Eriksson, Mikael

Image-derived artificial intelligence (AI) risk models have shown promise in short-term risk assessment for improving breast cancer (BC) screening. No image-derived long-term AI risk model for primary prevention has been developed and externally validated. Individuals aged 31 to 94 years were recruited between 2010 and 2020 in two population-based case cohorts (Olmsted County, Minnesota, US, and KARMA, Sweden) and one hospital-based case-control study (EMBED, Atlanta, US). Median follow-up in the case cohorts was 10 years, with BCs diagnosed through June 2022. Additional validation was performed in EMBED with 3-year follow-up. An AI risk model was developed in Sweden, and we report independent validation in Olmsted/KARMA/EMBED. Absolute 10-year risks were estimated at study entry. Time-dependent discriminatory performance [AUC( t )] and expected-to-observed events (E/O) were estimated. Comparisons were performed with clinical risk tools and the image-based Mirai tool. Across Olmsted/KARMA, 8696 individuals (mean age, 54.4±10.6) and 1633 individuals with incident BC (mean age, 57.0±10.6) were included. Average 10-year risks were 3.83 and 3.14%, and E/O ratios were 0.99 and 0.99. Ten-year AUC( t ) for invasive BCs were 0.72 in Olmsted and 0.72 in KARMA. Similar discriminatory results were observed in EMBED. Our AI risk tool performed significantly better than Mirai in Olmsted/KARMA/EMBED. In KARMA, in the top 10% of high-risk individuals, the AI risk tool predicted 33% of BCs compared with 23%/20%/24% predicted by Tyrer-Cuzick-v8/BCSC-v3/Mirai (all P < 0.01). The 10-year image-derived AI risk model showed significantly higher performance than clinical risk models and Mirai in diverse populations, supporting its clinical potential for primary prevention.

Background: Breast cancer polygenic risk scores (PRS) and traditional risk models (e.g., the Gail model [Gail]) are known to contribute largely independent information, but it is unclear how the overlap varies by ancestry, age, disease type (invasive breast cancer, DCIS), and risk threshold. Methods: In a retrospective case–control study, we evaluated risk prediction performance in 180,398 women (161,849 of European ancestry; 18,549 of Asian ancestry). Odds ratios (ORs) from logistic regression models and the area under the receiver operating characteristic curve (AUC) were estimated. Results: PRS for invasive disease showed a stronger association in younger (<50 years) women (OR = 2.51, AUC = 0.622) than in women ≥ 50 years (OR = 2.06, AUC = 0.653) of European ancestry. PRS performance in Asians was lower (OR range = 1.62–1.64, AUC = 0.551–0.600). Gail performance was modest across groups and poor in younger Asian women (OR = 0.94–0.99, AUC = 0.523–0.533). Age interactions were observed for both PRS (p < 0.001) and Gail (p < 0.001) in Europeans, whereas in Asians, age interaction was observed only for Gail (invasive: p < 0.001; DCIS: p = 0.002). PRS identified more high-risk individuals than Gail in Asian populations, especially ≥50 years, while Gail identified more in Europeans. Overlap between PRS, Gail, and family history was limited at higher thresholds. Calibration analysis, comparing empirical and model-based ROC curves, showed divergence for both PRS and Gail (p < 0.001), which indicates miscalibration. In Europeans, family history and prior biopsies drove Gail discrimination. In younger Asians, age at first live birth was influential. Conclusions: PRS adds value to risk stratification beyond traditional tools, especially in younger women and Asian ancestry populations.

Background: Image-derived artificial intelligence (AI) risk models have shown promise in identifying high-risk women in the short term. The long-term performance of image-derived risk models expanded with clinical factors has not been investigated. Methods: We performed a case–cohort study of 8110 women aged 40–74 randomly selected from a Swedish mammography screening cohort initiated in 2010 together with 1661 incident BCs diagnosed before January 2022. The imaging-only AI risk model extracted mammographic features and age at screening. Additional lifestyle/familial risk factors were incorporated into the lifestyle/familial-expanded AI model. Absolute risks were calculated using the two models and the clinical Tyrer–Cuzick v8 model. Age-adjusted model performances were compared across the 10-year follow-up. Results: The AUCs of the lifestyle/familial-expanded AI risk model ranged from 0.75 (95%CI: 0.70–0.80) to 0.68 (95%CI: 0.66–0.69) 1–10 years after study entry. Corresponding AUCs were 0.72 (95%CI: 0.66–0.78) to 0.65 (95%CI: 0.63–0.66) for the imaging-only model and 0.62 (95%CI: 0.55–0.68) to 0.60 (95%CI: 0.58–0.61) for Tyrer–Cuzick v8. The increased performances were observed in multiple risk subgroups and cancer subtypes. Among the 5% of women at highest risk, the PPV was 5.8% using the lifestyle/familial-expanded model compared with 5.3% using the imaging-only model, p < 0.01, and 4.6% for Tyrer–Cuzick, p < 0.01. Conclusions: The lifestyle/familial-expanded AI risk model showed higher performance for both long-term and short-term risk assessment compared with imaging-only and Tyrer–Cuzick models.

