were included. Of these patients, 717 were treated conservatively (lumpectomy and breast irradiation) and 1289 were treated with total mastectomy. All patients ...
Annals of Oncology 13: 1404–1413, 2002 DOI: 10.1093/annonc/mdf227
Original article
Predictive factors for local recurrence in 2006 patients with surgically resected small breast cancer R. Arriagada1,2*, M. G. Lê3, G. Contesso1, J. M. Guinebretière1, F. Rochard1 & M. Spielmann1 1
Institut Gustave-Roussy (IGR), Villejuif; 3Institut National de la Santé et de la Recherche Médicale (INSERM), Villejuif, France; 2Instituto de Radiomedicina (IRAM), Vitacura, Santiago, Chile Received 9 July 2001; revised 15 January 2002; accepted 11 February 2002
Background: Analyses of predictive factors for local recurrences are important, as an increasing number of patients with early breast cancer opt for a breast-conserving procedure. This study investigates whether factors predictive of local recurrence differ between patients treated with conservative or radical surgery. Patients and methods: Two thousands and six patients with invasive breast carcinoma (≤25 mm) were included. Of these patients, 717 were treated conservatively (lumpectomy and breast irradiation) and 1289 were treated with total mastectomy. All patients had axillary dissection and received lymph node irradiation if axillary nodes were positive. Most patients did not receive adjuvant chemotherapy or additive hormonal treatments. The mean duration of follow-up was 20 years. The main end point was the total local recurrence rate. The risk factors of local recurrence were estimated by multivariate analyses and interaction tests were used for intergroup comparisons. Results: Statistically significant predictive factors for mastectomized patients were histological grade, extensive axillary node involvement (10 nodes or more), and inner quadrant tumors, which were of borderline significance. Young age, however, was not a prognostic indicator for local recurrence. The main statistically significant factor for patients treated with a conservative approach was young age (≤40 years). These younger patients had a five-fold increased risk of developing a breast recurrence compared with patients older than 60 years. Conclusions: Younger patients with early breast cancer treated with breast-conserving surgery should in particular be followed up at regular intervals so that any sign of local failure can be diagnosed early. Key words: breast cancer, breast-conserving surgery, local recurrence, mastectomy, multivariate analysis, prognostic factors
Introduction Several publications have analyzed prognostic factors potentially related to the risk of local recurrence (LR) in early breast cancer [1–17]. By restricting their analysis to breastconserving surgery (BCS) series, some authors recently reported that the main factors significantly related to the risk of LR were young age [2, 6, 18–19], a high histological grade [11, 15, 18, 20] and an extensive intraductal component [7, 21–22]; and surgical margins [8, 11, 23–28] and local radiotherapy were treatment factors [29–36]. However, it is not clear whether prognostic factors for LR are similar for patients treated with BCS and those treated with total mastectomy. For instance, axillary lymph node involvement, which is a
powerful prognostic factor in non-irradiated mastectomy series [37–41], does not appear to have a significant local impact in BCS series [15–16]. A recent study on substantial randomized data concluded that age is a prognostic factor for LR in patients treated with BCS and that vascular invasion is prognostic for both BCS and mastectomy [17]. To search for additional information on this subject, we compared the factors related to the risk of LR in patients treated with a different surgical approach, in a series of operable breast cancer that originated from the Institut GustaveRoussy (IGR) database. To avoid biases related to tumor size, we limited the analysis to patients whose tumor was ≤25 mm.
Patients and methods Patients *Correspondence to: Dr R. Arriagada, Instituto de Radiomedicina (IRAM), Américo Vespucio Norte 1314, Vitacura, Santiago, Chile. Tel: +562-754-17-00; Fax: +562-207-64-81; E-mail: datacenter @gocchi.cl © 2002 European Society for Medical Oncology
From a cohort of 7711 patients with invasive breast carcinoma treated at the IGR between 1954 and 1983, we abstracted 2006 patients with the following characteristics: (i) age 61 years of age. However, the risk of LR was not significantly modified by age in patients treated by mastectomy (P = 0.30). The difference between the treatment groups with respect to the effect of young age is clearly shown by the interaction test, which was highly significant (P = 10–6). In addition, the effect of age in the BCS group was analyzed, taking into account seven age categories rather than only four. Results adjusted for all other significant factors are shown in
Figure 1, with observed and theoretical values calculated by the model. There were also small differences for two other factors, macroscopic tumor size (P = 0.06) and histological grade (P = 0.10), but they were of borderline significance as shown in Table 4. The relationship between the risk of LR and the location of the tumor, or the number of examined or involved lymph nodes was similar for both surgical approaches (P = 0.87, 0.65 and 0.20, respectively), and results shown include both surgical groups. It is, however, noteworthy that patients with extensive lymph node involvement (10 or more nodes) had a greater risk of LR despite postoperative radiotherapy.
