Sarah Cannon Research Institute Research Articles

Abstract PO1-19-08: Phase I trial of pegylated liposomal doxorubicin chemotherapy in combination with CD40 agonist and Flt3 ligand in metastatic triple-negative breast cancer

Abstract Background: Only a subset of patients with metastatic triple-negative breast cancers (TNBC) demonstrate response to FDA approved PD-1 immune checkpoint blockade (ICB), and few have durable responses. Data suggests that breast cancers have defects in antigen presentation and that antigen presenting cells especially the cDC1 subtype of dendritic cells (DCs) are required for response to ICB. CD40 agonists activate antigen presenting cells including DCs and B cells and repolarize macrophages to an anti-tumor phenotype. Flt3 ligand is a growth factor that increases differentiation and expansion of DCs. We recently demonstrated in pre-clinical TNBC models that the combination of liposomal doxorubicin chemotherapy, a CD40 agonist, and a Flt3 ligand improves outcomes compared to alternate combinations. Methods: This is a single arm phase I pilot study of liposomal doxorubicin, CDX-1140 (CD40 agonist monoclonal antibody), and CDX-301 (recombinant Flt3 ligand) combination therapy in patients with metastatic or unresectable locally advanced metastatic TNBC. Patients will be randomized to 3 lead-in arms (triplet therapy, doublet immunotherapy only, or liposomal doxorubicin only) for one cycle prior to receiving triplet therapy with tissue biopsies done before and after the lead-in treatment. CDX-301 will be given for only two cycles; liposomal-doxorubicin and CDX-1140 will be continued until disease progression or clinically limiting toxicities. Primary endpoint is determination of a recommended phase II dose based on treatment-related adverse events and dose-limiting toxicities. Secondary endpoints include anti-tumor immune response after triplet therapy, after immunotherapy alone, and after liposomal doxorubicin alone; median progression-free survival, overall response rate, duration of response, and clinical benefit rate. Key eligibility criteria are unresectable stage III or stage IV TNBC (ER ≤10%, PR ≤10%, HER2/neu negative), 1st to 3rd line treatment for metastatic disease (1st line patients need to be PD-L1 negative by 22C3 assay), measurable disease by RECIST 1.1 criteria, consent for pre-treatment and on-treatment biopsies of amenable soft tissue tumor lesions, no prior treatment with an anti-CD40 antibody or a Flt3 ligand, no anthracycline treatment in the metastatic setting, no prior progression while on anthracycline-based therapy or within 6 months of completing neoadjuvant chemotherapy, and no history of non-infectious pneumonitis or current pneumonitis. This trial will enroll up to 45 patients across multiple sites (NCT05029999) and is currently open at University of Texas Southwestern Medical Center, Texas Oncology, University of Chicago, University of Texas San Antonio, Sarah Cannon Research Institute, and Johns Hopkins. Citation Format: Sangeetha Reddy, Joyce O'Shaughnessy, Navid Sadeghi, Samira Syed, Cesar Santa-Maria, Virginia Kaklamani, Nan Chen, Denise Yardley, Yisheng Fang, Isaac Chan, Nisha Unni, Sarah Kashanian, Namrata Peswani, Shahbano Shakeel, Meredith Carter, Kelly Kyle, Farjana Fattah, Chul Ahn, Ina Patel, Joshua Gruber, Dawn Klemow, Glenda Delgado, Nicole Sinclair, Michael Yellin, Heather McArthur, Rita Nanda, Suzanne Conzen, Carlos Arteaga. Phase I trial of pegylated liposomal doxorubicin chemotherapy in combination with CD40 agonist and Flt3 ligand in metastatic triple-negative breast cancer [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-19-08.

Outpatient administration of mosunetuzumab in US community practice settings: perspectives and learnings from the Phase II MorningSun study

Outpatient administration of mosunetuzumab in US community practice settings: perspectives and learnings from the Phase II MorningSun study

Opening a Crack in the Door: An Interview with Haydar Frangoul

Opening a Crack in the Door: An Interview with Haydar Frangoul

Outcomes of patients (pts) with metastatic colorectal cancer (mCRC) participating in phase I clinical trials: The Sarah Cannon Research Institute UK experience.

