Amgen's Biosimilars Gambit

About a week ago, Amgen rocked the biotech industry's proverbial boat with their announcement that they'd be entering the biosimilars market. Multiple news outlets like Yahoo!, Forbes, and CNBC report that Amgen, starting in 2017, will be making six generic versions of blockbuster biologics:

• Abbvie's Humira
• Janssen's^{OLY 02/19/13} Remicade
• Roche's Avastin, Herceptin and Rituxan
• Eli Lilly's Erbitux

This comes as a surprise to many, because for years, Amgen has been saying that biologics really can't be copied.

You see, Amgen is facing patent expiration on it's blockbuster biologics. And unlike pharmaceutical companies, there really wasn't a need to be worried about patents expiring for biologics.

The U.S. Food and Drug Administration has long held that not only must the product meet quality specifications, but also the process that makes the drug must also rigorously meet process specifications.

Well, biologics are made by genetically-engineered microorganisms (cells) that conduct a symphony of biochemical reactions. You give these cells media; you control their environment (temperature, pH, dO2), and the cells do all the work making your protein out of the DNA.

Need to manage your cell culture data?

Since the process specification includes these proprietary cells, it stands to reason that no one can produce the drug product if they don't have your cells. For this reason, biotechnology companies have not been as worried about patent expiration.

Up until this announcement, Amgen has been focusing on defending their superior position by indicating how difficult it is to manufacture biologics and how consumers ought not to trust biosimilars:

• Amgen's Challenges of Manufacturing Biologics (YouTube)
• Amgen's BuildingBiologics.com website saying that comparing biologics is like comparing snowflakes.
• Amgen's membership of SafeBiologics, a lobby against biosimilars

"It's really hard to manufacture biologics."

"When copies of our drugs are made, you can't be certain of their safety/efficacy."

But all of the sudden, our ancient weapons dealer who has been selling us his impenetrable shields has a new offering: his spears that can penetrate anything: 自相矛盾.

Amgen sees the writing on the wall. The FDA is being forced to develop a biosimilars approval pathway as a part of Patient Protection and Affordable Care Act (aka, "ObamaCare"). By law, the only way to do this is to renege on the "product and process" cGMP principle.

by Oliver Yu

One of two things is going to happen:

1. The FDA is going to allow biosimilars into the US markets.
2. The FDA is NOT going to allow biosimilars into the US market.

Someone at Amgen put their brain on and decided no matter what happens, they were going to win.

Consistency issues aside, I think Amgen's gambit is genius.

Redskins Rule - Statistical Bunk

So in the case of the spurious relationship, you have two symptoms that appear when there is underlying disease.

1. Symptom A can indicate symptom B, but 
2. Symptom A does not cause symptom B.

Then there's nonsense like the Redskin Rule which is not even a spurious relationship... it's coincidence.

For those non-(American)-football fans, the Washington Redskins is a football team here in the US.  And since this team moved to Washington DC, an incredible coincidence started happening:

If the Redskins win their last home game before the election, the party that won the previous election wins the next election and that if the Redskins lose, the challenging party's candidate wins.

From 1932 to 2000, this "Rule" correctly predicted the outcome of the election.

Seriously?

This is why people say things like, "Lies, Damn Lies, and Statistics."

You can make the numbers "say" anything you want... especially when there's nothing to say.

If you do want to say something technically meaningful and relevant for solving short- and intermediate-term problems, here are some links to read:

• Correlation is not causation
• Multivariate Analysis in Cell Culture Manufacturing
• Cell Culture Multivariate Analysis

Ice Cream causes Swimming Pool Deaths!

by Oliver Yu

I see this proverbial "Ice cream causes swimming pool drownings" statement made in the world of economics and politics all the time.

It's so prevalent that there's a Wikipedia article on spurious relationships.

[Ice cream] sales are highest when the rate of drownings in city swimming pools is highest.

You can look at the data over and you'll see that this phenomenon happens like clockwork:

• Low ice cream sales... fewer swimming pool deaths.
• High ice cream sales... many swimming pool deaths.

So there's a correlation, right? Yes.

With that correlation, some go farther to allege that ice cream causes drownings or that drownings causes ice cream sales. (Ahem, no.)

To claim that ice cream sales is an indicator of drownings or vice versa also misses the point because ice cream sales and swimming pool deaths are both results of an underlying factor; a heat wave.

Unfortunately, this statement of two symptoms indicating one another is seen all the time in the world of cell culture analysis:

• Final ammonium (NH4+) is an indicator of culture performance
  - or -
• Final lactate (Lac) is an indicator of product titer

credit: The Usual Suspects MGM

Seriously, who here doesn't already know that cell growth impacts culture performance?  Or that cell metabolism impacts culture performance?

Yet we are still publishing papers on how final lactate is an indicator of product titer and concluding that cell metabolism impacts culture performance.

Final ammonium or final lactate are symptoms of cell culture metabolic conditions that produce higher titers.

Unless you can:

• Change media components
• Change a parameter setpoint (pH, temp, dO2)
• Change the timing of culture operations (temp shift, pH shift, timing of feeds...)

Essentially recommend specific changes the Production group can execute to improve culture conditions and you've simply uncovered a spurious relationship; there remains no action you can take to improve culture performance.

This is why it is best to start your multivariate analysis by picking actionable parameters to ensure that you have true factors.

When you pick actionable parameters to model as factors in your multivariate analysis, you have a shot at gaining control of an out-of-control campaign and meeting your Adherence-to-Plan, as Rob Johnson did.

If you're happy pontificating from ivory towers, keep making true-but-useless statements on how every time Y₁ happens that Y₂ also happens.

Otherwise

Thuperbowl1!!!! (Excel vs. PI continuous vs stepped)

I was actually pulling for the Niners yesterday primarily because I've lived in the SF Bay Area since 1999; but on balance, I didn't really care about the game: the Pixburgh Stillers weren't playing.

That and because the NFL actively supported SOPA, the bill that aims at restricting our internet freedom:

That said, here's a look how scatter plot of score vs game duration.


Source data:  Milwaukee-Wisconsin Journal Sentinel

The awesome thing about Excel is that you get to change the Y-axis increments (I made it 7 since that's how many points are in a touchdown)... as well you can configure the X-axis increments (I made it 15 since that's how many minutes are in a quarter).

It's a nice graph and all, but not terribly reflective of what actually happened.  You can see from the start of the game, there's a ramp from 0 to 7 for the Baltimore Ravens.  That's not what actually happened.  At 4:23 into the first quarter, the score was still 0.  At 4:24, the Ravens' score was 7.

Let's look at what this graph looks like in PI ProcessBook with correctly configured tags:

Here, is an accurate depiction of what the scores were at the time they happened:  the score goes up in increments forming what looks like "steps". 

When you have infrequent data, the most accurate visual representation of the data is to set the step point attribute equal to one (step=1).  For cell culture and fermentation, you do this for any offline measurement such as:

• viable cell count
• viability
• offline pH
• Osmo
• Glucose, Lactate, Ammonium, Sodium

If you have it configured any other way, it may look funky... like that first plot in Excel.