Unification of a common biochemical failure definition

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

Steve Jordan - 19 Mar 2008 22:47 GMT

This is directed toward RT but might be applicable to other txs.

"Three consecutive (PSA) rises (the ASTRO definition of biochemical
failure) yielded a 46% sensitivity and 84% specificity for predicting CF
(clinical failure)."

"CONCLUSIONS: Biochemical failure (BF) definitions applying a PSA
threshold at or after the nadir (e.g., threshold 3) demonstrated the
highest association with CF (clinical failure), OS (overall survival),
and CSS (cause-specific survival) for all assessed treatment modalities.
Definitions incorporating a PSA increase above the nadir value (e.g.,
nadir + 2 ng/mL) were also superior for all modalities. In general, BF
definitions have greater predictive capacity for clinical outcome with
brachytherapy or neoadjuvant AD than EBRT alone, possibly because of
less "noise" from production of background PSA."

Pub Med ID 16765527. Pub Med http://www.pubmed.gov is a service of the
US National Library of Medicine.

It appears that the ASTRO definition, which I have been unable to find
but have seen referenced many times, is fading away.

Regards,

Steve J

Reply to this Message

fred - 20 Mar 2008 02:47 GMT

Steve.....I have two doctorate level degrees, but I have read this 4
times and still have absolutely no clue what this is saying.

Would you mind translating into plain English please?

Fred

Reply to this Message

rosbif - 20 Mar 2008 08:57 GMT

>Steve.....I have two doctorate level degrees, but I have read this 4
>times and still have absolutely no clue what this is saying.
>
>Would you mind translating into plain English please?

Fred, from what I can gather from Steve's reference, the ASTRO
definition of biochemical failure - 3 consecutive PSA rises - results
in 14% false positives (specificity = 86%). There's a clue in this
article:-

http://professional.cancerconsultants.com/conference_astro_2003.aspx?id=57

"...The data from these patients indicated that PSA values could rise
on 3 consecutive tests and then stabilize, calling into question the
ASTRO definition, which declares 3 consecutive rises in PSA as a
marker for biochemical failure."

(It also suggests as many as 20-30% false positives but it's dated
2003 so ASTRO has improved it's status more recently!)

Steve's reference highlights better association using PSA = (nadir+X)
as a simple threshold test instead of the 3 consecutive rises. In the
example X= 2. The explanation apparently lies in the underlying noise
in the PSA readings which can confound the reliability of the
dynamics.

I'd imagine for this reason it's not relevant to those who've had
surgery since we get our nadirs down close to zero and are not
expecting much in the way of PSA noise. Hmm.

Reply to this Message

ron - 20 Mar 2008 03:46 GMT

The ASTRO definition of failure (DOF) is three consecutive increases
in PSA. In clinical trials, ASTRO would also call for the date of
failure to be backdated to the midpoint between the post-irradiation
nadir PSA and the first of the three consecutive rises. Numerous RT
studies have compared the ASTO definition to a variety of other DOFs.
In 2002 ASTRO accepted a committee recommendation to replace the ASTRO
DOF with the Houston DOF (biochemical failure is called when the PSA
is 2 ng/ml above the PSA nadir). The adoption process continues to
move forward...ron

Reply to this Message

JohnHace - 20 Mar 2008 23:40 GMT

> This is directed toward RT but might be applicable to other txs.
>
> "Three consecutive (PSA) rises (the ASTRO definition of biochemical
> failure) yielded a 46% sensitivity and 84% specificity for predicting CF
> (clinical failure)."

It's always curious to me that they never state the interval between
measurements. Many PSA bounces can last over a year. If you were
testing PSA every 3 months or less, it would be easy to get three
rises. I have to assume they are on six month intervals.

John

Reply to this Message

Rate this thread: