That being said, it’s important to take aim at the right type of lesion. Think palpable lesions. "Particularly in palpable lesions, fine-needle aspiration should be used as the initial approach," says Dr. Masood, who is professor and associate chair of the Department of Pathology at the University of Florida and chief of pathology and laboratory service and director of the breast center at the UF Health Science Center/Shands, Jacksonville.

Helen H. Wang, MD, DrPH, seconds that notion. "If there is a lesion that can be palpated by physical exam, it’s very easy to do a fine-needle aspiration first," says Dr. Wang, associate professor of pathology, Harvard Medical School, and director of cytopathology, Beth Israel Deaconess Medical Center, Boston. "You don’t lose anything, and you can get a diagnosis very quickly." Indeed, if a direct smear is made, the pathologist can offer a bedside interpretation. And while some detractors "talk about epithelial misplacement," says Dr. Wang, "you rarely compromise the lesion. Maybe a little bit, sometimes."

FNA is appropriate "if it’s a superficial tumor, if it is obviously benign, if it is obviously malignant," says Syed A. Hoda, MD. "Then the fine-needle aspiration can confirm the impression very quickly." FNA is also the best choice for patients who continue to present with huge tumors, many of which are considered to be inoperable. A quick FNA can make the call, enabling patients to start chemotherapy immediately. "But those cases are very rare, at least in our setting," says Dr. Hoda, associate professor of pathology, New York-Presbyterian Hospital, Weill Cornell Medical College of Cornell University.

Because FNA is minimally invasive, patients tolerate the procedure well. "It’s very atraumatic," Dr. Masood notes. Unlike core biopsy, it does not produce scarring or hemorrhaging, nor does it interfere with subsequent mammographic interpretation. FNA may also be preferable for sampling some nonpalpable lesions that are evident on ultrasound, she suggests, to reduce patient trauma and discomfort.

FNA is cheaper to do than core biopsies, hands down. As Dr. Wang puts it: "The core biopsy gun is a lot more expensive than a syringe and a needle."

In short, "If we can get a diagnosis with a simple, pretty noninvasive procedure, with minimal cost, then that’s wonderful," says Dr. Wang. "If that is not possible, then we move on to core biopsy."

In some quarters, that shift is coming preemptively. "It depends on who you are talking to," says Dr. Wang. The number of breast core-needle biopsies has surged in recent years, she says, unlike breast FNAs. "But some labs do show an increase [in FNAs], especially for the
palpable lesions."

Others report a marked decrease. Dr. Hoda recalls that 10 years ago, during his cytology fellowship, "almost every lesion used to be aspirated. Now, I think only about 25 percent of them are."

"The core biopsies are taking over," says Yolanda C. Oertel, MD. She sees it in her own practice, as director of the fine-needle aspiration service at Washington (D.C.) Hospital Center. "Breast aspirates used to be the most common aspirate in my practice, until about 1984. Then thyroid took over, and breast aspirates keep declining."

A similar shift occurred in prostate cancer, she notes. "Years ago, we used to do quite a few prostatic aspirates, and then core biopsy took over. So I am afraid that breast aspirates are going to go the way of the prostatic aspirate."

The reasons for FNA’s dwindling fortunes? Take your pick.

Though FNA is cheaper, "Cost doesn’t seem to be a primary concern in this country," says Dr. Wang—or at least not when the threat of legal action comes knocking. In medical malpractice, cases of missed carcinoma are "so expensive, people will do anything to avoid it," Dr. Wang says. "We want to make absolutely sure that we don’t miss a carcinoma, and so people go the full length to make sure we don’t." In that scenario, the more-expensive core biopsy procedure becomes attractive because it provides more material, permits more ancillary studies, and does not require cytology training to interpret it.

Lamenting the move away from FNA, Dr. Masood places part of the blame for its decline on pathologists. "They don’t want to take the time to work through the challenges of interpreting FNA biopsy."

