Non-Small Cell Lung Cancer
Conditions
Brief summary
This randomized phase 2 trial is studying the effect of adding denosumab to standard chemotherapy in the treatment of advanced lung cancer.
Detailed description
This is a global randomized double-blind placebo-controlled study in patients with Stage IV untreated non-small cell lung cancer (NSCLC) with or without bone metastasis. Eligible participants are to receive 4 to 6 cycles of a standard of care platinum-doublet chemotherapy regimen. Participants will be randomized in a 2:1 ratio to receive denosumab or matching placebo with the first investigational product dose coinciding with participant's first cycle of chemotherapy and continuing until the primary analysis, unacceptable toxicity, withdrawal of consent, death, or lost to follow-up. Participants who discontinued the investigational product early (ie, before primary analysis) were followed for disease status and survival. The primary analysis took place when 149 events of death had been reported. All participants were followed for 2 years after the last dose of blinded investigational product.
Interventions
DRUGDenosumab
Administered by subcutaneous injection once every 4 weeks (Q4W) plus one loading dose on study day 8; could be administered as often as every 3 weeks (Q3W) to participants receiving Q3W chemotherapy.
DRUGZoledronic acid
Administered by intravenous infusion in participants with bone metastasis upon investigative site request for IV bone-targeted therapy.
Administered by subcutaneous injection once every 4 weeks (Q4W) plus one loading dose on study day 8; could be administered as often as every 3 weeks (Q3W) to participants receiving Q3W chemotherapy.
DRUGStandard Chemotherapy
Standard of care chemotherapy consisting of pemetrexed or gemcitabine in combination with cisplatin or carboplatin administered according to local practice.
Administered by intravenous infusion in participants with bone metastasis upon investigative site request for IV bone-targeted therapy.
Sponsors
Amgen
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
TREATMENT
Masking
DOUBLE (Subject, Investigator)
Eligibility
Sex/Gender
ALL
Age
18 Years to 100 Years
Healthy volunteers
No
Inclusion criteria
* Histologically or cytologically confirmed stage IV non-small cell lung carcinoma (NSCLC), according to 7th Tumor/Node/Metastasis (TNM) classification (cytological specimens obtained by bronchial washing or brushing, or fine-needle aspiration are acceptable) * Subject has available and has provided consent to release to the sponsor (or designee) a tumor block with confirmed tumor content (or approximately 20 unstained charged slides \[a minimum of 7 slides is mandatory\]) and the corresponding pathology report * Planned to receive 4 to 6 cycles of pemetrexed or gemcitabine in combination with cisplatin or carboplatin • For subjects to receive pemetrexed, planned to receive vitamin B12 and folate per pemetrexed approved labeling * Radiographically evaluable (measurable or non-measurable) disease (according to modified Response Evaluation Criteria In Solid Tumors (RECIST) 1.1 criteria * Other inclusion criteria may apply
Exclusion criteria
* Known presence of documented sensitizing epidermal growth factor receptor (EGFR) activating mutation or echinoderm microtubule-associated protein-like 4-anaplastic lymphoma kinase (EML4-ALK) translocation (screening following local standards, but strongly encouraged in non-squamous histology) * Known brain metastases (systematic screening of patients not mandatory) * Any prior systemic therapy (before randomization) for the treatment of NSCLC (including chemoradiation), except if for non-metastatic disease and was completed at least 6 months prior to randomization * Planned to receive bevacizumab * Significant dental/oral disease, including prior history or current evidence of osteonecrosis/ osteomyelitis of the jaw, or with the following: * Active dental or jaw condition which requires oral surgery * Non-healed dental/oral surgery * Planned invasive dental procedures for the course of the study.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Overall Survival (OS) | From randomization until the end of study; median time on study was 9.64 months. | Overall survival was calculated as the time from the date of randomization to the date of death from any cause. Participants last known to be alive were censored at the last contact date. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Correlation of Tumor Tissue RANK Ligand Expression With Overall Survival | From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months. | To assess whether the treatment effect on overall survival was correlated with RANK ligand (RANKL) protein expression in tumor cells, RANKL expression in archival tumor samples was measured using immunohistochemistry. The intensity of stain in the cytoplasm was categorized as 0 (negative), 1+ (weak), 2+ (moderate) or 3+ (strong); All intensity is the sum of levels +1, +2 and +3. In addition, an H-score was calculated using the following formula: H-score=(percentage of cells of weak×1)+(percentage of cells of moderate×2)+(percentage of cells of strong×3). The maximum H-score was 300, corresponding to 100% of cells with strong intensity. The correlation between RANKL expression level and OS was evaluated using a Cox proportional hazard models that included RANKL expression level stratified by the randomization stratification factors in the corresponding treatment group. |
| Objective Response Rate | From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months. | Objective response rate was defined as the percentage of participants with a complete response (CR) or partial response (PR) based on modified RECIST 1.1 achieved over the study duration. CR: Disappearance of all target and non target lesions, normalization of tumor marker levels and no new lesions. PR: At least a 30% decrease in the size of target lesions with no progression of non-target lesions and no new lesions, or, disappearance of target lesions with persistence of one or more non-target lesions and/or maintenance of tumor marker levels above normal limits and no new lesions. Participants who underwent surgical resection while on study were not evaluated for response after the surgery. Participants who did not meet the criteria for an objective response by the analysis cutoff date were considered non-responders. |
| Correlation of Tumor Tissue RANK Expression With Objective Response Rate | From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months. | To assess whether the treatment effect on objective response rate (ORR) based on RECIST 1.1 was correlated with RANK protein expression in tumor cells, RANK expression in archival tumor samples was measured using immunohistochemistry. The intensity of stain in the cytoplasm, membrane, and total was categorized as 0 (negative), 1+ (weak), 2+ (moderate) or 3+ (strong); All intensity is the sum of levels +1, +2 and +3. In addition, an H-score was calculated using the following formula: H-score=(percentage of cells of weak×1)+(percentage of cells of moderate×2)+(percentage of cells of strong×3). The maximum H-score was 300, corresponding to 100% of cells with strong intensity. The correlation between RANK expression level and ORR was evaluated within each treatment group using a logistical regression model that included RANK expression value as an independent variable and stratified by the randomization stratification factors. The odds ratio and 95% confidence interval are reported. |
| Correlation of Tumor Tissue RANKL Expression With Objective Response Rate | From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months. | To assess whether the treatment effect on objective response rate (ORR) based on RECIST 1.1 was correlated with RANKL protein expression in tumor cells, RANKL expression in archival tumor samples was measured using immunohistochemistry. The intensity of stain in the cytoplasm was categorized as 0 (negative), 1+ (weak), 2+ (moderate) or 3+ (strong); All intensity is the sum of levels +1, +2 and +3. In addition, an H-score was calculated using the following formula: H-score=(percentage of cells of weak×1)+(percentage of cells of moderate×2)+(percentage of cells of strong×3). The maximum H-score was 300, corresponding to 100% of cells with strong intensity. The correlation between RANKL expression level and ORR was evaluated within each treatment group using a logistical regression model that included RANKL expression value as an independent variable and stratified by the randomization stratification factors. The odds ratio and 95% confidence interval are reported. |
| Correlation of Tumor Tissue RANK Expression With Overall Survival | From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months. | To assess whether the treatment effect on overall survival was correlated with receptor activator of nuclear factor (NF)-κB (RANK) protein expression in tumor cells, RANK expression in archival tumor samples was measured using immunohistochemistry. The intensity of stain in the cytoplasm, membrane, and total was categorized as 0 (negative), 1+ (weak), 2+ (moderate) or 3+ (strong); All intensity is the sum of levels +1, +2 and +3. In addition, an H-score was calculated using the following formula: H-score=(percentage of cells of weak×1)+(percentage of cells of moderate×2)+(percentage of cells of strong×3). The maximum H-score was 300, corresponding to 100% of cells with strong intensity. The correlation between RANK expression level and OS was evaluated using a Cox proportional hazard models that included RANK expression level stratified by the randomization stratification factors in the corresponding treatment group. |
