Interval cancer, sensitivity, and specificity comparing AI-supported mammography screening with standard double reading without AI in the MASAI study: a randomised, controlled, non-inferiority, single-blinded, population-based, screening-accuracy trial.

We aimed to compare the interval cancer rate in AI-supported mammography screening with standard double reading without AI.

我们的目标是比较人工智能支持的乳腺X线摄影筛查与没有人工智能的标准双重阅读的间期癌症发生率。

Methods

In this Swedish randomised, controlled, non-inferiority, single-blinded, population-based screening accuracy trial, participants were allocated in a 1:1 ratio to either AI-supported mammography screening (the intervention group) or standard double reading without AI (the control group).

在这项瑞典的随机、对照、非劣效、单盲、基于人群的筛查准确性试验中，参与者以1:1的比例被分配到人工智能支持的乳腺X线摄影筛查（干预组）或没有人工智能的标准双读（对照组）。

AI was used to triage examinations to single or double reading by radiologists and for detection support.

人工智能被用于将检查分为单读或双读，并为放射科医生提供检测支持。

This is a protocol-defined analysis of the primary outcome, interval cancer rate, with a 20% non-inferiority margin.

这是对主要结果——间隔癌发生率的协议定义分析，具有20%的非劣效性边界。

Secondary outcomes reported in this analysis are interval cancer characteristics, sensitivity, specificity, and sensitivity by age, breast density, and cancer type (in-situ and invasive).

本分析报告的次要结果包括间隔癌特征、灵敏度、特异性以及按年龄、乳腺密度和癌症类型（原位癌和浸润性癌）的灵敏度。

Other secondary outcomes from the trial that have been previously reported are referenced in the Methods section of this Article.

先前报告的试验其他次要结果在本文的方法部分有引用。

The trial is registered with ClinicalTrials.gov (NCT04838756) and is complete.

该试验已在ClinicalTrials.gov注册（NCT04838756），且已完成。

Results

Between April 12, 2021, and Dec 7, 2022, 105 934 women were randomly assigned to the intervention or control group, of whom 19 were excluded from the analysis.

2021年4月12日至2022年12月7日，共有105,934名女性被随机分配到干预组或对照组，其中19人被排除在分析之外。

Median age was 53·8 years (IQR 46·5-63·3) in the intervention group and 53·7 years (46·5-63·2) in the control group.

干预组的中位年龄为53.8岁（四分位数间距46.5-63.3岁），对照组的中位年龄为53.7岁（四分位数间距46.5-63.2岁）。

Interval cancer rates were 1·55 (95% CI 1·23-1·92) and 1·76 (1·42-2·15) per 1000 participants in the intervention and control group respectively, a non-inferior proportion ratio of 0·88 (95% CI 0·65-1·18; p=0·41).

干预组和对照组的间隔癌发生率分别为每1000名参与者1.55 (95% 置信区间 1.23-1.92) 和 1.76 (1.42-2.15)，非劣效比例比为0.88 (95% 置信区间 0.65-1.18; p=0.41)。

Descriptively, the intervention group had fewer interval cancers that were invasive (75 vs 89), T2+ (38 vs 48), or non-luminal A (43 vs 59) than the control group.

描述性地，干预组的侵袭性间隔癌(75 vs 89)、T2+ (38 vs 48)或非luminal A型(43 vs 59)的数量少于对照组。

Sensitivity was higher in the intervention group (80·5% [95% CI 76·4-84·2]) than the control group (73·8% [68·9-78·3]; p=0·031), an effect consistent across age and breast density, and for invasive cancer but not for in-situ cancer.

干预组的敏感性更高（80.5% [95% CI 76.4-84.2]）相较于对照组（73.8% [68.9-78.3]；p=0.031），这一效果在不同年龄和乳腺密度中保持一致，对于侵袭性癌症也是如此，但对于原位癌则不适用。

Specificity was 98·5% (95% CI 98·4-98·6) for both groups (p=0·88).

两组的特异性均为98.5%（95% CI 98.4-98.6）（p=0.88）。

interpretation

AI-supported mammography screening showed consistently favourable outcomes compared with standard double reading, with a non-inferior interval cancer rate, fewer interval cancers with unfavourable characteristics, higher sensitivity, and the same specificity, while also reducing screen reading workload.

AI支持的乳腺X线筛查与标准双读相比显示出一致的有利结果，间隔癌的发病率不逊色，具有不利特征的间隔癌更少，灵敏度更高，特异性相同，同时减少了筛查读片的工作量。

These findings imply that AI-supported mammography screening can efficiently improve screening performance compared with standard double reading and may be considered for implementation in clinical practice.

这些发现意味着AI支持的乳腺X线筛查可以有效地提高筛查性能，与标准双读相比，并且可以考虑在临床实践中实施。

funding

Swedish Cancer Society, Confederation of Regional Cancer Centres, Swedish governmental funding for clinical research.

瑞典癌症学会、瑞典区域癌症中心联盟、瑞典政府对临床研究的资助。