Search

Triple-negative breast cancer (TNBC) is persistently refractory to therapy, and methods to improve targeting and evaluation of responses to therapy in this disease are needed. Here, we integrate quantitative MRI data with biologically based mathematical modeling to accurately predict the response of TNBC to neoadjuvant systemic therapy (NAST) on an individual basis. Specifically, 56 patients with TNBC enrolled in the ARTEMIS trial (NCT02276443) underwent standard-of-care doxorubicin/cyclophosphamide (A/C) and then paclitaxel for NAST, where dynamic contrast-enhanced MRI and diffusion-weighted MRI were acquired before treatment and after two and four cycles of A/C. A biologically based model was established to characterize tumor cell movement, proliferation, and treatment-induced cell death. Two evaluation frameworks were investigated using: (i) images acquired before and after two cycles of A/C for calibration and predicting tumor status after A/C, and (ii) images acquired before, after two cycles, and after four cycles of A/C for calibration and predicting response following NAST. For Framework 1, the concordance correlation coefficients between the predicted and measured patient-specific, post-A/C changes in tumor cellularity and volume were 0.95 and 0.94, respectively. For Framework 2, the biologically based model achieved an area under the receiver operator characteristic curve of 0.89 (sensitivity/specificity = 0.72/0.95) for differentiating pathological complete response (pCR) from non-pCR, which is statistically superior (P < 0.05) to the value of 0.78 (sensitivity/specificity = 0.72/0.79) achieved by tumor volume measured after four cycles of A/C. Overall, this model successfully captured patient-specific, spatiotemporal dynamics of TNBC response to NAST, providing highly accurate predictions of NAST response. Significance: Integrating MRI data with biologically based mathematical modeling successfully predicts breast cancer response to chemotherapy, suggesting digital twins could facilitate a paradigm shift from simply assessing response to predicting and optimizing therapeutic efficacy.

Background and Purpose: There is currently lack of recognized imaging criteria for prediction of treatment response to NAST in breast cancer patients. And early identification of treatment response to neoadjuvant systemic therapy (NAST) in Triple Negative Breast Cancer (TNBC) patients is important for appropriate treatment selection and response monitoring. A novel MRI sequence, Magnetic Resonance Image Compilation (MagIC) is capable of simultaneous quantitation of several tissue water properties including longitudinal (T1), transverse (T2) relaxation times, and proton density (PD). In this study we evaluated the ability of a radiomic model extracted from a novel MagIC sequence acquired early during NAST to predict pathologic complete response to NAST in TNBC. Materials and Methods: This IRB approved prospective ARTEMIS trial (NCT02276443) included 184 women (122 training dataset, 62 testing dataset) diagnosed with stage I-III TNBC. All patients were scanned with MagIC on a 3T MRI scanner at baseline (184 patients), and after 4 cycles (156 Patients) of NAST. T1, T2 and PD maps were generated from the source images using SyMRI (SyntheticMR, Linkoping, Sweden). Histopathology at surgery was used to determine pathologic complete response (pCR) which was defined as absence of the invasive cancer in the breast and axillary lymph nodes. 3D contouring of the tumors was performed using an in-house toolbox. 310 (10 first-order, 300 GLCM) textural features were extracted from each map, with total of 930 features/patient. Radiomic features were compared between pCR and non-pCR using Wilcoxon Rank Sum test and Fisher’s exact test. To build a multivariate, predictive model, logistic regression with elastic net regularization was performed for texture feature selection. The tuning parameter was optimized using 5-fold cross-validation based on the average area under curve (AUC) of each fold of a cross-validation using training data. Then the testing data were used to compare model’s performance by AUC. Results: Univariate analysis found 23 PD, 17 T1 and 10 T2 radiomic features at C4 time point to be able to predict pCR status with AUC >70% in both training and testing cohort. The top performing radiomic features were Entropy, Variance, Homogeneity and Energy (Tables1-2). Multivariate radiomics models from C4-PD, and C4-T1 maps showed best performance during both cross validation and independent testing. The radiomic signature of C4-T1 map that included 27features had best performance, with an AUC of 0.77, 0.70 (95% CI: 0.571-0.868) in training and testing cohort respectively. C4-PD map radiomic signature that included 6features was able to predict the pCR status with AUC of 0.73, 0.72 (95% CI: 0.571-0.868) in training and testing cohort respectively. Conclusion: Our data found that MagIC-based radiomics signature could potentially predict pathologic complete response in TNBC early during NAST. This data shows the potential application of MagIC radiomic model for improvement of response assessment in TNBC. Table 1.Best performing radiomic features from PD map after 4 cycles of NAST in TNBC patients.FeatureTraining CohortTraining CohortTraining CohortTesting CohortTesting CohortTesting CohortNAUC95% CINAUC95% CIP-valuePD-mapAngular Variance of Sum entropy1060.73820.6437-0.8328500.73240.5895-0.8752 Table 2.Best performing radiomic features from T1-T2 maps after 4 cycles of NAST in TNBC patients.FeatureTraining CohortTraining CohortTraining CohortTesting CohortTesting CohortTesting CohortNAUC95% CINAUC95% CIP-valueT1-mapAngular Variance of Sum entropy1060.76530.6762-0.8544500.70510.5524-0.8579 Citation Format: Nabil Elshafeey, Ken-Pin Hwang, Beatriz Elena Adrada, Rosalind Pitpitan Candelaria, Medine Boge, Rania M Mahmoud, Huiqin Chen, Jia Sun, Wei Yang, Aikaterini Kotrotsou, Benjamin C Musall, Jong Bum Son, Gary J Whitman, Jessica Leung, Huong Le-Petross, Lumarie Santiago, Deanna Lynn Lane, Marion Elizabeth Scoggins, David Allen Spak, Mary Saber Guirguis, Miral Mahesh Patel, Frances Perez, Abeer H Abdelhafez, Jason B White, Lei Huo, Elizabeth Ravenberg, Wei Peng, Alastair Thompson, Senthil Damodaran, Debu Tripathy, Stacey L Moulder, Clinton Yam, Mark David Pagel, Jingfei Ma, Gaiane Margishvili Rauch. Radiomics model based on magnetic resonance image compilation (MagIC) as early predictor of pathologic complete response to neoadjuvant systemic therapy in triple-negative breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr PD11-06.

Background: TNBC constitutes an aggressive and heterogeneous group of tumors with variable response to neoadjuvant therapy (NAT) that currently lacks clinically available profiling strategies for prediction. We aimed to develop an integrated model based on imaging, pathological and clinical data capable to predict NAT response in TNBC early during therapy. METHOD AND MATERIALS:125 Stage I-III TNBC patients enrolled in an IRB approved prospective clinical trial (NCT02276433) who had DCE-MRI at baseline (BL) and post 2 cycles (C2) of NAT, and had surgery were included in this analysis. Tumor volume was calculated using 3D measurements at BL and C2 time points DCE-MRI. Percent tumor volume reduction (TVR) between BL and C2 was calculated. Demographic, clinical, and pathological data (age, T and N stage, histology, androgen receptor expression, Ki-67, stromal tumor infiltrating lymphocytes level (sTIL), and PD-L1 expression), and treatment response at surgery (pCR vs non-pCR) were documented. Recursive partitioning was used to identify TVR cutoff value. Multivariate logistic regression and ROC analysis were used to assess associations and build and evaluate predictive models. RESULTS: 61 (49%) TNBC pts showed pCR at surgery, and 64 (51%) non-pCR. Recursive partitioning analysis identified ≥ 55% TVR as the optimal cutoff values for pCR prediction at C2. TVR, N stage and sTIL were significantly associated with pCR in the multivariate analyses (p Citation Format: Gaiane Margishvili Rauch, Rosalind P. Candelaria, Mary Saber Guirguis, Medine Boge, Rania M. M. Mohamed, Nabil Elshafeey, Jia Sun, Gary J Whitman, Jessica Leung, Huong C Le-Petross, Lumarie Santiago, Deanna Lane, Marion Scoggins, David Spak, Miral M Patel, Frances Perez, Jason B. White, Elizabeth Ravenberg, Wei Peng, Debu Tripathy, Vicente Valero, Jennifer Litton, Lei Huo, Clinton Yam, Alastair Thompson, Jingfei Ma, Stacy L. Moulder, Wei Yang, Beatriz E. Adrada. Integrated model for early prediction of neoadjuvant systemic therapy response in triple negative breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr PD11-07.

