Pilot study for molecular screening in metastatic breast cancer. What did we learn?

Intro text: 


In 2014 the BIG network launched AURORA, the largest international research programme based on molecular screening and dedicated to improving our understanding of metastatic breast cancer. Prior to the launch of AURORA, BIG conducted a screening pilot study, involving 30 patients, to test the logistical and technical aspects of the programme’s infrastructure. This pilot, made possible through a grant from the Breast Cancer Research Foundation®, started in May 2013 and lasted about 1 year.

Prof. Sherene Loi* and Prof. Christos Sotiriou*, Principal Investigators of the pilot study, tell us more about the results and conclusions.

* Peter MacCallum Cancer Centre, Melbourne, Australia
* Institut Jules Bordet, Brussels, Belgium


BIG: What is unique about the AURORA program?

S. Loi: AURORA is unique as it offers targeted gene sequencing (TGS) for patients with metastatic breast cancer at a large number of sites across Europe. It will also collect clinical outcome data, which will be important for understanding the clinical relevance of genetic alterations in breast cancer.

C. Sotiriou: Thanks to the BIG network, approximately 80 European centers have accepted to join this challenging effort. Sample collection is at the heart of the programme, with a unique aspect being our collection of both primary and metastatic samples for comparison purposes. This, together with the collection of clinical outcome data over a period of 10 years, will not only help us to better understand these genomic aberrations, but it should help us develop better treatments for metastatic breast cancer in the future.  

Why was it important to run a pilot study?

S. Loi: Given that this is a multi-institutional study with a new type of technology, a pilot was necessary to make sure that the TGS was accurate and the results could be delivered in a reasonable time frame.

C. Sotiriou: It was also important to test the feasibility of obtaining biopsies of sufficient quality in order to perform 'real time' TGS as well as to challenge several aspects of the multicentric logistics. During this pilot phase, a specific IT infrastructure was also developed to support patient enrolment, logistics and transfer of molecular data. This platform was specifically designed for the physicians, biologists and bioinformatics professionals who are generating and gathering the results. This is now in use in AURORA!

Copy Number Aberration (CNA)* detection by TGS looks challenging. Why?

C. Sotiriou: There are several variables that may explain this. First of all, technical ones. For instance, the TGS requires a polymerase chain reaction (PCR) amplification step that could bias the CNA estimates. Another reason is the low coverage of the entire genome with TGS when compared to the DNA microarray technique called ‘SNP’*. Finally, the quality of the starting material is a non-negligible factor. Since DNA extracted from paraffin-embedded and frozen samples is degraded to some extent, TGS is not as optimal as SNP, which has been proven to work with even the most degraded samples.

S. Loi: Exome sequencing° data was not designed to accurately determine copy number. This may be helped by deeper sequencing or doing whole genome or SNP arrays. CNA is important information in breast cancer, however, and should not be discounted.

* CNAs are variations of the number of copies in one or more sections of the DNA. The gene copy can be either amplified (being duplicated and present more than the normal number; ex. A-B-C-C-D) or deleted (fewer than the normal number; ex. A-B-D [deletion of ‘C’]). These copy number aberrations can be detected by using microarray-based genetic analysis. Recent advances in DNA sequencing technology have further enabled the identification of CNAs by modern technologies, also called 'next-generation sequencing'. 
* SNP-array: technology in molecular biology used to amplify a single copy or a few copies of a piece of DNA
° Exome sequencing: technique for sequencing all the protein-coding genes in a genome (known as the exome)


What are the main conclusions of the AURORA pilot study?

C. Sotiriou: I think that we succeeded to demonstrate the feasibility of running a trans-European molecular screening program. Although the sequencing and clinical results across the different platforms were reasonably consistent, we did find some discrepancies. This highlights the need for better standardization and harmonization of the whole procedure, from the gene sequencing to bioinformatics and clinical interpretation.

What have you learned in the pilot that might have to be adapted for AURORA?

S. Loi: I think calling of clinically relevant variants is complex and can be misleading for clinicians and patients. A proper algorithm and/or validation system will need to be incorporated into AURORA. 

C. Sotiriou: We should limit the screening failure rate. Close monitoring of failures will be performed in AURORA in order to make necessary adjustments and improve sample quality overtime. 
Indeed, mutation calling and the clinical interpretation of the results remains an issue. Better bioinformatics and validation should be integrated into AURORA. As mentioned earlier, the estimation of CNAs from TGS is a concern. We are still looking into ways to improve this. This issue also needs to be considered in AURORA.

"The ultimate goal of AURORA is to demonstrate in a prospective manner that molecular screening programmes will improve the treatment care of our breast cancer patients."

Why will the results of AURORA be important for patients in the future?

S. Loi: Results of AURORA will be important as we will learn a lot about the clinical and biological importance of TGS sequencing of breast cancer and its clinical relevance.

C. Sotiriou: Well, very little is known regarding the molecular alterations that characterize metastatic breast cancer. We will learn a lot about breast cancer biology, its natural history, and response to standard and/or targeted agents, which will ultimately lead to better treatments and care.

Molecular screening programmes are now becoming very popular. Why has it become so important to screen breast cancer patients for molecular alterations?

S. Loi: We are still learning the clinical and biological importance of such screening. At present, the main aim is to understand prognosis and biology in a prospective manner as well as response to targeted drugs. In general, outside of AURORA, clinicians should not order such tests unless they know what they will do with the results. Usually, access to a drug if a targetable aberration is found is the critical issue when screening is used, and hence large tertiary referral centres with good phase I/II programmes are best placed to have screening programmes in place.

C. Sotiriou: I think that the ultimate goal of AURORA is to demonstrate in a prospective manner that molecular screening programmes will improve the treatment care of our breast cancer patients.



The pilot screening programme’s raison d’être

Advances in genomics have recently led to a drastic redefinition of the way cancers are classified. Cancer types that were once considered to be homogeneous subgroups are now known to be heterogeneous, with different molecular aberrations driving the disease and influencing the prognosis and the sensitivity to drugs. In parallel, drugs that specifically target some of the aberrations are being developed, sometimes showing remarkable efficacy. One consequence of this evolution is that clinical trials are increasingly conducted in small molecularly-defined subsets of patients.


The pilot study: all bases covered

All the relevant aspects of a molecular screening programme were covered in the pilot study, from the informed consent procedure, to patient recruitment and the collection of multiple biological samples, including paraffin-embedded and frozen core biopsies of the metastatic lesions as well as blood. The study also encompassed the shipping of samples and assays and reported the results to the treating clinicians involved. All results were uploaded, stored and assessed on the tailor-made IT infrastructure called the Molecular Screening Prototype Platform (MSPP). The MSPP’s user-friendliness was also evaluated, with a view towards using the screening results to identify patients for potential participation in future molecularly- defined clinical trials for metastatic breast cancer.


Who carried out the pilot study?

Four European laboratories were chosen to perform pathology and molecular tests, including next generation sequencing of metastatic lesion DNA: Institut Jules Bordet (Belgium), the Institute of Pathology and Genetics in Charleroi (Belgium), the European Institute of Oncology in Milan (Italy), and the Wellcome Trust Sanger Institute (UK). In addition, four European centres participated in the recruitment of patients: Institut Jules Bordet (Belgium), Vall d’Hebron hospital in Barcelona (Spain), the Offenbach clinic in Frankfurt (Germany) and Dundee University in the UK. The pilot study was sponsored and coordinated by Institut Jules Bordet, developed and run under the auspices of the Breast International Group, and funded by the Breast Cancer Research Foundation.