Prediction Of Response Research Articles

Abstract C053: Pancreatic cancer patient-derived organoids, preclinical tools for therapeutic profiling, patient response prediction, and tumor evolution

Abstract Pancreatic cancer (PC) is characterized by an aggressive biology and an exceptionally high tumor heterogeneity that causes considerable variations in response to chemotherapies, targeted agents, and immunotherapies. Patient-derived organoids (PDOs) accurately reflect parental tumor biological and molecular features and may represent powerful preclinical avatars to predict drug response and support clinical decision-making. We generated a living organoid biobank of >100 PC PDO lines from treatment-naïve and pretreated primary tumor and metastases with a reliable efficacy (60.8%), and previously developed a pharmacotyping-guided prediction model to prognosticate patient therapy response in an initial feasibility trial (Beutel AK et al, 2021). Real-life chemotherapy responses in 46 patients (for a total of 93 therapy lines) matched to pharmacotyping-informed predictions with overall 73.1% accuracy, 85.4% sensitivity (responsiveness), and 63.5% specificity (non-responsiveness). Overall, our model allowed a successful drug-response prediction in naïve patients with an accuracy of 73.0% and 70.0% for first and second-line regimens. Prediction power was nevertheless lower in pretreated and heavily pretreated patients with a precision of 68.2% and 50.0% for subsequent chemotherapy lines, respectively. Interestingly, PDO-based pharmacotyping also precisely predicted patient clinical outcome in the neoadjuvant and adjuvant settings, with respective accuracies of 83.3% and 71.4%. Retrospective analysis of patient clinical data in palliative setting finally showed that the administration of a regimen predicted to be efficient ultimately translated into a significantly longer progression-free survival. Implementing automation and drug screening miniaturization to our workflow also significantly enhanced our process capacity, reduced time before pharmacotyping, and improved compliance to internal quality controls. Tracking clonal evolution in longitudinal biopsies (a set of seven PDO pairs derived from the same patients at two distinct timepoints) revealed lower mutational burden and therapy-driven genetic alterations upon progression. Notably, this approach revealed a CHEK2-mutated patient responded over time to PARP inhibitor maintenance therapy, aligning with our predictions and underscoring the robustness of our method. Single nucleus-resolved RNA and ATAC-seq analysis of a unique longitudinally-collected case uncovered transcriptomic and epigenetic dynamics associated with an oncogenic FGFR2 fusion and with a subsequent resistance-mediating mutation upon targeted treatment with FGFR2 inhibitors. Particularly, FGFR2 constitutive activation correlated with aberrant epigenetic trace and transcription programs of downstream targets as PI3K-AKT and RAS family members and of RNA polymerase II transcription machinery. Here, we report a robust and clinically-relevant preclinical tool for drug-response prediction, one more step towards a PDO therapeutic profiling-guided personalized medicine in clinical routine. Citation Format: Johann Gout, Yazid J Resheq, Jessica Lindenmayer, Julian D Schwab, Elodie Roger, Johann M Kraus, Thomas Ettrich, Alica K Beutel, Lukas Perkhofer, Hans A Kestler, Thomas Seufferlein, Alexander Kleger. Pancreatic cancer patient-derived organoids, preclinical tools for therapeutic profiling, patient response prediction, and tumor evolution [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research; 2024 Sep 15-18; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84(17 Suppl_2):Abstract nr C053.

