Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Multicenter Study
. 2024 Nov 20;10(11):1832-1845.
doi: 10.3390/tomography10110134.

Tumor Morphology for Prediction of Poor Responses Early in Neoadjuvant Chemotherapy for Breast Cancer: A Multicenter Retrospective Study

Wen Li  1 , Nu N Le  1 , Rohan Nadkarni  1 , Natsuko Onishi  1 , Lisa J Wilmes  1 , Jessica E Gibbs  1 , Elissa R Price  1 , Bonnie N Joe  1 , Rita A Mukhtar  2 , Efstathios D Gennatas  3 , John Kornak  3 , Mark Jesus M Magbanua  4 , Laura J Van't Veer  4 , Barbara LeStage  5 , Laura J Esserman  2 , Nola M Hylton  1
Affiliations
Multicenter Study

Tumor Morphology for Prediction of Poor Responses Early in Neoadjuvant Chemotherapy for Breast Cancer: A Multicenter Retrospective Study

Wen Li et al. Tomography. .

Abstract

Background: This multicenter and retrospective study investigated the additive value of tumor morphologic features derived from the functional tumor volume (FTV) tumor mask at pre-treatment (T0) and the early treatment time point (T1) in the prediction of pathologic outcomes for breast cancer patients undergoing neoadjuvant chemotherapy.

Methods: A total of 910 patients enrolled in the multicenter I-SPY 2 trial were included. FTV and tumor morphologic features were calculated from the dynamic contrast-enhanced (DCE) MRI. A poor response was defined as a residual cancer burden (RCB) class III (RCB-III) at surgical excision. The area under the receiver operating characteristic curve (AUC) was used to evaluate the predictive performance. The analysis was performed in the full cohort and in individual sub-cohorts stratified by hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) status.

Results: In the full cohort, the AUCs for the use of the FTV ratio and clinicopathologic data were 0.64 ± 0.03 (mean ± SD [standard deviation]). With morphologic features, the AUC increased significantly to 0.76 ± 0.04 (p < 0.001). The ratio of the surface area to volume ratio between T0 and T1 was found to be the most contributing feature. All top contributing features were from T1. An improvement was also observed in the HR+/HER2- and triple-negative sub-cohorts. The AUC increased significantly from 0.56 ± 0.05 to 0.70 ± 0.06 (p < 0.001) and from 0.65 ± 0.06 to 0.73 ± 0.06 (p < 0.001), respectively, when adding morphologic features.

Conclusion: Tumor morphologic features can improve the prediction of RCB-III compared to using FTV only at the early treatment time point.

Keywords: breast cancer; magnetic resonance imaging; multicenter clinical trial; neoadjuvant therapy; residual cancer burden; tumor morphology.

PubMed Disclaimer

Conflict of interest statement

L.J.E. is on the Blue Cross Medical Advisory Panel, is an uncompensated board member of Quantum Leap Healthcare Collaborative, and is an Investigator who initiated trial for high-risk DCIS funded by Moderna for DCIS phase 1 study. L.J.v.V. is part-time employee and stocks Agendia NV, advisor and stock options Exai Inc. N.M.H. receives institutional research funding from NIH. B.N.J. received author royalties from UpToDate, received WorldClassCME honoraria for lectures, received Medicolegal consulting payment, serves as Board of Directors for Society of Breast Imaging, Board of Trustees for RSNA R&E Foundation, and Deputy Editor for Radiology: Imaging Cancer. All authors declare no conflicts of interest regarding this study.

