SEARCH

SEARCH BY CITATION

Keywords:

  • PSA;
  • kinetics;
  • prostate cancer;
  • prognosis;
  • diagnosis

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

OBJECTIVE

To evaluate whether the risk of having a positive repeat prostate biopsy is lower in patients with fluctuating prostate-specific antigen (PSA) levels than in patients with a steady or steadily increasing PSA level.

PATIENTS AND METHODS

Files were extracted from the 2000–2003 databases of two teaching hospitals; 191 patients who had a first negative biopsy followed by one or more sets of biopsies and at least two PSA measurements were included. A ‘fluctuating PSA level’ in a patient was defined as a PSA series including at least one PSA value lower than the one immediately preceding it.

RESULTS

The median PSA level at the first biopsy was 7 ng/mL, while that for the second, third and fourth biopsies were 8.0, 8.0 and 8.7 ng/mL, respectively. The median time between the first and second, and the second and third PSA tests was 290 and 317 days, respectively. Prostate cancer was eventually detected in 53 men (27.7%) in whom 39 it was at the first repeat biopsy. Among the 79 patients with a fluctuating PSA level, 17 (22%) had prostate cancer, vs 36 (32%) among the 112 with a ‘steady’ PSA level; the difference was not significant (P = 0.14). When considering the 53 patients diagnosed with prostate cancer, the 17 with a fluctuating PSA level and the 36 others had no significant difference in age, T stage, first PSA level and Gleason score.

CONCLUSION

In the present study, by contrast with the common and unfounded view, the risk of having a positive repeat prostate biopsy was no lower in men with a fluctuating PSA level than in those with a steady or steadily increasing PSA level. The practical and economical implications warrant further studies to confirm these findings.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Since the publication by Carter et al.[1], the concept that PSA velocity, or the rate of increase in PSA level, could help to enhance the specificity of total PSA in detecting prostate cancer, has been widely disseminated. However, several clinicians failed to reproduce these results with 1- and 2-year intervals between PSA determinations [2]. Importantly, in that study [1], PSA determinations were separated by ≥ 7 years. This explains, at least partly, why PSA velocity is not used routinely to enhance PSA specificity for diagnosing prostate cancer.

A frequent situation encountered in daily practice and yet poorly investigated is when a patient presents with fluctuating PSA values (up and down). Although fluctuating PSA levels cannot predict PSA velocity in the long term, these two concepts might well be confused in daily practice. As a result, it is tacitly thought that a fluctuating PSA level is a predictive factor for a negative prostate biopsy.

The aim of the present study was to evaluate whether the risk of having a positive repeat prostate biopsy was greater in patients with a steady or steadily rising PSA level than in those with a fluctuating PSA level.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Patients’ files were extracted from the 2000–2003 databases of two teaching hospitals in France. Patients were selected if they had a first negative prostate biopsy followed by at least one other biopsy, and the PSA level was measured before each biopsy in the same laboratory. All biopsies were taken transrectally under TRUS guidance. At least 12 cores were extracted per biopsy. A ‘fluctuating PSA level’ in a patient was defined as a PSA series with at least one PSA value lower than the one immediately preceding it. In all other cases, PSA was considered as ‘steady’ or steadily increasing. The PSA slope (ng/mL/month) was calculated from at least two PSA values, using the ‘slope function’ of MS Excel.

The results are expressed as the median (range), given the asymmetric data distribution. The Mann–Whitney nonparametric test was used to compare unpaired continuous values, and Fisher’s exact test to compare proportions. Logistic regression analysis was used to assess the association between PSA variation and prostate biopsy outcome, and the prognostic value of PSA variation in patients with prostate cancer; in all tests P < 0.05 was considered to indicate statistical significance.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

In all, 191 patients met the specified inclusion criteria, mainly at least two prostate biopsy sets per patient (58 had three sets and 10 had four), the first one being negative for cancer and each biopsy preceded by at least one PSA estimate. The median age of the men was 63.0 years; their median PSA levels before the first, second, third and fourth biopsy sets were 7.0, 8.0, 8.0 and 8.7 ng/mL, respectively. The median time between the first and the second PSA estimate (PSA2 − PSA1) was 290 days, while PSA3 − PSA2 and PSA4 − PSA3 were, respectively, 317 and 377 days. In all, 139 (72.8%) and 122 (63.9%) men had a PSA level of <10 ng/mL before the first and second biopsy, respectively.

The median prostate volume, as evaluated by TRUS, was 40 g. A median of 21 (12–21) sample cores were extracted per biopsy. Prostate cancer was eventually found in 53 men (27.7%), i.e. 39 at the second prostate biopsy set, 12 at the third and two at the fourth.

Patients with cancer were significantly older than those without (65.9 and 61.7 years) and their PSA density (total PSA/prostate volume) was higher (0.2 and 0.16, respectively). The PSA slope was significantly greater in men with cancer (20 × 10−4 and 1.5 × 10−4 ng/mL/month), but there was no difference between men with cancer or without, respectively in the first total PSA level (7.0 ng/mL in both), free PSA (14% and 15%) and the interval between the first and second biopsy (297 and 287 days).

