The GENIE BPC NSCLC Cohort: A Real-World Repository Integrating Standardized Clinical and Genomic Data for 1,846 Patients with Non-Small Cell Lung Cancer

PURPOSE

We describe the clinical and genomic landscape of the non-small cell lung cancer (NSCLC) cohort of the American Association for Cancer Research (AACR) Project Genomics Evidence Neoplasia Information Exchange (GENIE) Biopharma Collaborative (BPC).

EXPERIMENTAL DESIGN

A total of 1,846 patients with NSCLC whose tumors were sequenced from 2014 to 2018 at four institutions participating in AACR GENIE were randomly chosen for curation using the PRISSMM data model. Progression-free survival (PFS) and overall survival (OS) were estimated for patients treated with standard therapies.

RESULTS

In this cohort, 44% of tumors harbored a targetable oncogenic alteration, with EGFR (20%), KRAS G12C (13%), and oncogenic fusions (ALK, RET, and ROS1; 5%) as the most frequent. Median OS (mOS) on first-line platinum-based therapy without immunotherapy was 17.4 months [95% confidence interval (CI), 14.9-19.5 months]. For second-line therapies, mOS was 9.2 months (95% CI, 7.5-11.3 months) for immune checkpoint inhibitors (ICI) and 6.4 months (95% CI, 5.1-8.1 months) for docetaxel ± ramucirumab. In a subset of patients treated with ICI in the second-line or later setting, median RECIST PFS (2.5 months; 95% CI, 2.2-2.8) and median real-world PFS based on imaging reports (2.2 months; 95% CI, 1.7-2.6) were similar. In exploratory analysis of the impact of tumor mutational burden (TMB) on survival on ICI treatment in the second-line or higher setting, TMB z-score harmonized across gene panels was associated with improved OS (univariable HR, 0.85; P = 0.03; n = 247 patients).

CONCLUSIONS

The GENIE BPC cohort provides comprehensive clinicogenomic data for patients with NSCLC, which can improve understanding of real-world patient outcomes.

Accelerated curation of checkpoint inhibitor-induced colitis cases from electronic health records

Objective

Automatically identifying patients at risk of immune checkpoint inhibitor (ICI)-induced colitis allows physicians to improve patientcare. However, predictive models require training data curated from electronic health records (EHR). Our objective is to automatically identify notes documenting ICI-colitis cases to accelerate data curation.

Materials and Methods

We present a data pipeline to automatically identify ICI-colitis from EHR notes, accelerating chart review. The pipeline relies on BERT, a state-of-the-art natural language processing (NLP) model. The first stage of the pipeline segments long notes using keywords identified through a logistic classifier and applies BERT to identify ICI-colitis notes. The next stage uses a second BERT model tuned to identify false positive notes and remove notes that were likely positive for mentioning colitis as a side-effect. The final stage further accelerates curation by highlighting the colitis-relevant portions of notes. Specifically, we use BERT’s attention scores to find high-density regions describing colitis.

Results

The overall pipeline identified colitis notes with 84% precision and reduced the curator note review load by 75%. The segment BERT classifier had a high recall of 0.98, which is crucial to identify the low incidence (<10%) of colitis.

Discussion

Curation from EHR notes is a burdensome task, especially when the curation topic is complicated. Methods described in this work are not only useful for ICI colitis but can also be adapted for other domains.

Conclusion

Our extraction pipeline reduces manual note review load and makes EHR data more accessible for research.

Predicting immune checkpoint inhibitor-related pneumonitis using patient medical information

e13566

Background: Immune checkpoint inhibitors (ICI) have improved outcomes in tumor types allowing subgroups of patients to have longer, higher quality lives. However, potential life-threatening immunotoxicities can arise in susceptible patients, including pneumonitis. Identifying patients at high risk of immunotoxicity can help patients understand potential adverse events, improve clinical trial cohort selection, and inform therapy selection in clinical settings. Here, we use electronic health record (EHR) data to build a binary classification model that predicts the probability of developing pneumonitis after the first ICI administration. Methods: We utilized real-world EHR-derived structured and unstructured data from > 2,700 patients from Vanderbilt University Medical Center obtained prior to December 31, 2018. Unstructured data were transformed into structured variables by expert curators, including labels for pneumonitis episodes following ICI initiation. Feature engineering involved aggregating lab measurements over a 60-day time window before the first ICI; other features (conditions, smoking status, etc.) used a 1-year window. To build a small, easily deployable model and assess its performance robustly, we utilized a sequential process. In each step, we decided between two versions of a random forest model, one with the original feature set (M1) and one extended with a candidate feature (M2). We identified candidate features using 90% of the data. We performed nested cross-validation on this partition and compared the inner loop results. If M2 was significantly better, we tested whether it performed better on the 10% partition. If it did, we chose M2 and assessed its performance on the outer loop. This procedure was created as our dataset was rather small and noisy, which is typical for EHR-derived data. Results: All-cause pneumonitis incidence following ICI initiation was 8.4%. Our final model includes only six features: frequency of lung-related ICD-10 codes, frequency of C34 code, frequency of C78 code, smoking status, interaction between smoking and C34/C78 indicators, and median of blood oxygen saturation. This model achieved a mean AUC of 0.66 (SD: 0.07). Our analysis on the outer loop predictions showed that selecting 50% of patients with the lowest predicted probabilities reduced the occurrence of pneumonitis in the cohort to 5%, compared to 8.4%, when we select patients randomly. The model achieved a mean positive predictive value of 0.3 and negative predictive value of 0.96. Conclusions: We utilized a real-world EHR dataset to identify patterns in patient medical history that could predict the development of pneumonitis. We demonstrated that a small number of easily obtainable clinical covariates can result in meaningful predictions. This model illustrates potential future use for identifying the patients with the highest and lowest risks for pneumonitis during treatment.

Overcoming barriers in academic-industry partnerships to improve predictive modeling in immuno-oncology

e13581

Background: In the past decade, immunotherapies have revolutionized oncology practice by prolonging patient survival in previously rapidly fatal cancers. However, severe immune toxicities present a challenge, affecting ̃20% and up to 50% of patients on immune monotherapies and combination immunotherapies, respectively. Oncologists must balance toxicity risk with potential efficacy, and pharmaceutical companies have a vested interest in selecting patients with the highest benefit–risk ratio during trial enrollment. Predictive toxicity–efficacy modeling has the potential to guide trial subject selection and clinical care, yet there remains a need for predictive models that can be practically implemented in these settings. Methods: A common academic–industry contract data–transfer framework—wherein academic medical institutions and industry counterparts act in isolation—creates barriers to development of high-quality algorithms with practical applications. In this framework, 1) academic medical institutions provide patient data as a “data dump;” these data are static and cannot be refined—reducing opportunities for quality control; 2) predictive model outcomes may include artifacts that are not identified; 3) manual curation of patient data is required to accurately replicate the model in real-world settings; and 4) lack of clinician participation reduces the potential clinical applications of models and reciprocal benefit to the academic institution. We outline a more contemporary, engaged approach to unite strengths of both partners to achieve a common goal. Results: In 2019, Vanderbilt University Medical Center and GE Healthcare partnered with the goal of enabling safer and more precise immunotherapy use. As part of this work, we formulated a recipe for academic–industry partnerships that offers unique advantages over a static contract framework. In our iterative, interactive approach, 1) clinical and curation experts meet with industry modelers to dynamically refine deidentified data sets by resolving discrepancies in data from different sources (e.g., manually curated vs. structured data); 2) clinical experts iteratively review outputs of predictive models to identify potential artifacts and refine final models; 3) expert curators iterate with in-house machine-learning experts to create algorithms to automate curation of natural language elements from the identified EHR data; and 4) clinical and industry stakeholders participate in regular meetings with modelers to ensure clinical and trial utility of the modeling approach. Conclusions: Compared to data transfer-only relationships, this partnership framework offers an opportunity to develop more informed, higher quality immunotherapy models with clinical and industry applications.

