Determinants of axillary recurrence after axillary lymph node dissection for invasive breast cancer

EJSO 2001; 27: 250–255 doi:10.1053/ejso.2000.1111, available online at http://www.idealibrary.com on

Determinants of axillary recurrence after axillary lymph node dissection for invasive breast cancer A. C. Voogd∗, R. de Boer†, M. J. C. van der Sangen‡, R. M. H. Roumen¶, H. J. T. Rutten∀ and J. W. W. Coebergh∗,∗∗ ∗Comprehensive Cancer Centre South, P.O. Box 231, 5600 AE, Eindhoven, The Netherlands; †Academic Hospital Maastricht, P.O. Box 5800, 6202 AZ, Maastricht, The Netherlands; ‡Department of Radiotherapy, Catharina Hospital, P.O. Box 1350, 5602 ZA, Eindhoven, The Netherlands; ¶Department of Surgery, St. Joseph Hospital, P.O. Box 7777, 5500 MB, Veldhoven, The Netherlands; ∀Department of Surgery, Catharina Hospital, P.O. Box 1350, 5602 ZA, Eindhoven, The Netherlands; ∗∗Department of Public Health, Erasmus University Medical School, P.O. Box 1738, 3000 DR, Rotterdam, The Netherlands

Aim: This study was undertaken to gain insight into the risk factors for axillary recurrence among patients with invasive breast cancer who underwent breast-conserving treatment or mastectomy and axillary lymph node dissection. Methods: In a matched case–control design, 59 patients with axillary recurrence and 295 randomly selected control patients without axillary recurrence were compared. Matching factors included age, year of incidence of the primary tumour and postsurgical axillary nodal status. Results: For patients with negative axillary lymph nodes, those with a tumour in the medial part of the breast had a 73% (95% CI: 4–92%) lower risk of axillary recurrence compared to those with a tumour in the lateral part of the breast. For the patients with positive axillary lymph nodes the risk of axillary recurrence was 65% (95% CI: 16–86%) lower for those who had received axillary irradiation compared to those without axillary irradiation. Within the age group <50 years, the risk or axillary recurrence was 82% lower (95% CI: 45–94%) for patients with more than six lymph nodes found in the axillary specimen compared to those with six or less than six lymph nodes. Conclusions: Although based on a small number of patients, this study indicates that axillary irradiation is effective in reducing the risk of axillary recurrence for patients with positive lymph nodes. This favourable effect only applies to the subgroup with extranodal extension or nodal involvement in the apex of the axilla, as these were the only patients receiving axillary radiation during the study period.  2001 Harcourt Publishers Ltd

Key words: breast cancer; axillary recurrence; risk.

INTRODUCTION Axillary lymph node dissection (ALND) is a standard surgical procedure in the treatment of invasive breast cancer. The objectives of ALND are to stage the disease and prevent axillary recurrence (AR). Recently, less invasive techniques have been introduced to identify patients who do not need to undergo ALND. Of these, the sentinel node biopsy is the most promising.1–3 At present, most patients with invasive breast cancer still

Correspondence to: Adri C. Voogd, Comprehensive Cancer Center South, P.O. Box 231, 5600 AE Eindhoven, The Netherlands. Tel: +31 40 2971616; Fax: +31 40 2971610; E-mail: [email protected] 0748–7983/01/030250+06 $35.00/0

undergo ALND. In most series the risk of AR after full ALND or dissection of the lower and middle parts of the axilla (i.e. levels I and II) varied between 0.5 and 3.0%.4–11 To gain insight into the risk factors for AR large patients series are required. Most studies on AR are based on fewer than 30 patients, and often AR is not listed separately from other nodal sites of failure. By using follow-up data from a population-based cancer registry, 59 patients were identified with AR after ALND for the treatment of invasive breast cancer. No distant metastases were present at the time of diagnosis of AR. For each patient with AR five control patients without AR were selected with a follow-up of similar duration. The patients with and without AR were compared to  2001 Harcourt Publishers Ltd

DETERMINANTS OF AXILLARY RECURRENCE IN BREAST CANCER investigate potential case–control analysis.

