Circulating Tumor DNA May Be Predictive of Disease Progression in Patients with Rectal Cancer

PHILADELPHIA — Detection of circulating tumor DNA (ctDNA) after neoadjuvant chemoradiotherapy was associated with poor response and increased risk of metastases in patients with locally advanced rectal cancer, according to results published in Clinical Cancer Research, a journal of the American Association for Cancer Research.

“Responses to neoadjuvant chemoradiotherapy are variable in patients with rectal cancer,” said senior author David Cunningham, MD, director of clinical research and consultant medical oncologist at The Royal Marsden in London. “If we can accurately differentiate the good responders from the poor responders, we may be able to adapt treatment and move away from a one-size-fits-all approach.”

Current methods to assess responses to neoadjuvant chemoradiotherapy include imaging and direct examination of surgically removed tissue. However, traditional methods for assessing response on imaging can be challenging due to the scarring of tumor tissue arising from treatment. In addition, “15 to 20 percent of patients have no evidence of cancer remaining in the tissue after neoadjuvant treatment. Establishing whether these patients could have been spared from surgery if their responses had been detected beforehand is an area of active research,” explained Cunningham.

In this study, Cunningham and colleagues assessed the potential of ctDNA to predict tumor response and disease progression in patients with rectal cancer. The study enrolled patients with locally advanced rectal cancer, as well as patients with non-metastatic rectal cancer who were recommended to receive chemoradiotherapy. Blood samples were used to look for ctDNA and if present, to measure levels of ctDNA. Samples were collected before, during, and after chemoradiotherapy, as well as after surgery. For patients who did not receive surgery, blood samples were collected every three to six months after chemoradiotherapy until within three months of tumor regrowth.

Forty-seven patients were evaluable for ctDNA and primary tumor response. Thirty-three percent of patients with poor tumor responses, as measured by tumor regression grade on magnetic resonance imaging, had detectable levels of ctDNA after chemoradiotherapy, compared to 5 percent of patients with good tumor responses. All three patients with pathologic complete responses had detectable ctDNA prior to chemoradiotherapy and undetectable ctDNA from mid-treatment onwards. Detection of ctDNA was not associated with tumor response measured by RECIST.

Detection of ctDNA after chemoradiotherapy and persistence of ctDNA during treatment was also associated with the development of metastasis. Seven of the 10 patients (70 percent) with detectable ctDNA upon completion of chemoradiotherapy went on to develop metastases, compared to four of the 37 patients (11 percent) who did not have detectable ctDNA after chemoradiotherapy. Furthermore, metastasis-free survival was significantly shorter in patients with detectable ctDNA upon completion of chemoradiotherapy and in those with ctDNA that persisted throughout treatment. 

The authors also found that detection of ctDNA after surgery was associated with relapse. All three patients with detectable ctDNA post-surgery experienced relapse, compared to zero relapses among the 20 patients without detectable ctDNA post-surgery.

“Our results indicate that ctDNA can be used to identify patients who have a greater risk of developing metastases or experiencing relapse,” said first author Shelize Khakoo, MD, medical oncologist at The Royal Marsden. 

The authors are currently recruiting colon and rectal cancer patients into a multicenter study that will use the presence or absence of ctDNA to guide treatment decisions after surgery. “If the associations we’ve observed during the neoadjuvant period are confirmed, detection of ctDNA could also be used to tailor treatment for patients with rectal cancer in an attempt to prevent cancer spread,” said Cunningham.

A limitation of the study is the small sample size.

This study was supported by the National Institute for Health Research Biomedical Research Centre at The Royal Marsden/The Institute of Cancer Research. Cunningham has received grants from 4SC, Amgen, AstraZeneca, Bayer, Celgene, Clovis, Eli Lilly, Janssen, Medimmune, Merck, Merrimack, and Sanofi. Khakoo declares no conflicts of interest.