few cases show significantly more favourable treatment effects than trials in more developed countries and, on average, treatment effects are more favourable in less developed countries. These discrepancies may reflect biases in reporting or study design as well as genuine differences in baseline risk or treatment implementation and should be considers when generalising evidence across different settings.

Genome-wide Significant Associations for Variants With Minor Allele Frequency of 5% or Less—An Overview: A HuGE Review

The authors survey uncommon variants (minor allele frequency, ≤5%) that have reached genome-wide significance (P ≤ 10⁻⁷) in genome-wide association study(ies) (GWAS). They examine the typical effect sizes of these associations; whether they have arisen in multiple GWAS on the same phenotype; and whether they pertain to genetic loci that have other variants discovered through GWAS, perceived biologic plausibility from the candidate gene era, or known mutations associated with related phenotypes. Forty-three associations with minor allele frequency of 5% or less and P ≤ 10⁻⁷ were studied, 12 of which involved nonsynonymous variants. Per-allele odds ratios ranged from 1.03 to 22.11. Thirty-two associations had P ≤ 10⁻⁸. Eight uncommon variants were identified in multiple GWAS. For 14 associations, also other common polymorphisms with genome-wide significance were identified in the same loci. Thirteen associations pertained to genetic loci considered to have biologic plausibility for association in the candidate gene era, and mutations with related phenotypic effects were identified for 11 associations. Twenty-five uncommon variants are common in at least 1 of the 4 different ancestry samples of the International HapMap Project. Although the number of uncommon variants with genome-wide significance is still limited, these data suggest a possible confluence of rare/uncommon and common genetic variation on the same genetic loci.

A Genome-Wide Scan Identifies Variants in NFIB Associated with Metastasis in Patients with Osteosarcoma

UNLABELLED Metastasis is the leading cause of death in patients with osteosarcoma, the most common pediatric bone malignancy. We conducted a multistage genome-wide association study of osteosarcoma metastasis at diagnosis in 935 osteosarcoma patients to determine whether germline genetic variation contributes to risk of metastasis. We identified an SNP, rs7034162, in NFIB significantly associated with metastasis in European osteosarcoma cases, as well as in cases of African and Brazilian ancestry (meta-analysis of all cases: P = 1.2 × 10(-9); OR, 2.43; 95% confidence interval, 1.83-3.24). The risk allele was significantly associated with lowered NFIB expression, which led to increased osteosarcoma cell migration, proliferation, and colony formation. In addition, a transposon screen in mice identified a significant proportion of osteosarcomas harboring inactivating insertions in Nfib and with lowered NFIB expression. These data suggest that germline genetic variation at rs7034162 is important in osteosarcoma metastasis and that NFIB is an osteosarcoma metastasis susceptibility gene. SIGNIFICANCE Metastasis at diagnosis in osteosarcoma is the leading cause of death in these patients. Here we show data that are supportive for the NFIB locus as associated with metastatic potential in osteosarcoma.

Replication and predictive value of SNPs associated with melanoma and pigmentation traits in a Southern European case-control study.

Background Genetic association studies have revealed numerous polymorphisms conferring susceptibility to melanoma. We aimed to replicate previously discovered melanoma-associated single-nucleotide polymorphisms (SNPs) in a Greek case-control population, and examine their predictive value. Methods Based on a field synopsis of genetic variants of melanoma (MelGene), we genotyped 284 patients and 284 controls at 34 melanoma-associated SNPs of which 19 derived from GWAS. We tested each one of the 33 SNPs passing quality control for association with melanoma both with and without accounting for the presence of well-established phenotypic risk factors. We compared the risk allele frequencies between the Greek population and the HapMap CEU sample. Finally, we evaluated the predictive ability of the replicated SNPs. Results Risk allele frequencies were significantly lower compared to the HapMap CEU for eight SNPs (rs16891982 – SLC45A2, rs12203592 – IRF4, rs258322 – CDK10, rs1805007 – MC1R, rs1805008 - MC1R, rs910873 - PIGU, rs17305573- PIGU, and rs1885120 - MTAP) and higher for one SNP (rs6001027 – PLA2G6) indicating a different profile of genetic susceptibility in the studied population. Previously identified effect estimates modestly correlated with those found in our population (r = 0.72, P<0.0001). The strongest associations were observed for rs401681-T in CLPTM1L (odds ratio [OR] 1.60, 95% CI 1.22–2.10; P = 0.001), rs16891982-C in SCL45A2 (OR 0.51, 95% CI 0.34–0.76; P = 0.001), and rs1805007-T in MC1R (OR 4.38, 95% CI 2.03–9.43; P = 2×10−5). Nominally statistically significant associations were seen also for another 5 variants (rs258322-T in CDK10, rs1805005-T in MC1R, rs1885120-C in MYH7B, rs2218220-T in MTAP and rs4911442-G in the ASIP region). The addition of all SNPs with nominal significance to a clinical non-genetic model did not substantially improve melanoma risk prediction (AUC for clinical model 83.3% versus 83.9%, p = 0.66). Conclusion Overall, our study has validated genetic variants that are likely to contribute to melanoma susceptibility in the Greek population.

Time to Smoke First Morning Cigarette and Lung Cancer in a Case–Control Study

BACKGROUND Targeting smokers at higher lung cancer risk can improve efficiency and reduce false-positive detection in lung cancer screening. We evaluated whether time to first cigarette after waking (TTFC), a single-item measure of nicotine dependency, could improve stratification of lung cancer risk beyond standard smoking metrics (intensity, duration, and pack-years). METHODS In 3249 ever-smokers (n = 1812 case subjects; n = 1437 control subjects) from a population-based case-control study in Italy, we examined the association between TTFC and lung cancer using logistic regression and estimated lung cancer incidence by levels of TTFC, and intensity, duration, and pack-years using absolute risk regression. Significance tests were two-sided. RESULTS Compared with smokers with TTFC greater than 60 minutes, the lung cancer odds ratios for TTFC of 31 to 60 minutes, 6 to 30 minutes, and 5 or fewer minutes (by increasing dependency) were 2.57 (95% confidence interval [CI] = 2.03 to 3.26), 2.27 (95% CI = 1.79 to 2.88), and 3.50 (95% CI = 2.64 to 4.64), respectively (P trend < .0001). The average lung cancer incidence rates for smokers of 1 to 10, 11 to 20, 21 to 30 and more than 30 cigarettes per day were consistently higher among smokers with TTFC of 60 or fewer minutes vs more than 60 minutes (64.1 vs 11.7; 125.6 vs 28.6; 130.1 vs 40.7; and 260.8 vs 108.9 per 100000 person-years, respectively). The slopes of increase in lung cancer rates with smoking duration and pack-years were statistically significantly greater among smokers with higher dependency (P interaction < .001). CONCLUSIONS Lung cancer risk increases with shorter TTFC; this simple nicotine dependency measure increases lung cancer risk stratification beyond standard smoking measures. Assessing TTFC may improve lung cancer risk prediction and could be useful in lung cancer screening and smoking cessation programs.

Concordance of Sleep and Pain Outcomes of Diverse Interventions: An Umbrella Review

Background/Objective Pain influences sleep and vice versa. We performed an umbrella review of meta-analyses on treatments for diverse conditions in order to examine whether diverse medical treatments for different conditions have similar or divergent effects on pain and sleep. Methods We searched published systematic reviews with meta-analyses in the Cochrane Database of Systematic Reviews until October 20, 2011. We identified randomized trials (or meta-analyses thereof, when >1 trial was available) where both pain and sleep outcomes were examined. Pain outcomes were categorized as headache, musculoskeletal, abdominal, pelvic, generic or other pain. Sleep outcomes included insomnia, sleep disruption, and sleep disturbance. We estimated odds ratios for all outcomes and evaluated the concordance in the direction of effects between sleep and various types of pain and the correlation of treatment effects between sleep and pain outcomes. Results 151 comparisons with 385 different trials met our eligibility criteria. 96 comparisons had concordant direction of effects between each pain outcome and sleep, while in 55 the effect estimates were in opposite directions (P<0.0001). In the 20 comparisons with largest amount of evidence, the experimental drug always had worse sleep outcomes and tended to have worse pain outcomes in 17/20 cases. For headache and musculoskeletal pain, 69 comparisons showed concordant direction of effects with sleep outcomes and 36 showed discordant direction (P<0.0001). For the other 4 pain types there were overall 27 vs. 19 pairs with concordant vs. discordant direction of effects (P = 0.095). There was a weak correlation of the treatment effect sizes for sleep vs. headache/musculoskeletal pain (r = 0.17, P = 0.092). Conclusions Medical interventions tend to have effects in the same direction for pain and sleep outcomes, but exceptions occur. Concordance is primarily seen for sleep and headache or musculoskeletal pain where many drugs may both disturb sleep and cause pain.

Effect of bivalent human papillomavirus vaccination on pregnancy outcomes: long term observational follow-up in the Costa Rica HPV Vaccine Trial

Objective To examine the effect of the bivalent human papillomavirus (HPV) vaccine on miscarriage. Design Observational long term follow-up of a randomized, double blinded trial combined with an independent unvaccinated population based cohort. Setting Single center study in Costa Rica. Participants 7466 women in the trial and 2836 women in the unvaccinated cohort enrolled at the end of the randomized trial and in parallel with the observational trial component. Intervention Women in the trial were assigned to receive three doses of bivalent HPV vaccine (n=3727) or the control hepatitis A vaccine (n=3739). Crossover bivalent HPV vaccination occurred in the hepatitis A vaccine arm at the end of the trial. Women in the unvaccinated cohort received (n=2836) no vaccination. Main outcome measure Risk of miscarriage, defined by the US Centers for Disease Control and Prevention as fetal loss within 20 weeks of gestation, in pregnancies exposed to bivalent HPV vaccination in less than 90 days and any time from vaccination compared with pregnancies exposed to hepatitis A vaccine and pregnancies in the unvaccinated cohort. Results Of 3394 pregnancies conceived at any time since bivalent HPV vaccination, 381 pregnancies were conceived less than 90 days from vaccination. Unexposed pregnancies comprised 2507 pregnancies conceived after hepatitis A vaccination and 720 conceived in the unvaccinated cohort. Miscarriages occurred in 451 (13.3%) of all exposed pregnancies, in 50 (13.1%) of the pregnancies conceived less than 90 days from bivalent HPV vaccination, and in 414 (12.8%) of the unexposed pregnancies, of which 316 (12.6%) were in the hepatitis A vaccine group and 98 (13.6%) in the unvaccinated cohort. The relative risk of miscarriage for pregnancies conceived less than 90 days from vaccination compared with all unexposed pregnancies was 1.02 (95% confidence interval 0.78 to 1.34, one sided P=0.436) in unadjusted analyses. Results were similar after adjusting for age at vaccination (relative risk 1.15, one sided P=0.17), age at conception (1.03, P=0.422), and calendar year (1.06, P=0.358), and in stratified analyses. Among pregnancies conceived at any time from bivalent HPV vaccination, exposure was not associated with an increased risk of miscarriage overall or in subgroups, except for miscarriages at weeks 13-20 of gestation (relative risk 1.35, 95% confidence interval 1.02 to 1.77, one sided P=0.017). Conclusions There is no evidence that bivalent HPV vaccination affects the risk of miscarriage for pregnancies conceived less than 90 days from vaccination. The increased risk estimate for miscarriages in a subgroup of pregnancies conceived any time after vaccination may be an artifact of a thorough set of sensitivity analyses, but since a genuine association cannot totally be ruled out, this signal should nevertheless be explored further in existing and future studies. Trial registration Clinicaltrials.gov NCT00128661 and NCT01086709.