Prodduturi, Naresh

Abstract Deep learning has advanced digital pathology and multi-omics data analysis. However, there is still a critical need to integrate these modalities to better understand complex diseases. Deep learning models, such as Prov-Gigapath, a state-of-the-art pathology foundation model, have been developed for analyzing whole-slide images (WSI). Our previously published deep-learning framework, OmicsFootPrint (OFP), transforms high-dimensional multi-omics data into intuitive genomically-arranged circular images and utilizes transfer learning to overcome the challenges of high-dimensional, low-sample-size data. In this study, we aimed to combine WSI and multi-omics data to develop a framework capable of distinguishing between invasive ductal carcinoma (IDC) and invasive lobular carcinoma (ILC), which is clinically significant as it may help explain differential responses to therapies, such as tamoxifen versus letrozole. We enhanced the performance of our OFP model by dividing each multi-omics circular image (1024x1024x3 pixels) into 16 equal-sized patches (256x256x3 pixels). Each patch was trained using a separate EfficientNetV2 architecture with pre-trained ImageNet weights. The probability outcomes from these 16 models were subsequently combined through a meta-learner model using the AutoGluon framework. Concurrently, WSI images corresponding to the same patients were independently trained, maintaining consistent train/validation/test splits. The Prov-Gigapath model, which was trained using a dataset from the Providence Health System, was employed to extract patch-level embeddings from the WSI data. These embeddings were fine-tuned using a LongNet-based masked autoencoder to predict outcomes. Finally, predictions from the OFP model and the WSI-based Prov-Gigapath model were integrated through a meta-learner, preserving the original train/validation/test split. Model performances (AUC) were evaluated using the held-out test set. The gold standard used for the analysis was histological identification of IDC versus ILC, against which the performance of omics data alone, WSI data alone, and the multi-modal integration approach were compared. We evaluated our integrated model as a proof-of-concept using The Cancer Genome Atlas (TCGA) breast cancer cohort, comprising 92 ILC and 391 IDC samples. Multi-omics data, including gene expression, microRNA, copy number variation, and RPPA, were analyzed using the OFP framework, while WSI data were processed using the Prov-Gigapath method. The results demonstrated that the multi-modal integration approach significantly outperformed individual modalities, achieving an AUC of 0.98 compared to 0.87 for OFP (original), 0.95 for OFP (16-patches), and 0.92 for WSI. Integrating WSI and multi-omics data improves predictive accuracy, significantly improving the classification of breast cancer subtypes over histology alone. Citation Format: Naresh Prodduturi, Xiaojia Tang, Hamid R. Tizhoosh, Kevin J. Thompson, Richard Weinshilboum, Karthik V. Giridhar, Judy C. Boughey, Eric W. Klee, Liewei Wang, Matthew P. Goetz, Vera Suman, Krishna R. Kalari. A multi-modal learning framework to integrate digital pathology image and multi-omics data in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2429.

Abstract The OmicsFootPrint framework addresses the need for advanced multi-omics data analysis methodologies by transforming data into intuitive two-dimensional circular images and facilitating the interpretation of complex diseases. Utilizing deep neural networks and incorporating the SHapley Additive exPlanations algorithm, the framework enhances model interpretability. Tested with The Cancer Genome Atlas data, OmicsFootPrint effectively classified lung and breast cancer subtypes, achieving high area under the curve (AUC) scores—0.98 ± 0.02 for lung cancer subtype differentiation and 0.83 ± 0.07 for breast cancer PAM50 subtypes, and successfully distinguished between invasive lobular and ductal carcinomas in breast cancer, showcasing its robustness. It also demonstrated notable performance in predicting drug responses in cancer cell lines, with a median AUC of 0.74, surpassing nine existing methods. Furthermore, its effectiveness persists even with reduced training sample sizes. OmicsFootPrint marks an enhancement in multi-omics research, offering a novel, efficient and interpretable approach that contributes to a deeper understanding of disease mechanisms.

Abstract Breast cancer is one of the most common cancers in women. With early diagnosis, some breast cancers are highly curable. Classifying normal versus tumor breast tissues from microscopy images is an ideal case to use for deep learning and could help diagnose breast cancer with higher reproducibility. Since data preprocessing and hyperparameter tuning have impacts on breast cancer classification accuracies, training a classifier with appropriate data preprocessing and optimized hyperparameters could improve breast cancer classification accuracy. Using 12 combinations of model architectures, data preprocessing, and hyperparameter configurations, the validation accuracy was calculated using the BreAst Cancer Histology (BACH) dataset. The DenseNet201, a non-specialized model architecture, with transfer learning approach achieved 98.61% validation accuracy compared to only 64.00% for the digital pathology (DP)-specialized model architecture. The combination of image data preprocessing and hyperparameters have a profound impact on the performance of deep neural networks for image classification. To identify a well-performing neural network to classify tumor versus normal breast histology, researchers should not only focus on developing new models specifically for DP, since hyperparameter tuning for existing deep neural networks in the computer vision field could also achieve a better prediction accuracy.

Background Patients with peripheral artery disease are at increased risk for major adverse cardiac events, major adverse limb events, and all‐cause death. Developing tools capable of identifying those patients with peripheral artery disease at greatest risk for major adverse events is the first step for outcome prevention. This study aimed to determine whether computer‐assisted analysis of a resting Doppler waveform using deep neural networks can accurately identify patients with peripheral artery disease at greatest risk for adverse outcome events. Methods and Results Consecutive patients (April 1, 2015, to December 31, 2020) undergoing ankle–brachial index testing were included. Patients were randomly allocated to training, validation, and testing subsets (60%/20%/20%). Deep neural networks were trained on resting posterior tibial arterial Doppler waveforms to predict major adverse cardiac events, major adverse limb events, and all‐cause death at 5 years. Patients were then analyzed in groups based on the quartiles of each prediction score in the training set. Among 11 384 total patients, 10 437 patients met study inclusion criteria (mean age, 65.8±14.8 years; 40.6% women). The test subset included 2084 patients. During 5 years of follow‐up, there were 447 deaths, 585 major adverse cardiac events, and 161 MALE events. After adjusting for age, sex, and Charlson comorbidity index, deep neural network analysis of the posterior tibial artery waveform provided independent prediction of death (hazard ratio [HR], 2.44 [95% CI, 1.78–3.34]), major adverse cardiac events (HR, 1.97 [95% CI, 1.49–2.61]), and major adverse limb events (HR, 11.03 [95% CI, 5.43–22.39]) at 5 years. Conclusions An artificial intelligence–enabled analysis of Doppler arterial waveforms enables identification of major adverse outcomes among patients with peripheral artery disease, which may promote early adoption and adherence of risk factor modification.

Introduction: Predictive algorithms using multiple clinical variables can identify patients at higher risk for abdominal aortic aneurysms (AAA), however they are cumbersome and cannot used directly by patients. A simple, cheap, noninvasive, and readily available (virtual) screening tool for identification of AAA could aid in screening algorithms. Aims: We sought to determine if a machine learning algorithm could predict the presence of an AAA by using a image of the face Methods: Diagnostic imaging studies were extracted from the electronic health record (EHR) from 5/7/2018 through 1/1/2023 and analyzed by a natural language processing algorithm previously validated to identify abdominal aortic aneurysms. Patient EHR profile pictures were extracted and matched to imaging studies. Various deep neural network (DNN) architectures were explored, all trained as classifiers on the face images along with clinical variables to predict clinical outcomes. All models used were CNNs with an EfficientNet architecture. Model performance was evaluated by standard metrics such as the area under the curve (AUC), specificities (Sp), and sensitivities (Sn). All performance metrics reported results from the test subset of the primary dataset and did not contain any observations considered during training. Results: A total of 5522 patients with facial images and AAA diagnostic studies were analyzed, among whom 1314 (23.8%) had a AAA. The optimal model had an AUC of 0.72 (Sn 0.44, Sp 0.83) for predicting AAA and remained similar in women (AUC 0.7), men (AUC 0.65), age >65 (AUC 0.63), and age <=65 (AUC 0.72) (Table). Conclusions: Using only an image of a patients face, the presence of an AAA could be identified with moderate performance in men, women, and in those less than 65. The model’s performance in demographics not typically screened for AAA may allow for targeted screening programs in these groups.

Background: Patients with peripheral artery disease (PAD) are at increased risk for major adverse limb and cardiac events including mortality. Developing screening tools capable of accurate PAD identification is a necessary first step for strategies of adverse outcome prevention. This study aimed to determine whether machine analysis of a resting Doppler waveform using deep neural networks can accurately identify patients with PAD. Methods: Consecutive patients (4/8/2015 – 12/31/2020) undergoing rest and postexercise ankle–brachial index (ABI) testing were included. Patients were randomly allocated to training, validation, and testing subsets (70%/15%/15%). Deep neural networks were trained on resting posterior tibial arterial Doppler waveforms to predict normal (> 0.9) or PAD (⩽ 0.9) using rest and postexercise ABI. A separate dataset of 151 patients who underwent testing during a period after the model had been created and validated (1/1/2021 – 3/31/2021) was used for secondary validation. Area under the receiver operating characteristic curves (AUC) were constructed to evaluate test performance. Results: Among 11,748 total patients, 3432 patients met study criteria: 1941 with PAD (mean age 69 ± 12 years) and 1491 without PAD (64 ± 14 years). The predictive model with highest performance identified PAD with an AUC 0.94 (CI = 0.92–0.96), sensitivity 0.83, specificity 0.88, accuracy 0.85, and positive predictive value (PPV) 0.90. Results were similar for the validation dataset: AUC 0.94 (CI = 0.91–0.98), sensitivity 0.91, specificity 0.85, accuracy 0.89, and PPV 0.89 (postexercise ABI comparison). Conclusion: An artificial intelligence-enabled analysis of a resting Doppler arterial waveform permits identification of PAD at a clinically relevant performance level.

Abstract Background: Specific germline alterations within the TERT, EGFR, CCDC26, CDKN2A/B, PHLDB1, TP53 and RTEL1 regions are associated with development of glioma. While some of these germline variants are associated with all gliomas (e.g. rs2736100 in TERT), others are associated with a specific morphologic and/or molecular subtype. For example, the RTEL1 region variants are associated with primary glioblastoma and rs55705857 in CCDC26 is associated with 1p/19q co-deleted oligodendrogliomas and with IDH mutant astrocytic gliomas. We hypothesized that germline variants in these regions will be associated with other molecular subtypes. The Cancer Genome Atlas (TCGA) and other groups have described molecular subtypes of glioma based on acquired somatic mutation patterns. Herein, we grouped glioma patients into mutation-based molecular subgroups based on three molecular alterations and determined if germline polymorphisms are associated with these mutation-based molecular subtypes. Methods: Five molecular subtypes were defined based on combinations of the following molecular alterations: TERT promoter mutation (TERTmut vs. TERTwt), IDH1/2 mutation (IDH mut vs. IDH wt) and 1p/19q co-deletion (1p19qcodel vs. 1p19q noncodel). Germline single nucleotide polymorphisms (SNPs) were evaluated by custom Illumina array analysis of blood-derived DNA and 1K genome imputation. Results: One hundred ninety-nine gliomas (24 oligodendrogliomas, 59 mixed oligoastrocytomas, 41 grade 2-3 astrocytomas and 75 glioblastomas) had both SNP data and mutation data available in order to assign them into the five molecular subgroups. The prevalence of each of the molecular subtypes was as follows: TERTmut/IDHmut/1p19qcodel (20%), TERTmut/IDHmut/1p19qnoncodel (5%), TERTmut/IDHwt/1p19qnoncodel (32%), TERTwt/IDHmut/1p19qnondel (35%) and TERTwt/IDHwt/1p19qnoncodel (8%). The TERT SNP (rs2736100) was protective for TERTmut/IDHmut/1p19qnoncodel and TERTwt/IDHmut/1p19qnoncodel gliomas. The CCDC26 SNP (rs55705857) was strongly associated with the risk of developing TERTmut/IDHmut/1p19qcodel and TERTwt/IDHmut/1p19qnoncodel gliomas. The TP53 SNP (rs7837822) was strongly associated with risk of developing TERTmut/IDHmut/1p19qnoncodel and TERTwt/IDHmut/1p19qnoncodel gliomas. Interestingly, while one RTEL1 region SNP (rs6062297) was strongly associated with glioma risk, RTEL1 SNPs rs6010620 and rs2297440 were protective for the development of TERTmut/IDHwt/1p19qnoncodel gliomas. We are currently in the process of validating these results using the TCGA glioblastoma and low grade glioma data. Conclusions: Glioma patients can be meaningfully classified according to their acquired mutation subtype. Importantly, our results suggest that there are significant associations between germline polymorphisms and mutation-based glioma molecular subtypes. Citation Format: Jeanette E. Eckel-Passow, Thomas M. Kollmeyer, Gobinda Sarkar, Anisha Chada, Paul A. Decker, Matthew L. Kosel, Alissa A. Caron, Hugues Sicotte, Kannabiran Nandakumar, Naresh Prodduturi, Brian P. O'Neill, Daniel H. Lachance, Robert B. Jenkins. The association of glioma germline risk SNPs with mutation-based molecular subgroups. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4714. doi:10.1158/1538-7445.AM2014-4714