FNA is a deceptively simple technique, observes Dr. Oertel, which also works against it. "Many people, particularly surgeons, think it’s just a matter of sticking a needle in. It’s not." This has led to a too-high rate of unsatisfactory specimens, which further dampens interest in FNA. "We need to be mastering the technique and teaching surgeons how to do it better," Dr. Oertel says. Otherwise, she says, "Surgeons get disgusted with it, and they send the patient, even for palpable lesions, for a core biopsy." It goes without saying that those who don’t perform the technique won’t get better at it.

This catch-22 traps pathologists as well. "This is not something that someone can interpret if he or she does only a few specimens a month," says Dr. Wang. "It requires experience." It also requires precise understanding between those who render the cytology report and those who receive it. "This is extremely important when we are dealing with small specimens," she says. "People have to set their expectations accordingly—that there are limitations in interpretation because of the very small amount of material.

"They also have to understand there are overlapping features between benign and malignant lesions of the breast," she continues. "Certain malignant lesions can look very, very banal, and there are certain benign lesions that can look quite atypical. The clinician and the pathologist need to have the same understanding of the words and categories we use in
our reports."

With breast tumor size shrinking, FNA loses out a little more. Fifty years ago, women diagnosed with breast cancer had tumors the size of a small lemon, says Dr. Hoda. "But now, at least in our practice, the average tumor measures less than 2 cm. And with smaller lesions, the less you manipulate them, the better. So it’s best to get the diagnosis in one shot." Using core biopsy initially "reduces many steps in the diagnosis," he says.

FNA does have its limitations, a point readily conceded by most.

FNA cannot distinguish between in situ lesions and invasive carcinoma. "That is a problem, especially for small or nonpalpable lesions," acknowledges Dr. Wang.

Indeed, that’s the main problem with FNA, says Dr. Silverberg. On the other hand, he says, "In instances where clinically it’s obviously invasive, just knowing that it’s cancer is enough information to go on to plan treatment." Similarly, in a patient who has had breast cancer who presents with a new mass elsewhere—lymph node, skin, liver—simply obtaining malignant cells by FNA is sufficient to confirm metastatic disease. "So there are situations where FNA will probably do just as good a job as a core, and could be preferred because of the relative ease and economy," he says.

There are other chinks in FNA’s armor. Some of FNA’s weaknesses lie coiled within its strengths. Though using a fine needle minimizes trauma, it also means less material for ancillary studies, such as estrogen receptor/progesterone receptor and HER2/neu. That’s hardly a small consideration, especially for patients interested in preoperative adjuvant therapy. "We can probably get enough to do one additional study," Dr. Wang says. More than that would be difficult. "You’re talking about many passes," she says.

Dr. Oertel notes that lobular carcinoma is difficult to diagnose on aspirates. "The cells tend to be fewer and smaller, and many times you don’t get tumor cellularity," she explains.

But despite breast FNA’s limitations, and with firsthand knowledge of its declining use—80 percent of the aspirates she performs now are on thyroid—Dr. Oertel sees no reason to write off FNA. "Oh, absolutely not," she says.

"I hope breast FNA can hold its place," adds Dr. Wang.

It will have to make room for core-needle biopsy, however. Core biopsy has its advantages, and can be used to fill in some of the gaps FNA fails to fill. To wit:

Core-needle biopsy, like FNA, has suffered its share of bad press over the years. "One misunderstanding is that you can’t do ER and PR on core biopsy," says Dr. Hoda. "You can do it—we do it all the time." This misperception arose early on, when immunostaining techniques were less refined. "Now it can be easily done," he says. Performing HER2/neu immunohistochemistry studies on core biopsy material remains problematic, though, due to the considerable edge artifact present on the long, narrow specimens. "But if you were to employ FISH, it could easily be resolved on core biopsy material," Dr. Hoda says.

Others mistakenly assert core biopsy promotes tumor metastases, assuming that the greater degree of manipulation involved in the technique "feeds" the blood vessels. "But it has not been proven to be the case," says Dr. Hoda. However, repeated core biopsy—and repeated FNA, for that matter—can dislodge tumor and adjacent breast tissue.

