1
Genomic profiling defines variable clonal relatedness between invasive breast cancer
and primary ductal carcinoma in situ
Esther H. Lips
1*
, Tapsi Kumar
2,3*
, Anargyros Megalios
4*
, Lindy L. Visser
1
, Michael
Sheinman
5
, Angelo Fortunato
6,7
, Vandna Shah
4
, Marlous Hoogstraat
5
, Emi Sei
3
, Diego
Mallo
6,7
,
Maria Roman-Escorza
4
,
Ahmed A. Ahmed
4
,
Mingchu Xu
2
, Wim Brugman
8
**, Karen
Clements
9
, Helen R. Davies
10
,
Liping Fu
1
, Anita Grigoriadis
4
, Timothy M. Hardman
11
,
Lorraine M. King
11
, Marielle Krete
8
, Petra Kristel
1
, Michiel de Maaker
1
, Carlo C. Maley
7
,
Jeffrey R. Marks
11
, Brian Menegaz
12
, Lennart Mulder
1
, Frank Nieboer
1
, Salpie Nowinski
4
,
Sarah Pinder
4
, Jelmar Quist
4
, Carolina Salinas-Souza
4
, Michael Schaapveld
13
, Marjanka K.
Schmidt
1
, Abeer M. Shaaban
14
, Rana Shami
4
, Mathini Sridharan
4
, John Zhang
2
***, Hilary
Stobart
15
, Deborah Collyar
16
, Serena Nik-Zainal
10
, Lodewyk F.A.Wessels
5,17
, E. Shelley
Hwang
11
, Nicholas N. Navin
3
, Andrew Futreal
2
, Alastair Thompson
12#
, Jelle Wesseling
1,18,19#
,
Elinor J. Sawyer
4#
on behalf of the Grand Challenge PRECISION consortium
*Joint First Author,
#
Joint Senior Author
**authors listed from here to ***are in alphabetical order
Corresponding author: Elinor Sawyer, Breast Cancer Genetics, School of Cancer and
Pharmaceutical Sciences, Faculty of Life Sciences and Medicine, Guy’s Cancer Centre,
King’s College London, London, UK. elinor.sawyer@kcl.ac.uk
1
Division of Molecular Pathology, The Netherlands Cancer Institute, Amsterdam, the
Netherlands
2
Department of Genomic Medicine, MD Anderson Cancer Center, Houston, USA
3
Department of Genetics, MD Anderson Cancer Center, Houston, USA
4
School of Cancer and Pharmaceutical Sciences, Faculty of Life Sciences and Medicine,
Guy’s Cancer Centre, King’s College London, London, UK.
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.22.21253209doi: medRxiv preprint
NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.
2
5
Oncode Institute and Division of Molecular Carcinogenesis, The Netherlands Cancer
Institute, Amsterdam, the Netherlands
6
School of Life Sciences, Arizona State University, Tempe, AZ
7
Biodesign Center for Biocomputing, Security and Society, Arizona State University, USA
8
Genomics Core Facility, The Netherlands Cancer Institute, Amsterdam, The Netherlands
9
Screening Quality Assurance Service, Public Health England, UK
10
MRC Cancer Unit, University of Cambridge, Hutchison/MRC Research Centre and
Academic Department of Medical Genetics, Cambridge Biomedical Research Campus,
Cambridge, UK
11
Department of Surgery, Duke University School of Medicine, Durham, NC
12
Department of Surgical Oncology, Dan L Duncan Comprehensive Cancer Center, Baylor
College of Medicine, Houston, Texas, USA.
13
Division of Psychosocial research and Epidemiology, The Netherlands Cancer Institute,
Amsterdam, The Netherlands
14
Queen Elizabeth Hospital Birmingham and University of Birmingham, Birmingham, UK
15
Independent Cancer Patients' Voice, UK
16
Patient Advocates in Research, USA
17
Faculty of Electrical Engineering, Mathematics, and Computer Science, Delft University of
Technology, Delft, the Netherlands.