Discussion Studies evaluating factors predictive of cancer recurrence are important as they provide information on how the disease evolves after primary treatment. Such information may modify conventional treatment attitudes and this is why it is
1407 Table 2. Multivariate analyses: risk of local recurrence according to five factors in 1074 patients with a macroscopic tumor size ≤25 mm and negative axillary nodes Factors
Breast-conserving surgery (n = 452)
Mastectomy (n = 622)
n
RR (95% CI)
n
RR (95% CI)
61–75
106
1a
218
1a
51–60
133
1.4 (0.4–4.4)
175
0.3 (0.1–0.7)
41–50
147
1.6 (0.5–4.9)
176
0.6 (0.3–1.2)
66
2.8 (0.9–9.1)
53
0.9 (0.4–2.2)
P
P
Age (years)
≤40
0.07
0.29
Inner quadrant tumors No
348
1a
Yes
104
2.3 (1.2–4.6)
≤15 mm
233
1a
16–25
219
0.8 (0.4–1.6)
1
156
1a
2 and 3
296
1.5 (0.6–3.4)
113
1a
11–15
197
1.4 (0.5–4.0)
16–20
103
1.3 (0.4–3.9)
39
1.3 (0.3–5.2)
0.01
460
1a
162
0.8 (0.4–1.5)
248
1a
374
1.6 (0.9–2.9)
182
1a
440
2.1 (1.0–4.3)
0.58
Macroscopic tumor size
0.51
0.10
SBR grade
0.28
0.04
No. of axillary nodes examined ≤10
≥21
90
0.85
1a
205
2.5 (0.8–7.1)
186
1.8 (0.6–5.5)
141
1.7 (0.5–5.6)
0.99
a
Reference category. RR, relative risk of local recurrence in a multivariate analysis taking into account all factors included in the table; CI, confidence interval; SBR, Scarff–Bloom–Richardson.
critical that potential biases related to this type of analysis are reduced to a minimum [49–54]. In the present study, designed to identify differences in predictive factors for LR between patients treated with BCS and those treated with total mastectomy, a number of precautions were taken in the planned analyses. These precautions can be divided into those related to patient selection and those related to the analysis itself. Those relating to the former were: (i) the inclusion of patients with tumors of macroscopic size ≤25 mm (this tumor size cutoff was determinant in the decision for BCS or mastectomy at the IGR; BCS was initiated at this center in the early 1970s [11]); (ii) all patients had systematic axillary dissection; and (iii) all patients with N+ disease received lymph node irradiation. Precautions related to the analysis comprised: (i) defining LR as a total event rate, i.e. no censoring of other tumor events (this definition limits the risk of biases related to competing events [55–59]); (ii) systematically using multivariate analyses to identify the independent predictive effect of each factor; and (iii) avoiding false-positive results in subgroup analyses [60]. Interaction tests were applied for two major variables: axillary lymph node involvement and the surgical treatment. When interaction tests were non-significant, pooled
analyses were conducted (N– and N+ patients in Tables 4 and 5), and when they were significant (or of borderline significance), separate analyses were performed (type of surgical treatment according to age and two histological features in Table 5). These precautions do not totally eliminate the likelihood that results may be serendipitously misinterpreted, but they were taken to avoid interpreting errors as much as possible. Our findings are summarized and discussed according to each factor analyzed. The cutoff at 40 years, used to define young age, took into account the number of patients available (248 in total, 12% of the total population). The increased risk of LR (P = 0.07) for patients treated with BCS was demonstrated to be of borderline significance according to the trend tests for N– (Table 2) and N+ patients (Table 3). The possible lack of power in the statistical analysis of these subgroup analyses was confirmed by the highly significant trend test (P = 10–4) for the total number of BCS cases (717) as shown in Table 4. This effect of young age was totally different to that found in patients treated with mastectomy according to the test presented in Table 5, where both surgical approaches were compared directly (interaction test with a P value of 10–6).
