Stage pt0 After Radical Prostatectomy With Previous Positive Biopsy Sets: A Multicenter Study

Transcription

1 Stage pt0 After Radical Prostatectomy With Previous Positive Biopsy Sets: A Multicenter Study Thomas Bessède, Michel Soulié, Nicolas Mottet, Xavier Rebillard, Michaël Peyromaure, Vincent Ravery, Laurent Salomon* and the Cancerology Committee of the French Urological Association From the Centre Hospitalier Universitaire Henri Mondor (TB, LS), Créteil, Centre Hospitalier Universitaire Rangueil (MS), Toulouse, Clinique Mutualiste Saint-Etienne (NM), Saint-Etienne, Clinique Beausoleil (XR), Montpellier and Centre Hospitalier Universitaire Cochin (MP) and Centre Hospitalier Universitaire Bichat (VR), Paris, France Abbreviations and Acronyms ASAP atypical small acinar proliferation NHT neoadjuvant hormonal therapy PIN prostatic intraepithelial neoplasia PSA prostate specific antigen RP radical prostatectomy TURP transurethral prostate resection Submitted for publication June 18, * Correspondence: Department of Urology, Hospital Henri Mondor, 51 Avenue du Maréchal de Lattre de Tassigny, Créteil, France (telephone: 33 (0) ; FAX: 33 (0) ; laurent.salomon@hmn.aphp.fr). Purpose: We analyzed preoperative data, pathological results and followup of pt0 tumors after radical prostatectomy for prostate cancer diagnosed on previous positive biopsy. Materials and Methods: At 6 centers a total of 30 of 7,693 radical prostatectomy specimens were classified as pt0 despite prior biopsy proven prostate cancer. No patients were diagnosed after transurethral prostate resection or received neoadjuvant hormonal treatment. All biopsy cores and radical prostatectomy specimens were reanalyzed by a second pathologist. Followup comprised clinical examination and postoperative prostate specific antigen assay at 1 and 3 months, and every 6 months thereafter. Results: Median patient age was 63 years (range 46 to 73). Median preoperative prostate specific antigen was 7.4 ng/ml (range 1.3 to 23). Of the cases 24 were T1c and 6 were T2a. The median number of biopsy cores was 10 (range 6 to 21) with 1 positive (range 1 to 4). On biopsies median tumor length was 1 mm (range 0.3 to 18) and there was tumor in 11.1% (range 3.4% to 64%). In 25 cases (83.3%) there was only 1 positive biopsy. Gleason score was 3 3 in 23 cases and less than 6 in 5 with grade 4 in 2. Only 9 cases filled all nonsignificant tumor criteria. Median specimen weight was 61 gm (range 40 to 160). At a median 82-month followup (range 14 to 226) there was no biochemical progression. Conclusions: After biopsy proven cancer pt0 prostate cancer is an unpredictable pathological finding. Despite its excellent prognosis it has medicolegal repercussions that justify DNA based tissue analysis. There is no evidence that finding focal cancer after extensive prostate resection changes patient prognosis and postoperative treatment. Key Words: prostate; prostatectomy; prostatic neoplasms; neoplasm regression, spontaneous; biopsy TREATMENT in patients who have undergone RP for prostate cancer is based on specimen analysis and staging according to the 2002 TNM classification. 1 It includes stage pt0 for cases in which no residual prostate tumor is found, as may be expected after neoadjuvant treatment or TURP. 2,3 Apart from these 2 situations pt0 prostate cancer has been reported in patients with biopsy proven prostate cancer and no treatment before RP. 4 In single center studies the estimated incidence is between 0.15% and 2.9%, and the prognosis is excellent. 5,6 However, its significance and clinical con /10/ /0 Vol. 183, , March 2010 THE JOURNAL OF UROLOGY Printed in U.S.A by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. DOI: /j.juro

