Manufacturing Sciences: Campaign monitoring v. Process Improvement

by Oliver Yu

Manufacturing Sciences is the name of the department responsible for stable, predictable performance of the large-scale biologics process.

Manufacturing Sciences also describes the activities of supporting for-market, large-scale, GMP campaigns. The three main functions of Manufacturing Sciences are:

1. Campaign monitoring
2. Long-term process improvement
3. Technology Transfer

Within the department are:

1. Data analysis resources - responsible for campaign monitoring
2. Lab resources - responsible for process improvement

Figure 1: Flow of information within Manufacturing Sciences

The data group is responsible for monitoring the campaign and handing off hypothesis to the lab group.  The lab group is responsible for studying the process under controlled conditions and handing off plant trials back to the data group.

Campaign Monitoring

When a cGMP campaign is running, we want eyeballs watching each batch. There are automated systems in place to prevent simple excursions, but on a macro level, we still want human eyeballs. Eyeballs from the plant floor are the best. Eyeballs from the Manufacturing Sciences department are next best because they come with statistical process control (SPC) tools that help identify common and special cause.

Activities here involve:

• Defining key performance indicators (KPIs) for inoculum and production cultures.
• Control charting these KPIs
• Computing process capability.
• Running univariate and bivariate analysis

Ultimately, all this statistical process control enables data-based, defensible decisions for the plant floor and to production management, much of which will involve the right decisions for decreasing process variability, increasing process capability and reliability.

Long-term Process Improvement

The holy-grail of manufacturing is reliability/predictability. Every time we turn the crank, we know what we're going to get: a product that meets the exact specifications that can be produced with a known quantity of inputs within a known or expected duration.

Long-term process improvement can involve figuring out how to more product with the same inputs. Or figuring out how to reduce cycle time. Or figuring out how to make the process more reliable (which means to reduce waste or variability.

This is where we transition from statistical process control to actual statistics. We graduate from uni- and bivariate analysis into multivariate analysis because biologics processes have multiple variables that impact yield and product quality. To understand where there are opportunities for process improvement, we must understand the system rather than simple relationships between the parts. To get this understanding, we need to have a good handle on:

• Multivariate Analysis
• Multiple Linear Regression

Note: in order to have a shot at process improvement, you need variable data from large-scale. Meanwhile if you succeed at statistical process control, you will have eradicated variability from your system.

This is why a manufacturing sciences lab is the cornerstone of large-scale, commercial process improvement - so that you can pursue process improvement without sacrificing process variability and the results of your statistical process control initiatives.

Outsource Manufacturing Sciences

More Propaganda on Biologics

by Oliver Yu

From Forbes' article on biosimilars:

Compounding the complexity, even biologics that appear chemically the same have unique structural fingerprints as a result of the way protein structure assumes its functional shape, a process known as “folding.” Because biosimilars are so difficult to develop, they sometimes don’t work as effectively as the innovator biologic drug.

I wonder if the author knows how innovator biologic drugs are manufactured.  If he did, he'd know that these drugs are produced in batches and that controls must be in place to ensure that the drug product is the "same" from batch-to-batch.

Biologics are complex due to their size.  Their efficacy depends on not just their chemical composition, but on their shape.  But there are ways to provide a high degree of assurance that the drug product is effective.

In computer science, there's a concept called, "Checksum."  A checksum is a "fingerprint" for digital data... no two pieces of digital data produces the same "fingerprint."  In the world of QC drug testing, there is a concept called peptide mapping.  This is where you digest the protein with enzymes, chopping it at specific places.  Then you take the chopped-protein and run it through a chromatography column and see what order the parts come out.  The order the parts come out is a "fingerprint" for the molecule.

As for the activity of the drug product, you can test that a biosimilar trastuzumab binds to the HER2 protein.  I mean, that's how they tell if you are HER2+ in the first place.

I'm not saying that biosimilars ought to get rubberstamped with lower standards.  I'm saying that the vast majority of biologics manufacturing requirements of the innovator drugmaker can be ported over to biosimilars and that these putative hurdles described by our journalists are dated and not applicable given the modern technology available.

See also:

• SafeBiologics is the innovator biologic drugmaker's lobby. 
• FDA DRAFT Guidance on Biosimilars

OSI PI is Process Omniscience

Employee #2 at OSIsoft is this guy named Don Smith. And the first or second time I met him, I found out that he has an aquarium is hooked up to an OSI PI server and that he uses ProcessBook to trend things like pH, dO2, and temperature for his fish.

I've never seen the setup, but as he was telling me this story, it occurred to me that if his fish were theists, that Don would be their god. Not in the sense that he was their creator, but in the sense that at all times, Don knows every important thing that needs to be known about their "earth".

by Oliver Yu

At the time, I was a fermentation engineer for Genentech, so if Don was the omniscient presence for his fish, then I was the omniscient presence for these mammalian cells that were flying around these bioreactors.

Being the omniscient with respect to CHO cells or fish with the aid of a PI system is one thing. Knowing everything there is to know about a process is another.
There was this one time I got a call from the Instrumentation & Electrical (I&E) department asking me to see if I could tell if a probe got calibrated sometime between 1:00 and 1:15pm. I pulled up PI ProcessBook and was looking for the typical calibration characteristic of zeroing and spanning of the probe signal. I looked at the squiggly flat line and told those guys that there didn't seem to be any activity on the probe at all.

A week later, I find out that they fired an instrument technician on account of falsifying a work order. The trespass? Claiming that he executed the calibration when, according to PI data, nothing was going on at the time.

