Evaluation of Flexible Parametric Relative Survival Approaches for Enforcing Long-Term Constraints When Extrapolating All-Cause Survival.

OBJECTIVES: Parametric models are used to estimate the lifetime benefit of an intervention beyond the range of trial follow-up. Recent recommendations have suggested more flexible survival approaches and the use of external data when extrapolating. Both of these can be realized by using flexible parametric relative survival modeling. The overall aim of this article is to introduce and contrast various approaches for applying constraints on the long-term disease-related (excess) mortality including cure models and evaluate the consequent implications for extrapolation. METHODS: We describe flexible parametric relative survival modeling approaches. We then introduce various options for constraining the long-term excess mortality and compare the performance of each method in simulated data. These methods include fitting a standard flexible parametric relative survival model, enforcing statistical cure, and forcing the long-term excess mortality to converge to a constant. We simulate various scenarios, including where statistical cure is reasonable and where the long-term excess mortality persists. RESULTS: The compared approaches showed similar survival fits within the follow-up period. However, when extrapolating the all-cause survival beyond trial follow-up, there is variation depending on the assumption made about the long-term excess mortality. Altering the time point from which the excess mortality is constrained enables further flexibility. CONCLUSIONS: The various constraints can lead to applying explicit assumptions when extrapolating, which could lead to more plausible survival extrapolations. The inclusion of general population mortality directly into the model-building process, which is possible for all considered approaches, should be adopted more widely in survival extrapolation in health technology assessment.

Perils of Randomized Controlled Trial Survival Extrapolation Assuming Treatment Effect Waning: Why the Distinction Between Marginal and Conditional Estimates Matters.

OBJECTIVES: A long-term, constant, protective treatment effect is a strong assumption when extrapolating survival beyond clinical trial follow-up; hence, sensitivity to treatment effect waning is commonly assessed for economic evaluations. Forcing a hazard ratio (HR) to 1 does not necessarily estimate loss of individual-level treatment effect accurately because of HR selection bias. A simulation study was designed to explore the behavior of marginal HRs under a waning conditional (individual-level) treatment effect and demonstrate bias in forcing a marginal HR to 1 when the estimand is "survival difference with individual-level waning". METHODS: Data were simulated under 4 parameter combinations (varying prognostic strength of heterogeneity and treatment effect). Time-varying marginal HRs were estimated in scenarios where the true conditional HR attenuated to 1. Restricted mean survival time differences, estimated having constrained the marginal HR to 1, were compared with true values to assess bias induced by marginal constraints. RESULTS: Under loss of conditional treatment effect, the marginal HR took a value >1 because of covariate imbalances. Constraining this value to 1 lead to restricted mean survival time difference bias of up to 0.8 years (57% increase). Inflation of effect size estimates also increased with the magnitude of initial protective treatment effect. CONCLUSIONS: Important differences exist between survival extrapolations assuming marginal versus conditional treatment effect waning. When a marginal HR is constrained to 1 to assess efficacy under individual-level treatment effect waning, the survival benefits associated with the new treatment will be overestimated, and incremental cost-effectiveness ratios will be underestimated.

Sex-specific differences in the standard of care for infrarenal abdominal aortic aneurysm repair, and risk of major adverse cardiovascular events and death.

