вход по аккаунту



код для вставкиСкачать
The Natural History of Breast Carcinoma
What Have We Learned from Screening?
e write to support the view put forward in the report of the
Swedish Two-County Trial of breast screening1 that invasive
tumors must be detected before they progress to less differentiated
forms if the benefits of screening are to be realized fully. This
represents a broadening of the authors’ previous emphasis on the
importance of detecting “small” Grade 3 tumors (⬍ 15 mm) to
achieve early reduction of mortality.2 We have always upheld the
broader approach.
The Edinburgh Randomised Trial (ERT) involved women of a
more limited age range (45– 64 years at study entry), with the
screened arm offered biennial mammography, and it has achieved
comparability with the Two-County Trial through a mortality reduction of 29% at 14 years of follow-up.3 Any comparisons of
pathology features in the tumors in the screened and control arms
must avoid the influence of length bias. The ‘unbiased’ group,
comprised of tumors in previously screened women arising either
as interval cases or detected at incidence screens, constitutes a set
for comparison with controls. Such comparisons for the ERT cases
are provided in Table 1, to illustrate the significant differences in
the distribution of size, grade, and axillary lymph node positive
groups that occurred in the unbiased group of the ERT. We also
have demonstrated among all operable invasive tumors in our
randomized population a strong statistical relation between the
worsening of grade and increasing tumor size.4 This is not to be
construed as evidence for obligate progression to a worse grade for
all tumors with time. Rather, these findings from Sweden and
Scotland provide substantial evidence for the phenotypic drift of
some tumors over time, supporting the longitudinal French study
of breast tumors managed over a period that lacked substantial
mammographic intervention.5
The presentation of different mean sojourn times for the distinct
grades and histologic types of breast carcinoma in the Swedish study
may be of value, but the concept that they can be defined appears to
be a contradiction of belief in biologic shift with time. The provocative
statements with regard to women age ⬍ 50 years need greater substantiation than is provided in the article, and the authors’ conclusions regarding screening younger women remains a matter for debate. Furthermore, to our knowledge, the delays in mortality
reduction for such young women anticipated by the Swedish investigators have not been experienced in the U.K.3,6 The assumptions
involved in the mathematic modeling of the natural history of breast
carcinoma and screen detection need to be borne in mind by nonstatistical readers of this complex article. These procedures cannot
© 2000 American Cancer Society
Tumor Size, Histologic Grade and Lymph Node Positivity Proportions
in Women in the Screened and Control Groups, Avoiding Length Bias
ERT group
Screeneda (%)
Control (%)
Tumor size (mm)
ⱖ 50
0 positive
1–3 positive
⬎ 3 positive
(n ⫽ 212)
(n ⫽ 209)
(n ⫽ 192)
(n ⫽ 224)
(n ⫽ 196)
(n ⫽ 193)
ERT: Edinburgh Randomised Trial; ALN: axillary lymph nodes.
Detected at incidence screen or as interval cases in previously screened women.
P ⬍ 0.0001.
P ⬍ 0.01.
Borderline significance; chi-square test for proportions: P ⫽ 0.07, and for linear trend: P ⫽ 0.049.
provide a credible account of all details regarding the
natural history of breast carcinoma.
Tabar L, Duffy SW, Vitak B, Chen H-H, Prevost TC. The
natural history of breast carcinoma: what have we learned
from screening? Cancer 1999;86:449 – 62.
Tabar L, Fagerberg G, Duffy SW, Day NE, Gad A, Grontoft O.
Update of the Swedish Two-County Programme of Mammographic Screening for Breast Cancer. Radiol Clin North
Am 1992;30:187–210.
Alexander FE, Anderson TJ, Brown HK, Forrest APM, Hepburn W, Kirkpatrick AE, et al. 14 years of follow-up from the
Edinburgh randomised trial of breast-cancer screening.
Lancet 1999;353:1903– 8.
Anderson TJ, Alexander FE, Lamb J, Smith A, Forrest APM.
Pathology characteristics that optimise outcome prediction
of a breast screening trial. Br J Cancer. In press.
Tubiana M, Koscielny S. Natural history of human breast
cancer: recent data and clinical implications. Breast Cancer
Res Treat 1991;18:125– 40.
UK Trial of Early Detection of Breast Cancer Group. 16 year
mortality from breast cancer in the UK Trial of Early Detection of Breast Cancer. Lancet 1999;353:1909 –14.
Thomas J. Anderson, M.D.
Department of Pathology
Freda E. Alexander, Ph.D
Public Health Sciences
Patrick M. Forrest, M.D.