142 Background: In the last decade immunotherapy, BRAF inhibitors (inh) and trifluridine-tipiracil became new standard of care (SoC) options for selected mCRC pts through the drug development pathway. In this study we aimed to evaluate clinical characteristics and outcomes of pts with mCRC participating in phase I trials. Methods: Electronic medical records from 10/2011 to 07/2022 were reviewed; pts who received at least one dose of trial drug were included. Data were analysed with descriptive statistic. Results: 81 (100%) pts were identified; 38 (47%) female; median age: 59 (range 31-80); 31 pts (38%) - right sided; 35 (44%) - RASmut, 29 (36%) - BRAFmut, 16 (20%) - RAS/BRAFw; MSI-H status was detected in 11 pts. Eleven pts participated in >1 trial, for a total of 96 cases. Median follow-up was 7 months (m); overall response rate was 6.8%, mPFS: 2.9 m, mOS 18.8 m. Age<50 (P=0,02), low lymphocyte/monocyte ratio (LMR ≤2.8) (P=0.02), in-situ primary tumour (P<0.001), no adjuvant therapy (P=0,001), right sided tumour (P=0,02) were independently associated with shorter OS (univariate analysis); the last three variables remained significant in multivariate analysis. Outcomes according to 4 drug classes and others (chemotherapy-based +/- targeted agents) are shown. Conclusions: Phase 1 trials offer a reasonable therapeutic approach for pts with mCRC and early access to drugs with potential to become SoC in future. [Table: see text]

Implementation of a virtual, on-demand, molecular tumor board at a large, multi-clinic, community oncology practice.

96 Background: Despite the availability of molecularly-targeted agents for the treatment of many cancer types, gaps remain in integrating comprehensive precision oncology decision support tools and services into routine clinical practice. Molecular Tumor Boards (MTBs) have been shown to improve accurate incorporation of precision oncology and oncologic clinical trial enrolment into clinical practice. However, the traditional MTB model is a didactic meeting occurring at regular pre-scheduled cadences which may not align with treatment decisions or schedules of community-based general medical oncologists and advanced practice providers (APPs) without protected time away from clinic. Herein, we report on the utilization of an on-demand virtual MTB (vMTB) implemented at Tennessee Oncology (TO) powered by the Personalized Medicine (PM) team at the Sarah Cannon Research Institute (SCRI). Methods: “MolecularHelp” (MH) decision support services were implemented in September 2021 for oncology providers at TO – a network of over 100 oncologists and 86 APPs practicing across 34 clinics in Tennessee. The MH services request was a structured order that could be placed directly within the electronic health record (EHR) or through patient-protected email within the practice. MH orders initiated a virtual, on-demand interpretation of comprehensive genomic profiling (CGP) reports by a centralized vMTB run by SCRI's PM team and supported by Genospace, SCRI's precision medicine software platform. The PM team – comprised of pharmacologists, cell biologists, human geneticists, and molecular biologists – analyze CGP results and provide expert advice on both standard-of-care (SOC) and clinical research therapeutic options. Both SOC and clinical research targeted therapy options were relayed back to the treating physician and embedded within the EHR. Herein, we report key metrics including MH order frequency, average turnaround time, and subsequent clinical trial enrolment between September 2021 and March 2022. Results: CGP reports from 120 unique patients were reviewed by the vMTB during the collection period. MH orders were initiated by 30 TO providers from 14 different clinic locations across Middle and East Tennessee. The average turnaround time from referral to vMTB interpretation was less than 10 hours. Of the 120 patients reviewed, actionable mutations were identified by the MTB in 103 patients, of whom 27 subsequently enrolled onto clinical trials (15 phase 1 and 12 phase 2/3). Conclusions: An on-demand vMTB is feasible within an engaged community oncology practice with investments in bioinformatics, decision support software tools, and a team of precision oncology experts supported by a robust clinical trial menu. On-demand vMTBs can be widely adopted to enhance clinical trial enrolment. Future directions include studying the impact of vMTBs on patient outcomes over time.

Using Technology to Enhance Cancer Clinical Trial Participation.

The COVID-19 pandemic presented many challenges to health care systems, including oncology clinical research programs. There were substantial negative effects on oncology clinical trial screening, enrollment, and study activities that forced institutions and regulatory bodies to develop innovative solutions to maintain robust and equitable participation in these trials. Digital pathology innovations at Memorial Sloan Kettering Cancer Center have streamlined the diagnostic life cycle for patients with cancer, and the seamless integration of digital pathology services with next-generation sequencing and other molecular pathology services have accelerated the time to diagnosis and receipt of molecular results. Timely access to these results, coupled with Memorial Sloan Kettering Cancer Center's knowledge engine OncoKB, enhances patient clinical trial coordination precisely and efficiently. At the Sarah Cannon Research Institute, centralized remote clinical trial matching and screening, virtual molecular tumor boards, and centralized molecular interpretation support services have empowered clinic staff to identify more efficiently potential participants in clinical research, despite the COVID-19 pandemic. In addition, the U.S. Food and Drug Administration Oncology Center of Excellence has been involved in several efforts to address challenges for patients with cancer during the COVID-19 pandemic, including writing guidance documents and participating in efforts to modernize clinical trials. The enclosed personal experience of a patient with cancer currently participating in an oncology clinical trial emphasizes the need for continued decreasing of barriers to study participation. Clinical trial advances that were accelerated by the pandemic will ultimately help patients with cancer and the greater oncology health care community.

