Site Navigation Menu

7. PSA behaviour after radiotherapy - PSA bounce.
(Updated on 19th May 2006)

What goes up, must come down - hopefully!

In the absence of hormone therapy, PSA levels decline slowly after radiotherapy and can take several years to reach a nadir or minimum. In fact, the longer the time to reach the PSA nadir and the lower the value of the nadir, the better the chances are of remaining disease free. During this period, about a third of men experience an upward 'bounce' in the PSA levels which may last from a few months to about a year. With adjuvant hormone therapy, the PSA behaviour is more complex but upward excursions in the PSA level should not necessarily be taken as a sign of treatment failure.

PSA cartoonThe measurement of PSA levels after treatment is an important indicator of the outcome of treatment. In the case of surgical removal of the prostate (prostatectomy) in which all the prostate tissue is removed and where no prostate cells have metastasised to remote sites, the PSA should fall to an undetectable level in two or three months after the operation.

For radiotherapy, the behaviour is more complex even without any hormone use. The radiation does not of itself kill cells. Radiation works by damaging the DNA molecule in a cell. Generally, this is repairable but sometimes it produces fatal damage so that when the cell comes to divide into two cells (mitosis), these offspring cells have fatally flawed genetic information and so they are not viable cells. They either die at mitosis or peter out after a few generations. The time scale for prostate cells to reproduce is significant and may range from a few months up to periods of the order of a year or so - as evidenced by PSA doubling times. Now, the cells with fatally damaged DNA still continue to produce PSA and this only ceases when they die. As a result, PSA declines quite slowly after radiotherapy. Because healthy prostate cells resist damage from radiation better than cancerous cells, a patient at a lower level of risk will experience a slower rate of decay than someone with a higher level of risk. The link between the rate of decay and the risk of disease recurrence has been studied in an interesting paper by Ray et al (2006).They found that the time to reach the PSA minimum or nadir  (TnPSA) was an important predictor of disease free survival probability.  From a large cohort of patients with a wide range of risk categories going from low to high risk and all treated with ebrt without hormones and overall radiation doses of between 60Gy and 70Gy, they found the eight year disease free survival probability (failure being based on three consecutive PSA rises) to be strongly linked to the time to the PSA nadir - TnPSA. The table below summarises their results.

Time to minimum PSA after ebrt  (TnPSA)
Time to PSA nadir (months)
Greater than 24months
Between 12 and 24 months
Between 6 and 12 months
Less than 6 months
8 year disease free survival probability

Thus, a slowly declining PSA is generally a good sign. This is perhaps counter to intuition and it certainly puts a premium on the exercise of patience.

Ray et al (2006) also looked at the relationship between the PSA nadir (nPSA) and eight year disease free survival. The table below again summarises their results.

PSA nadir after ebrt  (nPSA)
PSA nadir range (ng/ml)
Less than 0.5 ng/ml
Between 0.5 and 1.0 ng/ml
Between 1.0 and 2.0 ng/ml
greater than 2.0 ng/ml
8 year disease free survival probability

Clearly the lower the final nadir, the better the long term prospects are of remaining disease free. It should be noted that these two tables are only a summary of their results and their paper needs to be consulted for more complete details.

It is also important to stress that even in the case of successful treatment, the decline of PSA is not always smooth but it may rise again for a period of a few months before declining towards its final minimum value - the nadir.This 'bounce' in PSA has been studied mainly in connection with brachytherapy but some data is available for external beam radiotherapy too - Zietman et al (2005). The sketch below shows a rather idealised version of PSA bounce after external beam radiotherapy without any hormone use.

PSA bounce

The reason for the bounce is not known for sure but somewhere between a third and a half of patients experience bounce and it may occur anywhere between about 1 to 3 years after treatment. The magnitude of the bounce lies in the range from about 0.5 ng/ml to 2 ng/ml and may last from a few months to around a year.

It should be stressed that PSA variation after treatment varies significantly from patient to patient and the above figures are only a guide. However, one should not jump too rapidly to the conclusion that treatment has failed because of upward variations in PSA values. Monitoring of PSA at three monthly intervals over a period of three years is probably necessary before any definite conclusions can be drawn about the effectiveness of the treatment - unless, of course, the PSA values rise in a sustained and significant way. It should also be noted that an analysis of PSA records of 4,839 patients from nine institutions in the US showed that there was no difference in long term survival or cure between patients with bounce and those without - Horwitz et al (2004).

With hormone treatment, the PSA behaviour after the radiation treatment is even more uncertain. With neo-adjuvant, concurrent and adjuvant hormone use, the PSA may fall to low or even undetectable levels a few months after radiation. If, then, a patient stops taking the hormones at some point, the PSA level will recover but it may take some months to do so and it may not do so smoothly. Even second 'bounces' occur in a number of cases. However, if the treatment has been effective, the PSA should finally level off and, as has already been discussed, the lower the final value, the better the prognosis.

To reiterate, the PSA nadir does not fall to 'zero' with external beam radiotherapy because the prostate is not entirely destroyed by the radiation. However, if the treatment has been successful, all the cancerous tissue will have been destroyed but 'healthy' prostate cells are more resistant to radiation damage and some of these will remain. The prostate gland will be reduced in volume but will still produce residual PSA. Clearly, the more of the prostate that remains, the greater the chance of a recurrence of the disease either through cancer re-developing in the remaining tissue or through the fact that not all the original cancerous tissue was destroyed. As the Ray et al (2006) results show, the lower the final PSA nadir and the longer it takes to reach this value, the better the chances are of remaining disease free.

Horwitz, E.M. et al (2004)
The biochemical and clinical significance of the post-treatment PSA bounce for prostate cancer patients treated with external beam radiation therapy alone: a multi-institutional pooled analysis.
Abstract 171, ASTRO 46th Annual Meeting.

Ray, M.E., Thames, H.D., Levy, L.B. et al (2006)
PSA nadir predicts biochemical and distant failures after external beam radiotherapy for prostate cancer: a multi-institutional analysis.
Int. J. Radiation Oncology Biol. Phys. Vol.62, No.3, pp.714-718.

Zietman, A.L., Christodouleas, J.P., and Shipley, W. (2005)
PSA bounces after neo-adjuvant androgen deprivation and external beam radiation : Impact on definitions of failure.
Int. J. Radiation Oncology Biol. Phys. Vol.64, No.4, pp.1140-1150.