Optimizing surveillance mammography following breast conservation surgery

EJSO 2002; 28: 11–13 doi:10.1053/ejso.2001.1188, available online at http://www.idealibrary.com on

Optimizing surveillance mammography following breast conservation surgery S. C. Weight, R. Windle∗ and A. T. Stotter∗ St Marks Breast Centre, 10 St Marks Road, Remuera, Auckland, New Zealand; ∗Department of Surgery, Glenfield Hospital, Groby Road, Leicester, UK

Aims: Prompt detection and treatment of local recurrence (LR) following breast conservation surgery (BCT) may improve subsequent survival. Following early reports demonstrating increased LR in the first years after surgery, a practice of surveillance mammography starting 1 year from diagnosis has become established. Increasing use of adjuvant chemotherapy with adriamycin-containing regimens has resulted in radiotherapy being postponed, so that the first mammogram coincides with the acute radiotherapy reaction, resulting in patient discomfort and poor quality films. We wished to determine if the 1-year mammogram could safely be omitted. Methods: We reviewed 1151 consecutive patients treated with BCT for in situ, stage I or II disease over a 10-year period. All patients had clear resection margins and, where indicated, underwent axillary surgery and adjuvant treatment. This consisted of radiotherapy (40 Gy with a 5 Gy boost), chemotherapy and/or tamoxifen. Results: Overall, the 5-year actuarial rate of local recurrence was 4.8%. The cumulative risk of LR at 12 months was 0.3% (three patients) and 1.9% (20 patients) at 24 months. This included all cases of high-grade DCIS (>30 mm in size) and all but one tumour with a Nottingham Prognostic index (NPI) of >6.0 which recurred locally. Conclusion: Even including those patients generally accepted to be at high risk of LR, the cumulative risk of LR was only 0.3% at 12 months from surgery. We now therefore schedule routine biennial mammography from time of surgery except for those with high risk of early local recurrence such as extensive, high-grade in situ disease.  2001 Harcourt Publishers Ltd Key words: breast conserving surgery; local recurrence; radiotherapy; surveillance mammography.

INTRODUCTION Breast conservation therapy (BCT) as primary treatment for breast cancer is well established and, in the UK, actively encouraged for T1 tumours under the quality assurance (QA) guidelines. That BCT with adjuvant radiotherapy (RT) for early stage breast cancer provides survival rates comparable to mastectomy is beyond question1,2 although there is an accepted locoregional failure rate of approximately 1% per annum. Regular follow-up of patients after BCT is important therefore because the early detection and salvage of local recurrence can be rewarded with 5-year survival rates approaching 70%.3,4 The detection of locoregional failure, as with primary disease, is facilitated by a combination of clinical examination, mammography and ultrasound.5 As with primary lesions, those detected mammographically appear to be associated with a lower incidence of

Correspondence to: S. C. Weight, St Marks Breast Centre, 10 St Marks Road, Remuera, Auckland, New Zealand. E-mail: [email protected] Published online 18 October 2001. 0748–7983/02/010011+03 $35.00/0

metastatic disease.6 Whilst local protocols regarding follow-up naturally vary, the practice of annual mammography, particularly during the early surveillance period, appears commonplace, possibly as a result of historical data suggesting an increased rate of local recurrence in the first 2 years after surgery.7,8 The widening indications for adjuvant systemic therapy frequently result in a hiatus of up to 9 months between surgery and the completion of RT. This is particularly a problem with the increasing use of anthracycline-based chemotherapy regimens, which necessitate a delay in RT to minimize the tissue reaction. A mammogram 1 year from diagnosis may thus coincide with the acute radiotherapy reaction, preventing effective breast compression with a concomitant decrease in mammographic quality. Although newer imaging modalities, including MRI, are becoming increasingly important,9 mammography remains the gold standard. Therefore a reasonable period between the cessation of RT and imaging is desirable to minimize discomfort, allow post-irradiation changes8 to mature and minimize false positive results.10 This reflects the common practice of scheduling annual mammography  2001 Harcourt Publishers Ltd

