aaavl2821

If you don't already plan to do so, you should interview bob Nelson at ARCH venture partners. They are a top biotech VC and one of the few VCs that has been active in antivirals the last few years (confounded Vir Biotech which is working on covid among other things)

Bob has been on top of the crisis from the very beginning and has a lot of interesting insight

am i reading correctly that you packaged synthetic viral RNA into viral particles and then suspended in transport medium? do you plan on studying actual patient samples and comparing to gold standard tests?

scientists have increasingly been doing this during covid by publishing on biorxiv and then discussing on other channels (twitter etc)

for ex, this paper [0] by UCSF analyzing 26 of the 29 viral proteins expressed by the virus was posted on biorxiv and publicized further via twitter [1]. they identified 69 FDA approved drugs that target these proteins, giving the medical and scientific community a massive head start on studying drugs that could potentially be near-term treatments

there are some parts of academic science that are stuck in the last century but scientists themselves are certainly not

for those interested in learning more about how science works and how scientists operate, the covid situation is a great way to see how its done. modern science is amazing

[0] https://www.biorxiv.org/content/10.1101/2020.03.22.002386v1

[1] https://twitter.com/kevansf/status/1241936588164063233

It may be that they just presented an abstract and the manuscript has not yet been submitted for publication or peer reviewed

didnt find a link to the abstract, seems the results are just being published

link to the trial on clinicaltrials.gov is here: https://clinicaltrials.gov/ct2/show/NCT04261517

patients in both tx and control arm fared better than in the french study from earlier this week, suggesting patients were healthier / lower risk at baseline

patients got slightly lower dose in this study (400 mg / day) vs the french study (600 mg / day)

because nothing is known to work yet for covid-19, doctors are doing the best they can with limited data. this means using unproven drugs for which, to the best of our knowledge, the risk-beneift is favorable. HCQ for covid-19 is an exampel of this: we dont know if it works, but it might, and it seems safe enough

the goal of studies like this are to strengthen the evidence base and remove those unknowns, so physicians can make evidence-based decisions. if the study is sufficiently poorly designed, it does not accomplish this goal, and can even lead to misinformation

You can't assume the top line effect size is real without critiquing the study design in detail. A lot of experienced drug developer think the data suggests that HCQ may be marginally effective, but Azt + HCQ seems promising. THere are others who are more skepticl

Check out this analysis: https://twitter.com/AppleHelix/status/1240495937522368512

Many doctors are already actively prescrbign this. I heard that it is basically de facto standard of care in France. Doctors recognize that when eveidence based medicine doesn't provide an adequate treatment, they need to improvise. Off-label prescribing is very common

The point of these studies is to help the evidence catch up to improvised clinical practice. If the studies are poorly designed, that goal is not achieved

I don't see a contradiction. The CDC is trying to maximize public health benefit. They recommend stronger protection for people with more exposure to higher levels of the virus. Coincidentally, the people with higher levels of exposure also happen to be health care workers, who need the best available protection to save lives

If you are not exposed to high levels and are not required to put yourself at risk to care for patients, you are at low risk. You can protect yourself with milder measures like social distancing and face protection that is not as effective but not in such short supply

The CDC does not say that only n95 or better are worth pursuing

https://www.cdc.gov/coronavirus/2019-ncov/hcp/ppe-strategy/f...

That is a misleading statement. The CDC has in fact changed its recommendations to recommend in a crises, even HCPs use home-made masks [0]

Health care providers, scientists and health officials are rapidly changing and adapting to the situation. It probably is a good idea for all people to wear homemade / improvised masks to reduce R0

But I'd still hope that civilians with large stores of N95 masks or other PPE would consider donating them to healthcare workers. The optimal scenario seems to be that non-HCPs stay home as much as possible, wash hands, sanitize and wear improvised masks when leaving home, and donate PPE to healthcare workers

9% of COVID-19 cases in italy are HCPs

[0] https://www.cdc.gov/coronavirus/2019-ncov/hcp/ppe-strategy/f...