risk

factors

in

a

matched

PATIENTS AND METHODS Patients Patients were selected from the population-based Eindhoven Cancer Registry, which serves a population of almost one million inhabitants in the southeastern part of The Netherlands (6% of the Dutch population). Data are collected by the cancer registry from copies of the pathology reports of 10 pathologists in three laboratories and from the medical records of eight community hospitals and one department of radiotherapy. During the period 1984–94 5408 patients with invasive breast cancer were documented, 4669 of whom (86.3%) underwent ALND. A total of 208 patients (4.5%) had either concurrent or sequential bilateral carcinomas, giving a total of 4877 ALNDs. In the treatment guidelines of the regional Breast Cancer Study Group, the borders of ALND consist of the latissimus dorsi muscle (dorsal), the thoracic wall below the major and minor pectoral muscles (ventral), and the lower border of the axillary vein (cranial). As a rule ALND was not always performed when the tumour measured 0.5 cm or less and was usually omitted in the case of irresectable cancer and for the very elderly. Irradiation of the axilla and the supraclavicular region was recommended for patients with inadequate ALND, extracapsular extension of tumour growth or nodal involvement in the apex of the axilla. The radiation dose and dose specification varied with time. The axillary field was usually exposed to 52.5 Gy (25 fractions) and the supraclavicular field to 40 Gy (20 fractions). The dose specification varied from 30 mm to 50 mm below the surface of the skin. During the study period adjuvant systemic treatment was recommended only for axillary node-positive patients. Chemotherapy was indicated for premenopausal node-positive patients and endocrine therapy (tamoxifen) for postmenopausal node-positive patients with an oestrogen or progesterone receptor-positive tumour. Since 1989 the Eindhoven Cancer Registry has recorded follow-up information on patients with breast cancer, including the date and site of local, regional and distant recurrence and the date of death. The information is provided by all pathologists, the radiotherapists and the majority of the surgeons. Until January 1998, 79 patients were reported with AR without concomitant or earlier metastases when the AR was diagnosed. After examination of the clinical records 20 patients were excluded: eight patients still appeared to have distant disease at the time of diagnosis of AR, in nine cases AR was reclassified as supraclavicular disease, and for three patients AR could not be confirmed. Of the remaining 59 patients, two had a concurrent recurrence in the

251

breast, four had an earlier recurrence in the breast and one had an earlier recurrence in the chest wall. For each of the 59 patients with AR five controls were randomly selected from the remaining cohort of patients without AR, given that they fulfilled the following four matching criteria: (1) follow-up time (i.e. each control had to have survived without distant disease for at least as long as the time between dissection of the primary tumour and the diagnosis of AR in the corresponding case), (2) age (plus or minus two years difference), (3) year of incidence of the primary tumour (plus or minus one year difference) and (4) postsurgical axillary nodal status (node-positive vs node-negative).

Statistical methods The relative risks of AR associated with disease and treatment characteristics were estimated by comparing the distribution of these characteristics between each patient with AR (i.e. the case) and the five matched controls in a conditional logistic regression analysis for individually matched pairs.12 Relative risk estimates (RR), two-sided P values, and 95% confidence intervals (95% CI) were calculated with the microcomputer program EGRET (SERC, Seattle, Washington, USA). Analyses were performed according to axillary nodal status. The following variables were analysed to assess their ability to predict the risk of AR: tumour site, histological type, tumour diameter, oestrogen receptor status, and number of lymph nodes found in the axillary specimen. For the patients with positive axillary lymph nodes the following factors were also studied: the number of positive lymph nodes, extranodal extension of tumour growth, nodal involvement of the apex of the axilla, axillary irradiation and adjuvant systemic treatment. Because of the potential confounding effect of age, a conditional logistic regression analysis was also performed according to age group (<50 vs [50 years of age).