12

METHODS From a single institutional database, a consecutive series of patients treated with primary BCT for DCIS and stage I or II invasive disease over a 10-year period (1989–1998) with at least 1 year of follow up (FU) were identified. This included symptomatic women and, from 1991, those seen as part of the National Breast Screening Programme. All patients underwent wide local excision, with either axillary sampling or level 1 dissection for invasive tumours, under the care of two surgeons (R.W. and A.S.). Selection for BCT was on a combination of patient preference and clinical indication (low tumour size to breast ratio, non-central location, uni-focal localized disease on imaging and no history of previous breast irradiation). Cases excluded from the dataset were those operated on elsewhere but followed up at the institution, phyllodes tumours, patients over 70 years of age at time of diagnosis (tamoxifen was used as first-line treatment) and those with recurrent disease undergoing further local surgery. Following surgery, patients in whom the margins were histologically close (<3 mm) underwent re-excision. Patients were referred for breast radiotherapy and received 40 Gy with a 5 Gy boost to the tumour bed, usually in 25 fractions. Adjuvant chemotherapy was dependent on the final histology, with the main indication being node-positive disease in pre-menopausal women and normally consisted of six cycles of CMF (cyclophosphamide, methotrexate and 5-folinic acid) as tolerated. More recent practice has seen expansion in the indications for adjuvant chemotherapy and an increase in the use of anthracycline-based chemotherapy regimens (e.g. adriamycin and cyclophosphamide (AC)). Both preand post-menopausal women were eligible for tamoxifen, 20 mg o.d. Adjuvant treatment was, in some cases, determined by patient entry into clinical trials. Patients were seen for follow-up at least annually, with surveillance mammograms at 1 and 2 years and biennial imaging thereafter. Local recurrence (LR), defined as cancer in the ipsilateral breast, was detected clinically, radiologically or via a combination of both modalities and confirmed with cytology or histology.

RESULTS There were 48 cases of LR (4.2%) from a total of 1151 patients treated with BCT over a 10-year period, with a median follow-up of 50 (95% CI 49.2–52.2) months (Table 1). The actuarial LR rate was 4.8% (n=36) at 5 years (Fig. 1). Overall three patients had LR at 12 months (cumulative risk 0.3%), two of which were detected

Table 1 Patient demographics and adjuvant treatment rates in the local recurrence and recurrence-free groups

Total, n (%) Actuarial 5-year local recurrence rate (%) Age, median (95% CI) Follow-up, median (95% CI) % radiotherapy % chemotherapy % tamoxifen

LR

Recurrence free

48 (4.2) 4.8

1103 (95.8)

52 (49.6–55.1) 62 (52.8–69.7)

55 (54.0–55.1) 48 (48.7–51.7)

80.4 19.2 58.7

90.3 18.7 75.8

100 Free of local recurrence (%)

from treatment cessation rather than diagnosis (see BASO guidelines). This study therefore evaluated if such a policy is prudent.

S. C. WEIGHT ET AL.

75

50

25 1151 0

24

579

421

261

56

48

60

72

96

Time from diagnosis (months)

Figure 1 Cumulative rate of local recurrence following wide excision for breast cancer or DCIS (Kaplan–Meier with SEM). The 5-year local recurrence rate was 4.8%. Numbers in italics refer to those patients at risk.

mammographically. By 24 months from diagnosis, a total of 20 patients had LR (cumulative risk 1.9%) and this group included all the cases of high-grade DCIS (>30 mm in size) which recurred locally and all but one of the cases with a Nottingham Prognostic index (NPI) of >6.0 which recurred locally. The LR patients were younger at diagnosis (P=ns, Mann–Whitney) and, probably as a consequence, were significantly less likely to have received adjuvant tamoxifen (P<0.05, Mann–Whitney) although there was no significant difference in the rates of adjuvant chemotherapy or radiotherapy. Overall, the actuarial survival was significantly worse in the LR group (P<0.01, log-rank test). At 5 years, the survival difference was 71.6% (LR) vs 86.8%, whilst by 10 years the gap was 64.4% (LR) vs 71.6% (Fig. 2).

DISCUSSION The local recurrence of breast cancer after conservation surgery, unlike post-mastectomy locoregional failure, is not a marker of systemic disease. It does nevertheless

OPTIMIZING SURVEILLANCE MAMMOGRAPHY

13

100

Survival (%)

75

50

25

0

48 1103

41 875 24

33 550

28 403

17 249

60

7 40 96

Time from diagnosis (months)

Figure 2 Overall cumulative survival from surgery in patients developing a local recurrence (–Ε–) and those remaining recurrence free (–Φ–) (P<0.01, log-rank test). Numbers in italics refer to those patients at risk (top figures=LR).

