Cancer Types


Site Map

Salvage  radiation results based on PSA level at time of relapse Journal of Clinical Oncology, Vol 19, Issue 4 (February), 2001: 1030-1039

PostOperative Radiation for a Rising PSA (or 'salvage' radiation
)  see video here

 see review here and here and current trial (RTOG 0534 here)

After having a radical prostatectomy the PSA level should fall to 0. Any persistent elevation is worrisome and any consistent rise in the PSA is considered evidence of a relapse. If the rise in the PSA is slow and occurs after a prolonged period, the site of relapse is generally at the site of the original tumor, and radiation to this area may still cure the patient. (see recent study in JAMA here and here and other recent studies, another study, a study of postop 3D conformal radiation.). There is evidence that salvage radiation increases survival (go here). There is evidence that a higher dose is better (Stanford study recommends 70Gy go here).

(For a discussion of immediate radiation after surgery go here.)   For a discussion of the significance of a rising PSA after surgery, go here.

There are ways to calculate the success of radiation in controlling patients who have a PSA relapse after surgery (go here) and as noted below. 
Collected data say the best results are when the PSA is the lowest and for every PSA rise of 0.1 the cure rate falls by 4% (go here) and here.

the cure rates are much lower if the cancer has obviously recurred in the gland (called a clinical local relapse see here.)

In general the higher the level of the PSA, the lower the chance that radiation will cure the patient. The NCCN guidelines recommended postOp XRT for patients with rising PSA's after surgery particularly if: rise occurs 1 year or later from surgery, seminal vesicles not involved, Gleason < 8 and the PSA < 2. Once the PSA is > 2.5 they recommend going with hormonal therapy  (see NCCN guidelines and flow chart


As noted in this study, the higher the PSA the smaller the chance of a cure:

PostOp Radiation for Recurrent Prostate  by PSA
PostOp PSA Level Cured at 5 Years
undetectable 71%
up to 2.7 48%
over 2.7 0%
Crane, Int J of Rad,Bio,Phys 1997;39:681

Improved Results with Radiation based on Level of PSA Prior to Therapy (5 series)
PSA Prior to Radiation Cured with Radiation (bNED)
<0.2 vs  0.3 - 2.0 vs > 2.0 77% vs 59% vs 10%
< 2.0 vs > 2.0 74% vs 22%
< 2.5 vs > 2.5 52% vs 8%
< 1.1 vs > 1.1 78% vs 18%
< 2.0 vs > 2.0 60% vs 21%

Valicenti. Int J of Rad, Bio, Phys 1998;42:501

Literature Review of All Studies using Radiation for Patients with Rising PSA's after Radical Prostatectomy

anscher_study.gif (8037 bytes)

(From Anscher, above) In 1997, a consensus panel was convened by the American Society for Therapeutic Radiology and Oncology (ASTRO) to formulate guidelines for the role of salvage radiotherapy for PSA failures after radical prostatectomy. No data from randomized trials were available. Based upon retrospective data, the panel could not identify any subgroup of patients with a higher or lower success rate from salvage RT. It was felt, however, that early treatment (i.e., before the PSA rose above 1.5 ng/mL) was more likely to be successful than delayed therapy. The panel also recommended utilizing doses = 64 Gy.

Our results do not support the findings of the consensus conference in that we could identify seven factors that had a significant relationship to DFS. As in our earlier report, pre-RT PSA =2.5 ng/mL was again a significant predictor of DFS, despite the fact that the composition of the present series differs somewhat from our original report. Since our previous analysis, we have been less likely to offer salvage RT to patients with prognostic factors that predict for the presence of occult distant metastases. This policy is reflected in the more favorable composition of the present series. For example, compared to our earlier analysis, the present group has a smaller percentage of patients with seminal vesicle involvement (34% vs. 49%) or positive nodes (3% vs. 6%). In addition, the median pre-RT PSA of the present series is lower than in our prior publication (1.4 vs. 2.5 ng/mL), reflecting a tendency to refer patients earlier on the part of urologists.

The present series, however, does support the recommendation of the consensus conference to utilize RT doses =64 Gy. In fact, the only significant variable related to DFS that we could identify on multivariate analysis was a RT dose >65 Gy. This finding is consistent with the report of Schild et al. in which a dose =64 Gy was positively correlated with DFS. Similar results have been reported by Forman and Velasco. In contrast, the median dose in our prior report was 61.2 Gy, which may, in part, be responsible for the poorer results in that series.