But core needle does not slip easily into the one-size-fits-all model, any more than FNA does. As with FNA, physicians need to understand the limitations of core-needle biopsy, says Dr. Masood. "We do not want to fall into the trap of false security with core-needle biopsy," she says.

Core biopsy cannot always distinguish in situ lesions from invasive carcinoma, says Dr. Wang, although it’s much better at doing so than FNA. And core biopsy has a greater risk of producing artifact, such as epithelial misplacement.

Less common, but a risk nonetheless, is what could be termed "disappearing lesions." Explains Dr. Wang: "There are cases where the lesion is so small, they can no longer find it after the core biopsy. You can’t estimate the size of the lesion using core, and when the lesion is small, the core can remove the carcinoma completely. I haven’t heard of that happening with fine-needle aspiration."

Core-needle biopsy fails to diagnose a handful of other lesions, Dr. Masood says. Among them: papillary lesions, atypical hyperplasia, lobular lesions, fibroepithelial lesions, and radial scars. "The majority of these lesions need to be followed by a needle localization excisional biopsy," she says.

Overcalling tumors is another pitfall of core biopsy. Radial scar and sclerosing adenosis are the worst culprits, according to Dr. Hoda.

And, finally, comes this warning. "What you don’t want is to put together a package of doing every technique and seeing what works," says Dr. Masood. "I don’t approve of that approach at all."

Says Dr. Hoda: "Doing FNA, followed immediately by a core without waiting for the results of the former, is double-dipping, and it’s probably abusive of both procedures."

The story could end there. FNA will probably not evolve technically, at least not with any great strides. The aim, after all, is to use a very small needle. "That’s the beauty of it," says Dr. Wang. "It’s meant to be a minimally invasive test; it’s not meant to become a complicated and complex technique."

Core-needle biopsy has more potential for technical improvement, and has already benefited from several technical advances over the years. The gauge of the needles has gradually increased, moving from, typically, a 22-gauge needle to a 14- or even 11-gauge needle, providing wider specimens. Vacuum-assisted techniques mean biopsies can be done more quickly, and with more cores. "It’s becoming easier for the pathologist to see the tissue better and make a diagnosis," says Dr. Silverberg.

There are limits, of course. Dr. Hoda describes a core-needle technique called ABBI (advanced breast biopsy instrument), which "has a needle that’s as large as a cigar, if you can believe it," he says. The idea is to combine diagnosis and treatment in one modality. "But it has not found application, at least not in the United States," says Dr. Hoda, whose colleagues at Cornell abandoned the procedure after only a few attempts. "It was very uncomfortable, it had complications, and invariably they had to go back and dig out more tissue," he says. "ABBI was, I think, before its time."

Pushed by breakthroughs in other areas, however, both FNA and core-needle biopsy may find their horizons expanding.

In core biopsy, the impetus could come from a technique called radioablation, which uses heat to destroy tumors. This has been used for years to ablate metastatic tumors, predominantly in the liver. Because of its success, it is now being looked at, with a rather hopeful eye, to treat primary tumors, in the breast as well as in prostate, brain, and lung.

Rache M. Simmons, MD, associate professor of surgery, New York-Presbyterian Hospital, Weill Cornell Medical College of Cornell University, is exploring this technique as part of a tri-institutional trial at Cornell, MD Anderson Cancer Center, and John Wayne Cancer Institute. The researchers are looking at primary tumors 2 cm or less in diameter, located 1 cm or more from the skin and from underlying breast muscle. Guided by ultrasound, surgeons insert a probe that contains wires, which are then extruded into the tumor. The wires create a frictional heat, which destroys the cancer. (Heat is only one of several intriguing possibilities, notes Dr. Simmons; other researchers are assessing whether freezing, cryoablation, or lasers can be used to destroy tumor without resorting to surgery.)

Currently, the researchers are following an ablate-and-resect protocol. "Once we have the tumor out, we’re looking at how effective the probe was in actually destroying the cancer," Dr. Simmons says. "In addition to the gross analysis and the H&E analysis, we’re also doing an NADH stain, looking for tissue viability."

So far, she says, "The results have been good, not fantastic."

