Screening breast cancer

Breast cancer is the most common type of cancer among women in the US other than skin cancer. Every year, more than 211,000 American women (and 1700 men) learn they have the disease. EHM spoke with Dr. Ellen B. Mendelson, Chief of Breast Imaging at the Lynn Sage Comprehensive Breast Center at Northwestern Memorial Hospital, and Professor of Radiology at Northwestern Feinberg School of Medicine about screening, symptoms, diagnosis, and treatment of breast cancer.

EHM. How does ultrasound in screening for and diagnosis of breast cancer compare against x-ray?

EM. The answer is forthcoming. From earlier individual studies that have been published we know that mammography has limitations in detection of masses predominantly in women who have dense fibroglandular breasts. Dense fibroglandular breast tissue translates into whiteness on an x-ray mammogram, and a mass that is a cancer or a cyst would also be white, and there’s no background contrast for a white mass to stand out against.

Ultrasound gives you a very small window into tissue. For breast tissue, the ultrasound studies that we do have a thin slice of breast tissue which is about five centimeters square at most. And that area will include all of the tissue at that sight. Ultrasound is an excellent way to characterize something that you may suspect is present on a mammogram or a palpable abnormality that either you yourself feel or your physician has observed when you come for a physical exam. Masses on ultrasound will look dark against the same fibroglandular background which is white or light grey on ultrasound.

What has been studied is how is there an adaptation of ultrasound that we can use for screening. We are now concluding a multi-centered trial conducted by ACRIN (American College of Radiology Imaging Network) and the National Cancer Institute. The study was for initial mammography and ultrasound, and then several follow-up exams. The study results, I would expect, although I haven’t seen the data yet, are going to support the adjunctive use of ultrasound for women who have dense breast tissue and are at high risk.

EHM. How do you assess the above against MRI?

EM. MRI is an entirely different kind of imaging modality. Recently, the American Cancer Society in a consensus document suggested a change in their guidelines for breast cancer screening to include annual MRI in addition to but not to replace mammography for women who have genetic abnormalities, or the familial pattern that suggests it, or women at very high risk for other reasons.

Whereas ultrasound, for example, has a small window into breast tissue, MRI will show you the whole picture. And it can show it to you in both breasts at the same time so that you have a larger field of view. You can reconstruct the images in any plane that you want.

The criticism of MRI is that it shows too much, that there are many false positives, and there are also some false negatives. It involves intravenous injection of contrast material, so there is a minor amount of risk, and there is also, lately, a warning about nephrogenic systemic fibrosis. MRI is another reason for why we need to look to ultrasound and perhaps modifications of the way ultrasound is performed, because we can’t afford to screen everyone that is at high risk with MRI. The studies are expensive. Unless your insurance covers it, and there have been some problems in coverage, the individual is billed at the retail price.

The numbers of magnets and access to these is also somewhat limited. If everyone, for example, with dense breast tissue would want to have their screening done along with mammography, there wouldn’t be enough time, there wouldn’t be enough MRI systems in this country, and the cost would be prohibiting. The applications of MRI that we are endorsing at this moment are for evaluation of extent of disease where you really don’t have a very good sense of what the breast is hiding if the tissue is dense either on the side of a new breast cancer that has been recently diagnosed or suspected in the other breast.

EHM. One of your interests is in imaging-guided percutaneous biopsy and ablative techniques. What have been major advancements in these fields?

EM. At Northwestern, we are multidisciplinary. We interact with our teammates in surgery and radiation, oncology and medical oncology on a daily basis. It is becoming the standard of care that any cancer that you need to diagnose should be diagnosed purcutaneously with image guidance or with palpation prior to going to surgery or selecting it.

What has also become standard of care in the US in the management and the therapy of breast cancer is to do sentinel node imaging. The greatest complications we see for breast cancer surgery is from axillary dissection or axillary sampling where lymphoedema can be a lifelong problem for that patient and for years to come. Sentinel node imaging, and it can be done a couple of different ways, is done on the day of surgery, and what accomplishes it is either uptake of a radionuclide or a vital dye, or both.

