Predict Tumor Growth Research Articles

Dynamic contrast-enhanced and diffusion-weighted MR imaging for predicting tumor growth of sporadic vestibular schwannomas: a prospective study.

Advanced MR imaging, such as diffusion-weighted (DWI) and dynamic contrast-enhanced (DCE) imaging, may provide valuable non-invasive information on intrinsic tumor biology. This study aims to evaluate apparent diffusion coefficient (ADC) and DCE-MRI-derived microvascular parameter values (Ktrans, ve, and vp) as potential imaging predictors for future sporadic vestibular schwannoma (VS) growth. In this prospective cohort study, patients with newly diagnosed unilateral sporadic VS and an initial wait-and-scan strategy were enrolled between January 2021 and January 2023. Patients underwent a single timepoint comprehensive MRI protocol, including DWI and DCE-MRI sequences. The estimated values of ADC, Ktrans, ve, and vp were calculated using established pipelines on a voxelwise basis within the delineated tumor region of interest. Associations of the estimated parameter values with volumetric growth were evaluated in uni- and multivariable logistic regression and survival analyses. Of the 110 analyzed patients, 70 (64%) exhibited growth during follow-up. A significant correlation was primarily observed between the DCE-MRI-derived parameters and VS growth. The combination of mean Ktrans (P<0.001) and ve (P<0.001) tumor values provided an internally validated model with an AUC of 0.85 for growth, yielding a sensitivity of 89% and specificity of 73% at the optimized cutoff value. Only the mean ADC values were found to be significantly higher in shrinking tumors (P=0.04). The strongly significant correlation observed between VS growth and Ktrans and ve tumor values indicates great potential of the non-invasive DCE-MRI for individualized VS management in clinical practice. External validation is needed to further substantiate these findings.

Computational Modeling of Drug Response Identifies Mutant-Specific Constraints for Dosing panRAF and MEK Inhibitors in Melanoma.

This study explores the potential of pre-clinical in vitro cell line response data and computational modeling in identifying the optimal dosage requirements of pan-RAF (Belvarafenib) and MEK (Cobimetinib) inhibitors in melanoma treatment. Our research is motivated by the critical role of drug combinations in enhancing anti-cancer responses and the need to close the knowledge gap around selecting effective dosing strategies to maximize their potential. In a drug combination screen of 43 melanoma cell lines, we identified specific dosage landscapes of panRAF and MEK inhibitors for NRAS vs. BRAF mutant melanomas. Both experienced benefits, but with a notably more synergistic and narrow dosage range for NRAS mutant melanoma (mean Bliss score of 0.27 in NRAS vs. 0.1 in BRAF mutants). Computational modeling and follow-up molecular experiments attributed the difference to a mechanism of adaptive resistance by negative feedback. We validated the in vivo translatability of in vitro dose-response maps by predicting tumor growth in xenografts with high accuracy in capturing cytostatic and cytotoxic responses. We analyzed the pharmacokinetic and tumor growth data from Phase 1 clinical trials of Belvarafenib with Cobimetinib to show that the synergy requirement imposes stricter precision dose constraints in NRAS mutant melanoma patients. Leveraging pre-clinical data and computational modeling, our approach proposes dosage strategies that can optimize synergy in drug combinations, while also bringing forth the real-world challenges of staying within a precise dose range. Overall, this work presents a framework to aid dose selection in drug combinations.

Classification of brain tumor using deep learning at early stage

Early detection and classification of brain tumors are crucial for patient survival. This study proposes a comprehensive deep learning approach for early brain tumor classification using medical imaging data. A diverse dataset encompassing various tumor types, stages, and healthy brain images is utilized. Preprocessing techniques like augmentation and normalization enhance data robustness. A convolutional neural network (CNN) architecture serves as the primary model, leveraging transfer learning from pre-trained models to extract relevant features even with limited data. The training process optimizes hyperparameters to prevent overfitting, and performance is evaluated using metrics like accuracy, precision, recall, F1 score, confusion matrices, and ROC curves on a separate test set. Focusing on early detection, the model explores predicting tumor growth trajectories and identifying subtle pre-tumor patterns, aligning with expert diagnoses and boosting real-world applicability. Ethical and regulatory guidelines are adhered to in data handling. Continuous improvement involves updating the model with new data and monitoring its clinical performance. This research contributes to advancing early tumor classification methods, potentially improving patient outcomes and treatment strategies.