| Progression-free Survival (PFS) | From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months. | Progression-free survival was defined as the time from randomization to the first observed disease progression per modified RECIST 1.1 criteria or death from any cause. Participants last known to be alive who did not experience disease progression were censored at their last imaging assessment date, last contact date if they were in the survival follow up phase, end of the study date, or the primary analysis cut-off date, whichever was first. If a participant underwent surgical resection while on study, the participant was censored at the last evaluable imaging assessment prior to the surgery. |
| Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Prior to dosing at day 8 and weeks 3, 6, 9, 12, 15, 18, 21 and 24. | Serum samples were analyzed for denosumab using enzyme-linked immunosorbent assay (ELISA) following a validated procedure. The lower limit of quantification for the assay was 20 ng/mL. |
| Serum Denosumab Trough Levels in Participants Who Received Q4W Dosing | Prior to dosing at day 8 and weeks 4, 8, 12, 16, 20 and 24 | Serum samples were analyzed for denosumab using enzyme-linked immunosorbent assay (ELISA) following a validated procedure. The lower limit of quantification for the assay was 20 ng/mL. |
| Number of Participants With Treatment-emergent Adverse Events | From first dose of study drug to the end of study date; the median (min, max) duration was 10.0 (0.2, 41.4) and 9.4 (0.2, 42.9) months for Placebo and Denosumab respectively. | An adverse event (AE) is defined as any untoward medical occurrence in a clinical trial participant. The event does not necessarily have a causal relationship with study treatment. A serious adverse event is defined as an AE that meets at least 1 of the following criteria * fatal * life threatening * requires in-patient hospitalization or prolongation of existing hospitalization * results in persistent or significant disability/incapacity * congenital anomaly/birth defect * other medically important serious event Treatment-related AEs include only TEAEs for which the investigator indicated there was a reasonable possibility they may have been caused by study drug. Fatal adverse events include only deaths reported on the Adverse Event Case Report Form. |
| Clinical Benefit Rate | From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months. | Clinical benefit rate was defined as the percentage of participants with an objective response (CR or PR) or stable disease (SD) or better for at least 16 weeks achieved over the study duration. If a participant underwent surgical resection while on study, the participant was not evaluated for response after the surgery. Participants who did not meet the criteria for clinical benefit by the analysis cutoff date were considered as non-responders. |
Countries
Australia, Canada, Czechia, France, Germany, Greece, Italy, Netherlands, United Kingdom, United States
Participant flow
Recruitment details
This study was conducted at 57 centers in 10 countries including the United Kingdom, Netherlands, France, Italy, Germany, Czech Republic, United States, Canada, Australia and Greece. Participants were enrolled from 31 December 2013 to 21 May 2015.
Pre-assignment details
Participants were randomized in a 2:1 ratio to receive denosumab or placebo. Randomization was stratified based on the presence of bone metastasis (yes or no), histology (squamous versus non-squamous), and geographic region (North America, Western Europe/Australia, and rest of world \[ROW\]).
Participants by arm
| Arm | Count |
|---|---|
| Placebo Participants were randomized to receive placebo matching to denosumab by subcutaneous injection once every 4 weeks (Q4W) plus one loading dose on study day 8 in addition to platinum-based standard chemotherapy. Participants may have received placebo as often as once every 3 weeks (Q3W) while receiving Q3W chemotherapy. Participants with bone metastases also received 4 mg zoledronic acid administered as an IV infusion Q4W or Q3W. | 78 |
| Denosumab Participants were randomized to receive 120 mg denosumab by subcutaneous injection once every 4 weeks plus one loading dose on study day 8 in addition to platinum-based standard chemotherapy. Participants may have received denosumab as often as Q3W while receiving Q3W chemotherapy. Participants with bone metastases also received placebo to zoledronic acid administered as an IV infusion Q4W or Q3W. | 148 |
| Total | 226 |
Withdrawals & dropouts
