Abstract 248P
Background
ANI remains a major prognostic factor but there is a need for robust real-world data to provide further insight on IBC tumor features and help clinicians select optimal adjuvant treatment.
Methods
LISE is an ongoing observational study comprehensively evaluating IBC characteristics, treatments and outcomes in 3,122 cases from 2010–2021 at Courlancy Cancer Institute (Reims, France). We evaluated the correlations between tumor features and clinical settings with ANI in an interim analysis on 1,412 new IBC cases.
Results
In our cohort, there was 997 (70.6%) pN0 and 415 (29.4%) pN+ tumors (pN1mic, 5.5%; pN1, 16.5%; pN2-3, 7.4%). ANI and other IBC features, with multivariate logistic regression for pN+ prediction, are in the table. Our presentation will introduce an innovative model for an easy-to-use ANI predictive score (pN0/pN1mic-N1/pN2-3). Table: 248P
| pN+, % | χ2 p-value | Fully adjusted OR | |
| Age | p=0.035 | ||
| < 40 | 37.7 | 0.034 | 1.64 |
| 40 – 49 | 35.2 | 1.47 | |
| 50 – 74 | 27.0 | Ref | |
| ≥ 75 | 30.3 | 1.17 | |
| Detection | p<0.001 | ||
| Screening | 23.1 | <0.001 | Ref |
| Clinical symptoms | 34.7 | 1.77 | |
| pT | p<0.001 | ||
| pT1a | 4.3 | <0.001 | Ref |
| pT1b | 12.5 | 3.16 | |
| pT1c | 26.1 | 7.76 | |
| pT2 | 49.0 | 21.12 | |
| pT3-4 | 75.0 | 66.86 | |
| SBR | p<0.001 | ||
| I | 20.4 | <0.001 | Ref |
| II | 30.4 | 1.70 | |
| III | 35.6 | 2.15 | |
| Ki-67 | p=0.029 | ||
| < 14% | 25.2 | 0.028 | Ref |
| ≥ 14% | 31.6 | 1.37 | |
| LVI | p<0.001 | ||
| No | 22.3 | <0.001 | Ref |
| Yes | 69.2 | 7.80 | |
| AJCC VIII Staging | p<0.001 | ||
| Stages I – II | 26.2 | <0.001 | Ref |
| Stages III – IV | 86.1 | 17.45 | |
| Molecular subtype | p=0.017 | ||
| Luminal A | 26.6 | 0.015 | Ref |
| Luminal B HER2- | 31.3 | 1.26 | |
| Luminal B HER2+ | 43.3 | 2.11 | |
| HER2+ | 40.0 | 1.84 | |
| TN | 23.7 | 0.86 | |
| BC relapse rate a | 6.6% (pN0 patients: 2.7%) | ||
| BC related death rate a | 5.4% (pN0 patients: 1.9%) |
a Median follow-up 46.0 monthsTN: triple negative
Conclusions
pN+ tumors were significantly correlated with young age, clinical detection, tumor size (pT), SBR grading, Ki-67, lymphovascular involvement (LVI), luminal B and HER2+ subtypes, with higher relapse and death rates. The multivariate model revealed a strong influence of BC detection type, SBR, Ki-67, molecular subtype and especially LVI. These results further support the comprehensive assessment of primary tumor features in clinical practice to guide treatment decisions.
Legal entity responsible for the study
Bruno Cutuli, Institut du Cancer Courlancy, Reims, France.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.