Computational modeling of drug response identifies mutant-specific constraints for dosing panRAF and MEK inhibitors in melanoma.

This study explores the potential of preclinical in vitro cell line response data and computational modeling in identifying optimal dosage requirements of pan-RAF (Belvarafenib) and MEK (Cobimetinib) inhibitors in melanoma treatment. Our research is motivated by the critical role of drug combinations in enhancing anti-cancer responses and the need to close the knowledge gap around selecting effective dosing strategies to maximize their potential. In a drug combination screen of 43 melanoma cell lines, we identified unique dosage landscapes of panRAF and MEK inhibitors for NRAS vs BRAF mutant melanomas. Both experienced benefits, but with a notably more synergistic and narrow dosage range for NRAS mutant melanoma. Computational modeling and molecular experiments attributed the difference to a mechanism of adaptive resistance by negative feedback. We validated in vivo translatability of in vitro dose-response maps by accurately predicting tumor growth in xenografts. Then, we analyzed pharmacokinetic and tumor growth data from Phase 1 clinical trials of Belvarafenib with Cobimetinib to show that the synergy requirement imposes stricter precision dose constraints in NRAS mutant melanoma patients. Leveraging pre-clinical data and computational modeling, our approach proposes dosage strategies that can optimize synergy in drug combinations, while also bringing forth the real-world challenges of staying within a precise dose range.

Fast model calibration for predicting the response of breast cancer to chemotherapy using proper orthogonal decomposition

Constructing digital twins for predictive tumor treatment response models can have a high computational demand that presents a practical barrier for their clinical adoption. In this work, we demonstrate that proper orthogonal decomposition, by which a low-dimensional representation of the full model is constructed, can be used to dramatically reduce the computational time required to calibrate a partial differential equation model to magnetic resonance imaging (MRI) data for rapid predictions of tumor growth and response to chemotherapy. In the proposed formulation, the reduction basis is based on each patient’s own MRI data and controls the overall size of the “reduced order model”. Using the full model as the reference, we validate that the reduced order mathematical model can accurately predict response in 50 triple negative breast cancer patients receiving standard of care neoadjuvant chemotherapy. The concordance correlation coefficient between the full and reduced order models was 0.986 ± 0.012 (mean ± standard deviation) for predicting changes in both tumor volume and cellularity across the entire model family, with a corresponding median local error (inter-quartile range) of 4.36 % (1.22 %, 15.04 %). The total time to estimate parameters and to predict response dramatically improves with the reduced framework. Specifically, the reduced order model accelerates our calibration by a factor of (mean ± standard deviation) 378.4 ± 279.8 when compared to the full order model for a non-mechanically coupled model. This enormous reduction in computational time can directly help realize the practical construction of digital twins when the access to computational resources is limited.

Initial Tumor Size and Narrow-Band Image Findings Estimate Growth Speed in Duodenal Tumors

Introduction: Recently, the detection of superficial non-ampullary duodenal epithelial tumors (SNADETs) including adenomas and superficial duodenal carcinomas has increased. Various endoscopic treatment methods have also been reported for SNADETs, but there are few reports on the natural history. The aim of this study was to analyze factors related to tumor growth and determine the characteristics of SNADETs which need early therapeutic intervention. Methods: A single-center, retrospective study was performed on the medical records of 309 patients with SNADETs who underwent endoscopic or surgical resection between January 2010 and May 2021. Of these, 41 patients who were followed up for more than 1 year by endoscopy were analyzed. The primary outcome was an analysis of the tumor growth speed. Secondary outcomes were the relationship between the tumor growth speed and mucin phenotype, tumor size and findings of magnifying endoscopy with narrow-band imaging (M-NBI). Results: The observation period was 24 months (13–182). Tumor growth speed was 1.1 mm/year (0–21.6). Tumor diameter ≥10 mm at first detection (p = 0.004; odds ratio 19.5 [2.03–186.96]) and mixed type by M-NBI (p = 0.036; odds ratio 9.69 [1.05–89.88]) were identified as risk factors of tumors growing at a rate of ≥3 mm/year. There was no statistically significant difference in the speed of tumor growth between the different mucin immunohistochemical phenotypes. Conclusion: Initial tumor size and findings of M-NBI are useful to predict tumor growth and consider early intervention.

Growth kinetics and predictive factors in renal angiomyolipomas.

Although renal angiomyolipomas (AMLs) are benign lesions, they can grow and cause serious complications. In this study, we aimed to determine the factors affecting the growth of renal AMLs. Patients followed up for AMLs between January 2014 and January 2024 were screened. By accepting 2.5mm/year as the limit for a significant growth rate, the patients were divided into two groups: those with and without significant growth. Demographic characteristics, tumor characteristics, and laboratory parameters, platelet-to-lymphocyte ratio (PLR), neutrophil-to-lymphocyte ratio (NLR), and aspartate aminotransferase-to-alanine aminotransferase (De Ritis) ratio, were compared between the groups. The study included a total of 98 patients. Of the entire cohort, 78.6% were women. Significant growth was detected in nine (9.2%) patients. Multivariate analysis revealed that the baseline scan tumor size, PLR, and De Ritis ratio were significant independent predictors of significant AML growth (p = 0.011, p = 0.017, and p = 0.030, respectively). In the receiver operating characteristic curve analysis, the cut-off value of PLR in predicting significant growth was 131.85 (sensitivity: 77.8%, specificity: 73%, area under the curve [AUC] 0754), while the cut-off value of the De Ritis ratio was 1.33 (sensitivity: 66.7%, specificity: 95.8%, AUC 0.721). Tumor size at the time of initial diagnosis, as well as PLR and De Ritis ratio, were found to be independent predictors of AML growth rate. The use of these factors in patient follow-up has the potential to assist clinicians in predicting tumor growth and related complications.

Deep learning for 4D modeling of glioblastoma multiforme with tumor treating fields (TTFields) therapy.

e14032 Background: Prediction of future tumor growth and recurrence are critical in the management of patients with glioblastoma multiforme (GBM). This study develops a deep learning method for estimating future areas of GBM spread across multiple timepoints using MRI in patients receiving tumor treating fields (TTFields) therapy. Methods: A single institutional database was queried to identify adult patients with histologically confirmed newly diagnosed and/or recurrent GBM undergoing TTFields therapy. For newly diagnosed GBM, patients received TTFields with temozolomide after maximal debulking surgery and chemoradiation therapy. For recurrent GBM, patients received TTFields as monotherapy. For each patient, all serial follow-up MRI exams were obtained, including T1 (pre-/post-contrast), T2, and T2/FLAIR sequences. On each exam, all regions of enhancing tumor core (excluding peritumoral edema, necrotic core, or resection cavity) were delineated and aligned across time points using nonlinear deformable registration. For any given pair of serial exams, a convolutional neural network (CNN) was designed to predict future GBM tumor on the follow-up exam given the precursor exam and time interval between the studies. The model is implemented as a 3D encoder-decoder architecture yielding a dense per-voxel prediction of future tumor probability optimized using a binary cross-entropy loss function. Class weights were used to develop high sensitivity and high positive predictive value (PPV) model variants. To generate final logit scores, time information is concatenated to the penultimate feature map as an additional feature channel, allowing the model to calibrate each estimate based on elapsed time between any pair of exams. Upon convergence, a 4D learned representation allows for prediction of spatial distribution of GBM tumor at any future time point. Results: A total of 123 patients (1112 total MR exams) were identified. For any given single patient, a median of 6 follow-up exams (IQR 2.5-12.5) at a median interval follow-up time of 46 days (IQR 15.75-63 days) between exams was observed. Upon five-fold cross-validation, the model demonstrated a 0.44 Dice score overlap between predicted and true areas of future GBM tumor growth. The high sensitivity model yielded a per-voxel sensitivity of 0.91 (IQR 0.77-0.99) and PPV of 0.26 (IQR 0.17 to 0.32), while the high PPV model yielded a per-voxel sensitivity of 0.14 (IQR 0.00 to 0.46) and PPV of 0.75 (IQR 0.56 to 0.94). Upon visual confirmation, model predictions across incremental time values for any given exam yielded expected gradual growth of tumor over time. Conclusions: A deep learning model can accurately predict future areas of GBM tumor growth in patients receiving TTFields therapy, with optimal performance that may be calibrated for high sensitivity or high PPV based on clinical use case.