It may be early to conclude that azithromycin has a complementary mechanism with HCQ, the study was too small and there are too many confounding factors. It is possible there is some synergy but also possible that there is not

Fwiw, Plaquenil is hydroxychloroquine, a less toxic derivative of chloroquine. It has a different safety profile than chloroquine

The French study of HCQ had several limitations. Including that it was very small and some patients who got HCQ and progressed to go to the ICU were excluded from the study. Thus making the drug look better than it is

HCQ is not likely to be a miracle cure. It may be useful in combo with other drugs

It's easy to make not great data look like great data to untrained eyes. And it's tempting to cut scientific corners to bring forward a drug for covid-19. But if the drug doesn't work, we should be honest about that and look for better drugs

Should at risk people take HCQ as a precaution until something better comes along? Definitely. but we need to be careful in talking about its potential

All of the things you've said about mask functionality, reusability and sterilization are very interesting. As is your experience with in-home caregiving.

Is there a quick, effective and easy sanitizing protocol for N95 masks that you think would work for hospitals that are short on masks?

I'm sure hospitals in Seattle, new Rochelle, Santa Clara and San Mateo counties, and LA county could use them. Really any county. 10 masks would be really helpful

That may be true. However, if you are a normal citizen (not a healthcare worker) and you have N95 masks, you should strongly consider donating them to healthcare providers right now

There are huge shortages of these masks in most US hospitals, even in areas with limited cases (cant speak for other coutnries but assume the same is true). HCPs are forced to either treat patients without wearing masks, guaranteeing high exposure to virus, or not care for patients

If you are sheltering in place (as you should be doing if you are healhty), N95 masks wont help you much. THey will 100% help HCPs

I agree, so many of these studies are not very high quality, and there is so much money behind the scenes trying to push agendas -- between pharma lobbies, hospital lobbies, payer lobbies, and politicians capitalizing on public anger at pharma and high healthcare costs in general -- that most of what we read seems to reflect political agendas more than facts

Many drug prices are too high but 80% of rx in the US are generic, and US is middle of the pack in terms of drug spend as a percent of HC spend compared to OECD countries

Drug prices "feel" higher bc copays are higher for drugs than office / hospital visits. If the goal is to reduce actual costs rather than perceived costs, we should look at hospital and provider costs, not just drug costs. But hospitals are huge employers with lots of political power, not to mention large lobbying budgets

Does the study also look at lower provider and hospital costs? Spending on hospital and providers makes up ~50% of US healthcare spend according to CDC. Rx spend is between 10-20% and admin overhead is 30% or so

I wonder why this paper didn't contemplate effect of Medicare for all on hospital and provider spend?

There are about 40-50 biotech startup IPOs every year. $14B was invested in biotech startups last year and ~75% of that goes to r&d

This article (like most articles about VC) doesn't mention biopharma VCs. Biopharma is the 2nd largest sector of VC after software -- $14B invested in 2019 [0] (disclosure: source is my site) -- and all they do is fund R&D heavy startups like Genentech

[0] https://www.baybridgebio.com/startup_database.html

this describes a t cell receptor that recognized and killed cells from several cancers while sparing healthy cells

theoretically this could may made into a therapy by taking t cells from patients, modifying their DNA to express this receptor, and then readministering the modified t cells to patients

there are two FDA approved drugs that use this basic approach. however, this only works currently in "liquid tumors", not "solid tumors" like breast cancer, lung cancer, prostate, etc

This article gives a nice, simple explanation as to why solid tumors are more difficult (scroll down to the chart): https://www.the-scientist.com/features/the-next-frontier-of-...

if this works, then the first challenge listed in the chart would be mitigated. however the challenges of a suppressive tumor environment and sufficient delivery to tumor cells is still a major unsolved challenge

Drug development is the second biggest sector receiving VC funding after software. Most biopharma startups are founded by PhDs and MDs

