Regulatory Bioinformatics with HIVE

Derisk the Submission of your Precision Medicine Blockbuster

Due to unreliable sequencing assays and the lack of a standardized approach to genomic analytics pipelines, FDA does not trust genomic data submissions and conducts its own independent analysis. This analysis is performed with Embleema’s HIVE [1], the only authorized tool used by the U.S. Food and Drug Administration to conduct regulatory data reviews involving NGS (Next Generation Sequencing).

What do we mean by critical data and analytics issues in genomics based therapeutics?

Genomic based therapeutics offer significant challenges in data integrity and analytics for the regulator. At the source, multiple NGS platforms employ different chemistries yielding different assay performances. Lack of NGS pipeline analysis standards lead to each bioinformatics team coming up with their own pipeline to answer the same question, and inclusion of proprietary pipelines not publicly available make it even more difficult for the regulator to assess the scientific validity of the analysis.

Because of the inherent complexity and unstability of NGS analytics, FDA conducts an independent analysis pertaining to the quality and consistency of the data and analytics, but also investigates potential patient safety and efficacy issues from the same data set. HIVE was built from 2012 onwards to conduct these regulatory analyses. 

A new entity for treating HEP-C, was the first instance of a NGS-based therapeutics presented to the FDA in 2013. HIVE was used for the first time by the FDA to conduct the quality control of the data submitted. By further analyzing the data with HIVE, FDA discovered a viral resistance [2], a major finding, that generated regulatory delays. 

In the case of an immuno-oncology drug submission, by using HIVE’s novel classification tool called RLDA which runs deeper multivariate pathway analyses than other tools available on the marketplace, FDA reviewers discovered a comprehensive genetic panel with +90% sensitivity which should be included in the diagnostics that were missed by the submitting sponsor which also had to change their label.

HIVE RLDA Classifier detected 4 additional genes that are also up-regulated and down-regulated and that should be included in the diagnostic pane

How does HIVE Submission Assurance work?

HIVE Submission Assurance provides a blinded, comprehensive analysis and problem determination of your clinical trial data set, including your STDM files, images, lab results, genomic data and algorithms used to analyze your data sets. 

Because HIVE is used by the FDA to conduct their regulatory analysis on your data, you will benefit by using the same Quality Control checks, ontology mapping, data provenance frameworks, analytics library and other tools used by the FDA to perform its quality control on NGS based data submissions.

Beyond retrospective quality control, HIVE also allows you to detect potential issues such as patient safety issues that would go undetected before you submit and that the FDA would identify during the regulatory review. 

To even further align with the FDA, HIVE can also package your NGS data set in accepted standard formats such as the HIVE Pack and BioCompute [3].

FDA Standard
Operating Procedures

Biological Sample Mapping

Quality Controls

Contaminations and wrong experiments

Alignment Hexagon

Classification MC-RLDA

Issues

Mislabelling, missing data, wrong formats

Low Quality NGS submissions

Contaminations and wrong experiments

Missing resistance, pathogens

Incomplete diagnostics panels

Demo

Wrong type of samples

Low quality NGS submissions

Mix-up of samples from different experiments

Revirulence and resistance detection

Immuno-oncology

Uniqueness of HIVE approach

BioSample and BioCompute FDA standards

Automated deep QC protocols

Integrated taxonomy identification

Sensitivity and performance

Multivariant pathway diagnostics

HIVE is an integrated bioinformatics platform that mirrors the FDA analytical steps to verify genomics data and analytics during a regulatory review

HIVE MultiQC verifies the quality of the sequencing data: the results on the left were satisfactory whereas the results on the right were rejected

HIVE Hexagon is the most sensitive alignment tool and detect mutations which would not appear in analyses perform by other tools such as BOWTIE or BWA

By adopting HIVE, you have the unique opportunity to speak the same bioinformatics language as the FDA & bring a much greater level of confidence on the acceptability of your data set.

Submit

Help define verification protocol, select the appropriate tools in HIVE and provide training to use HIVE for running the analyses

Solve

Provide comprehensive documentation for each issue and provide you a set of actionable actions to solve them

Our regulatory bioinformatics team led by Dr. Vahan Simonyan, former director of bioinformatics R&D at the FDA and co-author of HIVE, will help you define your verification protocol, select the appropriate tools in HIVE and train you to use HIVE for running the analyses. We will also provide a comprehensive documentation for each issue and provide you a set of actionable actions to solve them. 

Schedule a demo to see HIVE in action.












    [1] High-performance integrated virtual environment (HIVE): a robust infrastructure for next-generation sequence data analysis Vahan Simonyan, Konstantin Chumakov, […], and Raja Mazumder

    [2] Clinical evidence and bioinformatics characterization of potential hepatitis C virus resistance pathways for sofosbuvir. Donaldson EF1, Harrington PR, O’Rear JJ, Naeger LK.

    [3 ]h​ttps://biocomputeobject.org/about.html