Week 1 HW: Final Project Ideas

(1) What you would like to biomanufacture and why?

I want to make a modular production system for rare disease drugs. Rare disease, and their drugs known as orphan drugs are heavily financially disincentivized for big pharma since it’s not profitable. Why is it not profitable? High research price, specialized equipment needed to make it (big pharma are geared to make high demand drugs, and their production infrastructure reflects it), and low demand. Research price will possibly be reduced faster by AI use in biomedical research, but what about production costs? I believe that a modular production system based on living systems and synthetic biology principles could solve this problem. Lowering cost and making orphan drugs more accessible. Who is it for? Fundamentally it is for the patients and their loved ones, while this system could also be marketed to pharma companies to increase their outreach and scalability. I think mainly however this would be an open source platform so that more people could access it. Orphan drugs are very diverse however, and not all orphan drug types are feasible to be produced this way, at least meaningfully. Although theoretically I want this platform to be able to produce all orphan drugs in the world, however, early proof-of-concept prototypes must focus on an orphan drug(s) with these criterias:

1. The drug is part of a drug class that is used to treat diseases (not just rare ones); this is to ensure that the production system could be adapted to produce other drugs when demand is specifically low for a specific rare disease.
2. The drug is able to be produced from commonly obtainable precursors (no rare elements etc).
3. Current available drugs are already generics that could be produced by anyone.
4. Optional: the drug is a biologics

With these criterias in mind, I have decided to build a project about the bioproduction of Enzymes for Enzymes Replacement Therapy (ERT) of lysosomal storage disorders.

Why this disease specifically? These drugs are manufactured on the same CHO (Chinese Hamster Ovary) cell line platform and is a biologics, which means that different drugs only need different GOI being inserted into the cells. Additionally, the addressable price gap is huge, as patients are currently paying close to 200-500K yearly for ERT. Furthermore, a lot of the enzymes being used in ERT already have generic status and thus regulatory permission are less of a challenge compared to brand drugs.

(2) What living system would you use?

As these enzymes are currently produced on a CHO cell line platform, early prototypes could probably benefit in using the same platform. However, cell lines are significantly expensive compared to commonly used bacterial systems such as E. coli or cell-free systems. Therefore, in further iteration of this project, efforts to adapt the production circuit to bacterial and/or cell free systems should be done to increase scalability. This would include further research in trying to adapt the glycosylation pattern of the enzymes to match human glycosylation, adapting codon usage to mirror those used in bacteria/cell free systems, and also screen for potentially immunogenic/allergenic sequences and particles produced by the bacterial and cell free systems.

(3) How would you use automation?

Automation would be integral to both the production process and research process. A production line-style automated stepwise chemical modification of the biologics would be ideal, as the modular nature of the enzymes means that to add other chemical parts, we only need to add a new production line on top of the already existing ones. Furthermore, a module dedicated to the quality control, containing something like a living detection system (antibody-producing systems, etc) would be ideal to detect both enzyme quality, correct enzyme folding, glycosylation, presence of inhibitors/contaminants and so much more. However, there needs to be more research in terms of how specifically this kind of system could be implemented, which process is better automated, and which expression system is the best for this kind of setup.

(4) How much would you need to make, and how long would it take?

How much enzymes is needed to treat 1000 people? Let’s take the example of Gaucher’s disease, one of the diseases that could be treated using ERT. Gaucher’s dieases has a global prevalence of around 0.9 per 100.000 people, which means that globally there is an estimated 72,000 to 125,000 people with the disease worldwide. An ERT regiment for a patient with Gaucher’s disease according to FDA/EMA product label includes imiglucerase (Cerezyme) infusion every 2 weeks, with 15-60 units given per kilogram per infusion, depending on disease severity and response. THis would means that every patients would need to recieve 2275 - 9100 units of imiglucerase per month, or 27300 to 19200 units per year per person. For one person, this would means around 1-2.7 grams of pure protein per year.

1. To treat 1000 people therefore, the system would need to produce 1000 - 2700 grams of protein per year.
2. To treat an area around the size of Massachussets (assuming 7 million population, therefore around 63 patients), the system would need to produce 63 - 170 grams per year
3. To treat all Gaucher’s disease patients, the system would need to produce 72000 to 337500 grams of protein per annum.

How long would it need to achieve those numbers? Current production rate of imiglucerase is around 2.5 kg per year (estimated from sales, with an estimated of 1000-1500 patients on imiglucerase globally), this is not ideal if we’d like to treat all Gaucher’s disease patients. Assuming perfect efficiency, an E. coli production system could theoretically produce 100-150 mg of protein per liter per hour, assuming we’re using the Gingko RAC System which has around 35 L capacity, each reactor would be able to produce 84 - 126 grams per day, which means a maximum of 3780 grams per month and 45360 grams yearly. Which is already more than enough to treat all people with Gaucher’s disease in Massachussets. Two of such reactors with a total maximum output of ~90000 grams per annum would be enough to treat all Gaucher’s disease patients worldwide.

(5) What don’t you know yet? There are a few crucial things that is still unknown for this project to come into fruition.

1. Exact molecular mechanism and modularity of the ERT is still unknown and needs to be studied further.
2. Exactly how automation could help this production process is still not clear, this needs to be clarified before actually implementing the process.
3. Real-world inefficiencies are not taken into account for the calculation process and therefore the numbers could potentially not represent actual reality.