"Certainly the tissue you ablate is destroyed," she says. "There’s no question about that. The issue is making sure that you’re actually in,
the tumor."

Patients who receive neoadjuvant chemotherapy have posed a challenge. In a couple of these patients, the tumor shrunk to a patchwork of little pockets around the breast, rather than concentrically. "What happened was that one tumor ball was seen sonographically, and that was successfully ablated. But when they did the resection, they found other little pockets that weren’t detectable by imaging—and of course, they weren’t ablated," Dr. Simmons says.

In several other cases, the tumors were actually larger than they appeared to be on imaging; in another patient, what appeared to be tumor wasn’t. "What we ablated was destroyed, but it wasn’t the cancer," Dr. Simmons says.

Cautioning that the data are early, Dr. Simmons says, "This works well in killing cancer; the limitation is being able to say with confidence, ’This is all the tumor that there is.’ "

"It’s not a matter of improving the technique," she says. "It’s a matter of selecting out the appropriate patients for whom it will be successful." Adding MRI and PET scans to the currently used methods of ultrasound and mammogram may help physicians do just that.

That, in turn, could open the door for more core biopsies.

"You have to use core biopsies, because when you ablate the tumor, you lose the ability to get the ER, PR, and HER2/neu," Dr. Simmons says. "You also want to confirm that it is indeed cancer." Furthermore, she says, "If we ever move to an ablate-and-not-resect protocol, we may want to do cores after we’ve done the ablation, to see if there’s residual tumor."

FNA could also see expanded use—in risk assessment. But first, a detour through nipple aspiration and ductal lavage.

Ductal lavage is not a diagnostic technique; it’s a tool for assessing the level of risk for developing breast cancer in those who already fall into the "high risk" category.

Nor is ductal lavage a screening technique for the general population. However, the high-risk category alone comprises a "pretty significant population, because breast cancer is such a common disease," says Britt-Marie Ljung, MD.

The technique is an offshoot of nipple aspiration, which, up to now, has been used primarily as an investigative tool. "Nipple aspirations have been around for a long time," says Dr. Ljung, professor of pathology at the University of California in San Francisco. "Ductal lavage has really been built on nipple aspiration data," which show that women with abnormal cytology in nipple aspirates of breast fluid have an increased risk of breast cancer (Wrensch MR, et al. J Natl Cancer Inst. 2001;93: 1791-1798).

The problem is that the nipple aspiration samples—obtained via a small suction device placed on the nipple—are quite small. "You can’t do very much with it, and the material you get frequently does not contain the epithelial cells," says Dr. Ljung. Moreover, not everyone produces nipple aspiration fluid, although there is some indication that those who don’t produce fluid have a significantly lower risk of developing breast cancer. (On the other hand, just to give you an idea of how complicated this all is, women who’ve never had children—a risk factor for breast cancer—do not readily produce ductal lavage fluid.)

Ductal lavage can produce much more diagnostic material. In a recently published study that Dr. Ljung participated in (Dooley WC, et al. J Natl Cancer Inst. 2001;93:1624-1632), which looked at high-risk women, the median number of epithelial cells collected by nipple aspiration was 120 per breast, versus 13,500 per duct using ductal lavage. "I think there is significant possibility that we’ll be able to do a number of molecular studies on these specimens and more precisely pin down which patients are truly at the highest risk of developing cancer," says Dr. Ljung. "It’s also possible we can use ductal lavage to monitor patients on prophylactic treatment, such as tamoxifen, or possibly even to introduce treatment through this
access route."

The first step in ductal lavage is to produce nipple aspirate fluid—the droplets emerging on the surface indicate duct openings. Not all duct openings in a given nipple will produce fluid, though in patients with increased risk for breast cancer, 80 to 90 percent will produce fluid from at least one duct, says Dr. Ljung. A catheter is then inserted into the open duct to introduce saline, 2 to 4 mLs at a time. This is repeated until 10 to 20 mLs have been washed through the system.