| Period | Reason | FG000 | FG001 |
|---|---|---|---|
| Overall Study | Death | 61 | 120 |
| Overall Study | Lost to Follow-up | 0 | 1 |
| Overall Study | Sponsor Decision | 9 | 10 |
| Overall Study | Withdrawal by Subject | 8 | 17 |
Baseline characteristics
| Characteristic | Denosumab | Total | Placebo |
|---|---|---|---|
| Age, Continuous | 63.8 years STANDARD_DEVIATION 9.7 | 63.8 years STANDARD_DEVIATION 9.8 | 64.0 years STANDARD_DEVIATION 10.1 |
| Eastern Cooperative Oncology Group (ECOG) Performance Status 0 (Fully active) | 69 Participants | 102 Participants | 33 Participants |
| Eastern Cooperative Oncology Group (ECOG) Performance Status 1 (Restrictive but ambulatory) | 79 Participants | 124 Participants | 45 Participants |
| Geographic region North America/Australia | 63 Participants | 93 Participants | 30 Participants |
| Geographic region Rest of World | 18 Participants | 31 Participants | 13 Participants |
| Geographic region Western Europe | 67 Participants | 102 Participants | 35 Participants |
| Histology Non-squamous | 113 Participants | 173 Participants | 60 Participants |
| Histology Squamous | 35 Participants | 53 Participants | 18 Participants |
| Presence of Bone Metastasis No | 83 Participants | 126 Participants | 43 Participants |
| Presence of Bone Metastasis Yes | 65 Participants | 100 Participants | 35 Participants |
| Race/Ethnicity, Customized Black or African American | 7 Participants | 9 Participants | 2 Participants |
| Race/Ethnicity, Customized Others | 8 Participants | 11 Participants | 3 Participants |
| Race/Ethnicity, Customized White or Caucasian | 133 Participants | 206 Participants | 73 Participants |
| Sex: Female, Male Female | 54 Participants | 74 Participants | 20 Participants |
| Sex: Female, Male Male | 94 Participants | 152 Participants | 58 Participants |
| Time from Initial Diagnosis of NSCLC to Randomization | 6.32 months STANDARD_DEVIATION 16.96 | 4.99 months STANDARD_DEVIATION 14.17 | 2.46 months STANDARD_DEVIATION 5.28 |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 71 / 76 | 139 / 145 |
| serious Total, serious adverse events | 68 / 76 | 129 / 145 |
Outcome results
Primary
Overall Survival (OS)
Overall survival was calculated as the time from the date of randomization to the date of death from any cause. Participants last known to be alive were censored at the last contact date.
Time frame: From randomization until the end of study; median time on study was 9.64 months.
Population: All randomized participants
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Placebo | Overall Survival (OS) | 10.9 months |
| Denosumab | Overall Survival (OS) | 10.7 months |
p-value: 0.515795% CI: [0.81, 1.53]Log Rank
Secondary
Clinical Benefit Rate
Clinical benefit rate was defined as the percentage of participants with an objective response (CR or PR) or stable disease (SD) or better for at least 16 weeks achieved over the study duration. If a participant underwent surgical resection while on study, the participant was not evaluated for response after the surgery. Participants who did not meet the criteria for clinical benefit by the analysis cutoff date were considered as non-responders.
Time frame: From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months.
Population: Randomized participants with at least one baseline measurable lesion per modified RECIST 1.1.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Placebo | Clinical Benefit Rate | 53.9 percentage of participants |
| Denosumab | Clinical Benefit Rate | 47.9 percentage of participants |
p-value: 0.465495% CI: [0.46, 1.43]Regression, Logistic
Secondary
Correlation of Tumor Tissue RANK Expression With Objective Response Rate
To assess whether the treatment effect on objective response rate (ORR) based on RECIST 1.1 was correlated with RANK protein expression in tumor cells, RANK expression in archival tumor samples was measured using immunohistochemistry. The intensity of stain in the cytoplasm, membrane, and total was categorized as 0 (negative), 1+ (weak), 2+ (moderate) or 3+ (strong); All intensity is the sum of levels +1, +2 and +3. In addition, an H-score was calculated using the following formula: H-score=(percentage of cells of weak×1)+(percentage of cells of moderate×2)+(percentage of cells of strong×3). The maximum H-score was 300, corresponding to 100% of cells with strong intensity. The correlation between RANK expression level and ORR was evaluated within each treatment group using a logistical regression model that included RANK expression value as an independent variable and stratified by the randomization stratification factors. The odds ratio and 95% confidence interval are reported.
Time frame: From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months.
