Trial centers
The trial centers are St. Olavs University hospital, Trondheim, Norway; Oslo University Hospital, Oslo, Norway; Ålesund Hospital, Ålesund, Norway; Fondazione IRCCS Istituto Naxionale dei Tumori, Milan, Italy; Istituto Scientifico Romagnolo per lo studio e la cura dei tumori (IRST), Meldola (FC), Italy; Hospital Universitari Arnau de Vilanova, Lleida, Spain; Castle Hill Hospital, Cottingham, United Kingdom.
Outcome and statistical analyses
For all planned analyses baseline data will be presented with descriptive statistics; continuous variables as mean with standard deviation and categorical variables as frequency with percentages. A detailed overview of research questions and statistical analyses are described in the following section. All analyses will be performed using SPSS v 23 (IBM Corp. Armonk, NY) and STATA v 16 (Stata Corporation LP; College Station, TX, USA).
Research question 1: Which are the clinical predictors and biomarkers predictors of pain response to palliative RT for CIBP?
Objective: To obtain demographic, clinical data and biomarkers before start of RT and compare with RT response for pain in order to develop a classification system relevant for predicting RT response.
Outcome: Primary endpoint is response to RT defined as at pain reduction of worst pain score of two or more at the treated site on the 11-point NRS together with no increase in analgesic intake, or a reduction in opioid intake of at least 25% from baseline without an increase in worst pain score at the treated site [
18]. Patients with two or more radiation locations are defined as responders if they respond in one of the included sites. Patients who die before the first assessment (3 weeks after RT), will be defined as non-responders because these patients have not benefited from RT. Patients with missing data on outcome measurements including pain intensity at the treated site or the use of analgesic medications will not be included in the analyses.
Statistical method: Clinical variables (see previous chapter) and CRP will be included in a multivariate logistic regression model to predict potential factors for response to RT. The model will be adjusted by study centre. Regression diagnostics will be performed for all analyses adding interactions terms if necessary. Significant variables from the multivariable model will be presented as a response-score to determine the likelihood of response to RT. Second, we will perform analyses on inflammatory biomarkers and bone biomarkers correlated with response to RT. Only inflammatory biomarkers and bone biomarkers with levels of serum concentrations above the detection threshold and which have a variability between measurements will be analysed with a multivariate logistic regression model with response to RT as the dependant variable. Finally, an integrated regression model with response to RT including both the significant clinical variables and significant biological biomarkers will be conducted if appropriate based on results from the previous analyses.
Research question 2: Which inflammatory biomarkers are associated with change in pain intensity in patients with CIBP?
Objective: To investigate the correlation between pain intensity and inflammation. Patients included in the PRAIS study all have CIBP and will receive a standardized intervention expected to give pain relief in about 60% of the patients. This study can explore if there is a longitudinal relationship between changes in pain intensity and the detected level of inflammatory biomarkers.
Outcome: Serum concentrations of inflammatory biomarkers as described above associated with changes in pain intensity as measured by average pain intensity and worst pain intensity last 24 h (NRS 0–11) for a maximum of 1 year follow-up.
Statistical method: Longitudinal data analysis with repeated measurements using a liner mixed effect model or generalized estimating equation (GEE).
Research question 3: Which inflammatory substances are associated with cancer cachexia?
Objective: To gain insight into the role of inflammation in cancer cachexia.
Outcome: Cachexia is defined as a) weight loss > 5% over past 6 months (in absence of simple starvation); or b) body mass index (BMI) < 20 and any degree of weight loss > 2%; or c) appendicular skeletal muscle index consistent with sarcopenia (males < 7・26 kg/m2; females < 5・45 kg/m2) and any degree of weight loss > 2% [
38]. Cachexia severity is assessed as degree of weight loss and EORTC QLQ C15 PAL physical function and appetite loss.
Statistical method: Fist, we will perform a cross-sectional analysis in the baseline parameters using a logistic regression with cachexia as outcome, and the various inflammatory markers as explanatory variables. Second, we will investigate the association between changes in inflammatory markers over time and cachexia severity using mixed linear modelling.
Research question 4: Which inflammatory substances are associated with depression in cancer patients?
Objective: To explore the associations between serum concentrations of inflammatory substances and depression.
Outcome: Depression is measured by the PHQ-9 questionnaire [
57]. The PHQ-9 includes 9 items identical to the diagnostic criteria (i.e. the depression symptoms) for depression disorder. For the purpose of these analyses we will use depressive disorder as defined by the PHQ-9 recommended sum score calculation, and symptom by symptom in the PHQ-9 score in associations of serum concentrations of inflammatory biomarkers during a period of 1 year form inclusion.
Statistical method: We will use repeated measurements and a linear mixed effect model to perform the analyses.
Research question 5: Which are the clinical predictors and biomarker predictors of development of cachexia in patients with metastatic cancer disease?
Objective: To obtain demographic and clinical data and biomarkers in a homogenous population of patients with metastatic cancer, and within a prospective, longitudinal follow-up study observe which factors predicting the development of severe cachexia during one-year follow-up.
Outcome: Cachexia is defined as for research question 3.
Statistical method: As the purpose is to evaluate development of cachexia, only patients without cachexia at baseline will be included in this analysis. A Cox regression will be used to evaluate baseline predictors of cachexia development.
Research question 6: What is the relationship between pain reduction and depression in patients receiving RT for CIBP?
Objective: RT is expected to reduce pain in a substantial number of patients. This creates a possibility to study the relationship between depression and pain longitudinally in an experimental-like design in which one variable (pain) is manipulated and the effect on another related variable (depression) is studied.
Outcome: Depression is measured as for Research question 4.
Statistical method: Longitudinal analyses with repeated measurements analyzed with a linear mixed effect model.
Ethics
A signed consent will be obtained from all participants by an investigator at each site. The study will be carried out in accordance with ICH GCP and the World Medical Association Declaration of Helsinki (1964) and its’ revisions (Tokyo 1978, Venice 1983, Hong Kong 1989, South Africa 1996 and Edinburgh 2000). The study is approved by The Regional Committee for Medical and Health Research Ethics, REC Central Norway, and by the regulatory authorities at each trial site. If modifications to the protocol, amendments will be applied for to the regional ethics committee and after approval distributed to all local study investigators.
RT is indicated and will be given regardless of whether the patients choose to participate in the study or not. Thus, study participation means that patients consent to report symptoms and have their blood drawn. The volumes of blood samples are limited (less than 50 ml) and give no extra risk for anemia. Thus, the study includes no interventions that increase the risk for the patients. Data are handled anonymously. The database used for analyses only identify each patient by a study number. The linkage between study number and patient identity is in a document and / or memory stick stored in a safe at each study center.
Organizational issues
A Trial Steering Group will oversee the running of the trial. Members of the trial steering group include the chief principal Investigator, the principal Investigators of each centre, the Clinical Trial coordinator, and the trial statistician. After completion of the inclusions the Trial Steering Group will consists of the chief principal investigator, the principal investigators at all sites including 100 patients or more, and the trial statistician.
The coordinating centre will administer the study. This includes development and administration of Case report forms (CRFs), monitoring of data quality and preparation of the final study report. (CRFs) will be supplied by the coordinating centre. Specific queries about data will be addressed to the clinical trial coordinator at each study centre.
The results will be published in peer-reviewed journals. Authorship is based upon the Vancouver rules. All manuscript will be prepared by the researchers. The trial steering group will make decision related to use of data for publication. The principal investigator has access to all data. Access to researchers will be decided based upon the actual need for access relevant for analyses. Access to anonymously participant-level data set will be supplied depending on the journal policies.