The Importance of Routine Cavity Biopsy in Breast Conservation Surgery
Hewes JC, Imkampe A, Haji A, Bates T
Department of Breast Surgery, William Harvey Hospital, Ashford, Kent, UK
Abstract:
Introduction: The role of cavity biopsies (CB) at the time of wide local excision
(WLE) for primary breast cancer has not been fully evaluated. This study compared
four groups of patients who underwent surgery to determine the significance of
positive margins and CB on tumour characteristics and outcome.
Patients and Methods: A retrospective study of patients undergoing WLE and CB in
one institution over a 21 year period was carried out. Demographic data, tumour
characteristics and survival information were obtained. Four subgroups of patients
were compared according to their margin and cavity status (positive or negative).
Results: 957 patients underwent WLE and in 71% both margin and CB were tumour
negative. Median 10 year survival was 85.6% and breast cancer specific survival
(BCSS) 92.4%. Tumour size, grade, node and oestrogen receptor status were
independent predictors of survival. There was poor concordance between positive
resection margins and CB (32%), and a negative margin carried a 10.8% risk of
demonstrable residual disease. A positive CB but not a positive margin indicated
reduced overall and BCSS.
Conclusions: Cavity status was more significant with regards to survival than margin
status. CB is important in identifying residual and multifocal disease as margin and
cavity positivity are not concordant.
Keywords: Wide local excision, cavity biopsy, margins, survival
1
Introduction:
Breast conservation surgery (BCS) with wide local excision (WLE) is standard
practice in the treatment of patients with primary breast cancer and when combined
with adjuvant radiotherapy (RT) it has equivalent local recurrence (LR), disease-free
and overall survival (OS) rates when compared with mastectomy1-8.
The presence of unidentified tumour foci within the residual breast tissue however can
lead to local or systemic recurrence9. Margin positivity is predictive for residual
disease with rates of 29-66%9-11; however histologically negative margins do not
necessarily safeguard against residual disease, since LR rates of 7-27% following
BCS and RT in patients with initially negative margins have been described12-14.
Holland et al demonstrated further tumour foci in 20% of cases within 2cm of the
primary tumour and 43% of cases more than 2cm away15. They concluded that 7-9%
of patients would have had residual foci of invasive cancer left in the remaining breast
tissue if the patients with an initial invasive tumour size of less than 4cm had a
resection margin of 3-4cm.
Biopsies of the WLE cavity (CB) give additional information with regards to the
adequacy of the margins, and positive CB have been associated with a higher tumour
grade, extensive intraduct component, younger age and larger tumour diameter16.
Negative CB may render the overall final margin status histologically negative
thereby reducing the need for re-excision17. Cao et al however suggest that the lack of
concordance between cavity and margin status and the overall reduction in the need
for re-excision can be related to false positive margins18. This may be due to
2
inadequacy of the surgical procedure or poor specimen preparation - such as the
seepage of ink into crevices of the specimen, tumour friability promoting
displacement of tumour into the ink or the manipulation of specimens for radiographs.
It has been suggested that the application of routine tumour bed assessment with
selective re-excision may result in a lower LR rate19, although one study has shown an
increase in LR rates with no effect on OS in patients with positive CB20.
The role of CB has therefore yet to be fully established. This study was undertaken in
order to determine the margin and cavity status of patients within a large cohort
undergoing BCS. The relationship of these two factors to the tumour characteristics
and outcome was investigated to establish the value of routinely performing CB
during WLE.
3
Patients and Methods:
Patients:
A prospective computerised database was compiled that collected information on all
patients undergoing surgery for breast cancer in one institution (WHH, Ashford, UK)
from 1986 to December 2007 with near complete follow up. Data were entered at the
time of treatment. Retrospective data from consecutive patients who had undergone
WLE and CB for unilateral breast cancer over this 21 year period were then obtained.
Patients were included in the study if they had histological evidence of invasive
cancer within the main resection specimen. They were divided into one of four groups
according to whether or not the resection margin or CB from the initial operation were
infiltrated with tumour and the results then compared. The groups were labelled m+c+
(positive margins and CB), m+c- (positive margins, negative CB), m-c+ (negative
margins, positive CB) and m-c- (negative margins and CB).
