POSTER IN THE SPOTLIGHT 16 November 2022 13.30–14.20: Poster in the spotlight| Volume 175, SUPPLEMENT 1, S8-S9, November 01, 2022

An individualized breast cancer risk assessment model for women attending screening in BreastScreen Norway

      Background: The results of the effectiveness of population-based screening are controversial in terms of the balance between mortality reduction and adverse effects. In order to improve it, studies have proposed personalized screening strategies based on woman’s individual breast cancer (BC) risk. There is, therefore, a need to create individual risk prediction models through the analysis of large population-based databases. We developed a model that could be used to classify women targeted for mammography screening according to individual BC risk.
      Methods: We conducted a retrospective cohort study of 57,411 women screened in 4 counties of BreastScreen Norway between 2007 and 2019, and followed up until 2022. We used partly conditional Cox regression to estimate the adjusted hazard ratios (aHR) and the 95% confidence intervals (95%CI) for age, breast density, family history of BC, body mass index (BMI), age at menarche, alcohol habit, exercise, pregnancy, hormone replacement therapy (HRT) and benign breast disease (BBD). We calculated the 4-year absolute BC risk estimates, we validated the model using bootstrap resampling by means of the Expected-to-Observed ratio (E/O) and the area under the ROC curve (AUC) and we plotted the effect of each variable in the risk estimation.
      Results: Our results showed that all the variables included in the model explained part of the variability in BC risk. The 4-year BC risk varied between 0.22% and 7.43% with a median of 1.10%. The model slightly overestimated the risk with an E/O of 1.10 (95%CI: 1.09–1.11) and the AUC was 62.9% (95%CI: 60.8%–65.2%). Breast density was the variable that had a higher effect in the model.
      Conclusion: We developed and validated a risk prediction model to estimate the 4-year risk of BC in women eligible for mammography screening. All the ten variables used were found to significantly explain part of the variability in the BC risk. The model slightly overestimated the risk and had a similar discriminatory power than the usual BC risk prediction models. The model could be used to create individualized screening strategies aimed at improving the risk-benefit balance of mammography screening programs.
      Tabled 1
      aHR (95%CI)
      Age1.01 (1.00–1.03)
      BMI1.06 (1.04–1.08)
      Age at menarche0.95 (0.91–1.00)
      Breast density (VDG)
      10.59 (0.51–0.69)
      31.37 (1.20–1.56)
      41.71 (1.33–2.20)
      Family history of BC
      2nd degree1.17 (0.98–1.41)
      1st degree1.34 (1.10–1.63)
      Benign breast disease
      Self-reported1.55 (1.31–1.83)
      Clinician-reported1.42 (1.02–1.98)
      Alcohol habit/month
      No0.94 (0.76–1.16)
      <6 unitsRef.
      6–10 units1.06 (0.88–1.28)
      >10 units1.14 (0.96–1.36)
      <1 h0.80 (0.67–0.96)
      2–3h0.83 (0.70–0.97)
      >4 h0.65 (0.51–0.83)
      No1.10 (0.88–1.38)
      1 o 2Ref.
      >30.91 (0.79–1.04)
      Yes1.30 (1.13–1.48)
      No conflict of interest.