It’s time to quit the “war on cancer”

Todays big story, from Time, “Promising Results for mRNA Cancer Vaccine from Moderna and Merck:”

In a release, the two companies report that people with melanoma who received intismeran, jointly developed by Moderna and Merck, and Merck’s pembrolizumab (Keytruda), lived longer without any recurrence of their cancer, as well as had fewer incidences of their cancer spreading, compared to those receiving Keytruda alone, which is the current standard treatment. Combining the two therapies seems to lead to better outcomes in this later-stage trial. Intismeran is not yet approved, while Keytruda was first approved to treat melanoma in 2014, and in 2019 to treat melanoma in people after tumors were removed. The current study included people in this latter group with stage 2 to stage 4 melanoma who were treated with either intismeran and Keytruda or Keytruda alone after surgeons removed the bulk of their tumors.

Moderna stock surged 170% on this news, only to fall 25% the next day, giving up the bulk of its gains. My tweet yesterday warning to take profits proved prescient, in keeping with my track record (author has no position in Moderna):

As someone who invests, I have seen countless instances of pharma companies being pumped on the promise of cancer treatments, only to fall when sales disappoint or the drug is not as effective as hoped. Harnessing the immune system to treat cancer has been heralded as the “next big thing” for the past half century, and it seldom lives up to the media hype.

In my own online discussions about this, people always get mad when you set realistic expectations about this stuff. They always make it personal “my dad had cancer” implying that skeptics want to see people die, or accuse you of being ignorant. Ok, these are just the facts. From Claude:

Your numbers are basically right, and it’s worth stating plainly: oncology drug development has one of the lowest success rates of any therapeutic area. Multiple independent studies converge on roughly the same figure — the overall phase 1 to approval rate for all oncology-related drug development programs is 3.3%, with only 3.4 percent of all cancer drug development programs from 2000 to 2015 moved from phase 1 to regulatory approval, despite the fact that oncology accounted for 42 percent of all drug development programs during that window. That’s not a fringe stat — it’s the mainstream finding from MIT’s own tracking of the field, and it holds across multiple independent analyses using different databases and time periods. And it’s getting harder, not easier: success rates for oncology clinical trials have declined steadily from 2015 to 2023, even as investment has grown.

Survival rates were not published though. From the aforenoted Time article:

The companies did not provide specific details about the recurrence or survival rates, but earlier this year reported at ASCO, a major cancer conference, from an earlier stage study that showed that the combination lowered the risk of recurring melanoma by 49% and lowered the risk of distant spread of the cancer or death by 59%.

This is important, because “recurrence free” and “survival” are not the same thing. “How long does said treatment actually prolong life,” is an obvious question anyone asks when evaluating the efficacy of a cancer treatment. A cancer can appear to be in remission, only to return with vengeance when the treatment stops working, so “total survival time” does not improve much or at all.

I wrote about low-hanging fruit a few days ago. Despite generally being optimistic about technology, such as AI, I think trying to cure cancer is actually a poor use of resources, as we’re well past the point of diminishing returns, and the marginal returns are not that high relative to taxpayer costs (e.g. Medicare). Cancer patients tend to be disproportionately old and well past the most productive years of their life. Of course, there are plenty of examples of people who are not old dying of cancer, but were talking in the aggregate.

Much of the gains over the past 50 years have been in early detection and lifestyle modification (such as not smoking), not curing advanced cancers, yet we continue to shovel money at trying to cure it. Clearly, preventing cancer is much more cost effective.

Chemotherapy and surgery are effective against some cancers (certain leukemias and solid tumors), but progress has stalled for the rest. The published 5-20-percent 5-year survival for stage 4 cancer such as lung, colon, stomach, or pancreas hasn’t budged much in decades. On the other hand, there has been some progress, consulting Claude again:

On the stagnation claim: Distant-stage (metastatic) survival has actually moved more than “hasn’t budged” suggests, but from a brutally low base. Five-year relative survival for distant-stage disease rose from 17% to 35% between the mid-1990s and 2015-2021, driven heavily by a handful of cancer types — metastatic melanoma went from 16% to 35%, myeloma from 32% to 62%, liver cancer from 7% to 22%. That’s real, and it’s mostly immunotherapy and targeted-therapy driven, not prevention

But Adding 6-12 months of survival for stage 4 pancreatic cancer is heralded as a “huge advance,” but at what cost? Seems disappointing, TBH.

We’ve gotten much better at prolonging survival thanks to more drugs and surgeries, but not at curing it, so patients may die after 3-5 years instead of 1-2 years. How much is this worth? Probably not as much as it costs, especially when this money could be spent on people already healthy in the prime of their lives. For example, I think “free AI tokens” is a better use of money that can be put to actual productive use, than $20k/month of a treatment that may extend survival by a few months.