The cells that are retrieved from the duct linings can be prepared in any number of ways, ranging from Millipore filters to Cytospin to ThinPrep or AutoCyte Prep. Many give high marks to the liquid-based collection devices.

So far, so good.

"People generally have no problem identifying straightforward normal specimens—benign and malignant," says Dr. Ljung, who along with retired pathologist Eileen B. King, MD, developed the cytologic criteria for examining ductal lavage specimens; she also reviewed specimens as part of a Web-based training program run by the manufacturer of a ductal lavage device. (The company, Pro-Duct Health, was acquired by Cytyc Corp. late last year.) "When there is an abnormality, they tend to overinterpret a little bit—what they call ’marked changes’ we would downgrade to ’mild changes,’" she says. "And there are some cases where they call it ’mild changes,’ and there isn’t really anything."

Mild changes most likely represent a benign proliferative process of some sort, she says. "It’s not totally normal, but it’s nothing approaching carcinoma in situ." Benign papillary processes, such as papillomas and papillomatosis, account for a fairly large chunk of that category. "As we get more experience with these types of lesions, the features are actually quite specific," she says. Furthermore, there are data in the literature indicating papillary proliferations are associated with an increased risk of developing breast cancer. "So this is a meaningful finding in these specimens."

Marked changes are more severe, of course. "In that category, I would have significant concern about a neoplastic process going on," says Dr. Ljung. "But there aren’t enough features to make a definitive diagnosis." In a very small subset of patients, she adds, frank carcinoma can be seen in the specimens. But the "main purpose of the procedure is to risk stratify and find early changes before frank carcinoma is present," she says.

Now for the limitations.

"The problem with ductal lavage is that it’s new," says Seema A. Khan, MD, SM, associate professor of surgery, Northwestern Memorial Hospital, Northwestern University, Chicago. "Which, of course, is not a problem in and of itself, but as with anything new, we don’t have sufficient information to define where it fits in."

Dr. Khan, who participated in the Dooley study, says it’s reasonable to assume that the presence of abnormal cells retrieved by ductal lavage is an indicator of breast cancer risk. The next leap of logic—that the procedure can be used to detect breast cancer early in women who have normal physical exams and normal mammograms—is not clear-cut. "That’s because we don’t know the accuracy of these findings of atypical cells, in terms of being able to identify a real abnormality in the breast," she says.

One of the problems is the branches of the ductal system can occupy a fairly large volume of breast tissue. Within that large volume, physicians would need to localize the lesion producing the atypical lavage cells. "That has not been established yet," says Dr. Khan. "And we would need to do that before we can start assuming that finding abnormal cells by ductal lavage is a method of early cancer detection." MRI, ductograms, ductoscopy, or injected dye may provide means to do this, says Dr. Ljung, and all are being looked at.

Also yet to be determined: the correlation between ductal lavage cytology findings and histopathology findings in patients undergoing surgical procedures.

Ritu Nayar, MD, MIAC, who works with Dr. Khan and is director of cytopathology at Northwestern, raises practical issues related to handling ductal lavage samples.

It’s important to prevent air drying artifact, she says, which can interfere with interpretation, and maximize cell yield. And she echoes Dr. Ljung in cautioning against overcalling degenerative or reactive changes as atypia. In addition, she says, "Subjective differences in morphologic criteria and interpretation occur, and, at times, even intraobserver variability exists. It definitely helps to take the Web-based training and get a secondary review when starting to interpret these samples."

In women being lavaged for risk assessment, who do not present with clinical abnormalities, "Wet ducts often give a lot of histiocytes and/or few to no epithelial cells," she says. "This may indicate that wet ducts are often benign. However, we have picked up some significant disease, including high-grade DCIS, during risk assessment by this procedure."

Using ductal lavage for breast cancer risk assessment is based on the fact that atypia increases that risk, Dr. Nayar notes. But, she says, "We do not know if mild atypia can be equated to histologic ADH/low-grade DCIS. Equating marked atypia to ADH/DCIS is likely to be more reliable, but we need more followup studies to be sure." Northwestern, she says, is involved in cytology-histology correlation studies in women with radiologic or pathologic, or both, abnormalities.

