Individual patient data meta-analysis of randomized anti-epileptic drug monotherapy trials

How often can meta-analyses of individual-level data individualize treatment? A meta-epidemiologic study

BACKGROUND

One of the claimed main advantages of individual participant data meta-analysis (IPDMA) is that it allows assessment of subgroup effects based on individual-level participant characteristics, and eventually stratified medicine. In this study, we evaluated the conduct and results of subgroup analyses in IPDMA.

METHODS

We searched PubMed, EMBASE and the Cochrane Library from inception to 31 December 2014. We included papers if they described an IPDMA based on randomized clinical trials that investigated a therapeutic intervention on human subjects and in which the meta-analysis was preceded by a systematic literature search. We extracted data items related to subgroup analysis and subgroup differences (subgroup-treatment interaction p < 0.05).

RESULTS

Overall, 327 IPDMAs were eligible. A statistically significant subgroup-treatment interaction for the primary outcome was reported in 102 (36.6%) of 279 IPDMAs that reported at least one subgroup analysis. This corresponded to 187 different statistically significant subgroup-treatment interactions: 124 for an individual-level subgrouping variable (in 76 IPDMAs) and 63 for a group-level subgrouping variable (in 36 IPDMAs). Of the 187, only 7 (3.7%; 6 individual and 1 group-level subgrouping variables) had a large difference between strata (standardized effect difference d  ≥  0.8). Among the 124 individual-level statistically significant subgroup differences, the IPDMA authors claimed that 42 (in 21 IPDMAs) should lead to treating the subgroups differently. None of these 42 had d  ≥  0.8.

CONCLUSIONS

Availability of individual-level data provides statistically significant interactions for relative treatment effects in about a third of IPDMAs. A modest number of these interactions may offer opportunities for stratified medicine decisions.

Comparison of aggregate and individual participant data approaches to meta-analysis of randomised trials: An observational study

BACKGROUND

It remains unclear when standard systematic reviews and meta-analyses that rely on published aggregate data (AD) can provide robust clinical conclusions. We aimed to compare the results from a large cohort of systematic reviews and meta-analyses based on individual participant data (IPD) with meta-analyses of published AD, to establish when the latter are most likely to be reliable and when the IPD approach might be required.

METHODS AND FINDINGS

We used 18 cancer systematic reviews that included IPD meta-analyses: all of those completed and published by the Meta-analysis Group of the MRC Clinical Trials Unit from 1991 to 2010. We extracted or estimated hazard ratios (HRs) and standard errors (SEs) for survival from trial reports and compared these with IPD equivalents at both the trial and meta-analysis level. We also extracted or estimated the number of events. We used paired t tests to assess whether HRs and SEs from published AD differed on average from those from IPD. We assessed agreement, and whether this was associated with trial or meta-analysis characteristics, using the approach of Bland and Altman. The 18 systematic reviews comprised 238 unique trials or trial comparisons, including 37,082 participants. A HR and SE could be generated for 127 trials, representing 53% of the trials and approximately 79% of eligible participants. On average, trial HRs derived from published AD were slightly more in favour of the research interventions than those from IPD (HRAD to HRIPD ratio = 0.95, p = 0.007), but the limits of agreement show that for individual trials, the HRs could deviate substantially. These limits narrowed with an increasing number of participants (p < 0.001) or a greater number (p < 0.001) or proportion (p < 0.001) of events in the AD. On average, meta-analysis HRs from published AD slightly tended to favour the research interventions whether based on fixed-effect (HRAD to HRIPD ratio = 0.97, p = 0.088) or random-effects (HRAD to HRIPD ratio = 0.96, p = 0.044) models, but the limits of agreement show that for individual meta-analyses, agreement was much more variable. These limits tended to narrow with an increasing number (p = 0.077) or proportion of events (p = 0.11) in the AD. However, even when the information size of the AD was large, individual meta-analysis HRs could still differ from their IPD equivalents by a relative 10% in favour of the research intervention to 5% in favour of control. We utilised the results to construct a decision tree for assessing whether an AD meta-analysis includes sufficient information, and when estimates of effects are most likely to be reliable. A lack of power at the meta-analysis level may have prevented us identifying additional factors associated with the reliability of AD meta-analyses, and we cannot be sure that our results are generalisable to all outcomes and effect measures.

CONCLUSIONS

In this study we found that HRs from published AD were most likely to agree with those from IPD when the information size was large. Based on these findings, we provide guidance for determining systematically when standard AD meta-analysis will likely generate robust clinical conclusions, and when the IPD approach will add considerable value.

Phenobarbitone versus phenytoin monotherapy for epilepsy: an individual participant data review

BACKGROUND

This is an update of a Cochrane Review first published in 2001, and last updated in 2013. This review is one in a series of Cochrane Reviews investigating pair-wise monotherapy comparisons.Epilepsy is a common neurological condition in which abnormal electrical discharges from the brain cause recurrent unprovoked seizures. It is believed that with effective drug treatment, up to 70% of individuals with active epilepsy have the potential to become seizure-free and go into long-term remission shortly after starting drug therapy with a single antiepileptic drug in monotherapy.Worldwide, particularly in the developing world, phenytoin and phenobarbitone are commonly used antiepileptic drugs, primarily because they are inexpensive. The aim of this review is to summarise data from existing trials comparing phenytoin and phenobarbitone.

OBJECTIVES

To review the time to treatment failure, remission and first seizure with phenobarbitone compared with phenytoin when used as monotherapy in people with focal onset seizures (simple or complex focal and secondarily generalised), or generalised onset tonic-clonic seizures (with or without other generalised seizure types).

SEARCH METHODS

For the latest update, we searched the following databases on 21 August 2018: the Cochrane Register of Studies (CRS Web), which includes Cochrane Epilepsy's Specialized Register and CENTRAL; MEDLINE; the US National Institutes of Health Ongoing Trials Register (ClinicalTrials.gov); and the World Health Organization International Clinical Trials Registry Platform (ICTRP). We handsearched relevant journals and contacted pharmaceutical companies, original trial investigators, and experts in the field.  SELECTION CRITERIA: Randomised controlled trials comparing monotherapy with either phenobarbitone or phenytoin in children or adults with focal onset seizures or generalised onset tonic-clonic seizures.

DATA COLLECTION AND ANALYSIS

This was an individual participant data (IPD), review. Our primary outcome was time to treatment failure. Our secondary outcomes were time to first seizure post-randomisation, time to six-month remission and time to 12-month remission. We used Cox proportional hazards regression models to obtain trial-specific estimates of hazard ratios (HRs), with 95% confidence intervals (CIs), using the generic inverse variance method to obtain the overall pooled HR and 95% CI.

MAIN RESULTS

Individual participant data were obtained for five studies, which recruited a total of 635 participants, representing 80% of 798 individuals from all seven identified eligible trials. For remission outcomes, an HR of less than 1 indicates an advantage for phenytoin and for first seizure and treatment failure outcomes an HR of less than 1 indicates an advantage for phenobarbitone.Results for the primary outcome of the review were: time to treatment failure for any reason related to treatment (pooled HR adjusted for seizure type for 499 participants: 1.61, 95% CI 1.22 to 2.12, low-certainty evidence), time to treatment failure due to adverse events (pooled HR adjusted for seizure type for 499 participants: 1.99, 95% CI 1.37 to 2.87, low-certainty evidence), time to treatment failure due to lack of efficacy (pooled HR adjusted for seizure type for 499 participants: 1.87, 95% CI 1.32 to 2.66, moderate-certainty evidence), showing a statistically significant advantage for phenytoin compared to phenobarbitone.For our secondary outcomes, we did not find any statistically significant differences between phenytoin and phenobarbitone: time to first seizure post-randomisation (pooled HR adjusted for seizure type for 624 participants: 0.85, 95% CI 0.69 to 1.06, moderate-certainty evidence), time to 12-month remission (pooled HR adjusted for seizure type for 588 participants: 0.90, 95% CI 0.69 to 1.19, moderate-certainty evidence), and time to six-month remission pooled HR adjusted for seizure type for 588 participants: 0.91, 95% CI 0.71 to 1.15, moderate-certainty evidence).For individuals with focal onset seizures (73% of individuals contributing to analysis), numerical results were similar and conclusions the same as for analyses of all individuals and for individuals with generalised onset seizures (27% of individuals contributing to analysis), results were imprecise and no clear differences between the drugs were observed.Several confounding factors, most notably the differences in design of the trials with respect to blinding, were likely to have impacted on the results of the primary outcome 'time to treatment failure', and in turn, the treatment failure rates may have impacted on the secondary efficacy outcomes of time to first seizure and time to 12-month and six-month remission.

AUTHORS' CONCLUSIONS

Low-certainty evidence from this review suggests that phenytoin may be a more effective drug than phenobarbitone in terms of treatment retention (treatment failures due to lack of efficacy or adverse events or both). Moderate-certainty evidence from this review also indicates no differences between the drugs in terms of time to seizure recurrence and seizure remission.However, the trials contributing to the analyses had methodological inadequacies and methodological design differences that may have impacted upon the results of this review. Therefore, we do not suggest that results of this review alone should form the basis of a treatment choice for a patient with newly onset seizures. We recommend that future trials should be designed to the highest quality possible with consideration of masking, choice of population, classification of seizure type, duration of follow-up, choice of outcomes and analysis, and presentation of results.

Carbamazepine versus phenytoin monotherapy for epilepsy: an individual participant data review

BACKGROUND

This is an update of a Cochrane Review first published in 2002 and last updated in 2017. This review is one in a series of Cochrane Reviews investigating pair-wise monotherapy comparisons.Epilepsy is a common neurological condition in which abnormal electrical discharges from the brain cause recurrent unprovoked seizures. It is believed that with effective drug treatment, up to 70% of individuals with active epilepsy have the potential to become seizure-free and go into long-term remission shortly after starting drug therapy with a single antiepileptic drug in monotherapy.Worldwide, carbamazepine and phenytoin are commonly-used broad spectrum antiepileptic drugs, suitable for most epileptic seizure types. Carbamazepine is a current first-line treatment for focal onset seizures in the USA and Europe. Phenytoin is no longer considered a first-line treatment, due to concerns over adverse events associated with its use, but the drug is still commonly used in low- to middle-income countries because of its low cost. No consistent differences in efficacy have been found between carbamazepine and phenytoin in individual trials; however, the confidence intervals generated by these trials are wide, and therefore, synthesising the data of the individual trials may show differences in efficacy.

OBJECTIVES

To review the time to treatment failure, remission and first seizure with carbamazepine compared with phenytoin when used as monotherapy in people with focal onset seizures (simple or complex focal and secondarily generalised), or generalised onset tonic-clonic seizures (with or without other generalised seizure types).

SEARCH METHODS

For the latest update, we searched the following databases on 13 August 2018: the Cochrane Register of Studies (CRS Web), which includes the Cochrane Epilepsy's Specialised Register and CENTRAL; MEDLINE; the US National Institutes of Health Ongoing Trials Register (ClinicalTrials.gov); and the World Health Organization International Clinical Trials Registry Platform (ICTRP). We handsearched relevant journals and contacted pharmaceutical companies, original trial investigators, and experts in the field.

SELECTION CRITERIA

Randomised controlled trials comparing monotherapy with either carbamazepine or phenytoin in children or adults with focal onset seizures or generalised onset (tonic-clonic) seizures.

DATA COLLECTION AND ANALYSIS

This was an individual participant data (IPD) review. Our primary outcome was time to treatment failure. Our secondary outcomes were time to first seizure post-randomisation, time to six-month remission, time to 12-month remission, and incidence of adverse events. We used Cox proportional hazards regression models to obtain trial-specific estimates of hazard ratios (HRs), with 95% confidence intervals (CIs), using the generic inverse variance method to obtain the overall pooled HR and 95% CI.

MAIN RESULTS

IPD were available for 595 participants out of 1102 eligible individuals, from four out of 11 trials (i.e. 54% of the potential data). For remission outcomes, a HR greater than 1 indicates an advantage for phenytoin; and for first seizure and withdrawal outcomes, a HR greater than 1 indicates an advantage for carbamazepine. Most participants included in analysis (78%) were classified as experiencing focal onset seizures at baseline and only 22% were classified as experiencing generalised onset seizures; the results of this review are therefore mainly applicable to individuals with focal onset seizures.Results for the primary outcome of the review were: time to treatment failure for any reason related to treatment (pooled HR adjusted for seizure type for 546 participants: 0.94, 95% CI 0.70 to 1.26, moderate-certainty evidence); time to treatment failure due to lack of efficacy (pooled HR adjusted for seizure type for 546 participants: 0.99, 95% CI 0.69 to 1.41, moderate-certainty evidence); both showing no clear difference between the drugs and time to treatment failure due to adverse events (pooled HR adjusted for seizure type for 546 participants: 1.27, 95% CI 0.87 to 1.86, moderate-certainty evidence), showing that treatment failure due to adverse events may occur earlier on carbamazepine than phenytoin, but we cannot rule out a slight advantage to carbamazepine or no difference between the drugs.For our secondary outcomes (pooled HRs adjusted for seizure type), we did not find any clear differences between carbamazepine and phenytoin: time to first seizure post-randomisation (582 participants): 1.15, 95% CI 0.94 to 1.40, moderate-certainty evidence); time to 12-month remission (551 participants): 1.00, 95% CI 0.79 to 1.26, moderate-certainty evidence); and time to six-month remission (551 participants): 0.90, 95% CI 0.73 to 1.12, moderate-certainty evidence).For all outcomes, results for individuals with focal onset seizures were similar to overall results (moderate-certainty evidence), and results for the small subgroup of individuals with generalised onset seizures were imprecise, so we cannot rule out an advantage to either drug, or no difference between drugs (low-certainty evidence). There was also evidence that misclassification of seizure type may have confounded the results of this review, particularly for the outcome 'time to treatment failure'. Heterogeneity was present in analysis of 'time to first seizure' for individuals with generalised onset seizures, which could not be explained by subgroup analysis or sensitivity analyses.Limited information was available about adverse events in the trials and we could not compare the rates of adverse events between carbamazepine and phenytoin. Some adverse events reported on both drugs were abdominal pain, nausea, and vomiting, drowsiness, motor and cognitive disturbances, dysmorphic side effects (such as rash).

AUTHORS' CONCLUSIONS

Moderate-certainty evidence provided by this systematic review does not show any differences between carbamazepine and phenytoin in terms of effectiveness (retention) or efficacy (seizure recurrence and seizure remission) for individuals with focal onset or generalised onset seizures.However, some of the trials contributing to the analyses had methodological inadequacies and inconsistencies, which may have had an impact on the results of this review. We therefore do not suggest that results of this review alone should form the basis of a treatment choice for a person with newly-onset seizures. We did not find any evidence to support or refute current treatment policies. We implore that future trials be designed to the highest quality possible, with consideration of masking, choice of population, classification of seizure type, duration of follow-up, choice of outcomes and analysis, and presentation of results.

Carbamazepine versus phenobarbitone monotherapy for epilepsy: an individual participant data review

BACKGROUND

This is an updated version of the Cochrane Review previously published in 2016. This review is one in a series of Cochrane Reviews investigating pair-wise monotherapy comparisons.Epilepsy is a common neurological condition in which abnormal electrical discharges from the brain cause recurrent unprovoked seizures. It is believed that with effective drug treatment, up to 70% of individuals with active epilepsy have the potential to become seizure-free and go into long-term remission shortly after starting drug therapy with a single antiepileptic drug in monotherapy.Worldwide, carbamazepine and phenobarbitone are commonly used broad-spectrum antiepileptic drugs, suitable for most epileptic seizure types. Carbamazepine is a current first-line treatment for focal onset seizures, and is used in the USA and Europe. Phenobarbitone is no longer considered a first-line treatment because of concerns over associated adverse events, particularly documented behavioural adverse events in children treated with the drug. However, phenobarbitone is still commonly used in low- and middle-income countries because of its low cost. No consistent differences in efficacy have been found between carbamazepine and phenobarbitone in individual trials; however, the confidence intervals generated by these trials are wide, and therefore, synthesising the data of the individual trials may show differences in efficacy.

OBJECTIVES

To review the time to treatment failure, remission and first seizure with carbamazepine compared with phenobarbitone when used as monotherapy in people with focal onset seizures (simple or complex focal and secondarily generalised), or generalised onset tonic-clonic seizures (with or without other generalised seizure types).

SEARCH METHODS

For the latest update, we searched the following databases on 24 May 2018: the Cochrane Register of Studies (CRS Web), which includes Cochrane Epilepsy's Specialized Register and CENTRAL; MEDLINE; the US National Institutes of Health Ongoing Trials Register (ClinicalTrials.gov); and the World Health Organization International Clinical Trials Registry Platform (ICTRP). We handsearched relevant journals and contacted pharmaceutical companies, original trial investigators, and experts in the field.

SELECTION CRITERIA

Randomised controlled trials comparing monotherapy with either carbamazepine or phenobarbitone in children or adults with focal onset seizures or generalised onset tonic-clonic seizures.

DATA COLLECTION AND ANALYSIS

This was an individual participant data (IPD), review. Our primary outcome was time to treatment failure. Our secondary outcomes were time to first seizure post-randomisation, time to six-month remission, time to 12-month remission, and incidence of adverse events. We used Cox proportional hazards regression models to obtain trial-specific estimates of hazard ratios (HRs), with 95% confidence intervals (CIs), using the generic inverse variance method to obtain the overall pooled HR and 95% CI.