A Comparative Analysis of ARX and ANFIS Models for Tumor Growth Prediction Under Single and Multi-agent Chemotherapy.

Despite the advances in oncology and cancer treatment over the past decades, cancer remains one of the deadliest diseases. This study focuses on further understanding the complex nature of cancer by using mathematical tumor modeling to understand, capture as best as possible, and describe its complex dynamics under chemotherapy treatment. Focusing on autoregressive with exogenous inputs, i.e., ARX, and adaptive neuro-fuzzy inference system, i.e., ANFIS, models, this work investigates tumor growth dynamics under both single and combination anticancer agent chemotherapy treatments using chemotherapy treatment data on xenografted mice. Four ARX and ANFIS models for tumor growth inhibition were developed, estimated, and evaluated, demonstrating a strong correlation with tumor weight data, with ANFIS models showing superior performance in handling the multi-agent tumor growth complexities. These findings suggest potential clinical applications of the ANFIS models through further testing. Both types of models were also tested for their prediction capabilities across different chemotherapy schedules, with accurate forecasting of tumor growth up to five days in advance. The use of adaptive prediction and sliding (moving) data window techniques allowed for continuous model updating, ensuring more robust predictive capabilities. However, long-term forecasting remains a challenge, with accuracy declining over longer prediction horizons. While ANFIS models showed greater reliability in predictions, the simplicity and rapid deployment of ARX models offer advantages in situations requiring immediate approximations. Future research with larger, more diverse datasets and by exploring varying model complexities is recommended to improve the models' reliability and applicability in clinical decision-making, thereby aiding the development of personalized chemotherapy regimens.

Brain Tumor Growth Prediction Based on Segmentation Using Resolution Convolution Network

Abstract: This study presents a novel approach to brain tumor growth prediction using Resolution Convolution Network (RCN) segmentation. The method involves precise segmentation of brain tumor regions from magnetic resonance imaging (MRI) scans followed by application of RCN to predict growth. The RCN architecture integrates high-resolution convolutional layers, which facilitates the extraction of detailed features that are crucial for accurate segmentation and subsequent prediction. Performance evaluation is performed on a comprehensive dataset of MRI brain scans

3D extension of spatial patterning analysis of cellular ensembles (SPACE) to extract parameters for dynamic tumor modeling

Abstract Investigating the cellular interactions within three-dimensional (3D) tumor microenvironments is essential for understanding tumor growth and progression. Building upon the success of our previous computational framework, SPACE, designed for spatial analysis of high-plex tissue images, we are developing 3D-SPACE. This innovation extends spatial pattern analysis to 3D volumetric tissue images. Volumetric images more accurately capture nuances of cellular arrangements and tissue structures, offering a more realistic representation of the tumor microenvironment. Applying 3D-SPACE to analyze full 3D tumor datasets has the potential to reveal crucial parameters for understanding tumor outcomes. This approach not only elucidates previously undiscovered spatial patterns but also allows for the quantification of parameters such as contact probabilities between different cell types or the killing efficiencies of key effector cells in situ. These spatially-derived parameters can then be used in predictive equations that model and forecast tumor growth based on cellular interactions. Using a 3D tumor model and 3D-SPACE, we can quantify cell interactions and develop predictive equations for tumor growth dynamics, with the aim of deriving a predictive reproduction number (R₀) for tumor growth. Working towards a predictive R₀ for tumor growth offers a valuable approach for predicting tumor growth outcomes. This work was supported in part by the Intramural Research Program of NIAID, NIH.

Current Status and Future Directions: The Application of Artificial Intelligence/Machine Learning for Precision Medicine.

