Should axillary dissection be done? Richard Margolese, M.D., Herbert Black Professor of Surgical Oncology, McGill University, Montreal, Canada

Since the introduction of adjuvant chemotherapy for breast cancer, lymph node status has been a defining indicator for selection of patients and selection of therapy. Less invasive biochemical and cellular markers have been studied and proposed as alternatives but in every multivariate analysis lymph node status continues to be the most reliable indicator for the risk of future metastasis.

With the establishment of benefit of adjuvant chemotherapy for node negative women it appears that the majority of patients with breast cancer will benefit from chemotherapy. The patients who are unlikely to benefit from chemotherapy are those with small tumors with favorable histology discovered on mammography. The yield from axillary dissection runs between 5 and 15% on such tumors. If all such women were subjected to axillary node dissection and 10% had positive nodes and 20% of these had their prognosis altered by chemotherapy then 98% would have the procedure with no chance for benefit.

Sentinel node biopsy is a less invasive procedure with almost the same accuracy but the long term consequences of SNB are still to be determined by ongoing clinical trials. It is likely that refinement of prognostic features in genetic array analysis will ultimately provide the best indicators for who should receive chemotherapy and who is likely to benefit from it.

Axillary Sentinel Lymph Node Biopsy: Hints for the Neophyte Gordon F. Schwartz, M.D., MBA, Professor of Surgery, Jefferson Medical College, Philadelphia, PA

Few procedures have been so rapidly adopted into clinical practice as axillary sentinel lymph node biopsy (SLNB) in patients with breast cancer. Despite criticisms that SLNB has not been validated by clinical trials, its advocates maintain that, as a diagnostic procedure, it does not require the same randomized trials as new treatments would demand. Which of these positions is valid is moot, since SLNB has been adopted throughout the world, and current major concerns relate to perfecting its use.

Without preempting the subsequent discussion on technique, what are some guidelines for those surgeons who have not yet embraced SLNB, or for those who have had some difficulty in identifying “the” sentinel node(s)? The sentinel node is defined as the first node to which lymph drainage and metastasis from breast cancer occurs. Usually an axillary node in level I, it may, however, be behind the pectoralis minor muscle (level II), or even subclavicular (level III). It may be an intramammary node, an internal mammary node, or even a supraclavicular node, but these latter locations are rare.

With suitable training, SLNB is a replacement for axillary dissection as a staging and diagnostic procedure in patients with T-1 and T-2 (usually 3 cm or less), N-0, M-0 cancers. Identification of sentinel nodes in appropriately selected patients is >95%. When the sentinel node, as identified by radiocolloid or blue dye, is contiguous to node(s) that are clinically suspicious because of character or size, they should be removed with the sentinel node(s).

Although the pioneers in SLNB were self-taught, performing SLNB along with traditional axillary dissection to validate their own techniques, current clinical trials (e.g. NSABP, ACOGOG) mandate a training period with documentation of results before embarking on SLNB without concomitant axillary dissection. Hospitals have not addressed credentialing issues in SLNB as they have for other procedures, e.g., laparoscopic surgery or stereotactic breast biopsy.

Until surgeons have documented a detection rate of >90% and a false negative rate of <5%, they should perform concomitant levels I/II axillary dissection. When radiocolloid is used, institutional nuclear medicine teams must be involved, and in all cases, the surgical pathologist must adopt special techniques to handle these specimens.

The precise number of SLNB procedures with concomitant axillary dissection required to validate technique remains contentious; recommendations range from 10 to 100 procedures. Often neglected as these numbers are discussed is the important point that it is the number of patients with positive nodes that will determine the false negative rate. Since only ~30% of women with N-0 axillae will actually have node metastasis, it may take a very large number of patients for any single surgeon to validate his false-negative rate.

Whether radiocolloid, blue dye, or both are used to find sentinel nodes is an individual decision made by the surgical team. Although surgeons new to SLNB are advised to learn and use both radiocolloid and dye to achieve the best results, as experience is gained, one or the other may be used exclusively. (When a single technique is used, it is usually blue dye alone.) Peritumoral injection is the most commonly used site, but a combination of intradermal and peritumoral sites is proving most effective. Retroareolar injection is still considered experimental.

There are specific contraindications to SLNB. These include patients with clinically positive axillae (N-1), since the anatomy of the lymphatics may no longer be intact and lead the dye astray. Thus far, patients who undergo induction chemotherapy for large tumors, even with N-0 axillae, are not considered candidates for SLNB outside a clinical trial. Allergic reactions to both dye and radiocolloid have been reported; allergy to cosmetics containing blue dyes is, therefore, a relative contraindication. Until more data are available, SLNB should not be performed in pregnant women. Prior axillary procedures, such as axillary node biopsy, augmentation mammoplasty using an axillary incision, are relative contraindications. The performance characteristics of SLNB in women who have undergone prior reduction mammoplasty are unknown. A prior biopsy procedure, irrespective of its character, i.e., excisional, incisional, or core, that proved the diagnosis of cancer, is not a contraindication to SLNB and does not affect its success. This question had previously arisen as the practice of SLNB evolved. The role of SLNB in multicentric cancer has not been established. For two discontiguous tumors, two separate injections, one at each tumor site, have been used effectively.

Whether frozen section of lymph nodes is performed is an institutional decision. Most experienced surgeons use frozen sections to determine the presence of metastasis, so that if a more complete axillary dissection is required, it can be performed at the same time as the sentinel node biopsy. Each institution should have its own established pathology protocol. The College of American Pathologists has already promulgated one such protocol. Immunochemical (cytokeratin) staining of axillary nodes is often performed although this information should not be used to influence therapeutic decisions. Although “sub-microscopic” metastasis, i.e. CK-positivity only, has been proved, no data are available to suggest that the clinical outcome of patients with these sub-microscopic metastases is different from those with CK-negative nodes.

The role of SLNB in DCIS is another contentious point. DCIS alone, i.e., without evidence of microinvasion, is not an indication for SLNB. However, patients with DCIS and microinvasion (T-1-mic) should be considered for SLNB.

Patients undergoing mastectomy may be candidates for SLNB, using the same criteria as for patients undergoing breast conservation. The argument for frozen section is more valid in these patients, since completion axillary dissection following mastectomy is more technically difficult than it would be if it were done at the same time as mastectomy.

Technique of Sentinel Node Biopsy: Methods to Reduce False Negative Rates Patrick I. Borgen, M.D., Chief, Breast Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York City, NY

Sentinel lymph node biopsy has established itself as an acceptable standard of care, in the hands of the experienced surgeon, for staging the axilla in breast cancer. The basic hypothesis is that there is a first (sentinel) lymph node that represents the overwhelmingly most likely site of metastasis to the regional node-bearing area. The hypothesis contends that this node should be an accurate representative for the status of the balance of the nodes in the basin. That is, if this node is negative, it should not be necessary to proceed with removal of additional nodes. Sentinel lymph node studies to date, which include over 50 published institutional experiences using a variety of techniques and in a variety of disease situations, have strikingly similar results suggesting that the hypothesis has, in fact, been verified. In the United States, sentinel lymph node biopsy is rapidly gaining popularity and acceptance. Studies of sentinel lymph node biopsy focus on two outcomes. The success of a sentinel lymph node biopsy is defined as finding a hot or blue (or both) lymph node. In cases in which isotope is used, success is further defined by achieving a significant reduction in background isotope counts after removal of the sentinel lymph node. The second parameter, and in many ways the most important parameter, is the accuracy. The accuracy in sentinel lymph node biopsy is defined as the false-negative rate. It is very important to define the false-negative rate calculation as the cases with (missed) positive nodes divided by the number of cases of proven node negatives (in patients who have had an axillary clearance). The largest series published to date report false-negative rates in the 3-8% range. This range of false-negativity has been quite disturbing many. It is important, therefore, to remember that sentinel lymph node biopsy is not being compared against perfection. A full axillary dissection has a known 12-20% false-negative rate when lymph nodes that were signed out as negative are reevaluated more closely. Therefore, the false-negative rate with sentinel lymph node biopsy, in experienced hands, is lower than the false-negative rate associated with axillary node dissection. Strategies to reduce the false-negative rate can be divided into several categories. These are patient selection criteria, tracer blue dye selection/site of injection, intraoperative strategies and postoperative decision-making. Each of these is handled separately below.

