Predicting Molecular Subtype and Survival of Rhabdomyosarcoma Patients Using Deep Learning of H&E Images: A Report from the Children's Oncology Group

PURPOSE

Rhabdomyosarcoma (RMS) is an aggressive soft-tissue sarcoma, which primarily occurs in children and young adults. We previously reported specific genomic alterations in RMS, which strongly correlated with survival; however, predicting these mutations or high-risk disease at diagnosis remains a significant challenge. In this study, we utilized convolutional neural networks (CNN) to learn histologic features associated with driver mutations and outcome using hematoxylin and eosin (H&E) images of RMS.

EXPERIMENTAL DESIGN

Digital whole slide H&E images were collected from clinically annotated diagnostic tumor samples from 321 patients with RMS enrolled in Children's Oncology Group (COG) trials (1998-2017). Patches were extracted and fed into deep learning CNNs to learn features associated with mutations and relative event-free survival risk. The performance of the trained models was evaluated against independent test sample data (n = 136) or holdout test data.

RESULTS

The trained CNN could accurately classify alveolar RMS, a high-risk subtype associated with PAX3/7-FOXO1 fusion genes, with an ROC of 0.85 on an independent test dataset. CNN models trained on mutationally-annotated samples identified tumors with RAS pathway with a ROC of 0.67, and high-risk mutations in MYOD1 or TP53 with a ROC of 0.97 and 0.63, respectively. Remarkably, CNN models were superior in predicting event-free and overall survival compared with current molecular-clinical risk stratification.

CONCLUSIONS

This study demonstrates that high-risk features, including those associated with certain mutations, can be readily identified at diagnosis using deep learning. CNNs are a powerful tool for diagnostic and prognostic prediction of rhabdomyosarcoma, which will be tested in prospective COG clinical trials.

A cross-study analysis of drug response prediction in cancer cell lines

To enable personalized cancer treatment, machine learning models have been developed to predict drug response as a function of tumor and drug features. However, most algorithm development efforts have relied on cross-validation within a single study to assess model accuracy. While an essential first step, cross-validation within a biological data set typically provides an overly optimistic estimate of the prediction performance on independent test sets. To provide a more rigorous assessment of model generalizability between different studies, we use machine learning to analyze five publicly available cell line-based data sets: National Cancer Institute 60, ancer Therapeutics Response Portal (CTRP), Genomics of Drug Sensitivity in Cancer, Cancer Cell Line Encyclopedia and Genentech Cell Line Screening Initiative (gCSI). Based on observed experimental variability across studies, we explore estimates of prediction upper bounds. We report performance results of a variety of machine learning models, with a multitasking deep neural network achieving the best cross-study generalizability. By multiple measures, models trained on CTRP yield the most accurate predictions on the remaining testing data, and gCSI is the most predictable among the cell line data sets included in this study. With these experiments and further simulations on partial data, two lessons emerge: (1) differences in viability assays can limit model generalizability across studies and (2) drug diversity, more than tumor diversity, is crucial for raising model generalizability in preclinical screening.

Dynamic imaging of nascent RNA reveals general principles of transcription dynamics and stochastic splice site selection

The activities of RNA polymerase and the spliceosome are responsible for the heterogeneity in the abundance and isoform composition of mRNA in human cells. However, the dynamics of these megadalton enzymatic complexes working in concert on endogenous genes have not been described. Here, we establish a quasi-genome-scale platform for observing synthesis and processing kinetics of single nascent RNA molecules in real time. We find that all observed genes show transcriptional bursting. We also observe large kinetic variation in intron removal for single introns in single cells, which is inconsistent with deterministic splice site selection. Transcriptome-wide footprinting of the U2AF complex, nascent RNA profiling, long-read sequencing, and lariat sequencing further reveal widespread stochastic recursive splicing within introns. We propose and validate a unified theoretical model to explain the general features of transcription and pervasive stochastic splice site selection.

A Deep Learning Pipeline for Nucleus Segmentation

Deep learning is rapidly becoming the technique of choice for automated segmentation of nuclei in biological image analysis workflows. In order to evaluate the feasibility of training nuclear segmentation models on small, custom annotated image datasets that have been augmented, we have designed a computational pipeline to systematically compare different nuclear segmentation model architectures and model training strategies. Using this approach, we demonstrate that transfer learning and tuning of training parameters, such as the composition, size, and preprocessing of the training image dataset, can lead to robust nuclear segmentation models, which match, and often exceed, the performance of existing, off-the-shelf deep learning models pretrained on large image datasets. We envision a practical scenario where deep learning nuclear segmentation models trained in this way can be shared across a laboratory, facility, or institution, and continuously improved by training them on progressively larger and varied image datasets. Our work provides computational tools and a practical framework for deep learning-based biological image segmentation using small annotated image datasets. Published [2020]. This article is a U.S. Government work and is in the public domain in the USA.

