The Power of One: Using Single Molecule Sensitivity Diagnostics to Fight Cancer

GUEST AUTHOR: Oguzhan Atay, Co-founder & CEO of BillionToOne

For two million Americans every year, facing a cancer diagnosis is the most harrowing experience of their lives. And each of them wants to know what their prognosis is.

Despite commendable advancements in cancer treatment, however, oncologists can never offer 100% certainty of prognosis. Instead, oncology teams closely monitor patients during and after treatment to determine how the cancer is responding, so that they can continue, change, or escalate treatment.

Unfortunately, that is easier said than done using existing imaging tools. On the one hand, extremely small tumors, and undetectable remnants of surgically removed tumors are sometimes impossible to spot, allowing the cancer to grow. On the other hand, inflammation caused by the immune response, as well as necrotic tissue left over from previous treatments, can make tumors look larger than they are. In either case, diagnostic imaging is frequently misleading.

Conventional liquid biopsies (LB) offer another diagnostic method, but have lower sensitivity when tumor DNA circulating in the blood is in quantities less than 0.5% of the total cell-free DNA (cfDNA) in the sample. As a result of this relatively low sensitivity, conventional LBs miss cancer DNA in samples or fail to identify driver mutations that could be treated with targeted therapies.

For cancer patients, these diagnostic shortcomings mean that—on top of everything else that they must face—they must also live with the anxiety of uncertainty. And studies show that depression and anxiety are linked to lower cancer survival rates.

This is a challenge that we aimed to solve when we formed BillionToOne. The name refers to our ability to precisely identify and quantify cfDNA molecules in a blood sample, right down to a single molecule for each base pair (letter in our genetic code), amongst three billion others. To do that, we add bespoke synthetic molecule controls into samples before amplifying them for analysis. With machine learning, we can then distinguish the signals that we are looking for from the noise that is created in the sample amplification process. When we pinpoint a DNA mutation, we can tell if it is really from a tumor, or if it is simply an artifact of the amplification and sequencing process.

Single-molecule sensitivity and quantification of cfDNA in the blood of cancer patients means better therapy selection. In a head-to-head study presented at the American Society of Clinical Oncology (ASCO) conference, BillionToOne’s Northstar Select therapy test detected 50% more actionable mutations than conventional LBs. This means that more patients can receive a more effective therapy targeting the driver mutation for their cancer, or be selected for clinical trials of new therapies appropriate for their cancer’s mutational profile.

We also know with much higher precision than scans or conventional LBs if a tumor is growing or shrinking in response to therapy. Our most recent data shows that we can make this distinction as much as 6 months before the response is apparent in a scan.

Our most important tools in fighting cancer are early diagnosis and accurate treatment monitoring. While, at the moment, BillionToOne’s tests are available for Stage III and IV cancers, as this technology continues to rapidly advance, diagnoses will occur earlier. Within a decade, we expect to be able to accurately diagnose cancer at Stages I and II, even for asymptomatic patients in a general population setting, potentially saving the lives of millions worldwide whose diagnoses currently come too late.