Patient Selection Criteria

As the evidence mounts validating the sentinel hypothesis, a broader and broader category of patients appears to be appropriate candidates for sentinel lymph node biopsy. However, several groups of patients remain questionable candidates for this procedure. Patients with multiple invasive primary cancers in at least two separate quadrants may have a higher false-negative rate and current thinking is that they should be excluded from sentinel lymph node biopsy alone. Ongoing research may reveal that the entire breast has a sentinel node, in which case these cases would not be excluded. Secondly, patients with locally advanced cancer may have bulky axillary disease. It is likely that in some cases, a lymph node that is replaced by breast cancer will have reduced or non-existent lymphatic flow, thus increasing the likelihood of false-negativity. Similarly, patients who have had induction chemotherapy prior to surgery may have had the sentinel lymph node partially or totally sterilized by the chemotherapy, while other nodes in the axilla may contain viable metastatic disease. Along the same lines, patients with clinically suspicious ipsilateral axillary lymph nodes are also questionable candidates for a sentinel lymph node biopsy and at the very least, these nodes should be sampled if they are not the sentinel lymph node. Patients with tumors that are quite large and quite high in the axillary of Spence represent a special technical challenge as the distance to be mapped can be quite small. Tracer injection site overlay may mask the true sentinel node; blue dye may permeate the lower lymphatics and nodes.

Choice of blue dye/tracer and site of injection

Sentinel lymph node biopsy has been shown to be accurate using a variety of techniques and a variety of dyes and tracers. Blue dye and radiotracer used in Europe is often different from the blue dye and tracer used in the United States. Colloidal albumen is a radiotracer that has enjoyed great success in Europe but is not approved in the U.S. A number of individuals and centers are quite proficient with blue dye or tracer alone and have published excellent results with their technique. There is mounting evidence that the combination of blue dye and tracer results in the highest success rate and the lowest false-negative rate (McMasters, et al). Moreover, there is mounting evidence that intradermal injection site of tracer is associated with a significantly better sentinel lymph node identification rate than either subdermal or peritumoral injection (Linehan, McMasters). It is reasonable and likely that the combination of blue dye and tracer results in the largest number of nodes being identified. Our group has previously reported that 25% of disease in sentinel lymph nodes is found in the second third or fourth sentinel lymph node removed. The addition of radiotracer may play its biggest role in the identification of multiple sentinel lymph nodes. This procedure can certainly be done with blue dye, but in our hands is facilitated using the hand held gamma probe and the radiolabeled tracer. It is also true that substantial reduction in background counts, post-removal of all sentinel lymph nodes, is quite reassuring in terms of the search for additional lymph nodes. In our hands, this has had a beneficial effect on our published false-negative rates.

Intraoperative Decision Making/Technique

It is more valuable to identify a failed sentinel lymph node procedure than it is to confirm a successful one. When using radiotracer, a failed procedure is defined as one in which there is no identifiable radioactivity in the axilla or when there is diffuse radioactivity covering much of level I and/or level II. Fragmentation of the radiolabeled colloid can lead to a smaller particle size, which can permeate the axilla and make identification of the appropriate node problematic. Either of these scenarios should lead to consideration of a completion axillary dissection. In cases where tracer and blue dye are both used, we attempt to use blue dye to salvage the sentinel node biopsy procedure. If blue lymph nodes are identified, we will accept those as sentinel lymph nodes. The value of visually inspecting and digitally palpating the other nodes in the axilla cannot be overstated. It is very likely that lymph nodes that are replaced with breast cancer cells will have an attenuated lymphatic flow. This means that the nodes we are most interested in finding may not contain blue dye or tracer but may be palpable. We have documented a substantial reduction in our institutional false-negative rate with this simple maneuver. Any node that is felt to be unusually large or have an unusually large consistency is removed whether it is the sentinel lymph node or not. Palpating extra anatomic sites such as Rotter’s space in level II can also reduce false-negatives as sentinel lymph nodes in these locations have been reported. Patients with very large primary tumors, particularly high-grade cancers with extensive lymphovascular invasion, are also candidates for additional node removal (in addition to the sentinel lymph node). These are patients with a high likelihood of axillary disease and it is not unreasonable to sample additional lymph nodes in hopes of lowering the false-negative rate.

The most important technical lesson we have learned is to carefully proceed layer by layer in the axilla carefully controlling small perforator vessels. Blood staining the operating field makes identification of fine blue lymphatics very difficult and can even make the determination of whether a node is blue or not quite difficult. This should be a careful anatomic dissection, not an exercise in blunt dissection and ‘rooting’. Paying attention to hemostasis early in the case also significantly shortens operative time. Without question, the false negative rate can be impacted by following basic surgical principles.

Postoperative Decision Making

Unanticipated findings on the final pathology either on the breast tumor specimen or the axillary nodes may warrant further surgery. If we define a false-negative as leaving any positive lymph nodes in the axilla, then it would follow that any positive sentinel lymph node should lead to further axillary node removal. Twenty to thirty percent of patients with a positive sentinel lymph node are found to have additional positive non-sentinel lymph nodes in the nodal basin. At least six papers published to date have evaluated potential prognostic indicators for predicting non-sentinel lymph node positivity. All studies to date have been unable to reliably do so and all have concluded that a completion axillary dissection should remain the standard of care. A number of important ongoing trials are addressing this issue more closely and in more detail. Considerable debate exists about whether a lymph node with a metastasis seen on IHC only should lead to completion axillary burgery. This remains an unanswered question and again, ongoing trials have been designed to address the issue. Data will be presented concerning our own institutional experience with and approach to micrometastases.

1. Chu KU, Turner RR, Hansen NM, et.al. Do all patients with sentinel node metastasis from breast carcinoma need complete axillary node dissection? Ann Surg 1999; 229:536-541.
2. Weiser MR, Montgomery LL, Tan LK, et.al. Lymphovascular invasion enhances the prediction of non sentinel node metastases in breast cancer patients with positive sentinel nodes. Ann Surg Oncol 2001; 8:145-149.
3. Reynolds C, Mick R, Donohue JH, et.al. Sentinel lymph node biopsy with metastasis: can axillary dissection be avoided in some patients with breast cancer? J Clin Oncol 1999; 17:1720-1726.
4. Chu KU, Turner RR, Hansen NM, et.al. Sentinel node metastasis in patients with breast carcinoma accurately predicts immunohistochemi-cally detectable nonsentinel node metastasis. Ann Surg Oncol 1999; 6:756-761.
5. Turner RR, Chu KU, Qi K, et.al. Pathologic features associated with nonsentinel lymph node metastases in patients with metastatic breast carcinoma in a sentinel lymph node. Cancer 2000; 89:574-581.
6. Wong SL, Edwards MJ, Chao C, et.al. Predicting the status of the nonsentinel axillary nodes: a multicenter study. Arch Surg 2001; 136:563-568.
7. Linehan D, Hill A, Akhurst T, Yeung H, et a, Intradermal radiocolloid and intraparenchymal blue dye injection optimize sentinel node identification in breast cancer patients. Ann Surg Onc 1996(5):450-454.
8. Cody HS, Hill AD, Tran K, Brennan M, Borgen PI, Credentialing for breast lymphatic mapping - how many cases is enough? Ann of Surg 1999; 229(5):723-728
9. Derosis A, Fey J, Yeung H, et al, A trend analysis of the relative value of blue dye and isotope localization in 2000 consecutive cases of sentinel node biopsy for breast cancer. J Am Coll surg 2001; 193:473-478

The UK sentinel node trial (ALMANAC) Robert E. Mansel, MS, Professor of Surgery, University of Wales College of Medicine , Cardiff, UK

Sentinel node biopsy has become a popular technique in many breast units throughout the world, despite the fact that as yet no data has become available from randomised trials. Several large trials are underway in the US, notably the B-32 from the NSABP and American College of Surgeons trials Z10 and Z11. Within Europe the largest current randomised trial is the ALMANAC study, which had randomised over 500 patients by December 2001 in 14 UK centres. The trial is similar in structure to the B-32 with randomisation between conventional axillary surgery and sentinel node biopsy alone with further treatment been given only to the positive sentinel node containing axilla by either further surgery or radiation therapy.

The UK trial is unique in that all surgeons were required to pass a predetermined level of competence in performing sentinel node biopsy, after a period of in-hospital training by the principal investigator. Only those surgeons who achieved a greater than 90 percent localisation rate and a false negative rate of equal or less than 5% over 40 audit cases were permitted to enter patients into the randomised phase. In the preliminary audit phase of over 800 patients, in which all patients underwent formal axillary staging after sentinel node biopsy, it was notable that a small learning curve was recorded.

Primary endpoints of the trial are quality of life measures, health economics and arm morbidity. Local occurrence will also be a secondary endpoint. A new axillary subscale of the Fact B-4 quality of life questionnaire has been developed and validated for this trial. Early analysis of the audit data is is is shows the subscale is sensitive in discriminating between axillary clearance and sampling procedures.

We have found that an increased BMI and age are risk factors for failed localisation, but multifocal tumours or previous surgery had no effect. The randomised trial continues in progress.

Sentinel Node Biopsy - Early Results of a Randomized Trial Umberto Veronesi, M.D., Scientific Director, European Institute of Oncology, Milan, Italy

In most patients presenting today with small breast cancer the axillary lymph nodes are free of cancer cells at histological examination, so that routine axillary dissection appears an overtreatment. One objective of breast cancer research is to provide the surgeon with preoperative information on axillary node status, so that axillary dissection can be avoided if the nodes are negative.

The sentinel node biopsy (SNB) methodology has been developed for this purpose, and numerous studies have shown that the status of the biopsied node is an acceptably accurate predictor of the status of the other axillary nodes (1,2,3). Although these findings are encouraging, the method still requires validation in terms of efficacy and safety.

We therefore decided to carry out a randomized study comparing two series of patients, one treated with routine axillary dissection and the other with the SNB policy. This policy is that if the sentinel node is negative at intraoperative histological examination no further surgery to the axilla is performed.

The study was approved by the Ethics Committee of the European Institute of Oncology. It was a single center, randomized clinical trial with two arms. Patients with primary breast carcinoma less than 2 cm were randomly allocated, after breast conserving surgery, to either sentinel node biopsy and total axillary dissection (AD arm) or sentinel node biopsy followed by axillary dissection only if the sentinel node was metastatic ( SN arm). The SNB procedure was identical in both arms.

As SNB is a diagnostic and staging procedure, the primary end-point (a) of the study was to determine its staging power or percentage of cases with axillary involvement in relation to the percentage found by routine axillary dissection. Additional end-points were (b) patients’ quality of life in the two arms, (c) the number of overt axillary lymph node metastases appearing in SN arm patients with a negative sentinel node and, (d) long-term disease-free and overall survival in both arms.

Patients over 40 and less than 75 years of age, with invasive breast carcinoma less than 2 cm in maximum diameter and no previous history of malignancy, were eligible. Breast cancer was diagnosed from clinical examination, mammography or ultrasonography and, in most cases, a positive fine needle biopsy. Patients with multicentric cancer or who had undergone previous excisional biopsy were not eligible. Patients were randomized in the operating theater after the size and histology of primary carcinoma had been ascertained.

We considered a series of 649 consecutive patients initially. Of this initial series 532 were randomized. Of the 532 randomized patient 16 were not evaluable.

Most patients (410, 79%) were injected with radio-tracer the day before surgery, the remainder (106 patients, 21%) were injected the same day. Five to ten MBq of 99mTc-labeled colloidal human albumin particles (size range 50-200 nm) in 0.2 mL of saline were injected close to the tumor (4). Injection was subdermal if the tumor was superficial and peritumoral if it was deep. Anterior and anterior-oblique lymphoscintigraphic projections of the breast and axilla were subsequently obtained to reveal the dynamics of lymphatic flow (in particular whether more than one lymph node took up tracer) and to determine the exact position of the sentinel node. The skin projection of the sentinel node(s) was drawn to serve as landmark during biopsy. In eight (1.2%) of the 649 patients of the original series there was no uptake of the radiotracer at scintigraphy and these were not considered eligible for the trial.

Of the 516 evaluable patients, 292 had one sentinel node, 152 had two sentinel nodes, 47 had three sentinel nodes and in 25 more than three were identified. All axillary nodes taking-up radiotracer were removed and classified as sentinel nodes and all were evaluated histologically. The distribution of number of sentinel nodes found did not differ in the two arms. A total of 429 sentinel nodes were removed and examined from AD arm and 424 from SN arm. The total of 6200 axillary lymph nodes were removed (including sentinel nodes) from the 257 patients of AD arm (average 24 per axilla) and a total of 2249 were removed from the 93 patients of SN arm who had positive sentinel nodes (average 24 per axilla).

The first end-point of the study was to assess the ability of the sentinel node policy to screen cases with a positive axilla, in comparison with the detection rate in patients who received total axillary dissection. Among the 257 patients of AD arm, a sentinel node was positive in 83 (32.3%) cases and negative in 174 (67.7%); among the 259 patients of SN arm, a sentinel node was positive in 93 (35.9%) and negative in 166 (64.1%).

Among the 257 patients of AD arm all of whom received total axillary dissection, there were eight cases with metastatic axillary nodes but a negative sentinel node; in two of these cases there were micrometastases in one axillary node only. The overall accuracy of the SN technique in AD arm was therefore 96.9%, the positive predictive value was 100%, the negative predictive value was 95.4%, the sensitivity 91.2%, and the specificity 100%.

Out of 176 patients with metastatic sentinel nodes in 60 micrometastases only were found, defined as one focus of metastatic cells less than 2 mm in diameter. In 41 of these 60 cases the focus was smaller than 1 mm. In AD arm, 29 patients had a micrometastatic sentinel node, in 24 of which all the other axillary nodes were negative, while in five, one other axillary node was positive. In SN arm, 31 patients had a micrometastatic sentinel node, in 26 of which all other axillary nodes were negative, while in 5 there was one positive node. The median follow up for each patient is 26 months and the total person years on study is 552 in the AD arm and 566 in the SN Arm. There are 9 events associated with Breast Cancer to date, 8 in the AD arm and 1 in the SN arm. In the AD arm there are 2 controlateral breast cancers, 2 bone metastases, 2 lung metastases and 2 patients with multiple metastases. One patient in the SN arm had a bone metastases. Five patients, all in the AD arm, have died. Two due to metastases from breast cancer, one for carcinoma of the pancreas, one for endometrial carcinoma and one for intercurrent disease.

