Ductal Carcinoma
In Situ
395
therapy. Ciatto et al.
56
reported that infiltrating carcinoma
developed in the ipsilateral breast in 7 of 55 women (12.7%)
treated only by local excision or quadrantectomy. The DCIS
had been detected through routine examinations or mam-
mographic screening in Florence, Italy, from 1968 to 1988.
The length of follow-up was not stated.
Several population-based prospective analyses of exci-
sional surgery alone have been reported. The largest series,
from Denmark, consisted of 112 women with a median
follow-up of 53 months.
369
Recurrent invasive carcinoma
occurred in 5 women (4.4%), and 19 (17%) had recurrent
DCIS. The initial lesions ranged from 1 to 80 mm, with a
median size of 10 mm. Features favoring recurrence were
large nuclear size, lesion size greater than 10 mm, and the
presence of comedonecrosis regardless of the histologic sub-
type (solid, micropapillary, or cribriform). Papillary lesions,
of which there were few, had a high recurrence rate whether
or not comedonecrosis was present. Heterogeneity of growth
pattern was found in all but 3 of the 112 lesions. Margin sta-
tus, evaluable in only about one-third of the cases, did not
appear to be a good predictor of recurrence. Recurrences
occurred in 33% of cases with negative margins. Review of
132 patients with DCIS diagnosed in Malmo, Sweden, re-
vealed that 3 of 21 women (14%) treated by breast-conserv-
ing surgery alone developed ipsilateral invasive carcinoma
after a median follow-up of 7 years.
370
Two additional studies described the follow-up of women
with DCIS detected in regional mammography screening
programs and treated by excisional surgery alone. Arnes-
son et al.
371
identified 38 women with lesions detected with
a single-view mammography technique, who were treated
only by “sector resection” with negative margins. After a
median follow-up of 60 months, five (13%) patients had
recurrent carcinoma consisting of two invasive and three
intraductal lesions. The primary lesions associated with
recurrence were 3 to 15 mm in size. Cribriform DCIS pre-
ceded the two invasive lesions, whereas comedocarcinoma
was followed by recurrent DCIS. Carpenter et al.
372
reported
on 28 women with lesions detected through screening
mammography and clinical examination. Treatment con-
sisted of quadrantectomy or segmental resection. No data
were given about margin status. After a median follow-up
of 38 months, five recurrences detected mammographically
as microcalcifications in the region of prior excision con-
sisted of one invasive and four intraductal lesions. There
was no significant association between the development of
recurrent carcinoma and the size of the primary lesions, the
size of the excisional biopsy specimen, or the presence of
multifocality.
Schwartz et al.
373
selected patients with mammographi-
cally detected nonpalpable or incidentally discovered DCIS
for treatment by excision alone. Patients were eligible for
inclusion if the mammographic diameter of the area of cal-
cifications did not exceed 25 mm. Comedocarcinoma was
present to some extent in 51% of the lesions and was the pre-
dominant type in 29%. At least two subtypes of DCIS were
present in 41% of the cases. The excisions were not consis-
tently studied for margin status. After a median follow-up
recurrence 5 years after biopsy. The other patient was found
to have recurrent micropapillary carcinoma 8.8 years after
biopsy. In this series, the observed frequency of subsequent
carcinoma was 4.3 times (90% CI, 1.1 to 11.1) the expected
risk, somewhat higher for nonmicropapillary (5.4) than for
micropapillary (3.9) DCIS.
Data from a retrospective review of 1,877 biopsies clas-
sified as benign in the Nurses’ Health Study were reported
in 2007.
147
Previously unrecognized DCIS was identified in
13 specimens (0.7%). Four were classified as low, six as in-
termediate, and three as high nuclear grade. Architecturally,
seven were cribriform, and three were each solid and micro-
papillary. Carcinoma was clinically diagnosed subsequently
in the ipsilateral breast in 10 (77%) of the 13 patients after
intervals of 2 to 18 years. DCIS was diagnosed in four cases,
2 to 6 years after the initial biopsy that was interpreted as be-
nign, and six patients were found to have invasive carcinoma
after 4 to 18 years. Three patients who did not have clinically
diagnosed subsequent carcinoma had follow-up of 21 to 27
years. When compared with women with nonproliferative
fibrocystic changes, the odds ratio for the development of
subsequent carcinoma was 20.1 (95% CI, 6.1 to 66.5), and for
invasive carcinoma it was 13.5 (95% CI, 3.7 to 49.7).
Breast Conservation by Excision
Only—Prospective Data
Prior to the emergence of clinical trials, little information
was available prospectively about the treatment of DCIS by
breast-conserving excision. In 1982, Lagios et al.
267
reported
that 3 (15%) of 15 patients treated for DCIS by local exci-
sion developed recurrences in the ipsilateral breast during
follow-up averaging 44 months. An expanded series consist-
ing of 79 patients with average follow-up of 48 months was
reported in 1989.
366
Eight patients had developed recurrent
carcinoma, four entirely intraductal, and four invasive. Seven
of eight recurrences in the breast were in patients with com-
edocarcinoma or cribriform carcinoma with comedonecro-
sis. The eighth recurrence was associated with “intraductal
carcinoma with anaplasia.” Further information about this
series was reported in 1994.
367
At that time, the local failure
rate in the conserved breast was 14.7% after a mean follow-
up of 106 months. Half of the recurrences were described
as “minimally invasive carcinomas,” and the others were
intraductal. When correlated with histologic features of the
initial lesion, the recurrence rate for DCIS of high nuclear
grade with comedonecrosis was 30.5%, and for those with
intermediate nuclear grade it was 10%. There were no breast
recurrences in patients with low-grade DCIS.
After follow-up averaging 39 months, Fisher et al.
368
found a breast recurrence rate of 23% in 22 patients with
DCIS treated by excisional biopsy alone. These patients had
been entered into a clinical trial for invasive carcinoma in
which one of the randomized treatments was excision alone,
and the diagnosis was corrected to DCIS during a subsequent
pathology review. The same report described recurrences
in 2 (7%) of the 29 women with retrospectively diagnosed
DCIS who had been randomized to receive radiation