Framework for Implementing and Tracking a Molecular Tumor Board at a National Cancer Institute-Designated Comprehensive Cancer Center

BACKGROUND

Over the past few years, tumor next-generation sequencing (NGS) panels have evolved in complexity and have changed from selected gene panels with a handful of genes to larger panels with hundreds of genes, sometimes in combination with paired germline filtering and/or testing. With this move toward increasingly large NGS panels, we have rapidly outgrown the available literature supporting the utility of treatments targeting many reported gene alterations, making it challenging for oncology providers to interpret NGS results and make a therapy recommendation for their patients.

METHODS

To support the oncologists at Vanderbilt-Ingram Cancer Center (VICC) in interpreting NGS reports for patient care, we initiated two molecular tumor boards (MTBs)-a VICC-specific institutional board for our patients and a global community MTB open to the larger oncology patient population. Core attendees include oncologists, hematologist, molecular pathologists, cancer geneticists, and cancer genetic counselors. Recommendations generated from MTB were documented in a formal report that was uploaded to our electronic health record system.

RESULTS

As of December 2020, we have discussed over 170 patient cases from 77 unique oncology providers from VICC and its affiliate sites, and a total of 58 international patient cases by 25 unique providers from six different countries across the globe. Breast cancer and lung cancer were the most presented diagnoses.

CONCLUSION

In this article, we share our learning from the MTB experience and document best practices at our institution. We aim to lay a framework that allows other institutions to recreate MTBs.

IMPLICATIONS FOR PRACTICE

With the rapid pace of molecularly driven therapies entering the oncology care spectrum, there is a need to create resources that support timely and accurate interpretation of next-generation sequencing reports to guide treatment decision for patients. Molecular tumor boards (MTB) have been created as a response to this knowledge gap. This report shares implementation strategies and best practices from the Vanderbilt experience of creating an institutional MTB and a virtual global MTB for the larger oncology community. This report describe a reproducible framework that can be adopted to initiate MTBs at other institutions.

My Cancer Genome: Coevolution of Precision Oncology and a Molecular Oncology Knowledgebase

PURPOSE

The My Cancer Genome (MCG) knowledgebase and resulting website were launched in 2011 with the purpose of guiding clinicians in the application of genomic testing results for treatment of patients with cancer. Both knowledgebase and website were originally developed using a wiki-style approach that relied on manual evidence curation and synthesis of that evidence into cancer-related biomarker, disease, and pathway pages on the website that summarized the literature for a clinical audience. This approach required significant time investment for each page, which limited website scalability as the field advanced. To address this challenge, we designed and used an assertion-based data model that allows the knowledgebase and website to expand with the field of precision oncology.

METHODS

Assertions, or computationally accessible cause and effect statements, are both manually curated from primary sources and imported from external databases and stored in a knowledge management system. To generate pages for the MCG website, reusable templates transform assertions into reconfigurable text and visualizations that form the building blocks for automatically updating disease, biomarker, drug, and clinical trial pages.

RESULTS

Combining text and graph templates with assertions in our knowledgebase allows generation of web pages that automatically update with our knowledgebase. Automated page generation empowers rapid scaling of the website as assertions with new biomarkers and drugs are added to the knowledgebase. This process has generated more than 9,100 clinical trial pages, 18,100 gene and alteration pages, 900 disease pages, and 2,700 drug pages to date.

CONCLUSION

Leveraging both computational and manual curation processes in combination with reusable templates empowers automation and scalability for both the MCG knowledgebase and MCG website.

Landscape Analysis of Breast Cancer and Acute Myeloid Leukemia Trials Using the My Cancer Genome Clinical Trial Data Model

PURPOSE

The field of oncology is expanding rapidly. New trials are opening as an increasing number of therapeutic agents are being investigated before they can become approved therapies. Aggregate views of these data, particularly data associated with diseases, biomarkers, and drugs, can be helpful in understanding the trends in current research as well as existing gaps in cancer care.

