The research team, led by Qian Xing, Yong Cui, Xiao-Lei Gu and senior author Ying-Shi Sun, set out to do two things. First, they built a radiomics model—a machine learning pipeline that extracts hundreds of quantitative features from medical images, such as texture, intensity patterns and shape descriptors invisible to the human eye—to predict the tumor regression grade of individual liver metastases from gadoxetic acid-enhanced MRI. Gadoxetic acid is a liver-specific contrast agent that is taken up by functioning hepatocytes, allowing radiologists to image metastases with unusual clarity during the hepatobiliary phase of the scan. Second, and arguably more innovatively, they asked whether the heterogeneity of predicted responses across all of a patient’s lesions—some shrinking dramatically while others barely respond—could itself serve as a warning sign for future recurrence.

The study was a single-center retrospective analysis that included 295 liver lesions from 83 patients with colorectal liver metastases, split into a training group of 199 lesions and a validation group of 96 lesions. For each lesion, the researchers drew regions of interest on the latest preoperative MRI and computed radiomic features, ultimately screening down to 25 informative features that fed the predictive model. Rather than using the conventional five-tier TRG scale directly as the training label, the team used tumor residual rate—the percentage of original tumor tissue still alive after chemotherapy—as a continuous measure of response. This choice allowed the model to learn a graded, quantitative relationship between image texture and biological response rather than forcing predictions into coarse categories.

How well did the model perform? The correlation between the radiomics score and the actual tumor residual rate was moderate in both the training and validation groups, with coefficients of determination of 0.360 and 0.440 respectively. In plain terms, the AI could explain roughly a third to nearly half of the variation in how much tumor tissue survived chemotherapy, based purely on the appearance of the lesions on MRI. The authors describe this as modest predictive performance, and they are careful not to overclaim. Radiomics is not yet a substitute for pathology. But the fact that a noninvasive scan can capture a meaningful fraction of treatment response information at all is significant, because it opens the door to assessing response while the patient is still on chemotherapy, when the treatment plan can still be adjusted.

The truly headline-grabbing finding, however, concerns heterogeneity. The researchers defined radiomics-predicted TRG heterogeneity, abbreviated R_H, as the situation in which a single patient harbors both good predicted responses—predicted TRG 1 to 3, corresponding to a predicted tumor residual rate of 50 percent or less—and poor predicted responses—predicted TRG 4 to 5, with a predicted residual rate above 50 percent—across different liver lesions. This is the imaging equivalent of what pathologists have long observed under the microscope: metastases in the same liver do not respond uniformly to the same drugs. Some lesions melt away while others stubbornly persist, and this mosaic of sensitivity reflects the underlying biological diversity of the cancer.

To test whether this imaging-detected mosaic matters clinically, the team tracked local tumor disease-free survival, or LTDFS, a measure of how long patients remain free of locally recurring disease in the liver. Using Kaplan-Meier survival analysis, they compared the 13 patients whose lesions showed heterogeneity (R_H positive) with the 70 patients whose lesions were uniformly predicted to respond (R_H negative). The difference was stark. Patients with heterogeneous predicted responses had a median local tumor disease-free survival of just 3.0 months, while those without heterogeneity enjoyed a median of 10.2 months—a more than threefold difference that reached statistical significance with a p value of 0.002.

The researchers then pushed further, using Cox regression analysis to determine which factors independently predicted LTDFS when considered alongside clinical characteristics. Three variables emerged as independent predictors: the presence of radiomics-predicted TRG heterogeneity (p = 0.012), the best radiomics-predicted response across all lesions, termed R_min (p = 0.020), and whether the patient had undergone intraoperative radiofrequency ablation, a technique that destroys tumors with heat (p = 0.010). The appearance of R_min among the independent predictors is particularly intriguing. It suggests that the single best-responding lesion in a patient’s liver may carry more prognostic weight than the worst one, hinting that a cancer capable of a strong response somewhere may have a fundamentally more favorable biology than one that responds poorly everywhere.

Why should heterogeneity be so ominous? Tumor heterogeneity is one of the central challenges in modern oncology. A cancer that presents a mixed face to chemotherapy is likely to harbor resistant clones that survive treatment and seed recurrence. When those clones are distributed across multiple lesions in the liver, even complete surgical resection of the visible disease may leave behind microscopic resistant cells destined to regrow. Conventional imaging assessments, such as the RECIST criteria that track changes in lesion size, are poorly equipped to capture this biological diversity, because size is a crude proxy for response. A lesion can shrink substantially yet still be packed with viable tumor, while another may be replaced almost entirely by scar tissue. Radiomics, by quantifying internal texture and signal patterns, offers a window into these differences that simple measurements cannot provide.