1408 Table 3. Multivariate analyses: risk of local recurrence according to six factors in 932 patients with a macroscopic tumor size ≤25 mm and positive axillary nodes Factors
Breast-conserving surgery (n = 265)
Mastectomy + RT (n = 667)
n
n
RR (95% CI)
190
1a
197
1.2 (0.5–2.6)
195
1.0 (0.5–2.3)
85
1.3 (0.5–3.3)
RR (95% CI)
P
P
Age (years) 61–75
44
1a
51–60
89
1.4 (0.4–4.4)
41–50
88
1.6 (0.5–4.8)
≤40
44
2.9 (0.9–9.4)
0.07
0.69
Inner quadrant tumors 1a
No
212
Yes
53
2.5 (1.2–5.1)
89
1a
0.02
533
1a
134
2.2 (1.2–4.0)
164
1a
503
1.7 (0.7–3.8)
119
1a
548
2.7 (0.8–9.0)
0.01
Macroscopic tumor size (mm) ≤15 16–25
176
0.8 (0.4–1.4)
0.55
0.23
SBR grade 1 2 and 3
52 213
1a 1.5 (0.6–3.5)
0.29
0.09
No. of axillary nodes examined ≤10
31
1a
11–15
122
1.3 (0.4–3.9)
16–20
77
1.2 (0.4–3.7)
≥21
35
65
1a
198
0.8 (0.3–2.3)
201
0.7 (0.2–2.0)
1.3 (0.3–5.3)
203
0.6 (0.2–1.7)
1a
402
1a
199
1.9 (1.0–3.6)
66
4.2 (1.8–9.9)
0.77
0.16
No. of positive axillary nodes 1–3
204
4–9
46
1.6 (0.7–3.3)
≥10
15
0.7 (0.2–3.4)
0.68
0.0006
a
Reference category. RR, relative risk of local recurrence in a multivariate analysis taking into account all factors included in the table; RT, postoperative radiotherapy; CI, confidence interval; SBR, Scarff–Bloom–Richardson.
The effect of young age on the risk of LR for patients treated with BCS has been recognized in most studies on prognostic factors [2, 6, 15, 18, 19, 61]. The recent study by Voogd et al. [17], conducted in a randomized population of 1772 patients (879 treated with BCS and 893 with mastectomy), corroborated this fact. Two differences should be pointed out between this analysis and our own. The first is that Voogd et al. used separate tests for each age group whereas we used a trend test for all age groups. This latter analysis allowed us to test the whole series and not make assumptions about an age cutoff value. Indeed, both studies clearly indicate that there is no obvious age cutoff, and that a young age is rather related to a continuous trend towards a higher risk of LR. This is clearly observed in Figure 1, where there is only a decreased effect of age after 50 years. This change in the trend could be related to the menopausal status. The second difference is that Voogd et al. did not directly compare patients treated with BCS with those treated with mastectomy (despite the fact that data on
this specific treatment factor were randomized). In our study, we applied interaction tests to decrease the likelihood of biases related to the comparison of non-randomized data, which obviously attract criticism. The extremely low P value yielded by such a test in our series allows us to predict that similar results would have been obtained with the DanishEORTC (European Organization for Research and Treatment of Cancer) data. Few series have taken tumor location into account [5, 62, 63]. We included this factor as it was registered systematically in the IGR database. As shown in Tables 3–4, inner quadrant tumors were found to exhibit a trend towards a higher risk of LR. There was no difference between the surgical approaches, as shown in Table 5, using a non-significant interaction test (P = 0.87). Fowble et al. and Lohrisch et al. [62, 63] found no relationship between tumor location and local recurrence in their series.
1409 Macroscopic tumor size (pT) was also taken into account. Due to the initial patient selection, variation in tumor sizes was very limited. This factor was not predictive of the risk of LR for lesions measuring up to 25 mm in the BCS group. The effect of larger tumors (16–25 mm) in the mastectomy group was of borderline significance and it should probably be rejected, since the P value of the interaction test was also low (P = 0.06). Furthermore, in larger series of mastectomized patients in the same IGR database, which included all tumor sizes, we found that tumor size had no effect on local recurrence [64]. However, for patients treated with BCS it would be of interest to continue investigations into the potential effect of tumor size. These studies should be conducted in randomized trials evaluating BCS and including patients with a tumor >3 cm [30, 65]. Incidentally, the report by Voogd et al. [17] showed a non-significant trend towards a higher risk of LR for patients with tumors >1 cm in diameter.