Introduction:. Patients with locally advanced triple-negative breast cancer (TNBC) typically receive neoadjuvant therapy (NAT) to downstage the tumor and to improve the outcome of the subsequent breast conservation surgery. A critical unmet need is the lack of a method to accurately predict how a patient with TNBC will respond to NAT before surgery. In this work, we applied a clinical-computational framework to predict response of TNBC early in the course of NAT, by integrating quantitative MRI with mechanism-based mathematical modeling. Methods:. Patients and Data. Multiparametric quantitative MRI was acquired in patients (n = 46) before, and after 2 and 4 cycles of Adriamycin/Cyclophosphamide (A/C) regimen as part of the MD Anderson Cancer Center TNBC Moonshot Program. Within each imaging session, dynamic contrast-enhanced (DCE-), diffusion-weighted imaging (DWI), and a pre-contrast T1-map were acquired. Image processing. The processing pipeline consisted of three components. First, the images within each visit were registered to account for patient motion, and the parametric maps from the DCE and DWI images were computed. Second, inter-visit image registration was achieved by a non-rigid registration applied on breast, with a rigid penalty applied on the tumor region to preserve its size and shape. Third, post-processing was performed for preparation of modeling, including segmentation of the breast contour and tissues, and calculation of voxel-wise cellularity within tumors. Mathematical modeling. A predictive model was developed based on a reaction-diffusion equation (Eq. 1). The mobility of tumor cells is represented by diffusion coupled to mechanical properties of the tissue (Eq. 2), and the proliferation of the tumor is described with logistic growth. The injection and decay of administered therapies, inducing tumor cell death, is also represented in the model (Eq. 3). The variables and parameters used are listed in Table 1. Eq. 1: ∂N(x,t)/∂t = ∇⋅(D(x,t) ∇N(x,t)) + k(x) (1 - N(x,t)/θ)N(x,t) - (λ1(x,t) + λ2(x,t))N(x,t). Eq. 2: D(x,t) = D0 e-γσ(x,t). Eq. 3: λn(x,t) = αne-βn t C(x,t), n = 1, 2. For each patient, the domain and initial condition were generated from the pre-treatment images, and the images acquired during NAT were used for patient-specific calibration of parameters. The calibrated model was then used to predict the response to be observed at the end of NAT. We evaluated the model by comparing its predictions of tumor volume, longest axis, voxel-wise cellularity, and total tumor cellularity to the imaging measurements at the end of A/C. Results:. Our model predicted the tumor volume, total cellularity, and longest axis with a Pearson correlation coefficient (PCC) of 0.85, 0.80, and 0.60, respectively. The accuracy of voxel-wise cellularity achieved a PCC with the median (range) of 0.89 (0.77 - 0.93) between the prediction and the actual measurement. Moreover, we set criteria of 70% shrinkage of tumor volume to define response versus non-response cases, with which our model achieved a differentiation sensitivity/specificity of 0.90/0.73. Discussion:. Preliminary results of our study demonstrate the potential of the clinical-computational framework as a powerful tool for predicting response to NAT. Once validated, the method could also assist in optimizing treatment plans on a patient specific basis, or guiding patient selection in trials for novel NAT regimens. Table 1. Summary of the variables and parameters in the modelQuantitiesDefinition AssignmentDomainsΩbreast tissue domainGenerated from pre-treatment MRITEnd time point of NAT procedureDetermined from NAT schedulexCoordinate in breast tissueAssociated with spatial domain, ΩttimeAssociated with temporal domain, [0, T]VariablesN(x,t)Tumor cell numberInitialized from pre-treatment ADC, computed via Eq. 1D(x,t)Diffusive mobility of tumor cellsComputed via Eq. 2λn(x,t)Death rate induced by nth type of drugComputed via Eq. 3, n = 1 and 2 for A/Cσ(x,t)Von Mises stressComputed from gradient of N(x,t), based on Hormuth et al., 2018C(x,t)Spatiotemporal distribution of drugsAssigned based on NAT schedule and DCE imagesParametersk(x)Proliferation rate of tumor cellsLocally calibratedθTumor cells carry capacityGlobally calibratedαnEfficacy rate of nth type of drugGlobally calibratedβnDecay rate of of nth type of drugGlobally calibratedD0Diffusion coefficient of tumor cells in the absence of mechanical restrictionsGlobally calibratedγStress-tumor cell diffusion coupling constantAssigned based on Hormuth et al., 2018 Citation Format: Chengyue Wu, Angela M. Jarrett, Zijian Zhou, Nabil Elshafeey, Beatriz E. Adrada, Rosalind P. Candelaria, Rania M. Mohamed, Medine Boge, Lei Huo, Jason White, Debu Tripathy, Vicente Valero, Jennifer Litton, Stacy Moulder, Clinton Yam, Jong Bum Son, Jingfei Ma, Gaiane M. Rauch, Thomas E. Yankeelov. Forecasting treatment response to neoadjuvant systemic therapy in triple negative breast cancer viamathematical modeling and quantitative MRI [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P1-08-08.

PURPOSE To compare clinical outcomes in a cohort of patients with advanced non–small-cell lung cancer (NSCLC) with targetable genomic alterations detected using plasma-based circulating tumor DNA (ctDNA) or tumor-based next-generation sequencing (NGS) assays treated with US Food and Drug Administration–approved therapies at a large academic research cancer center. METHODS A retrospective review from our MD Anderson GEMINI database identified 2,224 blood samples sent for ctDNA NGS testing from 1971 consecutive patients with a diagnosis of advanced NSCLC. Clinical, treatment, and outcome information were collected, reviewed, and analyzed. RESULTS Overall, 27% of the ctDNA tests identified at least one targetable mutation and 73% of targetable mutations were EGFR-sensitizing mutations. Among patients treated with first-line epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitor (TKI) therapies, there were no significant differences in progression-free survival of 379 days and 352 days ( P value = .41) with treatment based on tissue (n = 40) or ctDNA (n = 40), respectively. Additionally, there were no differences in progression-free survival or objective response rate among those with low (n = 8, 0.01%-0.99%) versus high (n = 16, ≥ 1%) levels of ctDNA of the targetable mutation as measured by variant allele frequency (VAF). Overall, there was excellent testing concordance (n = 217 tests) of > 97%, sensitivity of 91.7%, and specificity of 99.7% between blood-based ctDNA NGS and tissue-based NGS assays. CONCLUSION There were no significant differences in clinical outcomes among patients treated with approved EGFR-TKIs whose mutations were identified using either tumor- or plasma-based comprehensive profiling and those with very low VAF as compared with high VAF, supporting the use of plasma-based profiling to guide initial TKI use in patients with metastatic EGFR-mutant NSCLC.