Population: All randomized participants with at least one baseline measurable lesion per modified RECIST 1.1 who had evaluable pre-treatment tumor RANK expression.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Placebo | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Total all intensity | 1.16 odds ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Membrane H-score | 1.14 odds ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Cytoplasm all intensity | 1.00 odds ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Cytoplasm H-score | 1.00 odds ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Membrane all intensity | 1.18 odds ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Total H-score | 1.14 odds ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Membrane all intensity | 0.82 odds ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Cytoplasm all intensity | 0.88 odds ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Total all intensity | 0.88 odds ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Cytoplasm H-score | 0.87 odds ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Membrane H-score | 0.82 odds ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Objective Response Rate | Total H-score | 0.87 odds ratio |
Comparison: The interaction of RANK expression level measured in the cytoplasm (all intensity) and treatment effect was evaluated using a logistic regression model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.476Regression, Logistic
Comparison: The interaction of RANK expression level measured in the membranes (all intensity) and treatment effect was evaluated using a logistic regression model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.2582Regression, Logistic
Comparison: The interaction of RANK expression level total (cytoplasm + membrane; all intensity) and treatment effect was evaluated using a logistic regression model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.223Regression, Logistic
Comparison: The interaction of RANK expression level measured in the cytoplasm (H-score) and treatment effect was evaluated using a logistic regression model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.4329Regression, Logistic
Comparison: The interaction of RANK expression level measured in the membranes (H-score) and treatment effect was evaluated using a logistic regression model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.262Regression, Logistic
Comparison: The interaction of RANK expression level total (cytoplasm + membrane; H-score) and treatment effect was evaluated using a logistic regression model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.2081Regression, Logistic
Secondary
Correlation of Tumor Tissue RANK Expression With Overall Survival
To assess whether the treatment effect on overall survival was correlated with receptor activator of nuclear factor (NF)-κB (RANK) protein expression in tumor cells, RANK expression in archival tumor samples was measured using immunohistochemistry. The intensity of stain in the cytoplasm, membrane, and total was categorized as 0 (negative), 1+ (weak), 2+ (moderate) or 3+ (strong); All intensity is the sum of levels +1, +2 and +3. In addition, an H-score was calculated using the following formula: H-score=(percentage of cells of weak×1)+(percentage of cells of moderate×2)+(percentage of cells of strong×3). The maximum H-score was 300, corresponding to 100% of cells with strong intensity. The correlation between RANK expression level and OS was evaluated using a Cox proportional hazard models that included RANK expression level stratified by the randomization stratification factors in the corresponding treatment group.
Time frame: From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months.
Population: All randomized participants who had evaluable pre-treatment tumor RANK expression.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Placebo | Correlation of Tumor Tissue RANK Expression With Overall Survival | Cytoplasm H-score | 0.83 hazard ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Overall Survival | Membrane all intensity | 0.72 hazard ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Overall Survival | Membrane H-score | 0.72 hazard ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Overall Survival | Total all intensity | 0.80 hazard ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Overall Survival | Total H-score | 0.80 hazard ratio |
| Placebo | Correlation of Tumor Tissue RANK Expression With Overall Survival | Cytoplasm all intensity | 0.84 hazard ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Overall Survival | Total H-score | 1.00 hazard ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Overall Survival | Membrane all intensity | 0.92 hazard ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Overall Survival | Total all intensity | 1.00 hazard ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Overall Survival | Cytoplasm H-score | 1.00 hazard ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Overall Survival | Membrane H-score | 0.93 hazard ratio |
| Denosumab | Correlation of Tumor Tissue RANK Expression With Overall Survival | Cytoplasm all intensity | 1.00 hazard ratio |