As shown in Table 4, SBR grade was a predictive factor for LR in mastectomized patients, as we and others have reported elsewhere [64, 66]. However, there is no demonstrated difference between BCS and mastectomy groups as shown in Table 5, where the interaction test was not significant (P = 0.10). In the trials reported by Voogd et al. [17], histological grade was of borderline significance in the BCS group (P = 0.07) when other factors such as vascular invasion were taken into account. A general effect may exist but very large numbers would be required to attain statistical significance. The number of axillary lymph nodes examined was not predictive of LR in any of the analyses, and its prognostic value should be rejected in patients with small tumors in whom the risk of having positive axillary nodes was low. The number of positive axillary nodes was not a predictive factor for LR in the BCS group with N+ patients, but it was a highly significant factor for mastectomized patients (Table 3;
Table 4. Multivariate analysis: risk of local recurrence according to six factors in 2006 patients with a macroscopic tumor size ≤25 mm Factors
Breast-conserving surgery (n = 717)
Mastectomy (n = 1289)
n
RR (95% CI)
n
RR (95% CI)
150
1a
408
1a
P
P
Age (years) 61–75 51–60
222
1.2 (0.6–2.6)
41–50
235
2.1 (1.1–4.0)
372
0.6 (0.3–1.0)
371
0.7 (0.4–1.2)
≤40
110
4.4 (2.2–8.7)
138
1.0 (0.5–1.9)
No
560
1a
993
1a
Yes
157
1.4 (0.9–2.3)
296
1.3 (0.9–2.1)
≤15
322
1a
412
1a
16–25
395
1.1 (0.8–1.7)
877
1.6 (1.0–2.6)
1
208
1a
301
1a
2 and 3
509
1.3 (0.9–2.1)
988
2.2 (1.2–4.1)
≤10
144
1a
155
1a
11–15
319
1.3 (0.7–2.2)
16–20
180
0.8 (0.4–1.5)
74
10–4
0.65
Inner quadrant tumors
0.11
0.18
Macroscopic tumor size (mm)
0.54
0.06
SBR grade
0.22
0.009
No. of axillary nodes examined
≥21
403
1.5 (0.7–3.1)
387
1.3 (0.6–2.7)
1.2 (0.6–2.6)
344
1.0 (0.5–2.3)
0.72
0.70
No. of positive axillary nodes 0
452
1a
622
1a
1–3
204
1.2 (0.8–1.8)
402
0.6 (0.3–0.9)
4–9
46
1.9 (0.9–4.0)
199
1.0 (0.6–1.8)
≥10
15
1.3 (0.3–5.3)
66
2.2 (1.0–4.7)
0.15
a Reference category. RR, relative risk of local recurrence in a multivariate analysis taking into account all factors included in the table; CI, confidence interval; SBR, Scarff–Bloom–Richardson.
0.41
1410 Table 5. Risk of local recurrence in 2006 patients with a macroscopic tumor size ≤25 mm: multivariate analysis comparing predictive factors according to surgical treatment when interaction tests had a P value of ≤0.10 n
Factors
RR (95% CI)
P value (Wald test)
P value (interactiona)
Age (years) Breast-conserving surgery 61–75
150
1b
51–60
222
1.5 (0.8–3.0)
41–50
235
2.7 (1.5–4.7)
≤40
110
5.3 (2.9–9.7)
10–6
10–6
Mastectomy 61–75
408
1b
51–60
372
0.5 (0.3–0.9)
41–50
371
0.6 (0.4–1.0)
≤40
138
0.9 (0.5–1.7)
≤15
322
1b
16–25
395
1.2 (0.8–1.7)
≤15
412
1b
16–25
877
1.5 (0.9–2.3)
1
208
1b
2 and 3
509
1.4 (0.9–2.1)
1
301
1b
2 and 3
988
2.0 (1.2–3.4)
0.30
Macroscopic tumor size (mm) Breast-conserving surgery
0.40
Mastectomy
0.06
0.07
SBR grade Breast-conserving surgery
0.13
Mastectomy
0.10
0.006
Inner quadrant tumorsc 1b
No
1553
Yes
453
1.3 (1.0–1.8)
≤10
299
1b
11–15
722
1.3 (0.9–2.0)
16–20
567
1.0 (0.6–1.6)
≥21
418
1.0 (0.6–1.7)
0.07
0.87
No. of axillary nodes examinedc
No. of positive axillary nodes 0
a
0.34 0.65
c
1074
1b
1–3
606
0.8 (0.6–1.1)
0.25
4–9
245
1.3 (0.8–2.1)
0.24
≥10
81
2.3 (1.2–4.5)
0.008
0.20
Interaction test comparing the two surgical approaches. Reference category. c All surgical treatments included. RR, relative risk of local recurrence in a multivariate analysis taking into account all factors included in the table; CI, confidence interval. b
1411 [35, 37–40, 72–78]. However, as systemic treatments also have a positive impact on local control [79, 80] it can be hypothesized that the current indications for these treatments in intermediate- and high-risk patients could lower the effect of age on tumor recurrence, as observed in the present series. It may prove increasingly important to analyze the long-term results (20 years or more) of all randomized trials comparing systemic treatment with no systemic treatment in patients having undergone BCS. At the present time and given the consistency of the results yielded by different studies on prognostic factors, surgeons and oncologists should be aware of the increased risk of local recurrence in younger patients with early breast cancer treated with BCS. In particular, these patients should be followed up on a regular basis for several years so that any sign of local failure can be diagnosed early.