Brentuximab Vedotin in Combination with Lenalidomide and Rituximab in Patients with Relapsed or Refractory Diffuse Large B-Cell Lymphoma (DLBCL) (ECHELON-3, Trial in Progress)

BackgroundThe majority of patients with relapsed/refractory (R/R) DLBCL who relapse after hematopoietic stem cell transplantation (HSCT) or chimeric antigen receptor T-cell (CAR-T) therapy, or who are not candidates for HSCT or CAR-T, have poor outcomes and need novel therapies (Crump 2017). Brentuximab vedotin (BV) was the first antibody-drug conjugate to be approved in multiple cancer types (Gauzy-Lazo 2020). The combination of a CD30-directed monoclonal antibody, a protease-cleavable linker, and the microtubule-disrupting agent monomethyl auristatin E drives the anticancer activity of BV by inducing cell cycle arrest and apoptosis (Sutherland 2006). Lenalidomide may enhance the activity of BV through immune-mediated mechanisms. In a phase 1 trial (NCT02086604) in which 37 patients with R/R DLBCL were treated with BV + lenalidomide, the objective response rate (ORR) was 56.7% (73.3% in CD30+ patients). The median duration of remission was 13.2 months in patients with a complete response (CR) or partial response (PR) and 11.7 months in patients with a CR, PR, or stable disease (SD) >6 months. The median progression-free survival (PFS) and overall survival (OS) were 10.2 months and 14.3 months, respectively, and results were similar in the CD30+ and CD30 <1% groups (manuscript submitted). The clinical activity and manageable safety profiles of BV, lenalidomide, and rituximab as single agents provide a strong rationale for this combination in multiple relapsed and heavily pretreated patients with R/R DLBCL.Study Design and MethodsECHELON-3 (NCT04404283) is a randomized, double-blind, placebo-controlled, active-comparator, multicenter phase 3 study designed to evaluate the efficacy of BV versus placebo, in combination with lenalidomide and rituximab, in patients with R/R DLBCL.Key eligibility criteria will include patients ≥18 years of age with R/R DLBCL with an eligible subtype, ≥2 prior lines of systemic therapy and ineligible for stem cell transplant or CAR-T therapy, Eastern Cooperative Oncology Group performance status 0 to 2, fluorodeoxyglucose-avid disease by positron emission tomography (PET), and bidimensional measurable disease of at least 1.5 cm by computed tomography (CT). Patients with a history of another malignancy within the past 2 years, any uncontrolled Grade ≥3 infection, Grade ≥2 peripheral neuropathy, or previous exposure to BV or lenalidomide will not be eligible.After completing the safety run-in period, patients (n=400) will be randomized 1:1 to receive either BV (1.2 mg/kg intravenously every 3 weeks [Q3W]) or placebo (intravenously Q3W) in combination with lenalidomide (20 mg orally daily) and rituximab (1400 mg subcutaneously Q3W from Cycle 2, with 375 mg/m 2 intravenously on Cycle 1 Day 1). Treatment may continue while there is clinical benefit (SD or better) without progression or unacceptable toxicity. Patients will be stratified by CD30 expression (≥1% [positive] vs <1% [negative]), prior allogeneic or autologous stem cell transplant therapy (received or not), prior CAR-T therapy (received or not), and cell of origin (germinal center B-cell [GCB] or non-GCB).The dual primary endpoints are PFS per blinded independent central review (BICR) in the intention-to-treat (ITT) and CD30+ populations. Key secondary endpoints are OS in the ITT and CD30+ populations, and ORR per BICR. Other secondary endpoints include CR rate, duration of objective response, and safety and tolerability of the combination. Disease response will be assessed by BICR and the investigator according to the Lugano Classification Revised Staging System for nodal non-Hodgkin and Hodgkin lymphomas (Cheson 2014). Radiographic disease evaluations, including contrast-enhanced CT and PET, will be assessed at baseline, then every 6 weeks from randomization until Week 48, then every 12 weeks thereafter. PET is not required after CR is achievedFor the analyses of PFS per BICR, the stratified log-rank test will be used to compare PFS between the 2 treatment groups. Hazard ratios will be estimated using the stratified Cox regression model. PFS will also be summarized using the Kaplan-Meier method. Similar methods will be used to estimate OS and other time-to-event efficacy endpoints.The study is currently enrolling and will be open in 16 countries. DisclosuresBartlett: Merck: Research Funding; Kite, a Gilead Company: Research Funding; Janssen: Research Funding; Genentech: Research Funding; Forty Seven: Research Funding; Celgene: Research Funding; Bristol Myers Squibb: Research Funding; Autolus: Research Funding; Seagen: Consultancy, Research Funding; Roche/Genentech: Consultancy; ADC Therapeutics: Consultancy, Research Funding; Millennium: Research Funding; Pharmacyclics: Research Funding. Yasenchak: Seagen Inc.: Research Funding. Harwin: Sarah Cannon Research Institute: Other: Grants, Research Funding. Sims: Seagen Inc.: Current Employment. Nowakowski: Celgene, MorphoSys, Genentech, Selvita, Debiopharm Group, Kite/Gilead: Consultancy, Membership on an entity's Board of Directors or advisory committees; Celgene, NanoString Technologies, MorphoSys: Research Funding.