Figures

Figure 1
Figure 1
Illustration of radiomic feature extraction from functional tumor volume (FTV) tumor mask in dynamic contrast-enhanced MRI. First, early percent enhancement (PE) and signal enhancement ratio (SER) thresholds were applied to generate the FTV tumor mask, from which the FTV was calculated. Second, multiple preprocessing steps were applied to the FTV tumor mask to fill small holes, smooth edges, and remove small clusters of connected voxels. Lastly, the Pyradiomics package was used to extract radiomic 3D shape features.
Figure 2
Figure 2
Data inclusion and exclusion. Patients were excluded from the analysis because of missing pathological outcomes, clinicopathologic data, or MRI or unusable imaging data.
Figure 3
Figure 3
Boxplots of area under the receiver operating characteristic curve (AUC) for prediction of residual disease with and without shape features. AUCs were evaluated by optimal machine learning models independently in 20 stratified subsamples of the analysis cohort (n = 910) for the prediction of RCB-III. Model—without shape: FTVR and clinicopathologic data were used in the predictive model. Model—with shape: shape features were added to the predictive model together with FTVR and clinicopathologic data.
Figure 4
Figure 4
Variable importance for prediction of RCB-III using shape features together with FTV ratio and demographic variables by random forest. Variable importance was ranked according to the mean decrease in accuracy (%) when a variable was excluded. A higher value means that a variable was more important. T0: pretreatment time point. T1: early treatment time point.
Figure 5
Figure 5
Beeswarm plots with overlaid boxplot of MRI features. (a) Functional tumor volume (FTV) ratio between pretreatment and early treatment. (b) Ratio of surface area to volume between pretreatment and early treatment. Number of patients in the RCB-III group was 141, with 769 patients in the nRCB-III group (RCB-0, -I, or -II).
Figure 6
Figure 6
Example cases. Representative slices of post-contrast dynamic contrast-enhanced MRI at pretreatment (T0) and early treatment (T1) time points are shown. Functional tumor volume (FTV) tumor masks were generated by voxels within the region-of-interest box in yellow that had an early percent enhancement (PE) above 70% and are shown superimposed on the representative slices. Colors within FTV tumor masks represent different levels of signal enhancement ratio (SER)—blue: 0 to 0.9; purple: 0.9 to 1.0; green: 1.0 to 1.3; red: 1.3 to 1.75; white: 1.75 and higher. Three-dimensional surface rendering of the preprocessed tumor mask is shown next to the representative slice. (a) An example of a patient with RCB-0. (b) An example of a patient with RCB-III.
Figure 7
Figure 7
Boxplots of area under the receiver operating characteristic curve (AUC) for prediction of residual disease with and without shape features. AUCs were evaluated by optimal machine learning models independently in 20 stratified subsamples of the analysis cohort (n = 910) for the prediction of RCB-III in (a) HR+/HER2−, (b) triple-negative, and (c) HR+/HER2+. Model—without shape: FTVR and clinicopathologic data were used in the predictive model. Model—with shape: shape features were added to the predictive model together with FTVR and clinicopathologic data.
See this image and copyright information in PMC

References

    1. Spring L.M., Fell G., Arfe A., Sharma C., Greenup R., Reynolds K.L., Smith B.L., Alexander B., Moy B., Isakoff S.J., et al. Pathologic Complete Response after Neoadjuvant Chemotherapy and Impact on Breast Cancer Recurrence and Survival: A Comprehensive Meta-Analysis. Clin. Cancer Res. 2020;26:2838–2848. doi: 10.1158/1078-0432.CCR-19-3492. - DOI - PMC - PubMed
    1. Yee D., DeMichele A.M., Yau C., Isaacs C., Symmans W.F., Albain K.S., Chen Y.-Y., Krings G., Wei S., Harada S., et al. Association of Event-Free and Distant Recurrence-Free Survival With Individual-Level Pathologic Complete Response in Neoadjuvant Treatment of Stages 2 and 3 Breast Cancer: Three-Year Follow-up Analysis for the I-SPY2 Adaptively Randomized Clinical Trial. JAMA Oncol. 2020;6:1355–1362. doi: 10.1001/jamaoncol.2020.2535. - DOI - PMC - PubMed
    1. Symmans W.F., Peintinger F., Hatzis C., Rajan R., Kuerer H., Valero V., Assad L., Poniecka A., Hennessy B., Green M., et al. Measurement of Residual Breast Cancer Burden to Predict Survival after Neoadjuvant Chemotherapy. J. Clin. Oncol. 2007;25:4414–4422. doi: 10.1200/JCO.2007.10.6823. - DOI - PubMed
    1. Provenzano E., Bossuyt V., Viale G., Cameron D., Badve S., Denkert C., MacGrogan G., Penault-Llorca F., Boughey J., Curigliano G., et al. Standardization of Pathologic Evaluation and Reporting of Postneoadjuvant Specimens in Clinical Trials of Breast Cancer: Recommendations from an International Working Group. Mod. Pathol. 2015;28:1185–1201. doi: 10.1038/modpathol.2015.74. - DOI - PubMed
    1. Bossuyt V., Provenzano E., Symmans W.F., Boughey J.C., Coles C., Curigliano G., Dixon J.M., Esserman L.J., Fastner G., Kuehn T., et al. Recommendations for Standardized Pathological Characterization of Residual Disease for Neoadjuvant Clinical Trials of Breast Cancer by the BIG-NABCG Collaboration. Ann. Oncol. 2015;26:1280–1291. doi: 10.1093/annonc/mdv161. - DOI - PMC - PubMed

Publication types