References

1. Giuliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg 1994; 220:391-401.
2. Krag D, Weaver D, Ashikana T, et al. The sentinel node in breast cancer - a multicenter validation study. New Engl J Med 1998; 339(14), 941-946.
3. Veronesi U, Paganelli G, Galimberti V, et al. Sentinel-node biopsy to avoid axillary dissection in breast cancer with clinically negative lymph-nodes. Lancet 1997; 349:1864-1867.
4. De Cicco C, Paganelli G, Cremonesi M, et al. Lymphoscintigraphy and radioguided biopsy of the sentinel axillary node in breast cancer. Eur J Nucl Med 1998; 39 (12): 2080-2084.

Identification of Metastatic Axillary and Internal Mammary Nodes with MRI and Contrast Media Steven E. Harms, MD, FACR, Professor of Radiology, University of Arkansas for Medical Sciences, Little Rock, AR, and Medical Director, Aurora Imaging Technology, North Andover, MA

Introduction:

The presence of axillary node metastasis is the most reliable predictor of of outcome in women with breast cancer. Therefore, axillary node dissections have become a routine part of breast cancer staging. With early cancer detection, the occurrence of positive lymph nodes is declining. Recent studies indicate that only about 10% of patients with invasive cancers of 1 cm or smaller in size will have metastatic nodes.1,2,3 This represents an improvement from the late 1960’s when about one third of small cancers had nodal metastases. Although node dissections have prognostic benefit, there is no therapeutic benefit for most patients. The relatively high morbidity of axillary dissection relative to its benefit have prompted efforts to avoid this surgery in women with other signs favoring a good prognosis.2,3 Sentinel node surgery is becoming a popular alternative to traditional axillary node dissections. A non-invasive method of accurately predicting nodal metastasis could further reduce costs and morbidity while potentially improving nodal sampling beyond the sentinel nodes. Ferumoxtran-10 (F-10) is a biodegradable, ultrasmall, superparamagnetic iron oxide covered with a low molecular weight dextran. When given intravenously, it shows active uptake by the reticuloendothelial system (liver, spleen, lymph nodes, and bone marrow). The particle size is such that there is preferential uptake by lymph nodes. After twenty-four hours, normal nodes contain a high concentration of F-10. This results in marked T2*-shortening (magnetic susceptibility effects) and a significant loss of signal within the node on a T2* weighted MRI image. Tumor-associated macrophages within and around primary and metastatic foci have less avid uptake and demonstrate low concentration T1 shortening effects. Metastatic foci within nodes increased intensity on T1-weighted images.4-7

Purpose:

The objective of this study is to determine the potential for ferumoxtran-10-enhanced MRI to localize axillary metastases in women with invasive breast carcinoma. Results of ferumoxtran-10 MRI will be compared with more conventional gadolinium-enhanced MRI for differentiating metastatic from nonmetastatic nodes, using histologic confirmation.

Methods:

Patient selection:
All women in the study were above 18 years of age and were recently diagnosed with a biopsy proven breast carcinoma less than or equal to 3 cm in size. All patients underwent breast conservation treatment and subsequent axillary lymph node sampling by conventional node dissection or by sentinel node biopsy. At this time a total of 27 patients from 7 institutions have been studied.

Imaging:
All studies in the multi-institutional trial employed routine 2D T1-weighted and T2 weighted spin-echo and T2* gradient echo sequences. In addition to the routine protocol, images at our institution were obtained on a 1.5 Tesla MRI imager (General Electric, Milwaukee) using the 3D RODEO (Rotating Delivery of Excitation Of-resonance) pulse sequence. The 3D scans provided improved spatial and contrast resolution to demonstrate the potential benefit of future image enhancements on overall contrast agent performance.

Drug administration:
Ferumoxtran-10 (2.6 mg Fe/kg) was administered intravenously after diluting the agent with 50 ml of normal saline (NaCl, 0.9%) at a rate of 4 ml/min. No adverse affects were reported.

Procedure:
Pre- and post-contrast gadolinium enhanced images were obtained in 5 cases. In all cases pre-, immediate, and 24-36 hour delayed, post-ferumoxtran-10 MRI images were obtained and analyzed. All lymph nodes were categorized according to size, anatomic location, and diagnosis (normal, metastatic, possibly metastatic) An axillary lymph node dissection bwas performed within 48 hours of administration of F-10.

Results:
Surgical specimens, obtained shortly after ferumoxtran-10 administration and examined with specific iron stains, demonstrated high concentrations of iron within normal nodes consistent with uptake of the agent . This high concentration of iron correlates with the intense T2* shortening (hypointensity) due to magnetic susceptibility effects on T2* weighted MRI scans. Metastatic and primary tumors showed low concentrations of iron within tumor associated macrophages(Figure 1). Low concentrations of iron result in hyperintensity on T1 weighted MRI scans.

The summary data from the multicenter trial of overread images combining the results of 2 blinded readers are available in table 1. Using size criteria alone, the widely used standard in CT and MRI currently, the specificity is high (96%) but the sensitivity is low (27%). The post-ferumoxtran-10 images provided nearly equivalent specificity (91%) but significantly improved sensitivity (68%) and accuracy (78%). This improvement is attributed to the ability to detect normal sized metastatic nodes. These results demonstrate the potential for ferumox-tran- 10 lymph node specific contrast to depict nodes that would otherwise be missed on conventional MRI scans.

Table 1

Despite the ability of ferumoxtran-10 to deliver sufficient contrast between normal nodal tissue and metastatic disease, a major limitation of current MRI is the ability to generate sufficient resolution to depict small lymph node metastases. The multicenter trial employed standard MRI pulse sequences with 5 mm thick sections . At UAMS, we also used the RODEO pulse sequence, which allowed the generation of 500 micron section thicknesses. In addition, the combined T1 and T2* weighting of RODEO further enhances the contrast mechanisms of ferumoxtran-10 (Figure 2).

Figure 1: Iron stain within tumor-associated macrophages

The Prussian blue stain (high power insert) demonstrates blue staining iron within macrophages surrounding and within an infiltrating breast carcinoma. This effect correlated with the hyperintense signal on T1 weighted MRI images from low concentration uptake of ferumoxtran-10.

Figure 3: Axial reformatted pre- (A) and post-gadolinium (B) RODEO images of the axilla, of the same patient as Figure 3, demonstrate multiple iso-intense to hyper-intense lymph nodes of various sizes but of similar morphology. Note the improvement in resolution compared to standard imaging techniques.

Immediately (C) and 24 hours (D) after the administration of F-10, there is progressive iron uptake within normal axillary lymph nodes resulting in low signal intensity (small yellow arrows). These smaller nodes are more clearly identified on the RODEO sequence. A malignant node (large red arrow) demonstrates no uptake and remains high in signal intensity. The largest node (blue arrow) was only partially involved with metastatic disease and has components of high and low signal.

Conclusion:
This study demonstrates the potential for ferumoxtran-10 enhanced MRI in the evaluation of axillary nodal metastases from breast cancer in normal sized nodes. When the contrast and spatial resolution is optimized with RODEO MRI, significantly improved nodal sampling can be achieved with section thicknesses approaching 500 microns. This capability could rival the sampling ability of conventional node dissections in a non-invasive examination.

Non-invasive node evaluations will be most helpful in patients with small infiltrating carcinomas and DCIS where the probability of node involvement is small. A non-invasive test that could exclude nodal metastases would lower morbidity and reduce health care costs.

If positive, newly developed MRI stereotaxic methods can be used to selectively sample nodes without a node dissection.

The use of ferumoxtran-10 enhanced RODEO MRI is expected to produce significant improvements in the management of breast cancer in the future.

Reference:

1. Reger V, Beito G, Jolly PC. Factors affecting the incidence of lymph node metastases in small cancers of the breast. Am J Surg 1989; 157:501-502.
2. Cady B, Stone MD, Wayne J. New therapeutic possibilities in primary invasive breast cancer. Ann Surg 1993; 338-349.
3. Silverstein MJ, Gierson ED, Waisman JR, et al. Axillary lymph node dissection for T1a breast carcinoma: Is it needed? Cancer 1994; 73:664-667.
4. Vassallo P, Matei C, Heston W, et al. AMI-227—enhanced MR Lymphography: Usefulness for Differentiating Reactive from Tumor-bearing Lymph Nodes. Radiology 1994; 193: 501-506.
5. Anzai Y, McLachlan S, Morris M, et al. Dextran-coated Superparamagnetic Iron Oxide: The First Human Use of a New MR Contrast Agent for Assessing Lymph Nodes in the Head and Neck. Amer. J. Neuroradiology 1994; 15: 87-94.
6. McLachlan S, Morris M, Lucas M, et al. Phase I Clinical Evaluation of a New Iron Oxide MR Contrast Agent. Magnetic Resonance Imaging 1994; 4: 301-307.
7. Anzai Y, Blackwell K, Hirschowitz S, et.al. initial clinical experience with dextran-coated superparamagnetic iron oxide for detection of lymph node metastases in patients with head and neck cancer. radiology 1994; 192: 709-715.

Journal of Clinical Oncology Special Issue Review Daniel F. Hayes, M.D., Director, Breast Oncology Program, University of Michigan Comprehensive Cancer Center, MI

I was asked by George Canellos to select the “Classic” papers regarding breast cancer published in the Journal of Clinical Oncology from 1990-2000 1. This was a daunting task for several reasons. 1) Over 500 breast-related articles were published in the journal during that time. 2) The definition of classic is very much operator dependent. I chose the following definitions: a) publication of the paper changed treatment (as far as I can tell); b) the concept presented a fundamental observation regarding the natural history of the disease; and/or c) the report represented the first observation (to my knowledge) of what would turn out to be an important area of research.

I divided the issue into the obvious categories (not unlike this symposium): Adjuvant Therapy, Bisphosphonates, Metastatic Chemotherapy, Metastatic Hormonal Therapy, Metastatic anti-HER2 therapy, Natural History, Primary Therapy, Prognostic and Predictive Factors, and Quality of Life. I am particularly pleased that at least three of these (bisphosphonates, anti-HER2 therapy, and QOL) would not have been considered in 1990. I excluded the reports that represented updates of previously published studies, Reviews, and ASCO Guidelinies publications (to my chagrin, but space constraints dictated this decision).

I included papers that others did not feel were classic, and I excluded some that are (to quote Yogi Berra, “Can’t win for losin’ “). Mortality from breast cancer is dropping in the Western World, and at the same time survivors are living better lives 2. This special issue of the Journal reflects some of the reasons for these advances, and we all owe thanks to the many investigators and patients who have worked so hard to produce these results.

1. Hayes DF: Classic Papers and Current Comments: Highlights of Breast Cancer Research, in Canellos GP (ed): Classic Papers and Current Comments from the Journal of Clinical Oncology. Baltimore, Lippincott Williams & Williams, 2001
2. Peto R, Boreham J, Clarke M, et al: UK and USA breast cancer deaths down 25% in year 2000 at ages 20-69 years. Lancet 355:1822., 2000

Current and Proposed Trials of Local Therapy Dr. Tim Whelan, B.M., B.Ch., M.Sc., Associate Professor, Dept. of Medicine and Director, Supportive Cancer Care Research Unit, McMaster University, Ontario, Canada

Following the seminal trials in the 1980’s demonstrating that breast conserving surgery is equivalent to mastectomy for women with early breast cancer, there have been a number of trials evaluating the role of breast irradiation following lumpectomy. These studies have uniformly demonstrated that radiation therapy reduces the risk of local recurrence. Recently there have been a number of trials evaluating the need for radiation in women at low risk of local recurrence following lumpectomy. NSABP B21 studied the need for radiation therapy in women with node negative breast cancer with primary tumours less than 1 cm treated with Tamoxifen. The study demonstrated that despite the use of Tamoxifen radiation resulted in a significant decrease in local failure. Two other trials have evaluated the role of radiation therapy in older women treated with Tamoxifen. The CALGB trial studied women 70 years or older with clinical Stage I disease and the Canadian trial studied women 50 years of age and older. The results of these studies are conflicting but follow up is still short.

Other recent trials have evaluated how best to deliver radiation therapy following lumpectomy including, accelerated treatment with increased fractionation, and the role of additional boost treatment in women with clear resection margins following lumpectomy.

Another area of recent interest has been the role of locoregional radiation therapy. A number of randomised trials have demonstrated that locoregional radiation following mastectomy in patients treated with adjuvant systemic therapy not only reduces the risk of local recurrence, but provides additional survival benefits in selected patients. These trials were performed largely in women treated with limited axillary dissection, who received non-anthracycline based chemotherapy or relatively short duration hormonal therapy. Recently, two randomized trials have been opened which should clarify the role of locoregional radiation in the modern management of breast cancer. The SWOG S9927 trial will evaluate locoregional radiation post-mastectomy in patients with 1 to 3 positive axillary nodes following adjuvant chemotherapy. The NCI(C) MA20 will evaluate regional radiation in addition to breast irradiation in patients with high risk node negative and node positive breast cancer.

In addition to these large phase III trials there are a number of important Phase I & II trials evaluating different ways of delivering adjuvant radiation therapy including brachytherapy, intraoperative radiation and the use of intensity modulated radiation therapy. Successful approaches will likely be taken into phase III testing in the next few years.

Current and proposed trials adjuvant therapy for breast cancer Debu Tripathy, M.D., UCSF Carol Franc Buck Breast Care Center, University of California, San Francisco, CA

Stepwise improvements in the care and outcomes of patients with breast cancer require a well thought out and deliberate set of clinical trials. These trials must be based on biological underpinnings as well as the experience from prior trials. Prioritization of trials depend on the degree of enthusiasm for the new drug being tested, such as the preliminary response rates seen in Phase II trials. However, issues such as feasibility (number of patients needed and safety of the drug/ease of administration) must be considered for the trial to finish quickly and for the results to be valid and useful. In the area of early stage breast cancer, several key issues are currently under investigation:

* Addition of taxanes
* Optimal duration of chemotherapy (number of cycles)
* Sequential versus alternating/combination chemotherapy
* Dose-dense chemotherapy
* Neoadjuvant chemotherapy (types of agents and sequence)
* High dose chemotherapy with stem cell rescue for high risk disease
* Role of aromatase inhibitors in early stage breast cancer
* Role of oophorectomy in premenopausal women in addition to chemotherapy
* Postmastectomy radiation * Adjuvant bisphosphonates
* Choice of chemotherapy based on HER2/neu status and other markers
* Adjuvant anti-HER2/neu antibody (Herceptin)
* Vaccines * Prostaglandin inhibitors
* Diet/lifestyle modifications

Specific strategies and trials are discussed below

Substituting taxanes for alkylators is being studied in order to eliminate some of the alkylator toxicities such as ovarian failure and leukemia risk. Example: Intergroup study comparing doxorubicin plus paclitaxel x 4 to standard cyclosphospamide plus doxorubicin. The study is completed and results are pending

Adding a taxane remains of uncertain value since its addition to doxorubicin plus cyclophosphamide resulted in improved disease-free survival, yet a retrospective analysis suggested that this benefit was confined to ER-negative cases. A confirmatory trial by the NSABP will reported soon. An ongoing Canadian trials is also comparing the addition of taxanes to an epirubicin-based regimen: CEF x 6 vs AC x 4 T x 4 vs EC x 6 T x 4 (C=cyclophosphamide; E=epirubicin; F=fluorouracil; A=doxorubicin; T=docetaxel;)

Prior trials have not clearly identified the optimal duration of anthracycline or taxane-based therapies. Hence a trial is being planned using a 2 x 2 designe that will compare AC x 4 vs AC x 6 vs weekly paclitaxel x 12 vs weekly paclitaxel x 18.

