Ductal Carcinoma
In Situ
397
6.1% (95% CI, 4.1% to 8.2%). With a median follow-up of
6.7 years, the incidence of recurrence for 105 patients in
the high-grade stratum was 15.3% (95% CI, 8.2% to 22.5%).
Patients with lower grade DCIS and clear margins 3 mm or
wider were deemed to have an acceptably low rate of ipsilat-
eral breast events at 5 years without irradiation. Patients with
high-grade lesions had a much higher rate of recurrence,
suggesting that excision alone would be inadequate.
Although patient outcome is maximized when treatment
is tailored to patient and disease characteristics, the num-
ber of events prevented per 1,000 radiation-treated women
is typically less than 10%.
390
Radiation treatment offers an
increase in the chances of successful breast conservation, but
this positive feature is somewhat offset by the likelihood that
mastectomy will be necessary in the event of recurrence, the
detection of a new primary carcinoma, or as a result of vari-
ous radiation-associated complications.
391
Accelerated partial breast irradiation, that is, brachy-
therapy using Mammosite (Hologic, Boxborough, MA),
has emerged as a substitute for whole-breast irradiation (in-
cluding three-dimensional conformal external beam) in the
treatment of breast carcinoma.
392,393
Preliminary outcome
results indicate similar results regardless of either radiation
modality used. A radiotherapy boost to the surgical cavity
improves local control in DCIS.
394
The interpretation of radiation-induced cytologic changes
in mammary epithelium may occasionally pose diagnostic
difficulty. It should be noted that recurrent disease exhibits
a histopathologic appearance that is generally similar to the
index lesion.
395
As such, histopathologic review of the previ-
ously diagnosed DCIS is critical for the optimal evaluation
of all “recurrences.”
Risk Factors for Breast Recurrence after
Conservation Surgery with Radiotherapy
Factors associated with an increased risk of breast recur-
rence after surgery alone are also significant for recurrence
after surgery and radiotherapy. Necrosis in DCIS or com-
edo-type DCIS imparts an especially high risk of breast re-
currence after breast conservation with radiotherapy. Solin
et al.
396
found that the presence of necrosis was a significant
risk factor when it occurred in DCIS with poorly differen-
tiated nuclear grade. Kuske et al.
397
reported significantly
poorer local control in patients with comedo (75%) than
in those with noncomedo (98%) carcinoma, but did not of-
fer a definition of “comedo” DCIS. In the NSABP B17 trial
comparing excision alone and excision plus radiotherapy,
features associated with increased risk of local recurrence
after either form of treatment were the presence of moderate
to marked comedonecrosis, regardless of histologic subtype,
margins that were positive or indeterminate, multifocality,
and a moderate to marked lymphocytic infiltrate.
398
The size
of the lesion (less than 10 or greater than or equal to 10 mm)
did not prove to be a statistically significant predictor for
breast recurrence. Necrosis proved to be the only statisti-
cally significant independent risk factor for recurrence in
both treatment groups in multivariate analysis.
randomized trials comparing the results of excision alone to
excision with radiotherapy. Initial reports published in the
1980s described selected patients and noted a recurrence
rate in the conserved breast of 10% or less after a median
follow-up of approximately 5 years.
381–383
Bornstein et al.
384
reported an actuarial 8-year breast recurrence rate of 27% in
a series of 38 selected patients. Five of the eight recurrences
were invasive, and one of these women developed metastatic
carcinoma. Solin et al.
385
identified 259 women treated in
nine institutions in the United States and Europe and found
a 10-year actuarial breast failure rate of 16%. Fifty percent
of the 28 recurrences were invasive, and four patients devel-
oped metastatic carcinoma. A later follow-up report by Solin
et al.
386
described an expanded cohort of 1,003 patients from
10 institutions with a median follow-up of 8.5 years (range,
0.2 to 24.6 years). Initial recurrences of carcinoma limited to
the breast were documented in 82 patients (8.2%). One addi-
tional patient had angiosarcoma (0.1%), and the nature of the
breast recurrence was unknown in two cases. The calculated
15-year rates of breast recurrence and systemic metastases
were 19% and 3%, respectively. The histology of the 82 initial
documented breast carcinoma recurrences was as follows:
invasive ductal 46 (57%), intraductal 34 (41%), and other 2
(2%). Five additional patients had invasive breast recurrences
concurrently with systemic recurrences. The risk of breast re-
currence was significantly lower in women who had a nega-
tive surgical margin or were 50 years of age or older at the
time of treatment. During the course of follow-up, contralat-
eral breast carcinoma was reported in 71 (7%) patients, and
56 (6%) women had a nonmammary malignant neoplasm.
Two large-scale randomized trials have compared local
control in the breast after excision alone and excision plus ra-
diotherapy. The NSABP B17 study revealed a 50% reduction
in breast recurrence when radiation was added to lumpec-
tomy.
387
The 12-year risk of breast recurrence was 31.7%
with surgery alone and 15.7% for surgery with radiotherapy.
A similar randomized trial by the European Organization
for Research and Treatment of Cancer (EORTC) yielded a
4-year breast recurrence rate of 16% after surgery alone and
9% when radiation was added to surgery.
388
Updated results
from the EORTC study with a median follow-up of 10.5 years
revealed a 10-year breast recurrence-free rate of 74% for exci-
sion alone and 85% for surgery followed by radiation.
296
The
risk reduction attributed to radiotherapy for recurrent DCIS
was 48%, and for invasive carcinoma it was 42%.
389
Patho-
logic features significantly associated with breast recurrence
in both treatment groups were intermediate to poorly differ-
entiated grade and cribriform or solid growth.
A low-risk subset of DCIS patients who could be spared
radiation treatment was characterized in a prospective East-
ern Cooperative Oncology Group and North Central Cancer
Treatment Group trial.
339
Patients with either low- or inter-
mediate-grade DCIS spanning less than 2.6 cm or high-grade
DCIS measuring less than 1.1 cm who had margin clearance
of more than 3 mm and no residual mammographic calcifica-
tions were eligible for this 1997 to 2002 trial. In a median fol-
low-up of 6.2 years, the 5-year rate of ipsilateral breast events
in 565 patients in the low- or intermediate-grade group was
