EP Vantage interview – Ariana’s algorithms could deliver bespoke cancer therapy


The advent of targeted cancer therapies has necessitated the development of companion diagnostics to identify those patients whose genetic makeup will allow them to benefit. But a patient’s gene signature is only one factor of many capable of affecting their response to a treatment.

There is a demand, says Mohammad Afshar, CEO of Ariana Pharma, for a tool that will analyse all of a patient’s characteristics at once – physiological, genetic, genomic, proteomic – and assign them the drug or procedure that will benefit them most. He tells EP Vantage: “What we’re working on here is the ability to look at all the data that is available on these particular tumours and patients, and combine this information together to identify the best treatment.”

Winther time

Spun out of the Institut Pasteur ten years ago, Ariana is at the forefront of bioinformatics, developing algorithms to help stratify patients for three distinct purposes. First, the firm can help pharma companies recruit the right patients to their clinical trials so that they can demonstrate that a compound works in patients with a particular mutation or other characteristic. Secondly, the technology can be used to aid the development of diagnostics.

And thirdly, the company is working towards the creation of a piece of software that doctors can use to guide patients towards the therapy that is best for them. To this end, it has just begun a 200-patient clinical trial in collaboration with the WIN Consortium, a non-profit organisation focused on genomic-based cancer research.

The trial, called Winther, spans five countries and will enrol patients with solid tumours that have been resistant to previous treatment, Mr Afshar says. “We are testing a new algorithm we’re developing in order to identify the best treatment for each of these patients. This is the first time we have a clinical trial of that size which is testing an algorithm.”

Currently, the choice of a particular treatment is based on factors such as the localisation of the tumour, its histology, and a number of single markers seen in blood or tissue samples, Mr Afshar says. “What we’re working on here is the ability to look at all the data that is available on these particular tumours and patients, combine it and then identify the best treatment. It’s trying to grasp this complexity and put it into something very simple that can be used by clinicians.”

The tool would be web-based, he says. “The clinicians enter a number of parameters, which could be a measurement, biochemical or a particular test that has been done. These are uploaded into the system which then securely delivers a decision.”

Service provider

Initial data from Winther could come next year. We’re going through the early recruitment phase now. We’re expecting to have finished the initial enrolment by April of next year, and then the patients will be followed for two years,” Mr Afshar says. Preliminary data should be available by the end of next year, with final results in 2015.

“A successful readout would be to show that the patients that follow the choice which is made with the algorithm have better overall survival or event-free survival compared with the standard treatment. That would be a major success.”

This treatment-deciding algorithm is still some way off commercialisation. Fortunately, Ariana has its other businesses: aiding pharma and diagnostics companies.

“We provide fee-for-service interactions with pharma and biotech and people developing diagnostic kits. Our short-term revenue is generated through the fee-for-service work we do for a number of clients in the US and Europe. Then we’re getting royalties on sales of diagnostic kits that have been put on the market by some of our partners, and the third line is the software that we’re developing that will be put on the market where the clinicians can access it directly.”

For example, Ariana worked with Fovea Pharmaceuticals to help identify responders to a drug in a phase II trial, Mr Afshar says, and to investigate the best measures of the efficacy of the drug. “That was extremely successful, and the company was subsequently acquired by Sanofi.”

The Paris company has also collaborated with diagnostic companies on multimarker tests, helping them develop a signature which is robust and can be confirmed in subsequent clinical trials. “Bio-Rad is a good example,” Mr Afshar says. “We started working with them a few years ago and developed a hepatitis C diagnostic kit which hopefully will be hitting the market sometime at the beginning of next year.”


Ariana has recently gained a huge fillip from the FDA in the shape of guidance on what is called clinical trial enrichment: the idea that if patients’ responses to a treatment can be better predicted, trials can be designed more efficiently and drugs get to market more quickly.

“When we started, the kind of work we used to do was seen as very esoteric,” Mr Afshar says. “Over the last five years there’s been a really dramatic change.

“This has been quite a big change in thinking since we started. It’s having a huge impact on our business. We’re now doing something which is accepted, and we’re one of the leaders in this field. The person from the FDA sent me the guidelines and said, ‘Mohammad, the first draft is out, just use it as your brochure’.”

Ariana’s business is having a moment in the sun, but more importantly the company’s work ought to bring tailored therapy closer to becoming the standard of care. “The Winther trial is the first time that we will see informatics helping the promise of personalised medicine actually happen. We’ve been talking about it for a long time and this is the first time we can see it really kicking in.”

Trial name Trial ID
Winther NCT01856296

To contact the writer of this story email Elizabeth Cairns in London at elizabethc@epvantage.com or follow @LizEPVantage on Twitter

Share This Article