Inflated expectations

I have had my inflation paper published in Contract Pharma.
I have long believed vendor contracts in the pharma industry were ill-thought-out and overly generous to vendors, and recently finally got around to spelling out why.
I reckon the timing was nice of the Rupee crisis, since I have a bit of a go at INR weakness in the article.

Royalty Pharma, Elan, and the oil & gas industry analogy

http://mobile.bloomberg.com/photo/2013-04-15/royalty-pharma-raises-offer-for-elan-to-as-much-as-12-per-adr

I can see why Royalty Pharma is interested in Elan's rump (back story here about Elan spinning off their R&D as Prothena), because to my mind this is just a terminal-value play like a reverse bond (see http://en.m.wikipedia.org/wiki/Intangible_asset_finance). I view it (the Elan rump property) as like a single oil field that is of known volume and largely played out, and now they are pumping water in to get the last squirts for a few years (my favourite analogy, more and more, of the oil and gas industry for pharma).

Prothena on the other hand is their small exploration unit that only ever found the one elephant (http://en.m.wikipedia.org/wiki/Natalizumab - ok, maybe there were others). However, $100M as a price for the company, when they have $100M net cash and that's enough to run the exploration unit for a couple/few years, seems a reasonable wager.

The philosophical question Elan raises is why wouldn't any given pharma company split off their exploration unit, and run the existing products into the ground as a terminal-value runoff deal that throws off cash until nobody buys the products any more? I suppose one reason is that this sort of idea is the pinnacle (nadir) of pessimism about the company, surely. Ironically, I think our self-hatred/pessimism complex as an industry is bad enough for this type of thinking: exhibit 'A' for the prosecution is that Elan have already done what they've done.

The reason ExxonMobil and RD Shell aren't doing this themselves is presumably because they still believe in their own exploration ability. And, they know that oil exploration and production is so expensive and so risky that you almost need to have a giant production unit that throws off insane amounts of cash, to fund the expensive exploration and dry holes. The smooth and giant cash flows fund the extremely lumpy dry- and wet-hole business. This is of course precisely the business model of pharma over the last 100+ years. I think one of our main challenges is simply having the balls and the existential confidence to stay the course.

Doctor Copper

I have always liked the sight of fresh copper guttering and downspouts, Swiss-style. I would be interested in knowing what the lifespan is, and cost, relative to baser metals. Clearly the Swiss think the longevity and aesthetics of copper outweigh the cost. Also, the country enjoys low enough crime for theft presumably to be less of an issue.

The photo is at Bruderholz tram stop a short stroll from our house.

Last days at Roche

It will be my last day here soon!

Possible HIV vaccine progress

http://m.sciencemag.org/content/early/2013/03/29/science.1234150

A clever story here where these folks out of Seattle and elsewhere have gone one step better on the lead of a well-known antibody that some people infected with HIV produce. The antibody, VRC01, neutralises most virus variants, but researchers have struggled to coax normal people to produce the antibody, since the necessary B-cell germ line is not easily roused to manufacture this antibody. At least not by simply dosing people with the gp120 viral protein that this antibody targets.

[A little background/reminder is that gp120, an HIV protein, binds CD4, a receptor on human T cells. It is the attendant destruction of such CD4-positive (CD4+ in the lingo) cells that ruins one's immune system: hence [virally] Acquired ImmunoDeficiency Syndrome.]

This VRC01 antibody—which would be key to a vaccine if one could force the body to produce it—binds gp120, but somehow in a different manner than CD4 binds it. It also importantly binds gp120 independently of the various variations in shape of the viral "spike" and its surrounding envelope constituents, thus maintaining its effectiveness in spite of all the devious manifold variations that HIV constantly throws up due to mutating rapidly.

Now then, the above researchers have taken gp120 and whittled it down cleverly to the tiniest fragment they could achieve that still binds VRC01. This little fragment should in theory then be a potent little immunogen since there's not much to it other than the bit that should stimulate VRC01 production.

Seemingly just whittling it down wasn't enough: these folks then assembled the fragment into virus-like nanoparticles of 60 or so fragments, and those pseudo-virii apparently stimulate, in vitro, the right kind of B cell! So, bravo, and their next step is to try this in animals that have been engineered to produce human type antibodies.

Why Nanoparticles?

One might ask: why the virus-like nanoparticle? I can only guess that this gives the collection of viral epitopes some biological plausibility to the B cells in charge of deciding whether to commit to production of antibody. You will frequently see cartoon models of antibody functioning, or cell receptor binding, that depict human cells as giant spheres, and proteins binding to receptors in a binary fashion as keys bind to a lock. But no doubt, this is a crude oversimplification. I can well imagine that one viral antigen fragment might not get a germ-line B-cell "excited", but a nanoparticle resembling a virus particle might interact differently with the surface of the cell recognising it. Perhaps the recognising cell wraps around it in a sort of pocketed embrace, and multiple receptors interact with the particle and realise it's a viroid. A sort of cellular version of the blind-men-and-the-elephant [weak] "joke".

I am just speculating above, but this could be a case of an intelligent breakthrough being based at least in part on a rejection of the cartoonish idea that human cells are spheres, and protein-protein interactions occur in separate, standalone binary fashion like a field full of independent padlocks and keys.

Plant expression of therapeutic proteins

What do you think of this? Butler presentation on plant-based protein expression coming of age.

It was news to me that this is possible (it looks like tobacco spp is the plant in question). And they cover how the plant glycosylation makes this crap for ADCC and the like, but that the glycosylation can be "humanised". Don't ask me how you do that, but I guess it's in vitro, post-purification chemical modification of the Ab? Or is it actually possible to tweak glycosylation by genetic modification of the organism producing the protein? I don't remember how all that works: endoplasmic reticulum blah blah blah.

It's possible that camel antibodies or camel-antibody fragments (http://en.wikipedia.org/wiki/Single-domain_antibody), with their easier folding, simpler structure, greater stability, etc, would actually be even far better candidates for plant expression than other antibodies and proteins. For example, it's possible that whatever relatively harsh extraction techniques are needed to get the Abs out of the plant mush are less likely to denature the camelbodies than other Abs. So it's possible you get both greater yield in the plant, from ease of expression and folding, and also greater yield out the other end of the purification process.

My other question would be is it possible to use another plant species that does a sort of compartmentalisation of tissues. For example, if it's possible to get corn plants to express a given protein at high volume into the corn kernels/fruit, then wouldn't it possibly be easier to harvest the protein, if you only have to purify it out of a fruit-like tissue (with hopefully simple biochemistry e.g. mostly starch) and can throw away at the beginning all the green stuff?