This paper was presented at the symposium on ‘Cancer patients follow-up in terms of medical and economical view’ at the 88th Annual Meeting of the Japanese Urological Association, 9 June 2000, Sapporo, Japan.
Medically and economically appropriate follow-up schedule for prostate cancer patients after radical prostatectomy
Article first published online: 23 MAY 2002
International Journal of Urology
Volume 9, Issue 3, pages 134–140, March 2002
How to Cite
NIWAKAWA, M., TOBISU, K.-I., FUJIMOTO, H., MATSUOKA, N. and KAKIZOE, T. (2002), Medically and economically appropriate follow-up schedule for prostate cancer patients after radical prostatectomy. International Journal of Urology, 9: 134–140. doi: 10.1046/j.1442-2042.2002.00435.x
- Issue published online: 23 MAY 2002
- Article first published online: 23 MAY 2002
- adjuvant hormone therapy;
- medical cost;
- radical prostatectomy
Background : Our goal was to determine the optimal frequency and method of follow-up after radical prostatectomy to minimize medical cost without adversely affecting patients.
Two hundred and twenty-one patients who underwent a radical prostatectomy with or without adjuvant androgen deprivation from 1989 to 1999 were selected for the study. Eighty percent of the patients received postoperative androgen deprivation. Tumor recurrence was strictly defined as detectable serum prostate specific antigen (PSA) and/or clinical findings such as local tumor detection or bone metastasis. Thirty of 221 patients experienced tumor recurrence. Risk of tumor recurrence, procedures for detection of recurrence, and PSA doubling time after biochemical failure were analyzed.
None of the 30 patients who were examined showed definitive local recurrence or metastatic sites on the imaging study at the time of initial PSA detection, and there were no observed recurrences in the absence of detectable serum PSA. In patients who showed elevated PSA within 12 months after radical prostatectomy, PSA levels rapidly increased with doubling times ranging from 1.2 to 13.7 months. Excluding those patients, the doubling time of PSA levels ranged from 2.8 to 31.5 months.
Prostate specific antigen screening is sufficient to detect treatment failure after radical prostatectomy, irrespective of adjuvant hormone therapy. Based on the calculated doubling time, the longest advisable interval between checks of PSA levels is estimated to be four months within the first year after radical prostatectomy, and biannually or annually thereafter. Continuously elevated PSA levels or clinical symptoms indicate surveys for local recurrences and distant metastases.
The number of men in the USA and Japan undergoing radical prostatectomies dramatically increased after the introduction of testing for serum prostate specific antigen (PSA).1 Unfortunately prostate cancer recurs in some patients after radical prostatectomy. Pound et al. reported that local recurrence developed in 2.97% and distant metastasis developed in 6.2% of their 1916 patients after radical prostatectomy was conducted. Few of their patients were treated with adjuvant therapy.2 Partin et al. reported recurrence rates of four percent for local recurrence, eight percent for distant metastases, and 23% for an isolated elevation of PSA only in their stage T1a to T3a prostatic cancer patients after radical prostatectomy.3 However, the appropriate postoperative follow-up schedule and examination methods, and their cost-effectiveness and clinical benefits are controversial issues.4
The frequency and intensity of follow-up for patients with potentially curative surgical treatment varies considerably. The follow-up schedules published by several prominent medical centers for PSA tests and digital rectal examination (DRE) are every 2–3 months for the first year, and every 4–6 months thereafter.5 However, surveys of American Urological Association (AUA) urologists suggest that for healthy patients who have undergone radical prostatectomy for stage T1 to T3aN0M0 prostate cancer follow-up should include an office visit with DRE, PSA, and urinalysis every 3 months in the first year, every 6 months in the second to fifth year, and annually thereafter.4 Bone scan, abdominal and pelvic computerized tomography (CT) and magnetic resonance imaging (MRI), are rarely recommended for healthy patients.2,4,5 These follow-up schedules are mainly for patients following radical prostatectomy without adjuvant hormone therapy. Few reports exist on the follow-up of patients after surgery with adjuvant androgen deprivation.
In the present review, we focus our attention on two questions concerning follow-up procedures after radical prostatectomy. First, is the determination of PSA levels a dependable indicator in patients who have undergone radical prostatectomy with adjuvant hormone therapy? Second, what is the appropriate follow-up procedure and frequency to minimize medical cost without missing signs of consequential recurrence and adversely affecting the patients?