RESULTS Risk factors for AR according to pathological node status (Table 1a and 1b) For patients with negative axillary lymph nodes, those with a tumour in the medial part of the breast had a significantly lower risk of AR compared to those with a tumour in the lateral part of the breast (RR=0.27; 95% CI: 0.08–0.96) (Table 1a). None of the other factors was found to be significantly associated with the risk of AR. For patients with positive axillary lymph nodes the relative risk of AR was 0.35 (95% CI: 0.14–0.84) for those who had received axillary irradiation, compared to those without axillary irradiation (Table 1b). Of the

252

A. C. VOOGD ET AL.

Table 1a Relative risks (RR) of axillary recurrence for patients with negative axillary lymph nodes (pN−) Variable Tumour site Lateral Medial Central Overlapping quadrants Histological type Ductal Lobular Other Tumour diameter (cm) Ζ1.0 1.1–3.0 >3.0 Oestrogen receptor status Positive Negative Number of lymph nodes examined [7 1–6

Cases/controls (34/170)

RR (95% CI)

P value

22/74 3/39 4/19 5/34

1.0 (–) 0.27 (0.08–0.96) 0.69 (0.20–2.42) 0.48 (0.17–1.43)

0.04 0.56 0.19

27/129 6/22 1/19

1.0 (–) 1.31 (0.49–3.51) 0.26 (0.03–2.00)

0.59 0.26

18/106 12/44 4/16

1.0 (–) 1.61 (0.71–3.62) 1.47 (0.46–4.77)

0.25 0.52

21/105 5/33

1.0 (–) 0.56 (0.18–1.73)

0.32

18/106 13/47

1.0 (–) 1.59 (0.69–3.67)

0.28

other potential risk factors, none appeared to be significantly associated with the risk of AR.

Risk factors for AR according to age group (Table 2) Within the age group <50 years, the relative risk of AR was 5.51 (95% CI: 1.83–16.57) for patients with less than seven lymph nodes found in the axillary specimen, compared to those with seven or more than seven lymph nodes. No such association was found within the age group [50 years. Tumour site, histological type, tumour diameter and oestrogen receptor status were not significantly associated with the occurrence of AR, among patients <50 years as well as those [50 years.

DISCUSSION By means of a case–control study, we were able to identify several risk factors for axillary recurrence (AR) in breast cancer patients who have undergone axillary lymph node dissection (ALND). For patients with negative axillary lymph nodes, the risk of AR appeared to be lower for those with a tumour in the medial part of the breast. For patients with positive lymph nodes, the risk was reduced by nodal irradiation. Patients with less than seven lymph nodes reported by the pathologist showed a higher risk of AR compared to those with more lymph nodes reported. However, this association was only found for patients under 50 years of age. Although the case–control design of the current study proved to be an efficient way of identifying breast cancer

patients with an increased risk for recurrence in the treated axilla, it has some limitations. In a case–control study, only relative risks can be calculated, not the absolute recurrence rates which are usually presented in cohort analyses. Moreover, by matching for age and nodal status it became impossible to study the prognostic effect of these factors. Finally, when considering the results it should be realized that the number of patients is small, so that we could not perform multivariate analyses and adjust for several confounding factors simultaneously, and which explains the broad confidence intervals of the estimated risks. There is convincing evidence in the literature that the risk of AR as the first site of failure is low for patients with an adequate ALND. Reported rates after complete or levels I and II ALND range between 0.5% and 3%.4–11 Much higher rates have been found after less extensive surgery in the axilla. In a study of Graversen et al., the 5-year probability of developing AR was 10% for patients with one or two excised negative lymph nodes and 5% for patients with 3–4 excised negative nodes.13 Wilking et al. reported a 5-year AR rate of 6% for patients with less than five nodes examined, compared to 2% for 5–9 nodes and 1% for 10 or more. These figures correspond with the five-fold increased risk found in our study for patients <50 years of age with less than seven nodes found in the axillary specimen. These data illustrate that most ARs arise from tumour tissue left behind in the axilla. Patients with a tumour in the medial part of the breast have been demonstrated to have a lower risk of axillary involvement,14,15 and this might explain our finding that among node-negative patients the risk of AR was