Technological innovations, such as artificial intelligence (AI) and machine learning (ML), have the potential to expedite the goal of precision medicine, especially when combined with increased capacity for voluminous data from multiple sources and expanded therapeutic modalities; however, they also present several challenges. In this communication, we first discuss the goals of precision medicine, and contextualize the use of AI in precision medicine by showcasing innovative applications (e.g., prediction of tumor growth and overall survival, biomarker identification using biomedical images, and identification of patient population for clinical practice) which were presented during the February 2023 virtual public workshop entitled "Application of Artificial Intelligence and Machine Learning for Precision Medicine," hosted by the US Food and Drug Administration (FDA) and University of Maryland Center of Excellence in Regulatory Science and Innovation (M-CERSI). Next, we put forward challenges brought about by the multidisciplinary nature of AI, particularly highlighting the need for AI to be trustworthy. To address such challenges, we subsequently note practical approaches, viz., differential privacy, synthetic data generation, and federated learning. The proposed strategies - some of which are highlighted presentations from the workshop - are for the protection of personal information and intellectual property. In addition, methods such as the risk-based management approach and the need for an agile regulatory ecosystem are discussed. Finally, we lay out a call for action that includes sharing of data and algorithms, development of regulatory guidance documents, and pooling of expertise from a broad-spectrum of stakeholders to enhance the application of AI in precision medicine.

Gompertzian tumor cell growth using a discrete dynamic model

Some benign tumors (made up of non-cancer cells) can become malignant (made up of cancer cells) over time, such as in our skin or colon. The dynamic interaction between cancer and non-cancer cells in tumors becomes interesting to analyze to predict tumor growth. This paper uses a predator-prey model to interpret the dynamic tumor growth, which uses cancer cells as a predator and non-cancer cells as prey. A Gompertzian growth function completes the model to describe natural tumor growth and add a treatment parameter. First, we investigate the stability criteria of each model equilibrium using an analytical approach. The eigenvalues ​​obtained in the stability analysis are used to construct a discrete dynamic model using the Non-Standard Finite Difference (NSFD) scheme. This scheme is implemented numerically in Python and is highly accurate since the maximum difference with the 4th Runge Kutta scheme is 0.42%. Finally, sensitivity parameter analysis shows that intensive treatment mostly affects the dynamic of the tumor cell population. The related party can use the simulation and analysis results to plan the proper and effective treatment to minimize tumor growth.

Prediction of tumor growth at the microscopic level through mathematical modeling

All articles must contain an abstract. The abstract text should be formatted using 10 point Times or Times New Roman and indented 25 mm from the left margin. Set the pre-paragraph to 0 pounds and the post-break to 22.7 pounds. Starting on the same page as the abstract. The abstract should give readers concise information about the content of the article and indicate the main results obtained and conclusions drawn. The abstract is not part of the text and should be complete in itself; no table numbers, figure numbers, references or displayed mathematical expressions should be included. It should be suitable for direct inclusion in abstracting services and should not normally exceed 200 words in a single paragraph. Since contemporary information-retrieval systems rely heavily on the content of titles and abstracts to identify relevant articles in literature searches, great care should be taken in constructing both. To gain valuable insights into the growth of microscopic tumors, which is crucial for advancing cancer research, scientists employ mathematical modeling techniques. These models help researchers to comprehend how cells interact, proliferate, and organize spatially within tumors, ultimately aiding in developing better treatment strategies, including personalized medicine. A recent study introduced a novel approach using the Occam Plausibility Algorithm (OPAL) for modeling tumor growth. OPAL is a probabilistic framework adept at generating and refining hypotheses based on complex datasets. In this study, researchers integrated two existing models, the Proliferation Invasion Model (PIM) and the Mathematical Phase Field Model (MPFM), to create a comprehensive understanding of microscopic tumor growth dynamics. In this research, we have employed a literature review methodology to explore how researchers utilize three various mathematical modeling techniques, including the Occam Plausibility Algorithm (OPAL), to delve deeply into the growth dynamics of microscopic tumors. The conclusions drawn emphasize the significant potential of mathematical modeling in advancing scientists understanding of microscopic tumor growth.

