Advanced imaging of colorectal cancer: From anatomy to molecular imaging

Enhancing Medical Imaging Segmentation with GB-SAM: A Novel Approach to Tissue Segmentation Using Granular Box Prompts

Recent advances in foundation models have revolutionized model development in digital pathology, reducing dependence on extensive manual annotations required by traditional methods. The ability of foundation models to generalize well with few-shot learning addresses critical barriers in adapting models to diverse medical imaging tasks. This work presents the Granular Box Prompt Segment Anything Model (GB-SAM), an improved version of the Segment Anything Model (SAM) fine-tuned using granular box prompts with limited training data. The GB-SAM aims to reduce the dependency on expert pathologist annotators by enhancing the efficiency of the automated annotation process. Granular box prompts are small box regions derived from ground truth masks, conceived to replace the conventional approach of using a single large box covering the entire H&E-stained image patch. This method allows a localized and detailed analysis of gland morphology, enhancing the segmentation accuracy of individual glands and reducing the ambiguity that larger boxes might introduce in morphologically complex regions. We compared the performance of our GB-SAM model against U-Net trained on different sizes of the CRAG dataset. We evaluated the models across histopathological datasets, including CRAG, GlaS, and Camelyon16. GB-SAM consistently outperformed U-Net, with reduced training data, showing less segmentation performance degradation. Specifically, on the CRAG dataset, GB-SAM achieved a Dice coefficient of 0.885 compared to U-Net's 0.857 when trained on 25% of the data. Additionally, GB-SAM demonstrated segmentation stability on the CRAG testing dataset and superior generalization across unseen datasets, including challenging lymph node segmentation in Camelyon16, which achieved a Dice coefficient of 0.740 versus U-Net's 0.491. Furthermore, compared to SAM-Path and Med-SAM, GB-SAM showed competitive performance. GB-SAM achieved a Dice score of 0.900 on the CRAG dataset, while SAM-Path achieved 0.884. On the GlaS dataset, Med-SAM reported a Dice score of 0.956, whereas GB-SAM achieved 0.885 with significantly less training data. These results highlight GB-SAM's advanced segmentation capabilities and reduced dependency on large datasets, indicating its potential for practical deployment in digital pathology, particularly in settings with limited annotated datasets.

Role of artificial intelligence in risk prediction, prognostication, and therapy response assessment in colorectal cancer: current state and future directions

Artificial Intelligence (AI) is a branch of computer science that utilizes optimization, probabilistic and statistical approaches to analyze and make predictions based on a vast amount of data. In recent years, AI has revolutionized the field of oncology and spearheaded novel approaches in the management of various cancers, including colorectal cancer (CRC). Notably, the applications of AI to diagnose, prognosticate, and predict response to therapy in CRC, is gaining traction and proving to be promising. There have also been several advancements in AI technologies to help predict metastases in CRC and in Computer-Aided Detection (CAD) Systems to improve miss rates for colorectal neoplasia. This article provides a comprehensive review of the role of AI in predicting risk, prognosis, and response to therapies among patients with CRC.

Diagnostic performance of [18F]-FDG PET/MR in evaluating colorectal cancer: a systematic review and meta-analysis

PURPOSE

To calculate the diagnostic performance of [¹⁸F]-FDG PET/MR in colorectal cancer (CRC).

METHODS

This study was designed following the PRISMA-DTA guidelines. To be included, published original articles (until December 31, 2021) that met the following criteria were considered eligible: (1) evaluated [¹⁸F]-FDG PET/MR as the diagnostic method to detect CRC; (2) compared [¹⁸F]-FDG PET/MR with histopathology as the reference standard, or clinical/imaging composite follow-up when pathology was not available; (3) provided adequate crude data for meta-analysis. The diagnostic pooled measurements were calculated at patient and lesion levels. Regarding sub-group analysis, diagnostic measurements were calculated in "TNM staging," "T staging," "N staging," "M staging," and "liver metastasis" sub-groups. Additionally, we calculated the pooled performances in "rectal cancer: patient-level" and "rectal cancer: lesion-level" sub-groups. A hierarchical method was used to pool the performances. The bivariate model was conducted to find the summary points. Analyses were performed using STATA 16.