Two hundred and twenty-one patients with clinical T1c, T2 and T3 prostate cancer who underwent a radical prostatectomy between June 1989 and May 1999, and were followed up at our hospital, were consecutively included in this retrospective study. The patients ranged from 48 to 84 (mean 61.3) years of age. Follow-up periods ranged from 0 to 144 (mean 46.3) months. Eighty percent of the patients in this series received neo-adjuvant hormone therapy for 3–12 (mean 5.8) months. Based on pathological results such as a high Gleason score, or high stage or positive lymph nodes, 177 patients also received continuous adjuvant hormone therapy after the operation6 with the intention of prolonging the time before treatment failure.3 Some patients had previously received neo-adjuvant hormone therapy at other institutions when they presented at our hospital. Consequently, initial PSA levels were not available.
After the operation, patients were followed up every 3–4 months by a PSA check (until 6.1997, Delfia PSA,7 lower limit = 0.1 ng/mL; 7.1997 to 3.1999, Lumipulse PSA,8 lower limit = 0.1 ng/mL; and after 3.1999, high sensitive Lumipulse PSA, lower limit = 0.03 ng/mL). Patients without detectable PSA levels were examined for recurrences annually by pelvic and abdominal CT, bone scan and chest X-ray. This occurred because 80.1% of the patients received continuous adjuvant hormone therapy, and PSA levels under hormonal regulation may not accurately reflect clinical findings.9 Tumor recurrence was defined as biochemical failure when detectable PSA (> 0.1 ng/ mL) was found in two consecutive assays, and/or as clinical failure when local recurrence or metastatic disease was detected. In the patients who exhibited biochemical failure, surveys of local recurrence by DRE and trans-rectal ultrasonography, as well as screening of distant metastasis by pelvic and abdominal CT, bone scan and chest X-ray were conducted.
To identify the ideal follow-up strategy after radical prostatectomy, pathological factors such as pT stage, status of surgical margin and Gleason’s score were analyzed in relation to the risk of tumor recurrence in all patients, including patients with adjuvant hormone therapy. The type of examination that first demonstrated tumor recurrence and the findings of other examinations at that time were analyzed. To determine the appropriate follow-up interval, PSA level doubling times after biochemical failure were also analyzed.
The pT stage is classified according to the TNM Classification of Malignant Tumors (5th edn, 1997).10 Surgical margin of the specimen was examined by 5-mm step sections transversely in the middle, and longitudinally in the apex and base of the prostate. Prostate specific antigen level doubling time was calculated by linear regression.11 Welch’s t-test and Wilcoxon’s test were used for statistical analysis.
Biochemical failure and clinical failure
Figure 1 shows the total number of patients without recurrence and the percentage of recurrence at each postoperative year. Tumor recurrence was identified in 30 patients (13.6%). Delayed recurrence (more than 5 years after operation) was detected in three patients (10%). These three patients received adjuvant endocrine therapy. In all 30 patients who experienced recurrence, biochemical failure was detected initially without clinical failure, despite that most of our patients underwent adjuvant hormone therapy. Of 30 patients with biochemical failure, six patients (20%) developed clinical failure (one lung metastasis, two bone metastases and three pathologically defined local recurrences) with PSA levels of 0.6–28.1 ng/mL. The other 24 patients (80%) still display elevated PSA levels as the sole sign of recurrence. For six patients with clinically recurrent disease, 3–36 (mean 15.5) months were required to definitively identify clinical failure following the initial detection of elevated PSA levels. In this series, no patients with undetectable PSA levels developed recurrent disease that was detected as clinical failure.
Risk factors for tumor recurrence
Pathological findings and the status of adjuvant hormone therapy in 221 patients are summarized in relation to tumor recurrence in Table 1. There was no significant correlation between tumor recurrence and pT stage, and the presence or absence of a positive surgical margin or Gleason’s score. Adaptation of adjuvant hormone therapy seems to have some benefit (P = 0.00018). Pathological findings and clinical features of patients with recurrent disease who had either undergone adjuvant hormone therapy or not are summarized in Table 2. There are significant differences in the two groups with regard to Gleason’s sum and the pathological stage of disease. There is no significant correlation between disease-free duration and the way recurrence was detected (biochemical failure or clinical failure).