Late complications in the present series, while common, were generally mild. A surprising finding was that the complication rate among patients treated using 3DRT was higher than in non-3D patients. There are two possible reasons for this phenomenon. First, the median dose in the 3D group was higher than in the non-3D group (66 Gy vs. 61 Gy). Second, it is possible that the use of 3D techniques resulted in the use of larger fields and, therefore, in the inclusion of larger volumes of bladder and rectum in the high-dose volume. Unfortunately, this latter question cannot be definitively addressed herein, due to the impossibility of accurately estimating volumes of irradiated bladder and/or rectum in the non-3D patients. The fact that the lateral field areas were similar between the 3D and non-3D patients, however, suggests that this difference in complications may be more related to the use of higher doses in the 3D group than to the inclusion of a larger volume of normal tissue in the irradiated fields.

Previous reports have also documented a low rate of serious late complications. Grade 3/4 genitourinary (GU) complications have been noted to occur in 0–6% of patients. Similarly, Grade 3/4 gastrointestinal (GI) complications have been described in 0–1% of patients receiving salvage postprostatectomy RT. Grade 1/2 GU and GI complications have been described in 0–21% and 2–17% of patients, respectively. Radiation-induced urinary incontinence has been reported to occur in 4–19% of patients, when administered after radical prostatectomy (most series report <10%). Recently, it has been suggested that radiation given in the adjuvant setting after radical prostatectomy, rather than for salvage of a rising PSA, is associated with a lower incidence of incontinence. The lower dose used to irradiate patients in the adjuvant setting probably accounts for this finding.

In summary, our intermediate-term data suggest that about 70% of patients receiving salvage pelvic RT for a rising PSA will achieve complete remission. Of those achieving remission, about 75% will remain disease-free at 4 years. Whether these remissions can be sustained beyond this time point will require additional follow-up. Doses above 65 Gy should be utilized. Early intervention, i.e., before the PSA gets much above 1 ng/mL is advocated.

ASTRO recently (J Clin Onc 1999;17:1155) published Guidelines for using Radiation for elevated PSA after Radical Prostatectomy and concluded that: radiation will be effective in 70% (response) but only 27-45% will be cured (have neg. PSA at 5 years); it is probably reasonable to wait until PSA rises slightly (level of 0.5) but should start XRT before PSA exceeds 1.5; dose of radiation should be 64Gy, and the role of combining XRT with hormonal therapy is unclear at this time. Specifically THE STATEMENT READ AS FOLLOWS:

PURPOSE: To develop evidence-based guidelines for (1) prostate re-biopsy after radiation and (2) radiation therapy with rising prostate-specific antigen (PSA) levels after radical prostatectomy in the management of patients with localized prostatic cancer.

DESIGN: The American Society of Therapeutic Radiology and Oncology (ASTRO) challenged a multidisciplinary consensus panel to address consensus on specific issues in each of the two topics. Four well-analyzed patient data sets were presented for review and questioning by the panel. The panel sought criteria that would be valid for patients in standard clinical practice as well as for patients enrolled in clinical trials. Subsequent to an executive session that followed these presentations, the panel presented its consensus guidelines.

RESULTS AND CONCLUSIONS: Based on the data presented, the prostate re-biopsy negative rates ranged from 62% to 80% for patients with stage T1-2 tumors. The panel judged that prostate re-biopsy is not necessary as standard follow-up care and that the absence of a rising PSA level after radiation therapy is the most rigorous end point of total tumor eradication. Further, the panel judged that re-biopsy may be an important research tool. Based on the data presented, the long-term (5 years or more) PSA remission rate after salvage radiation therapy ranges from 27% to 45%. The panel requested results from prospective randomized trials to evaluate optimally this information. The panel judged that the total dose of radiation should be 64 Gy or slightly higher and that, in patients with or without radiation therapy, there is no standard role for androgen suppressant therapy for rising PSA values after prostatectomy.


The panel was asked to reach a consensus, if possible, on the success of external-beam radiation as salvage therapy to cure patients with elevated PSA levels after radical prostatectomy, on the complete PSA remission rate and the durability of this remission after salvage radiation, the possible identification of subgroups with higher or lower probability of benefit, and the identification of optimal timing of salvage irradiation.

The rate of a complete biochemical remission after radiation therapy certainly depends on which patients are selected for this treatment. From the presented data, approximately 70% of treated patients had a complete PSA remission. The panel was unwilling to come to consensus now on the durability of a complete PSA remission after irradiation. The data presented from the four invited institutions ranged in median follow-up from 25 months to 67 months, and the actuarial biochemical failure-free survival rates at 5 years ranged from 27% to 45%. This range was judged to be an interim but an uncertain alternative to the results from prospective, randomized trials that are nearing completion or that are under way to evaluate optimally this information.

• If adjuvant or postoperative radiation treatment has been withheld because the pathology findings indicated local cure was likely, then waiting for secure evidence of a PSA failure (rising to a level of 0.5 ng/mL) likely does not decrease the probability of a patient being secondarily cured. In this instance clinicians can be more confident that the patient actually has relapsed disease and warrants treatment if they wait until PSA levels rise to this clearly measurable level. The panel noted that most data indicate the PSA level at time of salvage radiation may make a prognostic difference in outcome. The PSA level or threshold, based on the data presented, seemed to be 1.5 ng/mL.

