Study Investigating How Physicians Assess the Risk of Patients Developing Febrile Neutropenia During Chemotherapy.

Completed

Conditions: Chemotherapy-induced Febrile Neutropenia

Registration Number: NCT01813721

Lead Sponsor: Amgen

Brief Summary: This is a prospective observational study investigating how physicians assess the risk of febrile neutropenia (FN) developing in patients who will receive chemotherapy.

Approximately 150-200 investigators will take part in about 100 sites in Europe, Canada and Australia. Approximately 1000 subjects will be studied, all of whom will have non-small cell lung cancer (NSCLC), small cell lung cancer (SCLC), non-Hodgkin's lymphoma (NHL) or breast cancer and will be due to receive one of the specific chemotherapy regimens of interest.

Investigators' approach to FN risk assessment will be studied using lists of possible risk factors they may consider during their assessment. Investigators will be asked to select and rank the factors they consider the most important when assessing the overall FN risk of a subject and when making the decision whether to treat with granulocyte-colony stimulating factor (G-CSF) primary prophylaxis (PP). They will be asked to make these selections based initially on their own routine clinical practise and subsequently relating specifically to each subject recruited.

This is a non-interventional study that involves no procedures outside normal care for the subjects; all data collection will be completed prior to chemotherapy administration.

Detailed Description: Although a formal hypothesis will not be tested in this observational study, it is hypothesized that the clinical risk factors ranked as the most important when conducting FN risk assessments by investigators are aligned with international guidelines and published data. Also, that the investigator's decision to treat with G-CSF PP is influenced by clinical and non-clinical risk factors (such as distance from site, estimated subject compliance, and access to fully reimbursed G-CSF).

Study Design: Prior to identifying eligible subjects, Investigators will be registered and will record baseline information. During this Baseline Investigator Assessment investigators will be provided with two lists of risk factors. Investigators must rank selected risk factors that they consider to be the most important when assessing 1) overall FN risk (only scientific factors will be included), and 2) when deciding on whether G-CSF PP treatment will be used or not (this list will also contain non clinical factors). They will also record their own FN risk intervention threshold, which is the FN risk threshold score at which they would use G-CSF PP in their usual clinical practice.

Investigators will then prospectively and sequentially identify eligible subjects with NHL, breast or lung cancer who are due to initiate one of the permitted standard dose chemotherapy regimens listed in the protocol. The permitted chemotherapy regimens have an estimated intermediate FN risk (10%-20%) documented in published data and/or international guidelines.

For each enrolled subject, Investigators will complete a Subject Assessment prior to the start of their chemotherapy. They will be provided with the same two lists of risk factors as in the Baseline Assessment and asked to complete them based on each specific subject. Investigators must rank selected risk factors that they consider to be the most important when assessing 1) overall FN risk (only scientific factors will be included), and 2) when deciding on whether G-CSF PP treatment will be used or not. They will also document their final estimated FN risk score as a percentage based on the subject's medical history and standard of care (SOC) assessments (their routine practice for assessing this risk), and a decision as to whether G-CSF PP will be administered. Investigators will record which type of G-CSF they plan to use if one will be used.

End of Study for a subject will occur once these activities have been completed, and a prescription for the first cycle of chemotherapy has been written. The subject data collected will only be historical subject information and laboratory data from SOC assessments performed prior to beginning chemotherapy treatment. No data will be collected after the initiation of chemotherapy.

The approach to the statistical analysis will be generally descriptive in nature. The primary analysis will be conducted at two levels; investigator level and the subject level. It is expected that the opinions of investigators at a single site (that is, a department within a cancer treatment centre) will be correlated. Also, that the opinions about subjects from a single investigator will be more alike than subjects of other investigators; adjustments will be made in the analyses to account for this. Confidence intervals for the investigator level analysis and the subject level data will obtained from Multi-level Modelling (MLM) to allow for the expected intra-site and intra-investigator correlation of investigators within sites and subjects within investigator. In general, categorical data will be summarised by the number and percentage of subjects in each category. Continuous data will be summarised by mean, standard deviation, median, lower and upper quartiles, minimum and maximum values. Two-sided exact 95% confidence intervals (obtained using MLM) will be presented, where appropriate.