532 First-in-human phase 1 trial of SRK-181: a latent TGFβ1 inhibitor, alone or in combination with anti-PD-(L)1 treatment in patients with advanced solid tumors (DRAGON trial)

BackgroundTransforming growth factor-beta 1 (TGFβ1) plays an important role in mediating the primary resistance to PD-1/PD-L1 [PD-(L)1] blockade. SRK-181 is a fully human monoclonal antibody that selectively inhibits latent TGFβ1 activation. Mouse tumor models (bladder, melanoma, and breast cancer) demonstrated that treatment with SRK-181+anti-PD-1 overcame primary anti-PD-1 resistance. Four-week GLP nonclinical toxicology studies showed that SRK-181 has improved safety profile (no cardiotoxicities) compared to broad TGFβ pathway inhibition.MethodsThe DRAGON trial (NCT04291079) is an ongoing open-label, phase 1 study. Part A of the study follows a standard 3+3 dose escalation design to determine the dose for Part B. Part B (expansion phase) evaluates combination treatment in patients with non-small cell lung cancer (NSCLC), urothelial carcinoma, melanoma, or other advanced solid tumors. SRK-181 is administered IV every 3 or 2 weeks (Q3W/Q2W) alone in patients with advanced solid tumors (Part A1), or in combination with anti-PD-(L)1 in patients who did not respond to prior anti-PD-(L)1 therapy (Part A2/B).ResultsAs of 7 June 2021, 25 patients have enrolled to Part A; median 4 prior lines of therapies (range 1–9). Cancer types: colorectal, ovarian, prostate, and unknown primary (Part A1); liver, melanoma, NSCLC, oropharynx, renal cell carcinoma (RCC) and uterine (Part A2). In Part A1, 15 patients were treated with SRK-181 monotherapy at doses of 80, 240, 800, 1600, 2400, 3000mg Q3W, with no dose limiting toxicity (DLT) observed. The last cohort (2000mg Q2W) remains under evaluation. The most common treatment-related AEs (TRAE, &gt;10%) of any grade were decreased appetite and fatigue (each: 13.3%, n=2). Six patients had stable disease (SD) as best response (2/colorectal cancer, 1/prostate cancer, and 3/ovarian cancer). Three ovarian cancer patients were stable ≥153 days with tumor regressions. In Part A2, 10 patients were treated with SRK-181 at doses of 240, 800 and 1600mg Q3W+pembrolizumab. No DLT was observed up to 800mg. 1600mg Q3W is under evaluation. No TRAE (&gt;10%) of any grade were observed. One confirmed RECIST1.1 partial response (PR) was observed (800mg) in a patient with anti-PD-1 resistant RCC and 2 patients had best response of SD (1/oropharynx cancer, 1/liver cancer). The half-life of SRK-181 ranged from 3.9 to 19.3 days across the doses tested.ConclusionsAs of 7 June 2021, SRK-181 has been well tolerated as monotherapy and in combination with anti-PD-(L)1. One RECIST1.1 PR (800mg) was observed in a patient with anti-PD-1 resistant RCC. Next planned dose in Part A2 will be 2400mg Q3W.Trial RegistrationDRAGON trial (NCT04291079)Ethics ApprovalThe human study was approved by the Massachusetts General Hospital and Beth Israel Deaconess Medical Center (20–286), Sarah Cannon Research Institute (1276118), Mary Crowley Cancer Research (20–06), MD Anderson Cancer Center (2020–0110) Institutional Review Boards with written informed consent obtained from each participant and/or their legal representative, as appropriate.