MAIN RESULTS

We included 13 trials in this review and IPD were available for 836 individuals out of 1455 eligible individuals from six trials, 57% of the potential data. For remission outcomes, a HR of less than 1 indicates an advantage for phenobarbitone and for first seizure and treatment failure outcomes a HR of less than 1 indicates an advantage for carbamazepine.Results for the primary outcome of the review were: time to treatment failure for any reason related to treatment (pooled HR adjusted for seizure type for 676 participants: 0.66, 95% CI 0.50 to 0.86, moderate-quality evidence), time to treatment failure due to adverse events (pooled HR adjusted for seizure type for 619 participants: 0.69, 95% CI 0.49 to 0.97, low-quality evidence), time to treatment failure due to lack of efficacy (pooled HR adjusted for seizure type for 487 participants: 0.54, 95% CI 0.38 to 0.78, moderate-quality evidence), showing a statistically significant advantage for carbamazepine compared to phenobarbitone.For our secondary outcomes, we did not find any statistically significant differences between carbamazepine and phenobarbitone: time to first seizure post-randomisation (pooled HR adjusted for seizure type for 822 participants: 1.13, 95% CI 0.93 to 1.38, moderate-quality evidence), time to 12-month remission (pooled HR adjusted for seizure type for 683 participants: 1.09, 95% CI 0.84 to 1.40, low-quality evidence), and time to six-month remission pooled HR adjusted for seizure type for 683 participants: 1.01, 95% CI 0.81 to 1.24, low-quality evidence).Results of these secondary outcomes suggest that there may be an association between treatment effect in terms of efficacy and seizure type; that is, that participants with focal onset seizures experience seizure recurrence later and hence remission of seizures earlier on phenobarbitone than carbamazepine, and vice versa for individuals with generalised seizures. It is likely that the analyses of these outcomes were confounded by several methodological issues and misclassification of seizure type, which could have introduced the heterogeneity and bias into the results of this review.Limited information was available regarding adverse events in the trials and we could not compare the rates of adverse events between carbamazepine and phenobarbitone. Some adverse events reported on both drugs were abdominal pain, nausea, and vomiting, drowsiness, motor and cognitive disturbances, dysmorphic side effects (such as rash), and behavioural side effects in three paediatric trials.

AUTHORS' CONCLUSIONS

Moderate-quality evidence from this review suggests that carbamazepine is likely to be a more effective drug than phenobarbitone in terms of treatment retention (treatment failures due to lack of efficacy or adverse events or both). Moderate- to low-quality evidence from this review also suggests an association between treatment efficacy and seizure type in terms of seizure recurrence and seizure remission, with an advantage for phenobarbitone for focal onset seizures and an advantage for carbamazepine for generalised onset seizures.However, some of the trials contributing to the analyses had methodological inadequacies and inconsistencies that may have impacted upon the results of this review. Therefore, we do not suggest that results of this review alone should form the basis of a treatment choice for a patient with newly onset seizures. We recommend that future trials should be designed to the highest quality possible with consideration of masking, choice of population, classification of seizure type, duration of follow-up, choice of outcomes and analysis, and presentation of results.

Oxcarbazepine versus phenytoin monotherapy for epilepsy: an individual participant data review

BACKGROUND

This is an updated version of the Cochrane Review previously published in 2013. This review is one in a series of Cochrane Reviews investigating pair-wise monotherapy comparisons.Epilepsy is a common neurological condition in which abnormal electrical discharges from the brain cause recurrent unprovoked seizures. It is believed that with effective drug treatment, up to 70% of individuals with active epilepsy have the potential to become seizure-free and go into long-term remission shortly after starting drug therapy with a single antiepileptic drug in monotherapy.Worldwide, phenytoin is a commonly used antiepileptic drug. It is important to know how newer drugs, such as oxcarbazepine, compare with commonly used standard treatments.

OBJECTIVES

To review the time to treatment failure, remission and first seizure with oxcarbazepine compared to phenytoin, when used as monotherapy in people with focal onset seizures or generalised tonic-clonic seizures (with or without other generalised seizure types).

SEARCH METHODS

We searched the following databases on 20 August 2018: the Cochrane Register of Studies (CRS Web), which includes the Cochrane Epilepsy Group Specialized Register and the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE (Ovid, 1946 to 20 August 2018), ClinicalTrials.gov, and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP). We handsearched relevant journals and contacted pharmaceutical companies, original trial investigators and experts in the field.

SELECTION CRITERIA

We included randomised controlled trials comparing monotherapy with either oxcarbazepine or phenytoin in children or adults with focal onset seizures or generalised onset tonic-clonic seizures.

DATA COLLECTION AND ANALYSIS

This was an individual participant data (IPD) review. Our primary outcome was time to treatment failure and our secondary outcomes were time to first seizure post-randomisation, time to six-month and 12-month remission, and incidence of adverse events. We used Cox proportional hazards regression models to obtain trial-specific estimates of hazard ratios (HRs) with 95% confidence intervals (CIs), using the generic inverse variance method to obtain the overall pooled HR and 95% CI.

MAIN RESULTS

Individual participant data were available for 480 out of a total of 517 participants (93%), from two out of three included trials. For remission outcomes, a HR of less than one indicated an advantage for phenytoin; and for first seizure and treatment failure outcomes, a HR of less than one indicated an advantage for oxcarbazepine.The results for time to treatment failure for any reason related to treatment showed a potential advantage of oxcarbazepine over phenytoin, but this was not statistically significant (pooled HR adjusted for epilepsy type: 0.78 95% CI 0.53 to 1.14, 476 participants, two trials, moderate-quality evidence). Our analysis showed that treatment failure due to adverse events occurred later on with oxcarbazepine than phenytoin (pooled HR for all participants: 0.22 (95% CI 0.10 to 0.51, 480 participants, two trials, high-quality evidence). Our analysis of time to treatment failure due to lack of efficacy showed no clear difference between the drugs (pooled HR for all participants: 1.17 (95% CI 0.31 to 4.35), 480 participants, two trials, moderate-quality evidence).We found no clear or statistically significant differences between drugs for any of the secondary outcomes of the review: time to first seizure post-randomisation (pooled HR adjusted for epilepsy type: 0.97 95% CI 0.75 to 1.26, 468 participants, two trials, moderate-quality evidence); time to 12-month remission (pooled HR adjusted for epilepsy type 1.04 95% CI 0.77 to 1.41, 468 participants, two trials, moderate-quality evidence) and time to six-month remission (pooled HR adjusted for epilepsy type: 1.06 95% CI 0.82 to 1.36, 468 participants, two trials, moderate-quality evidence).The most common adverse events reported in more than 10% of participants on either drug were somnolence (28% of total participants, with similar rates for both drugs), headache (15% of total participants, with similar rates for both drugs), dizziness (14.5% of total participants, reported by slightly more participants on phenytoin (18%) than oxcarbazepine (11%)) and gum hyperplasia (reported by substantially more participants on phenytoin (18%) than oxcarbazepine (2%)).The results of this review are applicable mainly to individuals with focal onset seizures; 70% of included individuals experienced seizures of this type at baseline. The two studies included in IPD meta-analysis were generally of good methodological quality but the design of the studies may have biased the results for the secondary outcomes (time to first seizure post-randomisation, time to six-month and 12-month remission) as seizure recurrence data were not collected following treatment failure or withdrawal from the study. In addition, misclassification of epilepsy type may have impacted on results, particularly for individuals with generalised onset seizures.

AUTHORS' CONCLUSIONS

High-quality evidence provided by this review indicates that treatment failure due to adverse events occurs significantly later with oxcarbazepine than phenytoin. For individuals with focal onset seizures, moderate-quality evidence suggests that oxcarbazepine may be superior to phenytoin in terms of treatment failure for any reason, seizure recurrence and seizure remission. Therefore, oxcarbazepine may be a preferable alternative treatment than phenytoin, particularly for individuals with focal onset seizures. The evidence in this review which relates to individuals with generalised onset seizures is of low quality and does not inform current treatment policy.We recommend that future trials should be designed to the highest quality possible with regards to choice of population, classification of seizure type, duration of follow-up (including continued follow-up after failure or withdrawal of randomised treatment), choice of outcomes and analysis, and presentation of results.

Sodium valproate versus phenytoin monotherapy for epilepsy: an individual participant data review

BACKGROUND

Epilepsy is a common neurological condition in which abnormal electrical discharges from the brain cause recurrent unprovoked seizures. It is believed that with effective drug treatment up to 70% of individuals with active epilepsy have the potential to become seizure-free, and to go into long-term remission shortly after starting drug therapy with a single antiepileptic drug in monotherapy.Worldwide, sodium valproate and phenytoin are commonly used antiepileptic drugs for monotherapy treatment. It is generally believed that phenytoin is more effective for focal onset seizures, and that sodium pvalproate is more effective for generalised onset tonic-clonic seizures (with or without other generalised seizure types). This review is one in a series of Cochrane Reviews investigating pair-wise monotherapy comparisons. This is the latest updated version of the review first published in 2001, and updated in 2013 and 2016.

OBJECTIVES

To review the time to treatment failure, remission and first seizure of sodium valproate compared to phenytoin when used as monotherapy in people with focal onset seizures or generalised tonic-clonic seizures (with or without other generalised seizure types).

SEARCH METHODS

We searched the Cochrane Epilepsy Group's Specialised Register, the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, ClinicalTrials.gov and the World Health Organization (WHO) International Clinical Trials Registry Platform ICTRP on 19 February 2018. We handsearched relevant journals, contacted pharmaceutical companies, original trial investigators and experts in the field.

SELECTION CRITERIA

Randomised controlled trials (RCTs) comparing monotherapy with either sodium valproate or phenytoin in children or adults with focal onset seizures or generalised onset tonic-clonic seizures DATA COLLECTION AND ANALYSIS: This was an individual participant data (IPD) review. Our primary outcome was time to treatment failure and our secondary outcomes were time to first seizure post-randomisation, time to six-month, and 12-month remission, and incidence of adverse events. We used Cox proportional hazards regression models to obtain trial-specific estimates of hazard ratios (HRs) with 95% confidence intervals (CIs), using the generic inverse variance method to obtain the overall pooled HR and 95% CI.

