Nemat Hoda, Orr Martin, Barrow Lucy, Raobaikady Bindu, Matharu Sheila, Scerri Lisa, Banerji Udai, Costelloe Ceire
Health Informatics, Division of Clinical Studies, Institute of Cancer Research, London, UK.
Drug Development Unit, The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, London, UK.
BJC Rep. 2025 Aug 25;3(1):57. doi: 10.1038/s44276-025-00165-y.
Phase I trials include patients with metastatic cancer and complex health conditions. Understanding baseline comorbidity and demographic features is critical to improving trial design.
We used electronic patient records to study the association of comorbidity, polypharmacy, and demographic factors on trial recruitment, time on trial, and health service utilisation.
A cohort of 1671 patients was considered for allocation to a phase I study, of whom 518 patients were recruited to a phase I study and 1153 patients were not. A multivariable analysis revealed polypharmacy was associated with lower recruitment to phase I trials with an odds ratio of 0.95 (95% CI: [0.92, 0.99], p = 0.01), and a greater number of emergency admissions with a risk ratio of 1.1 (95% CI: [1.03, 1.17], p = 0.01). Interestingly, comorbidity was not associated with lower recruitment but was associated with a lower time on trial with a hazard ratio of 0.75 (95% CI: [0.62, 0.90], p ≤ 0.001). Demographic factors, including ethnicity, distance of residence from the hospital, and index of multiple deprivation, did not significantly influence these parameters.
Polypharmacy and comorbidity should be considered both in the design of phase I oncology trials and in planning for healthcare utilisation during these trials.
I期试验纳入了患有转移性癌症和复杂健康状况的患者。了解基线合并症和人口统计学特征对于改进试验设计至关重要。
我们使用电子病历研究合并症、多种药物治疗以及人口统计学因素与试验招募、试验时间和医疗服务利用之间的关联。
共有1671名患者被考虑分配到一项I期研究中，其中518名患者被招募进入I期研究，1153名患者未被招募。多变量分析显示，多种药物治疗与I期试验较低的招募率相关，比值比为0.95（95%置信区间：[0.92, 0.99]，p = 0.01），并且与更多的急诊入院相关，风险比为1.1（95%置信区间：[1.03, 1.17]，p = 0.01）。有趣的是，合并症与较低的招募率无关，但与较短的试验时间相关，风险比为0.75（95%置信区间：[0.62, 0.90]，p≤0.001）。包括种族、居住地与医院的距离以及多重贫困指数在内的人口统计学因素并未显著影响这些参数。
在I期肿瘤试验设计以及这些试验期间的医疗利用规划中，都应考虑多种药物治疗和合并症。