Seismic response prediction method of train-bridge coupled system based on convolutional neural network-bidirectional long short-term memory-attention modeling

Seismic response prediction is crucial for the safety analysis of train-bridge coupled systems. However, due to the complexity, suddenness, and high-risk nature of earthquakes, there are strong nonlinear relationships among different parts of bridges, making it challenging to express their spatial correlations using analytical models and traditional neural networks. To address this, this paper establishes a ballast track shaker scaling model and employs the grating monitoring measurement method to construct a spatial quasi-distributed monitoring system for the ballast track, thereby collecting seismic strain responses of the train-bridge coupled system under various seismic conditions. A hybrid neural network method is proposed for predicting the seismic responses of the train-bridge coupled system. This hybrid neural network integrates the features of a Convolutional Neural Network (CNN), Bidirectional Long Short-Term Memory Neural Network (BiLSTM), and the attention mechanism, thereby termed the CNN-BiLSTM-attention hybrid neural network. The model was validated using strain responses from 54 seismic scenarios. The results indicate that the model has a Mean Absolute Error (MAE) of 0.2349 and a coefficient of determination (R2) of 0.9446. Comparing the prediction results with those from RNN and LSTM models, it was found that the CNN effectively extracts features under various seismic parameters, while the BiLSTM better captures the temporal information of the strain responses, ensuring effective prediction regardless of the magnitude of strain responses. Therefore, the CNN-BiLSTM-attention hybrid neural network model is recommended for predicting seismic response.

State space model-based Runge–Kutta gated recurrent unit networks for structural response prediction

State space model-based Runge–Kutta gated recurrent unit networks for structural response prediction

Monte Carlo simulation of spatial frequency domain imaging for breast tumors during compression.

Near-infrared optical imaging methods have shown promise for monitoring response to neoadjuvant chemotherapy (NAC) for breast cancer, with endogenous contrast coming from oxy- and deoxyhemoglobin. Spatial frequency domain imaging (SFDI) could be used to detect this contrast in a low-cost and portable format, but it has limited imaging depth. It is possible that local tissue compression could be used to reduce the effective tumor depth. To evaluate the potential of SFDI for therapy response prediction, we aim to predict how changes to tumor size, stiffness, and hemoglobin concentration would be reflected in contrast measured by SFDI under tissue compression. Finite element analysis of compression on an inclusion-containing soft material is combined with Monte Carlo simulation to predict the measured optical contrast. When the effect of compression on blood volume is not considered, contrast gain from compression increases with the size and stiffness of the inclusion and decreases with the inclusion depth. With a model of reduction of blood volume from compression, compression reduces imaging contrast, an effect that is greater for larger inclusions and stiffer inclusions at shallower depths. This computational modeling study represents a first step toward tracking tumor changes induced by NAC using SFDI and local compression.

Deep Learning-Based Approaches Using Medical Imaging for Therapy Response Prediction in Breast Cancer: A Systematic Literature Review

Deep Learning-Based Approaches Using Medical Imaging for Therapy Response Prediction in Breast Cancer: A Systematic Literature Review

Simultaneous identification of propeller fluctuating forces and unknown parameters based on a Bayesian inversion approach

Accurate quantification of propeller excitation is essential for effective vibro-acoustic analysis and mitigation in marine vessels. This study introduces a Bayesian statistical framework that indirectly reconstruct uncertain forces and system parameters by fusing available prior information, forward modeling and measured vibration responses through Bayesian theorem. The Markov Chain Monte Carlo (MCMC) algorithm, enhanced with Polynomial Chaos Kriging (PC-Kriging) surrogates, is employed to facilitate efficient estimation of likelihood functions and robust uncertainty propagation, providing posterior probabilistic estimates and credible intervals for the identified forces. Additionally, the framework incorporates model class selection based on model evidence to determine the most plausible empirical spectrum model for the propeller broadband thrust spectrum. Specifically, this study proposes and evaluates two candidate empirical spectrum models, the Ornstein–Uhlenbeck (OU)-Gaussian and OU-Weibull models, which effectively capture the frequency–amplitude characteristics of unsteady thrust. Numerical results demonstrate substantial parameter uncertainty reduction, accurate thrust spectrum reconstruction, and effective quantification of statistical properties. Moreover, the OU-Gaussian model is revealed to be more plausible for characterizing the propeller thrust spectrum and response prediction. This framework may offer a practical solution for propeller excitation identification, enabling rapid vibro-acoustic performance assessments of marine vessels.