METHODS

In this paper, we performed a landscape analysis for breast cancer and acute myeloid leukemia related trials with structured, curated data from clinical trials using the My Cancer Genome clinical trial knowledgebase.

RESULTS

We have performed detailed analytics on breast cancer (N = 1,128) and acute myeloid leukemia trial sets (N = 483) to highlight the top biomarkers, drug classes, and drugs—thereby supporting a full view of biomarkers, biomarker groups, and drugs that are currently being explored in these respective diseases.

CONCLUSION

Analysis and data visualization of the cancer clinical trial landscape can inform strategic planning for new trial designs and trial activation at a particular site.

Deep analysis of breast cancer & AML trials using the my cancer genome model

Clinical and genomic predictors of brain metastases (BM) in non-small cell lung cancer (NSCLC): An AACR Project GENIE analysis

2032

Background: 30-50% of patients with non-early NSCLC will eventually develop BM, with a median survival of less than one year from BM diagnosis. There are no widely accepted clinical risk models for development of BM in patients without them at baseline. We predicted the binary risk of BM using clinical and genetic factors from a large multi-institutional cohort. Methods: Stage II-IV NSCLC patients from the AACR Project GENIE Biopharma Consortium dataset were eligible. This consisted of 4 academic institutions who curated clinical data of patients who had somatic next-generation tumor sequencing (NGS) between 2015-2017. We excluded patients who had BM at baseline, died within 30 days of NSCLC diagnosis, or did not undergo brain imaging. Covariates included demographics, anticancer therapies (received up to 90 days prior to BM development and within 5 years from NSCLC diagnosis), and NGS data; radiotherapy (RT) data were not available. NGS features included mutations and copy number alterations. These features were restricted to those classified as oncogenic by OncoKB. Univariate feature selection with Fisher’s test (p<.1) was performed on medication and genetic features. We compared 5 different machine learning models for prediction: random forest (RF), support vector machine (SVM), lasso regression, ridge regression, and an ensemble classifier. We split our data into training and test sets. 10-fold cross-validation was done on the training set for parameter tuning. The area under the receiver-operating curve (AUC) is reported on the test set. Results: 956 patients were included, 192 (20%) in the test set. Univariate features associated with BM were treatment with etoposide, Asian race, presence of bone metastases at NSCLC diagnosis, mutations in TP53 and EGFR, amplifications of ERBB2 and EGFR, and deletions of RB1, CDKN2A and CDKN2B. Univariate features inversely associated with BM were older age, treatment with nivolumab, vinorelbine, alectinib, pembrolizumab, atezolizumab, and gemcitabine, as well as mutations in NOTCH1 and KRAS. Ridge regression had the best AUC, 0.73 (Table). Conclusions: We achieved reasonable prediction performance using commonly obtained clinical and genomic information in non-early NSCLC. The biologic role of the associated alterations deserves further scrutiny; this study replicates similar findings for EGFR and KRAS in a much smaller cohort. Certain subsets of NSCLC patients may benefit from increased surveillance for BM and transition to drug therapies known to effectively cross the blood-brain barrier, e.g., nivolumab and alectinib. Inclusion of additional covariates, e.g., brain RT, may further improve model performance.[Table: see text]

The My Cancer Genome clinical trial data model and trial curation workflow

OBJECTIVE

As clinical trials evolve in complexity, clinical trial data models that can capture relevant trial data in meaningful, structured annotations and computable forms are needed to support accrual.

MATERIAL AND METHODS

We have developed a clinical trial information model, curation information system, and a standard operating procedure for consistent and accurate annotation of cancer clinical trials. Clinical trial documents are pulled into the curation system from publicly available sources. Using a web-based interface, a curator creates structured assertions related to disease-biomarker eligibility criteria, therapeutic context, and treatment cohorts by leveraging our data model features. These structured assertions are published on the My Cancer Genome (MCG) website.

RESULTS

To date, over 5000 oncology trials have been manually curated. All trial assertion data are available for public view on the MCG website. Querying our structured knowledge base, we performed a landscape analysis to assess the top diseases, biomarker alterations, and drugs featured across all cancer trials.

