Annals of Surgical Oncology

, Volume 16, Issue 9, pp 2442–2449

Survey of ASCO Members on Management of Sentinel Node Micrometastases in Breast Cancer: Variation in Treatment Recommendations According to Specialty

Authors

  • Nabil Wasif
    • John Wayne Cancer Institute at Saint John’s Health Center
  • Xing Ye
    • John Wayne Cancer Institute at Saint John’s Health Center
    • John Wayne Cancer Institute at Saint John’s Health Center
Breast Oncology

DOI: 10.1245/s10434-009-0549-7

Cite this article as:
Wasif, N., Ye, X. & Giuliano, A.E. Ann Surg Oncol (2009) 16: 2442. doi:10.1245/s10434-009-0549-7

Abstract

Background

American Society of Clinical Oncology (ASCO) guidelines recommend axillary dissection (ALND) as the primary treatment for sentinel node micrometastases (SNMM) in breast cancer. We hypothesized that variation exists in adherence to these guidelines and management of SNMM as a result of the medical specialty of the treating physician.

Methods

A questionnaire was e-mailed to 2511 active members of ASCO who specialize in breast cancer and who are board certified in surgery, medical oncology, or radiation oncology. Questions covered treatment of SNMM and factors influencing clinical decisions. Survey responses were analyzed by the Kruskal–Wallis and Fisher’s exact test.

Results

Of 612 questionnaires viewed by recipients, 537 (88%) were completed by 382 medical oncologists, 100 surgeons, and 55 radiation oncologists. Most (98.5%) regarded SNMM as clinically important. Unexpectedly, ALND for SNMM was recommended by only 23% of surgeons, 23% of medical oncologists, and 15% of radiation oncologists (no statistically significant intergroup difference). Factors identified as important in selecting patients for ALND were tumor size, age, tumor grade, lymphovascular invasion, and size of metastases. Most respondents in each specialty (89% overall) would consider adjuvant chemotherapy for primary tumors <1 cm if SNMM were present. Tumor size, age, size of metastases, estrogen receptor/progesterone receptor status, HER-2 neu status, and gene expression profile would influence this decision. Radiation oncologists (76%) were more likely than medical oncologists (57%) or surgeons (47%) to consider axillary radiation instead of ALND for SNMM (P = 0.0021).

Conclusions

Guidelines recommending ALND for SNMM are not being followed by most of the ASCO members we surveyed.

Traditional staging of the axilla in patients with breast cancer was performed with an axillary lymph node dissection (ALND). The introduction of the sentinel lymph node biopsy (SLNB) technique in the early 1990 s as an alternative to routine ALND for axillary staging dramatically altered the treatment of these patients.1 Since then, SLNB has been used with increasing frequency and been shown to be as effective as ALND in staging a clinically negative axilla with less morbidity.2 The sentinel lymph node is defined as the first node that receives lymphatic drainage from the breast, and hence the most likely to harbor metastases. SLNB also allows for more detailed histological examination of lymph nodes that are at highest risk of harboring metastases, including multiple sectioning with or without immunohistochemical staining, in addition to conventional hematoxylin and eosin staining.3 This focused examination of lymph nodes has in turn led to increased detection of low-volume metastases or micrometastases and isolated tumor cells.4,5

As a reflection of these changes, the American Joint Committee on Cancer updated the staging system for breast cancer in 2002 to categorize tumor deposits in lymph nodes by size: macrometastases (>2.0 mm), micrometastases (0.2–2.0 mm), and isolated tumor cells (ITC) (<0.2 mm). The corresponding N stage is N1-3 for macrometastases, N1mi for micrometastases, and N0(i+) for ITC. A new category of N0(mol+) was also introduced to reflect those nodes that are negative histologically but positive by molecular techniques such as reverse transcriptase-polymerase chain reaction (RT-PCR).5

Although the adverse prognostic impact of macrometastases is well recognized, the clinical significance of micrometastases remains uncertain, leading to variability in locoregional and systemic management.6 In an effort to standardize management of patients with node positive-disease and to define the role of SLNB in early breast cancer, the American Society of Clinical Oncology (ASCO) convened an expert panel of leaders in the field. On the basis of existing evidence, the panel recommended “routine ALND…for patients with micrometastases (0.2–2.0 mm) found on SLNB…, regardless of the method of detection.”7 This recommendation was based on the fact that in these patients, additional metastases are found in nonsentinel nodes (NSN) 20% to 35% of the time.7 The panel also noted that there was insufficient evidence to recommend radiation as an alternative to ALND for the treatment of SLNB micrometastases at this time.8

In a recent study, our group demonstrated that only 60% of patients with SLNB micrometastases were treated with an ALND in the general population.9 We hypothesized that some of this underuse could be explained by biases intrinsic to the specialty of the treating physician. To test our hypothesis that treatment decisions vary by specialty, we conducted a survey study among breast cancer specialists to assess management of SLNB micrometastases.

Methods

Survey Instrument

The survey used for this study was developed by the authors in consultation with board-certified medical and radiation oncologists at the John Wayne Cancer Institute (JWCI) who were also members of ASCO. Modification of the survey items was then performed by literature review, small focus group discussions, and multidisciplinary breast cancer tumor board. Pilot testing of the survey was performed at JWCI during a 2-month period for face and content validity and further refinement. The final construct was then developed; it contained 11 questions and took an average of 2 min to complete. Commercially available software (QuestionPro, Seattle, WA) was used for survey design and electronic dissemination. A cover paragraph accompanied the survey explaining the purpose of the study. A single reminder was sent to survey participants who did not respond to the initial request. No incentive was provided to complete the survey.

Demographic questions were asked about medical specialty and percentage of time devoted exclusively to management of patients with breast cancer. Other questions focused on the impact of micrometastases, ITC, and RT-PCR positivity after SLNB in clinical decision making regarding further treatment. Respondents were queried on how often they would recommend ALND, chemotherapy, or radiotherapy as a treatment for micrometastases. Respondents were also asked to identify which factors were important in helping them decide on treatment recommendations. The complete survey questionnaire is in Appendix 1.

Study Design and Population

Respondents were active members of the ASCO who listed breast cancer as their oncological specialty and had a board certification in surgery, medical oncology, or radiation oncology. Only physicians practicing in the United States were chosen. A total of 2511 ASCO members meeting inclusion criteria were chosen from the online ASCO membership directory. No unique information that could potentially identify a respondent was collected. Approval for the study was obtained from the institutional review board at JWCI. Completion of the survey was considered implied consent for participation.

Statistical Analysis

Analysis of the data was performed by the JWCI department of biostatistics. Descriptive analyses were performed by percentages. For comparison purposes, respondents were divided into surgery, medical oncology, and radiation oncology according to specialization. Differences in survey responses between specialties were analyzed by the Kruskal-Wallis and Fisher’s exact tests. All tests were two sided, and a P value of <0.05 was considered to be statistically significant. For multiple comparisons between the three groups, a Bonferroni-corrected significance level of 0.017 was used. All data were analyzed by SAS software (SAS Institute, Cary, NC).

Results

Of 612 e-mail questionnaires viewed by recipients, 537 (88%) were completed by 382 medical oncologists, 100 surgeons, and 55 radiation oncologists. Most respondents (81%) devoted >50% of their clinical practice to the treatment of breast cancer.

When questioned about the impact of low-volume metastases, i.e., N1mi, N0(i+), and N0(mol+) on clinical decision making, the responses were as in Fig. 1. Most respondents thought SLNB micrometastases were sometimes (43.8%) or always (54.7%) a factor in their clinical decision making (Fig. 1a). Surgeons seemed to place more emphasis on the presence of micrometastases, with 73% considering micrometastases as always significant clinically, compared with 51% and 44% of medical oncologists and radiation oncologists, respectively (P < 0.001 for both surgery versus medical oncology and surgery versus radiation oncology).
https://static-content.springer.com/image/art%3A10.1245%2Fs10434-009-0549-7/MediaObjects/10434_2009_549_Fig1_HTML.gif
Fig. 1

Clinical significance of: a N1mi, b N0(i+), and c N0(mol+)

Less importance was placed on the presence of ITC or RT-PCR positivity; 52.9% overall did not consider ITC to have an impact on clinical decision making, and 57% thought the same way for RT-PCR positivity. A high proportion of respondents were undecided on the clinical importance of ITC (28%) and of RT-PCR positivity (32%). No statistically significant intergroup differences were seen in these responses.