|Total cases||Cases with adjuvant hormone therapy|
|pT4 or pTx. pN1||49||9||18.4||48||9||18.8|
|With adjuvant hormone therapy||Without adjuvant hormone therapy||P-value|
|Age (average) (years)||64.1||61.5||0.280|
|Gleason’s sum (number of cases)||6 ≧||4||10||0.020|
|Stage (number of cases)||pT2||2||8||0.0012|
|pT4 or N1||9||0|
|Surgical margin (number of cases)||Negative||7||9||0.34|
|Duration of PSA failure free (average, months)||37.9||21.6||0.099|
|Type of recurrence (number of cases)||PSA elevation alone||13||11||0.897|
|Developed clinical failure||3||3|
Prostate specific antigen velocity and doubling time
Figure 2 shows PSA velocity after radical prostatectomy according to Gleason’s’ score (less than seven or more) and the status of adjuvant hormone therapy in patients who experienced recurrence. There was no definitive correlation between PSA velocity and these factors. Table 3 shows PSA doubling times for each patient group classified by Gleason’s sum, with or without adjuvant hormone therapy, and the time to recurrence. In patients who displayed elevated PSA levels within 12 months after radical prostatectomy, PSA levels rapidly increased with a doubling time of 1.2–13.7 (mean 4.7, median 2.0) months. Excluding these cases, the doubling time of serum PSA levels was 2.8–31.5 (mean 10.6, median 9.0) months (two-tailed P = 0.019).
To determine the appropriate postoperative follow-up strategy for patients who undergo radical prostatectomy, we need to consider several issues, including the survival benefit of earlier detection of recurrence, the duration of follow-up, the examination procedures, and the frequency and intensity of follow-up.
The first issue is whether follow-up is even necessary. If earlier detection of treatment failure after radical prostatectomy provides no survival benefit, strict follow-up of patients may be of no use to the urologist or the patient. Whether early detection and early start of adjuvant therapy in patients with small amounts of residual cancer will control local disease or prolong survival is currently unknown.12–14 Earlier detection of elevated PSA levels did, however, corresponded with smaller tumor volume. At this stage, residual cancer may be localized in the pelvic bed, and may respond more favorably to localized therapy with improved long-term survival.12 However, the prophylactic use of early, as opposed to delayed adjuvant radiation therapy after radical prostatectomy showed no difference in treatment outcome.15
The use of early, as opposed to delayed hormone therapy for rising PSA levels after radical prostatectomy is also a controversial issue.13 For some patients with Gleason’s score ≥ 8 and stage pT3b disease, early start of adjuvant therapy after radical prostatectomy provides survival benefits.14 Experiences at Mayo Clinic6,16 showed that early adjuvant androgen ablation therapy had a survival benefit for patients with pTxpN plus disease. In some instances, higher dose radiation therapy for pT3N0 prostate cancer showed durable efficacy when initiated at an early postoperative time.17 Prostate specific antigen tests enabled the early detection of persistent or recurrent prostate cancer, but the survival benefits of early initiation of subsequent therapy are unknown for prostate cancer that is pathologically organ-confined or specimen-confined at the time of radical prostatectomy. Whether or not early detection provides a survival benefit depends on the outcome of large, randomized trials such as EORTC 30943 (immediate versus deferred hormonal treatment for PSA recurrence after definitive local treatment), and EORTIC 22911 (postoperative external radiotherapy versus no immediate further treatment in patients with pT3pN0 prostatic adenocarcinoma).
The second issue for discussion is the length of the follow-up period. Radical prostatectomy offers good clinical results for patients with organ-confined disease. However, the possibility of delayed treatment failure exists, especially in patients who received adjuvant hormone therapy or radiation therapy, and patients with a low-grade tumor. It was reported that 25% of recurred patients had an undetectable serum PSA level at 5 years of follow-up but progression to biochemical and local disease recurrence at a later time.2 In our study, three of 30 patients (10%) experienced recurrence after five years or more. Some pathological and clinical factors (i.e. initial PSA level, PSA velocity, nadir PSA, Gleason’s score, pT stage and seminal vesicle involvement) predict the risk of treatment failure18–19 and distinguish local recurrence from distant metastases.3 However, these factors are not ultimate predictors for a complete lack of tumor recurrence. Our data suggest no significant correlation between tumor recurrence and pathological factors. This may be due to the application of postoperative androgen deprivation or the definition of tumor recurrence. The long-term benefit of postoperative surveillance and the value of the various types of therapy for recurrence is currently unclear.4 Urologists are therefore forced to continue follow-up of patients for a long time.