Application of a novel machine learning model for the prediction and optimization of anti-blast performance of sandwich honeycomb blast walls

Sandwich honeycomb blast wall with the superior energy absorption characteristics could be applied to anti-blast protection of offshore oil and gas platform. Recently, machine learning technique, as an alternative to traditional Finite Element Analysis (FEA), is becoming a powerful tool in understanding structural performance. The aim of this study is to propose a novel data-driven model for fast predictions of structure response of sandwich blast wall under explosion, including the maximum deformation δmax, the maximum internal energy Emax and the energy absorption ratio RE. FEA was firstly conducted to generate sample data for the proposed model by using LS-DYNA software. An improved artificial electric field algorithm (IAEFA) was combined with artificial neural network (ANN) to develop an IAEFA-ANN model. In case studies, the prediction results of IAEFA-ANN approximate the results of FEA well, and comparison results of 9 different ANN models demonstrate that IAEFA-ANN model has the best performance. Then, the effects of explosion parameters (including explosive charge mass Me, stand-off distance d and concave angles θ) on structural response prediction results were investigated by using the proposed model. Moreover, the IAEFA-ANN model was applied to conducting optimal design of the blast wall to obtain optimum anti-blast performance.

Quantitative SPECT/CT Metrics in Early Prediction of [177Lu]Lu-DOTATATE Treatment Response in Gastroenteropancreatic Neuroendocrine Tumor Patients.

Our objective is to explore quantitative imaging markers for early prediction of treatment response in patients with gastroenteropancreatic neuroendocrine tumors (GEP-NETs) undergoing [177Lu]Lu-DOTATATE therapy. By doing so, we aim to enable timely switching to more effective therapies in order to prevent time-resource waste and minimize toxicities. Methods: Patients diagnosed with unresectable or metastatic, progressive, well-differentiated, receptor-positive GEP-NETs who received 4 sessions of [177Lu]Lu-DOTATATE were retrospectively selected. Using SPECT/CT images taken at the end of treatment sessions, we counted all visible tumors and measured their largest diameters to calculate the tumor burden score (TBS). Up to 4 target lesions were selected and semiautomatically segmented. Target lesion peak counts and spleen peak counts were measured, and normalized peak counts were calculated. Changes in TBS (ΔTBS) and changes in normalized peak count (ΔnPC) throughout treatment sessions in relation to the first treatment session were calculated. Treatment responses were evaluated using third-month CT and were binarized as progressive disease (PD) or non-PD. Results: Twenty-seven patients were included (7 PD, 20 non-PD). Significant differences were observed in ΔTBSsecond-first, ΔTBSthird-first, and ΔTBSfourth-first (where second-first, third-first, and fourth-first denote scan number between the second and first, third and first, and fourth and first [177Lu]Lu-DOTATATE treatment cycles), respectively) between the PD and non-PD groups (median, 0.043 vs. -0.049, 0.08 vs. -0.116, and 0.109 vs. -0.123 [P = 0.023, P = 0.002, and P < 0.001], respectively). ΔnPCsecond-first showed significant group differences (mean, -0.107 vs. -0.282; P = 0.033); ΔnPCthird-first and ΔnPCfourth-first did not reach statistical significance (mean, -0.122 vs. -0.312 and -0.183 vs. -0.405 [P = 0.117 and 0.067], respectively). At the optimal threshold, ΔTBSfourth-first exhibited an area under the curve (AUC) of 0.957, achieving 100% sensitivity and 80% specificity. ΔTBSsecond-first and ΔTBSthird-first reached AUCs of 0.793 and 0.893, sensitivities of 71.4%, and specificities of 85% and 95%, respectively. ΔnPCsecond-first, ΔnPCthird-first, and ΔnPCfourth-first showed AUCs of 0.764, 0.693, and 0.679; sensitivities of 71.4%, 71.4%, and 100%; and specificities of 75%, 70%, and 35%, respectively. Conclusion: ΔTBS and ΔnPC can predict [177Lu]Lu-DOTATATE response by the second treatment session.

