Comprehensive Aneurysm Management Trial

The best management of patients with unruptured intracranial aneurysms is currently uncertain.

The time-honored approach to treating UIAs is with microsurgical clipping, Surgical clipping is widely considered to be a safe, durable means to treat UIAs. In recent years, endovascular treatment (EVT) has increasingly replaced surgical clipping. The efficacy of EVT in the prevention of SAH remains unknown. EVT is reputed to be potentially less effective than surgery because of long term angiographic recurrences. Neither surgery nor EVT have been shown superior to conservative management. Yet, both endovascular and surgical treatment are used every day in a large number of patients, without evidence they are doing more good than harm, and without evidence that one is better than the other.

There is currently no scientific evidence to support treatment of UIAs and there are no well-accepted guidelines. The overall annual rupture risk for conservatively managed lesions remains ~1-2% over a follow-up of 6-9 years, likely an underestimate. There is currently no all-inclusive care trial framework for UIA patients. There are also no randomized trials currently comparing active and conservative management for UIAs.

The uncertainty regarding the management of UIAs has not progressed in the last 30 years. Avoiding iatrogenia and the consequences of over-diagnosis is an increasingly important public health goal. In the spirit of care trials, this CAM trial offers a comprehensive framework, so that all patients confronted with the clinical dilemma can be offered participation.

Main objectives

1. To provide a care research framework to manage all patients with UIAs.
2. For patients for whom conservative management is contemplated, to validate previous observational studies claiming rupture risks are very low. This registry could be used in the future to recruit patients in trials assessing the potential value of imaging follow-up studies or trials assessing pharmacological prevention of ruptures (such as ASA or statins).
3. For each patient for whom curative treatment is contemplated, to provide a 50% chance of being spared a potentially useless or harmful treatment.
4. For each patient for whom surgical clipping is contemplated, to provide a 50% chance of being offered a less invasive alternative.

Principal questions to be addressed:

1. Can a care trial context assist clinicians and patients manage the UIA dilemma?
2. When conservative management is contemplated, is the risk of rupture as low as unvalidated observational studies currently suggest?
3. Do patients with UIAs, considered for curative treatments, have a better long-term clinical outcome with active treatment or conservative management? More specifically: 3.1 When active endovascular treatment is contemplated, do patients with UIAs have a better long-term clinical outcome with endovascular or conservative management? 3.2 When active surgical treatment is contemplated, do patients with UIAs have a better long-term clinical outcome with surgical or conservative management? 3.3 When active treatment is contemplated, and surgical and endovascular management are both judged reasonable, do patients with UIAs have a better long-term clinical outcome with (surgical/endovascular) or conservative management?
4. When patients are considered ineligible for conservative management, and surgical and endovascular management are both judged reasonable, do patients with UIAs have a better long-term clinical outcome with surgical or endovascular management?

Study design This study is designed as a pragmatic, comprehensive way to address the unruptured aneurysm clinical dilemma, combining large simple clinical care trials whenever patients are judged eligible for more than one management option, or otherwise a registry of each option. All patients with one or more UIAs will be eligible for participation in either a registry or one of the trials. A non-invasive (MRA or CTA) or catheter angiogram and a baseline CT-scan or MRI of the brain are required to enter the study. These studies should demonstrate the unequivocal presence of a saccular aneurysm, not recently ruptured (> 4 weeks since the last SAH). If they accept, all subjects will be registered, and when eligible, included in a specific randomized trial using the algorithm below which includes a combination of clinical judgment and randomized allocation. Conventional randomization will be performed whenever and wherever possible. A double consent pre-randomization design is possible wherever the method is accepted by the local IRB.

Planned algorithm for managing patients The algorithmic process combining clinical judgment and randomized allocation will proceed by asking the following questions to the treating physicians.

Q1: Is the patient being considered for treatment? Possible answers: YES or NO.

If the clinician selects NO, the patient will be included in the observational registry. If YES, the following question will be:

Q2: What treatment option is being considered for this patient? Clinicians will use clinical judgment to select one of three options: endovascular, surgical, or they will select that either treatment is reasonable.

Then comes the most important question:

Q3: Is conservative management a reasonable alternative given the lack of evidence that treatments are beneficial? Potential answers are YES or NO. If NO, they will be included in the CURES trial or in of the 2 treatment registries (surgical or endovascular). If YES, they will be included in the appropriate randomized trial (according to the answer to question 2) that compares endovascular to conservative management (TEAM, or trial on endovascular aneurysm management), surgical to conservative management (TSAM, or Trial on Surgical Aneurysm Management), or they will be included in the TOES arm (Trial on Observational, Endovascular or Surgical management), with potential for a second randomization, comparing surgical to endovascular treatment.