Axillary Dissection

Only 22.1% of the respondents would always recommend an ALND for micrometastases, although 73% would recommend it sometimes. When broken down by specialty, routine ALND for micrometastases was recommended by only 23% of surgeons, 23% of medical oncologists, and 14% of radiation oncologists (no statistically significant intergroup difference) (Fig. 2a). Factors identified as important in selecting patients for ALND were primary tumor size, age of patient, tumor grade, lymphovascular invasion, and size of the metastases. Histology, presence of lobular carcinoma-in situ, and family history did not influence decisions (Table 1). When a Bonferroni correction was applied to look for differences within groups, only the choice of tumor grade differed significantly between surgeons and radiation oncologists (69 vs. 87%, P = 0.012).
https://static-content.springer.com/image/art%3A10.1245%2Fs10434-009-0549-7/MediaObjects/10434_2009_549_Fig2_HTML.gif
Fig. 2

Treatment decisions by specialty

Table 1

Factors considered in selecting patients with SLNB micrometastases for ALND

Factor

All respondents (N = 537)

Surgery (N = 100)

Medical oncology (N = 382)

Radiation oncology (N = 55)

P value

Factors influencing selection for ALND (% answering yes)

    Primary tumor size

83.8

83

82

95

NS

    Size of micrometastases

78.8

81

77

85

NS

    Tumor grade

78.4

69

80

87

0.018

    Patient age

77.5

79

75

89

NS

    Lymphovascular invasion

70

56

69

76

0.036

Factors not important for ALND (% answering yes)

    Histology IDC versus ILC

19.7

27

19

15

0.0018

    DCIS

7.8

10

7

13

NS

    LCIS

3.5

4

3

5

NS

    Family history

12.5

9

13

18

NS

SLNB sentinel lymph node biopsy, ALND axillary lymph node dissection, IDC infiltrating ductal carcinoma, ILC infiltrating lobular carcinoma, DCIS ductal carcinoma-in situ, LCIS lobular carcinoma-in situ

Chemotherapy

Most respondents would recommend chemotherapy sometimes for SLNB micrometastases with a primary tumor <1 cm (89%), with no statistically significant difference between groups (Fig. 2b). Primary tumor size, patient age, lymphovascular invasion, size of metastases, HER-2 neu status, estrogen receptor/progesterone receptor status, and results of gene expression profile testing were factors selected as influencing chemotherapy decisions. Histology, family history, S phase fraction percentage, Ki-67 proliferation index, p53, and ploidy/DNA index were not important (Table 2). Interestingly, radiation oncologists placed more emphasis than either surgeons or medical oncologists on S phase fraction percentage, ploidy/DNA index, and p53 (P < 0.017 for all).
Table 2

Factors considered in selecting patients with SLNB micrometastases for chemotherapy

Factor

All respondents (N = 537)

Surgery (N = 100)

Medical oncology (N = 382)

Radiation oncology (N = 55)

P value

Factors influencing selection for chemotherapy (% answering yes)

    Primary tumor size

86.6

88

85

93

NS

    Patient age

88.3

87

87

98

NS

    Size of micrometastases

70

77

66

84

0.007

    Lymphovascular invasion

67.2

58

68

69

0.045

    Results of gene expression profile testing

69.3

66

70

73

NS

    HER-2 neu status

89.9

81

93

87

0.003

ER/PR status

90.9

895

91

95

NS

Factors not important for chemotherapy (% answering yes)

 

    Histology IDC versus ILC

14.9

12

16

15

NS

    Family history

12.9

9

13

18

NS

    S phase fraction

16.9

13

16

33

0.0037

    Ki-67 proliferation index

25.9

21

26

33

NS

    p53

8.6

7

7

24

0.0004

    Ploidy/DNA index

9.5

8

7

29

<0.001

ER estrogen receptor, PR progesterone receptor, IDC infiltrating ductal carcinoma, ILC infiltrating lobular carcinoma

