Up to 30% of patients who undergo colorectal cancer surgery also have occult liver metastases that correlate to an increased likelihood of dying from disseminated disease. Now Scottish researchers are taking another look at Doppler perfusion index (DPI) values to determine which patients should have post-surgical chemotherapy.
Laparotomy is considered the standard for detecting lesions, but the technique can miss smaller lesions. The same is true for other imaging methods, according to radiologist Dr. Edward Leen and colleagues from the Glasgow Royal Infirmary.
"Unfortunately, the early detection of these occult metastases is beyond the resolution of conventional imaging methods such as ultrasonography, computed tomography, and magnetic resonance imaging," the authors write in a recent issue of The Lancet (January 1, 2000 Vol. 355:9197, pp.34-37).
The Scottish researchers have been looking for several years at the feasibility of selecting patients to receive adjuvant chemotherapy based on a DPI ratio of hepatic arterial flow to total liver blood flow. The researchers had previously shown that liver metastases are associated with an increased ratio.
In the latest study, the researchers examined 120 patients who, based on laparotomy, ultrasound or CT findings, were thought to have undergone curative resection for colonic or rectal tumors. DPI measurements were done by a single observer with a Diasonics color duplex Doppler scanner with a 3.5 MHz convex, phased linear-array probe. The Doppler cursor was placed over the lumen of the common hepatic artery in its longitudinal axis. DPI values of 0.30 or higher were defined as abnormal compared to a control group of 50 healthy subjects with a DPI value of 0.26.
The patients were then followed up every three months until death or for at least five years. Of the 120 patients, 73 had abnormal DPI values and 53 of those developed recurrent disease. Of the 47 patients with normal DPI values, three had recurrent disease.
These results gave the DPI technique a sensitivity of 95%, specificity of 73%, and overall accuracy of 81%. This is lower than the 86% accuracy the same researchers found for DPI in an earlier study (Radiology, April 1995, Vol.195, No.1, pp.l13-116), but higher than the 65% to 75% they calculated for laparotomy, conventional ultrasound or CT in that study. DPI was also more predictive of recurrence than the Dukes' stage of the patient's cancer, they wrote.
"The ideal prognostic index for patients undergoing potentially curative resection of primary colorectal tumors should clearly differentiate between patients in whom cancer will eventually recur and those in whom cancer will not recur. This approach (DPI) is sufficiently robust for use in clinical practice," the authors concluded.
But there are drawbacks to the DPI technique. In an accompanying commentary, Dr. Yuman Fong of Memorial Sloan-Kettering Cancer Center in New York City pointed out that ultrasonography is operator-dependent, bringing reproducibility into question. Observer variability is one shortcoming; eliciting good patient cooperation is another because the DPI measurements must be done during respiratory suspension.
"Only a large study will be able to find out which combination of variables among many (nodal status, vascular invasion, molecular genetic factors, and DPI) should guide future adjuvant therapy," Fong wrote (The Lancet, Jan. 1, 2000, Vol. 355:9197, pp.5-6).
The Glasgow group did publish the results of a reproducibility study in August 1998. Using similar methodology, the DPI values were assessed by a radiologist with expertise in Doppler ultrasound, and a physicist with no previous Doppler ultrasound experience, according to an article in Radiology (August 1998, Vol. 208, No.2, pp.453-457).
"90% of the patients examined were given the same DPI status by both observers independently," Leen wrote.
However, Dr. Laurence Needleman, associate professor in the division of diagnostic ultrasound at Thomas Jefferson University in Philadelphia, said he would like to see the DPI method reproduced by a completely independent lab.
"What Leen really needed to do was teach someone else to do it and then look at the reproduciblity," Needleman said. "Without anybody else's experience, it's hard to know if it's a promising technique. It's not an entirely crazy idea because the belief is that the metastases may have already spread before the cancer is even found. He's either got a great idea and people are just not recognizing it, or it only works in his hands for now."
By Shalmali Pal
AuntMinnie.com staff writer
February 17, 2000
Copyright © 2000 AuntMinnie.com