Surgery:
The tumour was normally excised with a narrow overlying ellipse of skin and a
macroscopically clear margin down to fascia. Biopsies were then taken from the four
quadrants of the residual cavity, with the aim of sampling a wide area of the cut
surface with a minimum of tissue volume. Histological assessment was subsequently
undertaken by a dedicated breast pathologist. The margin was recorded as being
involved if tumour extended to the inked edge of the specimen, if it was clear by less
than 1mm or if the report was uncertain. Four lymph node axillary sampling was
performed routinely. This series documents the practice of two surgeons and in the
4
early years not all patients had salvage surgery for positive margins or cavity biopsies.
Patient choice was also a factor in variations from current practice.
Data collection:
Patient demographics and tumour characteristics included the size, histological type,
grade, oestrogen receptor (ER) status and the presence of vascular invasion (VI). The
nodal status of the patient was recorded and whether the tumour was multifocal or
unifocal. Multifocality was defined in the histological report if there was more than
one tumour focus within the specimen. The type of post-operative adjuvant therapy
(hormone manipulation, radiotherapy or chemotherapy) was noted. The timing and
nature of redo surgery and the pathological findings from these specimens were
documented. Disease recurrence was recorded as being local, to regional lymph nodes
or metastatic spread. Survival data were then obtained for each patient from the date
of surgery to the date of death or latest clinic or annual postal follow up. Overall
survival (OS) was recorded and breast cancer specific survival (BCSS) was calculated
when death was due to breast cancer disease progression. Survival data were censored
at 10 years.
Statistical analysis:
Categorical data were analysed with the Chi-square test and numerical data with the
Kruskal-Wallis test. Survival data were represented with Kaplan-Meier curves and
compared using the logrank test. Univariate and multivariate analyses were performed
on OS, BCSS and LR using Cox’s proportional hazards regression analysis.
Significance was assumed if p<0.01.
5
Results:
Patients:
957 patients underwent WLE with CB for invasive cancer within the study period.
The median age was 59 years, and follow up 45 months (range 0-253). The
distribution of patients into margin and cavity categories is shown in Table 1. Patients
in the m+c+ group were significantly younger than in the other groups (p=0.003). The
majority of patients had both negative margins and CB, with the remaining patients
approximately equally distributed amongst the other three groups. Figure 1 illustrates
the categorisation of patients within the m+ and c+ groups. 89 patients were m+c+ out
of 278 with either m+ or c+ (32%). 82/171 patients with positive CB had negative
margins (47.9%), and of the 761 patients with negative margins, 82 had positive CB
(10.8%).
Tumour characteristics:
The size of the tumour was larger in the m+c+ group. The type of tumour differed
according to the margin and cavity status as there were a higher proportion of patients
with infiltrating lobular carcinoma (ILC) in the m+c+ group compared with the m-cgroup. A reversal of this trend was seen with infiltrating ductal carcinoma (IDC).
There were a greater proportion of grade 2 tumours within the CB positive groups and
a higher percentage of grade 3 tumours in the m-c- group. Lymph node sampling was
performed in 939 cases (98.1%) and patients in the CB positive groups were more
likely to have positive nodes. ER status was recorded in 675 cases (70.5%) as this was
not routine practice in earlier specimen evaluation. A significantly greater number of
ER positive tumours were found in the m-c- group (p<0.001). The presence or
6
absence of VI was recorded in 893 cases (93.3%) and there were no significant
differences between the groups. The presence of multifocal disease however, did
show significant intergroup variation with a greater proportion in the c+ groups
(p<0.001).
Redo surgery:
The differences in the number of patients who underwent further surgery is
represented in Figure 2 together with the number who underwent adjuvant RT. 69
patients of the m+c+ group (77.5%) had a second operation, compared with 58
(54.2%) of the m+c- group, 54 (65.9%) of the m-c+ group and 42 (6.2%) of the m-cgroup; p<0.001. Median time to second surgery was 30 days. 14 patients (1.5%) had a
third operation at median 55 days after the first procedure with four patients in each of
the three positive margin and cavity groups, and two patients in the m-c- group
(p<0.001). The histology from the redo operation demonstrated a higher proportion of
patients in the CB positive groups with residual tumour, and ductal carcinoma in situ
(DCIS) in particular (Table 1).
Adjuvant therapy:
A higher proportion of patients in the CB negative groups underwent adjuvant RT
(Table 1). 249 patients were treated with post-operative chemotherapy with a higher
proportion in the m+c+ group. 817 patients had post-operative hormone manipulation
although there were no significant differences in the numbers of patients within each
group receiving this treatment.