Acknowledgements Figure 1. Relative risk of breast recurrence in 717 patients treated with breast-conserving surgery. Analysis was performed using seven age categories (≤35, 36–40, 41–45, 46–50, 51–55, 56–60 and 61–75 years). Results are adjusted for other prognostic factors. Younger patients had a higher risk of breast recurrence. The effect of age tends to disappear after 50 years. Observed values are similar to theoretical values calculated by the model assuming a continuous effect of age. Ninety-five percent confidence intervals are given for each observed value.
P = 0.0006). This effect has been reported previously [64]. However, this effect disappeared when N+ patients were pooled with N– patients (Table 4), and only extensive axillary involvement (10 or more nodes) maintained a prognostic value as seen in Tables 4 and 5. It is noteworthy that all N+ patients received loco-regional radiotherapy and this treatment could have canceled the prognostic effect of the number of axillary nodes involved. A similar effect was described in the report by Voogd et al. [17]. Other factors such as extensive intraductal carcinoma [7, 21–22] and surgical margins [8, 11, 23–28] were not analyzed as they were not systematically coded in the database. However, the treatment policy at the IGR was to perform total mastectomy when the first factor was present and to request adequate surgical margins, i.e. at least 1 cm of macroscopically normal breast tissue surrounding the tumor when BCS was performed. In summary, between the BCS and the mastectomy groups, the overriding differential effect predicting the risk of LR was young age. Conservatively treated patients aged 40 years or younger had a five-fold greater risk of LR compared with patients older than 60 years of age who had a baseline risk of LR of 6% at 10 years. This effect was not observed among mastectomized patients. Young age has also been described as a significant prognostic factor for overall survival [67–71]. That local recurrence may have an impact on distant metastasis and overall survival is also becoming widely recognized
The authors thank Ms Lorna Saint-Ange for editing of the manuscript.
References 1. Recht A, Silen W, Schnitt SJ et al. Time-course of local recurrence following conservative surgery and radiotherapy for early stage breast cancer. Int J Radiat Oncol Biol Phys 1988; 15: 255–262. 2. Fourquet A, Campana F, Zafrani B et al. Prognostic factors of breast recurrence in the conservative management of early breast cancer: a 25-year follow-up. Int J Radiat Oncol Biol Phys 1989; 17: 719–725. 3. Kurtz JM, Amalric R, Brandone H et al. Local recurrence after breastconserving surgery and radiotherapy: frequency, time course, and prognosis. Cancer 1989; 63: 1912–1917. 4. Haffty BG, Fisher D, Rose M et al. Prognostic factors for local recurrence in the conservatively treated breast cancer patient: cautious interpretation of the data. J Clin Oncol 1991; 9: 997–1003. 5. Kurtz JM. Factors influencing the risk of local recurrence in the breast. Eur J Cancer 1992; 28: 660–666. 6. Nixon AJ, Schnitt S, Connolly JL et al. Relationship of patient age to pathologic features of the tumor and the risk of local recurrence for patients with stage I or II breast cancer treated with conservative surgery and radiation therapy. Int J Radiat Oncol Biol Phys 1992; 24: 221–232. 7. Van Dongen JA, Bartelink H, Fentiman IS et al. Factors influencing local relapse and survival and results of salvage treatment after breast-conserving therapy in operable breast cancer: EORTC trial 10801, breast conservation compared with mastectomy in TNM stage I and II breast cancer. Eur J Cancer 1992; 28A: 801–805. 8. Borger J, Kemperman H, Hart A et al. Risk factors in breast-conservation therapy. J Clin Oncol 1994; 12: 653–660. 9. Harris JR, Gelman R. What have we learned about risk factors for local recurrence after breast-conserving surgery and irradiation. J Clin Oncol 1994; 12: 647–649. 10. Korzeniowski S, Dyba T, Skoyszewski J. Classical prognostic factors for survival and loco-regional control in breast cancer patients treated with radical mastectomy alone. Acta Oncol 1994; 33: 459–465. 11. Dewar JA, Arriagada R, Benhamou S et al. Local relapse and contralateral tumor rates in patients with breast cancer treated with con-