RESULTS

A total of 1534 patients from 18 studies were entered. The pooled sensitivities in CRC lesion detection (tumor, lymph nodes, and metastases) were 0.94 (95%CI: 0.89-0.97) and 0.93 (95%CI: 0.82-0.98) at patient-level and lesion-level, respectively. The pooled specificities were 0.89 (95%CI: 0.84-0.93) and 0.95 (95%CI: 0.90-0.98) at patient-level and lesion-level, respectively. In sub-groups, the highest sensitivity (0.97, 95%CI: 0.86-0.99) and specificity (0.99, 95%CI: 0.84-1.00) were calculated for "M staging" and "rectal cancer: lesion-level," respectively. The lowest sensitivity (0.81, 95%CI: 0.65-0.91) and specificity (0.79, 95%CI: 0.52-0.93) were calculated for "N staging" and "T staging," respectively.

CONCLUSION

This meta-analysis showed an overall high diagnostic performance for [¹⁸F]-FDG PET/MR in detecting CRC lesions/metastases. Thus, this modality can play a significant role in several clinical scenarios in CRC staging and restaging. Specifically, one of the main strengths of this modality is ruling out the existence of CRC lesions/metastases. Finally, the overall diagnostic performance was not found to be affected in the post-treatment setting.

1H-MRS application in the evaluation of response to photo-thermal therapy using iron oxide-gold core-shell nanoparticles, an in vivo study

BACKGROUND

Photo-thermal therapy (PTT) has been at the center of attention as a new method for cancer treatment in recent years. It is important to predict the response to treatment in the PTT procedure. Using magnetic resonance spectroscopy (MRS) can be considered a novel technique in evaluating changes in metabolites resulted from PTT.

METHODS

In the present project, we conducted an in vivo study to assess the efficacy of ¹H-MRS as a noninvasive technique to evaluate the response to treatment in the early hours following PTT. The BALB/c mice subcutaneously bearing tumor cells (CT26 cell line) were scanned by ¹H-MRS before and after PTT. Iron oxide-gold core-shell (Fe₃O₄@Au) as PTT agent was injected into intra-peritoneal at first and then irradiated by NIR laser. Single-voxel Point RESolved Spectroscopy (PRESS) sequence (TE = 144) was used, and metabolites alternations were evaluated by the non-parametric Wilcoxon test. Besides, Nanoparticle (NP) relaxometry was conducted for negative contrast agents' potentials.

RESULTS

MRS choline (Cho) peak dramatically reduced 24 h post-PTT (p = 0.01) and lipid peak as a marker for necrosis of tumor elevated (p = 0.01) just in group 3 (NPs injection + laser irradiation) 24 h after the procedure.

CONCLUSION

¹H-MRS showed its potential as a method in detecting the changes in metabolites and revealing the outcome accurately. Response to photo-thermal therapy evaluation was achievable only one day after PTT and proved by a 10-day follow-up of the tumor size. Iron oxide-gold core-shell can also be used as a negative contrast agent in MRI images during therapy.

Ribosome Biogenesis Alterations in Colorectal Cancer

Many studies have focused on understanding the regulation and functions of aberrant protein synthesis in colorectal cancer (CRC), leaving the ribosome, its main effector, relatively underappreciated in CRC. The production of functional ribosomes is initiated in the nucleolus, requires coordinated ribosomal RNA (rRNA) processing and ribosomal protein (RP) assembly, and is frequently hyperactivated to support the needs in protein synthesis essential to withstand unremitting cancer cell growth. This elevated ribosome production in cancer cells includes a strong alteration of ribosome biogenesis homeostasis that represents one of the hallmarks of cancer cells. None of the ribosome production steps escape this cancer-specific dysregulation. This review summarizes the early and late steps of ribosome biogenesis dysregulations described in CRC cell lines, intestinal organoids, CRC stem cells and mouse models, and their possible clinical implications. We highlight how this cancer-related ribosome biogenesis, both at quantitative and qualitative levels, can lead to the synthesis of ribosomes favoring the translation of mRNAs encoding hyperproliferative and survival factors. We also discuss whether cancer-related ribosome biogenesis is a mere consequence of cancer progression or is a causal factor in CRC, and how altered ribosome biogenesis pathways can represent effective targets to kill CRC cells. The association between exacerbated CRC cell growth and alteration of specific steps of ribosome biogenesis is highlighted as a key driver of tumorigenesis, providing promising perspectives for the implementation of predictive biomarkers and the development of new therapeutic drugs.

