KM Research Hub
← 返回论文库
已完成 入排:排除 图源:可得 入排置信度 0.85

Assessing Algorithmic Fairness With a Multimodal Artificial Intelligence Model in Men of African and Non-African Origin on NRG Oncology Prostate Cancer Phase III Trials.

Mack Roach, Jingbin Zhang, Osama Mohamad, Douwe van der Wal, Jeffry P Simko, Sandy DeVries, Huei-Chung Huang, Songwan Joun, Edward M Schaeffer, Todd M Morgan, Jessica Keim-Malpass, Emmalyn Chen, Rikiya Yamashita, Jedidiah M Monson, Farah Naz, James Wallace, Jean-Paul Bahary, Derek Wilke, Sonny Batra, Gregory B Biedermann, Sergio Faria, Lindsay Hwang, Howard M Sandler, Daniel E Spratt, Stephanie L Pugh, Andre Esteva, Phuoc T Tran, Felix Y Feng

JCO Clin Cancer Inform 9 e2400284

PMID40344545
PMCIDPMC12335010
DOI10.1200/CCI-24-00284
发表日期2025-05-09
入排依据:The record describes a prognostic analysis of an AI algorithm using data from Phase III randomized trials, but does not evaluate a drug intervention. It is an AI/diagnostic/prognostic study, not a drug efficacy trial.
摘要

摘要

PURPOSE: Artificial intelligence (AI) tools could improve clinical decision making or exacerbate inequities because of bias. African American (AA) men reportedly have a worse prognosis for prostate cancer (PCa) and are underrepresented in the development genomic biomarkers. We assess the generalizability of tools developed using a multimodal AI (MMAI) deep learning system using digital histopathology and clinical data from NRG/Radiation Therapy Oncology Group PCa trials across racial subgroups. METHODS: In total, 5,708 patients from five randomized phase III trials were included. Two MMAI algorithms were evaluated: (1) the distant metastasis (DM) MMAI model optimized to predict risk of DM, and (2) the PCa-specific mortality (PCSM) MMAI model optimized to focus on prediction death in the presence of DM (DDM). The prognostic performance of the MMAI algorithms was evaluated in AA and non-AA subgroups using time to DM (primary end point) and time to DDM (secondary end point). Exploratory end points included time to biochemical failure and overall survival with Fine-Gray or Cox proportional hazards models. Cumulative incidence estimates were computed for time-to-event end points and compared using Gray's test. RESULTS: There were 948 (16.6%) AA patients, 4,731 non-AA patients (82.9%), and 29 (0.5%) patients with unknown or missing race status. The DM-MMAI algorithm showed a strong prognostic signal for DM in the AA (subdistribution hazard ratio [sHR], 1.2 [95% CI, 1.0 to 1.3]; P = .007) and non-AA subgroups (sHR, 1.4 [95% CI, 1.3 to 1.5]; P < .001). Similarly, the PCSM-MMAI score showed a strong prognostic signal for DDM in both AA (sHR, 1.3 [95% CI, 1.1 to 1.5]; P = .001) and non-AA subgroups (sHR, 1.5 [95% CI, 1.4 to 1.6]; P < .001), with similar distributions of risk. CONCLUSION: Using cooperative group data sets with a racially diverse population, the MMAI algorithm performed well across racial subgroups without evidence of algorithmic bias.

KM 提取结果

2 个 Figure · 6 个 KM 面板

进入结果中心
KM Figure2 个面板

Figure 4a

Figure 4a and 4b: Estimated risks/cumulative incidence curves by racial subgroups for a) DM and b) DDM endpoint in the full cohort

A

DM

Figure 4a · Panel A

quality_warning 待 EXE 对齐 预期 2 条 已提取 1 条

Panel A 是同一 Figure 4a 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

质量提示疑似重复轨迹:同一条 KM 曲线可能被重复识别疑似图例或人工水平线:轨迹出现不合理的共同平台曲线缺失:预期 2 条,实际恢复 1 条疑似重复轨迹:同一条 KM 曲线可能被重复识别
查看结果
B