7
Disease recurrence:
The disease recurred locally in 31 patients (3.2%), and regionally in 17 (1.8%). There
was a higher incidence of metastatic recurrence in the CB positive groups (Table 2).
Survival:
160 patients died within the study period (16.7%), 138 of whom (86.2%) died within
10 years (Table 2). For reasons of clarity the survival figures were censored at 10
years although statistical analysis was applied to the full data set and significance was
not found to be altered. Overall 10 year survival was 85.6% and BCSS 92.4%. A
higher proportion of the deaths occurred in the CB positive groups. 73/138 (52.8%) of
deaths were due to breast cancer progression and again the differences between the
groups were significant and followed the trend for OS.
Figures 3 and 4 demonstrate OS and BCSS. Patients with positive CB (groups m+c+,
m-c+) had poorer survival when compared with patients with negative CB. Figure 5
shows that OS in this group of patients was related to whether or not they underwent
redo surgery or RT, with the best outcomes in patients who had surgery. It was also
seen that if the CB was positive for invasive cancer, as opposed to DCIS alone, there
was a non significant trend to a worse BCSS (p=0.1).
Multivariate analysis of OS showed that tumour size greater than 20mm; age over 70
years and positive CB, whether infiltrated with invasive tumour or DCIS alone, were
independent predictors for poorer overall survival. Positive margins however did not
reduce OS. Re-operation (either completion mastectomy or repeat WLE) for positive
8
margins or CB improved OS as did the presence of ER positive tumours. Adjuvant
RT conferred overall survival benefit, although chemotherapy did not.
Univariate analysis demonstrated a trend to worse BCSS with tumour size greater
than 20mm, grade 3 tumours, node and VI positive tumours. Positive CB was an
independent predictor of poorer BCSS although positive margins were not. Redo
surgery improved BCSS. Table 3 shows the multivariate analysis which in addition
demonstrated improved BCSS with ER positive tumours.
Analysis of factors
associated with improved LR demonstrated only the protective effect of adjuvant RT
(p<0.01). The presence of positive margins or CB did not confer significantly worse
LR.
9
Discussion:
The significance of the cavity status in patients undergoing BCS has not been fully
established and consequently there is no standard operative practice. Many surgeons
perform a wide local excision and direct further treatment according to the histology
of the resected specimen and lymph nodes. Others also take biopsies from the residual
cavity (CB) - either four quadrants (superior, inferior, medial and lateral), six
(including superficial and deep biopsies if the primary excision did not include the
skin or extend to the pectoralis fascia) or complete cavity margin excision17;21. The
differences in technique have led to difficulties in meaningful comparison between
studies as well as inconsistencies in terminology with the terms quadrant biopsy,
cavity shaving, cavity margin and tumour-bed biopsy all being used for similar
procedures.
Demographics and tumour characteristics:
The prevalence of positive CB in the present study (17.8%) is in agreement with 1739% reported in other studies11;16;17;19;21-23. This study demonstrated that these patients
had different demographic and tumour characteristics when compared with those with
negative CB. The patients were younger, which may be related to the greater disease
extent and multifocality being more difficult to detect during imaging of a dense
breast. The tumour type was more often ILC, which is a more extensive and
multifocal disease that is less easy to identify on imaging. CB positive tumours were
more often node positive, which may be related to younger patients with larger
multifocal tumours.
10
Recurrence:
Margin and cavity positivity were not concordant. Positive CB was found to be a
better predictor for residual disease than positive margins. CB positive patients were
more likely to have a completion mastectomy and this may account for the lack of
difference found in LR rates between the groups, most of whom had salvage surgery
and/or RT.
Survival:
Tumour grade, node and ER status as expected were independent predictors of
survival. Age of over 70 years also adversely affected OS although young age reduced
BCSS. Positive CB was found to be a predictor for poorer survival but margin
positivity was not. This may be as a positive CB either reflects incompletely excised
tumour (if the ipsilateral margins are also positive), or the presence of multifocal
disease (if the margins are negative), both of which carry a poorer prognosis. CB
positive patients also had a greater number of risk factors for reduced survival
including node positivity, a younger age and a larger tumour size; however the
negative effect on survival persisted on multivariate analysis. The adverse effect of
positive CB was in some cases reversed by redo surgery and RT, with no difference
demonstrated between mastectomy and repeat WLE.
The findings from this study emphasise the importance of performing CB at the time
of WLE for breast cancer, and that its routine practice should be recommended.
Standardisation of the nature of the procedure and terminology used is essential so
that accurate comparisons can be made between studies.