MAIN RESULTS

We included 11 trials in this review and IPD were available for 669 individuals out of 1119 eligible individuals from five out of 11 trials, 60% of the potential data. Results apply to focal onset seizures (simple, complex and secondary generalised tonic-clonic seizures), and generalised tonic-clonic seizures, but not other generalised seizure types (absence or myoclonus seizure types). For remission outcomes, a HR of less than 1 indicates an advantage for phenytoin, and for first seizure and treatment failure outcomes a HR of less than 1 indicates an advantage for sodium valproate.The main overall results were: time to treatment failure for any reason related to treatment (pooled HR adjusted for seizure type 0.88, 95% CI 0.61 to 1.27; 5 studies; 528 participants; moderate-quality evidence), time to treatment failure due to adverse events (pooled HR adjusted for seizure type 0.77, 95% CI 0.44 to 1.37; 4 studies; 418 participants; moderate-quality evidence), time to treatment failure due to lack of efficacy (pooled HR for all participants 1.16 (95% CI 0.71 to 1.89; 5 studies; 451 participants; moderate-quality evidence). These results suggest that treatment failure for any reason related to treatment and treatment failure due to adverse events may occur earlier on phenytoin compared to sodium valproate, while treatment failure due to lack of efficacy may occur earlier on sodium valproate than phenytoin; however none of these results were statistically significant.Results for time to first seizure (pooled HR adjusted for seizure type 1.08, 95% CI 0.88 to 1.33; 5 studies; 639 participants; low-quality evidence) suggest that first seizure recurrence may occur slightly earlier on sodium valproate compared to phenytoin. There were no clear differences between drugs in terms of time to 12-month remission (pooled HR adjusted for seizure type 1.02, 95% CI 0.81 to 1.28; 4 studies; 514 participants; moderate-quality evidence) and time to six-month remission (pooled HR adjusted for seizure type 1.05, 95% CI 0.86 to 1.27; 5 studies; 639 participants; moderate-quality evidence).Limited information was available regarding adverse events in the trials and we could not make comparisons between the rates of adverse events on sodium valproate and phenytoin. Some adverse events reported with both drugs were drowsiness, rash, dizziness, nausea and gastrointestinal problems. Weight gain was also reported with sodium valproate and gingival hypertrophy/hyperplasia was reported on phenytoin.The methodological quality of the included trials was generally good, however four out of the five trials providing IPD for analysis were of an open-label design, therefore all results were at risk of detection bias. There was also evidence that misclassification of seizure type may have confounded the results of this review, particularly for the outcome 'time to first seizure' and heterogeneity was present in analysis of treatment failure outcomes which could not be explained by subgroup analysis by epilepsy type or by sensitivity analysis for misclassification of seizure type. Therefore, for treatment failure outcomes we judged the quality of the evidence to be moderate to low, for 'time to first seizure' we judged the quality of the evidence to be low, and for remission outcomes we judged the quality of the evidence to be moderate.

AUTHORS' CONCLUSIONS

We have not found evidence that a significant difference exists between valproate and phenytoin for any of the outcomes examined in this review. However detection bias, classification bias and heterogeneity may have impacted on the results of this review. We did not find any outright evidence to support or refute current treatment policies. We recommend that future trials be designed to the highest quality possible with consideration of masking, choice of population, classification of seizure type, duration of follow-up, choice of outcomes and analysis, and presentation of results.

Lamotrigine versus carbamazepine monotherapy for epilepsy: an individual participant data review

BACKGROUND

This is an updated version of the original Cochrane Review published in Issue 11, 2006 of the Cochrane Database of Systematic Reviews.Epilepsy is a common neurological condition in which abnormal electrical discharges from the brain cause recurrent unprovoked seizures. It is believed that with effective drug treatment up to 70% of individuals with active epilepsy have the potential to become seizure-free, and to go into long-term remission shortly after starting drug therapy with a single antiepileptic drug (AED) in monotherapy.The correct choice of first-line AED for individuals with newly diagnosed seizures is of great importance. It is important that the choice of AEDs for an individual is made using the highest quality evidence regarding the potential benefits and harms of the various treatments.Carbamazepine or lamotrigine are recommended as first-line treatments for new onset focal seizures and as a first- or second-line treatment for generalised tonic-clonic seizures. Performing a synthesis of the evidence from existing trials will increase the precision of the results for outcomes relating to efficacy and tolerability and may assist in informing a choice between the two drugs.

OBJECTIVES

To review the time to treatment failure, remission and first seizure with lamotrigine compared to carbamazepine when used as monotherapy in people with focal onset seizures (simple or complex focal and secondarily generalised) or generalised onset tonic-clonic seizures (with or without other generalised seizure types).

SEARCH METHODS

We conducted the first searches for this review in 1997. For the most recent update, we searched the Cochrane Epilepsy Group Specialized Register, the Cochrane Central Register of Controlled Trials (CENTRAL) via the Cochrane Register of Studies Online (CRSO), MEDLINE, Clinical Trials.gov and the WHO International Clinical Trials Registry Platform on 26 February 2018, without language restrictions SELECTION CRITERIA: Randomised controlled trials comparing monotherapy with either carbamazepine or lamotrigine in children or adults with focal onset seizures or generalised onset tonic-clonic seizures DATA COLLECTION AND ANALYSIS: This was an individual participant data (IPD) review. Our primary outcome was time to treatment failure and our secondary outcomes were time to first seizure post randomisation, time to six-month, 12-month and 24-month remission, and incidence of adverse events. We used Cox proportional hazards regression models to obtain trial-specific estimates of hazard ratios (HRs) with 95% confidence intervals (CIs), using the generic inverse variance method to obtain the overall pooled HR and 95% CI.

MAIN RESULTS

We included 14 trials in this review. Individual participant data were available for 2572 participants out of 3787 eligible individuals from nine out of 14 trials: 68% of the potential data. For remission outcomes, a HR of less than one indicated an advantage for carbamazepine; and for first seizure and treatment failure outcomes, a HR of less than one indicated an advantage for lamotrigine.The main overall results were: time to treatment failure for any reason related to treatment (pooled HR adjusted for seizure type: 0.71, 95% CI 0.62 to 0.82, moderate-quality evidence), time to treatment failure due to adverse events (pooled HR adjusted for seizure type: 0.55 (95% CI 0.45 to 0.66, moderate-quality evidence), time to treatment failure due to lack of efficacy (pooled HR for all participants: 1.03 (95% CI 0.75 to 1.41), moderate-quality evidence) showing a significant advantage for lamotrigine compared to carbamazepine in terms of treatment failure for any reason related to treatment and treatment failure due to adverse events, but no different between drugs for treatment failure due to lack of efficacy.Time to first seizure (pooled HR adjusted for seizure type: 1.26, 95% CI 1.12 to 1.41, high-quality evidence) and time to six-month remission (pooled HR adjusted for seizure type: 0.86, 95% CI 0.76 to 0.97, high-quality evidence), showed a significant advantage for carbamazepine compared to lamotrigine for first seizure and six-month remission. We found no difference between the drugs for time to 12-month remission (pooled HR for all participants 0.91, 95% CI 0.77 to 1.07, high-quality evidence) or time to 24-month remission (HR for all participants 1.00, 95% CI 0.80 to 1.25, high-quality evidence), however only two trials followed up participants for more than one year so evidence is limited.The results of this review are applicable mainly to individuals with focal onset seizures; 88% of included individuals experienced seizures of this type at baseline. Up to 50% of the limited number of individuals classified as experiencing generalised onset seizures at baseline may have had their seizure type misclassified, therefore we recommend caution when interpreting the results of this review for individuals with generalised onset seizures.The most commonly reported adverse events for both of the drugs across all of the included trials were dizziness, fatigue, gastrointestinal disturbances, headache and skin problems. The rate of adverse events was similar across the two drugs.The methodological quality of the included trials was generally good, however there is some evidence that the design choice of masked or open-label treatment may have influenced the treatment failure and withdrawal rates of the trials. Hence, we judged the quality of the evidence for the primary outcome of treatment failure to be moderate for individuals with focal onset seizures and low for individuals with generalised onset seizures. For efficacy outcomes (first seizure, remission), we judged the quality of evidence to be high for individuals with focal onset seizures and moderate for individuals with generalised onset seizures.

AUTHORS' CONCLUSIONS

Moderate quality evidence indicates that treatment failure for any reason related to treatment or due to adverse events occurs significantly earlier on carbamazepine than lamotrigine, but the results for time to first seizure suggested that carbamazepine may be superior in terms of seizure control. The choice between these first-line treatments must be made with careful consideration. We recommend that future trials should be designed to the highest quality possible with consideration of masking, choice of population, classification of seizure type, duration of follow-up, choice of outcomes and analysis, and presentation of results.