Recruitment & Eligibility

Status: COMPLETED

Sex: All

Target Recruitment: 1007

Inclusion Criteria

Age ≥ 18 years old
Any stage NHL, small cell lung cancer (SCLC), non-small cell lung cancer (NSCLC), or breast cancer initiating a new chemotherapy course
Scheduled to receive one of the permitted standard dose chemotherapy regimens with an estimated intermediate (10%-20%) FN risk according to published data or guidelines (planned dose modifications +/-10% are allowable).
Before any study-specific procedure, the appropriate written informed consent must be obtained where this is required by local regulations

Exclusion Criteria

Ongoing or planned concurrent participation in any clinical study involving Investigational Product that has not been approved by the European Medicines Agency (EMA) or competent authority for any indication,
Ongoing or planned concurrent participation in any clinical study where the administration of Colony Stimulating Factor (CSF) is determined by the protocol (clinical trials on an approved drug and observational trials are permitted as long as these do not mandate how neutropenia should be treated)
Prior stem-cell transplantation (includes bone marrow transplantation)

Study & Design

Study Type: OBSERVATIONAL

Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Percentage of Participants for Whom Age and Chemotherapy Regimen Were Ranked as an Important Risk Factor	At enrolment, prior to chemotherapy initiation	Investigators ranked the risk factors they considered to be the most important when assessing the overall risk of febrile neutopenia for each participant. Only historical patient information recorded before the beginning of chemotherapy treatment were collected in this study. Age and chemotherapy regimen were specified in the protocol as risk factors of interest. Reported are age and chemotherapeutic agents ranked individually, chemotherapy agents detailed by specific factors, and age and chemotherapy agents jointly ranked. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals.
Percentage of Participants for Whom Each FN Risk Factor Was Ranked as Important	At enrolment, prior to chemotherapy initiation.	Investigators ranked the risk factors they considered to be the most important when assessing the overall risk of febrile neutopenia for each participant. Only historical patient information recorded before the beginning of chemotherapy treatment was collected in this study. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Investigators Who Ranked Each Factor as a Risk Factor for Febrile Neutropenia (FN)	Assessed at Baseline, prior to participant enrolment.	During the baseline investigator assessment (prior to identification of participants), investigators were provided a list of risk factors on a source document worksheet and asked to rank the risk factors that they considered to be the most important when assessing overall FN risk. To account for the expected correlation between investigators at the same sites, two-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group
Percentage of Investigators Who Ranked Age and Chemotherapy Regimen as a Risk Factor for Febrile Neutropenia	Baseline (prior to participant enrolment)	During the baseline investigator assessment (prior to identification of participants), investigators were provided a list of risk factors on a source document worksheet, and asked to rank the risk factors that they considered to be the most important when assessing overall febrile neutropenia (FN) risk. Age and chemotherapy regimen were specified in the protocol as risk factors of interest. Reported are age and chemotherapeutic agents ranked individually, chemotherapy agents detailed by specific factors, and age and chemotherapy agents jointly ranked. To account for the expected correlation between investigators at the same sites, two-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals.