1412
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
servative surgery and radiotherapy (Institut Gustave-Roussy 1970– 1982). IGR Breast Cancer Group. Cancer 1995; 76: 2260–2265. MacMillan RD, Purushotham AD, Cordiner C et al. Predicting local recurrence by correlating pre-operative mammographic findings with pathological risk factors in patients with breast cancer. Br J Radiol 1995; 68: 445–449. Magee B, Swindell R, Harris M, Banejee SS. Prognostic factors for breast recurrence after conservative breast surgery and radiotherapy: results from a randomised trial. Radiother Oncol 1996; 39: 223–227. Liljegren G, Lindgren A, Bergh J et al. Risk factors for local recurrence after conservative treatment in stage I breast cancer. Definition of a subgroup not requiring radiotherapy. Ann Oncol 1997; 8: 235–241. Arriagada R, Goset K, Koscielny S. Patterns of failure in small breast cancers treated by conservative surgery and radiotherapy at the Institut Gustave-Roussy. In Bland KI, Copeland EM (eds): The Breast. Philadelphia, PA: W. B. Saunders 1998; 1183–1190. Voogd AC, Peterse JL, Crommelin MA et al. Histological determinants for different types of local recurrence after breast-conserving therapy of invasive breast cancer. Eur J Cancer 1999; 35: 1828–1237. Voogd AC, Nielsen M, Peterse JL et al. Differences in risk factors for local and distant recurrence after breast-conserving therapy or mastectomy for stage I and II breast cancer: pooled results of two large European randomized trials. J Clin Oncol 2001; 19: 1688–1697. Kurtz JM, Jacquemier J, Amalric R et al. Why are local recurrences after breast-conserving therapy more frequent in younger patients? J Clin Oncol 1990; 8: 591–598. Nixon AJ, Neuberg D, Hayes EF et al. Relationship of patient age to pathologic features of the tumor and prognosis for patients with stage I or II breast cancer. J Clin Oncol 1994; 12: 888–894. Nixon AJ, Schnitt SJ, Gelman R et al. Ralationship of tumor grade to other pathologic features and to treatment outcome of patients with early stage breast carcinoma treated with breast-conserving therapy. Cancer 1996; 78: 1426–1431. Hurd TC, Sneige N, Allen PK et al. Impact of extensive intraductal component on recurrence and survival in patients with stage I or II breast cancer treated with breast conservation therapy. Ann Surg Oncol 1997; 4: 119–124. Schnitt SJ, Connolly JL, Harris JR et al. Pathologic predictors of early local recurrence in Stage I and II breast cancer treated by primary radiation therapy. Cancer 1984; 53: 1049–1057. Anscher MS, Jones P, Prosnitz LR et al. Local failure and margin status in early-stage breast carcinoma treated with conservation surgery and radiation therapy. Ann Surg 1993; 218: 22–28. Pittinger TP, Maronian NC, Poulter CA, Peacock JL. Importance of margins status in outcome of breast-conserving surgery for carcinoma. Surgery 1994; 116: 605–609. Schnitt SJ, Abner A, Gelman R et al. The relationship between microscopic margins of resection and the risk of local recurrence in patients with breast cancer treated with breast-conserving surgery and radiation therapy. Cancer 1994; 74: 1746–1751. Smitt MC, Nowels JW, Zdeblich MJ et al. The importance of the lumpectomy surgical margins status in long-term results of breast conservation. Cancer 1995; 76: 259–267. Wazer DE, Schmidt-Ulrich RK, Ruthazer R et al. Factors determining outcome for breast-conserving irradiation with margin-directed dose escalation to the tumor bed. Int J Radiat Oncol Biol Phys 1998; 40: 851–858. Peterson ME, Schultz DJ, Reynolds C et al. Outcomes in breast cancer patients relative to margin status after treatment with breast-
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39. 40.
41.
42.
43.
44.