DETERMINANTS OF AXILLARY RECURRENCE IN BREAST CANCER

253

Table 1b Relative risks (RR) of axillary recurrence for patients with positive axillary lymph nodes (pN+) Variable Clinical axillary nodal status (cN) cN− cN+ Tumour site Lateral Medial Central Overlapping quadrants Histological type Ductal Lobular Other Tumour diameter (cm) Ζ1.0 1.1–3.0 >3.0 Oestrogen receptor status Positive Negative Number of lymph nodes examined [7 1–6 Number of positive lymph nodes 1–3 >3 Extranodal tumour growth No Yes Nodal involvement of apex of axilla No Yes Axillary irradiation No Yes Adjuvant systemic treatment No Yes

Cases/controls (25/125)

RR (95% CI)

P value

16/83 6/27

1.0 (–) 1.08 (0.39–2.98)

0.89

10/58 3/17 6/16 6/34

1.0 (–) 1.09 (0.25–4.81) 2.40 (0.67–6.82) 1.03 (0.35–3.05)

0.91 0.20 0.96

21/97 3/24 1/4

1.0 (–) 0.58 (0.16–2.08) 1.21 (0.13–10.8)

0.40 0.87

6/33 8/36 11/53

1.0 (–) 1.22 (0.38–3.95) 1.14 (0.38–3.45)

0.74 0.81

13/75 10/34

1.0 (–) 1.71 (0.66–4.45)

0.27

15/90 9/24

1.0 (–) 2.18 (0.82–5.78)

0.12

15/66 9/53

1.0 (–) 0.74 (0.30–1.80)

0.51

16/55 9/67

1.0 (–) 0.47 (0.19–1.16)

0.10

17/81 7/43

1.0 (–) 0.79 (0.31–2.04)

0.63

16/46 9/77

1.0 (–) 0.35 (0.14–0.84)

0.02

6/23 19/102

1.0 (–) 1.46 (0.49–4.34)

0.50

lower for this group compared to the patients with a tumour in the lateral quadrants. To reduce the risk of AR, misclassification of patients with involved axillary lymph nodes as node-negative must be avoided and the extent of node positivity should be accurately determined. Alternatives to ALND need to be equally effective in staging the axilla. Sentinel lymph node biopsy seems to yield accurate assessment of lymph node status. In a crude overview of early experience with the sentinel lymph node biopsy the false-negative rate was found to be 6.2%.16 The risk of AR after sentinel node biopsy should not be higher than the risk after ALND. It remains unclear whether equal regional control rates can be obtained after sentinel node biopsy since follow-up of most series is still very short. Axillary irradiation significantly reduced the risk of AR

in patients with extensive regional disease. One should realize, however, that the absolute risk of AR is low, and that a large number of patients will need axillary irradiation to prevent a relatively small number of recurrences. The 65% risk reduction in the current study is in accordance with the results of others.17–20 The trials of Overgaard et al. and Ragaz et al. have demonstrated a significant decrease in the rate of regional and local recurrence and an improved survival due to postoperative loco-regional radiotherapy in addition to mastectomy including ALND and systemic treatment for patients with larger tumours and positive lymph nodes.17,18,20 In a retrospective analysis of 618 patients with 10 or more positive lymph nodes Diab et al. found the risk of loco-regional recurrence to be 61% lower in the subgroup that received adjuvant loco-regional irradiation.19 During our study period (i.e. 1984–97),

254

A. C. VOOGD ET AL.

Table 2 Relative risks (RR) of axillary recurrence, according to age group Variable

Age group [50 years

<50 years Cases/controls (18/90)

RR (95% CI)

P value

11/46 2/11 1/9 4/23

1 (–) 0.86 (0.16–4.45) 0.47 (0.05–4.34) 0.78 (0.23–2.66)

0.86 0.51 0.69

14/65 4/16 0/9

1 (–) 0.75 (0.23–2.45)

0.63

11/42 5/21 2/24

1 (–) 0.83 (0.26–2.68) 0.27 (0.05–1.61)

12/42 4/32

1 (–) 0.30 (0.08–1.20)