The third issue relates to types of postoperative examinations that should be performed. Determination of PSA levels is well documented as the most important indicator of tumor recurrence after radical prostatectomy without adjuvant androgen deprivation.2,20 Theoretically, the serum PSA should drop to undetectable levels, even when determined by an ultrasensitive assay for PSA.12 After radical prostatectomy, PSA levels of 0.4 ng/mL or more, strongly suggest the presence of residual disease5 and the probable development of subsequent local recurrence or distant metastasis.
Clinical failure without biochemical failure is rare and sporadic, irrespective of whether patients have received concomitant adjuvant hormone therapy.2,6,21 A survey of AUA urologists4 suggests that bone scan, CT and MRI should not be recommended for healthy patients with undetectable PSA levels, considering the cost and rare incidence of recurrence. The risk of clinical failure in the absence of detectable PSA has been reported in cases with very high grade or anaplastic tumors.5 In our series, no clinical failure with undetectable PSA levels was observed and follow-up by PSA tests was shown to be reliable for the early detection of treatment failure, even for patients under androgen deprivation. The interval to clinical detection of treatment failure from the initial elevation of postoperative PSA required several (3–36) months. Therefore PSA tests are the most powerful indicator of recurrence in patients after radical prostatectomy with adjuvant hormone therapy.
The final issue for discussion is the ideal frequency and intensity of follow-up. To define the ideal timing of PSA checks, the doubling time of PSA levels was analyzed. Prostate specific antigen doubling time was not significantly different between patients with or without adjuvant hormone therapy. The PSA doubling time correlated with earlier versus delayed detection of biochemical failure, but not with pathological results or the status of adjuvant hormone therapy. It has been reported that it is difficult to predict PSA doubling time in an individual with relapse after radical prostatectomy.21 Generally, a PSA level of 0.4 ng/dL or more after radical prostatectomy suggests the presence of residual disease.5 In our series, the lowest PSA level was 0.6 ng/mL when a local recurrence or remote metastasis was detected. The minimum PSA doubling time was 1.2 months when biochemical failure was detected within 1 year after the operation, and 2.8 months when it was detected after the first year.
To avoid overlooking the signs of clinical failure, at least 0.4 ng/mL PSA level should be detected. Based on the calculated PSA doubling time, the longest appropriate interval between PSA checks can be estimated as every 2.4 months within the first postoperative year, and every 5.6 months thereafter if the lower limit of the PSA test kit used is 0.1 ng/mL; and every 4.6 months for the first year, and every 10.4 months thereafter if a highly sensitive PSA test kit (lower limit = 0.03 ng/mL) is used.
In conclusion, PSA is the most powerful indicator of treatment failure for patients who received radical prostatectomy with or without adjuvant hormone therapy. The use of CT, bone scan and MRI should be extremely limited in patients with undetectable PSA and no clinical symptoms. Our data indicates that, from a medical and economic perspective, check-ups including PSA testing to detect signs of clinical failure should occur every 4 months within the first postoperative year and biannually or annually thereafter.
In the present series, 191 of 221 patients who showed negative PSA and no clinical failure nevertheless received CT and bone scans. In Japan, a single PSA assay costs approximate US$45, an abdominal enhanced CT scan costs US$184, and a bone scan costs US$295. Consequently, in the first year US$479 dollars for each patient, and a total of 191 × US$479 dollars were wasted by unnecessarily intensive follow-up examinations.
- 1Prostate carcinoma. In: Johnson FE, Virgo KS, eds. Cancer Patients Follow-Up. Mosby-Year Book, St Louis, 1997; 395 – 400.
- 5Prostate carcinoma. In: Johnson FE, Virgo KS, eds. Cancer Patients Follow-Up. Mosby-Year Book, St Louis, 1997; 378 – 89.
- 8[Evaluation of chemiluminescent enzyme immunoassay ‘Lumipulse PSA’.] Jpn. J. Medical Pharm. Sci. 1996; 36: 1395 – 402 (in Japanese).,
- 10ULCC TNM Classification of Malignant Tumors, 5th Edition, A John Wiley & Sons, 1997; 170 – 3.,