ASCO Annual Meeting 2018 Highlights

Date / Time

Session Title / Details

Friday, June 1st

1:20 p.m. – 1:40 p.m.

Clinical Problems in Oncology Session Rm. E253c

Clinical Challenges for Stem Cell Transplants: Selecting the Proper Patient, Timing, and Donor

· Stem Cell Transplant with Less Than a Perfect Matched Donor

· In this day and age, families are smaller and are often mixed race/ethnicity, which means there are less stem cell transplant donors.  

· Marcos J.G. De Lima, MD, will discuss the barriers surrounding stem cell transplants, and answer the questions: How do we make mismatched transplants work? And how do you choose a donor?

Saturday, June 2nd

8 a.m. – 11:30 a.m.

Poster: Hall A

Breast Cancer—Local/Regional/Adjuvant

·  Predicting neo-adjuvant chemotherapy response from pre-treatment breast MRI using machine learning and HER2 status

·  Currently, there is not a good way to tell if patients will respond well to chemotherapy, and giving patients ineffective treatment can make their cancer worse.

· Nathaniel Braman, PhD candidate in CWRU’s Biomedical Engineering Department, shows how combining two machine learning techniques and incorporating clinical information can better predict patient response than if used individually.

· The techniques used are radiomics (manually defined algorithms measuring certain aspects of a tumor's imaging appearance, which are then fed into a machine learning classifier and used to predict response) and deep learning (in which a neural network is trained to learn on its own what visual patterns indicate response vs. non-response).

Saturday, June 2nd

1:15 p.m. – 4:45 p.m.

Posters: Hall A

Cancer Prevention, Hereditary Genetics, and Epidemiology

· Comparing the prevalence of non-AIDS defining cancers by HIV status in the Ohio Medicaid population

· Siran M. Koroukian, PhD, MSN, looked at Medicaid data from Ohio and discovered that those with HIV were 4x more likely to have rectal cancer and 34x more likely to have anal cancer.  

Health Services Research, Clinical Informatics, and Quality of Care

·  Trial prospector update: A point of care automated clinical trials matching application

· Advances in oncological care and patient outcomes are commonly based on clinical trials, however only 3-7% of cancer patients participate. One major barrier to patient enrollment is the physician effort needed to identify clinical trials for which their patients are eligible.

· David Lawrence Bajor, MD, discusses Trial Prospector (TP) -- a web-based, point-of-care, automated clinical trials matching application.

Sunday, June 3rd

8:20 – 8:40 a.m.

Education Session: S100a

Thinking Beyond RECIST

·  Novel Quantitative Tumor Imaging: Techniques and Clinical Applications

· RECIST (Response Evaluation Criteria In Solid Tumors) is a set of published rules that define when cancer patients improve, stay the same, or worsen during treatments.

· Prateek Prasanna, PhD, discusses how a machine learning algorithm can not only classify tumors, but also has the ability to show sub features inside and on the surface of tumors which radiologists cannot see with the human eye.

·  In this educational session, he’ll talk about the applications of novel imaging biomarkers currently being developed in the context of evaluating response to therapy in brain, lung and breast cancer.

Sunday, June 3rd

8 a.m. – 11:30 a.m.

Poster: Hall A

Lung Cancer—Non-Small Cell Local-Regional/Small Cell/Other Thoracic Cancers

· Effects of comprehensive genomic testing in a large non-small cell cancer NSCLC cohort: Racial and survival impacts

· The clinical outcome of Non-Small Cell Cancer (NSCLC) has been advanced with molecular targeted therapy. To our knowledge, there is no data addressing targeted therapy outcomes based on race.

· Fatemeh Ardeshir-Larijani, MD, focuses on racial differences in response to targeted therapy in NSCLC.

Monday, June 4th

1:15 – 4:45 p.m.

Posters: Hall A

Tumor Biology

·  Computer extracted features of cancer nuclei from H&E stained tissues of tumor predicts response to nivolumab in non-small cell lung cancer

· Computer-extracted features relating to spatial arrangement of tumor infiltrating lymphocytes to predict response to nivolumab in non-small cell lung cancer (NSCLC)

· Xiangxue Wang, PhD candidate in CWRU’s Biomedical Engineering Department, describes how a machine learning algorithm can show which cells in the body are immune cells and which are cancerous. Leveraging this machine technology can help clinicians better diagnose tumors more clearly than just using the human eye, and better predict which patients will respond to immunotherapy drugs.