But, “personalized medicine” (PM – to my mind – the tight linkage of diagnostic results to therapeutic strategies) is an inevitable outcome of the current state-of-the-art knowledge of genomics, proteomics, combined with recent advances in computational, database, & networking capabilities; & that the practice of personalized medicine, at least at the institutional level, will soon be a requirement – not an option – for many types of treatment. The corollary of this is that institutions who implement these capabilities soon & well, will thrive – those that do not, will be seen as second-tier care centers. There are already numerous examples of PM that are being practiced today such as those defined by companies like Genomic Health who provide unique diagnostic assays for ER+ breast cancer or stage II+ colon cancer. Most large pharmas & even giants such as Medco are moving toward making PM pay.

What does, “tight linkage of diagnostic results to therapeutic strategies” really entail? I think that there are two key elements – technology & process. The technology piece primarily involves linking together multiple, disparate databases, providing the computational horsepower to rationally data mine the combined resources, & to provide the properly authenticated & authorized physicians, nurses, & researchers with timely access to the results.

None of the technology is particularly novel, but hooking up all of the pieces correctly is – & I can imagine that some parts of that process will be a bit painful.

I believe that an example of a good (but not perfect) model for the tech side of personalized medicine can be found at the Duke IGSP. Please see this URL if you’re interested:  http://www.genome.duke.edu/cores/

There are several ways to classify different cloud architectures, but the three most prevalent are:

1. Infrastructure as service         [example: Amazon EC2]
2. Platform as service         [example: Google API's]
3. Software as service        [examples:  Salesforce,  Google Apps]

The issues surrounding the cloud (besides the obvious one of vendors rapidly pushing incomplete software to market) are:

1. Reliability
2. Redundancy
3. Latency
4. Performance
5. Security

I’ll talk about each of these in turn in the next few posts. BTW… This website is hosted in the cloud – Amazon’s Elastic Compute Cloud (EC2) – to be more precise.