2021 World Conference on Lung Cancer.

Melissa Johnson (Sarah Cannon Research Institute, Nashville, TN, USA) presented results from a randomised, open-label, phase 3 study of 1013 patients with metastatic, EGFR and ALK wild-type non-small-cell lung cancer (NSCLC) comparing chemotherapy (type chosen on the basis of histology) plus durvalumab (1500 mg daily; n=338) versus chemotherapy plus durvalumab and tremelimumab (75 mg daily; n=338) versus chemotherapy alone (n=337) as first-line treatment. Median progression-free survival (co-primary endpoint) was 5·5 months (95% CI 4·7–6·5) in the durvalumab plus chemotherapy group versus 4·8 months (4·6–5·8) in the chemotherapy alone group (hazard ratio [HR] 0·74; 95% CI 0·62–0·89; p=0·0009) and median overall survival (co-primary endpoint) was 13·3 months (11·4–14·7) versus 11·7 months (10·5–13·1; HR 0·86; 0·72–1·02; p=0·076). Grade 3 or 4 treatment-related adverse events occurred in 149 (45%) patients in the chemotherapy plus durvalumab group, 171 (52%) in the chemotherapy plus durvalumab and tremelimumab group, and 148 (44%) in the chemotherapy alone group.

Impact of immune checkpoint inhibitor (CPI) and EGFR tyrosine kinase inhibitor (TKI) sequence on time to treatment failure (TTF) among EGFR plus NSCLC treated in a community-based cancer research network.

9099 Background: The development of CPIs and driver-targeted TKIs has transformed the treatment of NSCLC and increased survival rates. However, the role of CPIs in patients with oncogenic-driven NSCLC remains an area of investigation. We sought to examine the impact of CPI sequence on treatment response among patients with oncogenic-driver mutation-positive NSCLC. Methods: Patients with NSCLC being treated within the Sarah Cannon Research Institute network were identified through Genospace, Sarah Cannon’s clinico-genomic analytics platform. Advanced stage oncogenic-driven tumors (driver+) were defined as those with a record of receiving an FDA-approved TKI targeting EGFR, ALK, RET, ROS1, NTRK, MET, or BRAF. Kaplan-Meier estimates were used to examine TTF (defined as time from therapy start to start of next therapy, death, or loss to follow-up) and overall survival (OS). Results: We identified 12,352 patients with lung cancer and available therapy data (2005-2020), including 2,270 (18%) driver+ patients. Eleven percent (N=245) of driver+ patients received a CPI, including 120 (49%) with CPI prior to TKI, 122 (50%) with CPI post TKI, and 3 (1%) who received CPI both pre and post TKI. The CPI TTF was significantly longer for those who received CPI post TKI compared to those who received it prior (Table). EGFR+ tumors accounted for 82% (N=1,867) of driver+ patients, 10% of whom (N=188) received a CPI. Of the EGFR+/CPI+ patients, 78 patients (41%) received CPI prior to TKI, 107 (57%) received CPI post TKI, and 3 (2%) received CPI both pre and post TKI. EGFR+ tumors exposed to a CPI post TKI had a longer CPI TTF compared to patients who received it prior (Table). In contrast, there was no difference in length of benefit from TKI if it was received pre vs. post CPI (Table). There was also no difference in OS based on sequence of TKI and CPI (p=0.88). Larger sample sizes are needed for analysis of additional driver-stratified cohorts. Conclusions: Patients with oncogenic-driven NSCLC benefited from CPI longer when it was administered after TKI compared to before. Importantly, therapy sequence only affected length of benefit from CPIs and did not affect length of benefit from TKIs. This effect was present in EGFR+ NSCLC, but sample sizes were too small to determine if the same is true for other oncogenic-drivers. Therapy sequence had no impact on OS, indicating the presence of additional clinical, therapeutic, and/or genomic factors contributing to disease progression. Continued research is needed to better understand markers of CPI response in driver+ NSCLC.[Table: see text]

Famotidine Use Is Not Associated With 30-day Mortality: A Coarsened Exact Match Study in 7158 Hospitalized Patients With Coronavirus Disease 2019 From a Large Healthcare System

Famotidine Use Is Not Associated With 30-day Mortality: A Coarsened Exact Match Study in 7158 Hospitalized Patients With Coronavirus Disease 2019 From a Large Healthcare System

Barriers to Clinical Trial Enrollment in Patients With Pancreatic Adenocarcinoma Eligible for Early-Phase Clinical Trials.