18
Divisions of Diagnostic Oncology, The Netherlands Cancer Institute, Amsterdam, the
Netherlands
19
Department of Pathology, Leiden University Medical Center, the Netherlands
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.22.21253209doi: medRxiv preprint
3
Abstract:
Pure ductal carcinoma in situ (DCIS) is being diagnosed more frequently through breast
screening programmes and is associated with an increased risk of developing invasive
breast cancer. We assessed the clonal relatedness of 143 cases of pure DCIS and their
subsequent events using a combination of whole exome, targeted and copy number
sequencing, supplemented by single cell analysis. Unexpectedly, 18% of all invasive events
after DCIS were clonally unrelated to the primary DCIS. Single cell sequencing of selected
pairs confirmed our findings. In contrast, synchronous DCIS and invasive disease (n=44)
were almost always (93%) clonally related. This challenges the dogma that almost all
invasive events after DCIS represent invasive transformation of the initial DCIS and
suggests that DCIS could be an independent risk factor for developing invasive disease as
well as a precursor lesion. Our findings support a paradigm shift that confirms a more
complex role for DCIS than previously recognized, and that the future management of DCIS
should take into account both the precursor and risk factor implications of this diagnosis.
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.22.21253209doi: medRxiv preprint
4
Introduction
Approximately 60% of invasive breast cancers of ductal/no special type are associated with
synchronous ductal carcinoma in situ (DCIS)
1
. The majority of molecular studies looking at
these two synchronous components suggest that they are clonally related supporting the
hypothesis that DCIS is a non-obligate precursor of invasive breast cancer
2,3
. Recent single
cell analysis of DCIS that presented with synchronous invasive disease has shown that most
mutations and copy number aberrations have already evolved within DCIS prior to invasion
and suggest that multiple clones escape from the in situ component into the adjacent tissues
to establish invasive carcinomas
4
.
Less is known about the genomic evolution of invasive disease that occurs after pure DCIS
(i.e., DCIS without an invasive component) that has been treated with surgery with or without
radiotherapy. Pure DCIS is being diagnosed more frequently through breast screening
programmes and now accounts for ~20% of screen detected breast cancers. It carries a 4-
to 10-fold increased risk of invasive breast cancer with the highest risk being in women
under the age of 50
5-8
. It is presumed that the majority of these invasive lesions are clonally
related to the initial DCIS, particularly if treated with surgery alone. However, analysis of
clonal relatedness of subsequent invasive cancer arising after a diagnosis of pure DCIS has
not been widely performed, due to the difficulty of obtaining samples from a large cohort of
women with DCIS with adequate follow-up. Previously tumour grade, morphology and
immunohistochemistry have been shown to differ between DCIS and subsequent invasive
cancer, but it is not clear whether this is a result of tumour progression or represents the
development of a new primary invasive cancer
9,10
.
It is important to understand the clonal relatedness of DCIS and subsequent invasive
disease in order to assess the true recurrence rate of DCIS after different treatment
modalities and to be able to design effective strategies, not only to reduce subsequent
diagnoses but also overtreatment. To investigate the clonal relatedness between pure DCIS
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.22.21253209doi: medRxiv preprint
5
and subsequent events, we pooled samples from three countries resulting in the largest
cohort to date of DCIS cases with 5-17 years follow up and thus adequate time to develop
subsequent events (Supplementary Table 1).
Results
143 primary DCIS and their subsequent events were analysed. The median age at diagnosis
of primary DCIS was 57 years (range 34-87 years) and median time to subsequent event
was 4 years (0.4-17.5years). 54% were high grade, 67% were ER positive (ER+), 24% were
ER negative (ER-), 28% Her2 positive and 46% Her2 negative.
We studied three different types of subsequent events after a diagnosis of pure DCIS:
1. Pure DCIS that developed a subsequent ipsilateral invasive event (DCIS->ipINV; n=95)
2. Pure DCIS that developed a subsequent ipsilateral DCIS event (DCIS->ipDCIS; n=34).
3. Pure DCIS that developed a subsequent contralateral invasive event (DCIS->contraINV;
n=14)
We also analysed 44 pairs of synchronous DCIS and invasive disease (synDCIS&INV) in
order to assess whether clonal relatedness rates differed between synchronous and
subsequent invasive disease (Figure 1, Supplementary Table 2 for summary of sample pair
type, origin and molecular characterisation, Supplementary Tables 3a-c for clinical
characteristics).
Genomic features of primary DCIS and subsequent invasive disease
Whole exome sequencing (WES) (n=70), targeted sequencing (PanelSeq) (n=69) and copy
number analysis (n=118) revealed that pure DCIS and subsequent invasive disease (INV,
where INV includes both ipINV and contraINV) share similar genomic profiles (Figure 2).
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprintthis version posted March 26, 2021. ; https://doi.org/10.1101/2021.03.22.21253209doi: medRxiv preprint