Success in Meeting the Primary Endpoint in Phase III Trials: A Comparison of Industry and Cooperative Group Trials
Bibliographic record
Abstract
Abstract Abstract 508 Background: Evaluation of research effort, especially estimation of the proportion of treatment successes in randomized clinical trials (RCTs), has important ethical, scientific, and public policy implications. Whether commercial or public sector research programs generate higher discovery rate of new successful treatments when tested in cancer RCTs is not known. These research programs are postulated to be governed by two competing hypotheses. The “equipoise/uncertainty hypothesis” assumes that investigators cannot predict trial results in advance, and as a consequence, the rate of discovering new treatments is about 50%. In contrast, the “design bias hypothesis” assumes that researchers conduct only those RCTs which have high likelihood of success. We hypothesize that the public sector RCTs are governed by the equipoise hypothesis while the industry-sponsored (IS) RCTs are based on the design bias hypothesis. Here we conduct the comparative systematic assessment to investigate if IS RCTs are associated with higher success rates than publicly-sponsored trials (PS) according to design bias versus equipoise/uncertainty hypothesis, respectively. Methods: All consecutive, published and unpublished, phase III cancer RCTs assessing treatment superiority and conducted by Canada's NCIC Clinical Trials Group (NCIC CTG) and GlaxoSmithKline (GSK) from 1980 to June 2010 were included. All trial protocols from GSK and NCIC CTG were reviewed independently by two reviewers to determine their eligibility. Two reviewers independently extracted data from eligible study protocols and publications using a standardized form. Three metrics were extracted to determine treatment successes: (1) the proportion of statistically significant trials favoring new or standard treatments, (2) the proportion of the trials in which new treatments were considered superior according to the original investigators, and (3) quantitative synthesis of data for primary outcomes as defined in each trial. An experimental regimen (drug compound or combinations or procedures), which was not tested previously in an RCT involving a specific cancer population or for alleviation of symptoms was classified as a major innovation. If a drug or regimen was already tested in a specific cancer population and testing involved dose modifications or changes in route of administration, it was classified as a minor innovation. Results: Between1980 to 2010 NCIC CTG conducted 77 RCTs enrolling 33,260 patients while GSK conducted 40 cancer RCTs accruing 19,889 patients. Forty two percent (99%CI 24 to 60) of the results were statistically significant favoring experimental treatments in GSK versus 25% (99%CI 13 to 37) in the NCIC CTG cohort (p=0.04). Investigators concluded that new treatments were superior to standard treatments in 80% of GSK versus 44% of NCIC CTG RCTs (p<0.0001) The GSK investigators deemed 32% (99%CI 14 to 50; 14/44) of interventions as “breakthroughs” versus 10% (99%CI 1 to 18; 8/82) by NCIC CTG investigators (p=0.002). Pooled analysis for the primary outcome indicated higher success rate in GSK trials (odds ratio: 0.61 [99%CI 0.47–0.78]) versus NCIC trials (odds ratio: 0.86 [99%CI 0.74–1.00]) (p=0.003). Experimental treatments were considered as major innovations in 32% (99%CI 15 to 49; 16/50) of GSK vs. 93% (99%CI 86 to 100; 78/84) of NCIC CTG trials (p<0.0001). Increased success rate in IS RCTs was mainly due to testing of new palliative agents, while the research program of NCIC CTG largely focused on development of therapies to improve survival. Conclusions: This first study evaluating the treatment success and pattern of therapeutic discoveries in IS versus PS research showed that industry discovers more successful new treatments compared with public sector. However, industry appears to undertake RCTs with high likelihood of success. PS research had significantly high proportion of major innovations compared with IS research. Disclosures: No relevant conflicts of interest to declare.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.572 | 0.806 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.010 | 0.021 |
| Bibliometrics | 0.027 | 0.027 |
| Science and technology studies | 0.001 | 0.007 |
| Scholarly communication | 0.009 | 0.008 |
| Open science | 0.004 | 0.007 |
| Research integrity | 0.005 | 0.003 |
| Insufficient payload (model declined to judge) | 0.007 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".