Early-phase clinical trials are critical to the advancement of cancer care, especially in patients with pancreatic ductal adenocarcinoma, given its aggressive nature and limited available therapeutic options. A retrospective chart review of all patients with refractory or metastatic pancreatic ductal adenocarcinoma, referred to the Sarah Cannon Research Institute at HealthONE between 2014 and 2019, were reviewed. Patients who completed genomic profiling and qualified for a phase 1 trial (primarily 1a but some 1b) were identified to assess barriers to trial enrollment. Of 74 identified patients, 54 patients (73%) qualified for at least 1 clinical trial based on eligibility criteria and alterations detected via molecular profiling. Up to 40 industry-sponsored clinical trials were available during this time for consideration. Of the 54, 28 patients (52%) enrolled in a clinical trial, while 26 (48%) did not enroll. The most frequently cited barriers to enrollment were concerns regarding time commitment (12%), prolonged wait time for enrollment (12%), and fear of adverse events (8%). Seven of the 26 patients (27%) were lost to follow-up or had no stated reason for declining enrollment; others did not go on trial due to death/transition to hospice (n=5; 19%) or progression of disease/declining functional status (n=4; 15%). There were few statistically significant differences between patients who chose to go on trial and those who declined. An understanding of why eligible patients elect not to participate in early-phase clinical trials provides insight into the patient experience and can help identify misperceptions and areas for improvement in education and the clinical trial enrollment process.

Meeting Proceedings of the “Phase I: Where Science Becomes Medicine” Conference, Manchester, UK

Meeting Proceedings of the “Phase I: Where Science Becomes Medicine” Conference, Manchester, UK

Adoption and utilization of NGS-based molecular profiling in community-based oncology practices: Insights from Sarah Cannon.

34 Background: The price of NGS-based sequencing technologies is falling, and the adoption of NGS-based testing is increasing in oncology practices. To date, a survey of the adoption and utilization of NGS-based technologies as a part routine oncology clinical care has not been performed. Thus, a comprehensive analysis of physician adoption and utilization of commercial NGS testing in the non-academic medical center, community-setting between 2012 and 2018 was performed. Methods: Medical Oncologists in the Sarah Cannon Research Institute network ordered commercially-available NGS-based molecular profiling for their patients as standard of care. Data use agreements were initiated between SCRI, affiliated medical oncology practices, and commercial NGS testing providers, and patient NGS data was subsequently analyzed starting in 2012. Results: Community-based NGS testing rates with the Sarah Cannon network were 5.75/month in 2012 and 440/month in 2018. Plasma-based NGS testing began in 2014 and comprised 4.9% of total testing compared to 40.1% in 2018. The number of oncologist ordering molecular profiles increased from 11 in 2012 to 269 in 2018. Physician test utilization grew from an average of 6 tests/physician to 22 tests/physician in 2012 and 2018, respectively. NGS tests were performed on 34 different tumor types and biopsies were taken from both primary tumors (~40%) and metastatic sites (~60%). Tissue-based tests averaged 14 mutations/sample while plasma-based tests averaged 4 mutations/sample. There was a 74% decrease in median time between biopsy collection and NGS test results between 2012 and 2018 (131 and 34 days, respectively), indicating a shift toward the use of fresh – non-archival – tissue in recent years. Conclusions: These data establish NGS-based testing trends in community oncology practices and show that NGS-based tumor testing utilization has increased in the community-setting between 2012 and 2018. NGS testing is performed on a wide array of tumor types and oncologists order tests earlier and utilize fresher biopsies.

902O - An adaptive, biomarker directed platform study in metastatic urothelial cancer (BISCAY) with durvalumab in combination with targeted therapies

902O - An adaptive, biomarker directed platform study in metastatic urothelial cancer (BISCAY) with durvalumab in combination with targeted therapies

Automating incidental findings in radiology reports using natural language processing and machine learning to identify and classify pulmonary nodules.

