CT May Predict Response to Chemo for Pancreatic Cancer

CT scans showed pancreatic cancer response to chemotherapy can vary among patients

(RxWiki News) Pancreatic cancer is notoriously difficult to treat. Now doctors may be able to help pancreatic cancer patients better by more precisely targeting their medication therapy.

A new study has found that computed tomography (CT) can be used to see that chemotherapy is hitting its intended target and how well a tumor is absorbing the medication once it gets there.

CT is a widely used test that uses computer-processed X-rays to produce virtual slices of a body part.

This study was the first human trial to look at using CT scans to predict how much of a medication reaches the pancreatic tumor.

"If you have pancreatic cancer, talk to your doctor about the best treatment for you."

Jason Fleming, MD, a professor in surgical oncology at the University of Texas M. D. Anderson Cancer Center,was the corresponding author of this study. He and fellow researchers found that in addition to helping chemotherapy reach the tumor, CT scans may also predict how well the treatment will penetrate the tumor.

Previous studies in animals have already shown that these tumors are hard to treat because they contain things like nonfunctional blood vessels and a lot of fibrotic (scar) tissue. These combine to make pancreatic tumors hard for medications to reach and invade.

“We found that the distribution of intravenous dye used in CT scans is a surrogate for chemotherapy delivery in the tumor,” Dr. Fleming said in a press release.

Dr. Fleming and colleagues found that they could use data from routine CT scans and a mathematical formula that they themselves developed to predict how much the chemotherapy could penetrate and treat the tumor. They noted that their research could lead to better delivery of the chemotherapy.

These researchers enrolled 12 patients with pancreatic disease who were having surgical resections (partial removal of the pancreas). During surgery, each patient received an infusion of the chemotherapy medication gemcitabine. After surgery, they analyzed the tumor DNA to see how well the medication had penetrated.

This medication is known to be delivered to the tumor through a protein called the human equilibrative nucleoside transporter (hENT1). Some people have a of this protein, while others do not. When there is a lot, more of the tumor is benefiting from the treatment, and doctors can more easily see how much of the medication is reaching its target.

The researchers found that patients in this study who had dense, scarred tumors, and not much of the protein, did not have much of their tumor being treated with the chemotherapy.

Those in the study whose tumors were well-penetrated by the medication had better outcomes.

Based on this study, the researchers examined past CT scans from another group of patients to compare how well the medication reached its target. The researchers looked at the scans from 176 patients being treated for pancreatic cancer. The patients either received chemotherapy before the surgery, during the surgery or not at all.

These researchers noticed that the tumor appeared different when they analyzed the scans because the dye used in CT scans is absorbed differently. This led them to think that the dye used in scans shows absorption by the tumor much the same as chemotherapy would.

By considering facts that effect medication delivery to the tumor, the researchers found that there could be a six-fold difference in how well a patient’s tumor absorbs the medication.

Patients, therefore, may vary greatly in how well a medication may help them, the authors concluded.

Dr. Fleming said that “going forward, the implication is that molecular information from a biopsy of the tumor can be combined with data from a standard CT scan to place patients into categories that predict their response to therapy.” He acknowledged that further studies are needed.

The American Cancer Society estimates that more than 39,000 people will die from pancreatic cancer in 2014.

This study was published online on March 10 in the Journal of Clinical Investigation.

The authors reported no conflicts of interest.

Review Date: 
March 14, 2014