represent failure for the patient, in terms of both confidence and usually breast conservation. It may also be associated with a diminished actuarial survival rate from the time of recurrence4,11 although early detection and treatment can minimize this.6 Many of the factors associated with an increase risk of local recurrence after BCT have been delineated, including young age, the absence of post-operative radiotherapy,1 inadequate margins,12 extensive DCIS13 and multifocal disease.14 It was not the purpose of this study to re-examine these factors, rather to focus on whether biennial surveillance mammography was a reasonable policy given the demographics of local recurrence. Our policy of re-excision of the margins if clearance (invasive or in situ) were less than 3 mm has developed in tandem with improved standards in pathological reporting of the measured margins. Although many may consider this cut-off inadequate for confidence in the security of the margins, our data shows a LR rate of less than 1% p.a. which stands comparison with other published series. The overall rate of adjuvant RT is perhaps important with only 80% of the LR group treated. All but one of these patients had DCIS with secure margins, the other patient had an invasive ductal carcinoma (NPI 3.1) but declined RT. The apparent beneficial effect of tamoxifen in decreasing LR seen in our data may be artefactual, arising as a result of the age of the cohort treated with tamoxifen (who were almost entirely post-menopausal). These results demonstrate that only 0.26% of our patients treated with BCT developed local recurrence within 12 months of diagnosis, rising to just under 2% by 24 months. Patients with local recurrence had a significantly worse overall survival compared to those without LR (P<0.01) although it must be emphasized that this data was not matched and any difference may reflect,

at least in part, the different tumour characteristics (by definition) in each group. However, the comparable age and adjuvant chemotherapy demographics between the two groups suggests similar tumour prognostic indices. With the cumulative risk of LR being 0.26% at 1 year, it is difficult to justify routine annual mammograms on all patients after BCT. We have accordingly moved to a policy of biennial mammography from the time of diagnosis, although this is frequently little more than 1 year from the completion of adjuvant treatment. We hope this will minimize radiological false-positive rates. A degree of flexibility is essential, however, particularly when evaluating which patients are at high risk of early local recurrence, including those with extensive, highgrade DCIS and poor prognosis invasive tumours.

REFERENCES 1. Fischer B, Redmond C, Poisson R et al. Eight year results of a randomised clinical trial comparing total mastectomy and lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 1989; 320: 822–8. 2. Veronesi U, Salvadori B, Luini A et al. Conservative treatment of early breast cancer. Ann Surg 1990; 211: 250–9. 3. Kurtz J, Amalric R, Brandone H et al. Local recurrence after breastconserving surgery and radiotherapy. Cancer 1989; 63: 1912–7. 4. Stotter AT, Atkinson EN, Fairston BA et al. Survival following locoregional recurrence after breast conservation therapy for cancer. Ann Surg 1990; 212: 166–72. 5. Grosse A, Schreer I, Frischbier HJ, Maass H, Loening T, Bahnsen J. Results of breast conserving therapy for early breast cancer and the role of mammographic follow-up. Int J Radiat Oncol Biol Physics 1997; 38: 761–7. 6. Hassell PR, Olivotto IA, Mueller HA, Kingston GW, Basco VE. Early breast cancer: detection of recurrence after conservative surgery and radiation therapy. Radiology 1990; 176: 731–5. 7. Orel SG, Troupin RH, Patterson EA, Fowble BL. Breast cancer recurrence after lumpectomy and irradiation: role of mammography in detection. Radiology 1992; 183: 201–6. 8. Sardi A, Eckholdt G, McKinnon WM, Bolton JS. The significance of mammographic findings after breast-conserving therapy for carcinoma of the breast. Surg Gyn Obs 1991; 173: 309–12. 9. Kramer S, Schulz-Wendtland R, Hagedorn K, Bautz W, Lang N. Magnetic resonance imaging in the diagnosis of local recurrences in breast cancer. Anticancer Research 1998; 18(3C): 2159–61. 10. Stomper PC, Recht A, Berenberg AL, Jochelson MS, Harris JR. Mammographic detection of recurrent cancer in the irradiated breast. AJR 1987; 148: 39–43. 11. Early Breast Cancer Trialists’ Collaborative Group. Favourable and unfavourable effects on long-term survival of radiotherapy for early breast cancer: an overview of the randomised trials. Lancet 2000; 355: 1757–70. 12. Malik HZ, Purushotham AD, Mallon EA, George WD. Influence of tumour bed assessment on local recurrence following breastconserving surgery for breast cancer. Eur J Surg Oncol 1999; 25: 265–8. 13. Eberlein TJ, Connolly JL, Schnitt SJ et al. Predictors of local recurrence following conservative breast surgery and radiation therapy. Arch Surg 1990; 125: 771–7. 14. Voogd AC, van der Horst F, Crommelin MA et al. The relationship between findings on pre-treatment mammograms and local recurrence after breast conserving therapy for invasive breast cancer. Eur J Surg Oncol 1999; 25: 273–9. Accepted for publication 23 July 2001