Drug development is a very lengthy and expensive undertaking. Failure rate for novel drugs exceeds 95%. Therefore, successful drugs must cover the costs of these failures. As such, prescription drug prices have escalated at an alarming rate and show no signs of stopping. The need for successful drugs to cover failures also means that pharmaceutical companies primarily devote resources to pursuing drug candidates that have a large enough population to allow the company to earn a return on its investment. Thus, diseases that affect only a small portion of the populace are not investigated nearly as much as, say, oncology, cardiovascular or immunology. Current practice usually involves taking modeling techniques developed for small molecule research and trying to adapt them to biologics. However, this approach, more often than not, does not provide the scientist with predictions around feasibility and optimal drug properties, resulting in wasted effort pursuing leads that have no chance of making it through clinical trials, or to be reimbursed by payors. Applied BioMath has developed tools that address high value questions in the middle of the drug development pipeline. By coupling quantitative systems pharmacology techniques with high performance computing and sophisticated mathematical algorithms, we have proven an ability to predict optimal drug properties years before entering the clinic. For the past two years we have been offering our services to pharma and biotechs alike, to rave reviews. We have also been approached with inquiries to license our software. This project will fund the development of our proprietary algorithms and toolsets into a stable, standardized software platform, that can be automatically validated for GLP, for by biologics to develop their internal systems pharmacology models. At its heart, our toolsets are built on Kronecker Bio, an open source biophysical computational engine codeveloped by one of our Founders while pursuing his PhD in Biological Engineering with the Computer Science and Artificial Intelligence Lab from the Massachusetts Institute of Technology. This robust platform is currently in use, in its raw form, in the pharmaceutical industry but is limited in its adoption due to its lack of usability, quality contro and GLP validation. This project will focus on the application and presentation layer, allowing the underlying computational functionality to be easily accessed, utilized and understood, so capital requirements are less than a typical software development project. Achieving our goal of building this software platform is only the first step. What follows is a concerted push into the biologics segment, which we are currently seeding through our services offering and gaining a reputation as a firm that delivers high value on time. We have completed our second round of fundraising, raising a total of $1.8m between both rounds. This grant, plus the additional fundraising, will ensure that we are able to roll out our tools and assist drug companies in delivering bestinclass biologics, that meet unmet medical need, on an accelerated timeline to provide patients with a better quality of life. Better, faster, cheaper drugs... truly a winwinin.
To reduce the cost of drugs, identify fast failures, and accelerate the rollout of bestinclass biotherapeutics, analyses in the middle of the drug development pipeline, from LI to early clinical trials, based on mechanisms and biophysics, is needed. There are currently a number of open source tools that provide systems pharmacology models for use in research. However, a lack of a standardized and automated GLP validation scheme prevents the models from continuing through to development. A robust, systems pharmacology software platform, widely utilized by drug companies and upon which fit for purpose systems pharmacology models can be built, will greatly reduce the time required to get a drug to market, enable the development of best in class drugs, increase safety for patients and significantly reduce the development costs associated with biologics. We have developed proprietary mechanistic algorithms that we currently use to analyze assays and drug program data for our customers. These algorithms are naturally grouped into certain functions/toolsets. These tools are applicable, depending on what questions are being investigated, across most/all therapeutics areas. While our toolsets (InVitro Analyzer, Biologic Feasibility, Biologic Optimizer, Biologic PKPD, Biologic Mechanistic Covariants) are currently in alpha stage (being used to create fit for purpose models for our clients), they are being used in industry to fulfill our servces contracts. We are looking to create an industry standard platform, leveraging our proprietary algorithms and toolsets, that will allow drug scientists to create fit for purpose models that can be easily validated for GLP use in IND applications. We have sampled the market and estimated a potential target audience of over 100,000 users, worldwide. We arrived at this number by? reviewing the Top 100 Pharma companies by revenue, determining employee count from those companies' annual reports, using our Founders knowledge of select large and small pharma and biotechs and its employee count and number of employees that could use our software, segmenting the list into groups based on revenue, extrapolating on a percent basis each company's applicable target audience, assumed companies outside the Top 100 had one quarter (25%) of the applicable audience as the 100th largest company and then summed these estimates. This produced a number of over 100,000 potential customers. For validation, we determined that the number of people, globally, employed in the pharmaceutical industry is 4,500,000 (2012, www.statista.com). So our target market is less than 3% of total pharma employees and our business model shows market penetration of just over 5% of our target market by 2020, or less than 0.15% of total pharmaceutical employees.