BACKGROUND: This study investigated whether sex-specific differences in preoperative/perioperative standard of care (SOC) account for disparity in outcomes after elective infrarenal abdominal aortic aneurysm repair. METHODS: This was a retrospective cohort study of elective infrarenal abdominal aortic aneurysm repairs (2013-2020) using depersonalized patient-level National Vascular Registry data. SOC was defined for waiting times, preoperative assessment (multidisciplinary/anaesthetic review), cardiovascular risk prevention, and perioperative medication. The primary outcome was major cardiovascular event and/or death (MACED). RESULTS: Some 21 810 patients with an infrarenal abdominal aortic aneurysm were included, 2380 women and 19 430 men. Women less often underwent aneurysm repair within SOC waiting times (51.5 versus 59.3 per cent; P < 0.001), but were equally likely to receive preoperative assessment (72.1 versus 72.5 per cent; P = 0.742). Women were less likely to receive secondary prevention for known cardiac disease (34.9 versus 39.6 per cent; P = 0.015), but more often met overall cardiovascular risk prevention standards (52.1 versus 47.3 per cent; P < 0.001). Women were at greater risk of MACED (open: 12.0 versus 8.9 per cent, P < 0.001; endovascular: 4.9 versus 2.9 per cent, P < 0.001; risk-adjusted OR 1.33, 95 per cent c.i. 1.12 to 1.59). A significant reduction in the odds of MACED was associated with preoperative assessment (OR 0.86, 0.75 to 0.98) and SOC waiting times (OR 0.78, 0.69 to 0.87). There was insufficient evidence to confirm a significant sex-specific difference in the effect of SOC preoperative assessment (women: OR 0.69, 0.50 to 0.97; men: OR 0.89, 0.77 to 1.03; interaction P = 0.170) or SOC waiting times (women: OR 0.84, 0.62 to 1.16; men: OR 0.76, 0.67 to 0.87; interaction P = 0.570) on the risk of MACED. CONCLUSION: SOC waiting times and preoperative assessment were not met for both sexes, which was associated with an increased risk of MACED. Sex-specific differences in SOC attenuated but did not fully account for the increased risk of MACED in women.

Impact of a post-donation hemoglobin testing strategy on efficiency and safety of whole blood donation in England: A modeling study.

BACKGROUND: Deferrals due to low hemoglobin are time-consuming and costly for blood donors and donation services. Furthermore, accepting donations from those with low hemoglobin could represent a significant safety issue. One approach to reduce them is to use hemoglobin concentration alongside donor characteristics to inform personalized inter-donation intervals. STUDY DESIGN AND METHODS: We used data from 17,308 donors to inform a discrete event simulation model comparing personalized inter-donation intervals using "post-donation" testing (i.e., estimating current hemoglobin from that measured by a hematology analyzer at last donation) versus the current approach in England (i.e., pre-donation testing with fixed intervals of 12-weeks for men and 16-weeks for women). We reported the impact on total donations, low hemoglobin deferrals, inappropriate bleeds, and blood service costs. Personalized inter-donation intervals were defined using mixed-effects modeling to estimate hemoglobin trajectories and probability of crossing hemoglobin donation thresholds. RESULTS: The model had generally good internal validation, with predicted events similar to those observed. Over 1 year, a personalized strategy requiring ≥90% probability of being over the hemoglobin threshold, minimized adverse events (low hemoglobin deferrals and inappropriate bleeds) in both sexes and costs in women. Donations per adverse event improved from 3.4 (95% uncertainty interval 2.8, 3.7) under the current strategy to 14.8 (11.6, 19.2) in women, and from 7.1 (6.1, 8.5) to 26.9 (20.8, 42.6) in men. In comparison, a strategy incorporating early returns for those with high certainty of being over the threshold maximized total donations in both men and women, but was less favorable in terms of adverse events, with 8.4 donations per adverse event in women (7.0, 10,1) and 14.8 (12.1, 21.0) in men. DISCUSSION: Personalized inter-donation intervals using post-donation testing combined with modeling of hemoglobin trajectories can help reduce deferrals, inappropriate bleeds, and costs.

Prediction of Cardiovascular Disease Risk Accounting for Future Initiation of Statin Treatment.

Cardiovascular disease (CVD) risk-prediction models are used to identify high-risk individuals and guide statin initiation. However, these models are usually derived from individuals who might initiate statins during follow-up. We present a simple approach to address statin initiation to predict "statin-naive" CVD risk. We analyzed primary care data (2004-2017) from the UK Clinical Practice Research Datalink for 1,678,727 individuals (aged 40-85 years) without CVD or statin treatment history at study entry. We derived age- and sex-specific prediction models including conventional risk factors and a time-dependent effect of statin initiation constrained to 25% risk reduction (from trial results). We compared predictive performance and measures of public-health impact (e.g., number needed to screen to prevent 1 event) against models ignoring statin initiation. During a median follow-up of 8.9 years, 103,163 individuals developed CVD. In models accounting for (versus ignoring) statin initiation, 10-year CVD risk predictions were slightly higher; predictive performance was moderately improved. However, few individuals were reclassified to a high-risk threshold, resulting in negligible improvements in number needed to screen to prevent 1 event. In conclusion, incorporating statin effects from trial results into risk-prediction models enables statin-naive CVD risk estimation and provides moderate gains in predictive ability but had a limited impact on treatment decision-making under current guidelines in this population.

Optimizing Surveillance and Re-intervention Strategy Following Elective Endovascular Repair of Abdominal Aortic Aneurysms.

BACKGROUND: EVAR for abdominal aortic aneurysm has an initial survival advantage over OR, but more frequent complications increase costs and long-term aneurysm-related mortality. Randomized controlled trials of EVAR versus OR have shown EVAR is not cost-effective over a patient's lifetime. However, in the EVAR-1 trial, postoperative surveillance may have been sub-optimal, as the importance of sac growth as a predictor of graft failure was overlooked. METHODS: Real-world data informed a discrete event simulation model of postoperative outcomes following EVAR. Outcomes observed EVAR-1 were compared with those from 5 alternative postoperative surveillance and re-intervention strategies. Key events, quality-adjusted life years and costs were predicted. The impact of using complication and rupture rates from more recent devices, imaging and re-intervention methods was also explored. RESULTS: Compared with observed EVAR-1 outcomes, modeling full adherence to the EVAR-1 scan protocol reduced abdominal aortic aneurysm (AAA) deaths by 3% and increased elective re-interventions by 44%. European Society re-intervention guidelines provided the most clinically effective strategy, with an 8% reduction in AAA deaths, but a 52% increase in elective re-interventions. The cheapest and most cost-effective strategy used lifetime annual ultrasound in primary care with confirmatory computed tomography if necessary, and reduced AAA-related deaths by 5%. Using contemporary rates for complications and rupture did not alter these conclusions. CONCLUSIONS: All alternative strategies improved clinical benefits compared with the EVAR-1 trial. Further work is needed regarding the cost and accuracy of primary care ultrasound, and the potential impact of these strategies in the comparison with OR.

Evaluating the Cost-Effectiveness of Changes to the Surveillance Intervals in the UK Abdominal Aortic Aneurysm Screening Programme.

OBJECTIVES: To investigate the safety and cost-effectiveness of lengthening the time between surveillance ultrasound scans in the UK Abdominal Aortic Aneurysm (AAA) Screening Programme. METHODS: A discrete event simulation model was used to evaluate the cost-effectiveness of AAA screening for men aged 65, comparing current surveillance intervals to 6 alternative surveillance interval strategies that lengthened the time between surveillance scans for 1 or more AAA size categories. The model considered clinical events and costs incurred over a 30-year time horizon and the cost per quality-adjusted life year (QALY). The model adopted the National Health Service perspective and discounted future costs and benefits at 3.5%. RESULTS: Compared with current practice, alternative surveillance strategies resulted in up to a 4% reduction in the number of elective AAA repairs but with an increase of up to 1.6% in the number of AAA ruptures and AAA-related deaths. Alternative strategies resulted in a small reduction in QALYs compared to current practice but with reduced costs. Two strategies that lengthened surveillance intervals in only very small AAAs (3.0-3.9 cm) provided, at a cost-effectiveness threshold of £20 000 per QALY, the highest positive incremental net benefit. There was negligible chance that current practice is the most cost-effective strategy at any threshold below £40 000 per QALY. CONCLUSIONS: Lengthening surveillance intervals in the UK Abdominal Aortic Aneurysm Screening Programme, especially for small AAA, can marginally reduce the incremental cost per QALY of the program. Nevertheless, whether the cost savings from refining surveillance strategies justifies a change in clinical practice is unclear.

Safety of Men With Small and Medium Abdominal Aortic Aneurysms Under Surveillance in the NAAASP.

BACKGROUND: Population screening for abdominal aortic aneurysm (AAA) has commenced in several countries, and has been shown to reduce AAA-related mortality by up to 50%. Most men who screen positive have an AAA <5.5 cm in diameter, the referral threshold for treatment, and are entered into an ultrasound surveillance program. This study aimed to determine the risk of ruptured AAA (rAAA) in men under surveillance. METHODS: Men in the National Health Service AAA Screening Programme who initially had a small (3-4.4 cm) or medium (4.5-5.4 cm) AAA were followed up. The screening program's database collected data on ultrasound AAA diameter measurements, dates of referral, and loss to follow-up. Local screening programs recorded adverse outcomes, including rAAA and death. Rupture and mortality rates were calculated by initial and final known AAA diameter. RESULTS: A total of 18 652 men were included (50 103 person-years of surveillance). Thirty-one men had rAAA during surveillance, of whom 29 died. Some 952 men died of other causes during surveillance, mainly cardiovascular complications (26.3%) and cancer (31.2%). The overall mortality rate was 1.96% per annum, similar for men with small and medium AAAs. The rAAA risk was 0.03% per annum (95% CI, 0.02%-0.05%) for men with small AAAs and 0.28% (0.17%-0.44%) for medium AAAs. The rAAA risk for men with AAAs just below the referral threshold (5.0-5.4 cm) was 0.40% (0.22%-0.73%). CONCLUSIONS: The risk of rAAA under surveillance is <0.5% per annum, even just below the present referral threshold of 5.5 cm, and only 0.4% of men under surveillance are estimated to rupture before referral. It can be concluded that men with small and medium screen-detected AAAs are safe provided they are enrolled in an intensive surveillance program, and that there is no evidence that the current referral threshold of 5.5 cm should be changed.

Impact on survival of modelling increased surgical resection rates in patients with non-small-cell lung cancer and cardiovascular comorbidities: a VICORI study.

BACKGROUND: The impact of cardiovascular disease (CVD) comorbidity on resection rates and survival for patients with early-stage non-small-cell lung cancer (NSCLC) is unclear. We explored if CVD comorbidity explained surgical resection rate variation and the impact on survival if resection rates increased. METHODS: Cancer registry data consisted of English patients diagnosed with NSCLC from 2012 to 2016. Linked hospital records identified CVD comorbidities. We investigated resection rate variation by geographical region using funnel plots; resection and death rates using time-to-event analysis. We modelled an increased propensity for resection in regions with the lowest resection rates and estimated survival change. RESULTS: Among 57,373 patients with Stage 1-3A NSCLC, resection rates varied considerably between regions. Patients with CVD comorbidity had lower resection rates and higher mortality rates. CVD comorbidity explained only 1.9% of the variation in resection rates. For every 100 CVD comorbid patients, increasing resection in regions with the lowest rates from 24 to 44% would result in 16 more patients resected and alive after 1 year and two fewer deaths overall. CONCLUSIONS: Variation in regional resection rate is not explained by CVD comorbidities. Increasing resection in patients with CVD comorbidity to the levels of the highest resecting region would increase 1-year survival.

Mixed-effects models for health care longitudinal data with an informative visiting process: A Monte Carlo simulation study.

Electronic health records are being increasingly used in medical research to answer more relevant and detailed clinical questions; however, they pose new and significant methodological challenges. For instance, observation times are likely correlated with the underlying disease severity: Patients with worse conditions utilise health care more and may have worse biomarker values recorded. Traditional methods for analysing longitudinal data assume independence between observation times and disease severity; yet, with health care data, such assumptions unlikely hold. Through Monte Carlo simulation, we compare different analytical approaches proposed to account for an informative visiting process to assess whether they lead to unbiased results. Furthermore, we formalise a joint model for the observation process and the longitudinal outcome within an extended joint modelling framework. We illustrate our results using data from a pragmatic trial on enhanced care for individuals with chronic kidney disease, and we introduce user-friendly software that can be used to fit the joint model for the observation process and a longitudinal outcome.

Analysis of clinical benefit, harms, and cost-effectiveness of screening women for abdominal aortic aneurysm.

BACKGROUND: A third of deaths in the UK from ruptured abdominal aortic aneurysm (AAA) are in women. In men, national screening programmes reduce deaths from AAA and are cost-effective. The benefits, harms, and cost-effectiveness in offering a similar programme to women have not been formally assessed, and this was the aim of this study. METHODS: We developed a decision model to assess predefined outcomes of death caused by AAA, life years, quality-adjusted life years, costs, and the incremental cost-effectiveness ratio for a population of women invited to AAA screening versus a population who were not invited to screening. A discrete event simulation model was set up for AAA screening, surveillance, and intervention. Relevant women-specific parameters were obtained from sources including systematic literature reviews, national registry or administrative databases, major AAA surgery trials, and UK National Health Service reference costs. FINDINGS: AAA screening for women, as currently offered to UK men (at age 65 years, with an AAA diagnosis at an aortic diameter of ≥3·0 cm, and elective repair considered at ≥5·5cm) gave, over 30 years, an estimated incremental cost-effectiveness ratio of £30 000 (95% CI 12 000-87 000) per quality-adjusted life year gained, with 3900 invitations to screening required to prevent one AAA-related death and an overdiagnosis rate of 33%. A modified option for women (screening at age 70 years, diagnosis at 2·5 cm and repair at 5·0 cm) was estimated to have an incremental cost-effectiveness ratio of £23 000 (9500-71 000) per quality-adjusted life year and 1800 invitations to screening required to prevent one AAA-death, but an overdiagnosis rate of 55%. There was considerable uncertainty in the cost-effectiveness ratio, largely driven by uncertainty about AAA prevalence, the distribution of aortic sizes for women at different ages, and the effect of screening on quality of life. INTERPRETATION: By UK standards, an AAA screening programme for women, designed to be similar to that used to screen men, is unlikely to be cost-effective. Further research on the aortic diameter distribution in women and potential quality of life decrements associated with screening are needed to assess the full benefits and harms of modified options. FUNDING: UK National Institute for Health Research Health Technology Assessment programme.

Biases incurred from nonrandom repeat testing of haemoglobin levels in blood donors: Selective testing and its implications.

To help prevent anaemia, it is a requisite for blood donors to undergo a haemoglobin test to ensure levels are not too low before donation. It is therefore important to have an accurate testing device and strategy to ensure donors are not being inappropriately bled. A recent study in blood donors used a selective testing strategy where if a donor's haemoglobin level is below the level required for donation, then another reading is taken and if this occurs again, a third and final reading is used. This strategy can reduce the average number of readings required per donor compared to taking three measurements for all donors. However, the final decision-making measurement will on average be higher than a single measurement. In this paper, a selective testing strategy is compared against other strategies. Individual-level biases are derived for the selective strategy and are shown to depend on how close a donor's true haemoglobin level is to the donation threshold and the magnitude of error in the testing device. A simulation study was conducted using the distribution of haemoglobin levels from a large donor population to investigate the effects different strategies have on population performance. We consider scenarios based on varying the measurement device bias and error, including differential biases that depend on the underlying haemoglobin level. Discriminatory performance is shown to be affected when using the selective testing strategies, especially when measurement error is large and when differential bias is present in the device. We recommend that the average of a number of readings should be used in preference to selective testing strategies if multiple measurements are available.

Landmark Models for Optimizing the Use of Repeated Measurements of Risk Factors in Electronic Health Records to Predict Future Disease Risk.

The benefits of using electronic health records (EHRs) for disease risk screening and personalized health-care decisions are being increasingly recognized. Here we present a computationally feasible statistical approach with which to address the methodological challenges involved in utilizing historical repeat measures of multiple risk factors recorded in EHRs to systematically identify patients at high risk of future disease. The approach is principally based on a 2-stage dynamic landmark model. The first stage estimates current risk factor values from all available historical repeat risk factor measurements via landmark-age-specific multivariate linear mixed-effects models with correlated random intercepts, which account for sporadically recorded repeat measures, unobserved data, and measurement errors. The second stage predicts future disease risk from a sex-stratified Cox proportional hazards model, with estimated current risk factor values from the first stage. We exemplify these methods by developing and validating a dynamic 10-year cardiovascular disease risk prediction model using primary-care EHRs for age, diabetes status, hypertension treatment, smoking status, systolic blood pressure, total cholesterol, and high-density lipoprotein cholesterol in 41,373 persons from 10 primary-care practices in England and Wales contributing to The Health Improvement Network (1997-2016). Using cross-validation, the model was well-calibrated (Brier score = 0.041, 95% confidence interval: 0.039, 0.042) and had good discrimination (C-index = 0.768, 95% confidence interval: 0.759, 0.777).

Morphological suitability for endovascular repair, non-intervention rates, and operative mortality in women and men assessed for intact abdominal aortic aneurysm repair: systematic reviews with meta-analysis.

BACKGROUND: Prognosis for women with abdominal aortic aneurysm might be worse than the prognosis for men. We aimed to systematically quantify the differences in outcomes between men and women being assessed for repair of intact abdominal aortic aneurysm using data from study periods after the year 2000. METHODS: In these systematic reviews and meta-analysis, we identified studies (randomised, cohort, or cross-sectional) by searching MEDLINE, Embase, CENTRAL, and grey literature published between Jan 1, 2005, and Sept 2, 2016, for two systematic reviews and Jan 1, 2009, and Sept 2, 2016, for one systematic review. Studies were included if they were of both men and women, with data presented for each sex separately, with abdominal aortic aneurysms being assessed for aneurysm repair by either endovascular repair (EVAR) or open repair. We conducted three reviews based on whether studies reported the proportion morphologically suitable (within manufacturers' instructions for use) for EVAR (EVAR suitability review), non-intervention rates (non-intervention review), and 30-day mortality (operative mortality review) after intact aneurysm repair. Studies had to include at least 20 women (for the EVAR suitability review), 20 women (for the non-intervention review), and 50 women (for the operative mortality review). Studies were excluded if they were review articles, editorials, letters, or case reports. For the operative review, studies were also excluded if they only provided hazard ratios or only reported in-hospital mortality. We assessed the quality of the studies using the Newcastle-Ottawa scoring system, and contacted authors for the provision of additional data if needed. We combined results across studies by random-effects meta-analysis. This study is registered with PROSPERO, number CRD42016043227. FINDINGS: Five studies assessed the morphological eligibility for EVAR (1507 men, 400 women). The overall pooled proportion of women eligible (34%) for EVAR was lower than it was in men (54%; odds ratio [OR] 0·44, 95% CI 0·32-0·62). Four single-centre studies reported non-intervention rates (1365 men, 247 women). The overall pooled non-intervention rates were higher in women (34%) than men (19%; OR 2·27, 95% CI 1·21-4·23). The review of 30-day mortality included nine studies (52 018 men, 11 076 women). The overall pooled estimate for EVAR was higher in women (2·3%) than in men (1·4%; OR 1·67, 95% CI 1·38-2·04). The overall estimate for open repair also was higher in women (5·4%) than in men (2·8%; OR 1·76, 95% CI 1·35-2·30). INTERPRETATION: Compared with men, a smaller proportion of women are eligible for EVAR, a higher proportion of women are not offered intervention, and operative mortality is much higher in women for both EVAR and open repair. The management of abdominal aortic aneurysm in women needs improvement. FUNDING: National Institute for Health Research (UK).

Editor's Choice - Re-interventions After Repair of Ruptured Abdominal Aortic Aneurysm: A Report From the IMPROVE Randomised Trial.

OBJECTIVE/BACKGROUND: The aim was to describe the re-interventions after endovascular and open repair of rupture, and investigate whether these were associated with aortic morphology. METHODS: In total, 502 patients from the IMPROVE randomised trial (ISRCTN48334791) with repair of rupture were followed-up for re-interventions for at least 3 years. Pre-operative aortic morphology was assessed in a core laboratory. Re-interventions were described by time (0-90 days, 3 months-3 years) as arterial or laparotomy related, respectively, and ranked for severity by surgeons and patients separately. Rare re-interventions to 1 year, were summarised across three ruptured abdominal aortic aneurysm trials (IMPROVE, AJAX, and ECAR) and odds ratios (OR) describing differences were pooled via meta-analysis. RESULTS: Re-interventions were most common in the first 90 days. Overall rates were 186 and 226 per 100 person years for the endovascular strategy and open repair groups, respectively (p = .20) but between 3 months and 3 years (mid-term) the rates had slowed to 9.5 and 6.0 re-interventions per 100 person years, respectively (p = .090) and about one third of these were for a life threatening condition. In this latter, mid-term period, 42 of 313 remaining patients (13%) required at least one re-intervention, most commonly for endoleak or other endograft complication after treatment by endovascular aneurysm repair (EVAR) (21 of 38 re-interventions), whereas distal aneurysms were the commonest reason (four of 23) for re-interventions after treatment by open repair. Arterial re-interventions within 3 years were associated with increasing common iliac artery diameter (OR 1.48, 95% confidence interval [CI] 0.13-0.93; p = .004). Amputation, rare but ranked as the worst re-intervention by patients, was less common in the first year after treatment with EVAR (OR 0.2, 95% CI 0.05-0.88) from meta-analysis of three trials. CONCLUSION: The rate of mid-term re-interventions after rupture is high, more than double that after elective EVAR and open repair, suggesting the need for bespoke surveillance protocols. Amputations are much less common in patients treated by EVAR than in those treated by open repair.