Radiation

Radiation oncologists (76%) were more likely than medical oncologists (57%) or surgeons (47%) to consider axillary radiation instead of ALND for SLNB micrometastases (P ≤ 0.017 for both comparisons) (Fig. 2c). Factors influencing this decision are outlined in Table 3. The largest variation in responses was seen in this question. Significantly more emphasis was placed on patient age, primary tumor size, lymphovascular invasion, and size of metastases by radiation oncologists than surgeons and medical oncologists (P < 0.017 for all comparisons).
Table 3

Factors influencing treatment of SLNB micrometastases with axillary radiation as an alternative to ALND

Factors influencing use of radiation versus ALND (% answering yes)

All respondents (N = 537)

Surgery (N = 100)

Medical oncology (N = 382)

Radiation oncology (N = 55)

P value

ALND performed or not

85.1

88

83

96

0.02

Primary tumor size

51.8

30

54

78

<0.001

Size of micrometastases

49.3

46

46

82

<0.001

Patient age

43.4

37

41

69

<0.001

Lymphovascular invasion

40.4

35

38

69

<0.001

Tumor grade

38.7

25

41

45

<0.001

Histology IDC versus ILC

6.7

9

6

3

NS

Family history

4.7

5

4

4

NS

SLNB sentinel lymph node biopsy, ALND axillary lymph node dissection, IDC invasive ductal carcinoma, ILC invasive lobular cancer

Discussion

The aim of this study was to evaluate variation in management of SLNB micrometastases in breast cancer patients by physicians from different specialties and to assess adherence to current guidelines.

We found that most physicians would not recommend routine ALND for treatment of micrometastases, but would consider chemotherapy for these patients. Several factors influenced these decisions and were by and large consistent across specialties. Considering axillary radiation instead of ALND for micrometastases was a treatment option most favored by radiation oncologists than by either medical oncologists or surgeons.

Unexpectedly, ALND was routinely recommended by only 23% of surgeons specializing in breast cancer care. Most of the surgeon respondents (74%) would consider ALND as a treatment option, and some outliers would not recommend it at all (3%). It is possible that the wording of the answer choices as “always” and “sometimes” may have resulted in many respondents picking the “sometimes” option to account for patients in whom medical comorbidities may be prohibitive to performing an ALND, even though they would recommend it routinely. Both medical oncologists and radiation oncologists have potential alternative therapies to offer, so they might not be as inclined to recommend routine ALND for SLNB micrometastases. Nevertheless, the consensus guidelines state clearly that these patients should undergo ALND, and the responses reflect a departure from that recommendation.

Even though the prognostic significance of micrometastases remains controversial, there is more and more evidence to suggest that these patients may have a worse survival than truly node-negative patients, especially on longer follow-up.10,11 An analysis of the Surveillance, Epidemiology, and End Results Program database performed by our group showed that survival for N1mi disease was intermediate to N0 and N1, and that N1mi remained an important prognostic indicator by multivariate analysis.12 Furthermore, the presence of SLNB micrometastases is often a marker for additional metastases in NSN. A meta-analysis by Cserni et al. put the incidence at 20%, and our group showed that forgoing ALND risks leaving behind NSN metastases (mostly macrometastases) in 21% of patients.9,13 A conscious decision to omit ALND by the treating physician implies that the physician does not consider macrometastases in NSN to have any clinical significance in these patients—an assertion not supported by data. Even though NSABP-04 did not show a survival benefit for ALND, a subsequent Bayesian meta-analysis of six randomized trials with long-term follow-up showed an average survival benefit of 5.4% for patients undergoing ALND.14

Respondents identified primary tumor size, patient age, tumor grade, lymphovascular invasion, and size of metastases as factors influencing selection of ALND as treatment after identification of SLNB micrometastases. In studies assessing the likelihood of NSN metastases with SLNB micrometastases, primary tumor size, size of metastases, lymphovascular invasion, and grade have all been identified as risk factors.1517 Patient age can be considered a proxy for medical comorbidity, and physicians may also be reluctant to leave disease behind in patients who are younger and at risk for axillary recurrence later in life.

