Editor—I thank Bishop and Wallis for inquiring about the value of ipsilateral mammography every 6 months in patients with ductal carcinoma in situ treated with breast preservation. To my knowledge there are no prospective randomised data in the medical literature that prove the value of such frequent mammography. In the early 1980s, however, despite the lack of data, we simply made this our policy. The rationale was straightforward. If we were performing mammography yearly for healthy women why not perform it more frequently for patients with conservatively treated ductal carcinoma in situ?
Table.
Details of patients with recurrence of ductal carcinoma in situ. Values are number (percentage) unless stated otherwise
| Variable |
Non-palpable recurrence |
Palpable recurrence |
P value |
| Number of patients |
60 |
16 |
— |
| Median follow up (years) |
8.9 |
9.2 |
0.22 |
| Median time to local recurrence (years) |
2.3 |
5.1 |
0.0002 |
| No of invasive local recurrences |
22 (37) |
13 (81) |
0.003 |
| Median size of local recurrence (mm) |
19 |
49 |
0.0001 |
| Stage ⩾2 at time of invasive local recurrence |
9/22 (45) |
7/13 (54) |
0.46 |
| Local recurrence found by mammography (%) |
60 (100) |
12 (75) |
— |
| Local recurrence found by palpation (%) |
0 |
16 (100) |
— |
|
| No of patients with distant metastases |
2 |
3 |
0.03 |
| No of deaths from breast cancer |
2 |
2 |
0.14 |
Through 1998, we treated 480 patients with ductal carcinoma in situ with breast preservation (216 with excision plus radiation therapy and 264 with excision alone). Overall, 76 local recurrences occurred, 60 (79%) of which were non-palpable and detected by mammography alone. Sixteen (21%) local recurrences were detected by palpation; of these, 12 (75%) were also detected by mammography at the time of clinical presentation. The table compares the two groups at the time of recurrence.
The median follow up from the time of original diagnosis was about 9 years for each group. Non-palpable recurrences were detected on average about 3 years earlier than palpable recurrences. In patients with non-palpable recurrences there was a 37% chance that their lesions had progressed to invasive carcinoma compared with 81% in patients with palpable recurrences (P=0.003). The median size of an invasive recurrence was almost 30 mm larger than that of a non-invasive recurrence. Distant metastases developed in two patients with non-palpable invasive recurrences and three patients with palpable invasive recurrences (P=0.03). The difference was significant because there were many more patients in the non-palpable group. Four deaths occurred—two in each group.
An analysis of the initial patient demographics and tumour factors, including patient age, size of original lesion, margin status, comedonecrosis, classification, nuclear grade, palpability, mammographic presentation of original lesion, and percentage of patients receiving radiation therapy, showed no significant differences between the two groups.
Although these data were collected prospectively over a long period of time, they are not randomised and do not prove the superiority of mammography every 6 months. They do, however, suggest that non-palpable recurrences are more favourable than palpable recurrences.