DISCUSSION

Beyond curating commonly captured elements, such as disease and biomarker eligibility criteria, we have expanded our model to support the curation of trial interventions and therapeutic context (ie, neoadjuvant, metastatic, etc.), and the respective biomarker-disease treatment cohorts. To the best of our knowledge, this is the first effort to capture these fields in a structured format.

CONCLUSION

This paper makes a significant contribution to the field of biomedical informatics and knowledge dissemination for precision oncology via the MCG website.

KEY WORDS

knowledge representation, My Cancer Genome, precision oncology, knowledge curation, cancer informatics, clinical trial data model.

Opportunities and Challenges for Analyzing Cancer Data at the Inter- and Intra-Institutional Levels

PURPOSE

Our goal was to identify the opportunities and challenges in analyzing data from the American Association of Cancer Research Project Genomics Evidence Neoplasia Information Exchange (GENIE), a multi-institutional database derived from clinically driven genomic testing, at both the inter- and the intra-institutional level. Inter-institutionally, we identified genotypic differences between primary and metastatic tumors across the 3 most represented cancers in GENIE. Intra-institutionally, we analyzed the clinical characteristics of the Vanderbilt-Ingram Cancer Center (VICC) subset of GENIE to inform the interpretation of GENIE as a whole.

METHODS

We performed overall cohort matching on the basis of age, ethnicity, and sex of 13,208 patients stratified by cancer type (breast, colon, or lung) and sample site (primary or metastatic). We then determined whether detected variants, at the gene level, were associated with primary or metastatic tumors. We extracted clinical data for the VICC subset from VICC's clinical data warehouse. Treatment exposures were mapped to a 13-class schema derived from the HemOnc ontology.

RESULTS

Across 756 genes, there were significant differences in all cancer types. In breast cancer, ESR1 variants were over-represented in metastatic samples (odds ratio, 5.91; q < 10^-6). TP53 mutations were over-represented in metastatic samples across all cancers. VICC had a significantly different cancer type distribution than that of GENIE but patients were well matched with respect to age, sex, and sample type. Treatment data from VICC was used for a bipartite network analysis, demonstrating clusters with a mix of histologies and others being more histology specific.

CONCLUSION

This article demonstrates the feasibility of deriving meaningful insights from GENIE at the inter- and intra-institutional level and illuminates the opportunities and challenges of the data GENIE contains. The results should help guide future development of GENIE, with the goal of fully realizing its potential for accelerating precision medicine.

Conceptual Framework to Support Clinical Trial Optimization and End-to-End Enrollment Workflow

In this work, we present a conceptual framework to support clinical trial optimization and enrollment workflows and review the current state, limitations, and future trends in this space. This framework includes knowledge representation of clinical trials, clinical trial optimization, clinical trial design, enrollment workflows for prospective clinical trial matching, waitlist management, and, finally, evaluation strategies for assessing improvement.

Adapting crowdsourced clinical cancer curation in CIViC to the ClinGen minimum variant level data community-driven standards

Harmonization of cancer variant representation, efficient communication, and free distribution of clinical variant-associated knowledge are central problems that arise with increased usage of clinical next-generation sequencing. The Clinical Genome Resource (ClinGen) Somatic Working Group (WG) developed a minimal variant level data (MVLD) representation of cancer variants, and has an ongoing collaboration with Clinical Interpretations of Variants in Cancer (CIViC), an open-source platform supporting crowdsourced and expert-moderated cancer variant curation. Harmonization between MVLD and CIViC variant formats was assessed by formal field-by-field analysis. Adjustments to the CIViC format were made to harmonize with MVLD and support ClinGen Somatic WG curation activities, including four new features in CIViC: (1) introduction of an assertions feature for clinical variant assessment following the Association of Molecular Pathologists (AMP) guidelines, (2) group-level curation tracking for organizations, enabling member transparency, and curation effort summaries, (3) introduction of ClinGen Allele Registry IDs to CIViC, and (4) mapping of CIViC assertions into ClinVar submission with automated submissions. A generalizable workflow utilizing MVLD and new CIViC features is outlined for use by ClinGen Somatic WG task teams for curation and submission to ClinVar, and provides a model for promoting harmonization of cancer variant representation and efficient distribution of this information.

