NEW YORK (GenomeWeb) – Molecular Health is putting the finishing touches on a new product for clinical trials which it plans to launch in the first quarter of 2019.
The solution, called Molecular Health Guide Predictive Engine, uses machine learning methods to predict the potential outcomes and likelihood of success of clinical trials, Rudolf Caspary, the company's chief information officer, explained in an interview.
Caspary said that the product is intended for use by researchers in the pharmaceutical and biotechnology industry as well as clinical research organizations to help them make decisions about which clinical trials to pursue as well as which to stop. The software could also help potential investors identify trials that are likely to have a successful outcome and result in approved drugs.
Caspary also claimed that the company has benchmarked the software and that it has outperformed expectations. However, Molecular Health is not disclosing specific details about these benchmarks.
Molecular Health is still mulling pricing as well as the exact business model for MH Predictive Engine. Caspary said that the company is considering charging an unnamed cost per transaction as well as offering a volume-based pricing option. Like Molecular Health's other products, MH Predictive Engine will be marked under a software-as-a-service model.
In 2016, Molecular Health retired the wet laboratory component of its business opting to focus solely on selling clinical decision software solutions. That same year, the company released several products as part of its broader platform called Molecular Health Guide. These products use many of the same capabilities the company offered as part of its NGS-based oncology testing service. Its portfolio includes Molecular Health Guide Workbench Manager, which helps clinical laboratories manage pipelines and workflows for analyzing NGS data from tumor samples through to generating custom reports. A second product is the Molecular Health Guide Clinical Annotation service solution used by high-throughput labs to interpret and annotate actionable variants found in tumor samples.
A third product, launched earlier this month, is Molecular Health Mendel (MH Mendel) which is designed to analyze sequence data in the context of hereditary cancers and other germline disorders. The tool considers data from various sources to help geneticists assess and classify variants detected in patient data. Specifically, it uses information on population frequencies, reported clinical significance, and functional impact predictions to interpret variants. Geneticists can use the tool to manage variant classifications and interpretations as well as generate diagnostic reports for their cases.
All the company's products including the software it plans to release in January are based on its cloud-based Dataome infrastructure. Dataome comprises a series of databases and analytics capabilities that integrate data gleaned from various public and proprietary databases of molecular, clinical, and drug outcomes information. It includes data on protein interactions, biomarkers, cancer indications, drugs, clinical trials, variant classifications based on ACMG guidelines, and more. The next release of Dataome will include updates that enable users to better understand underlying molecular mechanisms and reactions, Caspary said. The company updates Dataome monthly.
Molecular Health's platform has been cleared for use as an in vitro diagnostic in Europe, which allows customers of the MH Guide seek reimbursement from insurance companies for use of the software. The company's products are used by customers in places like Charité University Hospital, University of Greifswald Hospital, and University of Hamburg Hospital.
Last year, Molecular Health announced that the US Food and Drug Administration had extended an ongoing research collaboration with the company. The product, which was initially called SafetyMap then renamed EngineusEffect and now called Molecular Health Effect (MH Effect), is used for analyzing drug-induced adverse events and for predicting possible safety issues of new drug candidates. The initial research agreement with the FDA's Center for Drug Evaluation and Research was signed in 2012. During the first five years of the collaboration, the FDA used MH Effect to evaluate potential mechanisms of emerging drug safety issues.
The updated agreement extends the collaboration for another five years. Under the terms of the new partnership, the FDA will continue to assess MH Effect's ability to predict adverse events and safety label changes. Molecular Health said that the results of the FDA collaboration will be incorporated into its technology improve its computational capabilities for use in drug development efforts. Molecular Health CEO Friedrich von Bohlen noted at the time that the FDA's decision to continue its partnership with the company around MH Effect highlights the value of the solution for predicting and characterizing drug adverse events.
The company points to papers such as this one published in 2016 by researchers from Reutlingen University, University of St. Gallen, and Novartis Institute for Biomedical research which pointed to efficacy and safety issues as some of the most probable reasons for failure in phase II and phase III drug development. "With MH Effect and our other products and technologies creating novel insights from data, we will continue to help improve the development of novel medicines," von Bohlen said in a statement.