Introduction CEST MRI permits quantitation of macromolecules such as amide proteins that are of interest in cancer metabolism. However, optimal CEST acquisition and analysis methods remain undetermined. In this study, we investigated CEST MRI as an imaging biomarker for early treatment response in 51 TNBC patients receiving NAST and compared the performance with two different CEST saturation power levels and two analysis methods. Methods A total of 51 stage I-III TNBC patients enrolled in the prospective ARTEMIS trial (NCT02276443) had CEST imaging performed on a 3T MRI scanner at baseline before NAST (BL, N = 51), after 2 cycles (C2, N = 37), and 4 cycles (C4, N = 44) of NAST. 33 of the 51 patients had imaging at all 3 time points. 29 of the 33 patients had pathological findings, with N = 16 with pathological complete response (pCR) and N = 13 with non-pCR. Two sets of CEST images using 0.9 and 2.0 µT saturation power levels were acquired and analyzed using the magnetization transfer ratio asymmetry (MTRasym) and the Lorentzian line fitting (Mag3.5) methods, for a total of 4 acquisition/analysis combinations. The group averaged CEST signals, MTRasym at 0.9 and 2.0 µT and Mag3.5 at 0.9 and 2.0 µT, at BL, C2 and C4 were determined and evaluated using unpaired (51 patients) and paired (33 patients) Kruskal-Wallis tests. The Mag3.5 at 0.9 µT and the MTRasym at 2.0 µT were further compared between pCR and non-pCR. The group averaged CEST signals at BL, C2, and C4 were evaluated using the Friedman test for the pCR and the non-PCR groups. Separately, the change in the CEST signal from BL to C2 and C4 was determined for each patient and evaluated using the Mann-Whitney test for both groups. P < 0.05 was considered statistically significant. Results The MTRasym at BL was higher at 2.0 µT than at 0.9 µT. In contrast, the Mag3.5 at BL was higher at 0.9 µT than at 2.0 µT. The MTRasym at 2.0 µT and the Mag3.5 at 0.9 µT decreased during treatment while the MTRasym at 0.9 µT and the Mag3.5 at 2.0 µT were similar. Both the unpaired and the paired Mag3.5 at 0.9 µT showed a significant decrease at C2 and C4 vs. BL (p < 0.01). The unpaired and paired MTRasym at 2.0 µT showed a decrease, although the change was not significant except for the unpaired data at C4. The decrease in the group averaged Mag3.5 at 0.9 µT was significant at C2 vs. BL for the pCR group (p = 0.04), while it was not significant for the pCR group at C4 vs. BL and for the non-pCR group at either C2 or C4 vs. BL. The group averaged MTRasym at 2.0 µT changes were not significant for either the pCR or the non-pCR groups. None of the CEST signal changes on a per patient basis at C2-BL, C4-BL and C4-C2 were significantly different between the pCR and the non-pCR groups. Further, none of the group averaged CEST signals at BL, C2 and C4 were significantly different between the pCR and the non-pCR groups. Conclusion Our study demonstrates that the CEST quantitation in TNBC patients undergoing NAST depends on acquisition and analysis. For a maximum change in the CEST effect, Lorentzian line fitting is better paired with acquisition at a low saturation power (0.9 µT) and MTRasym is better paired with acquisition at a high saturation power (2.0 µT). Further, a significant CEST signal decrease was observed in TNBC patients with pCR after NAST when a 0.9 µT saturation power and the Lorentzian line fitting were used. In comparison, the decrease was not significant in non-pCR patients using the same saturation power and analysis method. The results suggest that the CEST signal acquired at 0.9 µT saturation power and analyzed using Lorentzian line fitting may be able to differentiate between pCR and non-pCR among TNBC patients undergoing NAST. Additional studies with a larger patient population are ongoing to further validate our findings and their potential for determining pCR. Citation Format: Shu Zhang, Gaiane M Rauch, Beatriz E Adrada, Medine Boge, Rania MM Mohamed, Abeer H Abdelhafez, Jong Bum Son, Jia Sun, Nabil A Elshafeey, Jason B White, Deanna L Lane, Jessica WT Leung, Marion E Scoggins, David A Spak, Elsa Arribas, Elizabeth Ravenberg, Lumarie Santiago, Tanya W Moseley, Gary J Whitman, Huong Le-Petross, Benjamin C Musall, Mitsuharu Miyoshi, Xinzeng Wang, Brandy Willis, Stacy Hash, Aikaterini Kotrotsou, Peng Wei, Ken-Pin Hwang, Alastair Thompson, Stacy L Moulder, Rosalind P Candelaria, Wei Yang, Jingfei Ma, Mark D Pagel. Assessment of early response to neoadjuvant systemic therapy (NAST) of triple-negative breast cancer (TNBC) using chemical exchange saturation transfer (CEST) MRI: A pilot study [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS3-08.

PURPOSE Triple negative breast cancer (TNBC) is an aggressive and heterogeneous subtype of breast cancer. Pathologic complete response (pCR) to neoadjuvant systemic therapy (NAST) predicts better survival. Early prediction of the treatment response can potentially triage non-responding patients to alternative protocol treatments, spare them of the unneeded toxicity, and improve pCR. We evaluated the ability of radiomic textural analysis of intratumoral and peritumoral regions on the dynamic contrast enhanced (DCE) and diffusion-weighted imaging (DWI) MRI images obtained early during NAST to predict pCR. MATERIALS AND METHODS This IRB-approved prospective study (NCT02276443) included 182 patients with biopsy proven stage I-III TNBC who had multiparametric MRIs at baseline (BL), post 2 cycles (C2), and post 4 cycles (C4) of NAST before surgery. Tumors and peritumoral regions of 5 mm and 10 mm in thickness were segmented on the 2.5 minutes DCE subtraction images and on the b=800 DWI images. Ten histogram-based first order texture features including mean, minimum, maximum, standard deviation, kurtosis, skewness, 1st, 5th, 95th, and 99th percentile, and 300 radiomic Grey Level Co-occurrence matrix (GLCM) features along with their absolute and relative differences between the 3 imaging time points were extracted from the tumors and from the peritumoral regions with an in-house Matlab toolbox. Treatment response at surgery (pCR vs non-pCR) was documented. The samples were divided into training and testing datasets by a 2:1 ratio. Area under the receiver operating characteristics curve (AUC ROC) was calculated for univariate analysis in predicting pCR. Logistic regression with elastic net regularization was performed for texture feature selection. Parameter optimization was performed by using 5-fold cross-validation based on mean cross-validated AUC in the training set. RESULTS Of 182 TNBC patients, 88 (48%) had pCR and 94 (52%) did not achieve pCR. Eight multivariate models combining radiomic features from both DCE and DWI tumoral and peritumoral regions had AUC > 0.8 (0.807-0.831) with p-value < 0.001 in both training and testing sets. The highest AUC=0.831 was obtained from a model consisting of 15 radiomic features: tumor DWI (5 GLCM features) at C2, peritumoral region on DCE (skewness) at C2, tumor DCE (1st, 5th percentile) at C4, tumor DWI (3 GLCM features) at C4, peritumoral region DWI (1 GLCM feature) at C4, and the relative difference between C4/C2 on DCE (5th, 95th percentile and mean). CONCLUSION Multi-parametric MRI-based radiomics models from the tumor and the peritumoral regions showed high accuracy as potential early predictors of NAST response in TNBC patients. Citation Format: Rania M. Mohamed, Bikash Panthi, Beatriz Adrada, Rosalind Candelaria, Mary S. Guirguis, Wei Yang, Medine Boge, Miral Patel, Nabil Elshafeey, Sanaz Pashapoor, Zijian Zhou, Jong Bum Son, Ken-Pin Hwang, H. T. Carisa Le-Petross, Jessica Leung, Marion E. Scoggins, Gary J. Whitman, Zhan Xu, Deanna L. Lane, Tanya Moseley, Frances Perez, Jason White, Elizabeth Ravenberg, Alyson Clayborn, Mark Pagel, Huiqin Chen, Jia Sun, Peng Wei, Alastair M. Thompson, Stacy Moulder, Anil Korkut, Lei Huo, Kelly K. Hunt, Jennifer K. Litton, Vicente Valero, Debu Tripathy, Clinton Yam, Jingfei Ma, Gaiane Rauch. Multi-Parametric MRI-Based Radiomics Models from Tumor and Peritumoral Regions as Potential Predictors of Treatment Response to Neoadjuvant Systemic Therapy in Triple Negative Breast Cancer Patients [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P6-01-06.

