Assessment of the primary and secondary endpoints
The primary endpoint is neurocognition after cerebral irradiation in SCLC patients treated with WBRT or SRS, defined as a drop of at least 5 points from baseline in HVLT-R total recall at 3 months. Secondary objectives are intracranial progression (local tumor progression, number of new cerebral metastases), extracranial progression, overall survival, death due to brain metastases, local progression-free survival, progression-free survival, changes in other cognitive performance measures, quality of life, and toxicity.
Time to progression is defined as the number of days from randomization to the first occurrence of the respective event. Overall survival time is defined as number of days from randomization until death or end of follow-up. For patients alive at the end of the study, the overall survival time will be censored at the time of the last visit or follow-up contact.
Time to death due to BM is defined as number of days from randomization until death due to BM or end of follow-up, where death due to BM is defined as death with intracranial progression as a component of cause of death. Locally progression-free survival time is defined as number of days from randomization until local tumor progression, death without prior local progression, or end of follow-up. The occurring events of interest will be classified as (1) progression of cerebral metastases present at baseline only, (2) occurrence of new cerebral metastases only, (3) simultaneous detection of progression of cerebral metastases present at baseline and of new metastases, and (4) death without local progression.
Time to extracranial progression is defined as number of days from randomization until extracranial progression or end of follow-up, where extracranial progression is the first date on which progressive disease outside the brain occurred according to the physician treating the primary disease. Progression-free survival time is defined as number of days from randomization until the first occurrence of intracranial or extracranial progression, death without prior progression, or end of follow-up. Intracranial progression is defined as occurrence of progressive disease concerning the pre-existing BM or the occurrence of new BM.
Local tumor progression (progressive disease, PD) is defined as occurrence of intracranial progression) in the area of the SRS.
Progression in the WBRT area is defined as occurrence of intracranial tumor progression (progression of existing lesions and/or occurrence of new lesions).
The European Organization for Research and Treatment of Cancer (EORTC) Questionnaire including brain module (BN20) and the EORTC Paired Associated Learning Questionnaire will be used. This study will use the International Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 for toxicity and adverse event reporting.
Patients are followed within the trial protocol for 12 months after baseline visit (T0). After T0, patients are scheduled for follow-up visits every 3 months or as needed clinically, including contrast-enhanced MRI as well as thorough clinical-neurological assessment. Formal neurocognitive testing and quality of life instrument testing will be performed at baseline and during follow-up visits every 3 months. The computer-administered Cambridge Neuropsychological Test Automated Battery (CANTAB) will be used to examine specific components of cognition. The participants will be instructed to respond to stimuli presented on a computer screen by pressing a touch screen. Three tests were chosen to evaluate further neurocognitive changes: Paired Associated Learning for paired Associates Learning assesses visual memory and new learning [
21]; Reaction Time to provide assessments of motor and mental response speeds [
22], as well as measures of movement time, reaction time, response accuracy and impulsivity. Further, Spatial Working Memory will be used [
22], which requires retention and manipulation of visuospatial information. Staff members administering the tests were trained and approved by the clinical neuropsychologist in the project group.
For the last patient in, the final study visit will be 12 months after baseline. This is considered the final study visit (last patient out). All other patients will be followed regularly as described in detail until death or until 12 months after baseline. Beyond that, all patients can be followed within routine clinical visits according to national guidelines and survival and progression data will be documented until ‘last patient out’. If death occurs at less than 12 months or patients leave the study prior to 12 months, they will be still included into the intention-to-treat population.
Statistical analysis
The primary hypothesis of the trial is that there is a difference between the two treatment arms with respect to the primary endpoint, defined as a drop of at least 5 points from baseline in HVLT-R total recall at 3 months after baseline (T0). Chang et al. [
25] observed deterioration probabilities of 0.64 for SRS + WBRT and 0.20 for SRS alone at 4 months after baseline. Based on those results, assuming a deterioration probability of 0.20 for the SRS arm and 0.64 for the WBRT arm in our trial,
n = 19 patients per arm are required to demonstrate a difference between treatment arms applying a χ
2 test at a two-sided significance level of α = 0.05 with a probability of 1–β = 0.8. Assuming exponentially distributed survival times with a median of 6 months for both groups, 29.3% of all randomized patients are expected to have died before the measurement of the primary endpoint. Thus,
n = 28 patients per group are required to yield a sufficiently high power for a comparison of the deterioration rate within the two groups. Statistical analysis is based on the International Conference on Harmonization Guidelines Structure and Content of Clinical Study Reports and Statistical Principles for Clinical Trials. A detailed methodology for the statistical analysis will be described in the statistical analysis plan, which will be finalized before database lock. Statistical analysis will be performed using SAS v9.4 or higher.
Efficacy evaluation
The primary hypothesis of the trial is that there is a difference between the two treatment arms with respect to the primary endpoint, defined as a drop of at least 5 points from baseline in HVLT-R total recall at 3 months. With π
SRS being the deterioration probability in the SRS arm and π
WBRT being the deterioration probability in the WBRT arm, the null hypothesis H
0: π
SRS = π
WBRT is tested against its alternative H
1: π
SRS ≠ π
WBRT at a two-sided significance level of α = 0.05 using a Cochran–Mantel–Haenszel test adjusting for the confounder time of appearance (synchronous vs. metachronous). Missing data for the primary outcome variable will be replaced by using multiple imputation which takes the covariates of treatment group, time of appearance (synchronous vs. metachronous), and the baseline HVLT-R total recall score into account by application of the fully conditional specification method [
26]. For the secondary time-to-event endpoints overall survival, local progression-free survival, and locoregional progression-free survival, median event times and 1-year rates will be given with 95% confidence intervals and Kaplan–Meier curves will be calculated for both treatment groups. A (descriptive) log-rank test stratified for time of appearance (synchronous vs. metachronous) will be performed in order to assess differences between the two treatment groups, and a descriptive
P value will be given. A Cox proportional hazard model with overall survival as dependent variables well as treatment group and time of appearance as independent factors will be fitted to estimate the hazard ratio for the treatment group together with a 95% confidence interval.
All further secondary outcomes will be analyzed descriptively, and descriptive P values will be reported together with corresponding 95% confidence intervals.
The study protocol, Patient Information sheet, and Declaration of Informed Consent was approved by the Heidelberg University Ethics Committee (S-470/2017). The procedures described in the submitted study protocol regarding the performance, evaluation, and documentation of this study has been selected in such a way that the principles of the Good Clinical Practice (GCP) guidelines are observed. The regulations regarding medical confidentiality and data protection are fulfilled. Informed consent will be obtained from all participants in the study.
Concerning radiation protection law (StrSchV), the authors of this protocol presume that a submission to the Bundesamt für Strahlenschutz (BfS) is not required. To confirm this position, the investigators submitted this protocol to the expert commission of the German Society of Radiation Oncology ENCEPHALON Clinical Trial Protocol Version 1.0, Date 07/2017 (DEGRO No. 132) (Additional file
1; SPIRIT Checklist).