E-PINN: A fast physics-informed neural network based on explicit time-domain method for dynamic response prediction of nonlinear structures

The physics-informed neural networks (PINNs), serving as the surrogate models, have emerged in the dynamic response prediction of nonlinear systems. However, in the conventional PINN model, the global differential equation of motion of the nonlinear system is directly integrated into the loss function, which requires the network to capture the intricate global dynamic evolution mechanism of the system and thus poses significant challenges for network learning. In this study, inspired by the explicit time-domain method (ETDM), a novel PINN based on ETDM, called E-PINN, is proposed to address the challenges arising from the network learning of the existing PINNs. A lightweight long short-term memory (LSTM) module is trained to learn the nonlinear evolution mechanism of restoring forces, while a single convolutional layer (SCL) module is utilized to reflect the linear evolution mechanism of the primary structure under the combined action of the external excitations and the nonlinear restoring forces. The weights of the SCL module can be directly retrieved from the discrete convolutional formulation of ETDM, and only the parameters in the LSTM module need to be optimized through a loss function solely based on the nonlinear restoring forces, thereby enabling easy network learning of the proposed E-PINN by decoupling the linear and nonlinear evolution mechanisms embedded in the nonlinear system. Three numerical examples are investigated by the present approach, and the results demonstrate the superior training efficiency and prediction accuracy of E-PINN in dynamic response prediction of nonlinear systems.

Image-Based Artificial Intelligence in Psoriasis Assessment: The Beginning of a New Diagnostic Era?

Psoriasis, a chronic inflammatory skin disease, affects millions of people worldwide. It imposes a significant burden on patients' quality of life and healthcare systems, creating an urgent need for optimized diagnosis, treatment, and management. In recent years, image-based artificial intelligence (AI) applications have emerged as promising tools to assist physicians by offering improved accuracy and efficiency. In this review, we provide an overview of the current landscape of image-based AI applications in psoriasis. Emphasis is placed on machine learning (ML) algorithms, a key subset of AI, which enable automated pattern recognition for various tasks. Key AI applications in psoriasis include lesion detection and segmentation, differentiation from other skin conditions, subtype identification, automated area involvement, and severity scoring, as well as personalized treatment selection and response prediction. Furthermore, we discuss two commercially available systems that utilize standardized photo documentation, automated segmentation, and semi-automated Psoriasis Area and Severity Index (PASI) calculation for patient assessment and follow-up. Despite the promise of AI in this field, many challenges remain. These include the validation of current models, integration into clinical workflows, the current lack of diversity in training-set data, and the need for standardized imaging protocols. Addressing these issues is crucial for the successful implementation of AI technologies in clinical practice. Overall, we underscore the potential of AI to revolutionize psoriasis management, highlighting both the advancements and the hurdles that need to be overcome. As technology continues to evolve, AI is expected to significantly improve the accuracy, efficiency, and personalization of psoriasis treatment.

Creating cell-specific computational models of stem cell-derived cardiomyocytes using optical experiments.

