Learning Center
Plans & pricing Sign in
Sign Out
Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

Screen detected ductal carcinoma in situ DCIS overdiagnosis or


									J Med Screen 2001;8:149–151                                                                                                        149

                              Screen detected ductal carcinoma in situ (DCIS):
                              overdiagnosis or an obligate precursor of invasive
                              A J Evans, S E Pinder, I O Ellis, A R M Wilson

                              Abstract                                                recurrence of only 4 years.6 A more recent
                              Objectives—Ductal carcinoma in situ                     study has shown that DCIS with poorly diVer-
                              (DCIS) represents 20%–25% of malig-                     entiated cytonuclear morphology has a signifi-
                              nancy detected at mammographic screen-                  cantly higher risk for the development of inva-
                              ing. This study aims to clarify the value of            sive carcinoma than DCIS with well
                              detecting DCIS at mammographic                          diVerentiated cytonuclear appearance.7 High
                              screening by assessing its biological char-             grade DCIS is associated with high grade inva-
                              acteristics and by comparing screen de-                 sive cancer.8–10 High grade invasive cancer car-
                              tected DCIS with a series of symptomatic                ries a poor prognosis11 unless detected when
                              DCIS lesions.                                           less than 10 mm in size.12 High histological
                              Methods—222 Screen detected and 151                     grade DCIS and the presence of necrosis are
                              symptomatic cases of pure DCIS were                     known to predict higher rates of recurrence
                              identified. Their histological grade and                 after treatment by wide local excision.13 14
                              the prevalence of necrosis were ascer-                  About 50% of recurrent lesions after wide local
                              tained and compared.                                    excision of DCIS have invasive carcinoma and
                              Results—Of the screen detected lesions 28               there is a strong association between the grade
                              (13%) were low grade, 40 (18%) intermedi-               of the original lesion and the grade of the
                              ate grade, and 153 (69%) high grade. Of                 recurrent invasive carcinoma.
                              screen detected lesions 186 (87%) were                     Critics of breast screening often claim that
                              necrotic and 29 (13%) were not. Of the 151              the high rates of DCIS found represent
                              symptomatic lesions 24 (16%) were low                   overdiagnosis, many being lesions which would
                              grade, 34 (23%) intermediate grade, and                 never present clinically and threaten the wom-
                              89 (61%) high grade. Of symptomatic                     an’s life.15 Such lesions are also often exten-
                              lesions 112 (75%) were necrotic and 36                  sive16 and therefore often require mastectomy
                              (24%) were not necrotic. Screen detected                to obtain adequate excision. Such criticism
                              DCIS was more often necrotic (p=0.008)                  would be valid if screen detected DCIS lesions
                              than symptomatic DCIS.                                  were predominantly of low histological grade
                              Conclusions—As most DCIS detected at                    but would not be valid if screen detected DCIS
                              screening is high grade and necrotic,                   was predominantly of high histological grade.
                              aggressive investigation of suspicious mi-              The detection of high grade DCIS by screen-
                              crocalcification     at     mammographic                 ing is likely to prevent the development of high
                              screening is advocated. Given the biologi-              grade invasive cancer of poor prognosis, within
                              cal features of screen detected DCIS, the               a few years and could be important in produc-
                              existence of an upper limit for the detec-              ing part of the reduction in mortality found in
                              tion of DCIS by the NHS breast screening                randomised trials of mammographic screen-
                              programme seems to be inappropriate.                    ing.
                              (J Med Screen 2001;8:149–151)                              Very few, if any, clinicians think that they
                                                                                      should not treat symptomatic DCIS. We there-
                              Keywords: breast neoplasms; ductal carcinoma in situ;
                              screening; diagnosis                                    fore aim to compare the biological characteris-
Helen Garrod Breast
                                                                                      tics of screen detected DCIS with those of
Screening Unit, City
Hospital, Hucknall                                                                    symptomatic DCIS and allow the doubt that
Road, Nottingham              The introduction of mammographic screening              exists concerning the benefit of detecting
NG5 1PB, UK                   has led to an increase in the number of cases of        DCIS at screening to be seen within this
A J Evans, consultant         pure ductal carcinoma in situ (DCIS) diag-              context. This study aims to clarify the value of
radiologist                   nosed. Of screen detected breast cancers,               detecting DCIS at mammographic screening
A R M Wilson, consultant      20%–25%1 are DCIS compared with 5% of                   by assessing its histological grade and the
                              symptomatic breast cancer.2 Screening women             prevalence of necrosis and by comparing
Histopathology                under 50 years of age is associated with even           screen detected DCIS with a series of sympto-
Department                    higher proportions of pure DCIS lesions than            matic DCIS lesions.
S E Pinder, senior lecturer   those found when screening women over 50.3
in histopathology                Ductal carcinoma represents a range of
I O Ellis, reader in          disease. Low grade DCIS has only a 25%–50%              Methods
                              chance of developing into low grade invasive            All cases of DCIS diagnosed at our institution
Correspondence to:            cancer at 30 years4 and has low rates of local          since 1973 have had their pathological details
Dr A J Evans                  recurrence when treated by wide local exci-             recorded on a database. Each case was investi-        sion.5 High grade DCIS that is often associated         gated to see if the lesion was screen detected or
Accepted for publication      with necrosis, in one small series had a 75%            was symptomatic. This was normally done by
8 March 2001                  risk of invasive disease with a mean time to            consulting the breast screening computer

