drug development

In silico drug development won't be able to replace human and animal models for the next 50-100 years

Alex Soojung-Kim Pang — Thu, 25 Sep 2008 09:52:10 -0700

Description

The X2 Project conducted a workshop on the future of science at National University of Singapore on July 25, 2008.

During a discussion of in silico drug development, several people who work on whole system biological simulations and other complex computer simulations were very skeptical that we would build simulations that could replace animal models or human clinical trials in the next 50-100 years. They emphasized a couple things.

First, the incredible complexity of these systems makes this a daunting problem.
Second, the models that do exist of chemical or signal pathways, cells, etc., are created inductively, rather than deductively: they're not implementations of physical models or chemical formulae, but are hand-crafted around real data. This means that integrating separate models into larger models-- creating the pharmaceutical equivalent of a grand unified theory-- is impossible.
Finally, the accuracy rates of these simulations is lower than drug development requires: they're around 70-80%, but clinical trials need 99.5% reliability.

Biomedical Sciences and Biotechnology

Source

National University of Singapore expert workshop, July 25, 2008.

“Innovations in Bioinformatics - Emerging tools for drug discovery and development”

jorgemata — Wed, 07 May 2008 08:06:08 -0700

Description

The key findings of this report [1] are:

• The bioinformatics market is forecast to grow at a CAGR of 23% to $4.5bn by 2011, from a value of $1.6bn in 2006. The US led the geographical market with a share of 45% in 2006, but Europe’s CAGR of 24.6% will narrow the gap over the forecast period.
• Bioinformatics-enabled protein biomarker discovery will enable the development of safer and more effective drugs, targeted therapies and molecular diagnostics.
• Systems biology modelling is forecast to grow at a rate of 35% over the next five years, the highest growth of any bioinformatics sector. The potential for widespread integration throughout all stages of drug discovery will act as a catalyst for this expansion.
• Knowledge management is currently the leading bioinformatics market segment. This lead will be strengthened by mass adoption of Semantic Web technology and the increasing availability of software through the internet.
• Software for next-generation sequencing has vastly reduced time and cost constraints of DNA-sequencing. The miniaturization of reactions has increased their quality and density, subsequently lowering per-reaction costs.

Biomedical Sciences and Biotechnology

Source

[1] Innovations in Bioinformatics - Emerging tools for drug discovery and development. Business Insights, May 2008. http://www.decisionnewsmedia.com/nl/ARC/osm-0DRN507_2008.asp?c=bgxdtzpqrjkpate

Open Source Drug Discovery

Jean-Claude Bradley — Fri, 02 May 2008 13:31:59 -0700

Description

Seema Singh writes in a recent Cell article (1):

Open source software may have been around for 17 years, but using an open source model to speed up drug discovery is a relatively new idea. This month, India is launching a new open source initiative for developing drugs to treat diseases such as tuberculosis, malaria, and HIV.

Bernard Munos reports on a similar trend in a Nature review (2):

The low number of novel therapeutics approved by the US FDA in recent years continues to cause great concern about productivity and declining innovation. Can open-source drug research and development, using principles pioneered by the highly successful open-source software movement, help revive the industry?

Stephen Maurer (3) writes:

Despite their novelty and importance, open source methods have been largely been limited to software. However, scholars have long suggested that it would be logical to organize at least one other field - drug discovery - using open source principles. This paper reviews today's relatively tentative attempts to organize open source biology collaborations and argues that more ambitious projects are feasible. Five specific projects are proposed and analyzed in detail. The article concludes by examining the special legal problems of writing open source licenses in the patent-dominated field of biology.

The theme of Open Source Science used for the development of drugs is one that arises often lately. The value for neglected diseases could be immense.

In these reviews, most examples fall short of being truly open in the sense that one normally thinks of Open Source Software. If data are made public, it is generally after an embargo period. Or data are shared, but only within a group of collaborators, who may have to pay to join the consortium.

The signal here is the strong interest in the concept of Open Science as applied to drug development. I don't think that this will abate and look for bottom-up data sharing initiatives to realize some of the objectives of the movement.

Future of chemistry

Source

1) http://www.cell.com/content/article/abstract?uid=PIIS0092867408004510
2) http://www.nature.com/nrd/journal/v5/n9/abs/nrd2131.html
3) http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1114371

Prizes, not prices, to stimulate antibiotic R&D

Alex Soojung-Kim Pang — Wed, 26 Mar 2008 03:00:00 -0700

NOTE: This content was aggregated from RSS feed. Original source is http://www.scidev.net/en/science-and-innovation-policy/prizes-not-prices-to-stimulate-antibiotic-r-d-.html?utm_source=link&utm_medium=rss&utm_campaign=en_scienceandinnovationpolicy

James Love, director of Knowledge Ecology International, argues on Sci Dev that prizes could be used to encourage work on new antibiotics:

Many of the challenges associated with the development of new antibiotics and vaccines are familiar, and common to other medicines. Research and development (R&D) is expensive, particularly for clinical trials involving people, and product development can be a lengthy process — two unattractive features for most investors, who tend to be risk-averse.

Investors might also be deterred by patent thickets. Many of the scientific benefits of R&D, including those generated by failures, are difficult or impossible to appropriate under patent laws....

For an incentive system that efficiently rewards products that improve healthcare outcomes, and does not lead to rationing and ethical dilemmas over access, it is better to use prizes rather than prices.

In theory, prizes can dominate prices in every important policy area when implemented as part of a scheme that separates the market for innovation from the market for products....