Using drugs in combination compared to using them in sequence. Mathematical modeling has shown potential advantages to using drugs in a dose-dense sequential fashion. Example: NSABP B-30 - AC x 4 D x 4 vs AD x vs ADC (A=doxorubicin; D=docetaxel; C=cyclophosphamide)

New combinations using capecitabine are also being planned. The NSABP will compare AC D vs AC D + Capecitabine (A=doxoru-bicin; C=cyclophosphamide; D=docetaxel). A trial for older women will compare CMF or AC to capecitabine

Bisphosphonates will also be tested following standard chemotherapy and hormonal therapy. Observation will be compared to bisphosphonate (clodronate in the NSABP study and zolendronate in the SWOG study).

Given the survival benefit of adding Herceptin to chemotherapy in the advanced setting, several trials are ongoing to test Herceptin in the adjuvant (node+) setting.

Trial	Schema
NSABP B-31	AC x 4 Paclitaxel x 4 vs. AC x 4 Paclitaxel x 4 plus Herceptin weekly x 1 year
NCCTG 9831/ Intergroup	AC x 4Paclitaxel weekly x 12 vs. AC x 4 Paclitaxel weekly x 12 plus Herceptin weekly x 1 year AC x 4Paclitaxel weekly x 12 Herceptin weekly x 1 year
BCIRG 006	AC x 4 Docetaxel x 4 vs. AC x 4 Docetaxel x 4 plus Herceptin weekly x 1 year vs. Docetaxel + Carbo/cisplatin x 6 plus Herceptin weekly x 1 year
CALGB 9804	Neoadjuvant AC +/- dexrazoxane Weekly paclitaxel x 12 +/- Herceptin Surgery/radiation therapy+/- Herceptin weekly x 1 year (2x2x2 design)
HERA	Any chemo or XRTObservation vs Herceptin q 3 wk x 12 mo vs Herceptin q 3 wk x 24 mo

AC = doxorubicin 60 mg/m² plus cyclophosphamide 600 mg/m²
Paclitaxel 175 mg/m2 q 3 wk unless specified weekly, then 90 mg/m²
Docetaxel 100 mg/m² q 3 wk
Carbo/cisplatin - Carboplatin AUC 6 or cisplatin 75 mg/m² q 3 wk

A Visionary’s View of Breast Cancer Management in the Next Decade Umberto Veronesi, MD, Scientific Director, Istituto Europeo di Oncologia, Milano, Italy

A revolution in breast cancer treatment was ushered in the 1970s with the conservative approach. After 25 years survival, curves of the 701 cases enrolled in the Milan trial I show that quadrantectomy gives identical results to the Halsted mastectomy.

A more recent development is the extension of quadrantectomy to large tumors after initial (neo-adjuvant) chemotherapy to reduce the size of the mass and permit a safely conservative approach.

One of the most exciting conservative approaches in breast cancer surgery is sentinel node biopsy. In the first series of 376 patients we identified the first node draining the tumor area with the aid of the radiotracer (99Tc) and a gamma detecting used during surgery. All patient underwent complete axillary dissection. The study showed an overall accuracy of 96.8%, a sensitivity of 93.3% and a specificity of 100%. We have now more than two thousand cases of sentinel node biopsy including those in the randomized trial with the same accuracy rate, with a mean follow up of 30 months and only 1 case in which appeared axillary metastases.

By further exploiting the radiotracer and probe we developed a new technique for precisely localizing non palpable lesions. Called ROLL (radioguided occult lesion localization) we have used it on more than 1000 patients to remove the lesion with the maximum precision and accuracy as determine by intaoperative X-ray of the specimen.

Another important trial a the EIO recruited 436 patients with a clinically negative axilla and a breast cancers of less than 1.2 cm who did not received axillary dissection, to assess the protection afforded by prophylactic axillary irradiation. It was found that axillary dissection may be avoid without significant risks in such patients, as the 5-year survival was 98%.

We are now experimenting the intraoperative radiotherapy for limiting the RT to the tumor bed giving the maximum dose of radiation at the same time of operation without necessity of post operative therapy.

Also adjuvant treatments are reaching important goals by new and more selective drugs thanks to the improvement in molecular targeting with less side effects.

The new discipline of pharmacoprevention is emerging in breast cancer. Tamoxifen is significantly protective in high risk women but not in normal risk ones. HRT users may greatly benefit by tamoxifen and a new trial is in progress at the European Institute of Oncology. Finally, the use of vitamin A derivatives (fenretinide) can reduce the risk of developing breast cancer in premenopausal women.

The Use of the Decision Board for Breast Cancer Treatment Dr. Tim Whelan, B.M., B.Ch., M.Sc., Associate Professor, Dept. of Medicine and Director, Supportive Cancer Care Research Unit, McMaster University, Ontario, Canada

In recent years there have been major advances in the treatment of early stage breast cancer. The decisions a patient must make about the treatment are often difficult and complex. In the past physicians tended to make decisions for patients with little patient input. Many patients now often express a desire for more information about their disease and the need to be more actively involved in the decisions about their treatment. Researches have responded by investigating better ways of transferring information and involving patients in decision making. Decision aids have been defined as, “interventions designed to help people make specific and deliberative choices among options by providing information on options and outcomes relevant to a patient’s health”. The Decision Board consists of a visual aid and written material. The instrument is administered by a clinician during the patient consultation. Information from clinical trials on a patient’s choices, outcomes, probabilities of those outcomes and associated quality of life is presented in an interactive step-by-step fashion.

The Decision Board was initially developed to help women with node negative breast cancer to decide whether to receive adjuvant chemotherapy. It was found to be both acceptable and helpful in decision making. A cohort study of the Decision Board by patients deciding on radiation therapy following lumpectomy found that the instrument improved both patient knowledge and facilitated shared decision making. In another study it was found that the use of the Decision Board for women deciding between lumpectomy and mastectomy was well accepted by both patients and surgeons in the community. Recently a randomised trial of the Decision Board for chemotherapy in women with node negative breast cancer demonstrated that the instrument improved both patient knowledge and satisfaction with decision making. Currently a number of studies are on going including a randomised trial to evaluate the Decision Board process for women deciding on lumpectomy and mastectomy and the development and testing of computer versions of the Decision Board to facilitate the presentation of more complex decisions to women with breast cancer. A more versatile decision aid will respond to the diverse needs of patients and physicians and may facilitate its wider use throughout the medical community.