Background and Purpose Triple negative breast cancer (TNBC) is a biologically aggressive tumor and a refractory subtype of breast cancer due to the lack of therapeutic targets, such as estrogen receptors, progesterone receptors, and human epidermal growth factor receptor 2. In this study, we investigated the accuracy of radiomic models based on the dynamic contrast enhanced (DCE) MRI images obtained after the completion of NAST as discriminators of treatment response in TNBC patients. Materials and Methods This IRB-approved prospective study (ARTEMIS trial, NCT02276443) included 181 patients with biopsy proven stage I-III TNBC who Had MRIs after completion of NAST and before surgery. Patients were classified as pathologic complete response (pCR) and non-pCR at the surgery. Tumors were segmented on the 2.5 minutes DCE subtraction images. Regions with necrosis or clip artifacts were excluded from the contour. If tumors were not visible, the tumor bed was contoured. Whole-tumor histogram-based first order texture features (p=10) including mean, minimum, maximum, Standard deviation, kurtosis, skewness, 1st, 5th, 95th, and 99th percentiles, and radiomic (p=300) Grey Level Co-occurrence matrix (GLCM) features were extracted with an in-house Matlab toolbox. The samples were split into training and testing data sets by a 2:1 ratio. For univariate analysis area under the receiver operating characteristics curve (AUC ROC) was performed for pCR status prediction. For texture feature selection logistic regression with elastic net regularization was performed. Parameter optimization was performed by using 5-fold cross-validation based on mean cross-validated AUC in the training set. A P-value less than 0.05 was considered statistically significant. Results Of the total 181 patients, 88 (49%) had pCR and 93 (51%) had non-pCR. Univariate analysis identified 7 statistically significant first order imaging features (Minimum, Maximum, Mean, 1st Percentile, 5th Percentile, 95th Percentile, and 99th Percentile) with AUC >= 0.7 (p< 0.001), in both training and testing data sets. Percentile 5 showed highest AUC = 0.78 (p< 0.001). Two multivariate models were statistically significant at cross-validation with AUC>=0.7. The first model combined 2 first order data (Percentile 1 and Percentile 5) with AUC = 0.73 (p< 0.001). The second model combined 8 first order features (Percentile 1, 5, 95, 99, Mean, Minimum, Maximum, and Skewness) and 24 GLCM features with AUC = 0.7 (p=0.003). Conclusion DCE-MRI radiomic features from tumor and tumor bed regions in TNBC may be helpful imaging biomarkers for predicting treatment response after NAST. Citation Format: Rania M. Mohamed, Bikash Panthi, Beatriz Adrada, Rosalind Candelaria, Mary S. Guirguis, Wei Yang, Medine Boge, Miral Patel, Nabil Elshafeey, Sanaz Pashapoor, Zijian Zhou, Jong Bum Son, Ken-Pin Hwang, H. T. Carisa Le-Petross, Jessica Leung, Marion E. Scoggins, Gary J. Whitman, Zhan Xu, Deanna L. Lane, Tanya Moseley, Frances Perez, Jason White, Huiqin Chen, Jia Sun, Peng Wei, Jennifer K. Litton, Vicente Valero, Clinton Yam, Mark Pagel, Jingfei Ma, Gaiane Rauch. A Pre-operative Dynamic Contrast Enhanced MRI-Based Radiomics Models as Predictors of Treatment Response after Neoadjuvant Systemic Therapy in Triple Negative Breast Cancer Patients [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P6-01-35.

Background and Purpose Early prediction of neoadjuvant systemic therapy (NAST) response in triple negative breast cancer (TNBC) patients could potentially aid in the selection of alternative therapies and avoid unnecessary toxicity in patients unlikely to achieve pathologic complete response (pCR) with NAST. In this study, we investigated the radiomic features of the peritumoral and the tumoral regions from dynamic contrast enhanced (DCE) MRI acquired at different time points of NAST for early treatment response prediction in TNBC. Methods and Materials This study included 182 biopsy-confirmed stage I-III TNBC patients enrolled in an IRB approved prospective clinical trial (NCT02276433). All patients underwent DCE-MRI on a GE 3T MRI scanner at baseline (BL), after two (C2) and four (C4) cycles of doxorubicin/cyclophosphamide based chemotherapy and before surgery. The peritumoral and the tumoral regions were segmented manually by two fellowship-trained radiologists using early phase (2.5 min) DCE-MRI subtraction images. Ten first order radiomic features, 300 grey-level-co-occurrence matrix (GLCM) features along with their absolute and relative differences (C4/BL, C2/BL, C4/C2) between the 3 imaging time points were extracted from the peritumoral and the tumoral regions. Patients were randomly divided into training and testing sets in a 2:1 ratio. For univariate analysis, area under the receiver operating characteristics curve (AUC ROC) was measured to determine the features most predictive of pCR/non-pCR. Wilcoxon Rank Sum test was used to test the statistical significance of predictive performance. In multivariate analysis, radiomic models were established using logistic regression with elastic net regularization followed by 5-fold cross validation for performance assessment. Results Eighty-eight (48%) patients had pCR (59 training, 29 testing) and 94 (52%) patients had non-pCR (63 training, 31 testing). Twenty-five radiomic features (4 from peritumoral C4, 5 from tumoral C4, 4 from peritumoral C4/BL, 6 from tumoral C4/BL, 2 from peritumoral C4/C2 and 4 from tumoral C4/C2) were statistically significant with AUC ≥ 0.75 in both the training and the testing sets at the univariate analysis. The significant features at C4 had AUCs of 0.75-0.79 for the training set and 0.76-0.81 for the testing set. Changes measured between C4 and BL or C2 showed AUC of 0.76-0.84 in the training and 0.75-0.81 in the testing datasets. Eleven multivariate regression models comprised of radiomic features at BL, C2, C4 and their changes (C4/BL, C4/C2 and C2/BL) showed an AUC of 0.80-0.84 for cross validation and an AUC of 0.80-0.82 for independent testing. Conclusions Radiomic models using longitudinal DCE MRI parameters of peritumoral and tumoral regions during NAST have the potential to predict pCR in TNBC patients undergoing NAST. Citation Format: Bikash Panthi, Rania M. Mohamed, Beatriz Adrada, Rosalind Candelaria, Mary S. Guirguis, Wei Yang, Medine Boge, Miral Patel, Nabil Elshafeey, Sanaz Pashapoor, Zijian Zhou, Jong Bum Son, Ken-Pin Hwang, H. T. Carisa Le-Petross, Jessica Leung, Marion E. Scoggins, Gary J. Whitman, Zhan Xu, Deanna L. Lane, Tanya Moseley, Frances Perez, Jason White, Elizabeth Ravenberg, Alyson Clayborn, Mark Pagel, Huiqin Chen, Jia Sun, Peng Wei, Alastair M. Thompson, Stacy Moulder, Anil Korkut, Lei Huo, Kelly K. Hunt, Jennifer K. Litton, Vicente Valero, Debu Tripathy, Clinton Yam, Jingfei Ma, Gaiane Rauch. Longitudinal DCE-MRI Radiomic Models for Early Prediction of Response to Neoadjuvant Systemic Therapy (NAST) in Triple Negative Breast Cancer (TNBC) Patients [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P6-01-34.

The need to identify biomarkers to predict immunotherapy response for rare cancers has been long overdue. We aimed to study this in our paper, ‘Radiomics analysis for predicting pembrolizumab response in patients with advanced rare cancers’. In this response to the Letter to the Editor by Cunha et al, we explain and discuss the reasons behind choosing LASSO (Least Absolute Shrinkage and Selection Operator) and XGBoost (eXtreme Gradient Boosting) with LOOCV (Leave-One-Out Cross-Validation) as the feature selection and classifier method, respectively for our radiomics models. Also, we highlight what care was taken to avoid any overfitting on the models. Further, we checked for the multicollinearity of the features. Additionally, we performed 10-fold cross-validation instead of LOOCV to see the predictive performance of our radiomics models.

BackgroundWe present a radiomics-based model for predicting response to pembrolizumab in patients with advanced rare cancers.MethodsThe study included 57 patients with advanced rare cancers who were enrolled in our phase II clinical trial of pembrolizumab. Tumor response was evaluated using Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 and immune-related RECIST (irRECIST). Patients were categorized as 20 “controlled disease” (stable disease, partial response, or complete response) or 37 progressive disease). We used 3D-slicer to segment target lesions on standard-of-care, pretreatment contrast enhanced CT scans. We extracted 610 features (10 histogram-based features and 600 second-order texture features) from each volume of interest. Least absolute shrinkage and selection operator logistic regression was used to detect the most discriminatory features. Selected features were used to create a classification model, using XGBoost, for the prediction of tumor response to pembrolizumab. Leave-one-out cross-validation was performed to assess model performance.FindingsThe 10 most relevant radiomics features were selected; XGBoost-based classification successfully differentiated between controlled disease (complete response, partial response, stable disease) and progressive disease with high accuracy, sensitivity, and specificity in patients assessed by RECIST (94.7%, 97.3%, and 90%, respectively; pConclusionOur radiomics-based signature identified imaging differences that predicted pembrolizumab response in patients with advanced rare cancer.InterpretationOur radiomics-based signature identified imaging differences that predicted pembrolizumab response in patients with advanced rare cancer.