That is one reason that there has been a huge jump in diagnoses made with imaging guidance with purcutaneous techniques. An ultrasound is by far the easiest and best way for the patient as you are seeing in real time exactly what you are sampling as you go. And in core biopsy, where you get a sample that is analyzed histologically, so you see the architecture of the tissue, the results have been very good. The use of fine needles is probably a higher art in Europe than it is in this country where there is very variable expertise in cytopathology.

There are many different devices, some of them with different mechanics, that can be used, and both the spring-activated devices as well as the vacuum-assisted devices can be used for all imaging modalities, ultrasound, stereotactic guidance and MRI guidance.

EHM. Another one of your interests is in imaging protocols for follow-up of breast cancer patients treated conservatively. What have been improvements here?

EM. Improvements here are just the experience of all the breast imagers who see these patients – I’ve been studying it for 20 years. Initially, most sites in the US were still doing predominantly modified radical mastectomies for patients with breast cancer. The statistics that had been gathered were really in support of the equivalence, stage-for-stage, of modified radical mastectomy, lumpectomy and radiation therapy.

For invasive disease you do need radiation therapy. The recurrence rate without radiation therapy is much higher The follow-up of patients is made very difficult, not so much by the surgery, but by the radiation therapy. It takes much longer for whatever the resolution of some of the changes will be. The changes are breast swelling and edema, where there’s fluid retention in the breast, as well as the scar, which can simulate cancer, and the skin becomes thick.

In breast cancer treatment, what’s been found is that the cosmetic result is better if you allow for slow absorption of this fluid rather than aspirating it out and leaving a big cavity in the breast when you removed the fluid. And the numbers of abscesses and inflammatory processes after the surgery is very small. So there is very little risk of that for a given patient.

But the changes resolve very slowly, and some of the changes never go away for some of the patients who have been radiated. Think about slow motion. The radiation attacks the malignant process, and if there is still something that’s going to recur it makes it happen more slowly so that after radiation, for example, recurrence is usually after about three years.

Once in a while, not all that often, calcifications also occur after radiation therapy and after surgery too, and if that’s how the cancer has presented, it can be sometimes as difficult to differentiate whether the new calcifications are related to the treatment or whether they are representative of new disease.

It’s hard to see, and many of the changes mimic breast cancer. So an understanding of the temporal changes and when you would expect to see them is what is the tricky part of the follow-up of these patients. You don’t want to suggest they have recurrence when it’s just a post-treatment change that you’re seeing. For patients who have had breast cancer and have gone through a lot, you want to minimize your false positives as far as recurrences or new cancers are concerned.

There rarely is use for MRI, but we always use mammography. Breast ultrasound can be helpful, too. MRI for a mature scar after at least two years will be able to show enhancement if there is recurrence. The recurrence rate – at least at Northwestern – is about two percent, which is very low. You may use MRI for confirmation of a new abnormality, as a potential area of recurrence. But again, you need to biopsy to confirm it as an enhancing lesion after about two years.

EHM. Which are the areas where you would like to see progress being made?

EM. I’d like to see definition of how all of our various imaging modalities fit together. I’d like to see continued research on prevention and cause of breast cancer, and that may not involve imaging, but I think it will as molecular imaging becomes more mature and we can apply it clinically to breast cancer detection treatment. I’m looking ahead to another area of imaging where we will be satisfied that we have done as well as we can in causation, prevention and diagnosis and treatment.

I’d also like to see development in the immediate future – and we’re working on this – of an ultrasound method that takes some of the subjectivity out of ultrasound, which is an operator dependent small window type of examination, and make it a broader, automated type of procedure where we can make certain that we have imaged the entire breast tissue and have a rapid way – probably scrolling through it on a monitor – to be able to detect lesions that are occult to other imaging techniques, like mammography, and bringing an adjunctive screening method to mammography to increase the yield of breast cancers in women for whom mammography is limited. I think that we’re very close to this automated type of ultrasound where your eyes will be able to detect on representations of the tissue that show you the entire breast rather than depending on either someone else or yourself to detect at real time where a possible abnormality would be.

It’s difficult. Detection and characterization are two different things, and it would be very helpful if we were able to overcome some of the operator dependence of ultrasound and optimize it for screening.

Ellen B. Mendelson is a professor of radiology and Chief of Breast Imaging at the Lynn Sage Comprehensive Breast Center at Northwestern Memorial Hospital in Chicago.