Spatial analysis of tumor immune microenvironment (TIME) in patients treated with Bintrafusp alfa

3070

Background: Bintrafusp alfa is a first-in-class bifunctional fusion protein composed of the extracellular domain of TGF-βRII receptor (TGF-β “trap”) fused to a human IgG1 mAb blocking PD-L1. In preclinical models, bintrafusp alfa treatment promoted CD8+ T cell and NK cell activation, and both immune cell (IC) populations were required for optimal bintrafusp alfa mediated tumor control. However, the effect of bintrafusp alfa on TIME in humans has not been reported. Methods: In this unplanned interim analysis of a biomarker expansion cohort (NCT 02517398), patients (pts) with advanced non-small cell lung cancer (NSCLC) underwent paired biopsies (bx) before and on treatment with bintrafusp alfa (~ 50 days apart). The objective was to evaluate frequency and localization of tumor infiltrated ICs by IHC. Out of 12 pts, 7 had matched (Pre vs Post) tumor-containing specimens sufficient for multiplex immunofluorescence (MxIF) analysis of TIME. Four pts were excluded as Post bx histology for 3/12 [2 PR (partial response), 1 SD (stable disease)] was negative for tumor (necrosis or fibrosis) and 1/12 did not have a Post bx performed. Results: TIME study shows CD8 T cell infiltrates were increased in Post compared to Pre bx (median 161 vs 62/mm²; interquartile range [IQR] 65–396/mm² vs 31–135/mm²; p = 0·04). While M2 macrophages were also increased (median 800 vs 367/mm²; IQR 776–1131/mm² vs 171–831/mm²; p = 0·04), the ratio of M1/M2 was reversed in pts with SD (↑) compared to pts with PD (↓). Other ICs such as CD4, T-regs, NK cells and M1 macrophages were not changed. On average compared to baseline, M2 macrophages were > 2 fold closer to every other IC in pts with PD, but > 2 fold further from any IC in pts with SD. Tregs were relatively closer to other IC in PD pts. Linear Discriminant Analysis was also performed and results indicate that differential IC densities (mainly M1 macrophages and CD4 T cells) do perform as classifiers between long ( > 5 months) and short ( < 5 months) term responses. Conclusions: This study suggests that bintrafusp alfa not only can enhance intratumoral effector IC infiltrates (CD8) but also has a modulating effect on the spatial distribution of both M1/M2 macrophages within the NSCLC TIME. The differential proximity of M2 macrophages to other IC infiltrates and changes in M1/M2 ratios in association with response suggests that an M1/M2 macrophage balance is directly involved in response and/or resistance to bintrafusp alfa. Given the limited number of patients in this cohort, we intend to study effects of bintrafusp alfa in a larger cohort of patients. Clinical trial information: 02517398 .

AI Meets Exascale Computing: Advancing Cancer Research With Large-Scale High Performance Computing

The application of data science in cancer research has been boosted by major advances in three primary areas: (1) Data: diversity, amount, and availability of biomedical data; (2) Advances in Artificial Intelligence (AI) and Machine Learning (ML) algorithms that enable learning from complex, large-scale data; and (3) Advances in computer architectures allowing unprecedented acceleration of simulation and machine learning algorithms. These advances help build in silico ML models that can provide transformative insights from data including: molecular dynamics simulations, next-generation sequencing, omics, imaging, and unstructured clinical text documents. Unique challenges persist, however, in building ML models related to cancer, including: (1) access, sharing, labeling, and integration of multimodal and multi-institutional data across different cancer types; (2) developing AI models for cancer research capable of scaling on next generation high performance computers; and (3) assessing robustness and reliability in the AI models. In this paper, we review the National Cancer Institute (NCI) -Department of Energy (DOE) collaboration, Joint Design of Advanced Computing Solutions for Cancer (JDACS4C), a multi-institution collaborative effort focused on advancing computing and data technologies to accelerate cancer research on three levels: molecular, cellular, and population. This collaboration integrates various types of generated data, pre-exascale compute resources, and advances in ML models to increase understanding of basic cancer biology, identify promising new treatment options, predict outcomes, and eventually prescribe specialized treatments for patients with cancer.

Laparoscopic stereoscopic augmented reality: toward a clinically viable electromagnetic tracking solution

The purpose of this work was to develop a clinically viable laparoscopic augmented reality (AR) system employing stereoscopic (3-D) vision, laparoscopic ultrasound (LUS), and electromagnetic (EM) tracking to achieve image registration. We investigated clinically feasible solutions to mount the EM sensors on the 3-D laparoscope and the LUS probe. This led to a solution of integrating an externally attached EM sensor near the imaging tip of the LUS probe, only slightly increasing the overall diameter of the probe. Likewise, a solution for mounting an EM sensor on the handle of the 3-D laparoscope was proposed. The spatial image-to-video registration accuracy of the AR system was measured to be [Formula: see text] and [Formula: see text] for the left- and right-eye channels, respectively. The AR system contributed 58-ms latency to stereoscopic visualization. We further performed an animal experiment to demonstrate the use of the system as a visualization approach for laparoscopic procedures. In conclusion, we have developed an integrated, compact, and EM tracking-based stereoscopic AR visualization system, which has the potential for clinical use. The system has been demonstrated to achieve clinically acceptable accuracy and latency. This work is a critical step toward clinical translation of AR visualization for laparoscopic procedures.

