Disparities in Phase 3 Leukemia Trials from the Last Decade: A Long Way to Go
Bibliographic record
Abstract
Background Randomized clinical trials (RCTs) are crucial for medical advancements but often need more participant diversity, distorting our understanding of whether such advancements are effective across all populations. The lack of diverse representation of participants in RCTs can also perpetuate disparities in care. This study examines disparities in leukemia trials, underscoring the need to address equity in RCT representation. Methods: We analyzed phase 3 acute and chronic leukemia RCTs listed on ClinicalTrials.gov from January 2012 to March 2023. We selected these trials due to their potential to offer insights into the effectiveness and adverse effects of treatments. Our primary objective was to delineate disparities in trial demographics and characteristics, with a distinct focus on participant recruitment regions in relation to global leukemia incidence and prevalence. Two independent team members used a standardized form to extract study and participant demographic characteristics, treatment type, and outcomes data and classified participant countries based on the World Bank Ranking (WBR). The recruitment countries and participant numbers were gathered from the clinicaltrials.gov registry and further classified by the WBR into HIC-exclusive and LMIC-inclusive. Initially, we performed a descriptive analysis of all trials to delve into the characteristics of the RCTs. To measure completion rates, we divided the number of participants who finished by the number of participants who started. The second phase of our analysis was a more in-depth evaluation of published trials. We cross-referenced the registry for the corresponding publication. If the research paper was not found after an online search by two authors, it was excluded from the refined analysis. We collected information on the primary author (PA) and the number of authors affiliated with HIC and LMIC institutions. Additionally, we extracted the journal impact factor (JIF) from the year the paper was published. Results: Our analysis of 72 trials and 27,382 participants revealed an overrepresentation of participants from HICs (90.4%) compared to LMICs (9.6%). Most of the studies were drug-based (69.4%) and open-label (83.3%), focusing mainly on acute myelogenous leukemia (AML) and chronic lymphocytic leukemia (CLL). More males than females participated, with a median ratio of 1.44. Completion rates varied widely from 0% to 100% (median: 69%). A significant portion of these trials (61.1%) were industry-funded, and approximately 70% were published. Interestingly, some trials (12.5%) were non-randomized, despite being registered as phase 3. Of the 72 trials, 35 were exclusively from HICs, favoring AML, while LMIC-inclusive trials were more leaned towards CLL. The median number of LMICs per trial was considerably low (3 countries) compared to HICs (14 countries). The median number of total participants was lower in HIC-exclusive (275) compared to LMIC-inclusive (365) trials. This applied to both male and female patients: HIC-exclusive: 115 females, 144 males; LMIC-inclusive 127 females, 215 males. Only 3 trials exclusively involved LMIC participants, all of which were from China and focused on CML. These trials were non-randomized, industry-funded, and drug-focused. Two of these trials reported positive results, with one discussing global implementation, but none reported quality-of-life (QoL) outcomes. Of the 49 studies with published research, only a few studies (28%) discussed global implementation and considered QoL endpoints (8%). LMIC-inclusive trials reported more outcomes with positive connotations (60%) and frequently employed medical writers (93%). None of the LMIC-inclusive trials were investigator-initiated. The US was the predominant FA country. China was the only LMIC with a local FA country (n=3). Most authors came from HIC institutions. We found significant differences between HICs and LMICs, and moderate negative correlations between the number of LMIC authors and JIF. Conclusion: Our findings highlight a pressing need to improve LMIC representation in leukemia RCTs. Aligning trial demographics with global leukemia prevalence is vital since more than 80% of the global leukemia population resides in LMICs. Inclusive and diverse trial demographics are essential not just for statistical balance but for addressing enduring health disparities in underrepresented regions.
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 | 0.604 | 0.738 |
| Meta-epidemiology (narrow) | 0.001 | 0.002 |
| Meta-epidemiology (broad) | 0.007 | 0.006 |
| Bibliometrics | 0.014 | 0.019 |
| Science and technology studies | 0.004 | 0.014 |
| Scholarly communication | 0.019 | 0.022 |
| Open science | 0.006 | 0.015 |
| Research integrity | 0.008 | 0.013 |
| Insufficient payload (model declined to judge) | 0.009 | 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".