Human induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) have gained traction as a powerful model in cardiac disease and therapeutics research, since iPSCs are self-renewing and can be derived from healthy and diseased patients without invasive surgery. However, current iPSC-CM differentiation methods produce cardiomyocytes with immature, fetal-like electrophysiological phenotypes, and the variety of maturation protocols in the literature results in phenotypic differences between labs. Heterogeneity of iPSC donor genetic backgrounds contributes to additional phenotypic variability. Several mathematical models of iPSC-CM electrophysiology have been developed to help to predict cell responses, but these models individually do not capture the phenotypic variability observed in iPSC-CMs. Here, we tackle these limitations by developing a computational pipeline to calibrate cell preparation-specific iPSC-CM electrophysiological parameters. We used the genetic algorithm (GA), a heuristic parameter calibration method, to tune ion channel parameters in a mathematical model of iPSC-CM physiology. To systematically optimize an experimental protocol that generates sufficient data for parameter calibration, we created in silico datasets by simulating various protocols applied to a population of models with known conductance variations, and then fitted parameters to those datasets. We found that calibrating to voltage and calcium transient data under 3 varied experimental conditions, including electrical pacing combined with ion channel blockade and changing buffer ion concentrations, improved model parameter estimates and model predictions of unseen channel block responses. This observation also held when the fitted data were normalized, suggesting that normalized fluorescence recordings, which are more accessible and higher throughput than patch clamp recordings, could sufficiently inform conductance parameters. Therefore, this computational pipeline can be applied to different iPSC-CM preparations to determine cell line-specific ion channel properties and understand the mechanisms behind variability in perturbation responses.

Robust model averaging prediction of longitudinal response with ultrahigh-dimensional covariates

Robust model averaging prediction of longitudinal response with ultrahigh-dimensional covariates

Castration-resistant prostate cancer monitoring by cell-free circulating biomarkers.

Prostate cancer is the second leading cause of male cancer-related deaths in Western countries, which is predominantly attributed to the metastatic castration-resistant stage of the disease (CRPC). There is an urgent need for better prognostic and predictive biomarkers, particularly for androgen receptor targeted agents and taxanes. We have searched the PubMed database for original articles and meta-analyses providing information on blood-based markers for castration-resistant prostate cancer monitoring, risk group stratification and prediction of therapy response. The molecular markers are discussed along with the standard clinical parameters, such as prostate specific antigen, lactate dehydrogenase or C-reactive protein. Androgen receptor (AR) alterations are commonly associated with progression to CRPC. These include amplification of AR and its enhancer, point mutations and splice variants. Among DNA methylations, a novel 5-hydroxymethylcytosine activation marker of TOP2A and EZH2 has been identified for the aggressive disease. miR-375 is currently the most promising candidate among non-coding RNAs and sphingolipid analysis has recently emerged as a novel approach. The promising biomarkers have the potential to improve the care of metastatic prostate cancer patients, however, they need further validation for routine implementation.

Contribution of geometrical infill pattern on mechanical behaviour of 3D manufactured polylactic acid specimen: Experimental and numerical analysis

Additive manufacturing is regarded as a very efficient fabrication technique since it permits the manufacturing of any three-dimensional product. The present work determines the effect of various infill pattern on the mechanical properties in term of tensile strength, yield strength and hardness of poly-lactic acid (PLA) samples fabricated by fused deposition modeling method. The mechanical behaviour of the 3D-printed PLAs investigated using dog-bone specimens with six distinct infill patterns: line, triangle, tri-hexagon, cubic, octet, and gyroid. The mechanical characteristics were evaluated using the uniaxial tensile test and shore D type hardness tester. The strain and deformation criteria were employed to substantiate the ductile and brittle characteristics. The fractural surface morphology analyzed using the field emission scanning electron microscope. Nonlinear Finite Element Analysis (FEA) was employed to simulate the uniaxial tensile test and establish a Yeoh third order hyperelastic material model for the predictions of the stress-strain response. This model is chosen for its precise ability to predict the nonlinear stress-strain responses for significant deformation and is crucial in applications that involve high degrees of flexibility and elasticity, such as in tire modeling, polymer and elastomer analysis, sports equipment designing, 3D printed components etc. Results revealed that the cubic infill had a maximum tensile strength 32.648 ± 1.42 MPa and octet infill had a minimum tensile strength 22.373 ± 0.79 MPa. The majority of experimental data indicated a brittle behaviour for line-infilled, but triangular, trihexagonal, cubic, octet, and gyroid infill patterns demonstrated ductile behaviour. In comparison to other geometrical infills, cubic shown relatively superior mechanical responses. Consequently, the geometrical infill effect plays a significant role in finding the appropriate mechanical property for industrial applications. The developed material model possesses potential utility in non-linear FEA investigations pertaining to 3D printed PLA objects that are predicted to sustain tensile strength.