Nonetheless, says Dr. Ljung and others, "Ductal lavage is already clinically a useful test in a subset of patients." Patients with a Gail index of 1.7, who are considering tamoxifen treatment, "don’t have a lot of information to go on," she says. "And it’s not like tamoxifen is a no-brainer—there are plenty of side effects." Undergoing ductal lavage and finding abnormal cells, whether mild or marked, can improve the precision of risk estimation, giving these high-risk women information that will help them decide about risk-lowering medication.

With the fortunes of ductal lavage apparently on the rise, nipple aspiration cytology might be expected to remain on the back burner as little more than an investigational tool, despite the strength of the data presented by Dr. Wrensch and her fellow researchers.

Not so fast.

As it turns out, nipple aspiration fluid is extremely rich in protein, says Carol Fabian, MD, director of the breast cancer prevention center and professor of medicine at the University of Kansas Medical Center. And levels of estradiol, testosterone, and several other hormones are 10 to 15 times higher in nipple aspirate fluid than in serum.

Aromatase inhibitors "have recently been shown to be superior to tamoxifen as an adjuvant therapy for breast cancer," she says. "So what we are looking at is trying to determine whether we can see a response to these inhibitors in the prevention setting by measuring change in epithelial cell proliferation—obtained from FNA—as well as nipple aspirate fluid hormone levels."

This is one of several breast cancer prevention studies being conducted at the University of Kansas Medical Center. Additional upcoming trials include a multicenter National Cancer Institute-sponsored study to compare ductal lavage specimens with FNA.

Random periareolar FNA, to be specific. Unlike FNA done for diagnostic purposes, random FNA is used as a risk assessment tool. Explains Dr. Fabian: "About 12 years ago, we wanted to randomly sample breast tissue in high-risk women, with the idea that if the patient had widespread precancerous lesions in her breast, then we would be likely to observe these precancerous changes in the specimen by random periareolar FNA." The technique is not as random as it sounds—most of the terminal lobular duct units where breast cancers are thought to arise appear to be concentrated behind the nipple areolar area and in the upper outer quadrant.

In a study published in the Journal of the National Cancer Institute (Fabian CJ, et al. 2000;92:1217-1227), Dr. Fabian and colleagues performed random FNAs in 480 high-risk women, then followed them for development of breast cancer. "Finding atypical hyperplasia on random fine-needle aspiration was an extremely strong predictor for short-term risk for breast cancer," she says. Median followup was four years.

Their research has led to several NCI-sponsored chemoprevention trials, including the aforementioned study to compare ductal lavage and random FNA as possible methods for performing chemoprevention studies—in this case, celecoxib, an anti-inflammatory drug that may be useful against estrogen receptor-negative precancerous disease.

Random periareolar FNA will be the method used to provide cells for the primary endpoint in this study, as it has in their previous studies, says Dr. Fabian. "We know we can get cells at both baseline and at the end of the study with FNA, but no one knows if that can be done with ductal lavage." Her suspicion is that, although nipple aspirate fluid (which serves as the guide for lavage) can be obtained after 12 months of celecoxib, it may not be possible with antihormonal agents, which lower the level of estrogen or block estrogen receptor transcription in the breast tissue.

The FNA samples, though small, produce enough material to do cytology and several immunocytochemical biomarker studies, Dr. Fabian reports. "Remember, we’re aspirating both breasts, at two different sites, so there’s much more material than there usually is with a traditional diagnostic fine-needle aspiration." Interestingly, immunohistochemistry studies have not yet been done on ductal lavage specimens, though she and her colleagues are looking into it in preparation for the study.

It is too early to predict the path for many of these possible applications. Looking into the future, most offer only one solid prediction—that no one technique will outshine the rest. Each one will serve to complement, not replace, the others.

"It’s not as if we’re replacing methods that work particularly well," says Dr. Ljung. Despite the breakthroughs, despite the information overload, despite the winds of change, "There’s still a big vacuum that we’re trying to fill."

Karen Titus is CAP TODAY contributing editor and co-managing editor.