Personalizing cancer therapy

We have recently launched a novel personalised therapy clinical trial to treat patients with late-stage colorectal cancer at George Mason University's Center for Applied Proteomics and Molecular Medicine (CAPMM). This is one of the most cutting-edge personalised therapy clinical trials in the world for several reasons including:

• it uses protein information rather than genomic information to guide therapy

• the activation state of the protein, not its total amounts, is used to guide therapy

• therapy is tailored to the molecular signature of the metastatic lesion rather than the primary tumour

• the input to the analysis is tumour cells isolated from the metastatic lesion itself using a technique we invented called laser capture microdissection (LCM)

• it uses a novel protein microarray, the reverse phase array (RPMA), to map the activation state of the protein network. This is the first trial in the world where functional protein pathway network mapping is used to guide and tailor therapy.

In particular, the trial will be testing Novartis's Gleevec, which is FDA-approved for chronic myelogenous leukemia and gastrointestinal tumours. According to research conducted at CAPMM, Gleevec targets disease pathways in tumour cells that are among those found in typically fatal liver metastasis in colorectal cancer patients.

Unravelling the human genome

For our, trial, a patient's metastasis is sampled using a needle biopsy and the pure tumour cells are isolated using laser capture microdissection. They are then solubilised and analysed with the reverse phase protein microarray, and the activation states of the Gleevec drug targets (ckit, cabl and PDGFr) are assessed. Gleevec is then given to patients with highly activated signatures while patients whose network is not activated are given the standard care therapy.

Gleevec would not normally have been considered for treatment in these cases. In the new paradigm of personalised therapy, however, the site of the cancer does not matter because it is the molecular signature that drives selection — true personalised therapy.

The primary objectives of the trial are to see if we can implement this workflow and generate a Gleevec drug target activation score for each patient, see how that score changes before and after therapy, and determine the safety of Gleevec combined with an anti-EGFR therapy. The primary therapy endpoint is disease stabilisation as per RECIST (response evaluation criteria in solid tumours). We hope to have results on a statistically significant number of patients by the end of the year.

Reverse phase protein microarray technology

Key to the trial is the RPMA drug target mapping technology, which we invented at the NIH, and further developed and optimised for clinical applications at the CAPMM. We are now finding that cancer is a heterogenous disease and each patient's tumour is uniquely different at the molecular level. Because you don't know which pathways are activated and driving tumour growth, you need to measure many things to find the correct culprits — RPMA can do this using only a tiny specimen.

RPMA involves printing nanolitres of cellular lysates of cells and body fluids. Using sophisticated amplification techniques, the technology can achieve the sensitivities of measuring the level of a protein in less then a 1-cell equivalent. We invented the technology to do something that no other technology can do right now: quantitatively measure and map the activation levels of 200–300 key signalling molecules, (the drug targets themselves) from a tiny biopsy specimen. Personalised therapy will not happen with large pieces of tissue — the real world setting requires biopsy-sized specimens taken in a radiologic outpatient setting.

RPMA can also modularise with key types of cellular fractionation technology, such as LCM. We have published on the need to use LCM to procure pure tumour cell isolates from the biopsy specimen because we found that the molecular signature of what is activated in the tumour can be contaminated with non-tumour cell components, which may destroy the key drug target activation information.

Nearly all of the new classes of cancer therapy (molecularly-targeted inhibitors, such as Gleevec) that have been developed — and comprise nearly all the pharma's industry's current pipeline — target proteins, not genes. Indeed, the gene expression and mutational analysis that have underpinned previous and current personalised medicine trials measure factors that can be very far removed from the actual drug target. The genes are not the drug target — it's the proteins, and it's not just the protein, the drugs turn off the proteins only when they are activated or in use.

With RPMA, we wanted to develop a technology that could directly measure what we really want to see: the activity of the protein drug target itself in patients' tumours.

Further personalised projects

At CAPMM, we are also involved in other personalised medicine projects. We have just launched a personalised therapy trial for metastatic breast cancer patients where the RPMA data is used to personalise therapy to any FDA-approved targeted therapy (not just Gleevec). We also have a novel personalised therapy project for multiple myeloma where the tumour cells are taken directly from a bone marrow aspirate and treated ex vivo with different targeted therapies to determine which treatments could be personalised for each patient. Lastly, in collaboration with Italy's national health institute, Istituto Superiore di Sanita, we are working on personalised therapy projects for lung, colorectal, breast, prostate and ovarian cancers, leukaemia and multiple myeloma. We hope to launch the first protein pathway-based personalised medicine clinical trials in Italy focusing initially on metastatic colorectal cancer.

Dr Emanuel Petricoin and Dr Lance Liotta are Professors of Life Sciences and Co-Directors at the Center for Applied Proteomics and Molecular Medicine, George Mason University (VA, USA)