Research Projects
Circulating Tumor DNA
Detecting cancer directly from the blood through “liquid biopsies” has the potential to change the way cancer is diagnosed, monitored, and even treated. The most promising type of liquid biopsy depends on detection of cancer-derived DNA molecules seen in the blood plasma of patients, or circulating tumor DNA (ctDNA). However, detecting ctDNA can be challenging due to low abundance in the blood plasma of most patients. To address this, we have developed methods to detect multiple features of ctDNA, including mutations, copy number variations, and epigenetic changes, with a focus on improving limits of detection. These methods have applications for multiple types of cancers, including lymphoma, myeloma, and solid tumors.
Lymphoma Immunotherapy
Circulating tumor DNA analysis and other genomic technologies enable unprecedented platforms to understand mechanisms of response and resistance to targeted therapies. At the same time, novel immunotherapies for lymphomas and other cancers, such as chimeric antigen receptor T-cells, have emerged as effective treatments for many patients. However, matching therapy-to-patient based on their tumor characteristics remains a major challenge, hindering precision medicine approaches. To address this, we are utilizing ctDNA and other genomic techniques to study the molecular features of lymphoma patients respond – or not responding – to immunotherapies, we aim to better understand which patients will benefit from these novel treatments.
Clinical Risk Prediction
Despite the proliferation of “clinical risk prediction” scores like the International Prognostic Index, predicting outcomes for individual patients with cancer remains remarkably difficult. Most established risk prediction tools use a set of static biomarkers to divide patients into low and high risk categories. While these approaches can be useful for clinical trials, they are poorly predictive for individual patients. Moreover, they fail to utilize novel biomarkers such as liquid biopsies which can be assessed serially and repeatedly over time. To improve on this, we established the Continuous Individualized Risk Index, or CIRI. CIRI represents a novel Bayesian framework to integrate serial biomarkers over time, improving on static risk predictions.