Table 3 Summary of the clinical effectiveness papers by endpoint type
Primary or co-primary (n=45) | Secondary (n=48) | Exploratory (n=5) | Total (n=98) | |
|---|---|---|---|---|
Was a Kaplan-Meier curve for OS presented in the main text? | ||||
Yes | 40 (88.9%) | 41 (85.4%) | 5 (100.0%) | 86 (87.8%) |
No | 2 (4.4%) | 6 (12.5%) | 0 (0.0%) | 8 (8.2%) |
N/Aa | 3 (6.7%) | 1 (2.1%) | 0 (0.0%) | 4 (4.1%) |
Were the numbers at risk included in the Kaplan-Meier curve? | ||||
Yes | 36 (80.0%) | 38 (79.2%) | 5 (100.0%) | 79 (80.6%) |
No | 4 (8.9%) | 3 (6.3%) | 0 (0.0%) | 7 (7.1%) |
N/A | 5 (11.1%) | 7 (14.6%) | 0 (0.0%) | 12 (12.2%) |
Were the number of patients censored or number of events included in the at risk table | ||||
Yes | 9 (20.0%) | 3 (6.3%) | 1 (20.0%) | 13 (13.3%) |
No | 27 (60.0%) | 35 (72.9%) | 4 (80.0%) | 66 (67.3%) |
N/A | 9 (20.0%) | 10 (20.8%) | 0 (0.0%) | 19 (19.4%) |
Were confidence intervals included in the Kaplan-Meier curve? | ||||
No | 40 (88.9%) | 41 (85.4%) | 5 (100.0%) | 86 (87.8%) |
N/A | 5 (11.1%) | 7 (14.6%) | 0 (0.0%) | 12 (12.2%) |
Primary analysis | ||||
Hazard ratio and 95% CI from an unadjusted Cox model | 1 (2.2%) | 0 (0.0%) | 0 (0.0%) | 1 (1.0%) |
Hazard ratio and 95% CI from an adjusted Cox model | 1 (2.2%) | 1 (2.1%) | 0 (0.0%) | 2 (2.0%) |
Hazard ratio and P-value from a logrank test | 8 (17.8%) | 11 (22.9%) | 0 (0.0%) | 19 (19.4%) |
Hazard ratio and P-value from a stratified logrank test. | 3 (6.7%) | 3 (6.3%) | 0 (0.0%) | 6 (6.1%) |
Hazard ratio and p-value from an adjusted Cox model | 12 (26.7%) | 2 (4.2%) | 1 (20.0%) | 15 (15.3%) |
Hazard ratio and p-value from an unadjusted Cox model | 2 (4.4%) | 3 (6.3%) | 0 (0.0%) | 5 (5.1%) |
Hazard ratio from a Cox model and p-value from a logrank test | 3 (6.7%) | 7 (14.6%) | 0 (0.0%) | 10 (10.2%) |
Hazard ratio from an adjusted Cox model and p-value from a logrank test. | 2 (4.4%) | 4 (8.3%) | 0 (0.0%) | 6 (6.1%) |
Hazard ratio from an adjusted Cox model and p-value from a stratified logrank test. | 5 (11.1%) | 5 (10.4%) | 3 (60.0%) | 13 (13.3%) |
Hazard ratio from an unadjusted Cox model and p-value from a logrank test. | 1 (2.2%) | 4 (8.3%) | 0 (0.0%) | 5 (5.1%) |
Hazard ratio from an unadjusted Cox model and p-value from a stratified logrank test. | 0 (0.0%) | 2 (4.2%) | 0 (0.0%) | 2 (2.0%) |
N/Aa | 3 (6.7%) | 1 (2.1%) | 0 (0.0%) | 4 (4.1%) |
Other | 4 (8.9%) | 5 (10.4%) | 1 (20.0%) | 10 (10.2%) |
Were subsequent treatments mentioned in the paper? | ||||
Yes | 33 (73.3%) | 23 (47.9%) | 3 (60.0%) | 59 (60.2%) |
No | 12 (26.7%) | 25 (52.1%) | 2 (40.0%) | 39 (39.8%) |
Was the percentage or number of participants who received later lines included in the main text or appendix? | ||||
