结直肠息肉预防随机对照试验中的人口学代表性：系统评价和荟萃分析

Background: Colorectal cancer (CRC) is an increasing global health burden. With incidence rising, there is a need to develop effective preventive strategies. Polyp-prevention randomized controlled trials (RCTs) are central to testing CRC-preventive therapies, using colorectal polyp development as an established endpoint. Marked ethnic disparities in CRC incidence, particularly in early-onset disease, are well documented. However, no systematic evaluation of demographic representation in polyp-prevention RCTs exists. Methods: We conducted a systematic review and meta-analysis of RCTs assessing chemopreventive or immunopreventive interventions for polyp recurrence. The protocol was registered in PROSPERO (CRD420251102449). Eligible trials enrolled adults with prior polyps while excluding those with prior CRC, inflammatory bowel disease, or hereditary cancer syndromes. Of 9,127 screened records, 27 RCTs met inclusion. Demographic data including age, sex, geography, race, ethnicity, and socioeconomic status (SES) were extracted. Multi-arm data were pooled. Geography was grouped as Western, Non-Western, or multinational (Western and Non-Western). Race and ethnicity were harmonized to OMB-1997 categories. Age was meta-analysed using inverse-variance random-effects models. Sex, race, and ethnicity proportions were analysed using logit-transformed random-effects models. Temporal trends were visualised with sample-size-weighted bubble plots. All analyses were performed using R Statistical Software (v4.1.2; R Core Team 2024). Results: All trials reported age, geography, and sex. Other demographics were often incomplete or collapsed. Race was reported in 41% of trials, and only 11% reported SES. The pooled mean age was 60.7 years (95% CI 59.2-62.2). Participants were 67.9% male. Studies were predominantly Western (74.07%), with Non-Western (18.52%) and multinational studies (7.41%) comprising the remainder. Of the 12 trials reporting race, pooled representation was White 87.9% (95% CI 82.5-91.8), Black 6.8% (95% CI 3.5-12.8), and Asian 2.2% (95% CI 1.1-4.1). Only 8 trials reported ethnicity, with Hispanic representation of 5.0% (95% CI 3.1-8.1). Bubble plots demonstrated persistently low non-White representation across three decades. Conclusion: This review highlights inconsistent demographic reporting in CRC polyp-prevention RCTs. The persistent demographic underrepresentation, particularly of Black and Asian participants, fails to reflect the disproportionate disease burden. This limits the generalisability of prevention strategies and may worsen existing disparities in CRC outcomes. There is an urgent need for standardised demographic reporting and intentional equity-focused recruitment strategies to ensure that therapeutic prevention approaches are effective and applicable across all populations.