Lay Abstract

Desmoid tumors are aggressive tumors characterized by unpredictable behavior. Current scientific guidelines recommend following patients with a new diagnosis of desmoid tumor because some tumors may spontaneously shrink or remain stable. Treatments such as radiation, surgery, or systemic therapy are reserved for patients with growing tumors. However, delaying intervention in patients who ultimately require treatment may allow the tumor to progress, potentially resulting in locally advanced disease and a higher risk of treatment-related complications. Thus, predicting disease progression at the time of initial diagnosis is highly desirable.  

Human interpretation of magnetic resonance imaging does not predict the evolution of desmoid tumor. We propose to use a deep learning model, to extract the properties of images to uncover patterns and characteristics that cannot be appreciated by the naked eye. The objectives of this study are to collect demographics and imaging data of patients with desmoid tumors and develop a MMAI models to predict whether desmoid tumors will decrease in size, remain stable, or grow over time or if patients will eventually require definitive therapy or not. Once developed, future work will focus on validating this tool, with the goal of ultimately supporting clinical decision-making and patient counseling in the management of desmoid tumors.  

Scientific Abstract

Desmoid tumors (DT) are locally aggressive neoplasms characterized by an unpredictable behavior. Recent guidelines recommend approaching patients with newly diagnosed disease with an initial period of observation because of possible spontaneous tumor regression or stability. Systemic or definitive therapy are reserved for patients with tumor progression. Predicting disease progression at diagnosis is therefore of interest. However, we currently lack reliable methods to accurately predict disease progression. Standard radiological features with magnetic resonance imaging (MRI) are poor correlates of disease evolution.  

We therefore propose to develop a multimodal artificial intelligence (MMAI) model that integrates clinical and imaging-derived features to predict the evolution of desmoid tumors managed with initial observation. To achieve this, we will identify a cohort of patients with biopsy-confirmed desmoid tumors, all of whom underwent baseline diagnostic MRI and were subsequently followed over time. Clinical and imaging data will be collected, and outcomes during the subsequent two years will be categorized as follows: 1- Radiological outcomes assessed as per RECIST criteria and 2- Clinical outcomes defined as patients, requiring definitive treatment or not. 

An unsupervised deep learning model will be used to extract imaging features from MRI which will be combined with clinical variables to train the MMAI models using a supervised learning framework. The goal is to predict disease trajectory under observation, with the ultimate objective of validating the models in an external cohort and paving the way for a prospective clinical trial to assess its utility in guiding physician decision-making and patient counseling.

Lay Abstract

Desmoid tumors vary widely in behavior—some regress or remain stable, while others grow aggressively—but we currently lack tools to predict their course. To address this, a pilot study analyzed gene activity in individual cells from patient tumors and developed a novel method to identify cells carrying hallmark mutations in CTNNB1 or APC, allowing for precise distinction between tumor-specific signals and those from surrounding cells. Building on this work, the proposed study has three goals: (1) apply the improved approach to existing patient samples to identify gene activity patterns linked to tumor behavior and share these data through an interactive online portal; (2) recreate and manipulate these gene programs in laboratory-grown tumor models to test their roles in driving progression or regression; and (3) investigate how neighboring blood vessel and immune cells influence tumor behavior. Together, these efforts aim to uncover the biological drivers of desmoid tumor variability and lay the groundwork for targeted therapies and more personalized care strategies.

Scientific Abstract

Desmoid tumors, driven predominantly by CTNNB1 or APC mutations, display diverse clinical behaviors ranging from spontaneous regression to aggressive growth, yet the biological basis for this variability remains unclear. This project builds on a pilot study that used single-nucleus RNA sequencing with mutation annotation to identify transcriptional signatures unique to CTNNB1-mutant cells and developed a targeted amplification protocol to improve detection of these mutant populations. Preliminary findings suggest distinct transcriptional profiles and cellular compositions between regressing and progressing tumors. The central hypothesis is that transcriptional programs linked to progression or regression can be modeled and perturbed in vitro. Aim 1 applies enhanced genotyping and transcriptomic analysis across desmoid tumors to identify gene programs and microenvironmental features associated with clinical outcomes, with data shared through a public portal. Aim 2 tests key tumor-intrinsic hypotheses, including a profibrotic COL25A1-driven loop and a potential antifibrotic role for CD44. Aim 3 focuses on the tumor microenvironment, using computational and coculture approaches to identify supportive or inhibitory niches involving immune and endothelial cells. Together, this research will define the transcriptional and cellular mechanisms that drive desmoid tumor behavior and enable predictive and therapeutic advances using patient-derived models.

Lay Abstract

We have recently conducted a prospective study on patients affected by desmoid-type fibromatosis under active surveillance (AS), aiming to understand if changes in tumor immune microenvironment (TIME) and circulating immune populations correlate and influence spontaneous tumor regression. Our analysis suggested dysregulations of immune cells in TIME and peripheral blood in patients experiencing tumor regression. These findings support the hypothesis that the immune system may play a central role in inducing tumor regression. We are collecting additional blood and tumor tissues in a new prospective multicenter study in order to validate our previous findings in a larger patient population.

Our aim is to deeply characterize immune cells present in desmoid tumor tissue through spatial omic approaches to understand their cross-talk with tumor cells in the TIME. Additionally, we hypothesize that in patients who undergo a systemic therapy, changes in TIME and peripheral immune population recapitulate those observed during AS. Specifically, we want to test if methotrexate plus vinorelbine can induce changes in the peripheral immune population making patient peripheral immune profile similar to that of patients who experience spontaneous regression during AS. To test this hypothesis, we will conduct a prospective observation study to collect tissue and blood samples from patients on first-line treatment with methotrexate and vinorelbine. We expect that our findings by integrating data from patients who undergo spontaneous regression during AS with data from patients experiencing response to systemic treatments, will further show how the immune system contributes to tumor control and regression.

Scientific Abstract

Sporadic desmoid-type fibromatosis (DF) is a rare, locally aggressive mesenchymal tumor with an unpredictable clinical course. While some patients experience spontaneous regression (REG) under active surveillance (AS), others progress (PROG) and require treatment. In a recent prospective study at our institution, we analyzed the immune landscape in DF patients under AS and identified distinct immune profiles associated with diverse clinical outcomes. These findings globally suggest that immune homeostasis plays a central role in modulating DF behavior, promoting pro-resolving immunity in REG while instead sustaining a pro-inflammatory, pro-fibrotic phenotype in PROG.

Notably, these immune signatures can be detected in peripheral blood, offering potential for novel biomarkers to predict disease course and enable patient-tailored interventions.

Building on these insights, we propose a translational research project aimed at further dissecting the immune mechanisms underlying both spontaneous and treatment-induced regression in DF patients enrolled in a newly activated prospective international multicenter study, partially funded by an institutional grant. Specifically, we will:

1. Evaluate baseline and longitudinal blood samples to quantify and profile pro-resolving versus pro-inflammatory immune responses in both spontaneous and treatment-induced regression, compared to progression.

2. To deeply characterize immune cell populations within desmoid tumor tissue using spatial omic approaches and define their interactions with tumor cells in the TIME, with the goal of associating specific immune compositions with disease outcome (progression vs. regression).

3. Apply statistical approaches to integrate findings from Aim1 with clinical and tumor data (i.e. β-catenin mutation), as well as with systemic immune profile of patients under AS to identify common mechanisms of spontaneous and therapy-induced regression.

This comprehensive approach will enhance our understanding of DF immunopathogenesis and facilitate the development of targeted immunomodulatory strategies.