The study has limitations that the authors and readers should keep in view. It was retrospective and conducted at a single center, so the model will need external validation in independent cohorts before it can inform clinical decisions. The number of patients with heterogeneity was small—13 individuals—which, while sufficient for statistical significance, leaves room for uncertainty around the precise effect size. The predictive performance of the radiomics model itself was modest, meaning that refinement of feature selection, segmentation and modeling approaches will be needed. Nevertheless, the prognostic signal from heterogeneity was strong enough to stand out even with a first-generation model, which bodes well for improved versions.

The broader implications are considerable. If validated prospectively, a radiomics-based assessment of response heterogeneity could be performed on the same gadoxetic acid-enhanced MRI scans that patients already receive before liver surgery, adding no additional imaging burden. Surgeons could use it to decide which lesions merit resection versus ablation, and oncologists could identify patients whose mixed responses warrant intensified systemic therapy or closer surveillance. The study also adds to a growing body of evidence that artificial intelligence can extract clinically meaningful information from images that radiologists cannot see with the naked eye. For patients with colorectal liver metastases—a disease where the liver remains the dominant site of treatment failure and where every month of disease-free survival matters—a smarter read of a routine scan could ultimately help tailor therapy to the true, uneven biology of each person’s cancer.

Subject of Research: Radiomics-based prediction of tumor regression grade heterogeneity on MRI for prognostic assessment in colorectal cancer liver metastases after neoadjuvant chemotherapy

Article Title: Radiomics-predicted TRG heterogeneity based on gadoxetic acid-enhanced MRI and its prognostic value in patients with CRLM after neoadjuvant chemotherapy

Article References: Xing, Q., Cui, Y., Gu, X.-L., Zhu, H.-T., Li, X.-T., & Sun, Y.-S. (2026). Radiomics-predicted TRG heterogeneity based on gadoxetic acid-enhanced MRI and its prognostic value in patients with CRLM after neoadjuvant chemotherapy. BMC Medical Imaging. https://doi.org/10.1186/s12880-026-02879-y

Image Credits: AI Generated

DOI: 10.1186/s12880-026-02879-y

Keywords: radiomics, colorectal cancer, liver metastases, gadoxetic acid MRI, tumor regression grade, heterogeneity, neoadjuvant chemotherapy, machine learning, prognosis, disease-free survival, medical imaging, artificial intelligence

The research team embarked on a comprehensive retrospective analysis involving 151 patients diagnosed with LARC who received nCRT followed by surgical resection over a nine-year period from 2014 to 2023. Central to their investigation was the correlation between pre-treatment PET/CT metabolic data and tumor regression grade (TRG) post-surgery, categorized according to the Mandard classification system. Patients were dichotomized into complete responders (TRG1) and incomplete responders (TRG2–5), allowing for a focused assessment of clinical and pathological factors influencing outcomes.

Metabolic parameters derived from FDG PET/CT have long been hypothesized as potential biomarkers for tumor behavior and treatment sensitivity. SUVmax, referring to the peak uptake within the tumor, along with volumetric measures such as MTV and TLG, reflect both the metabolic activity and tumor burden. However, intriguingly, the study’s findings contradict expectations, as none of these PET/CT parameters demonstrated statistically significant predictive value for either pathological tumor regression or survival endpoints in the context of neoadjuvant therapy.

Instead, attention shifted to a host of clinico-pathological variables that emerged as robust predictors of therapeutic success and prognosis. Family history of cancer, baseline and post-treatment carcinoembryonic antigen (CEA) levels, radiotherapy-associated carbohydrate antigen 19-9 (CA19-9) dynamics, and the presence of concurrent chemotherapy (CC) stood out as critical factors determining complete response rates. These clinical indicators underscore the multifaceted nature of cancer progression and the limitations of relying solely on imaging data for comprehensive patient stratification.

The impact of histopathological features on treatment outcomes cannot be overstated. Perforation, lymphovascular invasion (LVI), and perineural invasion (PNI)—classic markers of aggressive tumor biology—invariably correlated with incomplete response and worse survival profiles. Their inclusion in multivariate analyses reinforced their independent prognostic significance, highlighting that microscopic tumor behavior ultimately governs response to nCRT beyond what can be visualized metabolically.