7/67 11/16

1 (–) 5.51 (1.83–16.57) 0.002

Tumour site Lateral Medial Central Overlapping quadrants Histological type Ductal Lobular Other Tumour diameter (cm) Ζ1.0 1.1–3.0 >3.0 Oestrogen receptor status Positive Negative Number of lymph nodes examined [7 1–6

axillary radiation was given to patients with inadequate ALND, extracapsular tumour extension or nodal involvement in the apex of the axilla. Recently, women with four or more positive lymph nodes have also received axillary radiation, which means that approximately 30% of patients with positive nodes now receive axillary radiation. Any further extension of the indications for axillary irradiation after ALND should be carefully weighed against the elevated risk of severe oedema of the arm which is associated with the combination of these treatments.21–26 Considering the indications for axillary irradiation, full ALND remains essential for patients with a positive sentinel node to determine the extent of regional disease. The role of radiotherapy as an alternative to full ALND in patients with a positive sentinel node remains to be investigated in future randomized controlled trials. In summary, AR is uncommon among patients with operable breast cancer when adequate axillary surgery is performed. Although based on a small number of patients, this study indicates that axillary irradiation is effective in reducing the risk of axillary recurrence for patients with positive lymph nodes. This favourable effect only applies to the subgroup with extranodal extension or nodal involvement in the apex of the axilla, as these were the only patients receiving axillary radiation during the study period.

Cases/controls (41/205)

RR (95% CI)

P value

21/86 4/45 9/26 7/45

1 (–) 0.37 (0.12–1.17) 1.42 (0.56–3.57) 0.66 (0.26–1.68)

0.09 0.46 0.38

34/161 5/30 2/14

1 (–) 0.78 (0.29–2.13) 0.68 (0.15–3.05)

0.63 0.62

0.76 0.15

13/97 15/59 13/45

1 (–) 2.07 (0.89–4.80) 2.54 (0.98–6.61)

0.09

22/138 11/35

1 (–) 1.81 (0.78–4.17)

0.16

26/129 11/55

1 (–) 0.94 (0.41–2.20)

0.89

0.09 0.05

ACKNOWLEDGEMENTS The authors wish to thank L. H. van der Heijden (Comprehensive Cancer Center South) for performing the random selection procedure, and M. A. Crommelin, M.D., for his contribution to the design of the study and comments on early drafts of the paper. Financial support was received from the Scientific Fund of the Catharina Hospital in Eindhoven.

REFERENCES 1. Giuliano AE, Dale PS, Turner RR, Morton DL, Evans SW, Krane DL. Improved axillary staging of breast cancer with sentinel lymphadenectomy. Ann Surg 1995; 222: 394–401. 2. Roumen RMH, Valkenburg JGM, Geuskens LM. Lymphoscintigraphy and feasibility of sentinel node biopsy in 83 patients with primary breast cancer. Eur J Surg Oncol 1997; 23: 495–502. 3. Veronesi U, Paganelli G, Viale G, et al. Sentinel lymph node biopsy and axillary dissection in breast cancer: results in a large series. J Natl Cancer Inst 1999; 91: 368–73. 4. Dewar JA, Sarrazin D, Benhamou E, et al. Management of the axilla in conservatively treated breast cancer: 592 patients treated at Institut Gustave-Roussy. Int J Radiat Oncol Biol Phys 1987; 13: 475–81. 5. Roukema JA, van Dongen JA. Surgical treatment of recurrent breast carcinoma in the axilla. Ned Tijdschr Geneeskd 1988; 132: 2060–2. 6. Fowble B, Solin LJ, Schultz DJ, Goodman RL. Frequency, sites of relapse and outcome of regional node failures following conservative surgery and radiation for early breast cancer. Int J Radiat Oncol Biol Phys 1989; 17: 703–10.