Secondary Outcome Measures

Name	Time	Method
Percentage of Participants for Whom Each Factor Was Ranked in the G-CSF PP Decision as Important by Number of Years in Clinical Practice in Oncology / Hematology	At enrolment, prior to chemotherapy initiation.	For each participant, the investigator ranked the risk factors that they considered to be the most important factors that they considered when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Investigators Who Ranked Each Factor in the G-CSF PP Decision as Important by Number of Years in Clinical Practice in Oncology / Hematology	Assessed at baseline, prior to participant enrolment.	During the baseline investigator assessment (prior to identification of participants), investigators were provided a list of factors on a source document worksheet, and asked to rank the factors that they considered to be the most important when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between investigators at the same sites, two-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. Subgroup analyses were performed where a subgroup contained at least 40 investigators. ECOG = Eastern Cooperative Oncology Group.
Percentage of Participants for Whom Each FN Risk Factor Was Ranked as Important by Clinical Specialty	At enrolment, prior to chemotherapy initiation.	Investigators ranked the risk factors they considered to be the most important when assessing the overall risk of febrile neutopenia for each participant. Only historical patient information recorded before the beginning of chemotherapy treatment was collected in this study. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Investigators Who Ranked Each Factor in the Granulocyte Colony Stimulating Factor (G-CSF) Primary Prophylaxis (PP) Decision as Important	Assessed at baseline, prior to participant enrolment.	During the baseline investigator assessment (prior to identification of participants), investigators were provided a list of factors on a source document worksheet, and asked to rank the factors that they considered to be the most important when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between investigators at the same sites, two-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Investigators Who Ranked Each Factor in the G-CSF PP Decision as Important by Institution Type	Assessed at baseline, prior to participant enrolment.	During the baseline investigator assessment (prior to identification of participants), investigators were provided a list of factors on a source document worksheet, and asked to rank the factors that they considered to be the most important when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between investigators at the same sites, two-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. Subgroup analyses were performed where a subgroup contained at least 40 investigators. ECOG = Eastern Cooperative Oncology Group.
Percentage of Participants for Whom Each Factor Was Ranked in the Granulocyte Colony Stimulating Factor (G-CSF) Primary Prophylaxis (PP) Decision as Important	At enrolment, prior to chemotherapy initiation.	For each participant, the investigator ranked the factors that they considered to be the most important factors that they considered when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Participants for Whom Each Factor Was Ranked in the G-CSF PP Decision as Important by Tumor Type	At enrolment, prior to chemotherapy initiation.	For each participant, the investigator ranked the risk factors that they considered to be the most important factors that they considered when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Investigators Who Ranked Each Factor in the G-CSF PP Decision as Important by Clinical Specialty	Assessed at baseline, prior to participant enrolment.	During the baseline investigator assessment (prior to identification of participants), investigators were provided a list of factors on a source document worksheet, and asked to rank the factors that they considered to be the most important when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between investigators at the same sites, two-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group. Subgroup analyses were performed where a subgroup contained at least 40 investigators. Results are reported for medical oncologists as this was the only specialty that contained at least 40 investigators.
Percentage of Participants for Whom Each Factor Was Ranked in the G-CSF PP Decision as Important by Country	At enrolment, prior to chemotherapy initiation.	For each participant, the investigator ranked the risk factors that they considered to be the most important factors that they considered when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Participants for Whom Each Factor Was Ranked in the G-CSF PP Decision as Important by Clinical Specialty	At enrolment, prior to chemotherapy initiation.	For each participant, the investigator ranked the risk factors that they considered to be the most important factors that they considered when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Investigators Who Ranked Each Factor as a Risk Factor for Febrile Neutropenia by Clinical Specialty	Assessed at Baseline, prior to participant enrolment.	During the baseline investigator assessment (prior to identification of participants), investigators were provided a list of risk factors on a source document worksheet, and asked to rankt the risk factors that they considered to be the most important when assessing overall FN risk. To account for the expected correlation between investigators at the same sites, two-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group. Subgroup analyses were performed where a subgroup contained at least 40 investigators. Results are reported for medical oncologists as this was the only specialty that contained at least 40 investigators.
Percentage of Investigators Who Ranked Each Factor as a Risk Factor for Febrile Neutropenia by Number of Years in Clinical Practice in Oncology / Hematology	Assessed at Baseline, prior to participant enrolment.	During the baseline investigator assessment (prior to identification of participants), investigators were provided a list of risk factors on a source document worksheet, and asked to rank the risk factors that they considered to be the most important when assessing overall FN risk. To account for the expected correlation between investigators at the same sites, two-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Participants for Whom Each FN Risk Factor Was Ranked as Important by Country	At enrolment, prior to chemotherapy initiation.	Investigators ranked the risk factors they considered to be the most important when assessing the overall risk of febrile neutopenia for each participant. Only historical patient information recorded before the beginning of chemotherapy treatment was collected in this study. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Participants With an Investigator-assessed FN Risk at or Above the Investigator Self-reported FN-risk Intervention Threshold Who Were Planned to Receive G-CSF PP	At Baseline and at enrolment, prior to chemotherapy initiation.	At Baseline investigators recorded the FN risk threshold score at which they would use G-CSF PP in their usual clinical practice. For each enrolled participant, the investigator documented their final estimated FN risk score as a percentage based on the participant's medical history and standard of care assessments (their routine practice for assessing this risk), and a decision as to whether G-CSF PP would be administered in Cycle 1.
Percentage of Participants for Whom Each Factor Was Ranked in the G-CSF PP Decision as Important by Institution Type	At enrolment, prior to chemotherapy initiation.	For each participant, the investigator ranked the risk factors that they considered to be the most important factors that they considered when deciding whether to use G-CSF PP treatment or not. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Investigators Who Ranked Each Factor as a Risk Factor for Febrile Neutropenia by Institution Type	Assessed at Baseline, prior to participant enrolment.	During the baseline investigator assessment (prior to identification of participants), investigators were provided a list of risk factors on a source document worksheet, and asked to rank the factors that they considered to be the most important when assessing overall FN risk. To account for the expected correlation between investigators at the same sites, two-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Participants for Whom Each FN Risk Factor Was Ranked as Important by Tumor Type	At enrolment, prior to chemotherapy initiation.	Investigators ranked the risk factors they considered to be the most important when assessing the overall risk of febrile neutopenia for each participant. Only historical patient information recorded before the beginning of chemotherapy treatment was collected in this study. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Participants for Whom Each FN Risk Factor Was Ranked as Important by Number of Years in Clinical Practice in Oncology / Hematology	At enrolment, prior to chemotherapy initiation.	Investigators ranked the risk factors they considered to be the most important when assessing the overall risk of febrile neutopenia for each participant. Only historical patient information recorded before the beginning of chemotherapy treatment was collected in this study. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.
Percentage of Participants for Whom Each FN Risk Factor Was Ranked as Important by Institution Type	At enrolment, prior to chemotherapy initiation.	Investigators ranked the risk factors they considered to be the most important when assessing the overall risk of febrile neutopenia for each participant. Only historical patient information recorded before the beginning of chemotherapy treatment was collected in this study. To account for the expected correlation between participants with the same investigators and between investigators at the same sites, three-level, empty (no explanatory variables) multilevel models were used in the estimation of the percentages and 95% confidence intervals. ECOG = Eastern Cooperative Oncology Group.