WEDNESDAY, APRIL 14, 1999
"Watchful Waiting or Serial Scanning: Which is Better Surveillance for Early Stage Breast Cancer?"

Surveillance is a very common and controversial issue in the treatment of breast cancer. Routine x-rays and serum markers have had varying degrees of success in predicting recurrence. Early stage could be defined as no recognized metastasis and having completed treatment. However, there may be a large period of time between when something is visible on a scan and metastasis develops. Some people may never develop metastasis, or have such long term responses that they are considered cured. Clearly there is a lot of controvery about surveillance. High risk and low risk groups have similar recurrence curves: both rise a bit for the first few years after completion of adjuvant therapy and then drop. After 10 years the risk of recurrence is about the same for both groups.

Common surveillance techniques include screening for new breast cancer. Catching cancer in the breast early is very useful, many randomized trials have shown this. Mammography and serum markers for metastatic cancer can give a lead time for diagnosis (about 6 months), but the outcome (average survival) is no different. The FDA has approved serum markers because they are highly predictive (85%) of recurrence. But most expert organizations do not recommend them because it doesn't help the patient. Dr. Tripathy thinks we should link results from serum markers with some kind of treatment. The problem here is that we don't know what the value of early treatment is. We are not yet sure of the best treatment for early stage breast cancer. The approach seems to be to try low intensity treatment first (patients who are HER2/neu positive should try Herceptin, ER/PR+ patients should try hormonal therapy).

A large American study to be presented at ASCO next month compares high dose chemotherapy and bone marrow transplant (BMT) with standard chemo for two years. Preliminary results show no difference. It is probably best to start with the least toxic therapy first. This will usually give the patient the best quality of life. Initial treatments usually have longer remission times than those you use later. Another study to be presented at ASCO compared high dose consolidation therapy with standard therapy. An example of consolidation therapy would be chemotherapy followed by bone marrow transplant. It has not shown any benefit to date with breast cancer, but has become a standard treatment for leukemia. Combination therapy (2 or 3 drugs used together) has been shown to be more effective than single drug treatment in early stage breast cancer. The 2 or 3 drugs are best used in sequence; this gives the best cell kill according to Dr.Larry Norton's research. Bonadonna's study has also shown that sequential adriamycin followed by CMF has better results that alternating the two chemotherapies. There is hope that new drugs like Herceptin will make great strides in treatment philosophies.

The question becomes what degree of sensitivity is best to improve outcome. If you cannot improve outcome, it is just a diagnostic test which is not very useful. So if the current state of things is that early detection does not improve outcome, we should not use it. Tests also have the downside of false positives--a real problem with serum markers. The value of screening depends on three things: sensitivity (does it pick up things), specificity (is it cancer), and prevalence (how common is the disease in the population). If mammography has a 2% false positive rate, one out of every 50 tests will be falsely positive. If you are screening a low risk (younger) population (2 out of 100 will get cancer), then half of all observed cancers will be false positives. This ratio is called the positive predictive value. If you are screening a high risk population (1 out of 10 has cancer) for every 2 that are false positive, you now have 8 cancers possible, so the positive predictive value is 80%. Screening a low risk population will yield a very low positive predictive value; too many unnecessary biopsies will be done to pick up a very few cancers. If you screen a group of patients in which the risk is high, high grade comedo DCIS with excision, the positive predictive value will be high. If you don't really want to have unnecessary procedures, screening should start around age 50 (with slightly higher risk). Risk factors could also include positive family history. As a general rule, the higher the risk, the higher the predictive value of screening tests, but in general risk factors for breast cancer are not strong. There are none which will more than double your risk.

Mammostatin, a peptide that inhibits the growth of breast cancer cells, is found in low levels in women with breast cancer and in higher levels in women without cancer. It is in clinical trials at M.D. Anderson. Before it can be considered a screening tool it will have to be tested on thousands of women for up to 10 years. Most of the serum markers that we use today are antibody-based tests that pick up proteins in the blood that are made by normal epithelial cells. In breast cancer cells the polarity is wrong; proteins normally in the base of the ducts are secreted in the bloodstream. This has not been sensitive enough to be a good screening tool in breast cancer . A high priority is to develop a good blood test for breast cancer, like the PSA test is for prostate cancer.

Biggest question with surveillance is what do you do with the information? We need to link early detection with effective treatment. This might be vaccine therapy. HER2/neu vaccines are being used on HER2/neu+ tumors.