The Computer Program, Adjuvant!, A Tool Assist In Adjuvant Therapy Decision Making for Early Breast Cancer. Peter M. Ravdin, MD., Ph.D. , Associate Professor of Medical Oncology, University of Texas, Health Science Center, San Antonio, TX

The decision of whether a breast cancer should receive adjuvant therapy involves weighing the risks and benefits of such therapy. The most common presentation of breast cancer today is as Stage 1 disease with a low risk of relapse and breast cancer related death. For many of these patients the benefit of adjuvant therapy is modest, and its use somewhat controversial.

To address this problem various panels have drawn up treatment guidelines. For example over the last year guidelines have been published by an NCI Consensus Conference, the NCCN, and from St. Gallen’s (an International Breast Cancer panel). A graphical summary of these guidelines is shown below. What is interesting about these recommendations is that there are a number of scenarios where they reach quite different recommendations. Take the example of patients with N0T1a estrogen receptor negative tumors. Treatment recommendations would range from mandated adjuvant chemotherapy (St. Gallen’s), to no therapy (NCCN).

What is missing from these Guidelines is some quantitative sense of what is gained or given up by getting selecting a given option. A evidence based tool, a computer program, Adjuvant!, was created to address this issue.

The steps to use the program, Adjuvant! are:

1. Entering the usual patient and tumor related data. With the program then projecting prognosis and competing mortality.
2. Selecting the adjuvant therapy that is to be considered. With Adjuvant! providing the efficacy estimates (proportional risk reductions), and projecting probably net benefit.
3. Printing out easy to read outcome sheets that might be used in patient consultations.
4. Printing out sheets with treatment schema and toxicity information.

Below is shown the Main Screen from Adjuvant!

In the extensive help files there are discussions of topics of special interest in adjuvant therapy, discussion of treatment guidelines, and links to help in the consideration of clinical trials

Although guidelines help standardize therapy, tools such as Adjuvant! have the power to better inform patients and make them true partners in weighing treatment options.

1. Adjuvant Therapy for Breast Cancer. NIH Consensus Statement Online 2000 November 1-3; 17(4): 1-23.
2. Carlson RW. Anderson BO. Bensinger W. Cox CE. Davidson NE. Edge SB. Farrar WB. Goldstein LJ. Gradishar WJ. Lichter AS. McCormick B. Nabell LM. Reed EC. Silver SM. Smith ML. Somlo G. Theriault R. Ward JH. Winer EP. Wolff A. National Comprehensive Cancer Network. NCCN Practice Guidelines for Breast Cancer. Oncology 14(11A):33-49, 2000 Nov.
3. Goldhirsch A, Glick JH, Gelber RD, Coates AS, and Senn H-J. Meeting Highlights: International Consensus Panel on the Treatment of Primary Breast Cancer. Journal of Clinical Oncology 19(18) 3817-3827, 2001.
4. Ravdin PM. Siminoff L A. Davis GJ. Mercer MB. Hewlett J. Gerson N. Parker HL. Computer program to assist in making decisions about adjuvant therapy for women with early breast cancer. Journal of Clinical Oncology. 19(4):980-91, 2001 Feb 15.

Communicating therapeutic goals: Enlisting the patient as partner Kathy D. Miller, M.D., Associate Professor of Medicine, Indiana University, Indianapolis, IN

In newly diagnosed patients, the risk of recurrence and death differs among patient subgroups, therefore the absolute reduction in relapse and death rates will be greatest in high-risk patients and smallest in low-risk patients. Because chemotherapy has toxicity, the art of medicine involves balancing risk and benefit for individual patients. With few exceptions, metastatic breast cancer is not curable making the goal of ALL therapy palliative (as opposed to potentially curative), requiring even greater patient inout into treatment decisions. Unfortunately clinical trials, by their nature, attempt to simplify through the rigorous application of study entry criteria. Consequently, virtually every trial is necessarily unrepresentative of the general population of breast cancer patients. In real life patients differ considerably from each other, and these differences have real consequences for the treating physician. These complexities include exposure to prior therapy, presence of co-morbid conditions, psychosocial circumstances, as well as emotional and spiritual needs. In real life clinical therapy frequently requires a series of negotiations between patient and physician; the patient’s needs and opinions definitely matter. The wise physician is aware of an old truth: in the exam room there are two experts, one an expert on the disease and the other an expert on the patient.

Research in complementary medicine Debu Tripathy, M.D., UCSF Carol Franc Buck Breast Care Center, University of California, San Francisco, CA

Complementary and alternative Medicine (CAM) has become increasing popular, especially in patients with cancer. While some CAM modalities have been practiced for centuries, very few studies have been done to rigorously assess the safety and effectiveness of these approaches. CAM includes many modalities - below is a partial list:

* Mind body medicine (meditation, yoga)
* Dietary approaches (macrobiotic diet, micronutrients)
* Herbal medicine
* Acupuncture
* Naturopathic medicine
* Homeopathy
* Reflexology
* Aromatherapy

Several reasons exist for the popularity of CAM in cancer:

1. Convention cancer treatment is only partially or minimally effective in many settings. Most common epithelial (solid) tumors in adults are not curable at the advanced stages. While many therapies may induce transient responses, often time resistance will develop. Furthermore, side effects of therapy can require discontinuation, or in some cases, patient refusal of therapy altogether. Even newer biologically targeted therapies tend to be effective only in a minority of treated patients, and resistance to these novel agents eventually develops as it does with conventional cancer drugs. Given the clear limitations of standard cancer treatment, many patients with cancer will opt to add CAM to their plan with the attitude that it cannot hurt and might help.

2. CAM practitioners are often in a position to spend more time with their patients and perhaps more time listening. This is in contrast to conventional medicine where recent trends in managed care and have forced physicians to spend less time with patients. The rewards that patients receive in attentiveness make a difference in their perception of trust and well being. This further popularizes CAM and may even produce a favorable psychological and even health benefit from using CAM. On the other hand, CAM usage may be a sign of distrust of conventional medicine as well as a higher degree of psychological stress and anxiety on the part of the patient.

3. Side effects of cancer and cancer treatment can have a significant impact on quality of life. Many CAM modalities are geared to treating symptoms and not necessarily the underlying disease. Conventional palliative medicines such as narcotics and anti-emetics themselves have undesirable side effects. Therefore, just as with therapy, many individuals view CAM treatments as being more natural and holistic.

4. While most patients use CAM therapies along with conventional treatment, many do not share this information with their physician. Similarly, most physicians do not ask their patients about CAM usage and are reluctant to discuss their views or opinions in this area. Several reasons exist for the lack of the integration of CAM into mainstream medicine. Most physicians lack training and exposure to CAM modalities. Also, there has been a growing emphasis on evidence-based medicine and very few CAM modalities have been subjected to rigorous laboratory and clinical study. Hence, there is very little representation of CAM in the medical literature, textbooks and scientific meetings.

All of these reasons can be considered valid from the standpoint of patients’ attitudes and the frustrations of the limitations in what can be done for some cancer patients. However, in many cases, there is also potential merit in CAM and there is a growing recognition that formal research of CAM needs to be performed in different cancer settings. In 1993, the U.S. Congress mandated the opening of the Office of Alternative Medicine in the National Institutes of Health. Five CAM research centers were then funded by this Office. In 1998, the Office was converted into the larger National Center for Complementary and Alternative Medicine (NCCAM) and yearly funding is now in excess of 60 million dollars a year. A special Office of CAM within the National Cancer Center has also been established. These organizations are actively releasing requests for applications in specific areas of CAM and developing mechanisms and review teams to fund meritorious work. There is a very slowly growing expertise of multidisciplinary investigators cross-trained in clinical research and CAM practice. Additionally, the National Cancer Institute has an information service (PDQ) regarding clinical trials and disease-specific information. Summaries about the background and scientific evidence for different types of CAM is being generated and reviewed to made readily available to the public. Surveys of both patients and physicians have demonstrated an interest in CAM research and in particular, broad support for clinical trials in this area.

