HealingWell
Search Close Search
☰

Recurrence rate and cure rate correlation

Support Forums
>
Prostate Cancer
✚ New Topic ✚ Reply
❬ ❬ Previous Thread |Next Thread ❭ ❭
Posted 5/20/2017 11:38 PM (GMT 0)
When Dr Katz cites that the actuarial 9 year freedom from biochemical failure was 89% for intermediate risk patients who were treated with SBRT, would it be accurate then to automatically say that 11% will suffer a biochemical failure within 9 years or are other factors involved that may make the flip side inaccurate? I guess another way of putting that is if the cure rate is 89%, does that always mean the recurrence rate would then be 11%?
Posted 5/20/2017 11:45 PM (GMT 0)
At least it would mean that the recurrence rate WAS 11% in his sample. If one samples enough patients, over time, maybe that will change, but it did equal that in the past.

Just splitting hairs about history as a predictor of the future. I'd say that the probability is high that 11% will suffer recurrence over 9 years, but those are just averages. Each patient has a 50/50 chance of suffering recurrence (either they will or they won't) but as a group, 11% will experience recurrence.

I could be all wet on the stats. I'm a financial guy, not a statistician.
Posted 5/21/2017 1:24 AM (GMT 0)
Yes, if 89% were free from recurrence, then 11% were not free from recurrence. In one of his papers he breaks down the GS 3+4 separately from the GS 4+3. Of course the GS 3+4 did much better. The last big 5-year study reported a bRFS of 97%.
Posted 5/21/2017 2:15 AM (GMT 0)

Paxton said...
... Each patient has a 50/50 chance of suffering recurrence (either they will or they won't) but as a group, 11% will experience recurrence.

I could be all wet on the stats. I'm a financial guy, not a statistician.

I heard that in Bones's voice, "darn it Jim! I'm a finance guy, not a statistician!"

[Edit: heh. I didn't say "darn". Forum software changed it.]

Minor quibble: just because there are only two outcomes actually doesn't imply a 50/50 chance of either one happening. The odds ratio can be anything from 0 to 100, even with only two outcomes. Just to clarify...
✚ New Topic ✚ Reply