| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
University of Oklahoma Breast Institute, 920 S. L. Young Boulevard, Williams Pavilion, Suite 2290, Oklahoma City, Oklahoma 73104
Correspondence: Address correspondence and reprint requests to: William C. Dooley, MD, FACS; E-mail: william-dooley{at}ouhsc.edu.
Khan et al.1 have reported a small series that raises questions about ductal lavage as a cancer detection method. Lavage was developed as a tool to enhance nipple aspiration in the detection of increased breast cancer risk. Almost from the very beginning, investigators, practicing physicians, and the lay public expected that such a straightforward and logical approach would offer an alternative to mammography for screening for breast cancer. As this article clearly shows, we are a long way from developing a powerful replacement for mammographic screening for breast cancer.
The first real question is why there was such a thirst for a new screening method. Mammography has dramatically decreased the average size of breast cancer at diagnosis in the last several years and, with this, has led to increases in survival. Why abandon something that is working? The answer lies in our increasing understanding of breast cancer biology. By the time a 6-mm breast cancer is seen on mammography, the cancer has already been invasive on average for longer than 6 to 10 years. If we really want to cure breast cancer with current technologies, we would have to diagnose virtually all as in situ or very early after invasion has occurred. No anatomical imaging method so far (mammography, magnetic resonance imaging, or ultrasonography) has any real hope of achieving this goal. With this gloomy realization comes desperation for a method that can exploit biology to reveal the earliest cancers.
The second question is why nipple aspiration failed as a method for cancer detection. This is now an old story—too few cells for cytopathologists to detect cancer and too few patients with cancer who produce enough fluid and cells at the nipple. Larger series of nipple aspiration were able to stratify patients into high versus low risk for cancer development.2 First, the ability to elicit nipple fluid was associated with a 1.88 relative risk. Second, if the fluid had atypical cells by cytopathology criteria, then the relative risk increased to 5.0. The initial lavage study was performed just to extend these observations to a new technique capable of eliciting higher cellular yields.3 Over the next decade, serial follow-up of lavage patients will tell us whether this ability to elicit cells from more patients has power the same as or greater than that in the original nipple aspiration series for risk assessment. Unfortunately, both nipple aspiration and lavage may fail to detect cancers by at least two mechanisms: failure of the involved duct to produce fluid and failure of the fluid to contain enough cytologically distinct cells to result in a pathologic call of malignancy. We have accepted that both occur and prevent the widespread use of nipple aspiration. Khan et al.1 show that both are still true for the current version of ductal lavage.
This series is much too small to tell us much about which mechanism contributes the most to the failure of either nipple aspiration or ductal lavage as a cancer detection method and how best to resolve this issue. Ductal endoscopy series from the last 15 years can help to clear up much of the confusion.4–11 First, not all cancers produce fluid at the time of cancer detection, but small tumors <3 cm and those associated with extensive intraductal carcinoma and atypical hyperplasia seem to be most likely to produce fluid. Although these series do not yet show a consensus on the exact percentage of fluid-producing early cancer ductal systems, the results strongly suggest that it is >70% overall. These series also clearly demonstrate that mammographical duct overlap would grossly underestimate the true percentage of connecting ducts. Cancers often develop very eccentric to the main branches of a ductal tree and may have mammographically unrecognized ductal carcinoma-in-situ that totally obstructs ducts for >2 cm.
Molecular studies on both lavage and ductoscopy series suggest strongly that cancers can be detected by relatively simple molecular methods—centromeric aneuploidy and gene hypermethylation.12 When the panel of markers is optimized for those present in the cancer, detection is >97%. The implication is strong that current cytopathology is too weak to detect rare cancer cells in these small specimens.
The future of the intraductal approach to breast cancer will lie with applying the new molecular tools with the new intraductal biopsy methods being developed by the sub–millimeter-scope companies. From these, we will learn about both the evolution of multifocality and the differential progression toward breast cancer within the ductal tree. Currently, even with a clear picture of a submillimeter mass and multiple.1-mm core biopsy samples, cytopathologists cannot render precise meaningful diagnoses. As molecular tools change this situation, we may see the development of a biologically rational approach to breast cancer screening that is not dependent on rather late anatomical changes such as microcalcifications and densities from increased scarring.
Received for publication February 13, 2005. Accepted for publication March 17, 2005.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
