“My first experience with cancer in the family occurred when I was studying statistics in India. The timing prompted me to start wondering about using quantitative approaches to discover the causes of cancer and how to prevent it. Now, I develop and apply such approaches to better predict how many types of risk factors affect the chances of developing cancer, alone or in combination with other risk factors.
“I feel privileged to work at the American Cancer Society, given the positive impact the ACS has on so many aspects of the cancer problem: research, advocacy, and patient and caregiver education and support. My research may inform prevention guidelines and help ACS achieve its mission ‘to save lives, celebrate lives, and lead the fight for a world without cancer.’”
As a Principal Scientist in Biostatistics, Parichoy Pal Choudhury, PhD, works with both Surveillance and Health Equity Science (SHES) and Population Science teams within the American Cancer Society (ACS).
He leads a research program focused on the development of statistical methods in risk prediction with applications in risk-stratified cancer prevention. He conducts research on mission-priority topics across the cancer continuum and actively collaborates with scientists across the ACS Discovery Pillar as well as with investigators who are not part of the ACS.
Pal Choudhury also holds an adjunct affiliation with the Division of Cancer Epidemiology and Genetics at the National Cancer Institute.
My research mainly focuses on developing methods and software tools for building and independently validating absolute risk-prediction models. I apply these models to predict absolute risk for breast cancer, lung cancer, and bladder cancer. The goal is to inform cancer prevention by stratifying the risks for developing cancer.
For example, for these complex diseases, I build models to assess how factors like genetic variants and non-genetic factors interact to affect the risk of disease and whether the risk-factor associations differ across biologically heterogeneous disease subtypes.
My recent statistical methodology research focused on how to precisely estimate metrics of model performance in a prospective cohort where certain expensive biomarkers may be available only on a subgroup of the study participants. I showed an application of this method in lung cancer risk prediction in a paper published in Cancer Epidemiology, Biomarkers & Prevention.
I have substantial experience in leading or co-leading large studies in a consortium setting. For example, I led a large validation effort within the Breast Cancer Association Consortium (BCAC) for a model predicting absolute risk of breast cancer. This model was validated in 15 prospective cohorts across 6 countries and incorporated risk factors and polygenic risk score (PRS) with relative risks derived from literature. The paper was published in the International Journal of Epidemiology with a commentary.
My research had been recognized with several recent awards for early-career scientists:
More recently, I am interested in the risk prediction of prostate cancer, ovarian cancer, association of infectious agents on hematologic malignancies, and cancer health disparities.
For a full list of Dr. Pal Choudhury’s publications, visit his Google Scholar page.
I live with my wife, Dr. Tushita Mukhopadhyaya, a material scientist at Georgia Tech, in the Brookhaven area in Atlanta. When I'm not doing research, I enjoy outdoor activities like hiking and kayaking and indoor ones like reading, watching movies, and listening to music.