Introduction: Patients with locally advanced, triple-negative breast cancer (TNBC) typically receive neoadjuvant therapy (NAT) to downstage the tumor and to improve the outcome of subsequent breast conservation surgery. There are currently no methods to accurately predict how a TNBC patient will respond to NAT before surgery. In this work, we applied a digital twin framework to address this unmet clinical need, by integrating quantitative magnetic resonance imaging (MRI) data with mechanism-based mathematical modeling. Methods: Multiparametric MRI was acquired in patients (N = 50) before, after 2 and 4 cycles of Adriamycin/Cyclophosphamide (A/C), and again after 12 cycles of Paclitaxel as part of the ARTEMIS (NCT02276433) trial. Within each imaging session, dynamic contrast-enhanced (DCE) MRI, diffusion-weighted imaging (DWI), and a pre-contrast T1-map were acquired. The images were processed by a pipeline consisting of motion correction, multiparametric image alignment, inter-visit image registration to align the tumor and surrounding breast tissue, tissue segmentation, and estimation of tumor cellularity from DWI. A mechanism-based mathematical model, a reaction-diffusion equation, is used to characterize the mobility of tumor cells via diffusion damped by mechanical tissue properties, tumor proliferation via logistic growth, and treatment-induced cell death via the delivery and decay of therapies. For each patient, pre-treatment images were used for model initialization. The model calibration and prediction were implemented with two strategies: 1) using images acquired during the A/C for calibration and predicting up to the end of A/C, and 2) using images acquired during and after the A/C for calibration and predicting up to the end of NAT. For strategy 1), we evaluated the model by comparing its predicted tumor volume and total tumor cellularity to the imaging measurements at the end of A/C. For strategy 2), we evaluated the model by comparing its predicted final response to the post-surgical pathological findings. Results: For strategy 1), our framework predicted the change of tumor volume and total tumor cellularity with Pearson correlation coefficients of 0.91 and 0.89, respectively. Regarding strategy 2), our framework achieved an area under the receiver operator characteristic curve of 0.88 for distinguishing pCR from non-pCR. As a comparison, imaging measurement of tumor volume at the end of A/C achieved an AUC of 0.79. Conclusion: Our approach successfully captures the patient-specific dynamics of TNBC response to NAT and provides an improved prediction of final response, which demonstrates the potential of a digital twin framework to be a powerful tool for predicting response to NAT. Once validated, the method will provide a unique opportunity for optimizing treatment plans on a patient-specific basis. Citation Format: Chengyue Wu, Angela M. Jarrett, Zijian Zhou, Nabil Elshafeey, Beatriz E. Adrada, Rosalind P. Candelaria, Rania Mohamed, Medine Boge, Lei Huo, Jason White, Debu Tripathy, Vicente Valero, Jennifer Litton, Clinton Yam, Jong Bum Son, Jingfei Ma, Gaiane M. Rauch, Thomas E. Yankeelov. Forecasting treatment response to neoadjuvant therapy in triple-negative breast cancer via an image-guided digital twin [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2736.

BACKGROUND: Increasing use of neoadjuvant systemic therapy (NAT) for early and locally advanced breast cancer led to critical need for development of tools capable of early treatment response assessment after NAT. Tc-99m sestamibi Molecular breast Imaging (MBI) as a functional imaging modality has a promise to detect changes in the tumor prior to anatomical changes detected by mammogram or ultrasound. PURPOSE: To evaluate the ability of quantitative MBI parameters to predict pathologic complete response (pCR) after completion of NAT in breast cancer patients. MATERIALS AND METHODS: Patients with invasive breast cancer (T1-T4, N0-N3, M0) planned for NAT followed by surgery were enrolled in a prospective IRB approved trial. MBI was performed at baseline and after two cycles of NAT. Patient demographic and tumor biology information (Ki-67, HER2, ER/PR) was collected. MBI images were quantified using a novel approach with corrections for scatter and attenuation and regions of interest (ROI) were drawn over tumors to compute three quantitative MBI uptake metrics for correlation with pathologic response: MBI-specific standardized uptake value (SUV), tumor to background ratio (TBR), and tumor volume. Pathologic complete response was determined based on final histopathology report at the time of surgery as absence of the invasive disease in the breast and axillary lymph nodes. MBI metrics at baseline, after 2 cycles of NAT and interval change were correlated with pCR and tumor biology using the Wilcoxon Rank Sum test, Kruskal-Wallis test or Fisher’s exact test. Statistical analysis was carried out using R (version 3.6.3, R Development Core Team). RESULTS: A total of 70 patients with median age 47.5 years (range 30-77) were included in the analysis. Breast cancer subtypes were: HER2 negative (ER/PR+) 35.7% (25/70), HER2 positive (ER/PR +/-) 35.7% (25/70), and triple negative (HER2-, ER/PR-) 28.6% (20/70). Change in SUV after 2 cycles of NAT was higher in patients with pCR compared to those who did not achieve pCR (mean decrease in SUV of 15.57 and 4.83 respectively, p Citation Format: Miral M Patel, Beatriz E Adrada, Benjamin Lopez, Rosalind P Candelaria, Jia Sun, Medine Boge, Rania M Mohamed, Nabil Elshafeey, Gary Whitman, MD, Huong T Le-Petross, Lumarie Santiago, Marion E Scoggins, Deanna Lane, Tanya Moseley, Galit Zylberman, Jerica Saddler, Jessica WT Leung, Wei T Yang, Vincente Valero, S Cheenu Kappadath, Gaiane M Rauch. Quantitative molecular breast imaging for early prediction of neoadjuvant systemic therapy response in locally advanced breast cancer patients [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P3-02-03.

Background and Purpose:There is currently a lack of recognized imaging criteria for prediction of treatment response to NAST in breast cancer patients with recent reports showing that breast MRI is the most accurate modality for evaluation of NAST response. DCE-MRI evaluates tumor perfusion that influences tumor enhancement at the post-contrast subtraction images and allows for more accurate measurement of changes in tumor volume during NAST. In this study, we evaluated the ability of tumor volumetric changes after 2 and 4 cycles of NAST by longitudinal ultrafast DCE-MRI to predict pathologic complete response (pCR) in TNBC undergoing NAST. Materials and Methods: Stage I-III TNBC patients enrolled in an IRB approved prospective clinical trial (ARTEMIS, NCT02276433) who had ultrafast DCE-MRI at baseline (BL, N=103), post 2 cycles (C2, N=59), and post 4 cycles (C4, N=103) of anthracycline-based NAST,and had surgery, were included in this analysis. Tumor volume was calculated using 3D measurements of the index lesion at BL, C2, and C4. Percent change of tumor volume (%TV) between BL, C2, and C4 was calculated at early (9-12 sec) and delayed (360-480 sec) phases of DCE-MRI. The largest lesion was used for analysis in patients with multicentric or multifocal disease. Demographic, clinical, and pathologic data and treatment response at surgery (pCR versus non-pCR) were documented. Receiver operating characteristics curve (ROC) analysis was performed for prediction of pCR status. Positive predictive value (PPV), negative predictive value (NPV) and Youden Index were used to select %TV cut-off thresholds for pCR prediction.Results: 103 patients (median age, 53 years; range, 24-79 years) were included, 48 (47%) had pCR, and 55 (53%) had non-pCR at surgical pathology. The %TV reduction at C2 DCE-MRI was predictive of pCR on both early phase DCE MRI (AUC, 0.873; CI:0.779-0.968, p < .0001) and delayed phase DCE MRI (AUC, 0.844; CI:0.742-0.947, p < .0001). Optimal thresholds were as follows: 70% TV reduction on early phase DCE MRI with Youden’s index of 1.58 was able to predict pCR correctly for 79% of patients with PPV of 81%; 75% TV reduction on delayed phase with Youden’s Index of 1.44 was able to predict pCR correctly for 71% of patients with PPV of 85%.%TV reduction was also predictive of pCR at the C4 time point on both early phase DCE MRI (AUC, 0.761; CI:0.665-0.856, p < .0001) and delayed phase DCE MRI (AUC, 0.737; CI:0.641-0.833, p < .0001). Optimal thresholds were as follows: 90% TV reduction on early phase DCE MRI with Youden’s index of 1.43 was able to correctly predict pCR in 72% of patients with PPV of 70%; and 90% TV reduction on delayed phase with Youden’s Index of 1.34 was able to predict pCR correctly in 68% of patients with PPV of 71%.Conclusion: Our data shows that percent tumor volume reduction by DCE-MRI after 2 and 4 cycles of NAST was able to predict pCR in TNBC with high accuracy and can be used as an early imaging biomarker of NAST response prediction. Volumetric changes by longitudinal DCE-MRI can be used to differentiate chemoresistant and chemosensitive TNBC patients as early as after 2 cycles of NAST, and can help to triage patients for treatment de-escalation or targeted therapy. Citation Format: Gaiane Margishvili Rauch, Adrada E Beatriz, Rosalind P Candelaria, Nabil Elshafeey, Abeer H Abdelhafez, Benjamin C Musall, Jia Sun, Medina Boge, Rania M.M Mohamed, Jong Bum Son, Shu Zhang, Jessica Leung, Deanna Lane, Marion Scoggins, David Spak, Elsa Arribas, Lumarie Santiago, Gary J Whitman, Huong T. Le-Petross, Tanya W Moseley, Jason B. White, Elizabeth Ravenberg, Ken-Pin Hwang, Peng Wei, Lei Huo, Jennifer K Litton, Vicente Valero, Debu Tripathy, Alastair M Thompson, Mark D Pagel, Jingfei Ma, Wei T Yang, Stacy Moulder. Volumetric changes on longitudinal dynamic contrast enhanced MR imaging (DCE-MRI) as an early treatment response predictor to neoadjuvant systemic therapy (NAST) in triple negative breast cancer (TNBC) patients [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD6-07.