Cardiac magnetic resonance in hypertrophic and dilated cardiomyopathies

Cardiomyopathies are a heterogeneous entity. The progress in the field of genetics has allowed over the years to determine its origin more and more often. The classification of these pathologies has changed over the years; it has been updated with new knowledge. Imaging allows to define the phenotypic characteristics of the different forms of cardiomyopathy. Cardiac magnetic resonance (CMR) allows a morphological evaluation of the associated (and sometimes pathognomonic) cardiac findings of any form of cardiomyopathy. The tissue characterization sequences also make magnetic resonance imaging unique in its ability to detect changes in myocardial tissue. This review aims to define the features that can be highlighted by CMR in hypertrophic and dilated forms and the possible differential diagnoses. In hypertrophic forms, CMR provides: precise evaluation of wall thickness in all segments, ventricular function and size and evaluation of possible presence of areas of fibrosis as well as changes in myocardial tissue (measurement of T1 mapping and extracellular volume values). In dilated forms, cardiac resonance is the gold standard in the assessment of ventricular volumes. CMR highlights also the potential alterations of the myocardial tissue.

Noninvasive assessment and therapeutic monitoring of drug-resistant colorectal cancer by MR molecular imaging of extradomain-B fibronectin

Antineoplastic resistance represents a multifaceted challenge for cancer therapy and diagnostics. Extensive molecular heterogeneity, even within neoplasms of the same type, can elicit distinct outcomes of administering therapeutic pressures, frequently leading to the development of drug-resistant populations. Improved success of oncotherapies merits the exploration of precise molecular imaging technologies that can detect not only anatomical but also molecular changes in tumors and their microenvironment, early on in the treatment regimen. To this end, we developed magnetic resonance molecular imaging (MRMI) strategies to target the extracellular matrix oncoprotein, extradomain-B fibronectin (EDB-FN), for non-invasive assessment and therapeutic monitoring of drug-resistant colorectal cancer (CRC). Methods: Two drug-resistant CRC lines generated from parent DLD-1 and RKO cells by long-term treatment with 5'-FU and 5'-FU plus CB-839 respectively, were characterized for functional and gene expression changes using 3D culture, transwell invasion, qRT-PCR, and western blot assays. Contrast-enhanced MRMI of EDB-FN was performed in athymic nu/nu mice bearing subcutaneous tumor xenografts with 40 µmol/kg dose of macrocyclic ZD2-targeted contrast agent MT218 [ZD2-N3-Gd (HP-DO3A)] on a 3T MRS 3000 scanner. Immunohistochemistry was conducted on patient specimens and xenografts using anti-EDB-FN antibody G4. Results: Analyses of TCGA and GTEx databases revealed poor prognosis of colon cancer patients with higher levels of EDB-FN. Similarly, immunohistochemical staining of patient specimens showed increased EDB-FN expression in primary colon adenocarcinoma and hepatic metastases, but none in normal adjacent tissues. Drug-resistant DLD1-DR and RKO-DR cells were also found to demonstrate enhanced invasive potential and significantly elevated EDB-FN expression over their parent counterparts. MRMI of EDB-FN with 40 µmol/kg dose of MT218 (60% lower than the clinical dose) resulted in robust signal enhancement in the drug-resistant CRC xenografts with 84-120% increase in their contrast-to-noise ratios (CNRs) over the non-resistant counterparts. The feasibility of non-invasive therapeutic monitoring using MRMI of EDB-FN was also evaluated in drug-resistant DLD1-DR tumors treated with a pan-AKT inhibitor MK2206-HCl. The treated drug-resistant tumors failed to respond to therapy, which was accurately detected by MRMI with MT218, demonstrating higher signal enhancement and increased CNRs in the 4-week follow-up scans over the pre-treatment scans. Conclusions: EDB-FN is a promising molecular marker for assessing drug resistance. MRMI of EDB-FN with MT218 at a significantly reduced dose can facilitate effective non-invasive assessment and treatment response monitoring of drug-resistant CRC, highlighting its translational potential for active surveillance and management of CRC and other malignancies.

Colorectal cancer: Parametric evaluation of morphological, functional and molecular tomographic imaging

Colorectal cancer (CRC) represents one of the leading causes of tumor-related deaths worldwide. Among the various tools at physicians’ disposal for the diagnostic management of the disease, tomographic imaging (e.g., CT, MRI, and hybrid PET imaging) is considered essential. The qualitative and subjective evaluation of tomographic images is the main approach used to obtain valuable clinical information, although this strategy suffers from both intrinsic and operator-dependent limitations. More recently, advanced imaging techniques have been developed with the aim of overcoming these issues. Such techniques, such as diffusion-weighted MRI and perfusion imaging, were designed for the “in vivo” evaluation of specific biological tissue features in order to describe them in terms of quantitative parameters, which could answer questions difficult to address with conventional imaging alone (e.g., questions related to tissue characterization and prognosis). Furthermore, it has been observed that a large amount of numerical and statistical information is buried inside tomographic images, resulting in their invisibility during conventional assessment. This information can be extracted and represented in terms of quantitative parameters through different processes (e.g., texture analysis). Numerous researchers have focused their work on the significance of these quantitative imaging parameters for the management of CRC patients. In this review, we aimed to focus on evidence reported in the academic literature regarding the application of parametric imaging to the diagnosis, staging and prognosis of CRC while discussing future perspectives and present limitations. While the transition from purely anatomical to quantitative tomographic imaging appears achievable for CRC diagnostics, some essential milestones, such as scanning and analysis standardization and the definition of robust cut-off values, must be achieved before quantitative tomographic imaging can be incorporated into daily clinical practice.