When chest x-rays and bone scans are done in addition to serum markers, markers tend to lead scans and x-rays by about 6 months in predicting recurrence. It does not seem to matter the grade or histology, only that the cancers are invasive.

If cancer is a very individualized disease as it seems to be, how can you treat it with general therapies? Therapies actually work on patients in an individualized way, we just don't know what that is. Some areas in which we have made headway in our understanding is estrogen receptor therapy (as in HER2/neu). There is a big initiative by the National Cancer Institute to begin to classify tumors according to their genetic fingerprint. This will take a large number of studies to actually get individualized information and then use it to tailor a therapy. Eastern medicine has always been highly individualized; depending on pulse and tongue diagnoses. Herbal therapies work this way. We are working to classify cancer according to their molecular lesion and then gearing therapy to the particular lesion.

Many of the HER2/neu tests have been considered defective. The protein may degrade in the tumor block making it more difficult to pick up the antibodies. The accuracy rate of testing in the best hands is only about 80%. One reassuring part is that most extremes will be picked up--very positive or totally negative. Discordance would most likely come from the people in the middle. Blood tests will pick up a highly positive HER2/neu. In designing individual therapies, it must be realized that no one factor (protein) is highly discriminant. Possessing a particular protein may give you an edge, i.e., you are 20% more likely to respond to Regimen A if you possess Protein X, but it doesn't guarantee a result. Those who are HER2/neu+ may have slightly worse prognosis, but you get a better result from adriamycin. There may be no way to design an individual therapy based on a single element. Not all risk factors benefit equally from a specific treatment. Very high risk people (those with family history) will always get more benefit from treatment (ex: tamoxifen shouldn't be taken by low risk people because of the possible side effects, but high risk people will certainly benefit from it).

Anything on the horizon to replace tamoxifen, without the side effects (uterine cancer)? Drugs like rolaxifene will require a large study over time. Two aromatase inhibitors, Letrozole and Arimidex, are being looked at for postmenopausal women. They seem to eliminate the increases in uterine cancer and hot flash side effects associated with tamoxifen, but add others like increases in osteoporosis and arthralgias. There are drugs known as biphosphonates which are used for osteoporosis which may lower the chance of bone metastasis, but they too have side effects like gastritis and reflux.

Surveillance with advanced cancer patients is not questioned. Some feel it is adequate to follow the patient clinically (standard of care). Most of the studies would suggest that the best way to treat someone with metastatic cancer is to use some amount of chemotherapy to get the best response with maximum induction, and then to stop the therapy. Studies on chemo forever vs. chemo for fixed time have not shown large differences on mortality. Scans can be helpful to determine when a particular regimen has stopped working and the patient may need to move on to another therapy. This is really highly individual.

What do people think of surveillance for early stage breast cancer after therapy? Polling the Forum, the number of people who would choose serum markers as surveillance vs. those who would not was even (8-8). Dr. Tripathy stated that in his experience, less than half the patients choose to get them. Baseline bone scans in the absence of symptoms are very low yield. If there are questionable areas on the scan, it would be helpful to have a baseline, but this would only be helpful in a very small number of people.

This evening's guest, Dr. Keith Block, runs an integrative cancer program in Evanston, Illinois. It combines conventional cancer therapy with a regimen of dietary supplements, nutrition, physical care, yoga, massage therapy, a stress care program which includes imagery, biofeedback, and meditation, all tailored to the individual. He has been working for the last 10 years on developing a metabolic fingerprint for patients. Each indivdual program is tailored to a variety of complicated diagnostic testing, from DNA, to lipids, to antioxidants, DHEA, etc. Because of this individualization of care, it is hard to do studies. He is trying to work with Dr. Tripathy to come up with a "smaller black box" or generic regimen. This is not ideal for patients, but it is a start. Currently there are 8 groups ongoing. The one that Dr. Tripathy is involved with is a 54 patient group ongoing since 1984 with Stage IV relapse breast cancer. Dr. Block's motivation to begin his program was to find help for his own health problems which included duodenal ulcers and migraines. He went on a macrobiotic diet and cleared up his symptoms within 3 months. His program started with food and exercise as a preventive regimen and introduced supplements later on. His plan is to work with Dr. Tripathy by coming every 4-8 weeks to do a 3-day training program on all the complementary components of his program, while the rest of the conventional care could be easily monitored here.