Ductal Lavage to Detect Breast Ca Questioned

May 1, 2004

NEW YORK-The lack of fluid yield from cancer-containing breast ducts may pose a significant barrier to using ductal lavage as a cancer-detection tool in women with mammographic calcifications, Seema A. Khan, MD, reported at the 57th Annual Cancer Symposium of the Society of Surgical Oncology (abstract 29). In a study of 20 patients with mammographic calcifications, there were 10 cases of ductal carcinoma in situ (DCIS), of which only 3 (30%) yielded fluid. "This is not a technique that should be offered to women with the hope that we might detect occult cancer," said Dr. Khan, director, Bluhm Family Breast Cancer Early Detection and Prevention Program, Northwestern Memorial Hospital.

NEW YORK—The lack of fluid yield from cancer-containing breast ducts may pose a significant barrier to using ductal lavage as a cancer-detection tool in women with mammographic calcifications, Seema A. Khan, MD, reported at the 57th Annual Cancer Symposium of the Society of Surgical Oncology (abstract 29). In a study of 20 patients with mammographic calcifications, there were 10 cases of ductal carcinoma in situ (DCIS), of which only 3 (30%) yielded fluid. "This is not a technique that should be offered to women with the hope that we might detect occult cancer," said Dr. Khan, director, Bluhm Family Breast Cancer Early Detection and Prevention Program, Northwestern Memorial Hospital.

Ductal lavage once held great promise as a potential method for cytological diagnosis of occult malignancies in the breast. These latest findings, however, follow earlier data discounting the promise of ductal lavage for detecting larger tumors, in a study also conducted by Dr. Khan and colleagues in 40 women with invasive cancers and DCIS (J Natl Cancer Instit, in press).

While the results are not encouraging, Dr. Khan said she was not yet ready to write off the potential for using this technology to diagnose breast cancer, since both of the studies were small.

In their previous study, Dr. Khan and her colleagues investigated the premise that ductal lavage could be used to detect known cancers in 40 women undergoing mastectomy. Results showed a specificity that was encouraging, according to Dr. Khan, but a "relatively poor" sensitivity ranging from 16% to 33%.

In that study, which included patients with large, palpable cancers (mean size of the invasive component, 3.2 cm), cancer detection was greater for earlier lesions. "The likelihood of detecting cancer using ductal lavage is far greater if there is significant DCIS, since the technique detects intraductal neoplasia," Dr. Khan explained.

Thus, the study presented at the Society of Surgical Oncology meeting looked at the "other end of the spectrum": women with microcalcifications (BIRADS 4 or 5) detected on screening. The 20 women in the study were scheduled to have ductal lavage prior to planned biopsy.

The lavage catheter was left in the duct, and radiographic contrast was instilled to the duct through the same catheter to confirm that the duct lavaged overlapped the area of calcification. If the duct did not overlap and there were additional fluid-yielding ducts, those additional ducts were lavaged to the extent possible. However, there were some constraints "related to patient willingness to undergo this procedure repeatedly, as well as timing and scheduling in the radiology suite," Dr. Khan reported.

Of 20 women entered into the study, four did not produce fluid (all had DCIS), and one patient did not have a ductogram. Of the 15 women who did complete that study, 4 demonstrated overlap between the area of calcification and the ductogram, indicating that the lavaged duct was the one that led to the lesion. The lesions were nonproliferative benign changes (2 cases) and atypical duct hyperplasia (1 case), along with DCIS (1 case); in this woman, the cytologic sample showed benign characteristics.

Of the eight women who showed no overlap between calcification and ductogram, DCIS was seen in four cases.

Three women demonstrated extravasation on the ductogram, so overlap could not be assessed; of those three patients, one had nonproliferative benign changes, one had atypical duct hyperplasia, and one had DCIS, Dr. Khan said.

Not all of the fluid-yielding ducts could be assessed in six women; in those patients, researchers correlated the location of the unlavaged duct with the calcifications. They found a match between duct and calcification in two women. Therefore, "there were an additional two women with DCIS who might have had lavage of the appropriate duct, but this could not be demonstrated for certain," she said.

Dr. Khan summed up: "Of the 10 patients with DCIS, one demonstrated overlap, and two additional patients may have overlapped (one matching duct extravasated and one matching duct was not lavaged), so that, at best, a total of 3 of 10 DCIS lesions were accessible with ductal lavage."