Background and Purpose:Image-based tumor phenotypes by using computer extraction techniques have been studied for evaluation of breast cancer invasiveness, stage, lymph node involvement, molecular subtypes and genomics. In this project we aimed to investigate ability of computer-extracted breast MR imaging radiomic phenotypes to predict nodal and distant metastasis in breast cancer patients. MATERIALS AND METHODS:This retrospective IRB approved study included 416 biopsy proven breast cancer patients who had pretreatment DCE MRI in a single institution between 2014 and 2018. Patient’s demographic, clinical data, pathology at diagnosis and surgery, nodal and distant metastasis (M1) at follow up were documented. Using QuantX imaging software, the tumor volume of interest was automatically-segmented using the multiple dynamic phases of DCE MRI. A total of 33 radiomic features describing tumor phenotype were extracted from each tumor site. A linear discriminant analysis (LDA) as a classifier with nested feature selection 10-fold cross validation was used to build the radiomic signature for prediction of nodal and distant metastasis occurrence. Receiver operating characteristic (ROC) and precision-recall analyses were used to evaluate performance, with 95% confidence intervals from 1000 bootstraps, and Kaplan-Meier was used to calculate the progression-free survival estimates and associated hazard ratio at the median cutpoint of the probability of metastasis calculated by the LDA in the 10-fold cross-validation. RESULTS:The quantitative DCE MRI radiomic model was able to differentiate between breast cancer patients with and without distant metastatic disease at follow up with area under the ROC of 0.75 (95% CI 0.65; 0.82) and precision-recall curves 0.46 (0.33;0.69), hazard ratio at median cut point is 3.76 (2.27; 6.24), p The DCE radiomic model was able predict presence of ipsilateral nodal disease (≥1 positive lymph nodes) at surgery with AUC 0.66 (95% CI: 0.60; 0.71), ≥4 positive lymph nodes at surgery with AUC 0.67 (95% CI: 0.60; 0.74), and N2/N3 disease with AUC 0.64 (95% CI: 0.56; 0.72). Effective radius was most important feature for nodal disease prediction. CONCLUSIONS:Our results show that DCE MRI based radiomic phenotypes were able to predict nodal involvement and distant metastasis in breast cancer patients. Quantitative breast DCE MRI radiomics shows promise for noninvasive image based phenotyping for prediction of nodal and distant metastatic disease in breast cancer patients. Citation Format: Gaiane Margishvili Rauch, Karen Drukker, Nabil Elshafeey, Rania M.m. Mohamed, Medina Boge, Beatriz E. Adrada, Rosalind P Candelaria, Mo Salama, Irene Shkatova, Maryellen Giger, Wei T Yang. Quantitative dynamic contrast-enhanced (DCE) MRI radiomic phenotypes for prediction of nodal and distal metastasis in breast cancer patients [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS3-01.

Background and Purpose:Early and accurate assessment ofbreast cancer response to NAST is important for patient management. In this study, we investigated the value of radiomic phenotypes derived from semi-quantitative and quantitative DCE-MRI parametric maps for early prediction of NASTresponse in TNBC patients. MATERIALS AND METHODS:This IRB approved study included 74 patients with stage I-III TNBC who were enrolled in the prospective ARTEMIS trial (NCT02276443). Pathologic complete response (pCR) and non-pCR were assessed by surgical histopathology after NAST (pCR=34; non-pCR=40).MRI scans were obtained at 3 time points during the NAST treatment with every 2-week anthracycline-based chemotherapy (AC): at baseline (BSL=74), post-2 cycles of AC (C2= 27) and post-4 cycles of AC (C4= 27). Patients went on to receive taxane-based chemotherapy prior to surgery. Tumor regions of interest (ROIs) were segmented by a breast radiologist at the early-phase subtractions of DCE-MRI scans using in-house developed software, followed by co-registration of the ROIs with quantitative (Ktrans, Veand Kep), and semi-quantitative DCE parametric maps (Maximum Slope Increase (MSI), Positive Enhancement Integral (PEI) and Peak Signal Enhancement Ratio (SER)).A total of 93 first order radiomic features were extracted from the tumor ROIs of each time point semi-quantitative DCE parametric map, while a total of 390 extracted radiomic features (first order-histogram features and second order grey-level-co-occurrence matrix) were extracted from each quantitative DCE parametric map using an in-house developed Matlab software.Radiomic features at each time point and changes between the 3 time points were compared between pCR and non-pCR using Wilcoxon Rank Sum test and Fisher’s exact test. Area under the receiver operating characteristics curve (AUC) was used to determine which features predicted pCR.Logistic regression was performed for feature selection, and used to build the radiomic phenotype model. The model performance was assessed by leave-one-out cross validation and 3-fold cross validation. RESULTS:Thirty-three radiomic features from PEI map were significantly different between pCR and non-pCR. The PEI most significant features were changesbetween BSL and C4 in skewness, mean and median (AUC=0.87, 0.85 and 0.87, p= Citation Format: Nabil Elshafeey, Beatriz E Adrada, Rosalind P Candelaria, Abeer H Abdelhafez, Benjamin C Musall, Jia Sun, Medine Boge, Rania M.M Mohamed, Hagar S Mahmoud, Jong Bum Son, Aikaterini Kotrosou, Shu Zhang, Jessica Leung, Deanna Lane, Marion Scoggins, David Spak, Elsa Arribas, Lumarie Santiago, Gary J. Whitman, Huong T Le-Petross, Tanya W Moseley, Jason B White, Elizabeth Ravenberg, Ken-Pin Hwang, Peng Wei, Jennifer K Litton, Lei Huo, Debu Tripathy, Vicente Valero, Alastair M Thompson, Stacy Moulder, Wei T Yang, Mark D Pagel, Jingfei Ma, Gaiane M Rauch. Radiomic phenotypes from dynamic contrast-enhanced MRI (DCE-MRI) parametric maps for early prediction of response to neoadjuvant systemic therapy (NAST) in triple negative breast cancer (TNBC) patients [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PD6-06.

