Here is the latest reading of my M-protein level from Monday (Day 41 after the CAR-T infusion):
It's now down to 0.39, which is a bit of a slowdown from previous readings. However, this probably means nothing. My doctor had already warned me that measurements from Kaiser Permanente—where I'm now getting tested—deviate slightly from measurements taken at City of Hope, allegedly because "the machines are different."
Weird, isn't it? In any case, I'm still on track to be at zero in a little more than a week. Keep your fingers crossed.
Attacking in the zero direction, in the parlance of Ukraine.
I'm pulling for you!
All paired body parts and digits crossed for ya Kevin. Keep up the good work! Congratulations!
That two, independent implementations measuring something biological might arrive at slightly different results? Doesn’t seem particularly weird.
Zero Covid = moronic
Zero M-protein = great!
Based upon not just on test results, but upon the number of posts over the weekend, it appears that Kevin Drum is doing very well indeed!
First, great news about your progress! Fingers crossed. The shape of the curve is pretty convincing, though.
Second, re tests. Here's a quick story for you. Decades ago, I went for a fasting cholesterol test two days in a row. (My doc ordered the first one; I ordered the second one and they let me do it (which they wouldn't now)) Same patient. Same hospital. Same lab. If tests were as precise as you seem to believe they are, the results should have been the same, but they weren't. They were about 10% different.
There's a margin of error to everything. Medical tests are no different. There's nothing weird about two tests giving close, but slightly different results.
Fully crossed, KD!
I’ve often wondered about the precision of these kinds of tests. I recently asked my doctor about this. He said they were imprecise but didn’t have a good answer.
I’d like to see error bars on all of these measurements. In my career I did a lot of quantitative analysis and had to sweat the details to reduce the error bars and to know what the error bars were.
I’m pretty sure the labs know the precision of these tests, they just don’t report it because most people aren’t interested.
I found this article where they sent synthetic samples of M-protein to 73 different Dutch labs. They found that within a lab, values were consistent but different labs varied quite a bit so Kevin's doctor is correct. https://www.degruyter.com/document/doi/10.1515/cclm-2020-1810/html
What are these numbers?
What units?
What they actually measure: total antibody or one specific?
As long as they keep dropping and you get to zero....
That's right. Who said it had to be linear? Lower is better.
🤞🤞🤞
Things are looking up. Reason to be optimistic. Very optimistic.
Re: machines differing. In astronomy in the last century, a common activity was to measure star positions very accurately. One observatory would take many measurements of a star, so that it's position in the sky might be known to be
RA 6 hours, 14 minutes, 12.000 seconds, with an accuracy of a thousandth of a second.
A second observatory would do the same measurement of the same star over many years, and they would get
RA 6 hours, 14 minutes, 14.500 seconds, with a plus or minus accuracy of a thousandth of a second. (These are just made up examples, but do represent what actually happened.)
Astronomers would attribute the huge differences between observatories to "instrumental error", because each time the observatories measured the star's position, they would get the same number to the same fine accuracy, but that number would be largely different between observatories.
🤞🏼🤞🏼🤞🏼