Endovascular Repair of Abdominal Aortic Aneurysm in Patients Physically Ineligible for Open Repair: Very Long-term Follow-up in the EVAR-2 Randomized Controlled Trial.

OBJECTIVE: The aim of the study was to compare long-term total and aneurysm-related mortality in physically frail patients with abdominal aortic aneurysm (AAA) randomized to either early endovascular aneurysm repair (EVAR) or no-intervention. SUMMARY BACKGROUND DATA: EVAR-2 remains the sole randomized trial to identify whether EVAR reduces mortality in patients physically ineligible for open repair. METHODS: Between September 1999 and August 2004, 404 patients from 33 centers in the United Kingdom aged ≥60 years with AAA >5.5 cm in diameter were randomized 1:1 using computer-generated sequences of randomly permuted blocks stratified by center to receive either EVAR (197) or no-intervention (207). The primary analysis compared total and aneurysm-related deaths in groups until June 30, 2015 (mean, 12.0 yrs; maximum 14.1 yrs). RESULTS: Mean follow-up until death or censoring was 4.2 years. There were 187 deaths (22.6 per 100 person-yrs) in the EVAR group and 194 (22.1 per 100 person-yrs) in the no-intervention group. By 12 years of follow-up the estimated survival was 5.3% [95% confidence interval (CI), 2.6-9.2] in the EVAR group and 8.5% (95% CI, 5.2-12.9) in the no-intervention group; there was no significant difference in life expectancy between the groups (both 4.2 yrs; P = 0.97). However, overall aneurysm-related mortality was significantly lower in the EVAR group [3.3 deaths per 100 person-yrs compared with 6.5 deaths per 100 person-yrs in the no-intervention group, adjusted hazard ratio 0.55 (95% CI, 0.34-0.91; P = 0.019)]. Patients surviving beyond 8 years were younger, with higher body mass index, estimated glomerular filtration rate, and forced expiratory volume in 1 second. CONCLUSIONS: EVAR does not increase overall life expectancy in patients ineligible for open repair, but can reduce aneurysm-related mortality.

Endovascular versus open repair of abdominal aortic aneurysm in 15-years' follow-up of the UK endovascular aneurysm repair trial 1 (EVAR trial 1): a randomised controlled trial.

BACKGROUND: Short-term survival benefits of endovascular aneurysm repair (EVAR) versus open repair of intact abdominal aortic aneurysms have been shown in randomised trials, but this early survival benefit is lost after a few years. We investigated whether EVAR had a long-term survival benefit compared with open repair. METHODS: We used data from the EVAR randomised controlled trial (EVAR trial 1), which enrolled 1252 patients from 37 centres in the UK between Sept 1, 1999, and Aug 31, 2004. Patients had to be aged 60 years or older, have aneurysms of at least 5·5 cm in diameter, and deemed suitable and fit for either EVAR or open repair. Eligible patients were randomly assigned (1:1) using computer-generated sequences of randomly permuted blocks stratified by centre to receive either EVAR (n=626) or open repair (n=626). Patients and treating clinicians were aware of group assignments, no masking was used. The primary analysis compared total and aneurysm-related deaths in groups until mid-2015 in the intention-to-treat population. This trial is registered at ISRCTN (ISRCTN55703451). FINDINGS: We recruited 1252 patients between Sept 1, 1999, and Aug 31, 2004. 25 patients (four for mortality outcome) were lost to follow-up by June 30, 2015. Over a mean of 12·7 years (SD 1·5; maximum 15·8 years) of follow-up, we recorded 9·3 deaths per 100 person-years in the EVAR group and 8·9 deaths per 100 person-years in the open-repair group (adjusted hazard ratio [HR] 1·11, 95% CI 0·97-1·27, p=0·14). At 0-6 months after randomisation, patients in the EVAR group had a lower mortality (adjusted HR 0·61, 95% CI 0·37-1·02 for total mortality; and 0·47, 0·23-0·93 for aneurysm-related mortality, p=0·031), but beyond 8 years of follow-up open-repair had a significantly lower mortality (adjusted HR 1·25, 95% CI 1·00-1·56, p=0·048 for total mortality; and 5·82, 1·64-20·65, p=0·0064 for aneurysm-related mortality). The increased aneurysm-related mortality in the EVAR group after 8 years was mainly attributable to secondary aneurysm sac rupture (13 deaths [7%] in EVAR vs two [1%] in open repair), with increased cancer mortality also observed in the EVAR group. INTERPRETATION: EVAR has an early survival benefit but an inferior late survival compared with open repair, which needs to be addressed by lifelong surveillance of EVAR and re-intervention if necessary. FUNDING: UK National Institute for Health Research, Camelia Botnar Arterial Research Foundation.

