Daily pill can double survival time for deadliest cancer, trial shows

I have quoted exactly the numbers that were written in the abstract of the article yesterday, at the same link that you have used.

What is very weird is that the abstract of the article has been changed, and now it is as you say.

So the numbers from today are worse than those from yesterday.

Perhaps the abstract of yesterday corresponded with a preliminary version of the study report, but meanwhile more patients have died, which has been taken into account in the final version, lowering both the median values for overall survival and for progression-free survival.

> Perhaps the abstract of yesterday corresponded with a preliminary version of the study report, but meanwhile more patients have died, which has been taken into account in the final version, lowering both the median values for overall survival and for progression-free survival.

No. The 'data cut' that these numbers are based on will have been taken (at least) weeks ago. You don't get (anything close to) real-time updates in the manner that you're implying. It was probably just a simple snafu somewhere along the line.

I'd argue that the numbers suggest that venture spend on info tech / AI / social media is at least of the same order of magnitude as new drug discovery, if not more.

It would be interesting to compare blockbuster drugs of the past 5--10 years or so to see what types of investment have been made there. mRNA / Covid19 vaccines, and GLP1s.

A Deloitte report gives oncology, infectious disease, and chronic conditions (particularly obesity and diabetes) as key areas.

It also gives some relevant numbers re investment and revenue:

The increase in IRR is driven by promising late-stage pipeline candidates and impressive trial outcomes (the average forecast peak sale has increased to US$510 million). However, high R&D costs, which reached an average of US$2.23 billion per asset in 2024, present a continuing challenge.

https://www.deloitte.com/ch/en/Industries/life-sciences-heal...>

(And if you're confused by exactly what that 'graph is actually communicating, I am too: "$2.23 billion per asset" is an odd and unclear phrasing.

From StartUs Insights, "Drug Discovery Market Report 2025":

The global drug discovery industry was valued at approximately USD 106.70 billion in 2025.

(Emphasis added.)

And...

The drug discovery sector includes 5370 companies with 990+ startups. The leading country hubs are the USA, the UK, India, China, and Japan, and major city hubs include San Diego, Cambridge, London, New York City, and San Francisco. ...

Over 9550 funding rounds closed with an average of USD 34 million per round. More than 7600 investors invested in over 2220 companies.

Doing the maths on that last: about $325 billion in funding.

https://www.startus-insights.com/innovators-guide/drug-disco...>

Also relevant is projected growth of industries.

AI, which leads infotech venture spend presently, is valued at $515 billion in 2026, and is projected to grow at 30.6% CAGR, reaching $1 trillion in 20209 and $3.5 trillion by 2033.

https://resourcera.com/data/artificial-intelligence/ai-marke...>

(Whether or not you, or I, believe these projections is less material than what the markets and investors believe, and they seem to be acting on the assumption that the market and asset values will grow.)

Pharma by contrast, is estimated at $1.7 billion 2025, growing at 6.08%, and reaching $2.8 trillion by 2033.

https://www.grandviewresearch.com/industry-analysis/pharmace...>

Which is to say that whilst projected AI and pharma markets in five years are comparable, growth in AI is seen as 5x the rate of pharma. And investment follows growth, not total value.

Pharma spend is largely capped by public and private insurance spend, which tends to be mature, already high relative to total economic activity, is constrained on multiple fronts, and divides results and rewards amongst multiple entities (both drugs/therapies and companies). Biggest growth is projected in Asia-Pacific.

Projections for AI spend are ... somewhat unhindered by empirical data or practical realities, and at least as told at investor storytime, promise both the heavens and stars and winner-take-all dynamics, such that money is chasing the possibility of landing on the top of the heap.

Again: quite probable that there's a lot more money available for AI moonshots than new drug discovery, though perhaps grounded more in anticipation and psychology than reality or net social benefit.

That’s a valid way of looking at things, but diminishing returns tells a different story.

People have been looking at treating the same disease with 100’s of billions / year across decades. What completely new ideas haven’t been explored with the 10’s of trillions of dollars we already spent? Vs deep learning or perhaps more narrowly LLM style AI where the total investment over the last 50 years is probably ~1% of what’s been spent on drugs.

So sure in 2026 things look heavily weighted to “junk” tech, but overall it’s probably at least a 2 orders of magnitude difference from a slightly different but still reasonable take. Similarly looking forward people are only going to toss 100’s of billions at it in 20 years if it’s actually had a massive impact, without that it’s a short term blip.