150                                                                                 Evans, Pinder, Ellis, et al

      system and looking at the clinical information     progression of 4 years.6 A more recent study
      given on diagnostic breast imaging reports         has shown that DCIS with poorly diVerenti-
      from the time of diagnosis (AJE). Lesions          ated cytonuclear morphology has a signifi-
      picked up on opposite breast follow up             cantly higher risk for the development of inva-
      mammography, family history screening, or          sive carcinoma than DCIS with well
      lesions found on symptomatic mammography           diVerentiated cytonuclear morphology.7 Evi-
      for symptoms in the other breast were also         dence from recurrence after breast conserving
      classified as screen detected. Population mam-      surgery for DCIS indicate that high grade
      mographic screening began in our unit in           DCIS and DCIS with necrosis represent a bio-
      1987.                                              logically aggressive subset of DCIS which show
         The symptomatic lesions were detected in        higher rates of invasive and in situ recurrence
      women with nipple discharge, Paget’s disease,      compared with low grade DCIS lesions with-
      or breast lumps. In our institution single duct    out necrosis.5 13 14 18 Invasive lesions with an
      nipple discharge is investigated by mammogra-      extensive intraductal component show a pre-
      phy and nipple discharge cytology. If either of    disposition to local recurrence after breast
      these is suggestive of malignancy either percu-    conserving therapy.19 The grade of DCIS asso-
      taneous or surgical biopsy is performed. Galac-    ciated with invasive cancers has been shown to
      tography is not used at our institution.           correlate with both disease free interval and
         All cases underwent pathological review to      survival.8 These data indicate the biological
      ascertain histological grade17 and the presence    importance of DCIS and show that the grade
      or absence of necrosis. The grade and inci-        of associated DCIS has biological relevance.
      dence of necrosis in DCIS cases that were          The strong associations that exist between the
      symptomatic and screen detected were com-          grade of invasive cancers and the grade of
      pared and statistical significance was ascer-       DCIS from which they arose may at least in
      tained with the 2 test and 2 test for trend.       part explain these associations. High grade
                                                         DCIS gives rise to high grade invasive cancer
      Results                                            whereas low grade DCIS gives rise to low
      Two hundred and twenty two screen detected         grade invasive cancer.8–10 The association be-
      (age 37–74, mean 58 years) and 151 sympto-         tween grade 3 invasive cancer and poorly
      matic (age 30–82, mean 54 years) cases of pure     diVerentiated DCIS is present whatever grad-
      DCIS were identified. Ductal carcinoma in situ      ing system is used. On average, with five diVer-
      represented 20% of screen detected cancers         ent grading systems, 67% of invasive cancers
      during the study period and was detected at a      associated with high grade DCIS were grade
      rate of 1.2/1000 women screened. The detec-        3.10 With the Van Nuys system as many as 75%
      tion rate of screen detected invasive cancer       of invasive cancers associated with high grade
      during this period was 4.6/1000. Over the          DCIS are grade 3.
      study period DCIS represented 4% of sympto-           Necrosis represents another marker of DCIS
      matic breast cancer.                               lesions with aggressive biological characteris-
         Of the screen detected lesions 28 (13%)         tics. Ductal carcinoma in situ with necrosis has
      were low grade, 40 (18%) intermediate grade,       been shown to have a poorer disease free
      and 153 (69%) high grade. One case was not         survival and a higher local recurrence rate than
      graded. Of screen detected lesions 186 (87%)       DCIS without necrosis:5 16 DCIS with necrosis
      were necrotic and 29 (13%) were not. In seven      has also been shown to be associated with
      the presence or absence of necrosis was not        microinvasion.
      known.                                                Previous studies that have compared the his-
         Of the 151 symptomatic lesions 24 (16%)         tological features of screen detected and symp-
      were low grade, 34 (23%) intermediate grade,       tomatic DCIS used DCIS classifications based
      and 89 (61%) high grade. Four cases were not       on architectural pattern.20 21 These studies
      graded. Of symptomatic lesions 112 (76%)           found that comedo DCIS was commoner in
      were necrotic and 36 (24%) were not necrotic.      screen detected lesions than in lesions present-
      The presence or absence of necrosis was            ing symptomatically.20 21 Since these studies
      unknown in three.                                  were published several new DCIS classifica-
         No significant diVerence was found between       tions have been proposed based on cytological
      the grade of screen detected and symptomatic       features and the presence of necrosis.5 17 22 23
      DCIS (p=0.08). Screen detected DCIS was            These classifications have become accepted as
      more often necrotic (p=0.008) than sympto-         being more reproducable and being more
      matic DCIS.                                        accurate in predicting the biological behaviour
                                                         of DCIS lesions.
      Discussion                                            This study has shown that DCIS detected by
      Long term studies on the progression to            mammographic screening is predominantly of
      invasive disease of DCIS lesions originally mis-   high nuclear grade and only 13% is low grade.
      diagnosed as benign indicate a 25%–50% rate        Screen detected DCIS is also more likely to
      of invasive disease at 30 years of follow up.4     contain areas of necrosis than symptomatic
      These studies consist almost entirely of low       lesions. The most likely explanation for these
      grade DCIS lesions, as such lesions are more       findings is suggested by a comparison of the
      likely to be misdiagnosed as benign lesions        radiological findings of diVerent DCIS sub-
      than high grade DCIS. Data on the natural          types. High grade DCIS more often shows
      history of high grade DCIS is limited but a        abnormal mammographic features than low
      small series showed a 75% rate of progression      grade DCIS, which is often mammographically
      to invasive disease with a mean time to            occult.16 24 The mammographic calcification