There was no significant difference between patients with a steady or steadily rising PSA level (112) and those with fluctuating PSA levels (79) in PSA density, free PSA level and the interval between biopsies. The percentage of cancer was lower in the fluctuating PSA group but the difference was not significant (Table 1). Patients in the fluctuating group were significantly younger and their first PSA level was higher. The PSA slope in the fluctuating group was negative, as opposed to that of the ‘steady’ group (Table 1).

Table 1.  The characteristics of the study group, and of the men with prostate cancer, according to PSA variation. Continuous values are given as the median (range)
VariableAll‘Steady’ PSAFluctuating PSAP
  1. PB, prostate biopsy.

N (%) 191 112 (58.6) 79 (41.4) 
Age, years 63.0 64.1 (47.2–81.2) 61.6 (48.4–80.1)0.02
PSA density  0.17  0.16 (0.05–0.86)  0.18 (0.04–11.2)0.2
PSA before 1st PB, ng/mL  7.0  6.2 (1.7–36.4)  8.1 (1.5–450)0.008
Highest free PSA % 15 15 (7–40) 15 (5–39)0.8
PB1 to PB2 interval, days290.0296.0 (24–2580)279.0 (51–2760)0.4
PSA before 2nd PB, ng/mL  8.0  8.5 (2.4–69.6)  7.5 (0.9–429)0.04
PSA slope, ng/mL/month  3.6 × 10−4  0.3 × 10−2 (−7.4 × 10−5 to + 0.12) −0.1 × 10−2 (−0.4 to + 0.01)<0.001
Men with prostate cancer
N (%) 53 (27.7) 36 (32.1) 17 (21.5)0.14
Age, years 65.9 66.0 (54.0–81.2) 64.5 (55.9–74.8)0.5
T1c stage, n (%) 39 25 (69) 14 (82)0.8
PSA preceding 1st PB, ng/mL  7.0  6.2 (1.7–36.4)  8.4 (1.5–450)0.1
Gleason score  6  6 (5–8)  6 (5–9)0.4

In the subgroup of patients with a PSA level of <10 ng/mL before both the first and second biopsy (114, 85 with a steady or rising PSA level and 29 with a fluctuating PSA level) the results of the statistical tests remained unchanged.

In the 53 patients diagnosed with prostate cancer, the 36 patients with a steady or rising PSA level and the 17 with a fluctuating PSA level were not significantly different in age and acknowledged prognostic factors, specifically T stage, Gleason score and PSA level (Table 1). However, in the subgroup of patients with a PSA level of <10 ng/mL before both the first and second biopsy (19 with a ‘steady’ and 10 with a fluctuating PSA level) the first PSA level was higher in the fluctuating group (median PSA 6.7 for the latter vs 5.0 ng/mL for the steady or rising PSA group; P = 0.03).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

While recognizing the importance of PSA in monitoring established disease and staging, many authors suggested that PSA would be of limited utility for screening or early detection, owing to a lack of specificity [3]. So-called PSA derivatives, including age-specific PSA levels, PSA density and PSA kinetics, as well as molecular forms of PSA, have been proposed to enhance the performance characteristics of PSA, but none of them is widely and routinely used. One of these methods is based on PSA velocity, first described by Carter et al.[1]. Measuring PSA velocity, or the change of PSA level over time, makes intrinsic sense. In men who progress to advanced prostate cancer, there is invariably an increase in PSA level, exceeding the generally slow rise in PSA in healthy men over time.

The concept of PSA velocity is well-known by urologists and is very likely to influence their reasoning in decisions about prostate biopsy, more specifically in the commonly encountered patients with fluctuating PSA levels on successive measurements. Although fluctuating PSA levels do not preclude a significantly increasing slope, these patients are more likely to be considered by the practitioner as having a greater chance of being cancer-free than patients with steadily increasing or even steady PSA levels. However, we found no published evidence to support this hypothesis, which led us to conduct the present study.

We chose a population with repeat biopsies for two reasons. First, there is no consensus about the indications for repeat biopsies, which makes it more difficult for the practising urologist. Second, there is less chance of missing a cancer after at least two sets of biopsies [4]. Unexpectedly, in the present study, although the PSA slope was slightly but significantly greater in the men with cancer, a fluctuating PSA level was not predictive of a negative biopsy. Moreover, in patients with prostate cancer, it was not associated with acknowledged prognostic factors, including total PSA level before biopsy, clinical T stage and Gleason score.