11
CB should not be considered as merely an extension of the margin, but also as a
sample of the remaining breast tissue to aid the detection of residual or multifocal
disease. A negative ipsilateral CB in some patients with a positive margin however
may avoid the need for further surgery. Margin status was shown to be an unreliable
determinant of residual disease and prognosis.
Acknowledgements:
The authors thank Mr N Griffiths, Consultant Breast Surgeon, whose patients were
included in the study. Mrs S Bendall and Mrs V Stevenson were also invaluable in the
collection and organisation of the data.
12
References:
1. Jacobson JA, Danforth DN, Cowan KH, d'Angelo T, Steinberg SM, Pierce L
et al. Ten-year results of a comparison of conservation with mastectomy in the
treatment of stage I and II breast cancer. N.Engl.J.Med 1995;332:907-911.
2. Veronesi U, Salvadori B, Luini A, Greco M, Saccozzi R, del Vecchio M et al.
Breast conservation is a safe method in patients with small cancer of the
breast. Long-term results of three randomised trials on 1,973 patients.
Eur.J.Cancer 1995;31A:1574-1579.
3. Veronesi U, Salvadori B, Luini A, Banfi A, Zucali R, del Vecchio M et al.
Conservative treatment of early breast cancer. Long-term results of 1232 cases
treated with quadrantectomy, axillary dissection, and radiotherapy. Ann Surg.
1990;211:250-259.
4. Fisher B, Anderson S, Redmond CK, Wolmark N, Wickerham DL, Cronin
WM. 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;333:14561461.
5. van Dongen JA, Voogd AC, Fentiman IS, Legrand C, Sylvester RJ, Tong D 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.
6. Blichert-Toft M, Rose C, Andersen JA, Overgaard M, Axelsson CK, Andersen
KW et al. Danish randomized trial comparing breast conservation therapy
with mastectomy: six years of life-table analysis. Danish Breast Cancer
Cooperative Group. J.Natl.Cancer Inst.Monogr 1992;19-25.
7. 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.
8. Clarke M, Collins R, Darby S, Davies C, Elphinstone P, Evans E et al. Effects
of radiotherapy and of differences in the extent of surgery for early breast
cancer on local recurrence and 15-year survival: an overview of the
randomised trials. Lancet 2005;366:2087-2106.
9. Haga S, Makita M, Shimizu T, Watanabe O, Imamura H, Kajiwara T et al.
Histopathological study of local residual carcinoma after simulated
lumpectomy. Surg.Today 1995;25:329-333.
10. Wazer DE, Schmidt-Ullrich RK, Schmid CH, Ruthazer R, Kramer B, Safaii H
et al. The value of breast lumpectomy margin assessment as a predictor of
residual tumor burden. Int.J.Radiat.Oncol.Biol.Phys. 1997;38:291-299.
13
11. Beck NE, Bradburn MJ, Vincenti AC, Rainsbury RM. Detection of residual
disease following breast-conserving surgery. Br.J.Surg. 1998;85:1273-1276.
12. Veronesi U, Luini A, del Vecchio M, Greco M, Galimberti V, Merson M et al.
Radiotherapy after breast-preserving surgery in women with localized cancer
of the breast. N.Engl.J.Med 1993;328:1587-1591.
13. Peterson ME, Schultz DJ, Reynolds C, Solin LJ. Outcomes in breast cancer
patients relative to margin status after treatment with breast-conserving
surgery and radiation therapy: the University of Pennsylvania experience.
Int.J.Radiat.Oncol.Biol.Phys. 1999;43:1029-1035.
14. Park CC, Mitsumori M, Nixon A, Recht A, Connolly J, Gelman R et al.
Outcome at 8 years after breast-conserving surgery and radiation therapy for
invasive breast cancer: influence of margin status and systemic therapy on
local recurrence. J.Clin.Oncol. 2000;18:1668-1675.
15. Holland R, Veling SH, Mravunac M, Hendriks JH. Histologic multifocality of
Tis, T1-2 breast carcinomas. Implications for clinical trials of breastconserving surgery. Cancer 1985;56:979-990.
16. Malik HZ, George WD, Mallon EA, Harnett AN, Macmillan RD,
Purushotham AD. Margin assessment by cavity shaving after breastconserving surgery: analysis and follow-up of 543 patients. Eur.J.Surg.Oncol.
1999;25:464-469.
