Patterns of Repeat Prostate Biopsy in Contemporary Clinical Practice Nitya E. Abraham, Neil Mendhiratta and Samir S. Taneja*,† From the Division of Urologic Oncology, Department of Urology, New York University School of Medicine, New York, New York

Abbreviations and Acronyms ASAP ¼ atypical small acinar proliferation CaP ¼ prostate cancer DRE ¼ digital rectal examination GS ¼ Gleason score HGPIN ¼ high grade prostatic intraepithelial neoplasia MRI ¼ magnetic resonance imaging PSA ¼ prostate specific antigen mp-MRI ¼ multiparametric MRI TRUS ¼ transrectal ultrasound UCSF-CAPRA ¼ University of California-San Francisco Cancer of the Prostate Risk Assessment Accepted for publication October 15, 2014. Study received local institutional review board approval. * Correspondence: Division of Urologic Oncology, Department of Urology, New York University Langone Medical Center, 150 East 32nd St., Suite 200, New York, New York 10016 (telephone: 646-825-6321; FAX: 646-825-6399; e-mail: [email protected]). † Financial interest and/or other relationship with Hitachi-Aloka, Healthtronics, Elsevier, Trod and Steba Biotech.

Purpose: The objectives of this study were to 1) describe the patterns of repeat prostate biopsy in men with a previous negative biopsy and 2) identify predictors of prostate cancer diagnosis on repeat biopsy in these men. Materials and Methods: From a university faculty group practice we identified 1,837 men who underwent prostate biopsy between January 1, 1995 and January 1, 2010. Characteristics of repeat biopsy were examined, including the indication for biopsy, the number of repeat biopsies performed, the number of cores obtained and total prostate specific antigen before biopsy. Features of prostate cancer diagnosed on repeat biopsy were examined, including Gleason score, number of positive cores, percent of tumor and treatment choice. Multivariable logistic regression was done to identify prostate cancer predictors. Results: Initial biopsy was negative in 1,213 men. In 255 men a total of 798 repeat biopsies were performed. Of the 63 men diagnosed with prostate cancer Gleason score was 6 or less in 33 (52%), 7 in 22 (35%) and 8e9 in 8 (13%). When categorized by Epstein criteria, the rate of clinically insignificant cancer diagnosis decreased substantially by the third and fourth repeat biopsies. Repeat biopsy in men older than 70 years, biopsies including more than 20 cores and the fourth repeat biopsy were associated with an increased likelihood of prostate cancer diagnosis. Conclusions: In men selected for multiple repeat biopsies clinically significant cancer is found at each sampling round. Given the continued likelihood of cancer detection even by the fifth biopsy, early consideration of saturation or image guided biopsy may be warranted in the repeat biopsy population. Key Words: prostate, prostatic neoplasms, biopsy, magnetic resonance imaging, diagnosis

CONTEMPORARY systematic biopsy technique relies on sampling efficiency for cancer detection. As such, the risk of missing cancer on initial prostate biopsy due to sampling error is substantial, often resulting in the need for repeat biopsy upon continued suspicion of CaP. While sampling efficiency can be improved by increased sampling, the cancer detection rate on repeat biopsy varies highly.1,2

1178

j

www.jurology.com

Previous studies show a decrease in the CaP diagnosis rate with each repeat biopsy whether using serial sextant biopsy3,4 or extended core sampling.5 Currently with the exception of men found to have ASAP on initial biopsy there is no clear consensus on the indications or techniques of repeat biopsy. Urologists typically rely on clinical judgment to determine

0022-5347/15/1934-1178/0 THE JOURNAL OF UROLOGY® © 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.

http://dx.doi.org/10.1016/j.juro.2014.10.084 Vol. 193, 1178-1184, April 2015 Printed in U.S.A.

PATTERNS OF REPEAT PROSTATE BIOPSY IN CONTEMPORARY PRACTICE

1179

the need for repeat sampling and on personal preference to determine the biopsy approach. Collectively the variable cancer detection rate of repeat biopsy and the absence of clear guidelines on indications lead to the potential for excessive repeat biopsies, subsequently escalating biopsy related complications and cost. We performed this study to evaluate the frequency, indications and outcomes of repeat prostate biopsy in a single institution, academic urological practice.

were investigated, including GS, number of positive cores, percent of tumor involvement, treatment choice and radical prostatectomy pathology if the patient was treated with surgery. We used multivariable logistic regression to identify factors associated with CaP detection. The primary dependent variable was CaP diagnosis. Covariates potentially associated with the CaP diagnosis included age at biopsy, reason for biopsy, TRUS volume, prebiopsy PSA, number of previous biopsies and total number of biopsy cores. Analysis was done with StataÒ 11.0. All hypotheses were 2-tailed and the critical a level was 0.05.

MATERIALS AND METHODS

RESULTS

After receiving local institutional review board approval we queried the billing records of a single university faculty group practice to identify men with no previous diagnosis of CaP who underwent prostate biopsy between January 1, 1995 and January 1, 2010. Cases after 2010 were excluded from analysis after integrating MRI guided strategies for prostate biopsy into our institutional clinical practice algorithms. Men were identified by the ICD-9 code 55700 in the billing database. At the time that ICD-9 code 185 was added to the record men were documented as diagnosed with CaP and, thus, excluded from additional analysis. Of men with an initial negative biopsy those without a subsequent 55700 code were assumed not to have undergone additional biopsy and were also excluded. Clinical characteristics at repeat biopsy were examined, including biopsy indication, total PSA, PSA density and PSA velocity before biopsy. Biopsy characteristics were also recorded, including number of repeat biopsies performed, interval between biopsies and number of cores obtained. Features of CaP diagnosed on repeat biopsy

During the study period 1,837 men underwent a total of 2,430 prostate biopsies. Initial biopsy was negative in 1,213 men. The treating urologist selected 255 men for repeat biopsy, in whom a total of 798 biopsies were performed during the study period (fig. 1). The remaining 958 men with negative biopsy were believed to have no clinical indication for rebiopsy or they did not return for followup care. Of 255 men selected for rebiopsy 140 (55%) underwent 2 or more repeat biopsies, including 5 or more in 18 (6.3%) (table 1). Prebiopsy PSA was 4 or less, 4.1 to 10 and greater than 10 ng/ml in 126 (16%), 443 (56%) and 151 biopsies (19%), respectively (table 1). Most repeat biopsies involved extended sampling (12 or more cores), including 107 (42%), 68 (50%), 50 (63%) and 24 (65%) on biopsies 2, 3, 4 and 5, respectively. The most common indication for biopsy in the repeat biopsy population was increasing/increased

Figure 1.

1180

PATTERNS OF REPEAT PROSTATE BIOPSY IN CONTEMPORARY PRACTICE

PSA (508 biopsies or 64%) followed by HGPIN (118 or 15%), combined increasing/increased PSA and HGPIN history (55 or 7%), ASAP (45 or 6%) and abnormal DRE (20 or 3%) (fig. 2). Median time to repeat biopsy was 12.8 months (IQR 0e32.4). Of the 45 rebiopsies performed after ASAP was identified 33 (73.3%) were done within 6 months of the initial

diagnosis. Most repeat biopsies performed due to increasing/increased PSA (226 of 290 or 77.9%), HGPIN (85 of 118 or 72.0%) and PSA plus HGPIN (45 of 55 or 81.8%) were done 12 months or longer after the prior biopsy (fig. 3). CaP was found in 63 repeat biopsies (8%), HGPIN was found in 140 (18%) and ASAP was found in 50 (6%). A decreasing CaP diagnosis rate was observed on serial repeat biopsy. The detection rate was 25 of 250 (10%), 15 of 137 (11%), 8 of 80 (10%) and 11 of 44 diagnoses (25%) on repeat biopsies 1, 2, 3, 4 and 5, respectively (fig. 1). Five men were diagnosed with CaP on repeat biopsy 5 or greater. Of men diagnosed with CaP on repeat biopsy 33 (52%), 22 (35%) and 8 (13%) were diagnosed with GS 6 or less, 7 and 8e9 disease, respectively (table 2). Only 1 core was positive in 30 biopsies (48%). Total tumor involvement was 5% or less in 37 biopsies (59%). When categorized into clinically insignificant cancer by Epstein criteria (PSA density less than 0.15 ng/ml, GS 6 or less, 2 or fewer positive cores and 50% or less involvement in a single core) and UCSF-CAPRA (score 2 or greater based on age, PSA at diagnosis, GS, clinical stage and percent of biopsy core involved with cancer),6 the rate of clinically insignificant cancer diagnosis decreased substantially by the third and fourth rebiopsies (fig. 1). Surgery was performed in 20 men diagnosed with CaP. In 6 of these men (30%) tumor volume was 5% or less (table 3). GS was 6 in 9 patients (45%), 7 in 10 (50%) and 8 in 1 (5%). Multivariable analysis revealed that CaP was significantly more likely to be diagnosed in men older than 70 years, biopsies including 20 cores or more and on repeat biopsy 4 (table 4). CaP was significantly less likely to be identified in biopsies of a prostate with a TRUS volume of greater than 50 cc.

Figure 2. Indication for 798 biopsies. Red bars indicate CaP. Blue bars indicate no CaP.

Figure 3. Time to repeat biopsy by indication

Table 1. Rebiopsy characteristics No./Total No. (%) No. rebiopsies (255 pts): 1 2 3 4 5 of Greater Biopsy reason: Increasing/increased PSA HGPIN Increasing/increased PSA þ HGPIN ASAP Abnormal DRE Missing Pts diagnosed with CaP Ca pos biopsies by biopsy reason: Increasing/increased PSA HGPIN Increasing/increased þ HGPIN ASAP Abnormal DRE Missing Greater than 12 cores (biopsy No.): 1 2 3 4 5 Prebiopsy PSA (ng/ml): 0e4 ng/ml Greater than 4e10 Greater than 10 Missing

798 115 59 35 28 18 798 508 118 55 45 20 52 63

(45.1) (23.9) (13.7) (11.0) (6.3) (63.6) (14.8) (6.9) (5.6) (2.5) (6.5) (24.8)

32/508 9/118 11/55 9/45 0/20 2/52

(6.3) (7.6) (20) (20)

23/250 107/250 68/137 50/80 24/44

(9.2) (42.8) (49.6) (62.5) (54.5)

126 443 151 78

(15.8) (55.5) (18.9) (9.8)

(3.8)

PATTERNS OF REPEAT PROSTATE BIOPSY IN CONTEMPORARY PRACTICE

Table 2. CaP features diagnosed on 63 rebiopsies

Table 4. Multivariable analysis of CaP detection predictors No. (%)

Clinical stage: T1c T2a Missing GS: 6 or Less 7 8e9 No. pos cores: 1 2 3 or Greater Missing % CaP cores: 5 or Less 10 Greater than 10 Max % CaP/1 core: 5 or Less 6e50 Greater than 50 Missing CaP laterality: Unilat Bilat No. prediagnosis rebiopsies: 1 2 3 4 or Greater CaP treatment: Radical retropubic prostatectomy External radiotherapy/brachytherapy Active surveillance High intensity focused ultrasound Watchful waiting Unknown

40 (63.5) 6 (9.5) 17 (27.0) 33 (52.4) 22 (34.9) 8 (12.6) 30 15 16 2

(47.6) (23.8) (25.4) (3.2)

37 (58.7) 12 (19.1) 14 (22.2) 21 24 15 3

(33.0) (38.1) (23.8) (4.7)

53 (84.1) 10 (15.9) 25 15 8 15

(39.7) (23.8) (12.7) (23.8)

20 17 6 2 2 16

(31.7) (27.0) (9.5) (3.2) (3.2) (25.4)

DISCUSSION Clinicians have yet to find the balance between avoiding the morbidity of repetitive prostate biopsy while still identifying clinically significant CaP in men with an initial negative prostate biopsy and persistent suspicion for CaP. Recommendations on when to repeat prostate biopsy have been reported in the literature. Presti recommended repeat biopsy if initial sampling was inadequate (ie sextant biopsy) using a low free-to-total PSA ratio as the Table 3. CaP features in 20 prostatectomy specimens

GS: 6 7 8 % Tumor vol: 5 or Less 10 Greater than 10 Missing Pathological stage: T2 T3

No. 1 Rebiopsy (%)

No. Greater Than 1 Rebiopsy (%)

7 2 (28.6) 5 (71.4) 0

13 7 (53.8) 5 (38.5) 1 (7.7)

2 2 2 1

(28.6) (28.6) (28.6) (14.3)

5 (71.4) 2 (28.6)

1181

4 3 5 1

(30.8) (23.1) (38.5) (7.7)

9 (69.2) 4 (30.8)

OR (95% CI) Age at biopsy (vs 40e60): 61e70 Greater than 70 No. previous biopsies (vs 1): 2 3 4 5 or Greater Biopsy reason (vs increasing/increased PSA): ASAP HGPIN PSA þ HGPIN Prebiopsy ng/ml PSA (vs less than 4): 4e10 Greater than 10 TRUS cc vol (vs less than 30): 31e49 50 or Greater No. biopsy cores (vs 12 or less): 13e20 Greater than 20

1.41 (0.65e3.06) 5.62 (2.52e12.5)* 0.80 1.04 2.98 1.46

(0.37 e1.72) (0.41e2.64) (1.13e7.82)* (0.44e4.86)

1.77 (0.68e4.66) 0.54 (0.21e1.37) 1.78 (0.76e4.14) 1.52 (0.52e4.41) 1.19 (0.34e4.19) 0.78 (0.31e1.95) 0.25 (0.09e0.73)* 1.72 (0.89e3.30) 3.77 (1.46e9.73)*

* p

Patterns of repeat prostate biopsy in contemporary clinical practice.

The objectives of this study were to 1) describe the patterns of repeat prostate biopsy in men with a previous negative biopsy and 2) identify predict...
522KB Sizes 2 Downloads 10 Views