Exploring changes over time and characteristics associated with data retrieval across individual participant data meta-analyses: systematic review

Objective To investigate whether the success rate of retrieving individual participant data (IPD) for use in IPD meta-analyses has increased over time, and to explore the characteristics associated with IPD retrieval.Design Systematic review of published IPD meta-analyses, supplemented by a reflection of the Cochrane Epilepsy Group's 20 years' experience of requesting IPD.Data sources Medline, CENTRAL, Scopus, Web of Science, CINAHL Plus, and PsycINFO.Eligibility criteria for study selection IPD meta-analyses of studies of all designs and all clinical areas published in English.Results 760 IPD meta-analyses which identified studies by systematic methods that had been published between 1987 and 2015 were included. Only 188 (25%) of these IPD meta-analyses retrieved 100% of the eligible IPD for analysis, with 324 (43%) of these IPD meta-analyses retrieving 80% or more of relevant IPD. There is insufficient evidence to suggest that IPD retrieval rates have improved over time. IPD meta-analyses that included only randomised trials, had an authorship policy, included fewer eligible participants, and were conducted outside of the Cochrane Database of Systematic Reviews were associated with a high or complete IPD retrieval rate. There was no association between the source of funding of the IPD meta-analyses and IPD retrieval rate. The IPD retrieval rate of the Cochrane Epilepsy Group has declined from 83% (up to 2005) to 65% (between 2012 and 2015) and the reported reasons for lack of data availability have changed in recent years.Conclusions IPD meta-analyses are considered to be the "gold standard" for the synthesis of data from clinical research studies; however, only 25% of published IPD meta-analyses have had access to all IPD.

Lamotrigine versus carbamazepine monotherapy for epilepsy: an individual participant data review

BACKGROUND

This is an updated version of the original Cochrane review published in Issue 1, 2006 of the Cochrane Database of Systematic Reviews.Epilepsy is a common neurological condition in which abnormal electrical discharges from the brain cause recurrent unprovoked seizures. It is believed that with effective drug treatment up to 70% of individuals with active epilepsy have the potential to become seizure-free and to go into long-term remission shortly after starting drug therapy with a single antiepileptic drug (AED) in monotherapy.The correct choice of first-line antiepileptic therapy for individuals with newly diagnosed seizures is of great importance. It is important that the choice of AEDs for an individual is made using the highest quality evidence regarding the potential benefits and harms of the various treatments. It is also important that the effectiveness and tolerability of AEDs appropriate to given seizure types are compared to one another.Carbamazepine or lamotrigine are first-line recommended treatments for new onset partial seizures and as a first- or second-line treatment for generalised tonic-clonic seizures. Performing a synthesis of the evidence from existing trials will increase the precision of the results for outcomes relating to efficacy and tolerability and may assist in informing a choice between the two drugs.

OBJECTIVES

To review the time to withdrawal, remission and first seizure with lamotrigine compared to carbamazepine when used as monotherapy in people with partial onset seizures (simple or complex partial and secondarily generalised) or generalised onset tonic-clonic seizures (with or without other generalised seizure types).

SEARCH METHODS

The first searches for this review were run in 1997. For the most recent update we searched the Cochrane Epilepsy Group Specialized Register (17 October 2016), the Cochrane Central Register of Controlled Trials (CENTRAL) via the Cochrane Register of Studies Online (CRSO, 17 October 2016) and MEDLINE (Ovid, 1946 to 17 October 2016). We imposed no language restrictions. We also contacted pharmaceutical companies and trial investigators.

SELECTION CRITERIA

Randomised controlled trials in children or adults with partial onset seizures or generalised onset tonic-clonic seizures comparing monotherapy with either carbamazepine or lamotrigine.

DATA COLLECTION AND ANALYSIS

This was an individual participant data (IPD) review. Our primary outcome was time to withdrawal of allocated treatment and our secondary outcomes were time to first seizure post-randomisation, time to six-month, 12-month and 24-month remission, and incidence of adverse events. We used Cox proportional hazards regression models to obtain trial-specific estimates of hazard ratios (HRs) with 95% confidence intervals (CIs), using the generic inverse variance method to obtain the overall pooled HR and 95% CI.

MAIN RESULTS

We included 13 studies in this review. Individual participant data were available for 2572 participants out of 3394 eligible individuals from nine out of 13 trials: 78% of the potential data. For remission outcomes, a HR < 1 indicated an advantage for carbamazepine and for first seizure and withdrawal outcomes a HR < 1 indicated an advantage for lamotrigine.The main overall results (pooled HR adjusted for seizure type) were: time to withdrawal of allocated treatment (HR 0.72, 95% CI 0.63 to 0.82), time to first seizure (HR 1.22, 95% CI 1.09 to 1.37) and time to six-month remission (HR 0.84, 95% CI 0.74 to 0.94), showing a significant advantage for lamotrigine compared to carbamazepine for withdrawal but a significant advantage for carbamazepine compared to lamotrigine for first seizure and six-month remission. We found no difference between the drugs for time to 12-month remission (HR 0.91, 95% CI 0.77 to 1.07) or time to 24-month remission (HR 1.00, 95% CI 0.80 to 1.25), however only two trials followed up participants for more than one year so the evidence is limited.The results of this review are applicable mainly to individuals with partial onset seizures; 88% of included individuals experienced seizures of this type at baseline. Up to 50% of the limited number of individuals classified as experiencing generalised onset seizures at baseline may have had their seizure type misclassified, therefore we recommend caution when interpreting the results of this review for individuals with generalised onset seizures.The most commonly reported adverse events for both of the drugs across all of the included trials were dizziness, fatigue, gastrointestinal disturbances, headache and skin problems. The rate of adverse events was similar across the two drugs.The methodological quality of the included trials was generally good, however there is some evidence that the design choice of masked or open-label treatment may have influenced the withdrawal rates of the trials. Hence, we judged the quality of the evidence for the primary outcome of treatment withdrawal to be moderate for individuals with partial onset seizures and low for individuals with generalised onset seizures. For efficacy outcomes (first seizure, remission), we judged the quality of evidence to be high for individuals with partial onset seizures and moderate for individuals with generalised onset seizures.

AUTHORS' CONCLUSIONS

Lamotrigine was significantly less likely to be withdrawn than carbamazepine but the results for time to first seizure suggested that carbamazepine may be superior in terms of seizure control. A choice between these first-line treatments must be made with careful consideration. We recommend that future trials should be designed to the highest quality possible with consideration of masking, choice of population, classification of seizure type, duration of follow-up, choice of outcomes and analysis, and presentation of results.

Individual participant data meta-analyses compared with meta-analyses based on aggregate data

BACKGROUND

Meta-analyses based on individual participant data (IPD-MAs) allow more powerful and uniformly consistent analyses as well as better characterisation of subgroups and outcomes, compared to those which are based on aggregate data (AD-MAs) extracted from published trial reports. However, IPD-MAs are a larger undertaking requiring greater resources than AD-MAs. Researchers have compared results from IPD-MA against results obtained from AD-MA and reported conflicting findings. We present a methodology review to summarise this empirical evidence .

OBJECTIVES

To review systematically empirical comparisons of meta-analyses of randomised trials based on IPD with those based on AD extracted from published reports, to evaluate the level of agreement between IPD-MA and AD-MA and whether agreement is affected by differences in type of effect measure, trials and participants included within the IPD-MA and AD-MA, and whether analyses were undertaken to explore the main effect of treatment or a treatment effect modifier.

SEARCH METHODS

An electronic search of the Cochrane Library (includes Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effectiveness, CENTRAL, Cochrane Methodology Register, HTA database, NHS Economic Evaluations Database), MEDLINE, and Embase was undertaken up to 7 January 2016. Potentially relevant articles that were known to any of the review authors and reference lists of retrieved articles were also checked.

SELECTION CRITERIA

Studies reporting an empirical comparison of the results of meta-analyses of randomised trials using IPD with those using AD. Studies were included if sufficient numerical data, comparing IPD-MA and AD-MA, were available in their reports.

DATA COLLECTION AND ANALYSIS

Two review authors screened the title and abstract of identified studies with full-text publications retrieved for those identified as eligible or potentially eligible. A 'quality' assessment was done and data were extracted independently by two review authors with disagreements resolved by involving a third author. Data were summarised descriptively for comparisons where an estimate of effect measure and corresponding precision have been provided both for IPD-MA and for AD-MA in the study report. Comparisons have been classified according to whether identical effect measures, identical trials and patients had been used in the IPD-MA and the AD-MA, and whether the analyses were undertaken to explore the main effect of treatment, or to explore a potential treatment effect modifier.Effect measures were transformed to a standardised scale (z scores) and scatter plots generated to allow visual comparisons. For each comparison, we compared the statistical significance (at the 5% two-sided level) of an IPD-MA compared to the corresponding AD-MA and calculated the number of discrepancies. We examined discrepancies by type of analysis (main effect or modifier) and according to whether identical trials, patients and effect measures had been used by the IPD-MA and AD-MA. We calculated the average of differences between IPD-MA and AD-MA (z scores, ratio effect estimates and standard errors (of ratio effects)) and 95% limits of agreement.

MAIN RESULTS

From the 9330 reports found by our searches, 39 studies were eligible for this review with effect estimate and measure of precision extracted for 190 comparisons of IPD-MA and AD-MA. We classified the quality of studies as 'no important flaws' (29 (74%) studies) or 'possibly important flaws' (10 (26%) studies).A median of 4 (interquartile range (IQR): 2 to 6) comparisons were made per study, with 6 (IQR 4 to 11) trials and 1225 (542 to 2641) participants in IPD-MAs and 7 (4 to 11) and 1225 (705 to 2541) for the AD-MAs. One hundred and forty-four (76%) comparisons were made on the main treatment effect meta-analysis and 46 (24%) made using results from analyses to explore treatment effect modifiers.There is agreement in statistical significance between the IPD-MA and AD-MA for 152 (80%) comparisons, 23 of which disagreed in direction of effect. There is disagreement in statistical significance for 38 (20%) comparisons with an excess proportion of IPD-MA detecting a statistically significant result that was not confirmed with AD-MA (28 (15%)), compared with 10 (5%) comparisons with a statistically significant AD-MA that was not confirmed by IPD-MA. This pattern of disagreement is consistent for the 144 main effect analyses but not for the 46 comparisons of treatment effect modifier analyses. Conclusions from some IPD-MA and AD-MA differed even when based on identical trials, participants (but not necessarily identical follow-up) and treatment effect measures. The average difference between IPD-MA and AD-MA in z scores, ratio effect estimates and standard errors is small but limits of agreement are wide and include important differences in both directions. Discrepancies between IPD-MA and AD-MA do not appear to increase as the differences between trials and participants increase.