For antibiotics, a reward system of cash prizes could value new products using economic models similar to those used to value stock options, inventories and other financial instruments. A new antibiotic would be valued not only for its use during the patent term, but as part of an ongoing portfolio of products needed for new diseases, conditions or resistance problems that are expected to emerge over time.

Prizes can be paid even in cases where current consumption is zero, or close to zero, as long as the new product enhances the security and sustainability of the treatment programme....

Finally, some smaller firms have expressed interest in the development of a system of prizes that rewards early stages of drug development.

Specifically, they propose a system of prizes to reward success in meeting benchmarks in product development, including the relatively early phase I or II clinical trials.

Study of successful drug targets could hasten development of new medications

Alex Soojung-Kim Pang — Thu, 31 Jan 2008 21:00:00 -0800

NOTE: This content was aggregated from RSS feed. Original source is http://www.eurekalert.org/pub_releases/2008-02/uocm-sos012108.php

EurekAlert reports:

Researchers from the University of Chicago Medical Center and Columbia University analyzed specific properties of the human genes and proteins that serve as targets for nearly a thousand FDA-approved drugs. They identified common characteristics of successful drug targets, especially those of high-revenue drugs. This data could speed up the process and cut down the cost of new drug development....

Guidance from an innovative computational approach could speed up the process and cut down the cost of new drug development.... The researchers analyzed specific properties of the human genes and proteins that serve as targets for nearly a thousand FDA-approved drugs. They identified a number of characteristics that were common among successful drug targets–and especially common among high-revenue drugs.

Promising Applications of Computational Biology

Alex Soojung-Kim Pang — Tue, 23 Oct 2007 11:10:30 -0700

Description

The tools of computational biology may be applied at an increasing rate to pharmaceutical innovation in the next 20 to 50 years, resulting in a faster, less costly, and more tailored approach to drug development.

Computer science and molecular biology have made some of the most significant contributions to science over the past 20 years, and computational biology (also known as bioinformatics) seeks to organize the multitude of activities that are emerging from new collaborations between the two fields. Computational biology makes use of advances in computing power, modelling, visualisation, genomics, protein chemistry, and information science, among others, to find relationships among biomarkers, genetics, pharmaceutical responses, normal responses, and diseases. For example, a computational biologist might search the human genome for particular patterns, analyse gene expression data for biologically relevant molecules, or develop models for visualising the interaction of DNA with other molecules. A loosely shared goal of computational biology is to bring the predictive power of mathematics and computer modelling to modern molecular biology and reign in the enormous amount of information produced by genomic sequencing.

Computational biology is showing preliminary signs of successful of applications. The applied subfields generating some of the greatest interest because of their potential impact on biomedicine are biosimulation and pharmacogenomics, and further research progress will come in these areas in the next 3 to 10 years.

Biosimulation is the computer modelling of biological processes and has the character of what some have called a 'laptop lab'. One hope is to use knowledge of the human genome and pharmaceutical chemistry to design new or more effective drugs that could then be 'tested' in computer models before attempting costly clinical trials, although this potential development is still years away.
Pharmacogenomics is the science of inherited variations in drug responses and promises better biomedicine through a personalized approach. The idea is that a patient's genome could be profiled to predict in advance the effectiveness of a particular drug or treatment. One of the few instances in which this approach has been demonstrated is with the cytochrome P450 (CYP) family of liver enzymes, which are involved in the metabolism of more than 30 different classes of drugs. Genetics tests have been developed to screen for variations and avoid drug overdoses. Another enzyme, thiopurine methyltransferase, has been shown to negatively influence chemotherapy treatments for childhood leukaemia in the rare patient who has a defective variant.

A new industry has developed around applications of computational biology in the last decade. Initial hopes have been tempered, however, and ethical concerns about privacy and property rights to genetic information have arisen. Nonetheless, many new computer applications to aid the drug development process are expected in the next decade. The larger goal of creating a fully predictive biomedicine with tailored treatments is still 20 to 50 years out."

This will be enabled by:

"Training of a new generation of scientists in computer science and biology
Continued investment by governments seeking to remain competitive in scientific research, especially biomedicine
New collaborations between the pharmaceutical industry and the biotechnology industry"

Early indicators include:

"Passage by Iceland in 1998 of the Health Sector Database Act, giving the DeCode exclusive rights to databases of genetic and medical information for the country's 270,00 citizens
Consideration by other countries, including the UK, of 'biobanks' or population databases
Issuance in 1999 by the Biomedical Information Science and Technology Initiative (BITSI) of the US National Institutes of Health of a report stating that the NIH should create between 5 and 20 National Programs of Excellence in Biomedical Computing and should develop a national computer infrastructure
Opening of research facilities related to the State of California's new initiative, the California Institute for Quantitative Biomedical Research, beginning in 2005
Choice by the Public Library of Science, a new open-access publisher of scientific and medical research, of Computational Biology to be its third journal and publication of the first issue in June 2005"

What to watch:

"New publicly and privately funded centres for biomedical computing open.
Life science research in universities reorganizes under the banner computational biology and bioinformatics.
Debates are waged over the merits of computer models verus clinical trials in providing evidence of toxicity or pharmacological efficacy.
A computer of model of a cell, perhaps a liver cell, is developed.

drug development

In silico drug development won't be able to replace human and animal models for the next 50-100 years

Description

Source

“Innovations in Bioinformatics - Emerging tools for drug discovery and development”

Description

Source

Open Source Drug Discovery

Description

Source

Prizes, not prices, to stimulate antibiotic R&D

Study of successful drug targets could hasten development of new medications

Promising Applications of Computational Biology

Description

Signals