conserving surgery and radiation therapy: The University of Pennsylvania experience. Int J Radiat Oncol Biol Phys 1999; 43: 1029–1035. Early Breast Cancer Trialists’ Collaborative Group. Effects of radiotherapy and surgery in early breast cancer. An overview of the randomized trials. N Engl J Med 1995; 333: 1444–1455. Fisher B, Anderson S, Redmond CK et al. Reanalysis and results after 12 years of follow-up in a randomized clinical trial comparing total mastectomy with lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 1995; 333: 1456–1461. Clark RM, Whelan T, Levine M et al. Randomized clinical trial of breast irradiation following lumpectomy and axillary dissection for node-negative breast cancer: an update. Ontario Clinical Oncology Group. J Natl Cancer Inst 1996; 88: 1659–1664. Forrest AP, Stewart HJ, Everington D et al. Randomised controlled trial of conservation therapy for breast cancer: 6-year analysis of the Scottish Cancer Trials Breast Group. Lancet 1996; 348: 708–713. Renton SC, Gazet JC, Ford HT et al. The importance of the resection margin in conservative surgery for breast cancer. Eur J Surg Oncol 1996; 22: 17–22. Liljegren G, Holmberg L, Bergh J et al. and the Uppsala-Orebro Breast Cancer Group. 10-year results after sector resection with or without postoperative radiotherapy for stage I breast cancer: a randomized trial. J Clin Oncol 1999; 17: 2326–2333. 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: 1757–1770. Holli K, Saaristo R, Isola J et al. Lumpectomy with or without postoperative radiotherapy for breast cancer with favourable prognostic features: results of a randomized study. Br J Cancer 2001; 84: 164–169. Arriagada R, Rutqvist LE, Mattsson A et al. Adequate locoregional treatment for early breast cancer may prevent secondary dissemination. J Clin Oncol 1995; 13: 2869–2878. Overgaard M, Hansen PS, Overgaard J et al. for the Danish Breast Cancer Cooperative Group 82b Trial. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who received adjuvant chemotherapy. N Engl J Med 1997; 337: 949–955. Arriagada R, Rutqvist LE, Le MG. Post-mastectomy radiotherapy: the randomized trials. Semin Radiat Oncol 1999; 9: 275–286. Overgaard M, Jensen MB, 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: 1641–1648. Recht A, Gray R, Davidson NE et al. Locoregional failure 10 years after mastectomy and adjuvant chemotherapy with or without tamoxifen without irradiation: experience of the Eastern Cooperative Oncology Group. J Clin Oncol 1999; 17: 1689–1700. Contesso G, Mouriesse H, Friedman S et al. The importance of histologic grade in long-term prognosis of breast cancer: a study of 1,010 patients, uniformly treated at the Institut Gustave-Roussy. J Clin Oncol 1987; 5: 1378–1386. Friedman S, Bertin F, Mouriesse H et al. Importance of tumor cells in axillary node sinus margins (‘clandestine’ metastases) discovered by serial sectioning in operable breast carcinoma. Acta Oncol 1988; 27: 483–487. Sarrazin D, Lê MG, Fontaine MF et al. Conservative treatment versus mastectomy in T1 or small T2 breast cancer—a randomized clinical trial. In Harris JR, Hellman S, Silen W (eds): Conservative Management of Breast Cancer. Philadelphia, PA: Lippincott 1983; 101–111.
1413 45. Cox JD. Regression models and life tables. J R Stat Soc 1972; 34: 187–220. 46. Bonetti M, Gelber RD. A graphical method to assess treatment– covariate interactions using the Cox model on subsets of the data. Stat Med 2000; 19: 2595–2609. 47. Russek-Cohen E, Simon RM. Evaluating treatments when a gender by treatment interaction may exist. Stat Med 1997; 16: 455–464. 48. Benhamou S, Reinikainen M, Bouchardy C et al. Association between lung cancer and microsomal epoxide hydrolase genotypes. Cancer Res 1998; 58: 5291–5293. 49. Armitage P, Gehan EA. Statistical methods for the identification and use of prognostic factors. Int J Cancer 1974; 13: 16–36. 50. Harrell FE, Lee KL, Califf RM et al. Regression modelling strategies for improved prognostic prediction. Stat Med 1984; 3: 143–152. 51. Simon R, Altman DG. Statistical aspects of prognostic factor studies in oncology. Br J Cancer 1994; 69: 979–985. 52. Clark GM, Hilsenbeck SG, Radvin PM et al. Prognostic factors: rationale and methods of analysis and integration. Breast Cancer Res Treat 1994; 32: 105–112. 53. Mazumdar M, Glassman JR. Categorizing a prognostic variable: review of methods, code for easy implementation and applications decision-making about cancer treatments. Stat Med 2000; 19: 113–132. 