Comparison: The interaction of RANK expression level measured in the cytoplasm (all intensity) and treatment effect was evaluated using a Cox proportional hazard model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.3759Cox propotional hazard model
Comparison: The interaction of RANK expression level measured in membranes (all intensity) and treatment effect was evaluated using a Cox proportional hazard model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.3666Cox propotional hazard model
Comparison: The interaction of RANK expression level total (cytoplasm + membranes; all intensity) and treatment effect was evaluated using a Cox proportional hazard model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.1917Cox propotional hazard model
Comparison: The interaction of RANK expression level measured in the cytoplasm (H-score) and treatment effect was evaluated using a Cox proportional hazard model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.3353Cox propotional hazard model
Comparison: The interaction of RANK expression level measured in membranes (H-score) and treatment effect was evaluated using a Cox proportional hazard model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.3179Cox propotional hazard model
Comparison: The interaction of RANK expression level total (cytoplasm + membranes; H-score) and treatment effect was evaluated using a Cox proportional hazard model that included treatment group, RANK expression level and treatment-by-RANK expression level interaction stratified by the randomization stratification factors.p-value: 0.1583Cox propotional hazard model
Secondary
Correlation of Tumor Tissue RANKL Expression With Objective Response Rate
To assess whether the treatment effect on objective response rate (ORR) based on RECIST 1.1 was correlated with RANKL protein expression in tumor cells, RANKL expression in archival tumor samples was measured using immunohistochemistry. The intensity of stain in the cytoplasm was categorized as 0 (negative), 1+ (weak), 2+ (moderate) or 3+ (strong); All intensity is the sum of levels +1, +2 and +3. In addition, an H-score was calculated using the following formula: H-score=(percentage of cells of weak×1)+(percentage of cells of moderate×2)+(percentage of cells of strong×3). The maximum H-score was 300, corresponding to 100% of cells with strong intensity. The correlation between RANKL expression level and ORR was evaluated within each treatment group using a logistical regression model that included RANKL expression value as an independent variable and stratified by the randomization stratification factors. The odds ratio and 95% confidence interval are reported.
Time frame: From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months.
Population: All randomized participants with at least one baseline measurable lesion per modified RECIST 1.1 who had evaluable pre-treatment tumor RANKL expression.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Placebo | Correlation of Tumor Tissue RANKL Expression With Objective Response Rate | Cytoplasm all intensity | 1.27 odds ratio |
| Placebo | Correlation of Tumor Tissue RANKL Expression With Objective Response Rate | Cytoplasm H-score | 1.25 odds ratio |
| Denosumab | Correlation of Tumor Tissue RANKL Expression With Objective Response Rate | Cytoplasm all intensity | 1.10 odds ratio |
| Denosumab | Correlation of Tumor Tissue RANKL Expression With Objective Response Rate | Cytoplasm H-score | 1.09 odds ratio |
Comparison: The interaction of RANKL expression level measured in the cytoplasm (all intensity) and treatment effect was evaluated using a logistic regression model that included treatment group, RANKL expression level and treatment-by-RANKL expression level interaction stratified by the randomization stratification factors.p-value: 0.4671Regression, Logistic
Comparison: The interaction of RANKL expression level measured in the cytoplasm (H-score) and treatment effect was evaluated using a logistic regression model that included treatment group, RANKL expression level and treatment-by-RANKL expression level interaction stratified by the randomization stratification factors.p-value: 0.4236Regression, Logistic
Secondary
Correlation of Tumor Tissue RANK Ligand Expression With Overall Survival
To assess whether the treatment effect on overall survival was correlated with RANK ligand (RANKL) protein expression in tumor cells, RANKL expression in archival tumor samples was measured using immunohistochemistry. The intensity of stain in the cytoplasm was categorized as 0 (negative), 1+ (weak), 2+ (moderate) or 3+ (strong); All intensity is the sum of levels +1, +2 and +3. In addition, an H-score was calculated using the following formula: H-score=(percentage of cells of weak×1)+(percentage of cells of moderate×2)+(percentage of cells of strong×3). The maximum H-score was 300, corresponding to 100% of cells with strong intensity. The correlation between RANKL expression level and OS was evaluated using a Cox proportional hazard models that included RANKL expression level stratified by the randomization stratification factors in the corresponding treatment group.
Time frame: From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months.
Population: All randomized participants who had evaluable pre-treatment tumor RANKL expression.
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Placebo | Correlation of Tumor Tissue RANK Ligand Expression With Overall Survival | Cytoplasm H-score | 0.79 hazard ratio |
| Placebo | Correlation of Tumor Tissue RANK Ligand Expression With Overall Survival | Cytoplasm all intensity | 0.77 hazard ratio |
| Denosumab | Correlation of Tumor Tissue RANK Ligand Expression With Overall Survival | Cytoplasm all intensity | 0.93 hazard ratio |
| Denosumab | Correlation of Tumor Tissue RANK Ligand Expression With Overall Survival | Cytoplasm H-score | 0.93 hazard ratio |
Comparison: The interaction of RANKL expression level measured in the cytoplasm (all intensity) and treatment effect was evaluated using a Cox proportional hazard model that included treatment group, RANKL expression level and treatment-by-RANKL expression level interaction stratified by the randomization stratification factors.p-value: 0.3946Cox propotional hazard model
Comparison: The interaction of RANKL expression level measured in the cytoplasm (H-score) and treatment effect was evaluated using a Cox proportional hazard model that included treatment group, RANKL expression level and treatment-by-RANKL expression level interaction stratified by the randomization stratification factors.p-value: 0.3735Cox propotional hazard model
Secondary
Number of Participants With Treatment-emergent Adverse Events
An adverse event (AE) is defined as any untoward medical occurrence in a clinical trial participant. The event does not necessarily have a causal relationship with study treatment. A serious adverse event is defined as an AE that meets at least 1 of the following criteria * fatal * life threatening * requires in-patient hospitalization or prolongation of existing hospitalization * results in persistent or significant disability/incapacity * congenital anomaly/birth defect * other medically important serious event Treatment-related AEs include only TEAEs for which the investigator indicated there was a reasonable possibility they may have been caused by study drug. Fatal adverse events include only deaths reported on the Adverse Event Case Report Form.
Time frame: From first dose of study drug to the end of study date; the median (min, max) duration was 10.0 (0.2, 41.4) and 9.4 (0.2, 42.9) months for Placebo and Denosumab respectively.
Population: All randomized participants who received at least one dose of study drug (denosumab or placebo)
| Arm | Measure | Group | Value (NUMBER) |
|---|---|---|---|
| Placebo | Number of Participants With Treatment-emergent Adverse Events | Treatment-related fatal adverse events | 0 participants |
| Placebo | Number of Participants With Treatment-emergent Adverse Events | AEs leading to discontinuation of study drug | 6 participants |
| Placebo | Number of Participants With Treatment-emergent Adverse Events | All adverse events (AEs) | 76 participants |
| Placebo | Number of Participants With Treatment-emergent Adverse Events | Serious adverse events | 68 participants |
| Placebo | Number of Participants With Treatment-emergent Adverse Events | Fatal adverse events | 53 participants |
| Placebo | Number of Participants With Treatment-emergent Adverse Events | Treatment-related adverse events (TRAEs) | 24 participants |
| Placebo | Number of Participants With Treatment-emergent Adverse Events | Treatment-related serious adverse events | 0 participants |
| Placebo | Number of Participants With Treatment-emergent Adverse Events | TRAEs leading to discontinuation of study drug | 0 participants |
| Denosumab | Number of Participants With Treatment-emergent Adverse Events | All adverse events (AEs) | 144 participants |
| Denosumab | Number of Participants With Treatment-emergent Adverse Events | Serious adverse events | 129 participants |
| Denosumab | Number of Participants With Treatment-emergent Adverse Events | Fatal adverse events | 113 participants |
| Denosumab | Number of Participants With Treatment-emergent Adverse Events | AEs leading to discontinuation of study drug | 20 participants |
| Denosumab | Number of Participants With Treatment-emergent Adverse Events | TRAEs leading to discontinuation of study drug | 7 participants |
| Denosumab | Number of Participants With Treatment-emergent Adverse Events | Treatment-related fatal adverse events | 0 participants |
| Denosumab | Number of Participants With Treatment-emergent Adverse Events | Treatment-related serious adverse events | 6 participants |
| Denosumab | Number of Participants With Treatment-emergent Adverse Events | Treatment-related adverse events (TRAEs) | 51 participants |
Secondary
Objective Response Rate
Objective response rate was defined as the percentage of participants with a complete response (CR) or partial response (PR) based on modified RECIST 1.1 achieved over the study duration. CR: Disappearance of all target and non target lesions, normalization of tumor marker levels and no new lesions. PR: At least a 30% decrease in the size of target lesions with no progression of non-target lesions and no new lesions, or, disappearance of target lesions with persistence of one or more non-target lesions and/or maintenance of tumor marker levels above normal limits and no new lesions. Participants who underwent surgical resection while on study were not evaluated for response after the surgery. Participants who did not meet the criteria for an objective response by the analysis cutoff date were considered non-responders.
Time frame: From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months.
Population: Randomized participants with at least one baseline measurable lesion per modified RECIST 1.1.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Placebo | Objective Response Rate | 43.4 percentage of participants |
| Denosumab | Objective Response Rate | 36.8 percentage of participants |
p-value: 0.349195% CI: [0.43, 1.35]Regression, Logistic
Secondary
Progression-free Survival (PFS)
Progression-free survival was defined as the time from randomization to the first observed disease progression per modified RECIST 1.1 criteria or death from any cause. Participants last known to be alive who did not experience disease progression were censored at their last imaging assessment date, last contact date if they were in the survival follow up phase, end of the study date, or the primary analysis cut-off date, whichever was first. If a participant underwent surgical resection while on study, the participant was censored at the last evaluable imaging assessment prior to the surgery.
Time frame: From randomization until the data cut-off date of 29 July 2016; median time on study was 9.64 months.
Population: All randomized participants
| Arm | Measure | Value (MEDIAN) |
|---|---|---|
| Placebo | Progression-free Survival (PFS) | 5.7 months |
| Denosumab | Progression-free Survival (PFS) | 5.2 months |
p-value: 0.736395% CI: [0.78, 1.43]Log Rank
Secondary
Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing
Serum samples were analyzed for denosumab using enzyme-linked immunosorbent assay (ELISA) following a validated procedure. The lower limit of quantification for the assay was 20 ng/mL.
Time frame: Prior to dosing at day 8 and weeks 3, 6, 9, 12, 15, 18, 21 and 24.
Population: Randomized participants who received denosumab every 3 weeks with at least one valid denosumab concentration measurement.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Dose 3 - Week 3 | 12200 ng/mL | Standard Deviation 5710 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Dose 4 - Week 6 | 19700 ng/mL | Standard Deviation 7350 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Dose 9 - Week 21 | 22100 ng/mL | Standard Deviation 11200 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Dose 10 - Week 24 | 23300 ng/mL | Standard Deviation 12700 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Dose 2 - Day 8 | 8590 ng/mL | Standard Deviation 4080 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Dose 5 - Week 9 | 19600 ng/mL | Standard Deviation 8270 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Dose 6 - Week 12 | 22800 ng/mL | Standard Deviation 10400 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Dose 7 - Week 15 | 24300 ng/mL | Standard Deviation 11500 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q3W Dosing | Dose 8 - Week 18 | 22700 ng/mL | Standard Deviation 10600 |
Secondary
Serum Denosumab Trough Levels in Participants Who Received Q4W Dosing
Serum samples were analyzed for denosumab using enzyme-linked immunosorbent assay (ELISA) following a validated procedure. The lower limit of quantification for the assay was 20 ng/mL.
Time frame: Prior to dosing at day 8 and weeks 4, 8, 12, 16, 20 and 24
Population: Randomized participants who received denosumab every 4 weeks with at least one valid denosumab concentration measurement.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q4W Dosing | Dose 8 - Week 24 | 15900 ng/mL | Standard Deviation 7810 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q4W Dosing | Dose 2 - Day 8 | 8990 ng/mL | Standard Deviation 2470 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q4W Dosing | Dose 3 - Week 4 | 10900 ng/mL | Standard Deviation 3170 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q4W Dosing | Dose 4 - Week 8 | 15700 ng/mL | Standard Deviation 6710 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q4W Dosing | Dose 5 - Week 12 | 14500 ng/mL | Standard Deviation 4940 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q4W Dosing | Dose 6 - Week 16 | 13000 ng/mL | Standard Deviation 4380 |
| Placebo | Serum Denosumab Trough Levels in Participants Who Received Q4W Dosing | Dose 7 - Week 20 | 15400 ng/mL | Standard Deviation 5620 |