Standardizing And Democratizing Access To Cancer Molecular Diagnostic Test Data From Patients To Drive Translational Research

In the last 3-5 years, there has been a rapid increase in clinical use of next generation sequencing (NGS) based cancer molecular diagnostic (MolDx) testing to develop better treatment plans with targeted therapies. To truly achieve precision oncology, it is critical to catalog cancer sequence variants from MolDx testing for their clinical relevance along with treatment information and patient outcomes, and to do so in a way that supports large-scale data aggregation and new hypothesis generation. Through the NIH-funded Clinical Genome Resource (ClinGen), in collaboration with NLM's ClinVar database and >50 academic and industry based cancer research organizations, a Minimal Variant Level Data (MVLD) framework to standardize reporting and interpretation of drug associated alterations was developed. Methodological and technology development to standardize and map MolDx data to the MVLD standard are presented here. Also described is a novel community engagement effort through disease-focused taskforces to provide usecases for technology development.

ClinGen Cancer Somatic Working Group - standardizing and democratizing access to cancer molecular diagnostic data to drive translational research

A growing number of academic and community clinics are conducting genomic testing to inform treatment decisions for cancer patients (1). In the last 3-5 years, there has been a rapid increase in clinical use of next generation sequencing (NGS) based cancer molecular diagnostic (MolDx) testing (2). The increasing availability and decreasing cost of tumor genomic profiling means that physicians can now make treatment decisions armed with patient-specific genetic information. Accumulating research in the cancer biology field indicates that there is significant potential to improve cancer patient outcomes by effectively leveraging this rich source of genomic data in treatment planning (3). To achieve truly personalized medicine in oncology, it is critical to catalog cancer sequence variants from MolDx testing for their clinical relevance along with treatment information and patient outcomes, and to do so in a way that supports large-scale data aggregation and new hypothesis generation. One critical challenge to encoding variant data is adopting a standard of annotation of those variants that are clinically actionable. Through the NIH-funded Clinical Genome Resource (ClinGen) (4), in collaboration with NLM's ClinVar database and >50 academic and industry based cancer research organizations, we developed the Minimal Variant Level Data (MVLD) framework to standardize reporting and interpretation of drug associated alterations (5). We are currently involved in collaborative efforts to align the MVLD framework with parallel, complementary sequence variants interpretation clinical guidelines from the Association of Molecular Pathologists (AMP) for clinical labs (6). In order to truly democratize access to MolDx data for care and research needs, these standards must be harmonized to support sharing of clinical cancer variants. Here we describe the processes and methods developed within the ClinGen's Somatic WG in collaboration with over 60 cancer care and research organizations as well as CLIA-certified, CAP-accredited clinical testing labs to develop standards for cancer variant interpretation and sharing.

Abstract LB-103: Landscape of somatic ERBB2 Mutations: Findings from AACR GENIE and comparison to ongoing ERBB2 mutant basket study

