University of Colorado School of Medicine researchers hope new research could prevent up to 130,000 unnecessary thyroid nodule fine-needle biopsies (FNA) and subsequent surgeries each year in the United States by better understanding the genetic risk associated with cancer of the thyroid.
Through an R21 grant from the National Institutes of Health, Nikita Pozdeyev, MD, assistant professor of biomedical informatics, Chris Gignoux, PhD, professor of biomedical informatics, and Bryan Haugen, MD, professor of medicine and chief of the Department of Endocrinology, Metabolism, and Diabetes, they will study new strategies that could pave the way for personalized thyroid nodule management, inform future mechanistic studies of thyroid cancer, and lead to a clinical trial of an ultrasound and genetic thyroid nodule classifier. This work aims to create a clearer diagnosis and better standard of care for thousands of patients presenting with a thyroid nodule that currently requires biopsy.
“Our ultimate goal is to better diagnose thyroid cancer,” says Pozdeyev, a trained endocrinologist in the Department of Biomedical Informatics who uses data to address clinical challenges.
Thyroid cancer is the most common endocrine malignancy, accounting for approximately 44,000 new cases and 1% of new cancer diagnoses each year. When a thyroid nodule is detected in a patient, it can be difficult to know whether it is benign or cancerous.
An FNA biopsy can help establish a diagnosis, but, ultimately, about 20 percent of biopsies return an inconclusive result, Pozdeyev says.
“We then order additional tests and often have to do diagnostic surgery, basically removing a person’s thyroid,” he explains. “And in some cases, we’re finding that we’ve gone through all that trouble to learn that it wasn’t necessary and that a thyroid nodule is benign. With this grant, we’re going to incorporate genetics to better quantify the risk that a particular person has thyroid Cancer.”
The power of data
Researchers will leverage biobanks around the world, including the biobank at the Colorado Center for Personalized Medicine, to create a dataset that would not otherwise be possible.
We have many institutions around the world who want to help us solve this problem. The study of human genetics is highly collaborative because we benefit from scenarios where we can look at hundreds and thousands to millions of people. This helps us to have thorough studies.”
Chris Gignoux, PhD, professor of biomedical informatics, University of Colorado School of Medicine
The data will help the team create a polygenic risk score (PRS), which Gignoux explains as a mechanism for measuring disease risk based on complex traits. Unlike some forms of cancer where a gene can determine the risk -? such as the BRCA gene in hereditary breast cancer – Thyroid cancer risk depends on a patchwork of genes that interact with each other.
To analyze the genetics behind thyroid cancer, the researchers will test genetic associations directly using a GWAS meta-analysis of 12,091 thyroid cancer cases, 56,949 patients with benign nodules, and nearly 1.8 million people without thyroid nodules as witnesses. They will also use a computational method to disentangle the signals that lead to thyroid cancer from other common characteristics of thyroid nodules, such as goitre.
In the end, research can inform more than just cancer.
“Our grant is focused on the end result of being able to say something about thyroid cancer, but to do that, we need to have a lot of data on a number of thyroid characteristics. For example, we’ve collected the largest data set on date on the genetics of hypothyroid,” says Gignoux. “This allows us to tease out the specific signal that predisposes people to thyroid cancer itself.”
“This is the future of personalized medicine research,” he continues. “We want to be able to take advantage of what the world of data can tell us collectively and then bring it back into an environment with domain experts to get the most out of it and ensure that our results translate into clinical and medical impact” .
The future of patient care
The course of treatment for a person with a thyroid nodule has evolved greatly over the past five decades.
“Before the 1980s, if a doctor felt a lump in a patient’s neck, they went almost straight to the operating room,” says Haugen, who works in the Department of Medicine and has seen patients with thyroid tumors for more than 30 years.
Then, the introduction of FNA biopsies allowed pathologists to know if a tumor was benign. It was a real improvement, Haugen says, and it cut unnecessary surgeries by about half, but there’s still more work to do to create better outcomes for patients, especially the 20 percent of people who get inconclusive biopsy results.
“The next level of innovation was better ultrasounds,” he says. “There are still a lot of people going to surgery who don’t need it. You don’t want to miss a cancer, but at the same time, you don’t want to send a bunch of people with benign nodules to surgery if they don’t need it.”
Thyroid surgery can come with risks of complications — while low, they still happen, Haugen says — and the possibility of needing medication for the rest of the patient’s life.
The success of the study means that in the future, a doctor could see a patient with a nodule, use an ultrasound, review the patient’s history and use the polygenic risk score to determine whether a biopsy or surgery is needed.
“This could reduce tens of thousands of unnecessary biopsies and subsequent surgeries,” he says. “It will be so beneficial for doctors and their patients to have another tool to guide management.”
