Hacker News Reader

Comment by sroussey | Hacker News Reader

Personalized per-patient treatments are going to need some manufacturing breakthroughs to make them not be cost prohibitive.

Is anyone working on that?

View on HN

I've mostly examined cell-based vectors, ie engineered viruses that grow on cells, or a bacterial/fungal vector for the neoepitopes. The major cost savings that I have seen is to shrink the size of the per-person manufacturing facility from a clean room that can hold a technician down to a box roughly the size of a microwave that uses robotics. This can shrink a 7 figure $ cost down to 5- to 6- figures.

There are all sorts of custom biological manufacturing companies, they are quite cheap. Like “i spent more at lunch” cheap for custom DNA sequences. When these kinds of things become routine the manufacturing cost gets very low. This particular technology may or may not still be expensive, but i doubt making it cheap will be a particularly interesting or important challenge. EDIT: maybe uninteresting and unimportant are the wrong ways to put it, it's amazing that I can get a completely custom plasmid for $30. I'm just betting the manufacturing is not going to be a major speed bump on the critical path

The big thing with pricing is that this is cancer, life-or-death. The companies know they can charge almost anything if it's highly effective. I wouldn't be surprised if the pricing is 100x or even 1000x the actual manufacturing cost. I'm sure some extra money is needed for investors and R&D, but in a lot of cases the pricing seems obscene for existing products/drugs even taking those into account.

The problem is how do you price fairly when hitting on a therapy that makes it to production is like winning the lottery because it's so rare. Then you get a monopoly with a patent and start thinking about shareholders and it's not like there's any sort of valid competition in breakthrough medicine. The whole industry is screwed up.

Gritstone Bio is working in this space and have in-house manufacturing capabilities. Still a ways from prime time, but fascinating technology and an innovative approach. [0]

[0] https://gritstonebio.com/

Some have been working to achieve single payer healthcare but unfortunately many voters have been sold stories to vote against their own interests. We have to realize we are at the point of the medical technology timeline where all the cheap lowhanging fruit have been picked, our next breakthroughs will therefore always be expensive and time consuming—short of finding something no one has seen that was already there (unlikely, considering everyone reads about the same sort of source material in a field). The best way to get these expensive treatments to more people who need them is to establish a public healthcare model that would use the national gdp to shoulder these medical costs for these rare diseases, that probably amount to a relative rounding error in the grand scheme of things.

> vote against their own interests

This is factually false for anyone who has access to a lot of money. Such people have access to much better care and would likely lose it in a single payer system. See: UK, canada

> would likely lose it in a single payer system. See: UK, canada

no they wouldn't, they'd still see private doctors in private hospitals, which is what rich people currently do in the uk as well.

they just might get taxed more for other people's health care.

So they'll have access to less money, so they won't have access to as good a care.

Only if they think only their own direct personal experiences impact their bottom line, and fail to see how rising tides lift all boats (including the value of their equities)

Luckily we do not need to believe you, we have data from UK and Canada, where people routinely die waiting for care, and get told that care would not improve their QoL enough to have it done. No thanks.

Why do you think people come to USA for the best care when money is no object? Because the profit motive drives the existence of said care.

Why do you think Americans get routine work done in Tijuana? To the point you can find billboards for these things all over san diego county?

Surely you understand the difference between routine ans specialty care?

Routine can be done anywhere, and people go "anywhere". Fine. Specialty is mainly best in USA [1][2][3][...many more...]. So let's not ruin it, maybe?

[1] https://www.beckershospitalreview.com/oncology/worlds-top-on...

[2] https://health.usnews.com/best-hospitals/pediatric-rankings/...

[3] https://www.magazine.medicaltourism.com/article/best-countri...

There are rumours about organ harvesting from China, driven by those "when money is no object".

Right, and they are called the 1% for a reason, if the rich voted for the rich and the poor for the poor the overwhelming win would be for the poor. Many Americans believe they are temporarily embarrassed millionaires.

Many americans correctly believe they have access to the best care in the world[1][2][3]. At the end of the day - better bankrupt than dead.

[1] https://www.beckershospitalreview.com/oncology/worlds-top-on...

[2] https://health.usnews.com/best-hospitals/pediatric-rankings/...

[3] https://www.magazine.medicaltourism.com/article/best-countri...

And so on...

If you're talking about the US, single payer healthcare would be nowhere close to a "rounding error". Medicare and Medicaid are already more than 5% of GDP.

Only 5% is not so bad at all honestly. Apollo program was like 5%. Especially when you consider the population eligible for medicare and medicaid is already paying the highest medical costs (the point of the system in the first place after all).

For cancer you are competing with some amazingly expensive existing treatments like complicated surgery. How does synthesizing a one-off mRNA compare with cutting open someone's brain?

from my understanding you need a sample of the tumor to design the vaccine, so it is not OR but AND

Yup. Cutting out as much as possible is going to continue to be part of the standard of care for the foreseeable future, too.

But I don't think that synthesizing some custom mRNA per-patient is at all cost prohibitive.

Formulating a lot of different batches of mRNA in lipid nanoparticles made with different mRNA might be a little complicated now, but I don't think it's an intrinsically terrible manufacturing problem.

It'll be better if this kind of technique is turns out to be applicable to more cancers, because you need to reach enough doses for economies of scale and manufacturing optimization to really kick in.

There's no way to determine the properties needed by imaging alone?

The immune system reacts to molecules on the surfaces of cells, not macro-level properties.

And even if you can get the immune system to react to the tumor, it's better if you've removed 99% of it, so that there's less opportunity for the tumor to evolve away from the immune reaction.

(Or, in the likely case that you just slow down growth of the tumor a lot: better to start out at a 99% smaller size).

Not a doctor (or even a biologist), but I wouldn't be surprised if it was a heavy lift to expect the immune system to break down a fruit-sized lump, or the lymphatic system to transport it away. So you may want to remove most of the mass physically, and just leave the immune system to mop up the remainder so it doesn't regrow.

Yeah, I would imagine they would do the minimally invasive surgey to remove as much tumor as possible and then do the immune treatment after. Maybe it would make sense to a biopsy and just the immune treatment, but my guess is it would be mkre effective the first way. At least in my personal experience with this kind of thing I would think this could improve survival rates and reduce disabilities caused or worsened by surgery by not having to achive clean margin removal of the tumor. But who knows. Maybe in the future they can extract enough tumor DNA from blood draws to create the target.

For this ourpose. Could a brain tumour be sampled by something less invasive like a needle biopsy?

The treatment is probably more effective if you capture most/all of the tumor you can. For one, maybe that alone is good enough. For two, theres the fact that many tumors are a heterogenous mixture of populations of cancer cells rather than one clonal mass of cells that are genetically identical. There is a chance you might "miss" and fail to capture a particularly malignant population that might be rare at the time of surgery but might have distant metastasis before long.

Probably not. Unfortunately it will become a mandatory step, which the MRI has substituted only in diagnosis.

we need it down to vending machine cost, stick your hand in and boom

Reply on news.ycombinator.com