10067 Background: Metastatic melanoma pt outcomes have been revolutionized by targeted therapy (TT) and immune checkpoint blockade (ICB), which are now being evaluated in the neoadjuvant (neoadj) setting. While tumor-based biomarkers may help predict response, predictors of response obtained by less invasive strategies could greatly benefit pt care and allow real-time treatment response monitoring. Radiomic signatures derived from computerized tomography (CT) images have recently been shown to predict response to ICB in stage IV pts. However, the association of radiomic features with pathological response following neoadj therapy has not been assessed. We sought to determine if radiomic assessment predicts pCR in pts receiving neoadj TT and ICB. Methods: We collected data for a cohort of melanoma pts with locoregional metastases who were treated with neoadj TT (n = 33) or ICB (n = 30). Pts received systemic therapy for 8-10 weeks prior to planned surgical resection. Responses were evaluated radiographically (RECIST 1.1) and via pathological assessment (evaluating for pathologic complete response; (pCR) versus < pCR). Thirty two pts (19 ICB; 13 TT) were included in the radiomics analysis based on the availability of appropriate CT imaging. A total of 310 unique radiomic features (10 histogram-based and 300 second-order texture features) were calculated from each extracted volume of interest (VOI). Feature extraction was performed on baseline and initial on-treatment pre-operative CT scans. Features associated with pCR were assessed using a feature selection approach based on Least Absolute Shrinkage and Selection Operator (LASSO). Selected features were used to build a classification model for prediction of pCR to ICB or TT. Leave-One-Out Cross-Validation was performed to evaluate the robustness of the estimates. Results: Out of 310 radiomic features, three features measured at baseline were able to predict a pCR to neoadj ICB or TT with sensitivity, specificity and accuracy of 100%, though these signatures were non-overlapping. In the on-treatment pre-operative scans, 3 distinct features (also non-overlapping and distinct from the predictive pre-treatment signatures) also predicted pCR to ICB and TT with 100% sensitivity, specificity and accuracy. Conclusions: Radiomic signatures in baseline and on-treatment CT scans accurately predict pCR in melanoma pts with locoregional metastases treated with neoadj TT or ICB. These provocative findings warrant further investigation in larger, independent cohorts.

66 Background: To predict responders versus non-responders to Pembrolizumab, an anti PD-1 monoclonal antibody, in patients with advanced rare cancers. Methods: The study included 58 patients with advanced rare cancers (eg. squamous cell carcinoma of the skin, adrenocortical carcinoma, carcinoma of unknown primary, and paraganglioma) who were enrolled in a phase 2 trial of Pembrolizumab. Tumor response was evaluated using Response Evaluation Criteria in Solid Tumors (RECIST) 1.1. Patients were categorized into: 21 responders (stable disease, partial response, and complete response) and 37 non-Responders (progressive disease). Target lesion(s) obtain from standard-of-care, pre-treatment contrast enhanced CT scans were segmented using 3D slicer v4.8.1. A total of 610 features (10 histogram-based and 600 second-order texture features) were calculated from each extracted volume of interest (VOI). Radiomic features were obtained using a feature selection approach based on Least Absolute Shrinkage and Selection Operator (LASSO). Selected features were used to build a classification model, using XGboost, for prediction of tumor response to Pembrolizumab. To evaluate the robustness of the estimates, Leave-One-Out Cross-Validation (LOOCV) was performed. Results: A total of 10 radiomic features were selected; the XGboost-based classification robustly differentiated between responders vs non-responders (area under the curve, sensitivity and specificity were 99%, 100%, and 95%, respectively [p

Purpose: Radiomics is the extraction of multidimensional imaging features, which when correlated with genomics, is termed radiogenomics. However, radiogenomic biological validation is not sufficiently described in the literature. We seek to establish causality between differential gene expression status and MRI-extracted radiomic-features in glioblastoma. Experimental Design: Radiogenomic predictions and validation were done using the Cancer Genome Atlas and Repository of Molecular Brain Neoplasia Data glioblastoma patients (n = 93) and orthotopic xenografts (OX; n = 40). Tumor phenotypes were segmented, and radiomic-features extracted using the developed radiome-sequencing pipeline. Patients and animals were dichotomized on the basis of Periostin (POSTN) expression levels. RNA and protein levels confirmed RNAi-mediated POSTN knockdown in OX. Total RNA of tumor cells isolated from mouse brains (knockdown and control) was used for microarray-based expression profiling. Radiomic-features were utilized to predict POSTN expression status in patient, mouse, and interspecies. Results: Our robust pipeline consists of segmentation, radiomic-feature extraction, feature normalization/selection, and predictive modeling. The combination of skull stripping, brain-tissue focused normalization, and patient-specific normalization are unique to this study, providing comparable cross-platform, cross-institution radiomic features. POSTN expression status was not associated with qualitative or volumetric MRI parameters. Radiomic features significantly predicted POSTN expression status in patients (AUC: 76.56%; sensitivity/specificity: 73.91/78.26%) and OX (AUC: 92.26%; sensitivity/specificity: 92.86%/91.67%). Furthermore, radiomic features in OX were significantly associated with patients with similar POSTN expression levels (AUC: 93.36%; sensitivity/specificity: 82.61%/95.74%; P = 02.021E−15). Conclusions: We determined causality between radiomic texture features and POSTN expression levels in a preclinical model with clinical validation. Our biologically validated radiomic pipeline also showed the potential application for human–mouse matched coclinical trials.

PURPOSE: The mainstay imaging technique in brain tumor is Magnetic resonance imaging (MRI). However, early detection of tumor cell proliferation using MRI is limited due to inapparent disruption of normal brain architecture. Radiomics and machine learning techniques can quantitate thousands of imaging features that can depict neoplastic changes in apparently normal brain. Herein, we investigate the potential role radiomics can play in early detection of tumor cell proliferation in apparently normal MRI using a preclinically trained radiomic. METHODS: Two glioblastoma stem-like cell lines were transformed to stably express luciferase under a constitutive promoter. A stereotactic injection of tumor cells was performed to generate orthotopic mouse models (N=48). Tumor cell engraftment and in-vivo proliferation were assessed using bio-luminescence imaging (BLI) along with a weekly MRI (Bruker 7T). Images were analyzed, and ROIs were placed using 3D slicer software and radiomic features were extracted using Matlab. ROIs (0.75 mm) were placed on tumor injection sites and normal appearing contralateral brain. Radiomic features were compared for their significant alterations over time using comparative marker selection (CMS). Genomics and Histopathology of tumors were performed ex-vivo. Validation was performed in a cohort of brain cancer patients. RESULTS: Three stages of post-implantation tumor cell presence and proliferation were identified: 1. Immediate post implantation lag/engraftment phase. 2. Linear cellular proliferation phase (normal on conventional MRI). 3. Exponential cellular proliferation phase (apparent tumor on conventional MRI). Our data showed that 43% of extracted radiomic features were significantly changing (P Conclusion: Radiomic texture analysis and machine learning detects tumor cell presence and proliferation in normal-appearing brain prior to tumor development on conventional imaging. CLINICAL RELEVANCE: Radiomics and machine learning algorithms are predictive of tumor presence in seemingly normal MRIs. Early detection of tumors can allow earlier intervention, more extensive radiation planning and appropriately dose chemotherapeutic regimens.

BACKGROUND: Radiomic texture analysis (TA) from standard MRI imaging may be able to discriminate between responders versus non-responders in glioblastoma patients treated with pembrolizumab immunotherapy. METHODS: 14 patients (5 males; mean age 58 years; range: 32–72 years), with pathologically-proven recurrent GBM, enrolled in a pembrolizumab clinical trial, were retrospectively evaluated. Immunotherapy Response Assessment in Neuro-Oncology(iRANO) were performed. Patients were categorized based on: 1) best response or 2) overall response (OR) using the iRANO status at the last scan time in the trial. Patients with progressive disease (PD) were classified as non-responders, while patients with partial response (PR) or stable disease (SD) were classified as responders. T2-FLAIR (edema/invasion) and post-contrast T1WI (enhancing tumor) of baseline scans were co-registered and segmented (3D Slicer, v.4.3.1) to create a volume of interest for Radiomic TA. A total of 4880 texture features were extracted. Feature selection was performed using Lasso regularization. For classification and predictive model building, gbtree booster of XGboost with Leave-One-Out Cross-Validation (LOOCV) was used on the selected texture features to build a binary logistic regression model and classify the patients into respective groups RESULTS: Using the best response classification, 10 patients were classified as non-responders and four patients classified as responders (1 SD; 3 PR). Using 13 radiomic features, these patients could be classified into their respective responding groups with a sensitivity, specificity and accuracy of 100%, p-value=0.0089. Based on OR, 12 patients were classified as non-responders and two as responders (2 SD). Seven features were able to differentiate the responding patients with a sensitivity, specificity and accuracy of 100%, p-value=0.0089. CONCLUSION: Radiomic TA was able to discriminate and predict those GBM patients that are responders versus non-responders to pembrolizumab with high robustness. Of note, given the small number of patients in this cohort, a larger cohort of patients is needed to minimize overfitting.