AUTHORS' CONCLUSIONS

IPD offers the potential to explore additional, more thorough, and potentially more appropriate analyses compared to those possible with AD. But in many cases, similar results and conclusions can be drawn from IPD-MA and AD-MA. Therefore, before embarking on a resource-intensive IPD-MA, an AD-MA should initially be explored and researchers should carefully consider the potential added benefits of IPD.

Treating Newly Diagnosed Epilepsy: The Canadian Choice

Background:

Choosing an antiepileptic medication to treat a patient with epilepsy can be a complicated process during which the treating physician must base her or his decision on efficacy and safety of each of many available drugs. The lack of comparative studies between medications is one of the reasons.

Methods:

We conducted a survey on the management of newly diagnosed epilepsy in adult patients. The surveyed were adult and pediatric neurologists with a subspecialty interest in epilepsy who were working in academic institutions or private practice across Canada. Scenarios presented were grouped in categories according to the epilepsy syndrome (absence epilepsy, juvenile myoclonic epilepsy, undetermined idiopathic generalized epilepsy, symptomatic or cryptogenic partial epilepsy, and unclassified epilepsy), the patient's gender and age. First and second step in medical treatment for status epilepticus were surveyed as well.

Results:

Forty one of 64 experts responded the survey (responder rate of 66%). The results revealed a consensus among Canadian epileptologists that the first choice of antiepileptic medication in generalized epilepsies was between valproate in men (chosen by 88% of respondents) and lamotrigine in women. In localization-related epilepsies, carbamazepine was the preferred drug of choice (chosen by 90% of respondents). In the treatment of status epilepticus, an initial intravenous dose of lorazepam (95% of respondents), followed by a second dose of lorazepam or intravenous phenytoin in case the initial dose of lorazepam failed, were the treatments preferred.

Sharing individual participant data from clinical trials: an opinion survey regarding the establishment of a central repository

BACKGROUND

Calls have been made for increased access to individual participant data (IPD) from clinical trials, to ensure that complete evidence is available. However, despite the obvious benefits, progress towards this is frustratingly slow. In the meantime, many systematic reviews have already collected IPD from clinical trials. We propose that a central repository for these IPD should be established to ensure that these datasets are safeguarded and made available for use by others, building on the strengths and advantages of the collaborative groups that have been brought together in developing the datasets.

OBJECTIVE

Evaluate the level of support, and identify major issues, for establishing a central repository of IPD.

DESIGN

On-line survey with email reminders.

PARTICIPANTS

71 reviewers affiliated with the Cochrane Collaboration's IPD Meta-analysis Methods Group were invited to participate.

RESULTS

30 (42%) invitees responded: 28 (93%) had been involved in an IPD review and 24 (80%) had been involved in a randomised trial. 25 (83%) agreed that a central repository was a good idea and 25 (83%) agreed that they would provide their IPD for central storage. Several benefits of a central repository were noted: safeguarding and standardisation of data, increased efficiency of IPD meta-analyses, knowledge advancement, and facilitating future clinical, and methodological research. The main concerns were gaining permission from trial data owners, uncertainty about the purpose of the repository, potential resource implications, and increased workload for IPD reviewers. Restricted access requiring approval, data security, anonymisation of data, and oversight committees were highlighted as issues under governance of the repository.

CONCLUSION

There is support in this community of IPD reviewers, many of whom are also involved in clinical trials, for storing IPD in a central repository. Results from this survey are informing further work on developing a repository of IPD which is currently underway by our group.

WITHDRAWN: Second or third additional chemotherapy drug for non-small cell lung cancer in patients with advanced disease

BACKGROUND

Randomized trials have demonstrated that adding a drug to a single-agent or to a two-agent regimen increased the tumor response rate in patients with advanced non-small cell lung cancer (NSCLC), although its impact on survival remains controversial.

OBJECTIVES

To evaluate the clinical benefit of adding a drug to a single-agent or two-agent chemotherapy regimen in terms of tumor response rate, survival, and toxicity in patients with advanced NSCLC.

SEARCH METHODS

There were no language restrictions. Searches of MEDLINE and EMBASE were performed using the search terms non-small cell lung carcinoma/drug therapy, adenocarcinoma, large-cell carcinoma, squamous-cell carcinoma, lung, neoplasms, clinical trial phase III, and randomized trial. Manual searches were also performed to find conference proceedings published between January 1982 and June 2006.

SELECTION CRITERIA

Data from all randomized controlled trials performed between 1980 and 2006 (published between January 1980 and June 2006) comparing a doublet regimen with a single-agent regimen or comparing a triplet regimen with a doublet regimen in patients with advanced NSCLC.

DATA COLLECTION AND ANALYSIS

Two independent investigators reviewed the publications and extracted the data. Pooled odds ratios (ORs) for the objective tumor response rate, one-year survival rate, and toxicity rate were calculated using the fixed-effect model. Pooled median ratios (MRs) for median survival also were calculated using the fixed-effect model. ORs and MRs lower than unity (< 1.0) indicate a benefit of a doublet regimen compared with a single-agent regimen (or a triplet regimen compared with a doublet regimen).

MAIN RESULTS

Sixty-five trials (13601 patients) were eligible. In the trials comparing a doublet regimen with a single-agent regimen, a significant increase was observed in tumor response (OR 0.42, 95% confidence interval [CI] 0.37 to 0.47, P < 0.001) and one-year survival (OR 0.80, 95% CI 0.70 to 0.91, P < 0.001) in favor of the doublet regimen. The median survival ratio was 0.83 (95% CI 0.79 to 0.89, P < 0.001). An increase also was observed in the tumor response rate (OR 0.66, 95% CI 0.58 to 0.75, P < 0.001) in favor of the triplet regimen, but not for one-year survival (OR 1.01, 95% CI 0.85 to 1.21, P = 0.88). The median survival ratio was 1.00 (95% CI 0.94 to 1.06, P = 0.97).

AUTHORS' CONCLUSIONS

Adding a second drug improved tumor response and survival rate. Adding a third drug had a weaker effect on tumor response and no effect on survival.

High-dose chemotherapy followed by autologous stem cell transplantation as a first-line therapy for high-risk primary breast cancer: a meta-analysis

Background and Objectives

Several trials have generated conflicting results about the results of high-dose chemotherapy followed by autologous stem cell transplantation (HDCT) for primary breast cancer. This meta-analysis summarizes the available evidence from all suitable studies.

Design and Methods

Prospective, randomized trials with HDCT as a first-line therapy for primary breast cancer were included in this meta-analysis. The primary outcome of interest for our analysis was survival (disease-free survival and overall survival); secondary endpoints included treatment-related mortality (TRM) and second (non-breast) cancers. We used a median age of 47, a PR positive rate of 50% and a premenopausal rate of 70% as cutoff values to complete the subgroup analyses, which were pre-planned according to the prepared protocol.

Results

Fourteen trials with 5747 patients were eligible for the meta-analysis. Compared with non-HDCT, non-significant second (non-breast) cancers (RR = 1.28; 95% CI = 0.82–1.98) and higher TRM (RR = 3.42; 95% CI = 1.32–8.86) were associated with HDCT for primary breast cancer. A significant DFS benefit of HDCT was documented (HR = 0.89; 95% CI = 0.79–0.99). No difference in OS (overall survival) was found when the studies were pooled (HR = 0.91; 95% CI = 0.82–1.00, p = 0.062). In subgroup analysis, age and hormone receptor status had a significant interaction with prolonged DFS and OS.

Conclusions

HDCT has a benefit on DFS and OS compared to SDC in some special patients with high-risk primary breast cancer.

Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves

Background

The results of Randomized Controlled Trials (RCTs) on time-to-event outcomes that are usually reported are median time to events and Cox Hazard Ratio. These do not constitute the sufficient statistics required for meta-analysis or cost-effectiveness analysis, and their use in secondary analyses requires strong assumptions that may not have been adequately tested. In order to enhance the quality of secondary data analyses, we propose a method which derives from the published Kaplan Meier survival curves a close approximation to the original individual patient time-to-event data from which they were generated.

Methods

We develop an algorithm that maps from digitised curves back to KM data by finding numerical solutions to the inverted KM equations, using where available information on number of events and numbers at risk. The reproducibility and accuracy of survival probabilities, median survival times and hazard ratios based on reconstructed KM data was assessed by comparing published statistics (survival probabilities, medians and hazard ratios) with statistics based on repeated reconstructions by multiple observers.

Results

The validation exercise established there was no material systematic error and that there was a high degree of reproducibility for all statistics. Accuracy was excellent for survival probabilities and medians, for hazard ratios reasonable accuracy can only be obtained if at least numbers at risk or total number of events are reported.

Conclusion

The algorithm is a reliable tool for meta-analysis and cost-effectiveness analyses of RCTs reporting time-to-event data. It is recommended that all RCTs should report information on numbers at risk and total number of events alongside KM curves.