Molecular endoscopic imaging: the future is bright

The prediction and final survival rate of gastrointestinal cancers are dependent on the stage of disease. The ideal would be to detect those gastrointestinal lesions at early stage or even premalignant forms which are difficult to detect by conventional endoscopy with white light optical imaging as they show minimum or no changes in morphological characteristics and are thus left untreated. The introduction of molecular imaging has greatly changed the pattern for detecting gastrointestinal lesions from purely macroscopic structural imaging to the molecular level. It allows microscopic examination of the gastrointestinal mucosa with endoscopy after the topical or systemic application of molecular probes. In recent years, major advancements in endoscopic instruments and specific molecular probes have been achieved. This review focuses on the current status of endoscopic imaging and highlights the application of molecular imaging in gastrointestinal and hepatic disease in the context of diagnosis and therapy based on recently published literature in this field. We also discuss the challenges of molecular endoscopic imaging, its future directions and potential that could have a tremendous impact on endoscopic research and clinical practice in future.

A radiologist's point of view in the presurgical and intraoperative setting of colorectal liver metastases

Multidisciplinary management of patients with metastatic colorectal cancer requires in each phase an adequate choice of the most appropriate imaging modality. The first challenging step is liver lesions detection and characterization, using several imaging modality ultrasound, computed tomography, magnetic resonance and positron emission tomography. The criteria to establish the metastases resectability have been modified. Not only the lesions number and site but also the functional volume remnant after surgery and the quality of the nontumoral liver must be taken into account. Radiologists should identify the liver functional volume remnant and during liver surgical procedures should collaborate with the surgeon to identify all lesions, including those that disappeared after the therapy, using intraoperative ultrasound with or without contrast medium.

Advanced Imaging Techniques in Evaluation of Colorectal Cancer

Imaging techniques are clinical decision-making tools in the evaluation of patients with colorectal cancer (CRC). The aim of this article is to discuss the potential of recent advances in imaging for diagnosis, prognosis, therapy planning, and assessment of response to treatment of CRC. Recent developments and new clinical applications of conventional imaging techniques such as virtual colonoscopy, dual-energy spectral computed tomography, elastography, advanced computing techniques (including volumetric rendering techniques and machine learning), magnetic resonance (MR) imaging-based magnetization transfer, and new liver imaging techniques, which may offer additional clinical information in patients with CRC, are summarized. In addition, the clinical value of functional and molecular imaging techniques such as diffusion-weighted MR imaging, dynamic contrast material-enhanced imaging, blood oxygen level-dependent imaging, lymphography with contrast agents, positron emission tomography with different radiotracers, and MR spectroscopy is reviewed, and the advantages and disadvantages of these modalities are evaluated. Finally, the future role of imaging-based analysis of tumor heterogeneity and multiparametric imaging, the development of radiomics and radiogenomics, and future challenges for imaging of patients with CRC are discussed. Online supplemental material is available for this article. ^©RSNA, 2018.

A guide to multimodal endoscopy imaging for gastrointestinal malignancy - an early indicator

Multimodality imaging is an essential aspect of endoscopic surveillance for the detection of neoplastic lesions, such as dysplasia or intramucosal cancer, because it improves the efficacy of endoscopic surveillance and therapeutic procedures in the gastrointestinal tract. This approach reveals mucosal abnormalities that cannot be detected by standard endoscopy. Currently, these imaging techniques are divided into those for primary detection and those for targeted imaging and characterization, the latter being used to visualize areas of interest in detail and permit histological evaluation. This Review outlines the use of virtual chromoendoscopy, narrow-band imaging, autofluorescence imaging, optical coherence tomography, confocal endomicroscopy and volumetric laser endomicroscopy as new imaging techniques for diagnostic investigation of the gastrointestinal tract. Insights into use of multimodal endoscopic imaging for early disease detection, in particular for pre-malignant lesions, in the oesophagus, stomach and colon are described.