e18093 Background: Pulmonary nodule incidental findings challenge providers to balance resource efficiency and high clinical quality. Incidental findings tend to be undertreated with studies reporting appropriate follow-up rates as low as 29%. Ensuring appropriate follow-up on all incidental findings is labor-intensive; requires the clinical reading and classification of radiology reports to identify high-risk lung nodules. We tested the feasibility of automating this process with natural language processing (NLP) and machine learning (ML). Methods: In cooperation with Sarah Cannon Research Institute (SCRI), we conducted a series of data science experiments utilizing NLP and ML computing techniques on 8,879 free-text, narrative CT (computerized tomography) radiology reports. Reports used were dated from Dec 8, 2015 - April 23, 2017, came from SCRI-affiliated Emergency Department, Inpatient, and Outpatient facilities and were a representative, random sample of the patient populations. Reports were divided into a development set for model training and validation, and a test set to evaluate model performance. Two models were developed - a “Nodule Model” was trained to detect the reported presence of a pulmonary nodule and a rules-based “Sizing Model” was developed to extract the size of the nodule in millimeters. Reports were bucketed into three prediction groups: &gt; = 6 mm, &lt; 6 mm, and no size indicated. Nodules were considered positives and placed in a queue for follow-up if the nodule was predicted &gt; = 6 mm, or if the nodule had no size indicated and the radiology report contained the word “mass.” The Fleischner Society Guidelines and clinical review informed these definitions. Results: Precision and recall metrics were calculated for multiple model thresholds. A threshold was selected based on the validation set calculations and a success criterion of 90% queue precision was selected to minimize false positives. On the test dataset, the F1 measure of the entire pipeline (lung nodule classification model and size extraction model) was 72.9%, recall was 60.3%, and queue precision was 90.2%, exceeding success criteria. Conclusions: The experiments demonstrate the feasibility of NLP and ML technology to automate the detection and classification of pulmonary nodule incidental findings in radiology reports. This approach promises to improve healthcare quality by increasing the rate of appropriate lung nodule incidental finding follow-up and treatment without excessive labor or risking overutilization.

Adoption and utilization of NGS-based molecular profiling in community-based oncology practices: Insights from Sarah Cannon.

e18064 Background: As the price of NGS-based sequencing technologies falls, adoption of those technologies is predicted to increase. To date, a survey of the adoption and utilization of NGS-based technologies as a part routine oncology clinical care has not been performed. Thus, a comprehensive analysis of physician adoption and utilization of commercial NGS testing in the community-setting between 2012 and 2018 was performed. Methods: Medical Oncologists in the Sarah Cannon Research Institute network ordered commercially-available NGS-based molecular profiling for their patients as standard of care. Data use agreements were initiated between SCRI, affiliated medical oncology practices, and commercial NGS testing providers, and patient NGS data were subsequently analyzed starting in 2012. Results: Community-based NGS testing rates within the Sarah Cannon network were 5.75/month in 2012 and 440/month in 2018. Plasma-based NGS testing began in 2014 and comprised 4.9% of total testing compared to 40.1% in 2018. The number of oncologists ordering molecular profiles increased from 11 in 2012 to 269 in 2018. Physician test utilization grew from an average of 6 tests/physician to 22 tests/physician in 2012 and 2018, respectively. NGS tests were performed on 34 different tumor types and biopsies were taken from both primary (~40%) and metastatic (~60%) sites . Tissue-based tests averaged 14 mutations/sample while plasma-based tests averaged 4 mutations/sample. There was a 74% decrease in median time between biopsy collection and NGS test results between 2012 and 2018 (131 and 34 days, respectively), indicating a shift toward the use of fresh – non-archival – tissue in recent years. Conclusions: These data establish NGS-based testing trends in community oncology practices and show that NGS-based tumor testing utilization has increased in the community-setting between 2012 and 2018. NGS testing is performed on a wide array of tumor types and oncologists utilize fresher biopsies.

Prospective analysis of 895 patients on a UK Genomics Review Board

BackgroundThe increasing frequency and complexity of cancer genomic profiling represents a challenge for the oncology community. Results from next-generation sequencing–based clinical tests require expert review to determine their clinical relevance and to ensure patients are stratified appropriately to established therapies or clinical trials.MethodsThe Sarah Cannon Research Institute UK/UCL Genomics Review Board (GRB) was established in 2014 and represents a multidisciplinary team with expertise in molecular oncology, clinical trials, clinical cancer genetics and molecular pathology. Prospective data from this board were collated.ResultsTo date, 895 patients have been reviewed by the GRB, of whom 180 (20%) were referred for clinical trial screening and 62 (7%) received trial therapy. For a further 106, a clinical trial recommendation was given.ConclusionsNumerous challenges are faced in implementing a GRB, including the identification of potential germline variants, the interpretation of variants of uncertain significance and consideration of the technical limitations of pathology material when interpreting results. These challenges are likely to be encountered with increasing frequency in routine practice. This GRB experience provides a model for the multidisciplinary review of molecular profiling data and for the linking of molecular analysis to clinical trial networks.