Clinical Health Sciences
University of Edinburgh
Edinburgh, Scotland
Author Reply
e first demonstrated the potential of tumors to
progress with respect to malignancy grade in
1991,1 and suggested that screening may arrest this
progression. It is gratifying that similar results now are
emerging from the Edinburgh Trial.
Our major interpretation of the lesser and delayed benefit observed in the 40 – 49 years age group
in our trial was that prevention of deaths from
Grade 3 aggressive ductal carcinoma confers an earlier mortality benefit than prevention of deaths in
less aggressive tumors because the former potentially are more rapidly fatal. This is intuitively reasonable and was demonstrated empirically in our
results. There was an early mortality benefit, and in
particular a reduction in mortality from Grade 3
ductal carcinoma, associated with screening in the
50⫹ years age group whereas there was a later benefit, and no reduction in overall mortality from
Grade 3 ductal carcinoma, in the 40 – 49 years age
group. Therefore we cannot agree with the assertion
of Anderson et al. that this is provocative.
Duffy SW, Tabar L, Fagerberg G, Gad A, Gröntoft O, South
MC, et al. Breast screening, prognostic factors and survivalresults from the Swedish Two-County Study. Br J Cancer
1991;64:1133– 8.
Laszlo Tabar, M.D.
Mammography Department
Central Hospital
Stephen W. Duffy, M.Sc.
MRC Biostatistics Unit
Institute of Public Health
University Forvie Site
Cambridge, United Kingdom
Determining the Site of the Primary
Cancer in Patients with Skeletal
Metastasis of Unknown Origin
A Retrospective Study
e read with interest the article by Katagiri et al.
on bone metastases of unknown primary site.1 A
few years ago, we conducted a similar study of 115
patients with bone metastases as a first manifestation
of malignancy, and we would like to point out some
CANCER April, 1 2000 / Volume 88 / Number 7
similarities and differences between the results reported by Katagiri et al. and our own.
In our work, medical history, physical examination, radiography (including chest X-ray), and laboratory investigation suggested the primary site in 53.9%
of our 115 patients and in 74.1% of those whose primary tumors were identified antemortem. Consequently, we agree that these initial examinations have
to be conducted first, and that only a few other investigations must be performed when no primary is suspected at this stage.
However, abdominal ultrasound examination was
performed on 51 of our patients with no suspected
primary, and revealed 12 (23.5%) primary lesions (10
renal cell, 1 hepatocellular, and 1 pancreatic carcinoma). Thus, we agree that abdominal explorations
should be conducted even if no primary site is suspected. Because we performed only 4 abdominal computed tomography (CT) scans, we do not have an
opinion concerning the respective usefulness of abdominal ultrasound and CT.
An interesting difference between the results of
Katagiri et al. and our own is that the primary malignancy was identified antemortem in 88% of their 64
patients and in only 70.4% of ours. Unlike Katagiri et
al., we excluded bone localizations of hematologic
malignancies, but we do not believe that that point
alone explains the differences in our respective positive diagnosis percentages. This difference is more
likely due to performance of thoracic CT scans on a
majority of the patients in the study of Katagiri et al.
and only one of ours. In favor of this hypothesis, lung
carcinomas were found in 35.9% of their patients and
only 17.4% of ours (while the respective percentages of
other primaries were quite similar). This suggests that,
as proposed by Katagiri et al., chest CT scan should be
performed on all patients.
Finally, we also believe that other explorations are
of little value when no primary site is suspected at this
stage. In our work, we performed 25 explorations of
the gastrointestinal tracts of patients with no suspected primary, and we found only 1 colon cancer.
However, we recommend that mammography be obtained for all women. We agree that breast carcinoma
in women with bone metastasis usually presents with
abnormality on physical examination. However, in our
experience, mammography was positive in 1 of 8 patients without clinical evidence of breast tumor, and
therefore had a significant effect on therapy and life
Katagiri H, Takahashi M, Inagaki J, Sugiura H, Ito S, Iwata H.
Determining the site of the primary cancer in patients with
skeletal metastasis of unknown origin: a retrospective study.
Cancer 1999;86:533–7.
Maillefert JF, Tebib J, Huguenin MC, Chauffert B, Pascaud F,
Peere T, et al. Les métastases osseuses révélatrices: recherche du cancer primitif. Etude d’une série de 115 cas. Sem
Hop Paris 1993;69:372– 8.
Jean Francis Maillefert, M.D.
Christian Tavernier, M.D.
