Novel Imaging Strategies for Breast Cancer Patients

Gisela de Lima Gomes de Menezes

Research output: ThesisDoctoral thesis 1 (Research UU / Graduation UU)

Abstract

Breast cancer is the most common malignant disease in women worldwide and approximately 1 in 8 women will be diagnosed with breast cancer during their lifetime. However, the mortality rates of this disease have been declining in the past decades, mostly due to a combination of advances in treatment and early detection by mammography. The three most important imaging modalities for detection and evaluation of breast abnormalities are mammography (MMG), ultrasound (US) and magnetic resonance imaging (MRI). This dissertation focused in these 3 imaging modalities and discussed recent developments and strategies for both conventional and novel breast cancer imaging techniques.

Nowadays, MMG is considered the best imaging technique for early detection of breast cancer in a population with an average risk for this disease. However, it is still difficult for radiologists to properly differentiate between benign and malignant calcifications, especially those classified as BI-RADS 3. Although BI-RADS 3 calcifications are considered probably benign (≤ 2% probability of malignancy), one of our studies showed a much higher prevalence of malignancy in these kinds of lesions (more than 30%). Therefore, we recommend that BI-RADS 3 calcifications should be considered suspicious and should undergo an immediate biopsy to avoid missing a malignant lesion.

MRI outperforms both MMG and US in terms of cancer detection, but the specificity of this technique is only low to moderate (approximately 70%). The dissertation discussed the most important indications to perform a breast MRI. One of these indications is to provide additional preoperative information about multifocal disease in patients diagnosed with invasive lobular carcinoma (ILC), due to the high incidence of ipsilateral and contralateral foci of disease in these tumors. One of the studies described in this dissertation found that even patients diagnosed with mixed ductal and lobular carcinomas have a very high incidence of ipsilateral and contralateral disease, irrespective of the amount of lobular component of the tumors. Therefore, the use of preoperative breast MRI in these patients deserves consideration.

We also discussednew breast imaging techniques that may improve breast cancer diagnostic accuracy, such as magnetic resonance (MR) spectroscopy, diffusion-weighted imaging (DWI) and 7 tesla (T) MRI. These are promising techniques, but they are not yet ready to be incorporated in clinical practice. Larger studies with standardized protocols and more statistical power are necessary to confirm the clinical value and the cost-benefit of these new imaging modalities.

Optoacoustic (OA) imaging is a new technology based on laser light excitation that, combined with conventional US (OA/US), enables simultaneous assessment of functional and anatomical information that may improve distinction between benign and malignant masses of the breast. A clinical trial to assess OA/US’s ability to correctly downgrade benign masses originally classified as BI-RADS 4a and 4b in US was discussed in this dissertation and the conclusion was that OA/US is a promising technique that has the potential to decrease the number of biopsies of benign lesions and short interval follow-up imaging exams. Future studies may help us to unfold the full potential of this new technology.
Original languageEnglish
Awarding Institution
  • University Medical Center (UMC) Utrecht
Supervisors/Advisors
  • Pijnappel, Ruud, Primary supervisor
  • van den Bosch, Maurice, Supervisor
  • Veldhuis, WB, Co-supervisor
Award date16 Nov 2017
Publisher
Print ISBNs78-90-393-6848-0
Publication statusPublished - 16 Nov 2017

Keywords

  • Breast cancer
  • magnetic resonance imaging
  • 7 tesla field
  • optoacoustic imaging

Fingerprint

Dive into the research topics of 'Novel Imaging Strategies for Breast Cancer Patients'. Together they form a unique fingerprint.

Cite this