54. Klinger A, Dannegger F, Ulm K. Identifying and modelling prognostic factors with censored data. Stat Med 2000; 19: 601–615. 55. Gelman R, Gelber R, Henderson IC et al. Improved methodology for analyzing local and distant recurrence. J Clin Oncol 1990; 8: 548–555. 56. Kramar A, Arriagada R. Analyzing local and distant recurrence (letter). J Clin Oncol 1990; 8: 2086–2087. 57. Arriagada R, Rutqvist LE, Kramar A, Johansson H. Competing risks determining event-free survival in early breast cancer. Br J Cancer 1992; 66: 951–957. 58. Veronesi U, Marubini E, Del Vecchio M et al. Local recurrences and distant metastases after conservative breast cancer treatments: partly independent events. J Natl Cancer Inst 1995; 87: 19–27. 59. Chapman JW, Fish EB, Link MA. Competing risk analyses for recurrence from primary breast cancer. Br J Cancer 1999; 79: 1508–1513. 60. Yusuf S, Wittes J, Probstfield J, Tyroler HA. Analysis and interpretation of treatment effects in subgroups of patients in randomized clinical trials. JAMA 1991; 266: 93–98. 61. Fowble BL, Schultz DJ, Overmoyer B et al. The influence of young age on outcome in early stage breast cancer. Int J Radiat Oncol 1994; 30: 23–33. 62. Fowble B, Solin LJ, Schultz DJ et al. Breast recurrence and survival related to primary tumor location in patients undergoing conservative surgery and radiation for early-stage breast cancer. Int J Radiat Oncol Biol Phys 1992; 23: 933–939. 63. Lohrisch C, Jackson J, Jones A et al. Relationship between tumor location and relapse in 6,781 women with early invasive breast cancer. J Clin Oncol 2000; 18: 2828–2835. 64. Arriagada R, Le MG. Adjuvant radiotherapy in breast cancer. The treatment of lymph node areas. Acta Oncol 2000; 39: 295–305.
65. Van Dongen JA, Voogd AC, Fentiman IS et al. Long-term results of a randomized trial comparing breast-conserving therapy with mastectomy: European Organization for Research and Treatment of Cancer 10801 trial. J Natl Cancer Inst 2000; 92: 1143–1150. 66. Rosen HR, Stierer M, Weber R. Nuclear polymorphism—a prognostic parameter to evaluate local recurrence of female breast cancer. Breast Cancer Res Treat 1990; 17: 9–14. 67. Bonnier P, Romain S, Charpin C et al. Age as a prognostic factor in breast cancer: relationship to pathologic and biologic features. Int J Cancer 1995; 62: 138–144. 68. Rutqvist LE, Wallgren A. Influence of age on outcome in breast carcinoma. Acta Radiol Oncol 1983; 22: 289–294. 69. De la Rochefordière A, Asselain B, Campana F et al. Age as a prognostic factor in premenopausal breast carcinoma. Lancet 1993; 341: 1039–1043. 70. Høst H, Lund E. Age as a prognostic factor in breast cancer. Cancer 1986; 57: 2217–2221. 71. Kroman N, Jensen MB, Wolfahrt J et al. Factors influencing the effect of age on the prognosis of breast cancer: population based study. Br Med J 2000; 320: 474–478. 72. Haffty BG, Goldberg NB, Fisher D et al. Conservative surgery and radiation therapy in breast carcinoma: local recurrence and prognostic implications. Int J Radiat Oncol Biol Phys 1989; 17: 727–732. 73. Kemperman H, Borger J, Hart A et al. Prognostic factors for survival after breast conserving therapy for stage I and II breast cancer. The role of local recurrence. Eur J Cancer 1995; 31A: 690–698. 74. Curcio LD, Chu DZJ, Ahn C et al. Local recurrence in breast cancer: implications for systemic disease. Ann Surg Oncol 1997; 4: 24–27. 75. Van Tienhoven G, Voogd AC, Peterse HJ et al. Prognosis after treatment for loco-regional recurrence after mastectomy or breast conserving therapy in two randomised trials (EORTC 10801 and DBCG-82TM): EORTC Breast Cancer Cooperative Group and Danish Breast Cancer Cooperative Group. Eur J Cancer 1999; 35: 32–38. 76. 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: 956–962. 77. Fortin A, Larochelle M, Laverdière J et al. Local failure is responsible for the decrease in survival for patients with breast cancer treated with conservative surgery and postoperative radiotherapy. J Clin Oncol 1999; 17: 101–109. 78. Koscielny S, Tubiana M. The link between local recurrences and distant metastases in human breast cancer. Int J Radiat Oncol Biol Phys 1999; 43: 11–24. 79. Goldhirsch A, Gelber RD, Price KN et al. Effect of systemic adjuvant treatment on first sites of breast cancer relapse. Lancet 1994; 343: 377–381. 80. Fisher B, Costantino J, Redmond C et al. A randomized clinicl trial evaluating tamoxifen in the treatment of patients with node-negative breast cancer who have estrogen-receptor-positive tumors. N Engl J Med 1989; 320: 479–484.