Treating osteoporosis in Canada: what clinical efficacy data should be considered by policy decision makers?

SAUMMARY: Using a Markov state-transition model, we estimated fractures averted with risedronate using two different types of clinical efficacy data. Summary data, as opposed to individual patient data (IPD), underestimated the number of fractures averted when applied in a specified high risk population. The choice of clinical efficacy data is an important consideration in health economic models evaluating osteoporosis therapies.

INTRODUCTION

This paper contrasts fracture reduction estimates for risedronate utilizing efficacy data from two approaches to meta-analysis: summary data versus individual patient data. We also examined differences in fracture reduction explained by varied cohort selection, especially the inclusion of low- versus high-risk populations.

METHODS

Using a Markov state-transition model, we compared fractures averted over 3 years in a hypothetical cohort by inputting fracture risk reduction estimates (risedronate versus placebo) from two data sources (summary data versus IPD). The cohort consisted of 100,000 Canadian women, age > or =65 years with osteoporosis (WHO criteria T-score < or = -2.5) and prevalent morphometric vertebral fracture.

RESULTS

Non-vertebral fractures averted with risedronate were: 3,571 and 6,584 per 100,000 women for summary data and IPD, respectively. For vertebral fractures, the numbers were 8,552 and 10,127. When IPD versus summary data was used, an additional 3,013 more non-vertebral fractures and 1,575 vertebral fractures were averted.

DISCUSSION

Relative risk estimates from IPD analyses were the best choice for modelling fracture outcomes when applied in a specified high-risk population. In addition to superior statistical methodology, they utilized RCT cohorts that are more representative of higher risk patients requiring treatment (osteoporotic women > or =65 years with a prevalent vertebral fracture).

Meta-analysis of summary survival curve data

The use of standard univariate fixed- and random-effects models in meta-analysis has become well known in the last 20 years. However, these models are unsuitable for meta-analysis of clinical trials that present multiple survival estimates (usually illustrated by a survival curve) during a follow-up period. Therefore, special methods are needed to combine the survival curve data from different trials in a meta-analysis. For this purpose, only fixed-effects models have been suggested in the literature. In this paper, we propose a multivariate random-effects model for joint analysis of survival proportions reported at multiple time points and in different studies, to be combined in a meta-analysis. The model could be seen as a generalization of the fixed-effects model of Dear (Biometrics 1994; 50:989-1002). We illustrate the method by using a simulated data example as well as using a clinical data example of meta-analysis with aggregated survival curve data. All analyses can be carried out with standard general linear MIXED model software.

A comparison of methods for fixed effects meta-analysis of individual patient data with time to event outcomes

BACKGROUND

Alternative methods for individual patient data (IPD) meta-analysis of time-to-event outcomes have been established and utilized in practice. The most common approach is a stratified log-rank analysis. The IPD approach is considered to be the gold standard approach for meta-analysis and is becoming increasingly more popular but the performance of different methods has not been studied previously.

PURPOSE

To compare commonly used methods for fixed effects meta-analysis of individual patient time-to-event data.

METHODS

The stratified log-rank analysis, an inverse variance weighted average of Cox model estimates, and the stratified Cox regression model are compared. First, a theoretical comparison of approaches is undertaken. Second, the bias and coverage are assessed for the pooled hazard ratio using simulated data under commonly encountered meta-analysis conditions. Finally, a comparison is presented using empirical data from four separate systematic reviews of anti-epileptic drug trials where IPD are available for two time-to-event outcomes.

RESULTS

For hazard ratio close to 1 with minimal heterogeneity between trials, theoretical results suggest similar results should be expected from all the three methods. Results for empirical and simulated data are in keeping with the theoretical results and show all the three methods perform well under these conditions. When there is no heterogeneity and the proportional hazards assumption holds, the stratified Cox model and inverse variance weighted average produce similar estimates of the pooled treatment effect and are to be preferred to the stratified log-rank analysis when the underlying treatment effect is large. Coverage values diminish for all the three methods and are below 95% for low or moderate heterogeneity. The low coverage values highlight the need for models that appropriately account for or explore the between trial variation.

LIMITATIONS

Until larger simulations can be undertaken, conclusions based on the simulated and empirical data should only be applied to small meta-analyses of four or five trials.

CONCLUSIONS

These investigations suggest that under normal conditions all three methods provide similar results. For moderate heterogeneity coverage for all the three fixed effects models depreciates.

Second or third additional chemotherapy drug for non-small cell lung cancer in patients with advanced disease

BACKGROUND

Randomized trials have demonstrated that adding a drug to a single-agent or to a two-agent regimen increased the tumor response rate in patients with advanced non-small cell lung cancer (NSCLC), although its impact on survival remains controversial.

OBJECTIVES

To evaluate the clinical benefit of adding a drug to a single-agent or two-agent chemotherapy regimen in terms of tumor response rate, survival, and toxicity in patients with advanced NSCLC.

SEARCH STRATEGY

There were no language restrictions. Searches of MEDLINE and EMBASE were performed using the search terms non-small cell lung carcinoma/drug therapy, adenocarcinoma, large-cell carcinoma, squamous-cell carcinoma, lung, neoplasms, clinical trial phase III, and randomized trial. Manual searches were also performed to find conference proceedings published between January 1982 and June 2006.

SELECTION CRITERIA

Data from all randomized controlled trials performed between 1980 and 2006 (published between January 1980 and June 2006) comparing a doublet regimen with a single-agent regimen or comparing a triplet regimen with a doublet regimen in patients with advanced NSCLC.

DATA COLLECTION AND ANALYSIS

Two independent investigators reviewed the publications and extracted the data. Pooled odds ratios (ORs) for the objective tumor response rate, one-year survival rate, and toxicity rate were calculated using the fixed-effect model. Pooled median ratios (MRs) for median survival also were calculated using the fixed-effect model. ORs and MRs lower than unity (< 1.0) indicate a benefit of a doublet regimen compared with a single-agent regimen (or a triplet regimen compared with a doublet regimen).

MAIN RESULTS

Sixty-five trials (13601 patients) were eligible. In the trials comparing a doublet regimen with a single-agent regimen, a significant increase was observed in tumor response (OR 0.42, 95% confidence interval [CI] 0.37 to 0.47, P < 0.001) and one-year survival (OR 0.80, 95% CI 0.70 to 0.91, P < 0.001) in favor of the doublet regimen. The median survival ratio was 0.83 (95% CI 0.79 to 0.89, P < 0.001). An increase also was observed in the tumor response rate (OR 0.66, 95% CI 0.58 to 0.75, P < 0.001) in favor of the triplet regimen, but not for one-year survival (OR 1.01, 95% CI 0.85 to 1.21, P = 0.88). The median survival ratio was 1.00 (95% CI 0.94 to 1.06, P = 0.97).

AUTHORS' CONCLUSIONS

Adding a second drug improved tumor response and survival rate. Adding a third drug had a weaker effect on tumor response and no effect on survival.

What is epilepsy? Clinical perspectives in the diagnosis and treatment

SUMMARY

The CINNR International Conference, "An Overview of Epilepsy Research: What, Where, When, and Why?," was designed to introduce epilepsy to the nonclinician interested in epilepsy research. This article discusses the clinical aspects of epilepsy, defines clinical terms associated with epilepsy and seizure disorders, and outlines the scope of the clinical problem and the issues that need clarification from a clinical perspective. Most importantly, it is hoped that this article will put a human face on this common disease.

Historical data in the design and interpretation of trials with newly diagnosed patients

An International League Against Epilepsy (ILAE) subcommission is exploring the possibility of utilizing historical data from treated and control (untreated or under-treated) patients who have been enrolled in monotherapy trials in newly diagnosed epilepsy. Active-control trials have not convincingly distinguished between equivalent effectiveness and equivalent ineffectiveness. Thus, there is insufficient historical evidence of sensitivity to drug effects. Noninferiority trials are based upon experience acquired with previous trials. Optimizing the exploitation of historical data obtained in newly diagnosed patients may be a way to improve the validity of future trials. A concern about use of historical controls is lack of assurance that historical populations and newly recruited populations are similar. The best approach to pooling and modeling data may be an Individual Patient Data (IPD) approach as (1) analysis of data from active-control trials, (2) demonstration that the subpopulation carrying an intermediate probability of recurrence is the most sensitive to drug effect, by comparison to similar subpopulations who received a placebo or a pseudoplacebo in historical studies. This subpopulation could become a reference for patient selection or stratification in future trials to improve assay sensitivity.

Lamotrigine versus carbamazepine monotherapy for epilepsy

BACKGROUND

The choice of an antiepileptic drug (AED) for any individual should take into account reliable information about seizure control, adverse effects and cost. Carbamazepine is the usual drug of choice for people with newly-diagnosed partial onset seizures. Lamotrigine is a relatively new AED which is licensed in many countries for use as an initial monotherapy.

OBJECTIVES

To review the best evidence comparing carbamazepine and lamotrigine when used as monotherapy in people with partial onset seizures, or generalized onset tonic-clonic seizures with or without other generalized seizure types.

SEARCH STRATEGY

We searched the Cochrane Epilepsy Group's Specialized Register (July 2005), the Cochrane Central Register of Controlled Trials (CENTRAL) (TheCochraneLibrary Issue 2, 2005), and MEDLINE (1966 to August 2005). No language restrictions were imposed. We also contacted pharmaceutical companies and trial investigators.

