Average Hazard Ratio Research Articles

C-Met as a prognostic marker in gastric cancer: a systematic review and meta-analysis.

Backgroundc-Met has been recognized as an important therapeutic target in gastric cancer, but the prognostic property of the c-Met status is still unclear. We aimed to characterize the prognostic effect of c-Met by systematic review and meta-analysis.MethodsWe identified 15 studies assessing survival in gastric cancer by c-Met status. Effect measure of interest was hazard ratio (HR) for survival. Meta-regression was performed to estimate the relationship between HR and disease stage. Random-effects meta-analyses were used to account for heterogeneity.Results15 eligible studies provided outcome data stratified by c-Met status in 2210 patients. Meta-analysis of the HRs indicated a significantly poorer Os in patients with high c-Met expression (average HR=2.112, 95%CI: 1.622–2.748). Subgroup analysis showed the prognostic effect of c-Met was identical in protein-level and gene-level based methodology. The same effect was also seen in Asian and Western ethnicity subgroup analysis. Meta-regression showed HR was not associated with disease stage.ConclusionsPatients with tumors that harbor high c-Met expression are more likely to have a worse Os, with this prognostic effect independent of disease stage. c-Met status should be evaluated in clinical prognosis.

Quantification of Survival Gain From Cardiac Resynchronization Therapy: Nonlinear Growth With Time, and Greater Gain in Low-Risk Patients, Make Raw Trial Data an Underestimate of Real-World Behavior

ObjectivesThe goal of this study was to examine the impact of calculation-window duration on lifespan gain (as observed in trials) and on who gains most.BackgroundThe landmark trials of biventricular pacing (cardiac resynchronization therapy [CRT]) typically ran for <1 device battery life, and they may therefore underestimate lifespan benefit over longer durations.MethodsWe conducted a meta-analysis of biventricular pacing trials to calculate lifespan gained: first, within the duration of randomized controlled trial data up to 2 years; second, over a 5-year typical battery life; and third, over >1 battery life. Importantly, we applied the Gompertz method for age-related increase in mortality from non–CRT-preventable causes.ResultsFive landmark trials (COMPANION [Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure], CARE-HF (CArdiac REsynchronization–Heart Failure), MADIT-CRT [Multicenter Automatic Defibrillator Implantation Trial With Cardiac Resynchronization Therapy], REVERSE [Resynchronization Reverses Remodeling in Systolic Left Ventricular Dysfunction], RAFT (Resynchronization–Defibrillation for Ambulatory Heart Failure)) provided data for 2 years (6,561 patients), with an average hazard ratio of 0.71. Lifespan gained across all trials increased nonlinearly with time from 0.1 month at 1 year, to 0.5 month at 2 years, and a projected 6.5 months at 5 years (65 times more than at 1 year). After multiple devices, it reached 14 months, involving on average 1.6 devices (i.e., 8.8 months per device implanted). Moreover, while over a short window (e.g., 2 years), lower-mortality patients may gain less than higher-mortality patients (1.4 vs. 2.3 months), their positions reverse by 15 years (16.0 vs. 13.7 months).ConclusionsLifespan gain from biventricular pacing rises nonlinearly with time. Early on, higher-risk patients exhibit more gain, but later, lower-risk patients exhibit more gain. Quantifying gain over less than a patient’s lifetime underestimates lifespan gain. Over the first 1 or 2 years, lower-risk patients may seem to gain less, although they may ultimately be the ones who gain the most.

Abstract 4842: Hazard ratio for colorectal cancer risk in Lynch syndrome is inversely associated with age.

Abstract Introduction: Carriers of germline mutations in mismatch repair genes (MLH1, MSH2, MSH6, PMS2) have a high risk of colorectal and other cancers (Lynch syndrome). Between 30% and 70% of carriers will be develop colorectal cancer by age 70 if they are not screened for the disease. What is not well understood is how the effect of mutations, in absolute and relative terms, varies by age. Methods: We conducted a literature review to identify studies of colorectal cancer that estimated age-specific hazard ratios for colorectal cancer for carriers of a mutation in MLH1 or MSH2 compared with the general population. Studies were excluded if they did not adjust correctly for ascertainment. We conducted a random effects meta-analysis to estimate the average age-specific hazard ratio. Annual risk (incidence) by multiplying the population incidence by the age-specific hazard ratio estimated by the meta-analysis. Results: We identified six studies comprising 673 families (282 MLH1, 391 MSH2). All studies observed that the hazard ratio for colorectal cancer was highest at young ages and decreased with age. This pattern was observed for both males and females and for both MLH1 and MSH2 mutation carriers. Overall, the hazard ratio dropped 6-8 fold over three decades (Table). Incidence of colorectal cancer for carriers increased until age 40 years for males and 50 years for females, and then plateaued (Table). Conclusion: At younger ages, in multiplicative terms, the magnitude of the increased risk conferred by the mutations is substantially greater than at older ages. The reason for this age-dependency is not known. It is, however, consistent with a multi-stage model of carcinogenesis for Lynch syndrome. These findings add to the complement of factors that complicate risk estimation for Lynch syndrome including: polygenic effects, non-genetic risk modifiers, and the possibility of a parent-of-origin effect. There is still much to learn about the clinical implications and carcinogenesis of Lynch syndrome. Meta-analysis estimates of hazard ratios and incidences for colorectal cancer by age group. 30-39 year olds 40-49 year olds 50-59 year olds 60-69 year olds Hazard ratio (95%CI): Risk of colorectal cancer in carriers/risk for the general population. Males 120 (60-243) 95 (39-226) 33 (14-78) 14 (4.2-51) Females 80 (42-154) 70 (35-139) 28 (19-41) 14 (7.4-26) Incidence: Proportion of carriers diagnosed with colorectal cancer per year. Males 1 in 146 1 in 50 1 in 39 1 in 38 Females 1 in 490 1 in 135 1 in 69 1 in 61 Citation Format: Mark A. Jenkins, Aung Ko Win, James G. Dowty, John D. Mathews, John L. Hopper. Hazard ratio for colorectal cancer risk in Lynch syndrome is inversely associated with age. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4842. doi:10.1158/1538-7445.AM2013-4842

Short-Course Antiretroviral Therapy in Primary HIV Infection

BackgroundShort-course antiretroviral therapy (ART) in primary human immunodeficiency virus (HIV) infection may delay disease progression but has not been adequately evaluated.MethodsWe randomly assigned adults with primary HIV infection to ART for 48 weeks, ART for 12 weeks, or no ART (standard of care), with treatment initiated within 6 months after seroconversion. The primary end point was a CD4+ count of less than 350 cells per cubic millimeter or long-term ART initiation.ResultsA total of 366 participants (60% men) underwent randomization to 48-week ART (123 participants), 12-week ART (120), or standard care (123), with an average follow-up of 4.2 years. The primary end point was reached in 50% of the 48-week ART group, as compared with 61% in each of the 12-week ART and standard-care groups. The average hazard ratio was 0.63 (95% confidence interval [CI], 0.45 to 0.90; P=0.01) for 48-week ART as compared with standard care and was 0.93 (95% CI, 0.67 to 1.29; P=0.67) for 12-week ART as compared with standard care. The proportion of participants who had a CD4+ count of less than 350 cells per cubic millimeter was 28% in the 48-week ART group, 40% in the 12-week group, and 40% in the standard-care group. Corresponding values for long-term ART initiation were 22%, 21%, and 22%. The median time to the primary end point was 65 weeks (95% CI, 17 to 114) longer with 48-week ART than with standard care. Post hoc analysis identified a trend toward a greater interval between ART initiation and the primary end point the closer that ART was initiated to estimated seroconversion (P=0.09), and 48-week ART conferred a reduction in the HIV RNA level of 0.44 log10 copies per milliliter (95% CI, 0.25 to 0.64) 36 weeks after the completion of short-course therapy. There were no significant between-group differences in the incidence of the acquired immunodeficiency syndrome, death, or serious adverse events.ConclusionsA 48-week course of ART in patients with primary HIV infection delayed disease progression, although not significantly longer than the duration of the treatment. There was no evidence of adverse effects of ART interruption on the clinical outcome. (Funded by the Wellcome Trust; SPARTAC Controlled-Trials.com number, ISRCTN76742797, and EudraCT number, 2004-000446-20.)

Does early intensive multifactorial treatment reduce total cardiovascular burden in individuals with screen-detected diabetes? Findings from the ADDITION-Europe cluster-randomized trial.

AimsTo describe the total cardiovascular burden (cardiovascular morbidity or mortality, revascularization or non-traumatic amputation) in individuals with screen-detected diabetes in the ADDITION-Europe trial and to quantify the impact of the intervention on multiple cardiovascular events over 5 years.MethodsIn a pragmatic, cluster-randomized, parallel-group trial in four centres (Denmark; Cambridge, UK; the Netherlands; and Leicester, UK), 343 general practices were randomized to screening plus routine care (n = 1379 patients), or screening and promotion of target-driven, intensive treatment of multiple risk factors (n = 1678). We estimated the effect of the intervention on multiple cardiovascular events after diagnosis of diabetes using the Wei, Lin and Weissfeld method.ResultsOver 5.3 years, 167 individuals had exactly one cardiovascular event, 53 exactly two events, and 18 three or more events. The incidence rates (95% CI) of first events and any event per 1000 person-years were 14.6 (12.8–16.6) and 20.4 (18.2–22.6), respectively. There were non-significant reductions in the risk of a first (hazard ratio 0.83, 95% CI 0.65–1.05) and second primary endpoint (hazard ratio 0.70, 95% CI 0.43–1.12). The overall average hazard ratio for any event was 0.77 (95% CI 0.58–1.02).ConclusionsEarly intensive multifactorial treatment was not associated with a significant reduction in total cardiovascular burden at 5 years. Focusing on first events in cardiovascular disease prevention trials underestimates the total cardiovascular burden to patients and the health service.

Non-parametric estimation of relative risk in survival and associated tests

We extend the Tarone and Ware scheme of weighted log-rank tests to cover the associated weighted Mantel-Haenszel estimators of relative risk. Weighting functions previously employed are critically reviewed. The notion of an average hazard ratio is defined and its connection to the effect size measure P(Y > X) is emphasized. The connection makes estimation of P(Y > X) possible also under censoring. Two members of the extended Tarone-Ware scheme accomplish the estimation of intuitively interpretable average hazard ratios, also under censoring and time-varying relative risk which is achieved by an inverse probability of censoring weighting. The empirical properties of the members of the extended Tarone-Ware scheme are demonstrated by a Monte Carlo study. The differential role of the weighting functions considered is illustrated by a comparative analysis of four real data sets.

Excision Repair Cross-Complementation Group 1 (ERCC1) Status and Lung Cancer Outcomes: A Meta-Analysis of Published Studies and Recommendations