Department of Rheumatology
Dijon University Hospital
Hôpital Général
Dijon, France
Jacques Tebib, M.D., Ph.D.
Department of Rheumatology
Lyon University Hospital
Lyon, France
Author Reply
e are pleased to respond to the comments of
Maillefert et al. with respect to the difference
between our study1 and their study2 of skeletal metastasis of unknown origin.
As in written in their comments, we believe that
thoracic CT scan should be performed as an initial
diagnostic study of all patients with skeletal metastasis
of unknown origin. The reasons for this are 1) lung
carcinoma is the most common primary lesion not
only in our study, but also in other previous reports;3–5
and 2) for the diagnosis of lung carcinoma, thoracic
CT scan is more accurate than plain chest X-ray film.
In our study, 4 (17%) of the 23 patients with lung
carcinoma were diagnosed as primary lung carcinoma
with chest X-ray film alone, but CT revealed lung
carcinoma that had not been seen on the chest X-ray
film in 7 patients (30%), and the lesion was seen with
chest X-ray film and identified as the primary lesion
with the support of CT findings in 11 (48%).1 Therefore, we suppose that the percentage of lung carcinomas in the study by Maillefert et al. (17.4%) would
have been higher if thoracic CT scan had been done in
most of their cases.2 Consequently, the percentage of
their patients in whom the primary malignancy was
identified antemortem (70.4%) would have been
higher as well.
We agree that abdominal examination focusing
on liver and kidney is necessary. However, we think
that abdominal CT is the most recommended modality as an initial abdominal study. In our study, as well
as in previous reports, CT is more useful than ultrasound examination in detecting abdominal lesions.6,7
It also would be more efficient with less suffering for
patients if the thoracic and abdominal CT were performed at the same time. However, ultrasound examination of the abdomen is noninvasive and useful for
those patients who are unsuitable for transport and
CT examination.
Maillefert et al. recommend performing mammography on all female patients. Heimann et al. reported that mammographically detected breast carcinomas were smaller than those detected clinically.8
This may be true; however, most of the patients with
breast carcinoma and bone metastasis present with
abnormality on physical examination. None of our
patients and only one of their patients required mammography to detect breast carcinoma. Therefore,
mammography should not be an initial study, but an
additional study in selected cases.
In conclusion, we recommend limiting the initial
imaging study to chest X-ray, thoracic and abdominal
CT scan, and bone scan. Other studies, such as mammography, pelvic and neck CT scan, and examination
of the gastrointestinal tract and gynecologic organs,
should be performed as an additional study in patients
who are candidates for palliative surgery and/or radiotherapy. The diagnostic strategy we suggested for
patients with skeletal metastases from an unknown
primary site1 are not only time- and cost-saving, but
also cause less suffering for the patient.
We thank Maillefert et al. for their interest.
Katagiri H, Takahashi M, Inagaki J, Sugiura H, Ito S, Iwata H.
Determining the site of the primary cancer in patients with
skeletal metastasis of unknown origin: a retrospective study.
Cancer 1999;86:533–7.
Maillefert JF, Tebib J, Huguenin MC, Chauffert B, Pascaud F,
Peere T, et al. Les métastases osseuses révélatrices: recherche du cancer primitif. Etude d’une série de 115 cas. Sem
Hos Paris 1993;69:372– 8.
Simon MA, Bartucci EJ. The search for the primary tumor in
patients with skeletal metastases of unknown origin. Cancer
1986;58:1088 –95.
Rougraff BT, Kneisl JS, Simon MA. Skeletal metastases of
unknown origin: a prospective study of a diagnostic strategy. J Bone Joint Surg Am 1993;75–A:1276 – 81.
Nottebaert M, Exner GU, von Hochstetter AR, Schreiber A.
Metastatic bone disease from occult carcinoma: a profile.
Int Orthop 1989;13:119 –23.
McMillan JH, Levine E, Stephens RH. Computed tomography in the evaluation of metastatic adenocarcinoma from
an unknown primary site. Radiology 1982;143:143– 6.
Kamin PD, Bernardino ME, Wallace S, Jing BS. Comparison
of ultrasound and computed tomography in the detection of
pancreatic malignancy. Cancer 1980;46:2510 –2.
Heimann R, Bradley J, Hellman S. The benefits of mammography are not limited to women of ages older than 50 years.
Cancer 1998;82:2221– 6.
Hirohisa Katagiri, M.D.
Department of Orthopaedic Surgery
Nagoya University School of Medicine
Nagoya, Japan
Без категории
Размер файла
52 Кб
Пожаловаться на содержимое документа