Abstract Recent studies have shown that associations between common genetic variation and gene expression in breast tumors provide insight into the functional mechanisms underlying breast cancer risk loci. However, the relationship between genetic variation, gene expression, and risk of triple negative (TN) breast cancer remains largely unexplored. We performed a genome-wide expression quantitative trait locus (eQTL) analysis of TN breast cancer using 668 formalin-fixed paraffin embedded TN tumors from the Triple Negative Breast Cancer Consortium (TNBCC) to identify novel genes relevant to TN breast cancer risk and to further explore the biology underlying known TN risk loci. Cis-eQTLs were defined as correlations between single nucleotide polymorphisms (SNP) and expression of genes within 1Mb, and trans-eQTLs were defined as the remainder of all SNP-gene correlations. In a genome-wide analysis, we identified 68,012 cis-eQTLs, representing correlations between 42,193 SNPs and 2,092 genes that were significant at a 10% false discovery rate (FDR). We also identified 70 trans-eQTLs across two unique genes at the same FDR threshold. Five cis-eQTLs involved SNPs previously associated with TN breast cancer risk (p&lt;5.0x10-5) in 3,700 TN cases and 4,700 controls from the TNBCC. Three of these, located on chromosomes 5p (cis-eQTL p=4.8x10-34; risk p=1.1x10-7), 7p (cis-eQTL p=6.1x10-10; risk p=1.3x10-7), and 17q (cis-eQTL p=1.8x10-15; risk p=2.9x10-7), have previously been reported in public databases of normal tissue. The remaining two TN breast cancer cis-eQTLs, located on chromosomes 8q (cis-eQTL p=6.3x10-6, risk p=8.3x10-6) and 14q (cis-eQTL p=3.1x10-6, risk p=1.2x10-5), have not been reported as eQTLs in other tissues. In addition, 12 of 25 known TN breast cancer risk loci (PEX14, MDM4, 2q31.1, ESR1, 11q13.1, 11q24.3, 12p13.1, PTHLH, NTN4, 12q24, 19p13.1, MLK1), contained at least one significant cis-eQTL. In summary, we have identified five cis-eQTLs that may identify novel risk loci for TN breast cancer. Citation Format: Kristen S. Purrington, Drakoulis Yannoukakos, Jane Carpenter, Heli Nevanlinna, Angela Cox, Gianluca Severi, Christine Ambrosone, Amanda Ewart Toland, Andrew K. Godwin, Hiltrud Brauch, Peter A. Fasching, Penelope Miron, Jenny Chang-Claude, Nicholas G. Martin, Grant W. Montgomery, Vessela Kristensen, Hoda Anton-Culver, Paul Goodfellow, Janet E. Olson, Hugues Sicotte, Naresh Prodduturi, Daniel W. Visscher, Jeanette E. Eckel-Passow, S. Keith Anderson, Seth Slettedahl, Curtis Olswold, Xianshu Wang, V. Shane Pankratz, Susan Slager, Wei Zheng, Arto Mannermaa, Ute Hamann, Diana M. Eccles, Celine M. Vachon, Fergus J. Couch. Expression quantitative trait locus analysis of triple negative breast cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3266. doi:10.1158/1538-7445.AM2014-3266