1430P - Cell-free circulating tumour DNA (ctDNA) in the management of patients with non-biopsiable advanced non-small cell lung cancer (NSCLC)

Background: Genomic profiling of cell-free ctDNA is a non-invasive method to guide personalised medicine. We assessed the utility of ctDNA in routine clinical practice where tumor biopsy were impossible to obtain, tissue insufficient or clinically contraindicated. Methods: A 73-gene panel using ctDNA NGS (Guardant360®) was offered to consecutive stage 4 NSCLC patients - results were discussed in our Genomics Review Board to assess potential actionable alterations and enrolment into clinical trials. Results: 50 pts (37F:13M; median 64 yrs) participated. ctDNA was obtained in 6 treatment (Tx) naïve patients (pts), in 22pts post 1st-line, in 7pts post 2nd-line, in 5 pts post 3rd-line and in 5 pts with > 4 Tx – in 5 pts prior Tx was unknown. EGFR status at Dx was known in 40 (23mt/17wt) and unknown in 10 pts. Of the pts with known EGFRmt, 6 pts progressed on gefitinib, 7 pts on afatinib, 8 pts on erlotinib, and 2 pts did not receive prior anti-EGFR Tx. ctDNA testing confirmed EGFRmt in 14 pts. In 9 pts with previously Bx-proven EGFRmt ctDNA did not detect an EGFRmt - the lag time between Bx and ctDNA was a median of 18 months (range 1-94 months). New EGFRmt were found in 3 pts with unknown EGFR status allowing access to anti-EGFR Tx. Acquired T790M were found in 4 pts progressing on prior anti-EGFR Tx, those pts received osimertinib. In 1 pt with ctDNA T790M+, a concomitant solid Bx was T790M-. Other alterations were, TP53 (n = 23), KRAS (n = 7), ERBB2 (n = 1), NTRK1 (n = 2), NF1 (n = 12), MET (n = 7), BRAF (n = 5), PIK3CA (n = 5), ALK (n = 2), BRCA1 (n = 1), BRCA2 (n = 2), PDGFRA (n = 3), AR (n = 4), TERT (n = 3), CDK4 (n = 3), CDK6 (n = 3), FGFR2 (n = 2), STK11 (n = 2), KIT (n = 2), SMAD4 (n = 3), CDH1 (n = 1), NOTCH1 (n = 2), RB1 (n = 3), TSC1 (n = 1), ERBB1(n = 1), MTOR (n = 3), MYC (n = 3), ARAF (n = 1), GNAS (n = 1), AKT1(n = 1), CDKN2A (n = 2), ARID1A (n = 2), CTNNB1 (n = 1), CCNE1 (n = 1). Critical review in the GRB meeting was fundamental in interpreting the genetic alterations of significance. Conclusions: We confirm the feasibility and clinical utility of ctDNA testing in NSCLC patients where tumor biopsies were insufficient, impossible or contraindicated and identified 7 pts (14%) who based on their ctDNA results received 1st/2nd-line anti-EGFR treatment, several more were recommended for clinical trials. Legal entity responsible for the study: Sarah Cannon Research Institute. Funding: Has not received any funding. Disclosure: P. Bennett: Employee: Sarah Cannon Molecular Diagnostics. I. Faull, R.B. Lanman: Employee: Guardant. All other authors have declared no conflicts of interest.

Sorafenib and continued erlotinib or sorafenib alone in patients with advanced non-small cell lung cancer progressing on erlotinib: A randomized phase II study of the Sarah Cannon Research Institute (SCRI)

PurposeTo evaluate the efficacy of erlotinib, continued after tumor progression, plus sorafenib versus sorafenib alone in patients with refractory metastatic non-small cell lung cancer (NSCLC) who previously benefitted from single-agent erlotinib. Patients and MethodsPatients with progressive refractory NSCLC who had previously benefitted from erlotinib (objective response or stable disease >8weeks) were randomized to receive treatment with either erlotinib and sorafenib (400mg orally twice daily) or sorafenib alone. Patients were evaluated for response every 8 weeks, and continued treatment until disease progression or intolerable toxicity. ResultsFifty-three patients were randomized (erlotinib/sorafenib, 25; sorafenib, 28) and 52 patients received study treatment. Patients in both groups received a median of 8weeks of treatment. The median PFS was 3.1months for erlotinib/sorafenib versus 1.7months for sorafenib alone; response rates were 8% and 4%, respectively. Both regimens were tolerable, but toxicity was more frequent with erlotinib/sorafenib. ConclusionsThese results do not suggest any benefit in continuing erlotinib after tumor progression in patients with refractory metastatic NSCLC. Both regimens tested had limited efficacy, consistent with results from other studies. ClinicalTrials.gov ID:NCT00609804