PurposeDespite discrepant results on clinical utility, several trials are already prospectively randomizing non-small cell lung cancer (NSCLC) patients by ERCC1 status. We aimed to characterize the prognostic and predictive effect of ERCC1 by systematic review and meta-analysis.MethodsEligible studies assessed survival and/or chemotherapy response in NSCLC or SCLC by ERCC1 status. Effect measures of interest were hazard ratio (HR) for survival or relative risk (RR) for chemotherapy response. Random-effects meta-analyses were used to account for between-study heterogeneity, with unadjusted/adjusted effect estimates considered separately.Results23 eligible studies provided survival results in 2,726 patients. Substantial heterogeneity was observed in all meta-analyses (I2 always >30%), partly due to variability in thresholds defining ‘low’ and ‘high’ ERCC1. Meta-analysis of unadjusted estimates showed high ERCC1 was associated with significantly worse overall survival in platinum-treated NSCLC (average unadjusted HR = 1.61, 95%CI:1.23–2.1, p = 0.014), but not in NSCLC untreated with chemotherapy (average unadjusted HR = 0.82, 95%CI:0.51–1.31). Meta-analysis of adjusted estimates was limited by variable choice of adjustment factors and potential publication bias (Egger's p<0.0001). There was evidence that high ERCC1 was associated with reduced response to platinum (average RR = 0.80; 95%CI:0.64–0.99). SCLC data were inadequate to draw firm conclusions.ConclusionsCurrent evidence suggests high ERCC1 may adversely influence survival and response in platinum-treated NSCLC patients, but not in non-platinum treated, although definitive evidence of a predictive influence is lacking. International consensus is urgently required to provide consistent, validated ERCC1 assessment methodology. ERCC1 assessment for treatment selection should currently be restricted to, and evaluated within, clinical trials.

Estimating Covariate-Adjusted Log Hazard Ratios for Multiple Time Intervals in Clinical Trials Using Nonparametric Randomization Based ANCOVA

The hazard ratio is a useful tool in randomized clinical trials for comparing time-to-event outcomes for two groups. Although better power is often achieved for assessments of the hazard ratio via model-based methods that adjust for baseline covariates, such methods make relatively strong assumptions, which can be problematic in regulatory settings that require prespecified analysis plans. This article introduces a nonparametric method for producing covariate-adjusted estimates of the log weighted average hazard ratio for nonoverlapping time intervals under minimal assumptions. The proposed methodology initially captures the means of baseline covariables for each group and the means of indicators for risk and survival for each interval and group. These quantities are used to produce estimates of interval-specific log weighted average hazard ratios and the difference in means for baseline covariables between two groups, with a corresponding covariance matrix. Randomization-based analysis of covariance is applied to produce covariate-adjusted estimates for the interval-specific log hazard ratios through forcing the difference in means for baseline covariables to zero, and there is variance reduction for these adjusted estimates when the time to event has strong correlations with the covariates. The method is illustrated on data from a clinical trial of a noncurable neurologic disorder.

Estimation of the 2-sample hazard ratio function using a semiparametric model

The hazard ratio provides a natural target for assessing a treatment effect with survival data, with the Cox proportional hazards model providing a widely used special case. In general, the hazard ratio is a function of time and provides a visual display of the temporal pattern of the treatment effect. A variety of nonproportional hazards models have been proposed in the literature. However, available methods for flexibly estimating a possibly time-dependent hazard ratio are limited. Here, we investigate a semiparametric model that allows a wide range of time-varying hazard ratio shapes. Point estimates as well as pointwise confidence intervals and simultaneous confidence bands of the hazard ratio function are established under this model. The average hazard ratio function is also studied to assess the cumulative treatment effect. We illustrate corresponding inference procedures using coronary heart disease data from the Women's Health Initiative estrogen plus progestin clinical trial.

The estimation of average hazard ratios by weighted Cox regression

Often the effect of at least one of the prognostic factors in a Cox regression model changes over time, which violates the proportional hazards assumption of this model. As a consequence, the average hazard ratio for such a prognostic factor is under- or overestimated. While there are several methods to appropriately cope with non-proportional hazards, in particular by including parameters for time-dependent effects, weighted estimation in Cox regression is a parsimonious alternative without additional parameters. The methodology, which extends the weighted k-sample logrank tests of the Tarone-Ware scheme to models with multiple, binary and continuous covariates, has been introduced in the nineties of the last century and is further developed and re-evaluated in this contribution. The notion of an average hazard ratio is defined and its connection to the effect size measure P(X<Y) is emphasized. The suggested approach accomplishes estimation of intuitively interpretable average hazard ratios and provides tools for inference. A Monte Carlo study confirms the satisfactory performance. Advantages of the approach are exemplified by comparing standard and weighted analyses of an international lung cancer study. SAS and R programs facilitate application.

