The Count Me In initiative demonstrates how new technologies and approaches can be incorporated into large-scale, patient-driven research projects, an idea known as convergence. The initiative started with the Metastatic Breast Cancer Project, which has enrolled thousands of patients in an effort to generate a large, public database of clinical and genomic data.

In recent years, cancer research has increasingly moved into the realm of big data, with projects tapping new technologies to build large, public databases that researchers can use to guide their work. One of the ideas driving this change is convergence, a topic explored at the 30th Anniversary American Association for Cancer Research (AACR) Special Conference, Convergence: Artificial Intelligence, Big Data, and Prediction in Cancer.

“Convergence is the idea of bringing together work and technology and new approaches from very different parts of science to try to answer challenging questions in cancer,” explained William Hahn, MD, PhD, of Dana-Farber Cancer Institute in Boston, MA. For Phillip Sharp, PhD, of the Massachusetts Institute of Technology in Cambridge, MA, who co-chaired the meeting with Hahn, convergence incorporates physics, computation, big data, and engineering into biomedical science.

An initiative called Count Me In, which was outlined at the conference, demonstrates how convergence can apply to research. The initiative consists of projects in metastatic breast and prostate cancers, angiosarcoma, and gastroesophageal cancer that are enrolling as many patients as possible to create extensive online databases of clinical and genomic information.

More than 80% of U.S. adults with cancer are treated at community oncology centers, limiting their opportunity to contribute to research, said presenter and Count Me In Director Nikhil Wagle, MD, of Dana-Farber. The Count Me In projects, however, leverage social media to engage large numbers of patients who can then sign up online. Once enrolled, patients can “share whatever they want, whether it's their medical information, their saliva, their tumors, their blood samples, and their patient-recorded data,” Wagle said.

“By having these projects, we create the opportunity for people to participate in research, and for us to build these data in a bigger, faster way,” said Adam Bass, MD, also of Dana-Farber, who leads the recently launched Gastroesophageal Project.

Wagle piloted the concept with the Metastatic Breast Cancer (MBC) Project, which, since 2015, has enrolled 3,000 patients from more than 1,000 institutions. Researchers have received medical records from 570 patients, along with 1,800 saliva samples, 400 tumor samples from more than 280 patients, and 850 blood samples. They have performed tumor and germline exome sequencing on 300 tumor and saliva samples, and whole-genome sequencing on all the blood samples. Genomic data from the first 180 patients—including 237 tumor exomes—is available online, and new data will be released every 6 months.

“This is a really good approach to study outliers,” like exceptional responders, young people, men, and patients with rare subtypes or unusual sites of metastasis, Wagle said. “Because we're casting a wide net, we might be able to find patients with characteristics that would be difficult to find with traditional approaches or at any one institution.” The combination of clinical and genomic data may also provide insights into risk factors, patterns of metastases, response to therapies, and real-world clinical practice.

As Count Me In grows, Wagle and his team hope to use machine learning to analyze medical records, digitized slides of patient tumors, and radiology data. They also aim to link their data with those of other large-scale projects, such as the AACR's Project GENIE.

Matthew Ellis, PhD, of Baylor College of Medicine in Houston, TX, noted that because metastatic breast cancer is highly heterogeneous, studying a large patient cohort is essential to reach statistically significant conclusions. He considers the MBC Project “a good segue into the next steps of understanding metastatic biology, drug resistance, and cancer mortality,” because its data can inform research and validate others' findings. For example, Ellis aims to incorporate MBC Project data on the prevalence of NF1 mutations into his research.

Building on the success of the MBC Project, Count Me In has used the same model to launch the Angiosarcoma Project, the Metastatic Prostate Cancer Project, and, most recently, the Gastroesophageal Project. Researchers plan to launch similar projects in brain, pancreatic, and ovarian cancers over the next few years. –Catherine Caruso

©2018 American Association for Cancer Research.
2018
American Association for Cancer Research.