• There are presently no specific subgroups of patients, based on either the pathologic information from the prostatectomy specimen or the preprostatectomy PSA levels, who indicate a higher or lower success rate with salvage radiation therapy.

• The highest dose of radiation therapy that can be given without morbidity is justifiable. Based on the data presented at this conference, the dose should be 64 Gy or slightly higher with standard fractionation (1.8 or 2.0 Gy per fraction).

• There is no standard role for androgen suppression therapy in patients with or without radiation therapy for rising PSA values after prostatectomy. The panel judged that the use of hormonal therapy in this setting is still investigational. As such, a cautious approach to the wholesale use of hormone therapy could help reduce cost in the managed-care setting.

The current RTOG trial (96-01) for T3 lesions waits until the PSA is between 0.5 and 4.0 and is comparing postOp radiation alone (64.8G) or combined with Casodex for 2 years.

Background Information from RTOG 96-01 Trial
Radical prostatectomy is an effective treatment for prostate cancer when the tumor is found pathologically to be confined to the gland. However, when on pathologic examination tumors are found to extend through the prostatic capsule and/or to have a positive surgical (inked) margin, these patients are likely to develop recurrent disease. The median time to develop clinical recurrence with positive surgical margins is approximately seven years. Serum prostatic specific antigen (PSA) has been evaluated in many centers as an early predictor of clinical recurrence following primary therapy. A rising PSA following radical prostatectomy occurs earlier than clinical disease recurrence. The frequency of PSA progression following radical prostatectomy for margin positive patients at 3-4 years after surgery is 50-70% which is similar to the long term failure suggesting that PSA progression is an important predictor of surgical treatment failure. Local clinical recurrence following radical prostatectomy is a poor prognostic finding with a low long term disease free salvage rate following either radiation or hormonal therapy.Treatment of clinically recognized metastatic prostate cancer by hormonal therapy results in a median survival of less than 3 years. There are data suggesting that early or adjuvant treatment of patients at high risk for PSA progression (those with pathologic T3 disease and/or positive surgical margins) can significantly lower the rate of PSA progression and possibly the clinical course. However, at many institutions, adjuvant radiation therapy is not given to those patients at substantial risk of local failure based on their pathologic staging, following radical prostatectomy. Instead, treatment by either radiation or hormonal therapy is reserved if PSA progression subsequently occurs. The justification is that a blanket policy of adjuvant radiation for all pathologic stage T3 patients would needlessly risk irradiating men who may either have no disease or occult metastatic disease, neither of whom would stand to benefit. In addition, the use of stringent disease recurrence of criteria (by PSA progression) now allows the early detection of relapse . This has allowed for more rapid detection than previously of surgical treatment failure and thus has encouraged this wait and watch policy.

A substantial proportion of patients (at least 80%) with PSA progression and a negative metastatic workup will respond to local external beam radiation therapy.However, while 50% or more of men will respond completely with PSA returning to undetectable levels, the durability of this response is uncertain. A number of reports suggested that only one quarter to one third of these men will remain free of a second biochemical progression at five years after irradiation.When patients with node positive disease are excluded, the actuarial three year freedom from second PSA progression is higher at 48%. Further exclusion of those patients with histologic evidence of seminal vesical invasion, who may be at greater risk for having had distant dissemination, takes the figure still higher to 56%. A recent article from Washington University reported an even higher figure of 68% for a 3 year freedom from a second PSA progression in men given external beam radiation for an isolated PSA failure. However, there is no long term follow-up for these reports attempting "early" external beam irradiation to attempt to cure surgical failures detected by PSA progression. This is because clinical "serial PSA testing" after prostatectomy has only been available for six years. Nevertheless, with even this moderate follow-up, there is significant PSA relapse indicating that "early" radiation therapy in an attempt to salvage these patients is not always curative.

While many controversies still exist in virtually all areas of the treatment of prostate cancer, several conclusions can be made: 1) a significant portion of patients who have radical prostatectomy with pathologic stage T3 disease will fail; 2) PSA elevation (either as persistence following radical surgery or as progression) is an early predictor of treatment failure; and 3) hormonal therapy only at the time of clinically documented metastases does not prevent death from prostate cancer and the survival interval is inversely related to the extent of the disease treated. Above recent data suggests that: 1) radiation therapy is locally effective in controlling locally recurrent disease following radical prostatectomy and 2) there may be advantages to combining early hormonal therapy with radiation therapy in other settings of localized prostate cancer not confined to the organ. In an early adjuvant setting the optimal method of hormonal therapy has not been defined.