Toxicity-dependent feasibility bounds for the escalation with overdose control approach in phase I cancer trials.

Phase I trials of anti-cancer therapies aim to identify a maximum tolerated dose (MTD), defined as the dose that causes unacceptable toxicity in a target proportion of patients. Both rule-based and model-based methods have been proposed for MTD recommendation. The escalation with overdose control (EWOC) approach is a model-based design where the dose assigned to the next patient is one that, given all available data, has a posterior probability of exceeding the MTD equal to a pre-specified value known as the feasibility bound. The aim is to conservatively dose-escalate and approach the MTD, avoiding severe overdosing early on in a trial. The EWOC approach has been applied in practice with the feasibility bound either fixed or varying throughout a trial, yet some of the methods may recommend incoherent dose-escalation, that is, an increase in dose after observing severe toxicity at the current dose. We present examples where varying feasibility bounds have been used in practice, and propose a toxicity-dependent feasibility bound approach that guarantees coherent dose-escalation and incorporates the desirable features of other EWOC approaches. We show via detailed simulation studies that the toxicity-dependent feasibility bound approach provides improved MTD recommendation properties to the original EWOC approach for both discrete and continuous doses across most dose-toxicity scenarios, with comparable performance to other approaches without recommending incoherent dose escalation. © 2017 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

The use of repeated blood pressure measures for cardiovascular risk prediction: a comparison of statistical models in the ARIC study.

Many prediction models have been developed for the risk assessment and the prevention of cardiovascular disease in primary care. Recent efforts have focused on improving the accuracy of these prediction models by adding novel biomarkers to a common set of baseline risk predictors. Few have considered incorporating repeated measures of the common risk predictors. Through application to the Atherosclerosis Risk in Communities study and simulations, we compare models that use simple summary measures of the repeat information on systolic blood pressure, such as (i) baseline only; (ii) last observation carried forward; and (iii) cumulative mean, against more complex methods that model the repeat information using (iv) ordinary regression calibration; (v) risk-set regression calibration; and (vi) joint longitudinal and survival models. In comparison with the baseline-only model, we observed modest improvements in discrimination and calibration using the cumulative mean of systolic blood pressure, but little further improvement from any of the complex methods. © 2016 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

Modelling semi-attributable toxicity in dual-agent phase I trials with non-concurrent drug administration.

In oncology, combinations of drugs are often used to improve treatment efficacy and/or reduce harmful side effects. Dual-agent phase I clinical trials assess drug safety and aim to discover a maximum tolerated dose combination via dose-escalation; cohorts of patients are given set doses of both drugs and monitored to see if toxic reactions occur. Dose-escalation decisions for subsequent cohorts are based on the number and severity of observed toxic reactions, and an escalation rule. In a combination trial, drugs may be administered concurrently or non-concurrently over a treatment cycle. For two drugs given non-concurrently with overlapping toxicities, toxicities occurring after administration of the first drug yet before administration of the second may be attributed directly to the first drug, whereas toxicities occurring after both drugs have been given some present ambiguity; toxicities may be attributable to the first drug only, the second drug only or the synergistic combination of both. We call this mixture of attributable and non-attributable toxicity semi-attributable toxicity. Most published methods assume drugs are given concurrently, which may not be reflective of trials with non-concurrent drug administration. We incorporate semi-attributable toxicity into Bayesian modelling for dual-agent phase I trials with non-concurrent drug administration and compare the operating characteristics to an approach where this detail is not considered. Simulations based on a trial for non-concurrent administration of intravesical Cabazitaxel and Cisplatin in early-stage bladder cancer patients are presented for several scenarios and show that including semi-attributable toxicity data reduces the number of patients given overly toxic combinations. © 2016 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.