PURPOSE Image-derived artificial intelligence–based short-term risk models for breast cancer have shown high discriminatory performance compared with traditional lifestyle/familial-based risk models. The long-term performance of image-derived risk models has not been investigated. METHODS We performed a case-cohort study of 8,604 randomly selected women within a mammography screening cohort initiated in 2010 in Sweden for women age 40-74 years. Mammograms, age, lifestyle, and familial risk factors were collected at study entry. In all, 2,028 incident breast cancers were identified through register matching in May 2022 (206 incident breast cancers were found in the subcohort). The image-based model extracted mammographic features (density, microcalcifications, masses, and left-right breast asymmetries of these features) and age from study entry mammograms. The Tyrer-Cuzick v8 risk model incorporates self-reported lifestyle and familial risk factors and mammographic density to estimate risk. Absolute risks were estimated, and age-adjusted AUC model performances (aAUCs) were compared across the 10-year period. RESULTS The aAUCs of the image-based risk model ranged from 0.74 (95% CI, 0.70 to 0.78) to 0.65 (95% CI, 0.63 to 0.66) for breast cancers developed 1-10 years after study entry; the corresponding Tyrer-Cuzick aAUCs were 0.62 (95% CI, 0.56 to 0.67) to 0.60 (95% CI, 0.58 to 0.61). For symptomatic cancers, the aAUCs for the image-based model were ≥0.75 during the first 3 years. Women with high and low mammographic density showed similar aAUCs. Throughout the 10-year follow-up, 20% of all women with breast cancers were deemed high-risk at study entry by the image-based risk model compared with 7.1% using the lifestyle familial-based model ( P < .01). CONCLUSION The image-based risk model outperformed the Tyrer-Cuzick v8 model for both short-term and long-term risk assessment and could be used to identify women who may benefit from supplemental screening and risk reduction strategies.

AbstractBackgroundReproductive factors have been shown to be differentially associated with risk of estrogen receptor (ER)-positive and ER-negative breast cancer. However, their associations with intrinsic-like subtypes are less clear.MethodsAnalyses included up to 23 353 cases and 71 072 controls pooled from 31 population-based case-control or cohort studies in the Breast Cancer Association Consortium across 16 countries on 4 continents. Polytomous logistic regression was used to estimate the association between reproductive factors and risk of breast cancer by intrinsic-like subtypes (luminal A-like, luminal B-like, luminal B-HER2–like, HER2-enriched–like, and triple-negative breast cancer) and by invasiveness. All statistical tests were 2-sided.ResultsCompared with nulliparous women, parous women had a lower risk of luminal A-like, luminal B-like, luminal B-HER2–like, and HER2-enriched–like disease. This association was apparent only after approximately 10 years since last birth and became stronger with increasing time (odds ratio [OR] = 0.59, 95% confidence interval [CI] = 0.49 to 0.71; and OR = 0.36, 95% CI = 0.28 to 0.46 for multiparous women with luminal A-like tumors 20 to less than 25 years after last birth and 45 to less than 50 years after last birth, respectively). In contrast, parous women had a higher risk of triple-negative breast cancer right after their last birth (for multiparous women: OR = 3.12, 95% CI = 2.02 to 4.83) that was attenuated with time but persisted for decades (OR = 1.03, 95% CI = 0.79 to 1.34, for multiparous women 25 to less than 30 years after last birth). Older age at first birth (Pheterogeneity &lt; .001 for triple-negative compared with luminal A-like breast cancer) and breastfeeding (Pheterogeneity &lt; .001 for triple-negative compared with luminal A-like breast cancer) were associated with lower risk of triple-negative breast cancer but not with other disease subtypes. Younger age at menarche was associated with higher risk of all subtypes; older age at menopause was associated with higher risk of luminal A-like but not triple-negative breast cancer. Associations for in situ tumors were similar to luminal A-like.ConclusionsThis large and comprehensive study demonstrates a distinct reproductive risk factor profile for triple-negative breast cancer compared with other subtypes, with implications for the understanding of disease etiology and risk prediction.

Abstract Background Rigorous evaluation of the calibration and discrimination of breast-cancer risk-prediction models in prospective cohorts is critical for applications under clinical guidelines. We comprehensively evaluated an integrated model incorporating classical risk factors and a 313-variant polygenic risk score (PRS) to predict breast-cancer risk. Methods Fifteen prospective cohorts from six countries with 239 340 women (7646 incident breast-cancer cases) of European ancestry aged 19–75 years were included. Calibration of 5-year risk was assessed by comparing expected and observed proportions of cases overall and within risk categories. Risk stratification for women of European ancestry aged 50–70 years in those countries was evaluated by the proportion of women and future cases crossing clinically relevant risk thresholds. Results Among women &lt;50 years old, the median (range) expected-to-observed ratio for the integrated model across 15 cohorts was 0.9 (0.7–1.0) overall and 0.9 (0.7–1.4) at the highest-risk decile; among women ≥50 years old, these were 1.0 (0.7–1.3) and 1.2 (0.7–1.6), respectively. The proportion of women identified above a 3% 5-year risk threshold (used for recommending risk-reducing medications in the USA) ranged from 7.0% in Germany (∼841 000 of 12 million) to 17.7% in the USA (∼5.3 of 30 million). At this threshold, 14.7% of US women were reclassified by adding the PRS to classical risk factors, with identification of 12.2% of additional future cases. Conclusion Integrating a 313-variant PRS with classical risk factors can improve the identification of European-ancestry women at elevated risk who could benefit from targeted risk-reducing strategies under current clinical guidelines.

Abstract We evaluated the joint associations between a new 313-variant PRS (PRS313) and questionnaire-based breast cancer risk factors for women of European ancestry, using 72 284 cases and 80 354 controls from the Breast Cancer Association Consortium. Interactions were evaluated using standard logistic regression and a newly developed case-only method for breast cancer risk overall and by estrogen receptor status. After accounting for multiple testing, we did not find evidence that per-standard deviation PRS313 odds ratio differed across strata defined by individual risk factors. Goodness-of-fit tests did not reject the assumption of a multiplicative model between PRS313 and each risk factor. Variation in projected absolute lifetime risk of breast cancer associated with classical risk factors was greater for women with higher genetic risk (PRS313 and family history) and, on average, 17.5% higher in the highest vs lowest deciles of genetic risk. These findings have implications for risk prevention for women at increased risk of breast cancer.