Screen detected ductal carcinoma in situ (DCIS)                                                                                                              151

                              found in high grade DCIS and DCIS with                              7 Eusebi V, Feudale E, Foschini MP, et al. Long term
                                                                                                     follow-up of in situ carcinoma of the breast. Semin Diagn
                              necrosis is more characteristic of malignancy,                         Pathol 1994;11:223–35.
                              often showing rod and branch shapes. The                            8 Lampejo OT, Barnes DM, Smith P, et al. Evaluation of infil-
                                                                                                     trating ductal carcinomas with a DCIS component: corre-
                              granular-punctate calcifications found in low                           lation of the histologic type of the in-situ component with
                              grade DCIS24 are non-specific and are more                              grade of the infiltrating component. Semin Diagn Pathol
                              likely to be confused with benign processes and                     9 Cadman BA, Ostrowski JL, Quinn CM. Invasive ductal car-
                              not recalled at mammographic screening.                                cinoma accompanied by ductal carcinoma in situ (DCIS):
                                                                                                     comparison of DCIS grade with grade of invasive
                                 The subtype profile of screen detected DCIS                          component. Breast 1997;6:132–7.
                              suggests that most lesions would progress to                       10 Douglas-Jones AG, Gupta SK, Attanoos RL, et al. A critical
                                                                                                     appraisal of six modern classifications of ductal carcinoma
                              high grade invasive disease within 5–10 years.                         in situ of the breast (DCIS): correlation with grade of asso-
                              Preventing such high grade invasive disease is                         ciated invasive carcinoma. Histopathology 1996;29:397–
                              likely to have a considerable impact on breast                     11 Elston CW, Ellis IO. Pathological prognostic factors in
                              cancer mortality. The small proportion of low                          breast cancer. I. The value of histological grade in breast
                                                                                                     cancer: experience from a large study with long term follow
                              grade, non-necrotic DCIS lesions found at                              up. Histopathology 1991;19:403–10.
                              mammographic screening indicates that diag-                        12 Tabar L, DuVy SW, Vitak B. The natural history of breast
                                                                                                     carcinoma: what have we learned from screening? Cancer
                              nosis of the more indolent forms of DCIS is                            1999;86:449–62.
                              not common. This finding indicates that over-                       13 Silverstein MJ, Lagios MD, Craig PH , et al. A prognostic
                                                                                                     index for ductal carcinoma in situ of the breast. Cancer
                              diagnosis of DCIS at mammographic screen-                              1996;77:2267–74.
                              ing is not a major problem.                                        14 Fisher ER, Dignam J, Tan-Chiu E, et al. Pathological
                                                                                                     findings from the national surgical adjuvant breast project
                                 On the basis of this evidence we advocate                           (NSABP): 8 year update of protocol B-17. Cancer 1999;86:
                              that the aggressive investigation of suspicious                        429–438.
                                                                                                 15 Baum M. Review of ABC of breast disease. J Med Screen
                              microcalcification at mammographic screening                            1995;2:233–4.
                              should be encouraged. The consequent detec-                        16 Holland R, Hendricks JHCL, Verbeek ALM, et al. Extent,
                                                                                                     distribution and mammographic/histological correlations
                              tion of predominantly high grade DCIS at                               of breast ductal carcinoma in situ. Lancet 1990;335:519–
                              screening also enables the diagnosis of other-                         22.
                                                                                                 17 NHS breast screening programme. Pathology reporting in
                              wise occult coexisting small grade 3 invasive                          breast cancer screening April 1995. ISBN 1 871997 22 4