The Utility of Cloud Computing in Analyzing GPU-Accelerated Deformable Image Registration of CT and CBCT Images in Head and Neck Cancer Radiation Therapy

The images generated during radiation oncology treatments provide a valuable resource to conduct analysis for personalized therapy, outcomes prediction, and treatment margin optimization. Deformable image registration (DIR) is an essential tool in analyzing these images. We are enhancing and examining DIR with the contributions of this paper: 1) implementing and investigating a cloud and graphic processing unit (GPU) accelerated DIR solution and 2) assessing the accuracy and flexibility of that solution on planning computed tomography (CT) with cone-beam CT (CBCT). Registering planning CTs and CBCTs aids in monitoring tumors, tracking body changes, and assuring that the treatment is executed as planned. This provides significant information not only on the level of a single patient, but also for an oncology department. However, traditional methods for DIR are usually time-consuming, and manual intervention is sometimes required even for a single registration. In this paper, we present a cloud-based solution in order to increase the data analysis throughput, so that treatment tracking results may be delivered at the time of care. We assess our solution in terms of accuracy and flexibility compared with a commercial tool registering CT with CBCT. The latency of a previously reported mutual information-based DIR algorithm was improved with GPUs for a single registration. This registration consists of rigid registration followed by volume subdivision-based nonrigid registration. In this paper, the throughput of the system was accelerated on the cloud for hundreds of data analysis pairs. Nine clinical cases of head and neck cancer patients were utilized to quantitatively evaluate the accuracy and throughput. Target registration error (TRE) and structural similarity index were utilized as evaluation metrics for registration accuracy. The total computation time consisting of preprocessing the data, running the registration, and analyzing the results was used to evaluate the system throughput. Evaluation showed that the average TRE for GPU-accelerated DIR for each of the nine patients was from 1.99 to 3.39 mm, which is lower than the voxel dimension. The total processing time for 282 pairs on an Amazon Web Services cloud consisting of 20 GPU enabled nodes took less than an hour. Beyond the original registration, the cloud resources also included automatic registration quality checks with minimal impact to timing. Clinical data were utilized in quantitative evaluations, and the results showed that the presented method holds great potential for many high-impact clinical applications in radiation oncology, including adaptive radio therapy, patient outcomes prediction, and treatment margin optimization.

Nitric oxide synthase expression is downregulated in basal cell carcinoma of the head and neck

The small molecule nitric oxide (NO) has many actions, most of which are poorly understood. Recently, NO and related compounds have been implicated in skin damage caused by ultraviolet light although their exact role is not clear. We undertook an immuno histochemical study to assess the expression of type II NO synthase (NOS2) and type III (NOS3) in basal cell carcinomas (BCCs) of the head and neck. In all 48 cases studied, NOS2 was found in the basal cell layer of the skin at the tumour margin but it w as significantly reduced in the tumour epithelial cells (P=0.001). NOS3 was localized to the endothelium of the blood vessels in both skin and tumour in all cases, and it was not seen in the tumour epithelial cells. The results suggest that expression of NOS is down-regulated in basal cell carcinomas.

Allergy to local anaesthetic: the importance of thorough investigation

A case report is presented which highlights the importance of a good history in arriving at the correct diagnosis in cases where allergy to local anaesthetic is suspected. Management of the patient is discussed and the topic of 'adverse reaction' briefly reviewed.

A double-blind comparative study of soluble aspirin and diclofenac dispersible in the control of postextraction pain after removal of impacted third molars

The analgesic efficacy and patient acceptability of soluble aspirin and diclofenac dispersible were compared in patients with postoperative pain after removal of impacted third molars. A total of 136 patients were randomly allocated to receive soluble aspirin 600 mg tds or diclofenac dispersible 50 mg tds after extraction under local anaesthesia of impacted third molars on one side of the mouth. The medication, which was both patient and operator blind, was reversed after extraction of the contralateral third molars 3 weeks later, the patients acting as their own controls in assessing postoperative pain, pain relief, and interincisal opening. Patients receiving diclofenac dispersible recorded significantly lower pain levels; pain relief was significantly greater and the patients' assessment significantly favoured diclofenac dispersible. Interincisal opening throughout the study period was significantly increased in the diclofenac dispersible group. The surgeons' postoperative assessment of extraction sites showed no significant difference between the two treatment groups in rate of healing. Two patients reported side-effects while taking soluble aspirin, and eight while taking diclofenac dispersible, two of whom discontinued treatment.

Excess vitamin A injures the liver

Chronic vitamin A intoxication in a 56-year-old female is reported. Some abnormal blood chemistries included elevated transaminase and alkaline phosphatase, increased cerebrospinal fluid and portal pressure, and elevated vitamin A in blood and liver. A liver biopsy indicated histologic evidence of perisinusoidal collagen deposition and noncoalescent fat droplets in Ito cells. Caution against the misdiagnosis of alcoholic cirrhosis for vitamin A intoxication is recommended.