The study, conducted by researchers at the Fox Chase Cancer Center, also uncovered evidence that not all circulating tumour cells are the same, and some may predict worse outcomes than others.
Benjamin P. Negin, M.D., a medical oncology fellow at Fox Chase, said: "Eventually, we hope we can use changes in number of circulating tumour cells to make real-time treatment decisions, instead of having to wait weeks for radiological scans.
"This is one of the first studies looking at the role of circulating tumour cells in pancreatic cancer."
Previous work has demonstrated that prostate, colon, or breast cancer patients who have more circulating tumour cells have poorer survival than patients with fewer circulating tumour cells.
To learn whether a similar scenario occurs in pancreatic cancer patients, Negin and colleagues enrolled 48 patients with pancreatic cancer in the current study (additional patients still being added).
The team used an immunomagnetic separation system to isolate circulating tumour cells from patients' blood at three time points.
The time points included study entry, seven days after the start of therapy, and at the time of their first radiological evaluation, which was six to ten weeks after initiating therapy.
The team found that 50 per cent of the patients had one or more circulating tumour cell per 7.5 mL peripheral blood at baseline.
That proportion decreased to 40 per cent after seven days of therapy and to 32 per cent % at the time of the first scan.
Moreover, the presence of circulating tumour cells correlated with patient outcomes.
Patients with circulating tumour cells at study entry had a median overall survival of 191 days compared with 269 days for those without circulating tumour cells, although the difference did not reach statistical significance in the small study population.
A similar trend appeared for progression-free survival, with a median of 85 days for those with detectable circulating tumour cells compared with 137 days for those without.
Patients who had circulating tumour cells at either point after therapy initiation also trended towards poorer overall or progression-free survival.
Interestingly, the investigators found that some circulating tumour cells may be worse than others.
Patients whose circulating tumour cells expressed the MUC1 protein, which has been associated with more aggressive pancreatic cancer, trended towards a shorter median overall survival than those whose circulating tumour cells did not express the protein, at 85.5 and 310 days, respectively.
Negin said: "We find that having circulating tumour cells is bad.
"But the more interesting story that appears to be coming out of our study is that not all circulating tumour cells are equal. The MUC1 cells seem particularly bad, suggesting that there is a difference in the biology of these tumours and providing some insight into how these tumours function."
Negin will present the study at the 46th Annual Meeting of the American Society of Clinical Oncology on June 6.