If you're interested I'm running a program called Biotech Startup School on how to start biotech companies: https://www.baybridgebio.com/biotech-startup-school.html

that last quote seems strange to me. There are certainly instances where fast followers price lower and got significant share. The hepatitis c market is a good recent example. Gilead had the first drug and then two other companies followed shortly thereafter pricing at ~20-50% discounts and got significant share

The phenomenon that a small percentage of patients comprise the majority of health costs is not unique to the US, nor is it unique to this moment in time. There are many studies from other countries -- in Asia, Europe etc -- and even of the US decades ago -- that all highlight the same phenomenon

I believe first round did a study that showed that companies founded by people who'd previously had a successful exit had better outcomes, but they also raised at higher valuations, and those effects canceled each other. So returns from investing in successful serial founders vs new founders are similar

One example of a type of company where decacorn-or-nothing investing doesn't work is biotech / pharma, which is the second biggest VC sector after pharma

Cash on cash returns from seed investing in the biggest biotech companies are an order of magnitude lower than tech. Series a investments in the biggest biotech startups are about half that of tech. This is despite the fact that the companies grow to comparable sizes on comparable amounts of capital [0]

Value inflection happens later in biotech than software. Software startups can get product market fit on seed capital, but the biggest value inflection in biotech is human proof of concept, which costs tens or hundreds of millions

If you invest in biopharma you should focus on lower loss rates (ie do good technical diligence) and concentrate bets in winning companies

[0] https://www.baybridgebio.com/blog/anatomy_of_a_decacorn.html

Re akuous and decibel, I don't know too much about their programs

Re eye diseases / blindness, there is at least one approved gene therapy for a form of congenital blindness, luxturna: https://www.fda.gov/news-events/press-announcements/fda-appr...

the eye is one of the areas where gene therapy delivery is currently feasible. other areas include the CNS using AAV9. Perhaps the most impactful gene tx yet approved is Avexis' Zolgensma (Avexis was acq by Novartis for $8-9B in 2018). Zolgensma is a potential cure for spinal muscular atrophy, the leading genetic cause of infant death. Zolgensma did $160M in its first full quarter on the market

Other areas where we will likely see approved gene therapies soon include blood diseases like hemophilia, beta thalessemia. In addition to blood, eye and motor neurons, the liver is a popular target for gene therapy as all of these tissues are easier to deliver gene tx to with current technology

Delivering gene tx to other tissues is not really feasible as far as i know at this point

The most effective drugs for lung cancer approved in the last few years are PD-1 and PD-L1 inhibitors. The leading drug is mercks pembrolizumab

Here are some of the most widely cited studies of pembro in lung cancer that illustrate its clinical benefit:

Open label pembro vs chemo in first line non small cell lung cancer: https://www.nejm.org/doi/full/10.1056/NEJMoa1606774

Development of PD-L1 biomarkers in 495 patient phase 1 study: https://www.nejm.org/doi/full/10.1056/nejmoa1501824

Pembro + chemo in lung cancer: https://www.nejm.org/doi/full/10.1056/NEJMoa1801005

Pembro + docetaxel for previously treated lung cancer: https://mdanderson.elsevierpure.com/en/publications/pembroli...

There are biotech startups working on all of those already

There are a couple interesting gene therapy companies workout on hearing (akuous and decibel among others). Lots of ophthalmology gene therapy companies also, as well as more traditional companies. Hair regeneration has been a popular area of research for a while

Most companies work on biomarkers as part of the drug dev process. However building a business around just biomarkers is hard -- you need to develop your own drugs based on those biomarkers. Diagnostics is really tough bc reimbursement and pop health is tough bc incentive alignment is nearly impossible in many contexts

I just meant zero overlap among the top VCs. YC is doing some really cool things in biotech but they are pretty new to it. In terms of dollars invested, exits and returns, the top biotech VCs are funds like the column group, orbimed, 5am, versant, foresite, flagship, third rock, atlas, arch, sofinnova. These funds only do biotech and have been doing it for decades