Background: AACR GENIE is an international data-sharing project that aggregates clinical-grade cancer genomic data. As a demonstration of utility, we evaluated the landscape of ERBB2 mutations in the first 18,486 patients included in this registry and compared it to the first 100 patients enrolled in an ongoing international Phase 2 SUMMIT ‘basket’ study of the pan-HER inhibitor neratinib in ERBB2 mutant solid tumors (NCT01953926). Results: ERBB2 mutations were identified in 2.8% (519/18,486) of patients in the GENIE cohort and observed at all participating centers. In total, there were 482 missense, 66 indels, 19 truncating mutations, and 14 structural variants. A total of 263 unique missense mutations were observed including 12 at previously identified hotspots which accounted for 69.2% of all missense mutations. 35 unique cancer types were represented. The tumor types with the highest proportion of ERBB2 mutations were bladder (12.8%, 82/641), breast (3.9%, 87/2230), colorectal (3.3%, 70/2102), and NSCLC (3%, 90/3006). Among patients with copy number data available (91%) 11% had concurrent ERBB2 amplification, most often in breast cancer. The most frequently observed alterations in ERBB2, adjusted for differing exon coverage between panels, was S310F/Y in 0.46% of the GENIE cohort (12.6% of samples with ERBB2 alterations), Y772_A775dup in 0.21% (6.9%), R678Q in 0.17% (4.5%), L755S in 0.16% (5.2%), V777L in 0.12% (3.8%), and V842I in 0.09% (3.1%). The distribution of specific ERBB2 variants differed significantly by tumor type with exon 20 insertions being most common in NSCLC (44.4%, 40/90), L755S (18.9%, 11/92) in breast, S310F/Y (26.9%, 28/104) in bladder, and V842I (13.9%, 10/72) in colorectal cancer. Structural variants included intragenic deletions (n=4) and fusions involving various partners including GRB7 (n=2), and one each of C1orf87, PPIL6, HEXIM2, THRA, ASIC2, BCA3, WIPF2. The frequencies of ERBB2 mutant cancer types observed in the GENIE cohort were generally comparable to those enrolled to the neratinib basket study including NSCLC (17 vs 22%, respectively), breast (16.4 vs 24%), bladder (15.5 vs 14%), colorectal (13.2 vs 17%), and endometrial (4.2 vs 6%). At the variant level, S310F/Y was less prevalent in GENIE compared to the neratinib study (12.6 vs 24%) while all other mutations were generally similar including L755S (5.2 vs 9%), R678Q (4.5 vs 2%), Y772_A775dup (6.9 vs 13%), V777L (3.8 vs 9%), and V842I (3.1 vs 6%). Conclusion: GENIE confirms that a diversity of ERBB2 mutations are prevalent across a variety of tumor types in patients with advanced cancer. The genomic landscape of ERBB2 mutations was largely similar in the population based GENIE cohort and the neratinib SUMMIT study, providing the first direct evidence that basket study enrollment accurately reflects the true landscape of the target alteration.

Citation Format: Alison Schram, Helen H. Won, Fabrice Andre, Monica Arnedos, Funda Meric - Bernstam, Philippe L. Bedard, Kenna R. Shaw, Hugo Horlings, Christine Micheel, Ben Ho Park, Grace Mann, Alshad S. Lalani, Lillian Smyth, David B. Solit, Deborah Schrag, Mia A. Levy, Barrett J. Rollins, Mark Routbort, Charles L. Sawyers, Eva Lepisto, Michael F. Berger, David M. Hyman, on behalf of the AACR Project GENIE Consortium. Landscape of somatic ERBB2 Mutations: Findings from AACR GENIE and comparison to ongoing ERBB2 mutant basket study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-103. doi:10.1158/1538-7445.AM2017-LB-103

Utility of adding clinical data to a molecular results portal for improving clinical trial prescreening efficiency

e18182

Background: Bioportals that aggregate patient genomic results and diagnosis data elements can be used as a tool for identifying potentially eligible patients for molecularly-driven clinical trials. However, over time, increasing numbers of patients will be deceased, making this process less efficient. Methods: We sought to evaluate the addition of minimal clinical data to a clinical trial prescreening workflow utilizing these bioportals. We selected three molecularly-driven clinical trials currently enrolling patients at Vanderbilt-Ingram Cancer Center and evaluated the incremental contribution of genomic and clinical data to refinement of cohort identification. Utilizing data from the enterprise data warehouse (EDW), we assessed the potentially eligible patient population after addition of gene-level, alteration-level, vital status (known to be deceased), and date of last contact data elements to the data extraction query. Results: Utilizing gene-level and diagnosis data elements only, 68 potentially eligible patients were identified for these trials from a total of 7,200 patients whose NGS data was added to the EDW between 2010 and 2016. Addition of alteration-level detail eliminated 29% of these patients. Of the 53 remaining patients, incorporating vital status resulted in paring the potentially eligible cohorts by an additional 42%. Conclusions: This study demonstrates the added value of querying structured clinical and molecular data stored in the EDW to improve prescreen workflow efficiency and decrease manual review requirements. [Table: see text]