Intratumor heterogeneity: models of malignancy emergence and evolution

Intratumor heterogeneity: models of malignancy emergence and evolution

NIMG-86. ESTIMATING BRAIN TUMOR RELATED EDEMA FROM NON-INVASIVE IMAGING IN A CLINICALLY RELEVANT TIMELINE

Abstract Image-based mathematical modeling is emerging as an advantageous tool to integrate into the care routine for patients with brain tumors. These tools can be used for a variety of tasks including predicting tumor growth, indicating spatial genomic alterations, and guiding radiation therapy. Often, the methods to produce these maps are developed in silos, using cohorts of retrospective patients and switching to a per patient map calculation can be an afterthought. Automatic pipelines are needed to enable creating mathematical maps in a clinically-relevant workflow. To this end, we developed a pipeline to create an image-based mathematical map of edema abundance based on T2-weighted (T2W) magnetic resonance imaging that includes preprocessing steps, abnormality and tissue type segmentation, brain extraction, map generation, and conversion between image file formats. The python-based pipeline was designed to take in T1-weighted (T1W), T1W with gadolinium contrast (T1Gd), T2W, and fluid attenuated inversion recovery (FLAIR) DICOM images for a single patient as input and convert them to NIfTIs for downstream steps. The pipeline preprocessed images by correcting bias field fluctuations using the N4 algorithm and normalizing intensities. The T1W, T1Gd, and FLAIR are then registered to the T2W. T1Gd and FLAIR are used in a deep learning algorithm to delineate the brain and abnormality including enhancing tumor, necrosis, and peripheral edema. Normal tissue segmentations are performed using the statistical parametric mapping (SPM) segmentation routine. The segmentations and preprocessed images are all utilized in the edema mathematical model. The resulting edema map is then reformatted as a DICOM and output for further use by clinicians and clinical imaging systems. Our future goals include integrating our pipeline in a clinical workstation and sending final maps to our hospital’s picture archiving and communication system (PACS) to allow for deployment of these models for patient care in a clinically efficient timeline.

THU087 Natural History Of Nonfunctioning Pituitary Macroadenomas Followed Without Intervention

Abstract Disclosure: I. Ayalon-Dangur: None. E. Robenshtok: None. H. Duskin-Bitan: None. G. Tsvetov: None. A. Gorshtein: None. A. Akirov: None. I. Shimon: None. Objective: The treatment strategy for non-functioning pituitary adenomas (NFPA) includes surgery, radiotherapy, medication, or follow-up. Data on non-operated patients with macroadenomas are sparse. This study sought to investigate the natural history of NFPAs followed without treatment. Design and patients: The database of a tertiary medical center was retrospectively searched (1995-2020) for adult patients with NFPA ≥10 mm who were naïve to surgery or medical treatment and were followed for ≥12 months after diagnosis. Patients with chiasmal threat were excluded. Follow-up was terminated if the patient underwent surgery, received cabergoline, or was lost to follow-up. Measurements: Data on tumor characteristics and size on imaging by MRI, symptoms including visual disturbances, and pituitary hormone levels were collected from the medical files. Tumor growth was defined as a maximal increase of &amp;gt;2 mm in diameter. Results: The cohort included 49 patients, 30 male and 19 female, of mean age 68.0±12.0 years. Median tumor size at diagnosis was 16 mm (range, 10-35), and median duration of follow-up was 3 years (range, 1-23(. Tumor’s size increased in 16 patients (33%), by a mean of 5.1±4.4 mm, and decreased in 10 patients (20%), by a mean of 3.5±1.3 mm; in 23 patients, the tumor remained stable. Overall, 33 patients (67%) were observed without intervention, 3 underwent surgery, and 13 received cabergoline. None of the parameters evaluated, including age, gender, baseline tumor size, tumor invasiveness, visual disturbances, or hypopituitarism at diagnosis, predicted tumor growth. Conclusions: Observation of NFPAs without surgery or medication is a reasonable approach in selected patients. In our study, there was no association of patient- or tumor-related parameters with tumor growth. Presentation: Thursday, June 15, 2023