17. Keskek M, Kothari M, Ardehali B, Betambeau N, Nasiri N, Gui GP. Factors
predisposing to cavity margin positivity following conservation surgery for
breast cancer. Eur.J.Surg.Oncol. 2004;30:1058-1064.
18. Cao D, Lin C, Woo SH, Vang R, Tsangaris TN, Argani P. Separate cavity
margin sampling at the time of initial breast lumpectomy significantly reduces
the need for reexcisions. Am.J.Surg.Pathol. 2005;29:1625-1632.
19. Malik HZ, Purushotham AD, Mallon EA, George WD. Influence of tumour
bed assessment on local recurrence following breast-conserving surgery for
breast cancer. Eur.J.Surg.Oncol. 1999;25:265-268.
20. Taylor I, Mullee MA, Carpenter R, Royle G, McKay CJ, Cross M. The
significance of involved tumour bed biopsy following wide local excision of
breast cancer. Eur.J.Surg.Oncol. 1998;24:110-113.
21. Macmillan RD, Purushotham AD, Mallon E, Ramsay G, George WD. Breastconserving surgery and tumour bed positivity in patients with breast cancer.
Br.J.Surg. 1994;81:56-58.
22. Barthelmes L, Al Awa A, Crawford DJ. Effect of cavity margin shavings to
ensure completeness of excision on local recurrence rates following breast
conserving surgery. Eur.J.Surg.Oncol. 2003;29:644-648.
14
23. Macmillan RD, Purushotham AD, Mallon E, Love JG, George WD. Tumour
bed positivity predicts outcome after breast-conserving surgery. Br.J.Surg.
1997;84:1559-1562.
24. Campbell ID, Theaker JM, Royle GT, Coddington R, Carpenter R, Herbert A
et al. Impact of an extensive in situ component on the presence of residual
disease in screen detected breast cancer. J.R.Soc.Med. 1991;84:652-656.
25. Scopa CD, Aroukatos P, Tsamandas AC, Aletra C. Evaluation of margin
status in lumpectomy specimens and residual breast carcinoma. Breast J.
2006;12:150-153.
15
Table 1: Demographic data of the four margin and cavity groups
including tumour characteristics and adjuvant therapy
Category
Number
m+c+
%
89
9.2
m+c-
%
107
11.2
m-c+
%
82
8.6
m-c-
%
679
71.0
Total
%
957
100.0
P
Median Age (range)
52(36-84)
58(28-81)
58(26-85)
59(30-88)
59(26-88)
0.003*
Median tumour size (mm)
20(1.4-80)
17(1.1-65)
17(2.5-50)
15(1.2-60)
16(1.1-80)
0.002*
Tumour type
IDC
55
61.8
85
79.4
62
75.6
559
82.3
761
79.5
ILC
28
31.5
16
15.0
13
15.9
56
8.2
113
11.8
6
6.7
6
5.6
7
8.5
64
9.4
83
8.7
1
21
23.6
18
16.8
16
19.5
208
30.6
263
27.5
2
46
51.7
50
46.7
49
59.8
283
41.7
428
44.7
3
20
22.5
20
18.7
14
17.1
177
26.1
231
24.1
Node positive
34
38.2
25
23.4
26
31.7
171
25.2
256
26.8
0.04
ER positive
49
55.1
50
46.7
47
57.3
431
63.5
577
60.3
<0.001
VI positive
30
33.7
29
27.1
31
37.8
194
28.6
284
29.7
0.37
Multifocal
28
31.5
19
17.8
20
24.4
36
5.3
103
10.8
<0.001
Radiotherapy
40
44.9
87
81.3
46
56.1
595
87.6
768
80.3
<0.001
Chemotherapy
36
40.4
27
25.2
24
29.3
162
23.9
249
26.0
0.008
Endocrine therapy
72
80.9
90
84.1
73
89.0
582
85.7
817
55.4
0.47
<0.001
Other
<0.001
Tumour grade
<0.001
Redo surgery tumour type†
No residual tumour
32
46.4
42
72.4
24
44.4
32
76.2
130
58.3
IDC
9
13.1
9
15.6
10
18.5
3
7.2
31
13.9
ILC
12
17.4
2
3.4
5
9.3
2
4.7
21
9.4
DCIS
15
21.7
5
8.6
15
27.8
2
4.7
37
16.6
Other
1
1.4
0
0
0
0
3
7.2
4
1.8
Key: p values determined using Chi-Square test apart from * when Kruskal-Wallis test used. †Percentages calculated from
the number of patients undergoing redo surgery within each group.