The metabolic profiles of endogenous and exogenous substances in a poor metabolizer of humanized CYP2D6 model

Background: Species differences in CYP2D6 drug metabolism complicate the extrapolation of in vivo pharmacokinetic data to humans and impact the prediction of drug responses. This study aimed to develop an in vivo model to predict human responses to CYP2D6 metabolized compounds and to evaluate medication risks and disease development.Methods: We used embryonic stem cell (ES) targeting and CRISPR-Cas9 technology to create a humanized CYP2D6 mouse model by inserting the human wild-type CYP2D6 gene and knocking out the mouse Cyp2d locus. Metoprolol was used as the substrate probe to examine the pharmacokinetic properties of exogenous substances, tissue distribution, and in situ metabolism of CYP2D6. Untargeted and quantitative metabolomics analyses compared endogenous substance metabolism between different species of CYP2D6 enzymes.Results: No significant differences in CYP2D6 homologous protein distribution and expression of primary metabolic organs were found between humanized CYP2D6 mice and wild-type (WT) mice. The activity and metabolic capacity of CYP2D6 in humanized mice were substantially lower than homologous Cyp2d22 of WT mice in metabolizing metoprolol. The levels of several glycerolipids and glycerophospholipid-related metabolites were down-regulated in humanized CYP2D6 mice. Triglyceride TG (14:0_22:6_22:6) was significantly downregulated in male and female humanized mice, suggesting a strong association with reduced CYP2D6 activity.Conclusions: This study established a robust animal model to investigate human CYP2D6-mediated metabolic profiles of exogenous and endogenous compounds, predict medication risks, and explore the potential roles of CYP2D6 in organ-specific toxicity and disease development.

The Role of Circulating Tumor DNA in Ovarian Cancer.

Ovarian cancer is the deadliest of all gynecological diseases because its diagnosis and treatment still pose many problems. Surgical excision, hormone therapy, radiation, chemotherapy, or targeted therapy for eradicating the main tumor and halting the spread of metastases are among the treatment options available to individuals with ovarian cancer, depending on the disease's stage. Tumor DNA that circulates in a patient's bodily fluids has been studied recently as a possible novel biomarker for a number of cancers, as well as a means of quantifying tumor size and evaluating the efficacy of cancer therapy. The most significant alterations that we could find in the ctDNA of ovarian cancer patients-such as chromosomal instability, somatic mutations, and methylation-are discussed in this review. Additionally, we talk about the utility of ctDNA in diagnosis, prognosis, and therapy response prediction for these patients.

Influence of reinforcement reductions on the response of reinforced concrete half-joints

Reinforced concrete half-joints are susceptible to deterioration mechanisms primarily due to the leakage of surface water, often contaminated with chlorides from de-icing salts, through the expansion joint. Due to the joint geometric form, favourable conditions can be created inside the joint for the corrosion of steel reinforcement. To this end, seven reinforced concrete half-joints were tested to failure under three-point bending to investigate the effect of reinforcement reductions that simulate common deterioration scenarios arising from the corrosion of reinforcement. The results were then compared to the predictions of nonlinear finite element analyses and the strut-and-tie method. The results demonstrated that a reduction in the amount of reinforcement close to the re-entrant corner has major influence on the load-deflection response of reinforced concrete half-joints. The largest reductions in load-carrying capacity were obtained from the half-joints with a full removal of diagonal bars and a combined removal of diagonal and U-bars, both showing reductions in load capacity of more than 50 %. The failure mode in each case of simulated deterioration scenarios involved a shear failure at the nib, except one case which displayed a shear failure in the full depth section due to the reduced anchorage of the bottom reinforcement. Furthermore, it was shown that nonlinear finite element analyses provide accurate predictions of the load-deflection response and failure crack patterns. The strut-and-tie method was found to result in overly conservative predictions, with an average observed-to-predicted load ratio of 1.48 and a coefficient of variation (COV) of 10.2 %. This is considerably higher than an average load ratio of 1.05 and a COV of only 3.1 % obtained from the nonlinear finite element analyses.

