Continuous cancer monitoring in vivo

Materials World magazine
1 Jul 2009
In vivo images for a control (a) and tumor (b) device three days after being implanted. Superimposed over the device is a pseudo-colourised map of the transverse relaxation time (T2) within the device (colour bar on the left). The T2 of the sample device is lower than the control device, indicating the presence of human chorionic gonadotrophic at the tumor site

Non-invasive, continuous, in vivo monitoring at the site of cancerous tumours may eventually be possible thanks to research at MIT, USA. Scientists are looking into what they say is the first implantable device that can fulfil this function, superparamagnetic iron oxide nanoparticles.

They believe the instrument could overcome the limitations of biopsies, which only offer a snapshot of the tumour at a single moment in time. Spotting tumour growth after surgery, they say, is difficult and takes longer using standard magnetic resonance imaging (MRI), because these growths are not distinguishable from benign lesions caused by chemoradiotherapy.

Implanting the MIT device during tumour resection could enable reporting on the local environment in the following weeks and months.

The prototype, about the size of a grain of rice, is made up of a biocompatible injection moulded high-density polyethylene (HDPE) substrate with a reservoir cut into it that is covered by a semi-permeable polycarbonate membrane with pores of about 10nm and 6x108 pores/cm2.

Superparamagnetic iron oxide nanoparticles are enclosed in the reservoir and act as magnetic relaxation switches (MRSw). ‘We attach antibodies for specific molecular targets onto the surface of the nanoparticles,’ explains Professor Michael Cima at MIT. ‘The antibodies then bind to specific epitopes on the target [biomarker] and cause the nanoparticles to be aggregated to [them].’ These aggregated particles cause a decrease in the transverse relaxation time (T2) – a signal that is picked up by MRI.

While the biomarkers can diffuse into the device, the nanoparticles cannot leak into the surrounding tissue, preventing an immune response. This means any fluctuations in T2 due to MRSw concentrations are solely attributed to aggregation in sensing the targetted molecule.

Target practice

Proof of principle tests in mice have involved implanting xenographic tumours that were known to secrete human chorionic gonadotrophic (HCG) – an indicator of testicular and ovarian cancer in increased concentrations. In comparison to the control devices in mice without tumours, those implanted near a tumour showed a decrease in the T2 (see image above).

The nanoparticles could be tailored to detect if chemotheraphy drugs are reaching tumours, pH or oxygen levels (which reveal tumour metabolism) and tumour metabolites.

Personal touch

Professor Tony Cass, Deputy Director of the Institute of Biomedical Engineering at Imperial College London, UK, believes this is an ‘interesting example of the personalisation of medical treatment to individual response’.

He says that the focus on detecting ‘biochemical changes and [whether] residual pieces of the tumour are growing again is novel, combining well established imaging technology with knowledge about molecular markers. It also offers an interesting approach to combining imaging with sensing technology’.

Cass believes that the key issues for clinical practice are sensitivity to biomarkers at low concentrations for early detection and long-term stability in the body due to the fragility of some antibodies.

The team at MIT report that the T2 signal measured represents total exposure to the analyte over time and so could be made more sensitive to lower concentrations, which are currently detectable but at a slower rate.

Cima adds, ‘We have only explored stability from the HCG device for six weeks, which was observed to be the limit for a good signal. We believe that is due to the antibody, although there may also be a problem with the chemistry used to link [it]’.

He says, ‘Some applications require only a few weeks of response [and] the HCG marker for cancer is not common [enough] to make it a major focus for long-term stability. It would be nice to have a full year for other applications. This depends on the antibody with some being much more stable.

‘The pH sensor has no such issues [as] there is no need for an antibody. The physical chemistry of the solution changes the relaxivity’.

The researchers are also exploring the potential for a wireless implant that can be read by a hand-held device.