Q&A: Ray DuBois on Screening, Early Detection

Throughout his career, Raymond DuBois, MD, PhD, has been interested in finding ways to pinpoint cancer in its earliest stages or prevent it altogether. “Almost 90% of patients diagnosed with any type of localized cancer can survive for 5 years, versus only 21% of those diagnosed with advanced metastatic disease,” he notes. “Early detection is key from my perspective.”

As such, DuBois, who directs the Medical University of South Carolina's Hollings Cancer Center in Charleston, strongly advocates effective cancer screening—Pap smears have reduced cervical cancer mortality in the United States by about 70% and, since 1990, mammograms have cut breast cancer mortality by 40%—and expanding its availability.

Speaking with Alissa Poh for Cancer Discovery, DuBois offered further insights and discussed his recent testimony before a U.S. House subcommittee, as well as Hollings's mobile screening program.

Some key House subcommittee members, having lost loved ones to cancer, want expanded screening for subsets of individuals. These include younger women who develop lung cancer without prior tobacco exposure—they're missed by current screening criteria, which are based on smoking history and age. I was invited to provide input on proposed legislation (Cancer Prev Res 2021;14:1053–4).

I tried to help them understand that it wouldn't be wise to screen all young women who could be at risk of lung cancer; it's still an uncommon disease in this subset. We must be smart about stratifying who could benefit, perhaps using whole-genome sequencing as it gets cheaper, along with other 'omics methods and deep learning algorithms, once fully validated.

Yes. I hate to say this, but unfortunately, screening in this country depends on having some sort of insurance coverage. One goal of Medicaid expansion under the Affordable Care Act was to make screening more equitably available to underserved populations. It's had a positive impact where adopted. Screening is now covered for more individuals, and studies have shown that the result is reduced mortality.

This is a bit of a political hot potato. Deciding against expansion meant many Southern states didn't receive billions of federal dollars in health care support, which would have helped improve outcomes for their most vulnerable residents. I hope more people come around to the fact that there are certain essential services we have to provide as part of regular health care, including screening, or we'll pay a steeper price in the long run, caring for sicker patients.

One possible solution being discussed is developing a relatively low-cost pan-cancer blood test that could be done at home.

It's emerging as one of the most promising biomarkers for early cancer detection. Preliminary results with ctDNA tests suggest reasonable specificity in the 99% range. Sensitivity, however, depends on cancer type and stage—it's approximately 80% for stage III disease, but only 40% for stage I, which is too low. There's room for improvement. We also haven't seen data from a large enough sample size to better understand real-world test performance. Ultimately, demonstrating reduced mortality in prospective randomized screening trials will be necessary for full validation. So, this isn't quite ready for prime time, but it will be exciting to see how things develop—maybe we'll need both DNA methylation and mutational analyses to get the best, most accurate tests.

Yes, for the last 20 years, we've focused on areas of persistent poverty, where families average $20,000 or less a year for five to six people, usually with no health insurance. We started with a basic mammography van, and now we have a state-of-the-art unit equipped with 3D digital breast tomosynthesis technology. The secret to success is building long-term, trusting community relationships and being sensitive to the needs of individuals, with churches, particularly pastors, as partners. Our vans show up regularly to provide free screening for breast, cervical, and skin cancers—and offer vaccines to prevent human papillomavirus–related cancers. Our dream is to outfit an 18-wheeler with a low-dose CT scanner for lung cancer screening.

Some places we serve still lack reliable Wi-Fi, making it difficult to download images into our database for analysis. One of the Biden administration's infrastructure goals is to improve broadband coverage in rural areas.

We could repurpose drugs for prevention through increased knowledge of cancer mechanisms. A example is aspirin, which we know can decrease colon cancer mortality by 30% to 40%, likely by inhibiting chronic inflammatory processes. Data are emerging about how aspirin also affects macrophages and other immune cells in the tumor microenvironment.

Hopefully, too, the NCI-sponsored PreCancer Genome Atlas, which is using 'omics to characterize common premalignant lesions, will shed light on markers for early detection. If we disrupt oncogenesis by hitting these new molecular targets, we might truly alleviate the burden of cancer. ■