In order to assess these therapies and begin to integrate them into conventional cancer care, stepwise clinical trials will need to be done in a similar fashion to the process that is needed for currently available therapies and new experimental treatments. However, several investigative challenges are posed by the nature of some CAM approaches: These include:

* There are few preclinical (laboratory) models or well-documented clinical series upon which to base or prioritize clinical trials.
* CAM approaches are often individualized, so applying the same experimental variable to all subjects may not be in keeping with the CAM modality. For example, herbal regimens are often complex and highly individualized based not only on the disease but many other characteristics of the patient.
* CAM therapies are usually geared towards many outcomes such as tumor response, quality of life and emotional well being. From a statistical standpoint, it is very difficult to demonstrate efficacy in all of these endpoints.
* There is a lack of funding, cross-trained investigators and qualified reviewers for research grants and publications

CAM research needs to follow the same basic principles of any biomedical investigation. There should be a sufficient background and scientific rationale for pursuing the treatment in question. The design of the study must contain the same experimental and control groups and must use standard statistical data analysis tools for definitive conclusions to be made.

Laboratory methodology in CAM may not always apply given the nature of some CAM approaches. For example, mind-body medicine is difficult to model and study in the laboratory, although in human subjects, laboratory studies can be done, for example assessing immune function or cytokine release after prayer or meditation. In other cases, laboratory studies are necessary prior to clinical investigation. Herbal therapies can be tested against cancer cell lines if an anti-proliferative effect is being sought. It is well recognized that these models are not very accurate in predicting clinical success and may not be able assess host-tumor interactions, but nevertheless, they can serve a useful role in prioritizing and justifying clinical trials.

Our group has begun investigative work in CAM, using herbal therapy as a model since there is a basis for the use of botanical products in cancer. First, these are known to possess biological activity and in fact, many modern drugs are derived from herbal components. Second, there is some degree of consistency in the types of herbs used for specific indications and there is also licensure and training of herbalists in many states. Finally, herbal therapy represents one of the more commonly used modalities for cancer. Below are several studies in progress or completed by our group. In all cases, Investigation New Drug Licenses (INDs) have been obtained from the FDA for these IRB-approved studies.

1. Combination Herbal Therapy to Alleviate Side Effects of Doxorubicin and Cyclophosphamide Chemotherapy for Early Stage Breast Cancer. A 21-herb combination that has been used by many practitioners was formulated and standardized. This pilot study required a placebo arm in order to differentiate potential herbal toxicities from chemotherapy side effects and also to obtain an estimate of the effect size in reducing side effects. Validated tools to assess symptoms and quality of life are being used in this 60 patient study
2. Pilot Phase I Study of Tibetan Herbal Therapy for Advanced Breast Cancer. This study was designed to preserve the individualization of care by an expert Tibetan physician. Patients with minimally symptomatic measurable advanced breast cancer were treated with Tibetan herbal formula as their only treatment. The actual formula was determined by the Tibetan physician based in his history and examination. Standard safety assessments and tumor measurements were performed and patients were treated until tumor progression. No Grade III/IV toxicities were seen related to therapy and one of 9 evaluable patients had a 4-month partial response.
3. A Phase I/II study of Herba Scutellaria Barbatae (HSB) for Advanced Breast Cancer. This study is similar to #2, but all patients receive an extract of HSB, which emerged from our laboratory program screening herbal extracts for antiproliferative activity on breast cancer cell lines

Breast Cancer Online: the internet as a source of information for professionals J. F. R. Robertson, Professor of Surgery, City Hospital, Nottingham, UK

There are several problems facing a researcher, clinician or patient, in using the Internet to source information on breast cancer. Firstly one must contend with the massive information overload that the World Wide Web presents to the individual. This is compounded by the lack of a proper, consistent peer-review process for material made available over the Internet. It can therefore be extremely dangerous to rely on data sourced purely off the net.

There are several strategies that one can adopt to search for material, and the most common starting point is the large search engines, such as Altavista or Google. The advantage of these search engines is that they can conduct a comprehensive search across the Internet for key phrases in seconds. However the disadvantages are often evident when too many pages of dubious quality are returned. It is possible to refine ones search and reduce the number of items returned but as yet no method of verifying their quality exists.

A second option in commencing an Internet search is to use one of the bibliographic databases, many of which are exclusively medical. Leading examples include Medline and the cancer specific database CancerLit. Such sources as entry points into online information offer the user some confidence that the comprehensive search will be limited to published material. Much of this will have undergone the necessary process of peer-review as quality control. There still remains one major disadvantage in that the user will often only have access to the abstracts of material. Full-text papers, if available, usually involve some kind of rights fee.

Another strategy would be to use one of the many news groups, which are available to individuals with a specific interest. These are essentially chat rooms. Within medicine, good examples are alt.support.cancer.breast and sci.med.diseases.cancer. However once again the user cannot be assured of the quality of the posted messages which may be open to abuse and inaccuracy. On the other hand within specialist areas, the members of the group will be known and their reputation acts as an endorsement of quality. In a similar vein, new initiatives based on the concept of the news group (such as clinical.netprints.org) provide researchers with the opportunity to post research ahead of publication whilst protecting copyright and inviting informal peer-review. There are, however, currently no similar initiatives specific to the breast cancer research community.

One of the most useful general methods for accessing information on breast cancer from the web is via one of the online medical directories. These are essentially compendiums of links to relevant web sites and are a shortcut to the less specific Internet search engines. Examples include Cancer Index and Medical Matrix. In these the user can be confident that there has been a degree of quality control of the links that are available. Additionally, these directories list the web sites they include under specific categories and make the task of identifying relevant sites much easier.

The last strategic option for the healthcare professional sourcing information from the Internet is the specialist, peer-reviewed, portal site. This combines the advantages of being comprehensive in content and links with the quality control which is normally only associated with the established print-on-paper journals. One of these specialist sites is Breast Cancer Online (http://www.bco.org/) that is specific to the needs of breast cancer professionals. It contains peer-reviewed articles that are quality controlled by an international editorial board, in much the same way as a traditional peer-reviewed journal. Indeed, Breast Cancer Online and the more research orientated Breast Cancer Research site, both have unique identifiers, and can thus both be cited and indexed. The editorial board also vets the links section and thus obviates the process of assessment the user.

For the present, the best practical solution for the health care professional seeking reliable information from the Internet on breast cancer is to use a combination of the options described. To rely on the general search engines is unwise and wasteful when compared to the quality controlled medical directories or the peer-reviewed portal sites, such as Breast Cancer Online. To search published papers, the best of the bibliographic databases is PubMed but this does exclude many journals (especially non-US ones) and it is not always possible to access full text directly. For access to full-text papers, the best source is to go to the relevant journal’s homepage (for example, http://www.bmj.com/, which is one of the best adaptations of a print on paper journal to online) or the “clearing houses” for journal subsidiary rights, such as High Wire or Catchword/Ingenta.

As the quantity of information on the Internet continues to grow exponentially it seems the development of portal sites, such as Breast Cancer Online, within all the medical subspecialties will provide a pivotal role in allowing researchers and clinicians to find their way through the jungle of information rapidly and with confidence.

Back | Top of Page