With the results of Carter et al.[1], relying on the PSA velocity to indicate a prostate biopsy is far from being in systematic use, one reason being that other studies failed to reproduce their results [3,5,6]. These discrepancies can probably be explained by considering that the velocity in the study of Carter et al. was calculated using long intervals, much longer than those considered in routine practice; the PSA determinations were separated by ≥ 7 years. Intra-individual PSA variability, even when PSA is measured in the same laboratory, is now established [3]. The studies by Nixon et al.[7,8] showed that before a change in PSA level could be ascribed to a real prostatic disease process, there must be a 25% increase. This calculation takes into account biological variation and laboratory issues. For free PSA, an even greater change of 36% must be recorded. Many studies have confirmed the intra-individual PSA variability; Kristal et al.[9] studied 3341 men who participated in the placebo arm of the Prostate Cancer Prevention Trial and were free of prostate cancer. These authors found that several demographic and lifestyle factors were associated with PSA level, and concluded that the magnitude of these associations was small and was unlikely to bias the interpretation of one PSA test. By contrast, race, smoking, age, energy intake, calcium supplement use, and weight change were associated with substantial differences in PSA velocity, and the clinical interpretation of PSA velocity might be biased. Eastham et al.[10] found, in a retrospective analysis of an unscreened population of 972 men, that 44% of 154 participants with a PSA level of >4 ng/mL had a normal PSA level at one or more subsequent visits during a 4-year follow-up. They concluded that an isolated increase in PSA level should be confirmed several weeks later before proceeding with further testing, including prostate biopsy. However, these studies did not assess the impact on the risk of having prostate cancer when PSA values fluctuated in the same patient.

The weakness of the present study is that it was retrospective; moreover, for reasons described above, we only included the subset of patients who had a first negative biopsy. Obviously, the strength of evidence does not equal that from a prospective study with a longitudinal follow-up. However, until such a study is conducted, our results, based on a database of consecutive patients, are the first to our knowledge to provide elements to answer the question of whether it should be reassuring for the patient and the practitioner when successive PSA levels ‘go up and down’. The study made no case for an affirmative answer, as there was no significant difference between these patients and those with a steady or steadily increasing PSA level in the outcome of the prostate biopsy, nor any association with prognostic factors in men with prostate cancer. These findings should be confirmed, as they have economic and strategic implications. They question the interest of confirming in the short-term a 4–10 ng/mL PSA value. This issue was also raised recently by a British study presented at the 2006 congress of the AUA [11]; those authors found that using thresholds of 3 and 4 ng/mL, 16.9% and 22.4%, respectively, of men with diagnosed prostate cancer had PSA levels below the threshold on a repeated measurement.

In conclusion, in the present study, by contrast with the common and yet unfounded view, the risk of having a positive repeat prostate biopsy was no lower in men with a fluctuating PSA level than in those with a steady or steadily increasing PSA level. The practical and economic implications warrant further studies to confirm these findings.

CONFLICT OF INTEREST

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

None declared. Source of funding: University Hospital.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES
  • 1
    Carter HB, Morrell CH, Pearson JD et al. Estimation of prostatic growth using serial prostate-specific antigen measurements in men with and without prostate disease. Cancer Res 1992; 52: 33238
  • 2
    Porter JR, Hayward R, Brawer MK. The significance of short-term PSA change in men undergoing ultrasound guided prostate biopsy. J Urol 1994; 151: 293A
  • 3
    Brawer MK. Prostate-specific antigen: current status. CA Cancer J Clin 1999; 49: 26481
  • 4
    Djavan B, Remzi M, Schulman CC, Marberger M, Zlotta AR. Repeat prostate biopsy: who, how and when? A review. Eur Urol 2002; 42: 93103
  • 5
    Roobol MJ, Kranse R, De Koning HJ, Schroder FH. Prostate-specific antigen velocity at low prostate-specific antigen levels as screening tool for prostate cancer: results of second screening round of ERSPC (Rotterdam). Urology 2004; 63: 30915
  • 6
    Raaijmakers R, Wildhagen MF, Ito K et al. Prostate-specific antigen change in the European Randomized Study of Screening for Prostate Cancer, section Rotterdam. Urology 2004; 63: 31620
  • 7
    Nixon RG, Wener MH, Smith KM, Parson RE, Strobel SA, Brawer MK. Biological variation of prostate specific antigen levels in serum: an evaluation of day-to-day physiological fluctuations in a well-defined cohort of 24 patients. J Urol 1997; 157: 218390
  • 8
    Nixon RG, Wener MH, Smith KM, Parson RE, Blase AB, Brawer MK. Day to day changes in free and total PSA: significance of biological variation. Prostate Cancer Prostatic Dis 1997; 1: 906
  • 9
    Kristal AR, Chi C, Tangen CM, Goodman PJ, Etzioni R, Thompson IM. Associations of demographic and lifestyle characteristics with prostate-specific antigen (PSA) concentration and rate of PSA increase. Cancer 2006; 106: 3208
  • 10
    Eastham JA, Riedel E, Scardino PT et al.; Polyp Prevention Trial Study Group. Variation of serum prostate-specific antigen levels: an evaluation of year-to-year fluctuations. JAMA 2003; 289: 2695700
  • 11
    Hamdy F, Dedman D, Oliver S et al. The value of repeat serum PSA measurements in the detection of prostate cancer – experience from the feasibility phase of the UK Protect Study. J Urol 2006; 175 (Suppl.): Abstract 472