Ongoing research

For more information than what is given below, please contact Per Hall (

The Karma Cohort

The Karma Cohort is a prospective screening cohort consisting of 70,877 women, initiated in 2011 [1]. Participating women have donated blood that is stored at the Karolinska Institutet Biobank []. Participating women have also repeatedly answered a detailed questionnaire and allowed matching to health care registers and medical records. Every time a participant preforms a mammogram the images are stored. Current rate is approximately 25,000 images / month. Approximately 20,000 Karma participants have been genotyped with the iCOGS or Oncoarry chip as part of a collaboration with the Breast Cancer Association Consortium [as an example please see 2]. For a detailed description of the Karma Cohort, please see [1].

  1. Gabrielson M, Eriksson M, Hammarström M, Borgquist S, Leifland K, Czene K, Hall P. Cohort profile: The KarolinskaMammography Project for Risk Prediction of Breast Cancer (KARMA).Int J Epidemiol. 2017 Feb 9. doi: 10.1093/ije/dyw357. PMID: 28180256
  2. Michailidou K, .., Czene K, ..Hall P, ….., Easton DF. Association analysis identifies 65 new breast cancer risk loci.Nature. 2017 Oct 23. doi: 10.1038/nature24284.


Karma Intervention Study – Karisma

Karisma (the Karma Intervention Study) aims to reduce the incidence of breast cancer by preventing breast cancer using the anti-estrogen drug tamoxifen. Tamoxifen has been used to prevent breast cancer recurrence in patients for 30 years. Healthy women who take tamoxifen have a 40-50% reduced risk of breast cancer, but the medicine may cause side effects. The side effects of tamoxifen are hot flashes, sweating, mode swings, weight gain, vaginal discharge, etc. There is also a risk of more severe side effects such trombo-embolism. The aim of the Karisma trial is to find out if lower doses of tamoxifen have fewer side effects without lowering the risk-reducing effect.

In February 2017 recruitment of 1,440 women started at the South General Hospital, Stockholm. Women attending the national mammography-screening program are invited to participate. After informed consent participants are randomised to 20, 10, 5, 2.5, 1 mg of tamoxifen and placebo. Outcome is change in mammographic density and women are treated for 6 month. Mammograms are taken at baseline and end of treatment. In addition to the randomized controlled trial we measure metabolites of tamoxifen, polymorphism of key metabolising genes (e.g CYP2D6) and female sex hormones.

We anticipate ending recruitment by December 2018 and having the first results by early fall 2019.


Karma CREME – A Phase 2 trial on topical endoxifen

Our overall aim of the Karisma trial is to reduce the risk of breast cancer using a lower dose of tamoxifen than the accepted 20 mg. A lower dose would probably come with lower levels of side effects. An alternative approach to lowering the dose would be to apply tamoxifen directly to the skin of the breast expecting the compound to penetrate in to the glandular tissue. The challenge is that tamoxifen is more or less inactive and needs to be metabolised to any of the two active metabolites afimoxifen or endoxifen.

In collaboration with the US company Atossa Genetics [] we will launch a pilot study, testing the hypothesis that topical application of endoxifen can generate an individual change in mammographic density.

Ninety participants will be randomized to one of three groups (one placebo group and two groups on different doses of topical endoxifen). The objective of the study is to determine the effect size of breast density between the topical and active groups, which will permit sample size calculations in a future Phase III study.

The study will open for inclusion during the second half of 2018.


Karma Risk – Individual risk prediction of breast cancer 

Good risk prediction models can be used in clinic to predict the individual risk of developing breast cancer. Most current models for breast cancer include lifestyle factors and family history of breast cancer. Some include mammographic density or genetic determinants.

We have developed a new model for individualised short-term risk prediction, called CAD2Y. The prediction model has been developed using mammographic density, computer-aided detection of microcalcifications and masses, use of menopausal hormone therapy, family history of breast cancer, menopausal status, age, and body mass index.

Applying CAD2Y to the Karma population enabled early identification of women within the mammography-screening program at such high risk of breast cancer that additional examinations are warranted. In contrast, women at low risk could probably be screened less intensively.

We are currently working on adding more risk factors to the model to further improve the predictive power.

For a detailed description of the CAD2Y breast cancer risk prediction model, please see [3].

  1. Eriksson M, Czene K, Pawitan Y, Leifland K, Darabi H, Hall P. A clinical model for identifying the short-term risk of breast cancer.Breast Cancer Res. 2017 Mar 14;19(1):29. doi: 10.1186/s13058-017-0820-y.PMID: 28288659


Karma Risk Communication – Acceptability and view of individual risk prediction and risk based screening among professionals and healthy women

Additional knowledge of breast cancer risk factors has meant that we are currently exploring risk-based screening, i.e. determining screening strategies based on women’s varying levels of risk. This also enables risk management through primary prevention strategies, e.g. lifestyle programme or risk-reducing medication. However, future implementation of risk-based screening and prevention will warrant significant changes in current practice and policy. In collaboration with researchers in the Netherlands (Radboud university medical center; Dutch Expert Center for Screening), and UK (Manchester University NHS Foundation Trust), the Karma group investigates women’s, and health professionals, perceptions of risk prediction , prevention and risk based screening. This to map organisational needs and preferences to ultimately optimise future acceptability and uptake.

First results under review and submission is planned to fall 2018.


Karma MRI – Abbreviated MRI protocol 

At South General Hospital, Stockholm, the Karma Group is testing if a short protocol for magnetic resonance imaging (MRI) is as good as a full protocol. The sensitivity of a mammogram in a woman with very dense breast is low and thus the number of false negatives increased. An MRI examination has an excellent sensitivity but takes approximately an hour to perform and interpret. There are recent studies describing an “abbreviated” MRI protocol, where examination time is reduced to 15 minutes without loosing sensitivity or specificity. Using a case control approach, including 500 breast cancer cases and healthy controls we will test if the short and full protocols are comparable.

Recruitment is ongoing. First results will come during  2019.


Karma STRATUS – High throughput measurement of mammographic density

The white, dense, part of a mammogram is a reflection of the stroma and epithelial content of the breast. The higher the density, the higher is the risk of breast cancer. Breast density can be measured through visual scores or by semi-automatic, computer aided, techniques.

In Karma, more than 3000 mammograms are collected every week and a fully automatic measuring system for mammographic density is therefore a necessity. Using the software ImageJ (, and in collaboration with colleagues at Karolinska Institutet and Agency for Science, Technology and Reserch (A*STAR) in Singapore, we have previously developed an automatic measuring model of mammographic density [4].

By measurements from raw images using FDA approved software iCAD wehave now developed a new algorithm called STRATUS, which measures density on all type of images, regardless of vendor, andcontrols for non-biological differences seen in time series ofmammograms from the same women [5]. STRATUS has the potential to become a useful tool for epidemiological studies and clinical follow-up.

  1. Eriksson M, Czene K, Pawitan Y, Leifland K, Darabi H, Hall P. A clinical model for identifying the short-term risk of breast cancer.Breast Cancer Res. 2017 Mar 14;19(1):29. doi: 10.1186/s13058-017-0820-y. PMID: 28288659
  2. Eriksson M, Li J, Leifland K, Czene K, Hall P. A comprehensive tool for measuring mammographic density changes over time.Breast Cancer Res Treat. 2018 Jun;169(2):371-379. doi: 10.1007/s10549-018-4690-5. Epub 2018 Feb 1. PMID: 29392583


Karma Normal – Analyses of normal breast tissue

Karma Normal is a biopsy-based study of 161 healthy Karma participants without prior history of breast cancer. The overall aim of Karma Normal is to understand the biological mechanisms underlying mammographic density as a risk factor for breast cancer.

By analysing the histological composition and expression of hormone receptors in the healthy breast we found that high mammographic density was associated with higher amount of stroma and epithelium and less amount of fat, but was not associated with a change in epithelial proliferation or receptor status [3]. Increased expressions of both epithelial PR and stromal ER were associated with a greater proportion of stroma, suggesting hormonal involvement in regulating breast tissue composition. We have also found that reproductive risk factors significantly influence the epithelial tissue compartment and expression of hormone receptors in later life, and that these changes remain after menopause [4].

These studied provide deeper insights of the biological mechanisms by which risk factors influence breast tissue composition and expression of hormone receptors, and as a consequence the risk of breast cancer.

  1. Gabrielson M, Chiesa F, Paulsson J, Strell C, Behmer C, Rönnow K, Czene K, Östman A, Hall P. Amount of stroma is associated with mammographic density and stromal expression of oestrogen receptor in normal breast tissues.Breast Cancer Res Treat. 2016 Jul;158(2):253-61. doi: 10.1007/s10549-016-3877-x. Epub 2016 Jun 27. PMID: 27349429
  2. Gabrielson M, Chiesa F, Behmer C, Rönnow K, Czene K, Hall P. Association of reproductive history with breast tissue characteristics and receptor status in the normal breast.Breast Cancer Res Treat. 2018 Mar 30. doi: 10.1007/s10549-018-4768-0. PMID: 29603032


Karma Plasma Proteomics and Hormone Metabolomics – Identifying circulating markers of breast cancer risk

Using the rich collection of plasma in Karma we are conducting proteomic profiling to identify putative proteins and biomarkers associated with mammographic density. Plasma samples from 1329 women without prior history of breast cancer have been profiled with antibody suspension bead array assays [5].Screening proteins in plasma indicates associations between breast density and processes of tissue homeostasis, DNA repair, cancer development and/or progression in breast cancer.

Further studies comparing proteomic profiles of cases and controls in relation to mammographic density and breast cancer risk factors are ongoing.

Some sex hormones and their metabolites are associated with mammographic density and breast cancer risk. To better understand these associations we are studying a large panel of circulating plasma hormones in 700 cases and 1,400 control women.

  1. Byström S, Eklund M, Hong MG, Fredolini C, Eriksson M, Czene K, Hall P, Schwenk JM, Gabrielson M. Affinity proteomic profiling of plasma for proteins associated to area-based mammographic breast density.Breast Cancer Res. 2018 Feb 14;20(1):14. doi: 10.1186/s13058-018-0940-z


Karma iCOGS – Identification of genetic determinants of breast cancer

Genetic alterations that predispose women for later developing breast cancer are studied in collaboration with the Breast Cancer Association Consortium (BCAC) (, the University of Cambridge and the European Commission funded COGS project ( consortium includes more than 550 researchers at 300 research facilities across the globe.

Currently DNA from more that 20,000 Karma participants is genotyped together with over 275,000 non – Swedish samples. The samples are genotyped with a 570K single nucleotide polymorphism (SNP) markers custom array manufactured by Illumina, the so-called iCOGS OncoArray (

Professor Douglas Easton, head of BCAC –

A complete list of Breast Cancer Association Consortium publications with Karma samples can be found here.

Comments are closed.