SELECTION CRITERIA

Randomized controlled trials, blinded or unblinded, in which children or adults with partial onset seizures or generalized onset tonic-clonic seizures were randomized to monotherapy with either carbamazepine or lamotrigine.

DATA COLLECTION AND ANALYSIS

This was an individual patient data review. Outcomes were (1) time to treatment withdrawal, (2) time to first seizure post randomization, and (3) seizure freedom at six months. Time to event data were analysed using a stratified logrank analysis with results expressed as hazard ratios (HR) and 95% confidence intervals (95% CI); binary data were expressed as relative risks (RR) and 95% confidence intervals (95% CI). A HR or a RR greater than 1 indicated an event was more likely on lamotrigine than carbamazepine.

MAIN RESULTS

Individual patient data were available for 1384 participants (100% of total randomized) from the five trials that met our inclusion criteria. The main results (HR (95% CI)) were (1) time to treatment withdrawal 0.55 (0.35 to 0.84) (random-effects), (2) time to first seizure post randomization 1.14 (95% CI 0.92 to 1.43), and (3) seizure freedom at six months RR 0.92 (95% CI 0.81 to 1.04). The review suggested that time to treatment withdrawal was significantly improved with lamotrigine compared to carbamazepine, while time to first seizure and seizure freedom at six months favoured carbamazepine although the results were not statistically significant.

AUTHORS' CONCLUSIONS

Lamotrigine was significantly less likely to be withdrawn than carbamazepine but results for time to first seizure suggested that carbamazepine may be superior in terms of seizure control. Trials were of too short a duration to measure important seizure outcomes such as time to 12 month remission. Further trials are needed in which longer-term outcome is assessed as well as measures such as psychosocial outcome and quality of life.

An overview of methods and empirical comparison of aggregate data and individual patient data results for investigating heterogeneity in meta-analysis of time-to-event outcomes

Combining the results of individual studies using meta-analysis may be undertaken using either aggregate data (AD) or individual patient data (IPD). In any meta-analysis it is important to consider statistical heterogeneity between studies. Potential sources of heterogeneity can be explored using regression models with either AD or IPD. An overview of approaches and empirical assessment of how the results and conclusions differ from these analyses is undertaken using a meta-analysis of five randomized controlled trials comparing two antiepileptic drugs with time-to-event outcomes. Alternative meta-regression models using AD are compared to stratified Cox regression models using IPD. Age as a potential cause of heterogeneity is detected by both AD and IPD regression models. Time from first ever seizure to randomization is only identified by some AD models. A more thorough explanation of heterogeneity is obtained from the model using IPD but further empirical evidence comparing IPD and AD results are needed.

Treatment of epilepsy in adults: expert opinion, 2005

RATIONALE

Over the past decade, there has been a proliferation of new therapies for the treatment of epilepsy. Faced with this growing list of options, clinicians must decide what therapy, or combination of therapies, is best for a given individual. Although controlled clinical trials exist for each treatment option, the answer to these questions may remain unclear. In 2000, a survey of expert opinion was done to address questions concerning which treatment options might be best in a number of clinical situations. We surveyed a group of US epileptologists again in 2004 and compared the results of the two surveys.

METHODS

We sent a questionnaire on the treatment of adolescent and adult epilepsy syndromes to a group of opinion leaders in the field of epilepsy. The questions were formatted to simulate real-world clinical situations in the treatment of symptomatic localization related epilepsy (SLRE) and idiopathic generalized epilepsy (IGE). The experts were asked to rate treatment options based on a modified RAND 9-point scale (with "9" most appropriate and "1" least appropriate). Statistical analysis of data was performed as defined by the expert consensus method. The results were used to develop user-friendly recommendations concerning overall treatment strategies and choice of specific medications.

RESULTS

Of the 48 experts to whom the survey was sent, 43 (90%) responded; 29 (67%) of the respondents had also participated in the first survey. For initial monotherapy for IGE (generalized tonic-clonic [GTC], absence, and myoclonic seizures), valproate was rated as treatment of choice. For IGE-GTC seizures, lamotrigine and topiramate were also identified as usually appropriate for initial monotherapy. For IGE-absence seizures, ethosuximide was also a treatment of choice, and lamotrigine was usually appropriate. For SLRE, the experts were again asked to rate treatment options based on seizure type: simple partial seizures (SPS), complex partial seizures (CPS), and secondarily generalized tonic-clonic seizures (SGTC). In SLRE-SPS and SLRE-SGTC, carbamazepine and oxcarbazepine were treatments of choice, with lamotrigine and levetiracetam also usually appropriate. In SLRE-CPS, carbamazepine, lamotrigine. and oxcarbazepine were treatments of choice, while levetiracetam was also usually appropriate. For women who are pregnant or trying to conceive, lamotrigine was treatment of choice for both syndrome types. In the elderly, whether medically stable or ill, the treatment of choice was lamotrigine, while levetiracetam was also usually appropriate (along with gabapentin for persons with comorbid medical illness). In persons with HIV and epilepsy, lamotrigine and levetiracetam were usually appropriate. In people with both epilepsy syndromes who have depression, lamotrigine was treatment of choice. In a person with seizures and renal disease, lamotrigine was usually appropriate for both syndromes, with valproate also usually appropriate for IGE. In patients with hepatic disease, levetiracetam and lamotrigine were usually appropriate for IGE; in SLRE, levetiracetam was treatment of choice, with gabapentin also usually appropriate.

CONCLUSIONS

Although the panel of experts reached consensus on many treatment options, there are limitations to these types of data. Despite this, the expert consensus method concisely summarizes expert opinion, and this opinion may be helpful in situations in which the medical literature is scant or lacking. The information in this report should be evaluated in conjunction with evidence-based findings.

Tratamento de epilepsia: consenso dos especialistas brasileiros

Epilepsia é uma condição muito freqüente em todo o mundo. Na última década, várias opções terapêuticas surgiram ou foram aprimoradas. O principal método utilizado para decisão terapêutica baseia-se nos estudos randomizados, que representam o maior nível de evidência. Entretanto, mesmo estes estudos são passíveis de críticas e em alguns casos o tratamento de escolha permanece controverso. Nestas situações, a opinião dos especialistas, na área da epileptologia, com maior experiência clínica, passa a ter grande valor. O presente estudo tem como principal objetivo elaborar um consenso de tratamento das epilepsias, através da opinião de experts brasileiros no assunto. Este consenso poderá auxiliar na criação de manuais e estratégias para o tratamento de determinadas síndromes epilépticas, de acordo com os padrões socioeconômicos brasileiros.

Aggregate data meta-analysis with time-to-event outcomes

In a meta-analysis of randomized controlled trials with time-to-event outcomes, an aggregate data approach may be required for some or all included studies. Variation in the reporting of survival analyses in journals suggests that no single method for extracting the log(hazard ratio) estimate will suffice. Methods are described which improve upon a previously proposed method for estimating the log(HR) from survival curves. These methods extend to life-tables. In the situation where the treatment effect varies over time and the trials in the meta-analysis have different lengths of follow-up, heterogeneity may be evident. In order to assess whether the hazard ratio changes with time, several tests are proposed and compared. A cohort study comparing life expectancy of males and females with cerebral palsy and a systematic review of five trials comparing two anti-epileptic drugs, carbamazepine and sodium valproate, are used for illustration.

Individual patient-versus literature-based meta-analysis of survival data: time to event and event rate at a particular time can make a difference, an example based on head and neck cancer

The objective of this study is to compare the results of an individual patient-based and a literature-based meta-analysis in chemotherapy in head and neck cancer and to identify the sources of difference. For all head and neck cancer randomized controlled clinical trials comparing chemotherapy and loco-regional treatment with loco-regional treatment alone, both the literature data and the individual patient data are retrieved and meta-analyses performed and compared. Only survival data are used as outcome, although both time to death and mortality at specific time points are considered in different analyses. There are substantial differences between the individual patient-based and the literature-based meta-analyses. The most important reason for the differing results is that the individual patient-based meta-analysis is based on a time to event analysis, whereas the literature-based meta-analysis is based on mortality at a specific time point. Mortality can change substantially with follow-up time. The absolute survival differences in the case study, for instance, increase from 2.6% at 2 years to 5.6% at 5 years.

Meta-analyses of randomised clinical trials in oncology

A meta-analysis is a quantitative synthesis of randomised clinical trials, used to evaluate moderate treatment effects in oncology. It is complementary to large-scale trials. We describes the principles, methods, and limits of meta-analyses. The gold standard for a meta-analysis is to obtain individual patient data directly from each principal investigator, but this is time-consuming and costly. The main steps of a meta-analysis using individual patient data are described. Multidisciplinary collaboration is needed for clinical insight and critical review of the data and results. Meta-analysis should include an evaluation of the trial quality, a quantification of the overall treatment effect, a study of the variations seen in this effect between trials, and pre-planned exploratory analyses to identify groups of patients who may benefit more from the treatment. Statistical methods are explained using real working examples. Since literature-based meta-analysis can lead to seriously biased assessments, meta-analyses of individual patient data should be undertaken systematically when long-term follow-up is needed, when a detailed analysis is important, or when the literature-based meta-analyses are not in agreement. The main factors which influence the quality of a meta-analysis are discussed.