Features of aminothiol metabolism and progression of breast cancer

Background. Imbalance of aminothiol metabolism is a potential risk factor for malignant transformation of cells and caner development, including breast cancer, which is the most commonly diagnosed cancer in the world. The purpose of the study was to summarize the available data on the characteristics of thiol metabolism as one of the factors contributing to the progression of breast cancer. Material and Methods. Data were searched from 1999 to 2022 using the Web of Science, Scopus, MedLine, The Cochrane Library, PubMed databases, which made it possible to assess the role of thiol-dependent metabolic disturbances in the regulation of tissue redox balance in breast cancer genesis. Results. The review considers the results of both our own data and international studies on breast cancer, which suggest that an imbalance of thiol compounds necessary to maintain a moderately reducing cellular environment that counteracts oxidative stress during cellular metabolism and detoxifcation under conditions of tumor progression can provoke reprogramming of the leading links of antiblastoma resistance, contributing to cancer progression. Conclusion. A more detailed study of the mechanisms of aminothiol metabolism in breast cancer emphasizes their particular importance for stabilizing the cellular genome and providing antitoxic protection of the cell, as well as understanding the important role of thiols as a coordination center in redox signaling. Disturbances at any stage of thiol metabolism can play an etiological role in oncogenetic pathologies, while the role of thiols as signaling molecules and the regulation of their metabolism should not be generalized for the entire group of diseases. Determination of serum markers of the redox state in patients with breast cancer, especially during antitumor therapy, can serve for an objective assessment of the effectiveness of treatment and the adaptive capabilities of the body, as well as predicting tumor growth and optimizing the program for screening and preventing cancer.

Utility of Targeted Positron Emission Tomography Imaging to Predict Schwannoma Growth in a Murine Tumor Model.

To identify if targeted positron emission tomography (PET) imaging with radiolabeled antibodies can predict tumor growth rate and ultimate tumor size in a murine flank schwannoma model. Animal research study. Rat schwannoma cells were cultured and implanted into 40 athymic nude mice. Once tumors reached 5 mm in diameter, 30 mice were injected with zirconium-89 labeled antibodies (HER2/Neu, vascular endothelial growth factor receptor 2 (VEGFR2), or IgG isotype). PET/CT was performed, and standardized uptake values (SUV) were recorded. Tumors were serially measured until mice were sacrificed per IACUC protocol. Statistical analysis was performed to measure correlations between SUV values, tumor size, and growth. Mean tumor sizes in mm3 on Day 0 were 144 ± 162 for anti-HER2/Neu, 212 ± 247 for anti-VEGFR2, and 172 ± 204 for IgG isotype groups respectively. Mean growth rates in mm3 /day were 531 ± 250 for HER2, 584 ± 188 for VEGFR2, and 416 ± 163 for the IgG isotype group. For both initial tumor size and growth rates, there was no significant difference between groups. There were significant correlations between maximum tumor volume and both the SUV max in the HER2 group (p = 0.0218, R2 = 0.5020), and we observed significant correlations between growth rate, and SUV values (p = 0.0156, R2 = 0.5394). Respectively, in the anti-VEGFR2 group, there were no significant correlations. In a murine schwannoma model, immunotargeted PET imaging with anti-HER2/Neu antibodies predicted tumor growth rate and final tumor size. Laryngoscope, 134:1372-1380, 2024.

TGM-Nets: A deep learning framework for enhanced forecasting of tumor growth by integrating imaging and modeling

Prediction and uncertainty quantification of tumor progression are vital in clinical practice, i.e., disease prognosis and decision-making on treatment strategies. In this work, we propose TGM-Nets, a deep learning framework that combines bioimaging and tumor growth modeling (TGM) for enhanced prediction of tumor growth. This proposed framework, developed based on physics-informed neural networks (PINNs), is capable of integrating the TGM and sequential observations of tumor morphology for patient-specific prediction of tumor growth. The novelties of the design of TGM-Nets include the employment of Fourier layers to extract the features of the input images as well as the utilization of sequential learning and fine-tuning with physics for extrapolation to improve the prediction accuracy. The validity of TGM-Nets for tumor growth forecasting is verified by testing the model performance on synthetic and in-vitro datasets, respectively. Our results show that the TGM-Nets not only can track the growth rates of the mild and aggressive tumors but also capture their detailed morphological features within and outside the training domain. In particular, TGM-Nets can be used to predict the long time dynamics of tumor growth in mild and aggressive cases. Our results show that the parameters inferred from the TGM-Nets can be used for long-time prediction for up to 4 months with a maximum error of ∼ 4%. We also systematically study the effects of the number of training points and noisy data on the performance of TGM-Nets as well as quantify the uncertainty of the model predictions. We show that TGM-Nets can integrate the biomedical images to predict the growth of the in-vitro cultured pancreatic cancer cells and identify the associated growth rates, demonstrating the possibilities of using TGM-Nets in clinical practice. In summary, we propose a new deep learning model that combines imaging and TGM to improve the current approaches for predicting tumor growth and thus provide an advanced computational tool for patient-specific tumor prognosis.