Department of Urology, University of Minnesota , Minneapolis , Minnesota.
School of Public Health, University of Minnesota , Minneapolis , Minnesota.
J Urol. 2019 Aug;202(2):385-391. doi: 10.1097/JU.0000000000000294. Epub 2019 Jul 8.
We performed a multiregistry analysis to assess relative differences in accrual sufficiency and race/ethnicity reporting in trials of common urological cancers and other nonurological solid organ tumors.
We queried ClinicalTrials.gov and the ISRCTN (International Standard Randomised Controlled Trial Number) Registry for closed phase III and IV trials focused on prostate, colorectal, kidney, bladder, testicular, breast and lung cancer. Identified trials were cross-verified with appropriate published data sources. Comparative accrual sufficiency and rates of race/ethnicity reporting were calculated. Multivariable logistic regression analysis was performed to determine factors associated with accrual status and race/ethnicity reporting.
A total of 326 trials were identified based on our prespecified criteria, of which 63% reported sufficient accrual by time of closure and 58% reported data by race/ethnicity. Nonurological trials were significantly more likely to mention race data than urological trials (OR 3.25, 95% CI 1.24-8.55, p = 0.02). Industry sponsored trials were more likely to meet accrual targets than government funded projects (OR 5.44, 95% CI 1.64-18.20, p = 0.001). Although funding source did not influence race reporting, the reported recruitment of participants of African ethnicity was lower in industry sponsored trials (11.49% vs 3.18%, p <0.01). Two-thirds of the studies did not report baseline characteristics by African American race/ethnicity.
Insufficient accrual and inadequate race/ethnicity reporting are prevalent issues, limiting interpretation of the results of clinical trials of major solid organ malignancies. Addressing these shortcomings would enhance result validity by raising statistical power and improving the transparency of reporting to better evaluate the generalizability of results.
我们进行了一项多注册分析，以评估常见泌尿系统癌症和其他非泌尿系统实体肿瘤临床试验中累积充足性和种族/民族报告的相对差异。
我们在 ClinicalTrials.gov 和 ISRCTN（国际标准随机对照试验编号）注册中心查询了针对前列腺癌、结直肠癌、肾癌、膀胱癌、睾丸癌、乳腺癌和肺癌的已完成的 III 期和 IV 期试验。确定的试验与适当的已发表数据源进行了交叉验证。计算了累积充足性和种族/民族报告率的比较。进行多变量逻辑回归分析，以确定与累积状态和种族/民族报告相关的因素。
根据我们预先设定的标准，共确定了 326 项试验，其中 63%的试验在关闭时报告了充足的累积量，58%的试验报告了种族/民族数据。非泌尿系统试验比泌尿系统试验更有可能提到种族数据（比值比 3.25，95%置信区间 1.24-8.55，p = 0.02）。与政府资助项目相比，工业赞助的试验更有可能达到累积目标（比值比 5.44，95%置信区间 1.64-18.20，p = 0.001）。尽管资金来源不影响种族报告，但工业赞助的试验中招募非洲裔参与者的比例较低（11.49%比 3.18%，p <0.01）。三分之二的研究没有按非裔美国人种族/民族报告基线特征。
累积不足和种族/民族报告不足是普遍存在的问题，限制了对主要实体恶性肿瘤临床试验结果的解释。解决这些缺陷将通过提高统计效力和提高报告的透明度来增强结果的有效性，从而更好地评估结果的普遍性。