                              carcinomas, which have already arisen within                       18 Solin LJ, I-Tien Yeh, Kurtz J, et al. Ductal carcinoma in situ
                                                                                                     (intraductal carcinoma) of the breast treated with breast
                              the DCIS lesions.25 The increased availability                         conserving surgery and definitive irradiation. Cancer 1993;
                              of stereotactic core biopsy, especially with                           71:2532–42.
                                                                                                 19 Van Dongen JA, Fentiman IS, Harris JR, et al. In situ breast
                              digital imaging, has led to good diagnostic                            cancer: the EORTC consensus meeting. Lancet 1989;ii:
                              results when sampling mammographic                                     25–7.
                                                                                                 20 Bellamy OC, Mc Donald C, Salter DM, et al. Non-invasive
                              calcifications.26–28 This should mean that an                           ductal carcinoma of the breast: the relevence of histological
                              aggressive approach to mammographic calcifi-                            categorization. Hum Pathol 1993;24:16.
                                                                                                 21 Evans A, Pinder S, Ellis I, et al. Screening detected and
                              cation should not give rise high rates of surgical                     symptomatic ductal carcinoma in situ: mammographic fea-
                              benign biopsy. Given the nature of screen                              tures with pathologic correlation. Radiology 1994;191:237–
                              detected DCIS, the existence of an upper limit                     22 Holland R, Peterse JL, Millis RR, et al. Ductal carcinoma in
                              for the detection of DCIS by the NHS breast                            situ: a proposal for a new classification. Semin Diagn Pathol
                              screening programme seems to be inappropri-                        23 Poller D, Silverstein M, Galea M, et al. Ductal carcinoma in
                              ate.29                                                                 situ of the breast: a proposal for a new simplified histologi-
                                                                                                     cal classification. Mod Pathol 1994;7:257–62.
                                                                                                 24 Evans AJ, Pinder SE, Wilson ARM, et al . Ductal carcinoma
                               1 Andersson I, Aspergren K, Janzon L et al. Mammographic              in situ of the breast: correlation between mammographic
                                  screening and mortality from breast cancer: the Malmo              and pathologic findings. AJR Am J Roentgenol 1994;162:
                                  mammographic screening trial. BMJ 1988;297:943–8.                  1307–11.
                               2 Smart C, Myers M, Gloecker L. Implications from SEER            25 Evans AJ, Pinder SE, Wilson ARM et al. The detection of
                                  data on breast cancer management. Cancer 1978;41:787–9.            DCIS at mammographic screening enables the diagnosis of
                               3 Evans WP, Starr AL, Bennos ES. Comparison of the relative           small, grade 3 invasive tumours Br J Cancer 1997;75:542–
                                  incidence of impalpable invasive breast carcinoma and              4.
                                  ductal carcinoma in situ in cancers detected in patients       26 Liberman L, LaTrenta LR, Van Zee KJ, et al. Stereoractic
                                  older and younger than 50 years of age. Radiology                  core biopsy of calcifications highly suggestive of malig-
                                  1997;204:489–91.                                                   nancy. Radiology 1997;203:673–7.
                               4 Page DL, Dupont WD, Rogers LW, et al. Continued local           27 Britton PD, Flower CDR, Freeman AH, et al. Changing to
                                  recurrence of carcinoma 15–25 years after a diagnosis of           core biopsy in an NHS breast screening unit. Clin Radiol
                                  low grade ductal carcinoma in situ of the breast treated           1997;52:764–86.
                                  only by biopsy. Cancer 1995;76:1197–200.                       28 Whitlock JPL, Evans AJ, Burrell HC, et al. Digitally
                               5 Silverstein MJ, Poller DN, Waisman JR et al. Prognostic             acquired imaging improves upright stereotactic core biopsy
                                  classification of breast duct carcinoma in situ. Lancet 1995;       of mammographic microcalcifications. Clin Radiol 2000;
                                  345:1154–7.                                                        55:374–7
                               6 Dean L, Geshchicter CF. Comedocarcinoma of the breast.          29 NHS breast screening programme. Quality assurance guide-
                                  Arch Surg 1938;36:225–34.                                          lines for radiologists. May 1997. ISBN 1 871997 92 5.


To top