Hypotheses:

The primary hypothesis for patients allocated to at least 2 options, one of which is conservative management is: The 10 year combined neurological morbidity and mortality (mRS >2) will be reduced from 24% to 16% (beta 80%; alpha 0.048; sample size 961 patients (836 plus 15% losses to FU and cross-overs) with active treatment.

Outcomes The primary outcome is survival without neurological dependency (mRS<3) at 10 years.

Secondary outcome measures are needed to monitor patient outcomes during the duration of the trial.

Secondary outcomes will be: 1/ The incidence of SAH during follow-up and related morbidity and mortality; 2/ The morbidity and mortality related to EVT or surgical treatment of the UIA at one year; 3/ Overall mortality at 1, 5 and 10 years; 4/ Overall morbidity (mRS>2) at 1, 5 and 10 years; 5/ Length of hospitalization; 6/ Discharge to location other than home.

This pragmatic care trial has been designed to potentially include all patients with UIAs. Inclusion criteria are as broad as possible and exclusions as few as possible.

Frequency and duration of follow-up All the investigations proposed in the UIA care trial constitute part of routine, standard care in the treatment of unruptured intracranial aneurysms.

Patients who undergo treatment will be seen in clinic at approximately 6-12 weeks after treatment as part of routine follow-up care. Patients will be followed with another routine clinic visit at one year, and at 5 and 10 years thereafter. All treated patients will have non-invasive imaging (CTA or MRA) at 12 ± 2 months post-treatment to determine the presence of a major, saccular aneurysm recurrence.

Number of patients To assess whether the 10 year combined neurological morbidity and mortality (mRS >2) will be reduced from 24% to 16% with active treatment (beta 80%; alpha 0.048) will require at least 961 patients (836 plus 15% losses to FU and cross-overs).

Planned Statistical Analyses The 4 stratified trial arms (TSAM, TEAM II, TOES, and CURES) and the registries will be analyzed separately. Combined exploratory analyses will also be performed, looking for differences in group characteristics and in treatment outcomes between RCTs and registries, attempting to better understand the scope and limits of trial results. As with any trials of surgery, early risks may be followed by later benefit. Therefore the hazard ratio will be unfavorable during the recruitment years, while interventions are being performed. Initially, the DSMC will assure that treatment-related complications and hemorrhagic events are within confidence intervals compatible with a safe and meaningful trial. In order to describe how and when hemorrhagic events occur, analyses will include Kaplan-Meyer life-table methods to assess the 1, 5-, and 10-year survival, and survival without hemorrhage, among all those allocated immediate treatment (including the few who did not undergo it) and all those allocated deferral of any intervention (including the few who will eventually be treated). The 'survival' functions will be compared graphically and using a log-rank statistic. The main statistical tests will involve comparisons between the probabilities of mortality and morbidity-mortality 1/ from hemorrhage, excluding peri-operative complications, 2/ from hemorrhage or from complications of treatment, or 3/ comparisons of the 1, 5 and 10-year probabilities of combined disease/treatment-related mortality/morbidity, in the absence of other causes of death or disability. Descriptive statistics will be done on demographic variables and potential risk factors to compare the groups at baseline. Means, standard deviations and range will be presented for quantitative variables such as size of aneurysms and frequency tables for categorical variables (such as number of patients with a previous history of SAH, or multiple aneurysms). Statistics will be broken down by center and by treatment arm. Comparability of the groups will be assessed through independent ANOVAs (quantitative data) or Mantel-Haentzel and X2 tests (categorical data). Assuming comparability of groups across centers, the primary outcome, disease or treatment-related combined morbidity/mortality (for both intent-to-treat and per-protocol populations) will be compared between groups through a Fisher's Exact Test for independent proportions at 1, 5, and 10 years. Similar analyses will be done for disease or treatment-related mortality combined mortality-morbidity. Secondary outcomes and overall morbidity will be compared between groups through independent t-tests (quantitative variables) or X2 statistics (categorical data). The analyses of neurological data at follow-up will control for baseline data using logistic regression, ANOVA or Cox regression multivariate models. All tests will be interpreted with adjustment for multiplicity to have the 0.05 level of confidence at 10 years only. Finally, a logistic regression will be used to find baseline variables capable of predicting hemorrhages or complications in both groups. The method planned is a stepwise forward on a likelihood ratio with a probability of 0.05 to enter a predictor (PIN in) and a probability of 0.10 to exclude a predictor (PIN out). Possible predictors include the size of the aneurysm, location, patient age, status of the aneurysm and previous history of SAH versus unruptured only) as well as other baseline characteristics. All analyses of the primary endpoint will be performed taking into account the multiplicity of tests and any interim analyses. Secondary analyses will be performed using an alpha of 5%.