OncoGenomX, Inc., Allschwil CH - Leading the way to more successful drug development with predictive analytics.

Pharmaceutical companies have been resorting to machine-made intelligence since the 1960s. What is remarkable is the increased attention artificial intelligence (AI) has received from pharmaceutical companies since 2016.

This newfound interest was largely sparked by the groundbreaking advancements that have been made in the AI field. Since then, AI has revolutionized pharmaceutical research and development in every way.

Artificial intelligence and drug development today

Artificial intelligence (AI) has already shown its transformative capabilities in widening the scope of drug targets, molecular structures, therapeutics, identifying biomarkers, and many others. It is used in many aspects of drug development including:
(i) Computer-aided design of molecules with desired properties. or repurposing existing molecules.
(ii) Making predictions to help with the effective synthesis and production of drugs.
(iii) Large-scale prediction and comparison of drug activity spectra.
(iv) Data mining.

The success of AI in solving R&D problems depends on high data quality. The present challenge is to establish the causal link between the mechanism of treatment, validation models, and clinical goals to achieve sustained disease remission if not complete tumor elimination.

Pharmaceutical R&D is fraught with expensive failures

Over 90% of pharmaceutical R&D efforts end in failure. 60% of failures are attributed to unsuccessful clinical trials. Out of five FDA submissions four fail due to insufficient evidence of antitumor efficacy or serious side effects. Three out of four commercially available cancer drugs are withdrawn from the market after they are approved because they do not meet the clinical expectations of prescribers and patients.

Moreover, innovative anticancer drugs typically require 12 years to reach the market. 4 to 5 years for preclinical development and 6-8 years for clinical development.

The average development cost is 750 million USD. Development failures included the costs per successful new drug application amount to 3 billion USD. Preclinical research accounts for 30% of the budget, while clinical development accounts for 65% of the budget.

These statistics point to the urgent need for innovative approaches and solutions to make drug development clinically successful.

AI and pharmaceutical R&D - Pulling the right lever

Today’s chemistry, data mining, and process centric improvement measures did not help much with the goal of fast and successful low-cost R&D. Firstly, because the overall cost of drug research and development is insignificantly decreased. Secondly, and more importantly the clinical success rate is barely affected.

We believe that the challenges in cancer drug discovery & development can only be overcome if we could predict earlier and better which innovative drugs are most likely to be effective and in which patient tumors. More than on quantity clinical R&D success hinges on quality.

Bender & Cortes-Ciriano research supports this conclusion as well: Even a small improvement in quality would make a significant difference to the overall speed, cost-effectiveness, and success of pharmaceutical cancer drug development.Clinical success requires clinical relevance

Oncology offers a wide range of molecularly targeted anticancer drugs. But these are exceptional situations where the drug-specific target is also the causal mechanistic driver of the target tumor.

So, what is needed to advance anticancer drugs with high clinical success potential?

(i) We need a clearly defined clinical problem. Tumors are distinct biologic entities with specific clinical problems. Therefore, drug targets and treatment mechanisms need to be tumor and problem specific.
(ii) A molecular target that not only impels the clinical problem but qualifies for a pharmacological manipulation likely to eliminate the clinical problem.
(iii) A molecular target that is integral to killing all cancer cells in a tumor to facilitate complete tumor elimination.
(iv) A granular description of the patient tumors for which an innovative drug is likely to work.
(v) A list of preferred treatment mechanisms, standard or investigational, likely to maximize the clinical effectiveness of an innovative drug.

OncoGenomX: Precise and accurate predictions for clinical R&D success

We at OncoGenomX are committed to assisting cancer therapeutics companies on their path to fast, clinically successful, and low-cost R&D. To do this, we've assembled a team of specialists who have more than 80 years of complementary industry experience between them. Our value proposition is that we offer clinic and success-oriented predictions throughout the pharmaceutical value chain.

Predictive computational cancer models

Our computational cancer models are rooted in human data. Informed by science, broad research, experience, and appreciation of clinical needs. We blend cancer target biology, clinical goal, and mechanistic understanding of the levers of clinical success into predictive cancer models.

OncoGenomX embraces a multiplicity of perspectives, offering decision guidance for drug developers through a hybrid of artificial intelligence, machine learning, causal prediction, and benchmarking with outcome experience.

Mediating between different data worlds

It is not enough to rely on AI-assisted predictions to forecast the likelihood of an innovative drug going to work and for which tumors.

We have the tools necessary to identify preclinical models that mirror the clinical problem to validate the intended treatment hypothesis through experimental verification.

Reimagining cancer drug development

We don't expect our approach to eliminate all of today’s R&D challenges. But we believe that innovative prediction technology must be brought to the decision-making process to advance more successful anticancer drugs.

A 20% improvement in clinical success makes a difference for pharmaceutical giants in their bottom line. While it could make the difference between demise or survival for a small pharmaceutical enterprise.

OncoGenomX - aiR&D support to foster pharmaceutical R&D success

Making drug development decisions of sufficient quality is all but trivial. Not always is the knowledge base compatible with the goal of quality R&D decisions. If you learned of an opportunity to reduce the risk of late R&D failures, what reason is there not to take advantage of it?

To know more about our aiR&D support service, you may contact us at info@oncogenomx.ch.