Tumor infiltrating lymphocytes and change in tumor load on MRI to assess response and prognosis after neoadjuvant chemotherapy in breast cancer

L M Janssen; B B L Penning de Vries; M H A Janse; E van der Wall; S G Elias; R Salgado; P J van Diest; K G A Gilhuijs

doi:10.1007/s10549-024-07484-7

Tumor infiltrating lymphocytes and change in tumor load on MRI to assess response and prognosis after neoadjuvant chemotherapy in breast cancer

Breast Cancer Res Treat. 2024 Sep 16. doi: 10.1007/s10549-024-07484-7. Online ahead of print.

Authors

L M Janssen¹, B B L Penning de Vries², M H A Janse¹, E van der Wall³, S G Elias², R Salgado^{4

5}, P J van Diest⁶, K G A Gilhuijs⁷

Affiliations

¹ Image Sciences Institute, University Medical Centre Utrecht, Utrecht University, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands.
² Julius Center for Health Sciences and Primary Care, University Medical Centre Utrecht, Utrecht University, Utrecht, The Netherlands.
³ Department of Medical Oncology, University Medical Centre Utrecht, Utrecht University, Utrecht, The Netherlands.
⁴ Department of Pathology, ZAS Hospitals, Antwerp, Belgium.
⁵ Division of Research, Peter Mac Callum Cancer Centre, Melbourne, Australia.
⁶ Department of Pathology, University Medical Centre Utrecht, Utrecht University, Utrecht, The Netherlands.
⁷ Image Sciences Institute, University Medical Centre Utrecht, Utrecht University, Heidelberglaan 100, 3584 CX, Utrecht, The Netherlands. K.G.A.Gilhuijs@umcutrecht.nl.

PMID: 39285068
DOI: 10.1007/s10549-024-07484-7

Abstract

Purpose: In this study, we aimed to explore if the combination of tumor infiltrating lymphocytes (TILs) and change in tumor load on dynamic contrast-enhanced magnetic resonance imaging leads to better assessment of response to neoadjuvant chemotherapy (NAC) in patients with breast cancer, compared to either alone.

Methods: In 190 NAC treated patients, MRI scans were performed before and at the end of treatment. The percentage of stromal TILs (%TILs) was assessed in pre-NAC biopsies according to established criteria. Prediction models were developed with linear regression by least absolute shrinkage and selection operator and cross validation (CV), with residual cancer burden as the dependent variable. Discrimination for pathological complete response (pCR) was evaluated using area under the receiver operating characteristic curves (AUC). We used Cox regression analysis for exploring the association between %TILs and recurrence-free survival (RFS).

Results: Fifty-one patients reached pCR. In all patients, the %TILs model and change in MRI tumor load model had an estimated CV AUC of 0.69 (95% confidence interval (CI) 0.53-0.78) and 0.69 (95% CI 0.61-0.79), respectively, whereas a model combining the variables resulted in an estimated CV AUC of 0.75 (95% CI 0.66-0.83). In the group with tumors that were ER positive and HER2 negative (ER+/HER2-) and in the group with tumors that were either triple negative or HER2 positive (TN&HER2+) separately, the combined model reached an estimated CV AUC of 0.72 (95% CI 0.60-0.88) and 0.70(95% CI 0.59-0.82), respectively. A significant association was observed between pre-treatment %TILS and RFS (hazard ratio (HR) 0.72 (95% CI 0.53-0.98), for every standard deviation increase in %TILS, p = 0.038).

Conclusion: The combination of TILs and MRI is informative of response to NAC in patients with both ER+/HER2- and TN&HER2+ tumors.

Keywords: Breast cancer; Magnetic resonance imaging; Neoadjuvant chemotherapy; Pathological complete response; Tumor infiltrating tumor cells.

Abstract

Grants and funding