I thought it just meant rising PSA after treatment.  Depending on
the treatment type and the practitioner, the amount and duration of
the rise that constitutes "failure" are sometimes defined
differently (e.g., it is usually defined as a higher number for
radiation than surgery, and different rad oncs disagree on the
numbers and frequency of rises that constitute failure.)

It's "biochemical" in that it's just the detection of this chemical
(PSA) in the bloodstream.  It doesn't strictly prove that a person
has cancer.  No biopsy has been done that found any cancer.  But it's
what all the clinical trials uses as an endpoint because:

1. It's easy to test.

2. It's cheap.

3. It happens long before any other symptoms appear, giving advance
warning that something isn't right.

   Alan

Alan is right.  "Biochemical Failure" does not refer to failure of the
body, such as cells proliferating, as one would think..  I once
complained about the use of medical terminology that is either
damaging or incomprehensible to cancer patients in general, and used
this term as an example.  One of the replies I got was from, from MR,
a scientist.  He told me privately that "biochemical failure" is one
term that "should not have escaped the labarotory.".

believe I can shed a little light on that since I have a background in
biomedical sciences. I was a microbiologist and have had considerable
interaction
with the clinical/medical community. I have been a medical school
instructor and have supervised a hospital laboratory. I have also been
involved in several
basic research projects, including one on leukemia. Of course, my own
experience as a 3-year PCa survivor has added to that.

=snip=

"One aspect of the terminology problem is that certain words are used
differently within the medical community, than they are by the non-
medical public.  It's when the terms get out into public use and
acquire emotional baggage, they become troublesome.

"The term "biochemical failure" is another example. Medically, it is
used to distinguish it from other means of detecting recurrence, such
as by biopsy or
scans. Since detecting PSA in the blood is a biochemical process, the
term has a very specific, limited clinical meaning. Again it is used
to make a very
clear distinction.

"The problem arises when the word failure shows up in public use, with
its negative implications, etc.

Leah.

Good question.

You are quite correct that this term is loosely defined.

Statistical studies require an objective rule for classifying surgical
patients but many do not state what they mean by "biochemical failure".

I've seen a couple of studies that classify patients in the "biochemical
failure" category if post surgical PSA is greater than or equal to 0.07.
This is the lowest value I found that was used.   "Recurrence" is commonly
defined as PSA greater than or equal to 0.30 measured more than once at
least 2 weeks apart measured in the same lab.  Many studies use this
threshold to define "biochemical failure".  I found another study that used
0.40 as a threshold, another that used 0.02 and others that use 0.50 as a
breakpoint.

The definition may include the words "and rising".  A PSA of 0.22 or some
such number that is stable could conceivably be the expression of benign
prostate tissue not excised during surgery but post-operative PSA increasing
from say 0.02 to 0.15 to 0.22 might be classified as biochemical failure by
some.

Bottom line - so far as I can see, there is no single universally accepted
threshold but it is in the range of 0.07 to 0.5 and most commonly 0.3
nanograms/litre.