A new technology developed by Harvard Medical School researchers at the Massachusetts General Hospital Center for Systems Biology allows the simultaneous analysis of hundreds of cancer-related protein markers from minuscule patient samples gathered through minimally invasive methods. This powerful and sensitive technology uses antibodies linked to unique DNA ‘barcodes’ to detect a wide range of target proteins.
It could serve as a tool to help clinicians gain insights into the biology of cancer progression as well as determine why certain cancer therapies stop working or are ineffective to begin with.
Minimally invasive techniques—such as fine-needle aspiration or circulating tumour cell analysis—are increasingly employed to track treatment response over time in clinical trials, as the tests can be simple and cheap to perform. Fine-needle aspirates are also much less invasive than core biopsies or surgical biopsies, since very small needles are used. The challenge has been to comprehensively analyse the very few cells that are obtained via this method.
‘What this study sought to achieve was to vastly expand the information that we can obtain from just a few cells,’ explained Cesar Castro, HMS instructor in medicine at Mass General and a co-author of the paper. ‘Instead of trying to procure more tissue to study, we shrank the analysis process so that it could now be performed on a few cells.’
Until now, pathologists have been able to examine only a handful of protein markers at a time for tumour analyses. With this new technology, the researchers have demonstrated the ability to look at hundreds of markers simultaneously, down to the single-cell level.
‘We are no longer limited by the scant cell quantities procured through minimally invasive procedures,’ Castro said. ‘Rather, the bottleneck will now be our own understanding of the various pathways involved in disease progression and drug target modulation.’
The new method uses an approach known as DNA-barcoded antibody sensing, in which unique DNA sequences are attached to antibodies against known cancer marker proteins. The DNA ‘barcodes’ are linked to the antibodies with a special type of glue that breaks apart when exposed to light. When mixed with a tumour sample, the antibodies seek out and bind to their targets; then a light pulse releases the unique DNA barcodes of these bound antibodies that are subsequently tagged with fluorescently labelled complementary barcodes. The tagged barcodes can be detected and quantified via imaging, revealing which markers are present in the sample.
After initially demonstrating and validating the technique’s feasibility in cell lines and single cells, the team tested it on samples from patients with lung cancer. The technology was able to reflect the great heterogeneity—differences in features such as cell-surface protein expression—of cells within a single tumour and to reveal significant differences in protein expression between tumours that appeared identical under the microscope. Examination of cells taken at various time points from participants in a clinical trial of a targeted therapy drug revealed patterns that distinguished those who did and did not respond to treatment.
‘We showed that this technology works well beyond the highly regulated laboratory environment, extending into early-phase clinical trials,’ said Castro, who is also a medical oncologist in the Mass General Cancer Center and director of the Cancer Program within the hospital’s Center for Systems Biology. ‘In this era of personalised medicine, we could leverage such technology not only to monitor but actually to predict treatment response. By obtaining samples from patients before initiating therapy and then exposing them to different chemo-therapeutics or targeted therapies, we could select the most appropriate therapy for individual patients.’ Harvard Medical School