DETERMINANTS OF AXILLARY RECURRENCE IN BREAST CANCER 7. Siegel BM, Mayzel KA, Love SM. Level I and II axillary dissection in the treatment of early-stage breast cancer. An analysis of 259 consecutive patients. Arch Surg 1990; 125: 1144–7. 8. Recht A, Pierce SM, Abner A, et al. Regional nodal failure after conservative surgery and radiotherapy for early-stage breast carcinoma. J Clin Oncol 1991; 9: 988–96. 9. Cabanes PA, Salmon RJ, Vilcoq JR, et al. Value of axillary dissection in addition to lumpectomy and radiotherapy in early breast cancer. The Breast Carcinoma Collaborative Group of the Institut Curie. Lancet 1992; 339: 1445–8. 10. Renolleau C, Merviel P, Clough KB, Asselain B, Campana F, Durand JC. Isolated axillary recurrences after conservative treatment of breast cancer. Eur J Cancer 1996; 32A: 617–21. 11. Vicini FA, Horwitz EM, Lacerna MD, et al. The role of regional nodal irradiation in the management of patients with early-stage breast cancer treated with breast-conserving therapy. Int J Radiat Oncol Biol Phys 1997; 39: 1069–76. 12. Breslow NE, Day NE. Statistical methods in cancer research. Volume I. The analysis of case-control studies. IARC Scientific Publications, No. 32. Lyon: International Agency for Research on Cancer, 1980. 13. Graversen HP, Blichert-Toft M, Andersen JA, Zedeler K. Breast cancer: risk of axillary recurrence in node-negative patients following partial dissection of the axilla. Eur J Surg Oncol 1988; 14: 407–12. 14. Gann PH, Colilla SA, Gapstur SM, Winchester DJ, Winchester DP. Factors associated with axillary lymph node metastasis from breast carcinoma. Descriptive and predictive analyses. Cancer 1999; 86: 1511–9. 15. Voogd AC, Coebergh JWW, Repelaer van Driel OJ, et al. The risk of nodal metastases in breast cancer patients with clinically negative lymph nodes: a population based analysis. Breast Cancer Res Treat 2000; 62: 63–9. 16. Sounding Board. Sentinel-lymph-node biopsy for breast cancer. Not yet the standard of care. N Engl J Med 1998; 339: 990–5. 17. Overgaard M, Hansen PS, Overgaard J, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer

18. 19.

20.

21.

22. 23. 24. 25.

26.

255

who received adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med 1997; 337: 949–55. Ragaz J, Jackson SM, Le N, et al. Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer. N Engl J Med 1997; 337: 956–62. Diab SG, Hilsenbeck SG, de Moor C, et al. Radiation therapy and survival in breast cancer patients with 10 or more positive axillary lymph nodes treated with mastectomy. J Clin Oncol 1998; 16: 1655–60. Overgaard M, Jensen M-B, Overgaard J, et al. Postoperative radiotherapy in high-risk postmenopausal breast-cancer patients given adjuvant tamoxifen: Danish Breast Cancer Cooperative Group DBCG 82c randomised trial. Lancet 1999; 353: 1641–8. Larson D, Weinstein M, Goldberg I. Edema of the arm as a function of the extent of axillary surgery in patients with stage I-II carcinoma of the breast treated with primary radiotherapy. Int J Radiat Oncol Biol Phys 1986; 12: 1575–82. Kissin MW, Querci della Rovere G, Easton D, Westbury G. Risk of lymphoedema following the treatment of breast cancer. Br J Surg 1986; 73: 580–4. Delouche G, Bachelot F, Premont M, Kurtz JM. Conservation treatment of early breast cancer: long-term results and complications. Int J Radiat Oncol Biol Phys 1987; 13: 29–34. Senofsky GM, Moffat FL Jr, Davis K, et al. Total axillary lymphadenectomy in the management of breast cancer. Arch Surg 1991; 126: 1336–41. Halverson KJ, Taylor ME, Perez CA, et al. Regional nodal management and patterns of failure following conservative surgery and radiation therapy for stage I and II breast cancer. Int J Radiat Oncol Biol Phys 1993; 26: 593–9. ¨ rebro Breast Cancer Study Liljegren G, Holmberg L, The Uppsala-O Group. Arm morbidity after sector resection and axillary dissection with or without postoperative radiotherapy in breast cancer stage ¨ rebro Breast Cancer I. Results from a randomised trial. Uppsala-O Study Group. Eur J Cancer 1997; 33: 315–20.

Accepted for publication 11 January 2001