June 2, 2017

Time for stakeholder participation?

Q: Did you see `young blood’ cuts cancer and Alzheimer’s risk?

A: That’s the headline, yes.

Q: This is the Silicon Valley startup that’s transfusing young people’s blood into older people?

A: Well, Monterey rather than Silicon Valley, but yes.

Q: Isn’t it a pity we used up all the vampire jokes on Theranos?

A: I’m sure they aren’t really dead, just sleeping.

Q: Ok, back to business. How much less cancer was there in the treatment group than the control group?

A: There wasn’t any cancer. Also, no control group.

Q: No cancer? That sounds like a success.

A: It was a one-month trial.

Q: So why are they saying “cuts cancer risk”?

A: Some proteins in the blood that are correlated with cancer risk were at lower concentrations after than before treatment.

Q: And this wasn’t just because they got diluted?

A: The story doesn’t actually say, but I’m pretty sure they’ve thought of that possibility.

Q: How good are these proteins at predicting cancer? Are they used for cancer screening?

A: Not that good, no.

Q: And the Alzheimer’s risk?

A: One person with early-stage Alzheimer’s improved after treatment. And people had lower levels of amyloid protein in their blood.

Q: Haven’t we seen that sort of thing before in failed drug trials?

A: Yes.

Q: The story at Quora says “20% fewer amyloid plaques”

A: It does, but there isn’t a source given other than New Scientist, who don’t say that. And amyloid in the blood was one of the planned measurements, but amyloid plaques weren’t

Q: Where did you find these planned measurements?

A: The record at clinicaltrials.gov lists 112 `primary outcomes’ for the trial

Q: That’s quite a lot. So you might expect some of them to look as if they got better just randomly?

A: You might indeed.

Q: That clinicaltrials.gov link says 600 people in the trial, but the story only says 70. What happened to the other 530?

A: They haven’t been recruited yet. The trial is supposed to take two years and they’ve only had one year so far.

Q: Isn’t 70 less than half of 600?

A: Yes, but things often start up slowly, even when you’re not trying to charge US$8000 to participate

Q: They still shouldn’t be reporting conclusions about cancer risk with only about 10% of the study done.

A: Indeed not, but they probably shouldn’t be reporting conclusions about cancer risk without a control group and measures of cancer risk, either.

Q: So, why wasn’t there a control group?

A: Because participants are being charged US$8000, and no-one’s going to pay that sort of money to be in a proper controlled trial.

Q: Is plasma so expensive?

A: The cost to a US hospital was recently estimated as about $400/unit, so about $3200 for the two litres the participants are getting.

Q: Plus all the costs of storing and testing for safety and transfusing and so on, I suppose.

A: No, the $400 included all that. Though measuring 112 primary outcome variables and managing the data will cost a lot.

Q: And a commercial startup needs to make a profit, of course

A: Yes, like Uber and Snap

avatar

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »