William G. Nelson, MD, PhD Photo by Joe Rubino

IN THE LATE 1950s, breast surgeon Bernard Fisher oversaw a clinical trial that tested whether adding chemotherapy after mastectomy could improve outcomes in women with breast cancer. This study found the addition of chemotherapy after surgery resulted in fewer recurrences and increased overall survival for a subset of the women who had cancer in four or more lymph nodes.

Since then, this practice of using additional cancer treatment after primary cancer treatment, such as surgery or radiation, to reduce the risk of cancer recurrence has become known as adjuvant therapy. Beginning with that initial trial, researchers have developed and refined adjuvant therapy regimens for many types of cancers to help provide patients with longer-lasting remissions and even cures.

Yet a conundrum remains: Although adjuvant therapy clearly extends remissions in some patients, the additional treatment may not always be needed to achieve a cure and may sometimes even be harmful. For others, the added treatment may do little to prevent cancer from coming back.

Traditionally, doctors decide when patients should receive adjuvant therapy based on the size and extent of cancer growth (stage), as well as the appearance of the cancer under the microscope (tumor grade). Over the years, other more precise tools have helped to guide these treatment decisions. Tests that measure proteins, such as estrogen or progesterone receptors inside cancer cells, can determine whether people with breast cancer should take hormone therapy. Other tests that measure gene expression, such as Oncotype DX and MammaPrint, can help determine the likelihood of recurrence and the need for additional treatment.

Now, new tools, such as DNA sequencing-based tests that detect circulating tumor DNA (ctDNA), are poised to better define who might benefit from adjuvant therapy. The new tests do more than just catalog defects in cancer genes. They collect the entire universe of DNA sequence changes that accumulate throughout the cancer cell genome and use this information to create an individualized “ultrasensitive” test.

For example, one company, called Personalis, offers a ctDNA test that incorporates as many as 1,800 DNA sequence changes based on the unique DNA sequence of each patient’s cancer. The test is highly specific for ctDNA and has remarkable sensitivity. When used to detect persistent ctDNA in the blood after surgery for breast, lung and colorectal cancers—called minimal residual disease—it reliably predicts eventual disease relapse, while the absence of ctDNA is accompanied by a very good prognosis.

Individualized ultrasensitive ctDNA tests like this one also have the potential to detect cancer recurrence far sooner than a scan. In addition, a decrease in ctDNA can reassure patients and physicians that treatments are, in fact, beneficial.

If these tests help to indicate more clearly who needs adjuvant therapy and who does not, then the decisions to begin adjuvant treatments will more closely align with the precision medicine ideal: the right treatment, at the right time, to the right patient.

William G. Nelson, MD, PhD, DSc, is the director of the Johns Hopkins Kimmel Cancer Center in Baltimore.