From a survival perspective, TRG1 patients enjoyed markedly improved overall survival (OS) and disease-free survival (DFS), reflecting the clinical importance of achieving a complete pathological response. Kaplan–Meier survival curves neatly illustrated these disparities, reinforcing TRG1 as a pivotal biomarker for patient prognosis. Moreover, multivariate Cox regression models revealed that TRG1 status, administration of concurrent chemotherapy, and absence of tumor perforation independently predicted improved survival outcomes.

The lack of correlation between pre-treatment FDG PET/CT parameters and treatment response represents a significant paradigm shift. This revelation urges oncologists and radiologists to reconsider the predictive weight traditionally assigned to metabolic imaging in the neoadjuvant setting for rectal cancer. It further emphasizes the necessity of integrating a more holistic clinical and pathological approach when designing personalized treatment strategies.

Neoadjuvant chemoradiotherapy is a cornerstone in managing LARC, frequently enabling tumor downstaging and sphincter-sparing surgeries. However, individualized responses vary greatly, posing challenges to optimizing treatment. The study’s integrative methodology, combining imaging, biochemical markers, and histological data, provides a robust framework that could streamline prognostication and allow for more nuanced clinical decision-making.

CEA and CA19-9, both serum tumor markers, have long been employed in gastrointestinal oncology for monitoring disease progression and recurrence. Their demonstrated associations with TRG and survival in this study reaffirm their clinical utility in tandem with conventional diagnostic tools. Importantly, post-treatment CEA levels emerged as a particularly powerful indicator, reflecting residual tumor burden and potential microscopic disease persistence.

Concurrent chemotherapy, when administered alongside radiotherapy, was associated with better pathological response and improved survival metrics. This finding aligns with existing literature advocating for combination regimens to enhance radiosensitization and maximize tumor eradication. It validates current therapeutic protocols and paves the way for further investigations into optimizing concurrent treatment modalities.

The study also explored the role of family history, traditionally a surrogate for genetic predisposition and shared environmental risk factors, in modulating treatment response. The strong association observed suggests that genetic and epigenetic influences may play a more significant role in therapeutic sensitivity than previously recognized, warranting further molecular profiling studies to elucidate underlying mechanisms.

Importantly, the presence of perforation—an indicator of tumor invasion into surrounding tissues—was strongly predictive of poorer outcomes. This emphasizes the aggressive nature of certain rectal cancers and the need for early detection and intervention. Histopathological assessment after surgery remains indispensable for accurate staging and prognosis.

Perineural invasion and lymphovascular invasion, both indicative of tumor dissemination pathways, independently impacted treatment response and survival, reaffirming their value as critical prognostic factors. Their detection necessitates thorough pathological evaluation to guide adjuvant therapy considerations.

While FDG PET/CT remains invaluable for cancer staging and detecting distant metastases, this study cautions against overreliance on baseline metabolic parameters as solitary prognostic tools in the context of nCRT for LARC. A multimodal diagnostic strategy incorporating serum markers, histopathological examination, and clinical context remains paramount.

The findings presented in this study have profound implications for clinical practice. They suggest that clinicians should place greater emphasis on molecular and histopathological markers when predicting response to nCRT in rectal cancer. Additionally, the study advocates for continued research into novel biomarkers and imaging techniques that might better capture the complexity of tumor biology and treatment effects.

This comprehensive investigation lays the foundation for refining patient selection criteria and tailoring neoadjuvant treatment protocols. By recognizing the limited prognostic role of baseline FDG PET/CT metabolic parameters and amplifying the significance of clinical-pathological features, oncologists can enhance therapeutic precision, improve survival rates, and ultimately advance patient care in rectal cancer management.

Subject of Research: Predictive value of baseline F-18 FDG PET/CT metabolic parameters and clinico-pathological variables in tumor regression and survival outcomes in locally advanced rectal cancer patients undergoing neoadjuvant chemoradiotherapy.

Article Title: Evaluation of pre-treatment F-18 FDG PET/CT according to Mandard classification in locally advanced rectal cancer patients undergoing neoadjuvant chemoradiotherapy.

Article References:
Aras, F., Parvizi, M., Nalbant, O.A. et al. Evaluation of pre-treatment F-18 FDG PET/CT according to Mandard classification in locally advanced rectal cancer patients undergoing neoadjuvant chemoradiotherapy. BMC Cancer 25, 1262 (2025). https://doi.org/10.1186/s12885-025-14659-y

Image Credits: Scienmag.com

DOI: https://doi.org/10.1186/s12885-025-14659-y