Personalized Treatment in High-Risk Mammary Cancer - the role of medical imaging
Breast cancer is the most common cancer among women in Norway, with 3000 new cases diagnosed annually, including 5 - 10% stage III diagnosis. Successful tumor downstaging by pre-surgical medical therapy, currently performed through medical therapy with chemotherapeutics or endocrine drugs, significantly lowers the risk of a locoregional relapse.Project Rationale: Individual monitoring of response to therapy is an integral part of neoadjuvant breast cancer treatment, where early response assessment allows rapid discontinuation or change of drug treatment for non-responders. Medical imaging offers non-invasive, quantitative information which offers potential for identification of predictive biomarkers. Recruitment: Currently, 120 women have been recruited into the PeTreMac study across at participating sites. Recruitment is expected to continue until spring 2019. Imaging Recruitment in Trondheim: As of January 2018, 22 patients have been approached and recruited for the quantitative imaging protocol, with 11 having attended at least one imaging session and the majority with 2+. Hybrid PET-MR data collected includes 18-FDG uptake, gadolinium-enhanced MRI, morphological MR imaging (T1 and T2), and the extended diffusion-weighted imaging protocol developed at NTNU in collaboration with St. Olavs hospital. Recruitment for MR imaging in additional centres (Stavanger, Bergen) for a compatible diffusion imaging protocol is also underway, and the collected data will be transferred to NTNU for inclusion in the analysis. Progress and Output: Analysis of imaging biomarkers from a related repeatability cohort acquired in conjunction with PeTreMac study has provided a structure for the analysis pipeline of data for PeTreMac data, including incorporation of distortion correction for diffusion images and advanced modelling of diffusion signals. Retrospective reconstruction of dynamic PET data successfully matches diffusion and Gd-enhanced MR images with intrinsic registration and acquisition time. Analysis of the repeatability (and thus utility) of diffusion-derived biomarkers, together with an assessment of physiological noise present in diffusion data, have formed the basis of two research abstracts that have been submitted for the annual meeting of the international society for magnetic resonance in medicine by the post-doctoral researcher working on the PeTreMac study. Under the supervision of the same post-doctoral researcher, a student in the group has developed an exploratory mixture model technique for leveraging the specificity and low background of PET data to inform lesion delineation in diffusion images. This work has also been submitted as an abstract to the same conference, and demonstrates good agreement of lesion diffusion coefficients for several longitudinal patients in the PeTreMac cohort.
Breast cancer is the most common type of cancer among women in the Western world, including Norway, where nearly 3000 new cases are diagnosed annually. Within this group, between 5 and 10% are diagnosed at stage III. These patients have a poor prognosis. Moreover, pre-surgical tumor downstaging by medical therapy prior to surgery significantly lowers the risk of a locoregional relapse. This is currently done through medical therapy with chemotherapeutics or endocrine drugs.The prospective, multicenter PeTreMaC study aims at identifying the role of novel, targeted drugs in neoadjuvant breast cancer therapy. Based on novel stratification criteria, the outcome of personalized therapy regimens will be evaluated and genetic information will be collected in order to increase our understanding of factors that contribute to response and resistance to treatment. Since response and resistance to targeted drugs and chemotherapy cannot be accurately predicted using current biomarkers, individual monitoring of response to therapy is an integral part of neoadjuvant breast cancer treatment. Early assessment of response can identify non-responders and allow rapid discontinuation or change of drug treatment. Medical imaging offers noninvasive assessment of the disease and extraction of quantitative information through systematic application of standardized imaging protocols in clinical trials has been proposed as a novel strategy for identification of predictive biomarkers. A hybrid PET/MRI examination, reporting on tissue microstructure and tumor metabolism, represents the most comprehensive imaging modality to date and is therefore considered likely to provide clinically relevant, quantitative biomarkers. The increasing use of targeted drugs which may initiate late response in size reduction compared to traditional chemotherapy warrants research into functional, preferably noninvasive, measures of treatment response. PET and MRI are suitable medical imaging modalities for this, and the PeTreMaC study is well designed to address this challenge. Hypothesis of the image part of the PETREMAC study: The outcome of neoadjuvant therapy with targeted drugs can be predicted from pre-treatment or early assessed imaging biomarkers of response using PET, DW-MRI and/or DCE-MRI. Objectives: To identify predictive molecular markers for treatment response/resistance To assess potential factors predicting relapse-free and overall survival at follow-up To develop strategies for noninvasive therapy monitoring using MRI and PET The clinical study is led by Hans Petter Eikesdal, Haukeland University Hospital, and patients will be recruited from several participating hospitals. Steinar Lundgren is the local clinical PI. The research part focusing on quantitative imaging is led from Trondheim (PI Tone F.Bathen). Currently, about 40 patients have been recruited to the study, the first patients from St. Olavs Hospital were recruited in January 2017. A Post Doc experienced in clinical MRI and image analysis started November 2016. PET/MRI protocols are implemented, and the work regarding image analysis pipelines are in progress.