I&E must have had their suspicions, but when they confirmed with the all-knowing process guru (and had me print out a screenshot), they had enough to let the guy go.

When it comes to operating in a cGMP environment, it really pays to have process omniscience... like an OSIsoft PI system recording every last detail about your process.

Well, this is flattering...

Here is a competitor's activities on my website:





They're searching virtually every OSI PI error code and concept there is on supporting and administrating OSIsoft PI servers.

This is actually more common than you'd think.  There are a lot of automation contractors that are experts at PLC and DCS control systems.  And when there are big automation projects, these guys land the big jobs with the process historian as the afterthought.

by Oliver Yu

But if you're a company that understands that your manufacturing data is the crown jewels of your continuous process improvement initiatives (be it SPC, Six Sigma, Lean, Class A...), you can see from my website analytics that even good automation engineers need a lot of help when it comes to OSIsoft PI support.

My awesome customers understand this and can trust that when they deploy an OSIsoft PI system that it was installed and configured by a battle-tested manufacturing sciences engineer with a proven record supporting commercial GMP campaigns.

OPC Interface, OPC Server

Interoperability in automation is done with OPC. OP-wut?

OPC = Ole for Process Control

A open, proprietary protocol for how automation systems can talk to each other.

In the old days, automation systems were monolithic. Once you standardized on a vendor, you basically had to stick with that vendor.

Since then, the big names in the factory automation decided to form a consortium manage the OPC standard for interoperating with each other...

And since then, OPC has basically taken over the market.

How it works is this:

You have PLCs and DCSs for process control over decades of operations. And you need data from each of these to monitor your process. These PLC and DCS vendors will write OPC Servers. In some cases, 3rd party vendors will write OPC Servers for these devices.

What's an OPC Server?

OPC Server

Software that outputs (serves) data in OPC.

By having an OPC Server, these devices are no longer speaking their native tongue... they speak "OPC-ese." The reason we want them to speak OPC is because OPC clients - such as an OPC interface - can receive this data and translate it.

OPC Interface

Software that can "listen" to OPC and translate it into the native language of another piece of automation software.

For example:

You're trying to connect your OSI PI system to an Emerson DeltaV system. Since an Emerson DeltaV system has an embedded (crippled) OSI PI system inside, you need just a PItoPI interface to connect the two automation systems and send the data from DeltaV to OSI PI.

But suppose your factory has other pieces of equipment that don't write natively to OSI PI... or more likely, you're not interested in buying the an interface for each flavor of automation that you own.

This is where OPC really shines. What you do is, you try to get OPC Servers from the automation vendors. Then you purchase exactly one OPC interface from OSIsoft. And for each OPC server, you configure and instance of the OPC interface.

by Oliver Yu

And now, you eliminate a lot of future headache because you've standardized on OPC and eliminated having to support an interface per automation vendor. You can train on the widely supported OPC standard. You have access to native OPC Servers or in cases where they don't exist, you can purchase third-party OPC Servers that do the heavy lifting.

Commercial Cell Culture, cGMP and 483s

So when I was at Cornell studying to be a biological engineer and then switching over to chemical engineering, I figured that my career in life would be a biochemical engineer. That was the hot new thing and even the School of Chemical Engineering within the College of Engineering at Cornell changed their name to the The School of Chemical and Biomolecular Engineering.

You learned about fermentation, cell culture, bioreactors and weird hosts like plant-cell culture and insect baculovirus. But what they never got to was cGMP... they had classes on it, but it was very much an outsider's perspective.

GMP Regulations in the real-world

When you get to the real world, drug firms manufacturer under the reign of terror known as cGMPs. The way it works is this:

• By law, drug manufacturers whose products make it into the US markets must register with the FDA.
• FDA issues a Form 482 and presents credentials when they inspect your plant
• Should there be inspectional observations, the FDA inspectors issue a Form 483 detailing your manufacturing risks
• If these observations are not addressed and the problems sufficiently serious, the FDA issues a warning letter.
• If the issues remain unaddressed, the FDA gets a consent-decree to seize product or shutdown your plant.

If you run cell culture processes to make food or drug products sold in the US... and you are conscientious about getting your product to market, you have no choice but to maintain a high degree of compliance with the GMP regulations as well as be aware of the FDA's emphasis.

Where to put your resources?

The key is to focus on items as far upstream of the consent decree as possible. So between addressing the warning-letters vs. the non-binding 483 observations, the vast majority of QA Pharma professionals focus on the dreaded 483.

The deal with the 483s is that the FDA issues a lot of them and cannot release unredacted versions of the inspectional observations. Requesting 483s from the FDA can take months.

FDAzilla offers FREE 483s!!

This is where FDAzilla steps in; FDAzilla is a website that provides FDA intelligence and is coming up with a 483 store and service. They've already organized all the free 483s from www.fda.gov into a Dropbox folder.

I checked it out over the weekend on my iPad. You can read Novartis 483s as fast as you read emails:


Here's the Genzyme's infamous 2009 Form 483. The free ones are the most-requested or deemed a public interest by the FDA.


Here's another view of a (Gilead) 483:


There are over 500 documents in this Dropbox folder and access is free (well, you need to leave an email address to whom they grant the access).

Sign up for FREE FDA Form 483s

by Oliver Yu

From the QA professional that needs these resources to the kept-in-the-dark cell culture engineer, if you work in the cGMP field, you really ought to look into access to this FDAzilla resource.

p.s. - Tony, CEO of FDAzilla, tells me there's a surprise in store, so stay tuned.