16
Table 2: Recurrence and mortality data of the four margin and cavity groups
Category
m+c+
%
m+c-
%
m-c+
%
m-c-
%
Total
%
89
9.2
107
11.2
82
8.6
679
71.0
957
100.0
Local
5
5.6
7
6.5
3
3.7
16
2.4
31
3.2
Regional
4
4.5
2
1.9
1
1.2
10
1.5
17
1.8
0.23
Metastatic
21
23.6
10
9.3
16
19.5
47
6.9
94
9.8
<0.001
10 yr mortality
24
27.0
14
13.1
19
23.2
81
11.9
138
14.4
<0.001
10 yr BCRM
18
20.2
9
8.4
12
14.6
34
5.0
73
7.6
<0.001
Number
P
Recurrence
Key: p values determined using Chi-Square test. BCRM: breast cancer related mortality
17
0.07
Figure 1: Number of patients within the margin positive (m+)
and cavity positive (c+) groups
m+ (n=196)
107
c+ (n=171)
89
18
82
Figure 2: Flowchart showing the numbers of patients undergoing redo
surgery and RT within the four margin and cavity groups
WLE
957
m+c+
89
Mast
54
Repeat
WLE
15
m+c107
No Op
20
Mast
12
Mast
4
RT
13
RT
12
Repeat
WLE
46
Mast
3
RT
15
RT
3
m-c+
82
No Op
49
Mast
40
Repeat
WLE
1
RT
40
Repeat
WLE
14
m-c679
No Op
28
Mast
10
Mast
1
Mast
4
RT
44
RT
12
RT
10
Key: Mast: mastectomy, No Op: no redo surgery, RT: Radiotherapy
19
Repeat
WLE
23
RT
24
RT
2
Axillary
surgery
9
No Op
637
Repeat
WLE
1
RT
19
RT
8
RT
566
Figure 3: 10 year OS of the four margin
and cavity groups (p=0.02)
100
m+c+
m+cm-c+
80
Percent survival
m-c60
40
20
0
0
30
60
90
120
Months post op
No at risk
m+c+
89
79
73
70
65
m+c-
107
102
98
97
93
m-c+
82
79
74
69
63
m-c-
679
653
625
607
598
20
Figure 4: 10 year BCSS of the four margin
and cavity groups (p<0.001)
100
Percent survival
80
60
m+c+
40
m+c-
20
m-c-
m-c+
0
0
30
60
90
120
Months post op
No at risk
m+c+
89
81
76
74
71
m+c-
107
102
100
99
98
m-c+
82
81
76
74
70
m-c-
679
668
650
646
645
21
Figure 5: 10 year OS of CB positive patients with or without redo
surgery or RT after the initial operation (p<0.0001)
100
Percent survival
80
Op+ RT+
60
Op+ RTOp- RT+
40
Op- RT20
0
0
30
60
90
120
Months post op
No at risk
Op+RT+
47
47
46
46
45
Op+RT-
76
74
71
64
61
Op-RT+
39
31
24
23
18
Op-RT-
9
6
6
6
4
Key: Op+: redo surgery, Op-: no redo surgery, RT+: radiotherapy, RT-: no radiotherapy
22
Table 3: Multivariate analysis of BCSS using
Cox’s proportional hazards regression
957 subjects with 82 events
Deviance (likelihood ratio) chi-square = 95.9; df = 22; p<0.0001
Parameter
Hazard Ratio
95% CI
P
Age 50-69
1.08
0.63-1.87
0.76
Age >70
0.73
0.28-1.89
0.5
Tumour size 20-50mm
2.01
1.19-3.39
0.008
Tumour ER positive
0.39
0.19-0.81
0.01
Tumour grade 2
2.89
1.23-6.77
0.01
Tumour grade 3
3.78
1.51-9.44
0.004
Node positive
1.90
1.07-3.38
0.02
VI positive
1.62
0.93-2.81
0.08
Multifocal tumour
1.91
0.91-4.01
0.09
Cavities DCIS
2.78
1.27-6.08
0.01
Cavities invasive
3.10
1.56-6.14
0.001
Margin positive
1.24
0.81-1.90
0.31
Chemotherapy
0.81
0.42-1.52
0.5
Radiotherapy
1.54
0.67-3.52
0.30
Subsequent WLE
0.15
0.04-0.51
0.003
Subsequent mastectomy
0.38
0.15-0.95
0.04
23