Response to Brown et al., ‘Incident and prevalent herpes simplex virus type 2 infection increases risk of HIV acquisition among women in Uganda and Zimbabwe’

The recent paper by Brown and colleagues [1] concerning the impact of incident and prevalent herpes simplex virus type 2 (HSV-2) on the acquisition of HIV among women in Uganda and Zimbabwe provides further evidence for the role of the former in the spread of the latter and confirms the results of an earlier longitudinal study among female sex workers [2] and a recent meta-analysis [3]. Of particular importance is the relative contribution of incident and prevalent herpes simplex virus infection. Brown and colleagues [1] show that incident HSV-2 infection has a greater impact on the risk of HIV acquisition than does prevalent HSV-2 infection. The hazard ratio for the risk of acquiring HIV among individuals with incident and prevalent infection does not, however, quite reach significance at the 5% level in either of their two data sets. In Figure 1 we compare the data given by Brown and colleagues [1] with the corresponding data from South Africa [2].Fig. 1: The hazard ratio for the risk of HIV acquisition among women with incident and prevalent herpes simplex virus type 2 infection comparing women in the period after herpes simplex virus type 2 seroconversion to women who were herpes simplex virus type 2 positive at the start of the study. The hazard ratios are South Africa 6.0 (2.6–14.0) [2], Uganda women 1.64 (0.67–4.01) [1], Zimbabwe women 1.95 (0.98–3.91) [1], weighted average 2.46 (1.53–3.95).The three estimates do not differ significantly and the weighted average hazard ratio suggests that women with incident HSV-2 infections are 2.46 (1.53–3.95) times more likely to be infected with HIV-1 than women with prevalent HSV-2 infections. These data suggest that the risk of acquiring HIV-1 infections wanes over time after infection with HSV-2, but none of the studies provide sufficient data to measure the rate at which this happens directly. The South African study may tend to give a higher estimate than the studies carried out in Uganda and Zimbabwe because of the much shorter time between visits (one month versus 3 months), which would mean that women were identified as being HSV-2 positive sooner after being infected in South Africa than in Uganda. Furthermore, the South African women were sex workers and at a higher overall risk with an HSV-2 incidence over the course of the study of 35% per year compared with 10 and 9% in Uganda and Zimbabwe, respectively, [1]. In a study in Carletonville, South Africa, among young adults from the general population [4] HIV infection was very strongly associated with HSV-2 infection; the odds ratios for HIV infection among those with and without HSV-2 infection was 5.3 (2.7–10.3) for men and 8.4 (4.9–14.2) for women. Furthermore, by the time they reached 24 years of age, 89% of women and 51% of men were infected with HSV-2. The high population prevalence and the very high incidence of HSV-2 among women between the ages of 15 and 25 years adds to the urgency of finding ways to prevent and manage HSV-2 infection. We fully endorse the conclusions drawn by Brown and colleagues [1] concerning the need to find ways to control HSV-2 as part of the overall strategy for controlling HIV. If, as seems to be the case, the effect of HSV-2 on HIV acquisition wanes over time then prevention of HSV-2 is likely to be even more effective for HIV prevention than cure or symptomatic treatment of HSV-2. The study in South Africa adds important evidence in support of their arguments, especially in protecting young women from being infected with HIV. Acknowledgements We thank Peter Ghys for helpful comments and suggestions during the preparation of this manuscript.

Estimation of the average hazard ratio

SUMMARY Simple estimators of the average hazard ratio are obtained for the two sample problem with censored failure time data. These estimators are compared with estimators arising out of the partial likelihood within the proportional hazards class. Efficiency results are found to be generally quite favourable provided the hazard ratio is not too large. Some comments are made on extensions to the k sample problem.