396
Chapter 11
(24%) and 12 recurrences were invasive. Substantial necrosis
was present in the original DCIS in 70% of the cases with re-
currence and in 83% with invasive recurrence. Necrosis was
present in only 25% of DCIS not followed by recurrence.
MacDonald et al.
377
retrospectively reviewed 445 patients
with DCIS treated by excision alone between 1972 and 2004.
In this nonrandomized series of selected patients, 70 (17%)
of the women had local breast recurrences, including 26 (6%)
that were invasive. Seventy-two of the 79 breast recurrences
(91%) were in the same quadrant as the original DCIS. The
median follow-up period for all patients was 57 months, and
the median time to breast recurrence was 26 months. One pa-
tient died of metastatic breast carcinoma, and there were 23
deaths due to other causes. Significant risk factors for local re-
currence were margin width of less than 10 mm, age less than
40 years, and high nuclear grade. The hazard ratio (HR) for
breast recurrence was 5.39 times greater if the margin widths
were less than 10 mm than if they were greater than 10 mm.
Wong et al.
378
described the results of surgery alone with
a final margin clearance of at least 10 mm for DCIS in a sin-
gle-arm prospective trial. The study was limited to patients
selected on the basis of the following criteria: predominant
grade 1 or 2; mammographic size of less than or equal to 2.5
cm; and final lumpectomy margins clear by 10 mm or more or
noDCIS in a reexcision specimen. After amedian follow-up of
40 months, 13 of 158 patients (8.2%) enrolled in the study had
experienced a breast recurrence. Four recurrences (31%) were
invasive, and the remaining recurrences were intraductal. Ten
recurrences (77%) were in the same quadrant as the original
DCIS. The breast recurrence rate was 2.4% per patient-year
with a projected 5-year breast recurrence rate of 12%.
When DCIS has been treated by excision alone, the fore-
going data indicate that lesion size greater than 2.5 cm, ne-
crosis, high nuclear grade, and a clear margin of less than
10 mm are factors that predispose to recurrence in the
breast. However, as demonstrated by Wong et al.,
378
local
breast recurrence may occur after excisional surgery alone,
even with a clear margin that exceeds 10 mm.
A commercially available RT-PCR-based multigene assay
(Oncotype DX, Genomic Health Inc, Redwood City, CA) is
available for DCIS. This test generates a “DCIS score” rang-
ing from 0 to 100. The “score” quantifies the likelihood of
10-year recurrence for
in situ
or invasive carcinoma.
379
The
test uses a subset of the 21 genes assessed for the Oncotype
DX breast cancer assay for invasive breast carcinoma. It can
be performed on formalin-fixed paraffin-embedded tissue
using manual microdissection for lesional tissue. The likeli-
hood of local recurrence at 10 years increases continuously
with increases in score, and the latter information could
potentially help in management decisions.
380
Breast-Conserving Surgery with Radiotherapy
Radiation therapy has been employed in conjunction with
excisional surgery in an effort to improve local control af-
ter breast conservation therapy for DCIS. Data are avail-
able from a number of prospective investigations of patients
treated by excisional surgery with radiation therapy, and from
of 47 months, 11 recurrences were detected in the ipsilat-
eral breasts of 70 women (15.3%), consisting of three inva-
sive and eight intraductal lesions. All recurrent DCIS were
detected mammographically because of the appearance of
calcifications. Comedocarcinoma was present in 10 of 11
lesions followed by recurrence (one was papillary), and all
recurrent DCIS were of the comedo type. There was no cor-
relation between the number of duct cross sections with
DCIS in the primary lesion and recurrence.
Another series of clinically selected patients treated by ex-
cision alone was reported by Hetelekidis et al.
374
The group
consisted of 59 women, almost all of whom had mammo-
graphically detected lesions. Local recurrence in the breast
was detected in 10 women (17%) 5 to 132 months after exci-
sion, with a median interval of 37 months and a 5-year recur-
rence rate of 10%. Four recurrent lesions were invasive, and
six were intraductal. Eight recurrences were at the site of prior
excision. Factors associated with local recurrence were high
nuclear grade, lesions occupying more than five low-magni-
fication (4×) microscopic fields and tumor 1 mm or less from
the margin. Lesion size was the only statistically significant
indicator of recurrence. The local recurrence rate was 18% for
lesions with poor nuclear grade and less than 10% when nu-
clear grade was intermediate or well differentiated (
p
=
NS).
DCIS involving fewer than five low-power fields had a 3% re-
currence rate compared with more extensive lesions with a
17% recurrence rate (
p
=
0.02). The local recurrence rates for
negative and close margins were 8% and 25%, respectively.
The foregoing prospective studies suggest that DCIS can
be treated successfully by excisional breast-conserving sur-
gery in some cases. Although the majority of these patients
will not develop a breast recurrence, about 20% of recurrent
lesions are invasive and, therefore, carry the added risk of
metastatic spread.
Risk Factors for Breast Recurrences after
Conservation Surgery Alone
Data presented in many reports on breast-conserving sur-
gery suggest that pathologic features of DCIS might con-
tribute to the success of treatment by excisional surgery
alone. A case-control study by Badve et al.
375
examined the
value of five histologic classification schemes for predicting
local recurrence in the breast after excisional surgery alone.
The exercise involved reviewing slides of excisional biopsies
from 43 patients who developed recurrences and from 81
controls matched for age at diagnosis who did not develop a
recurrence. The median time to recurrence was 39 months,
and for recurrence-free controls median follow-up was 68
months. None of the classifications systems was clearly su-
perior for predicting local recurrence. The characteristics of
DCIS most strongly associated with recurrence classifica-
tion were poorly differentiated histologic grade, the pres-
ence of necrosis, and poorly differentiated nuclear grade.
Goonewardene et al.
376
studied the significance of necro-
sis as a risk factor for local recurrence in 166 women who had
been treated for DCIS by excision alone. After an average fol-
low-up of 6.5 years, recurrences were detected in 40 patients