Treatment response and survival after bevacizumab failure remains poor in patients with glioblastoma. Several recent publications examining glioblastoma patients treated with bevacizumab have described specific radiographic patterns of disease progression as correlating with outcome. This study aims to scrutinize these previously reported radiographic prognostic models in an independent data set to inspect their reproducibility and potential for clinical utility. Sixty four patients treated at MD Anderson matched predetermined inclusion criteria. Patients were categorized based on previously published data by: (1) Nowosielski et al. into: T2-diffuse, cT1 Flare-up, non-responders and T2 circumscribed groups (2) Modified Pope et al. criteria into: local, diffuse and distant groups and (3) Bahr et al. into groups with or without new diffusion-restricted and/or pre-contrast T1-hyperintense lesions. When classified according to Nowosielski et al. criteria, the cT1 Flare-up group had the longest overall survival (OS) from bevacizumab initiation, with non-responders having the worst outcomes. The T2 diffuse group had the longest progression free survival (PFS) from start of bevacizumab. When classified by modified Pope at al. criteria, most patients did not experience a shift in tumor pattern from the pattern at baseline, while the PFS and OS in patients with local-to-local and local-to-diffuse/distant patterns of progression were similar. Patients developing restricted diffusion on bevacizumab had worse OS. Diffuse patterns of progression in patients treated with bevacizumab are rare and not associated with worse outcomes compared to other radiographic subgroups. Emergence of restricted diffusion during bevacizumab treatment was a radiographic marker of worse OS.

INTRODUCTION Hypermutation is the excessive accumulation of DNA mutation in cancer cells. This specific hypermutated genotype has been reported in low and high grade gliomas, specifically post-temozolomide treatment and is associated with treatment-resistance. Herein, we sought to identify an imaging-based signature for hypermutated gliomas using a radiomics-based approach. MATERIALS AND METHODS In this IRB-approved retrospective study, we analyzed a total of 101 patients with primary gliomas from the University of Texas MD Anderson Cancer Center. Next generation sequencing (NGS) platforms (T200 and Foundation 1) were used to determine the Mutation burden status in post-biopsy (stereotactic/excisional). Patients were dichotomized based on their mutation burden; 77 hypomutated (=30 mutations or RESULTS We found 100 radiomic features that can discriminate between hypermutated versus hypomutated gliomas, AUC 96.3% (CI: 90.2%-98.9%), Sensitivity 100%, Specificity 95%, p-value=3.769e-6. CONCLUSION Hypermutated gliomas has a unique radiomic quantitative signature that can be used to predict mutation burden regardless of tumor grade or histopathology. Citation Format: Islam Hassan, Aikaterini Kotrotsou, Carlos Kamiya Matsuoka, Kristin D. Alfaro-Munoz, Nabil Elshafeey, Nancy El Shafei, Pascal O. Zinn, John F. de Groot, Rivka R. Colen. A radiomic-based MRI phenotype is uniquely associated with hypermutated genotype in gliomas [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2955.

BACKGROUND: Treatment-related imaging changes are often difficult to distinguish from true tumor progression. Treatment-related changes or pseudoprogression (PsP) subsequently subside or stabilize without any further treatment, whereas progressive tumor requires a more aggressive approach in patient management. Pseudoprogression can mimic true progression radiographically and may potentially alter the physician's judgment about the recurrent disease. Hence, it can predispose a patient to overtreatment or be categorized as a non-responder and exclude him from the clinical trials. This study aims at assessing the potential of radiomics to discriminate PsP from progressive disease (PD) in glioblastoma (GBM) patients. METHODS: We retrospectively evaluated 304 GBM patients with new or increased enhancement on conventional MRI after treatment, of which it was uncertain for PsP versus PD. 149 patients had the histopathological evidence of PD and 27 of PsP. Remaining 128 patients were categorized into PD and PsP based on RANO criteria performed by a board-certified radiologist. Volumetrics using 3D slicer 4.3.1 and radiomics texture analysis were performed of the enhancing lesion(s) in question. RESULTS: Using the MRMR feature selection method, we identified 100 significant features that were used to build a SVM model. Five texture features (E, CS, SA, MP, CP) were found to be most predictive of pseudoprogression. On Leave One Out Cross-Validation (LOOCV), sensitivity, specificity and accuracy were 97%, 72%, and 90%, respectively. Using 70% of the patient data for training and 30% for validation, an AUC of 94% was achieved, with the sensitivity of 97% and specificity of 75%. CONCLUSION: 3D radiomic texture features of conventional MRI successfully discriminated pseudoprogression from true progression in a large cohort of GBM patients. Citation Format: Srishti Abrol, Aikaterini Kotrotsou, Ahmed Hassan, Nabil Elshafeey, Tagwa Idris, Naveen Manohar, Anand Agarwal, Islam Hassan, Kamel Salek, Nikdokht Farid, Carrie McDonald, Shiao-Pei Weathers, Naeim Bahrami, Samuel Bergamaschi, Ahmed Elakkad, Kristin Alfaro-Munoz, Fanny Moron, Jason Huse, Jeffrey Weinberg, Sherise Ferguson, Evangelos Kogias, Amy Heimberger, Raymond Sawaya, Ashok Kumar, John de Groot, Meng Law, Pascal Zinn, Rivka R. Colen. Radiomics discriminates pseudo-progression from true progression in glioblastoma patients: A large-scale multi-institutional study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3040.

2031Background: Radiomics derives quantitative features from medical images to reveal novel information about imaging phenotypes. Although significant research has been conducted on the application...

2022Background: Hypermutation is defined as the excessive accumulation of DNA mutation in cancer cells and is reported in several forms of cancer including low and high grade gliomas. Incidence of ...

2015 Background: Treatment-related imaging changes are often difficult to distinguish from true tumor progression. Treatment-related changes or pseudoprogression (PsP) subsequently subside or stabilize without any further treatment, whereas progressive tumor requires a more aggressive approach in patient management. Pseudoprogression can mimic true progression radiographically and may potentially alter the physician’s judgment about the residual disease. Hence, it can predispose a patient to overtreatment or be categorized as a non-responder and exclude him from the clinical trials. This study aims at assessing the potential of radiomics to discriminate PsP from progressive disease (PD) in glioblastoma (GBM) patients. Methods: We retrospectively evaluated 304 GBM patients with new or increased enhancement on conventional MRI after treatment, of which it was uncertain for PsP versus PD. 149 patients had the histopathological evidence of PD and 27 of PsP. Remaining 128 patients were categorized into PD and PsP based on RANO criteria performed by a board-certified radiologist. Volumetrics using 3D slicer 4.3.1 and radiomics texture analysis were performed of the enhancing lesion(s) in question. Results: Using the MRMR feature selection method, we identified 100 significant features that were used to build a SVM model. Five texture features (E, CS, SA, MP, CP) were found to be most predictive of pseudoprogression. On Leave One Out Cross-Validation (LOOCV), sensitivity, specificity and accuracy were 97%, 72%, and 90%, respectively. Conclusions: 3D radiomic texture features of conventional MRI successfully discriminated pseudoprogression from true progression in a large cohort of GBM patients.

INTRODUCTION: Patterns of progression following bevacizumab (bev) treatment and associated outcomes remain poorly characterized. In patients (pts) with glioblastoma (GB) treated with bev, we describe radiographic patterns of progression and their association with outcome. METHODS: 64 pts treated at MD Anderson matched the predetermined inclusion criteria. Tumor progression after bev treatment was assessed according to the RANO criteria and pts categorized into groups based on previously published data: Group1:exclusively T2-diffuse hyperintense tumor (T2-diffuse), Group2:initial decrease and subsequent flare-up of contrast enhancement (CE) at progression (cT1 Flare-up), Group3:no decrease in CE or development of new lesions at first follow-up imaging (non- responders), Group4:exclusively T2-circumscribed hyperintense tumor progression (T2-circumscribed). In addition, we screened for new diffusion-restricted lesions or pre-contrast T1-hyperintense lesions or both (double-positive). RESULTS: Pts were categorized into Group1:11%, Group2: 33%, Group3: 45%, Group4: 11%. 16 pts had T1-hyperintense lesions and 37 had restricted diffusion;10 pts had double-positive lesions. There was no significant difference in time-to-initiation of bev treatment in the 4 groups. After starting bev, median OS and PFS (months) was Group1:8.6, 4.2 Group2:12.3,3.9 Group3:5.6,1.4 and Group4:7.0,3.2 respectively. Comparing non-responders vs the rest of the groups (responders), OS from initiation of bev was 5.6 vs 10 months (p =