Yes | 31 (68.9%) | 18 (37.5%) | 1 (20.0%) | 50 (51.0%) |
No | 1 (2.2%) | 5 (10.4%) | 2 (40.0%) | 8 (8.2%) |
N/A | 13 (28.9%) | 25 (52.1%) | 2 (40.0%) | 40 (40.8%) |
Was a breakdown or summary of the number of subsequent treatments (two, three, four) included in the main text or appendix? | ||||
Yes | 3 (6.7%) | 1 (2.1%) | 0 (0.0%) | 4 (4.1%) |
No | 28 (62.2%) | 22 (45.8%) | 3 (60.0%) | 53 (54.1%) |
N/Ab | 14 (31.1%) | 25 (52.1%) | 2 (40.0%) | 41 (41.8%) |
Was a breakdown or summary of the type of subsequent treatments (treatment 1, treatment 2, treatment 3) included in the main text or appendix? | ||||
Yes | 26 (57.8%) | 18 (37.5%) | 1 (20.0%) | 45 (45.9%) |
No | 5 (11.1%) | 5 (10.4%) | 2 (40.0%) | 12 (12.2%) |
N/Ab | 14 (31.1%) | 25 (52.1%) | 2 (40.0%) | 41 (41.8%) |
Was additional analysis conducted to account for subsequent treatment lines? | ||||
Yes | 2 (4.4%) | 3 (6.3%) | 2 (40.0%) | 7 (7.1%) |
No | 30 (66.7%) | 20 (41.7%) | 1 (20.0%) | 51 (52.0%) |
N/Ab | 14 (31.1%) | 25 (52.1%) | 2 (40.0%) | 41 (41.8%) |
Were subsequent treatment lines mentioned in the discussion? | ||||
Yes | 22 (48.9%) | 15 (31.3%) | 2 (40.0%) | 39 (39.8%) |
No | 11 (24.4%) | 8 (16.7%) | 1 (20.0%) | 20 (20.4%) |
N/A | 12 (26.7%) | 25 (52.1%) | 2 (40.0%) | 39 (39.8%) |
How were subsequent treatments mentioned in the discussion? | ||||
The uptake of subsequent treatments is given as a reason for the discrepancy between OS and surrogate endpoints. | 0 (0.0%) | 2 (4.2%) | 0 (0.0%) | 2 (2.0%) |
The uptake of subsequent treatments is given as a reason for better OS results or a reduced event rate than expected. | 4 (8.9%) | 1 (2.1%) | 1 (20.0%) | 6 (6.1%) |
The uptake of subsequent treatments is given as a reason for reduced OS effect/stated as may have affected the results/used to caveat the results. | 6 (13.3%) | 7 (14.6%) | 1 (20.0%) | 14 (14.3%) |
The lack of uptake of subsequent treatments is given as a reason for no OS effect/may have negatively affected OS. | 4 (8.9%) | 0 (0.0%) | 0 (0.0%) | 4 (4.1%) |
Subsequent treatments are stated to not have affected the OS results as an OS benefit was observed or OS was similar between those who did and did not receive a subsequent treatment. | 2 (4.4%) | 1 (2.1%) | 0 (0.0%) | 3 (3.1%) |
Randomisation/balance of subsequent treatment lines is given as a reason as to why subsequent treatment lines will not have affected the OS results. | 2 (4.4%) | 2 (4.2%) | 0 (0.0%) | 4 (4.1%) |
None or limited options of subsequent treatment lines for patients are given as a reason as to why subsequent treatment lines will not have affected the OS results. | 0 (0.0%) | 2 (4.2%) | 0 (0.0%) | 2 (2.0%) |
Other | 4 (8.9%) | 0 (0.0%) | 0 (0.0%) | 4 (4.1%) |
N/A | 23 (51.1%) | 33 (68.8%) | 3 (60.0%) | 59 (60.2%) |