An optimal MRI method for detection of intratumoral cancer cell dissemination doorways
openNCI - National Cancer Institute
Project Summary/Abstract
Cancer cell dissemination often occurs at specialized intravasation sites on blood vessel walls called “Tumor
Microenvironment of Metastasis” (TMEM) Doorways. These structures, composed of a tumor cell, a perivascular
macrophage, and an endothelial cell, facilitate the transient and localized opening of the blood vessel, a
process known as TMEM-Associated Vascular Opening (TAVO), eventually leading to entry of prometastatic
tumor cells into the peripheral circulation. Prior research from our group has amply demonstrated that TMEM
doorway activity is associated with increased metastatic risk in breast cancer patients, and that it can predict
prometastatic responses elicited by cytotoxic neoadjuvant chemotherapy, thus offering the potential of exploitation
to promote personalized treatment. To this end, we previously developed a non-invasive imaging technique,
TMEM Doorway Activity-MRI (TDAM), utilizing dynamic contrast-enhanced MRI to measure the activity
of TMEM doorways (i.e., TAVO events) in preclinical models of breast carcinoma and human patients. The use
of animal models in this study is essential, as TMEM doorway biology represents a highly dynamic multicellular
process that cannot currently be mechanistically interrogated or experimentally manipulated in human patients.
Specifically, the MMTV-PyMT mouse model of breast carcinoma is (patho)physiologically relevant to human
breast cancer and further permits controlled therapeutic interventions under standardized conditions, enabling
rigorous optimization of TDAM features prior to clinical translation. To outline a clear pathway for the development
and clinical integration of the TDAM assay, this proposal aims to refine and validate TDAM, seeking to
predict metastatic risk and monitor therapeutic response in breast cancer patients, thus potentially transforming
it into a companion diagnostic tool. Our study is divided into two primary aims. Aim 1 focuses on refining TDAM
measurement features using orthogonal methodologies in the MMTV-PyMT mouse model, to increase its sensitivity
and predictive accuracy. To achieve this, we will adopt a data-driven machine-learning approach to generate
a classifier model and calculate contrast response curves within MRI voxels of mouse breast carcinomas.
Aim 2 seeks to validate these measurements in a clinical setting with a cohort of breast cancer patients. Specifically,
the clinical study will assess changes in TMEM doorway activity before and after neoadjuvant chemotherapy,
using TDAM features correlated with established metastatic dissemination endpoints, such as circulating
tumor cells. TDAM represents a significant advance over existing imaging modalities by distinguishing potentially
lethal cancers with high metastatic risk from less aggressive forms. Importantly, TDAM uses standardof-
care MRI data and advanced computational analyses to measure multiple tissue properties associated with
TMEM activity, which could markedly improve non-invasive metastatic risk assessment in breast cancer management.
In conclusion, the successful implementation of TDAM could provide a critical tool for predicting
chemotherapy outcomes non-invasively, thus guiding more personalized and effective treatment strategies.
Up to $432K
health research