Most respondents agreed on offering adjuvant chemotherapy to patients with SLNB micrometastases. As reflected by the responses, current decisions on systemic therapy in breast cancer are usually made on the basis of characteristics of the primary tumor and not on the presence or absence of micrometastases.6 In clinical practice, there will be a small subset of patients in which the decision to provide adjuvant chemotherapy will hinge only on the presence of micrometastases. In these cases, the guidelines suggest they should be considered node positive and treated as such. Clinical decision making is more complex in the presence of ITC or RT-PCR positivity, and little information exists on the subject. Currently, there is little evidence to suggest that either ITC of RT-PCR positivity has an adverse effect on survival in breast cancer patients.6

Not surprisingly, the use of axillary radiation as an alternative to treat SLNB micrometastases instead of ALND was an option most often selected by radiation oncologists (71%), even though many surgeons (48%) and medical oncologists (56%) would consider it “sometimes.” Several studies have reported on the use of radiotherapy instead of ALND for SLNB-positive nodes, although the data are difficult to interpret because of the use of systemic therapy and breast tangential radiation. In many cases, most level I and II axilla can be included in a breast-conservation field if a high-tangent technique is used.18 NSABP B-04 randomly assigned patients to modified radical mastectomy, total mastectomy plus radiation, or total mastectomy alone. At 25-year follow-up patients with positive nodes had an axillary recurrence rate of 11% if treated with radiation compared with 8% for those treated with axillary dissection (P = NS).19 Another study involved 658 patients with clinically negative nodes and primary tumor <3 cm randomly assigned to axillary dissection or axillary radiation after breast-conservation therapy and breast radiation. Recurrences in the axillary node were less frequent in the axillary dissection group at 15 years (1 vs. 3%; P = 0.04), even though survival was equivalent for both groups.20 For patients with micrometastases only, Langer et al.21 reported on 27 patients with micrometastases who did not undergo ALND or axillary radiation and who did not have any axillary recurrence at a median follow-up of 42 months. Axillary radiation may therefore be considered for patients with a positive SLNB who are poor surgical candidates or who decline ALND, but at this point, it cannot be routinely recommended as an alternative to ALND.

Limitations of our study should be noted. Electronic dissemination of questionnaires enables a cheap and instantaneous way to reach respondents. However, many of our requests were undeliverable as a result of outdated e-mail addresses or inquiries that were bounced back from e-mail filters. Similarly, even though 2511 subjects met the inclusion criteria, only 612 actually viewed the e-mail. We postulate that in most cases, the e-mail was delivered to the junk folder, and thus discarded or never read. The high response rate (88%) of those who did open the e-mail shows that once the survey reached respondents, most went on to complete it. Because the responses are self-generated, it is possible that misclassification may have occurred, although the survey was anonymous to enable accurate reporting.

To conclude, we show that great variation among specialties exists in the management of SLNB micrometastases in breast cancer. Until data from randomized trials become available, these patients should be treated with ALND and chemotherapy when medically appropriate, in line with current consensus guidelines. Omission of ALND should be in the setting of a clinical trial or for comorbidity only because we have yet to identify a subset in which this can be safely performed.

Acknowledgment

Supported by grants from the Gonda (Goldschmied) Research Laboratories of the John Wayne Cancer Institute at Saint John’s Health Center. Supported by funding from QVC and the Fashion Footwear Association of New York Charitable Foundation (New York, NY), the Margie and Robert E. Petersen Foundation (Los Angeles, CA), Mrs. Lois Rosen (Los Angeles, CA), and the Associates for Breast and Prostate Cancer Studies (Santa Monica, CA). Supported by funding from the Family of Robert Novick (Los Angeles, CA), Ruth and Martin H. Weil Fund (Los Angeles, CA), and the Wrather Family Foundation (Los Alamos, CA). Special thanks to Gwen Berry for editorial assistance.

Copyright information

© Society of Surgical Oncology 2009