The Role of Predictive and Prognostic MRI-Based Biomarkers in the Era of Total Neoadjuvant Treatment in Rectal Cancer.

The role of magnetic resonance imaging (MRI) in rectal cancer management has significantly increased over the last decade, in line with more personalized treatment approaches. Total neoadjuvant treatment (TNT) plays a pivotal role in the shift from traditional surgical approach to non-surgical approaches such as 'watch-and-wait'. MRI plays a central role in this evolving landscape, providing essential morphological and functional data that support clinical decision-making. Key MRI-based biomarkers, including circumferential resection margin (CRM), extramural venous invasion (EMVI), tumour deposits, diffusion-weighted imaging (DWI), and MRI tumour regression grade (mrTRG), have proven valuable for staging, response assessment, and patient prognosis. Functional imaging techniques, such as dynamic contrast-enhanced MRI (DCE-MRI), alongside emerging biomarkers derived from radiomics and artificial intelligence (AI) have the potential to transform rectal cancer management offering data that enhance T and N staging, histopathological characterization, prediction of treatment response, recurrence detection, and identification of genomic features. This review outlines validated morphological and functional MRI-derived biomarkers with both prognostic and predictive significance, while also exploring the potential of radiomics and artificial intelligence in rectal cancer management. Furthermore, we discuss the role of rectal MRI in the 'watch-and-wait' approach, highlighting important practical aspects in selecting patients for non-surgical management.

An efficient method based on shear models for structural seismic response prediction considering hysteretic characteristics

An efficient method based on shear models for structural seismic response prediction considering hysteretic characteristics

Effect of concrete cover on the pure torsional behavior of reinforced concrete beams: Analytical model

Experimental data on RC members subjected to pure torsion is compiled from published studies so far. Interestingly, the past studies have concentrated largely on members with small to moderate concrete covers, with handful data on members with thick concrete cover. Widely accepted torsion models including the diagonal compression field theory (CFT), softened truss model (STM) and softened membrane model for torsion (SMMT) are verified on the domains of experimental data missing-in specimens with thick concrete cover. With increasing demand for durability design of RC structures use of thick concrete cover, say up to 75 mm, is being introduced in practice. Recent experimental investigation by the authors demonstrated that spalling of concrete cover, particularly in cases of thick covers, significantly impacts the torsional behavior of RC members. The present study uses these set of experiments to look at the response prediction capability of the existing advanced models and assess the reliability of existing concrete cover spalling theories. For cases with thick cover, the existing models either did not adequately capture the ultimate capacity or erroneously predicted the torsional behavior of the members due to the different mechanics between RC beams with thin and thick covers. Similarly, the spalling theories failed to fully explain the physically observed spalling behavior. Guided by the recent experimental observations and the apparent gaps, the study provides a rational theory for the spalling of concrete cover. The initiation and gradual evolution of concrete cover spalling is traced by observing the acting spalling moment and spalling resistance formulated in the proposed spalling model. The proposed spalling model inherently assumes members susceptibility to spalling of cover concrete and is governed by, the thickness of the concrete cover, tensile strength of concrete, presence of rebar cage and their location, and size of the member. The theory is unified and can explain the spalling of cover due to torsion as well as shear. Its capability is examined by integrating the approach into existing truss model. When compared with state-of-the-art models, the proposed method provides consistent prediction with a relatively smaller scatter, for cases with thick concrete cover as well.