Assessment of actionability of cancer genomic testing panels based on a structured clinical trial knowledge base

6533

Background: Today’s oncologist is responsible for choosing appropriate cancer genomics tests to inform patient treatment from multiple available platforms, weighing cost, availability, sensitivity and specificity, and clinical actionability. Knowledge-driven clinical decision support tools can assist clinicians in choosing the panel that is most informative in a given clinical space. Methods: Using a queryable knowledgebase of >1800 active clinical trials containing structured eligibility criteria curations for diagnosis and genomic alterations, we compared two CLIA-regulated genomic panels for clinical actionability over the landscape of solid, breast, and lung cancer clinical trials. Results: The larger panel (73 genes) was more actionable than the smaller panel (62 genes) in the breast cancer (10x more trials returned) and solid tumor (2.7x more trials returned) clinical trial space, while the smaller panel returned 1.2x more trials in the lung cancer space (see table). Conclusions: This analysis demonstrates that patient diagnosis has a significant effect on the potential clinical actionability of a given genomic panel. Further, this analysis demonstrates the clinical utility of knowledge-driven clinical decision support tools for test selection, especially given the often-limited tumor sample available, cost of genomic panel testing, and continuously shifting trial landscape. [Table: see text]

Somatic cancer variant curation and harmonization through consensus minimum variant level data

BACKGROUND

To truly achieve personalized medicine in oncology, it is critical to catalog and curate cancer sequence variants for their clinical relevance. The Somatic Working Group (WG) of the Clinical Genome Resource (ClinGen), in cooperation with ClinVar and multiple cancer variant curation stakeholders, has developed a consensus set of minimal variant level data (MVLD). MVLD is a framework of standardized data elements to curate cancer variants for clinical utility. With implementation of MVLD standards, and in a working partnership with ClinVar, we aim to streamline the somatic variant curation efforts in the community and reduce redundancy and time burden for the interpretation of cancer variants in clinical practice.

METHODS

We developed MVLD through a consensus approach by i) reviewing clinical actionability interpretations from institutions participating in the WG, ii) conducting extensive literature search of clinical somatic interpretation schemas, and iii) survey of cancer variant web portals. A forthcoming guideline on cancer variant interpretation, from the Association of Molecular Pathology (AMP), can be incorporated into MVLD.

RESULTS

Along with harmonizing standardized terminology for allele interpretive and descriptive fields that are collected by many databases, the MVLD includes unique fields for cancer variants such as Biomarker Class, Therapeutic Context and Effect. In addition, MVLD includes recommendations for controlled semantics and ontologies. The Somatic WG is collaborating with ClinVar to evaluate MVLD use for somatic variant submissions. ClinVar is an open and centralized repository where sequencing laboratories can report summary-level variant data with clinical significance, and ClinVar accepts cancer variant data.

CONCLUSIONS

We expect the use of the MVLD to streamline clinical interpretation of cancer variants, enhance interoperability among multiple redundant curation efforts, and increase submission of somatic variants to ClinVar, all of which will enhance translation to clinical oncology practice.

Two-dimensional nanoparticle arrays show the organizational power of robust DNA motifs

The bottom-up spatial organization of potential nanoelectronic components is a key intermediate step in the development of molecular electronics. We describe robust three-space-spanning DNA motifs that are used to organize nanoparticles in two dimensions. One strand of the motif ends in a gold nanoparticle; only one DNA strand is attached to the particle. By using two of the directions of the motif to produce a two-dimensional crystalline array, one direction is free to bind gold nanoparticles. Identical motifs, tailed in different sticky ends, enable the two-dimensional periodic ordering of 5 and 10 nm diameter gold nanoparticles.