PSA doubling time with fluctuating data

[FSB12]

I think that I mentioned that I am a working statistician. My husband appears to be in BCR with PSAs going up at what looked like 4 - 6 months doubling time. Then came a fairly good sized blip down and now a smallish blip up. He has a total of 5 numbers spanning roughly one year. I plugged them into the various calculators and came up with around 5-6 months PSADT. But then I decided to fit it to get the 95% confidence bound for the doubling time. This was very interesting - between 4 months and 13 months. So my take-away from that is that simply plugging into the calculators may be unnecessarily depressing with small amounts of fluctuating data. We will not know for sure if these blips mean anything good but for now we will be a little happy.

[LearnAll]

Today I had a talk with my oncologist about frequently and closely monitoring PSA. She said just watching PSA too closely is like "tubular vision" One has to take a composite picture of all different markers and all other parameters to asses true picture.

Besides monitoring PSA, testo, AST/ALT, Alk Phos, BP, BS, Weight,S Calcium, creatinine .GFR and scans, clinical status (symptoms and functioning level) , general sense of well being and so on..have to be watched and one needs to take a composite, BIG picture.

[Hidden in reply to LearnAll]

I agree, it's easy to get too focused on the PSA and miss the bigger picture. My doctor reminds me that we are not treating the PSA, we are treating the cancer.

[Tall_Allen]

The MSK calculator is widely used. They fit to a curve. It is only valid for 3 or more recent PSAs:

mskcc.org/nomograms/prostat...

PSADT is notoriously unstable:

ncbi.nlm.nih.gov/pmc/articl...

[FSB12 in reply to Tall_Allen]

I fit the same curve and got a similar answer. It would be nice if they included some measure of uncertainty with the estimate

[PhilipSZacarias]

Need to obtain more frequent PSA results, perhaps 1 per month initially, along with alkaline phosphatase and lactate dehydrogenase. If lymphocyte and neutrophil counts can be obtained, then the neutrophil/lymphocyte ratio, a marker of inflammation, can be calculated. My results have been less bouncy, trends are usually slow either way...so far. Cheers, Phil

[j-o-h-n]

youtube.com/watch?v=11QrHJG...(Not working as expected?)

Good Luck, Good Health and Good Humor.

j-o-h-n Thursday 07/25/2019 8:06 PM DST