DDM

Figure 4a · Panel B

quality_warning 待 EXE 对齐 预期 2 条 已提取 1 条

Panel B 是同一 Figure 4a 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

质量提示疑似重复轨迹:同一条 KM 曲线可能被重复识别疑似图例或人工水平线:轨迹出现不合理的共同平台曲线缺失:预期 2 条,实际恢复 1 条疑似重复轨迹:同一条 KM 曲线可能被重复识别
查看结果
KM Figure4 个面板

Figures 5a and 5b:

Figures 5a and 5b: Risk stratifications of MMAI algorithms within racial subgroups a) DM-MMAI for DM endpoint and b) PCSM-MMAI for DDM endpoint in the development and test cohorts

DM-MMAI_for_DM_endpoint_-_Development

DM

Figures 5a and 5b: · Panel DM-MMAI_for_DM_endpoint_-_Development

quality_warning 待 EXE 对齐 预期 6 条 已提取 1 条

Panel DM-MMAI_for_DM_endpoint_-_Development 是同一 Figures 5a and 5b: 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

质量提示疑似重复轨迹:同一条 KM 曲线可能被重复识别疑似图例或人工水平线:轨迹出现不合理的共同平台曲线缺失:预期 6 条,实际恢复 1 条疑似重复轨迹:同一条 KM 曲线可能被重复识别
查看结果
DM-MMAI_for_DM_endpoint_-_Test

DM

Figures 5a and 5b: · Panel DM-MMAI_for_DM_endpoint_-_Test

quality_warning 待 EXE 对齐 预期 6 条 已提取 1 条

Panel DM-MMAI_for_DM_endpoint_-_Test 是同一 Figures 5a and 5b: 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

质量提示疑似重复轨迹:同一条 KM 曲线可能被重复识别疑似图例或人工水平线:轨迹出现不合理的共同平台曲线缺失:预期 6 条,实际恢复 1 条疑似重复轨迹:同一条 KM 曲线可能被重复识别
查看结果
PCSM-MMAI_for_DDM_endpoint_-_Development

DDM

Figures 5a and 5b: · Panel PCSM-MMAI_for_DDM_endpoint_-_Development

quality_warning 待 EXE 对齐 预期 6 条 已提取 2 条

Panel PCSM-MMAI_for_DDM_endpoint_-_Development 是同一 Figures 5a and 5b: 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

质量提示疑似重复轨迹:同一条 KM 曲线可能被重复识别疑似图例或人工水平线:轨迹出现不合理的共同平台曲线缺失:预期 6 条,实际恢复 2 条疑似重复轨迹:同一条 KM 曲线可能被重复识别
查看结果
PCSM-MMAI_for_DDM_endpoint_-_Test

DDM

Figures 5a and 5b: · Panel PCSM-MMAI_for_DDM_endpoint_-_Test

quality_warning 待 EXE 对齐 预期 6 条 已提取 2 条

Panel PCSM-MMAI_for_DDM_endpoint_-_Test 是同一 Figures 5a and 5b: 中的独立 KM 子图;人群、终点或分析条件以本 Panel 图注语义为准。

质量提示疑似重复轨迹:同一条 KM 曲线可能被重复识别疑似图例或人工水平线:轨迹出现不合理的共同平台曲线缺失:预期 6 条,实际恢复 2 条疑似重复轨迹:同一条 KM 曲线可能被重复识别
查看结果
其他 Figure(3,默认折叠)

这些图片未产生 KM 曲线结果,不在主页面重复展示。

Figure 1Figure 1: Flow diagram representing clinical trial pooling for algorithm development and testing
非 KM Figure
Figure 2Figure 2: Distributions of MMAI scores between racial subgroups in the development and test cohorts
非 KM Figure
Figure 3aFigure 3a and 3b: Univariable model results in racial subgroups for a) DM-MMAI and b) PCSM-MMAI in the development and test cohorts
非 KM Figure