2 STAGE pt0 AFTER PROSTATECTOMY WITH PREVIOUS POSITIVE BIOPSY 959 sequences remain unclear for pathologists and urooncologists. In what to our knowledge is the largest multicenter series of confirmed pt0 stage in RP cases after biopsy diagnosed prostate cancer we analyzed preoperative data, pathological results and followup of pt0 tumors after RP. MATERIALS AND METHODS From 1998 to 2006 a total of 30 of 7,693 RPs (0.39%) at 6 centers were classified as pt0 (TNM 2002). All cancer cases were diagnosed by ultrasound guided prostate biopsy. None were diagnosed after TURP. No neoadjuvant hormonal treatment was delivered before RP. Each specimen was analyzed according to the Stanford protocol. 7 Patient age, clinical stage, preoperative serum PSA, prostate biopsy results (number of sets, number of cores and positive cores, tumor length and percent per core, and Gleason score), specimen weight and analysis report were noted. After the first specimen analysis all biopsy cores and RP specimens were reanalyzed by a second pathologist at each center. Cases were selected after confirmation of pt0 status. Postoperative followup comprised clinical examination and PSA assay at 1 and 3 months, and every 6 months thereafter. Biochemical progression was defined as postoperative PSA greater than 0.2 ng/ml and increasing. 8 RESULTS The table lists patient characteristics. In all cases prostate cancer on positive biopsy diagnosis and RP specimen pt0 were confirmed by a second pathologist even after reprocessing. Median patient age was 63 years (range 46 to 73). Median preoperative PSA was 7.4 ng/ml (range 1.3 to 23). Of the patients 28 (93%) had increased serum PSA and digital rectal examination was abnormal in 6 (20%). A total of 27 patients (90%) were diagnosed on the first biopsy set and others had 2 to 5 sets. Gleason score of positive biopsies was 6 (3-3) in 23 cases (77%) and less than 6 in 5 (17%), and there was grade 4 in 2 (7%). On biopsies median tumor length was 1 mm (range 0.3 to 18) and there was tumor in 11.1% (range 0.3% to 60%). According to Epstein criteria only 9 cases were considered probably nonsignificant tumor with T1c clinical stage, PSA less than 10 ng/ml, 1 positive biopsy, tumor length less than 3 mm, percent of tumor less than 15% and Gleason score less than 7 with no grade 4. Median specimen weight was 61 gm (range 40 to 160). PIN and ASAP were present in 17 and 3 cases, respectively. At a median 82-month followup (range 30 to 241) no clinical or biological recurrence was detected. Pt No. Age Preop Stage PSA (ng/ml) No. Biopsy Sets No. Cores/ No. Pos Gleason Score Tumor Length Specimen mm % Biopsies Wt (gm) Report Followup (mos) 1 63 T1C / High grade PIN T1C / T2A / T1C / T2A / Dysplasia T1C / ASAP T1C / High grade PIN T1C / T1C / T1C / T1C / PIN T1C / T1C / PIN T1C / PIN T2A / PIN T1C / PIN T1C / PIN T1C / T1C / PIN T2A / PIN T2A / PIN T1C / High grade PIN T1C / T1C / High grade PIN ASAP T1C / High grade PIN T1C / High grade PIN T1C / T1C / T1C / PIN T2A / PIN 30

3 960 STAGE pt0 AFTER PROSTATECTOMY WITH PREVIOUS POSITIVE BIOPSY DISCUSSION The 2002 TNM classification for prostate cancer includes T0 for absent primitive tumor in a specimen. Previously ct0 was described in patients with positive prostate biopsies performed at their request in the absence of clinical or biological abnormality. 9 Stage pt0 has been noted after NHT, TURP and/or positive biopsy. All cases in our study were diagnosed on a biopsy set followed by RP only, thus avoiding specimen changes due to the effects of hormonal therapy. Depending on duration NHT could induce pt0 status in 2.25% to 15% of RP specimens. 10,11 There was no favorable prognosis in PSA relapse-free survival in post-nht pt0 cases. 12 In a retrospective analysis an Italian group concluded that pt0 cases after NHT may have a worse clinical outcome than untreated pt0 cases. 13 After a diagnosis of prostate cancer in the TURP specimen the risk of pt0 in the RP specimen was estimated at 6% to 25% in ct1a and 3% to 9% in ct1b cases. 3 No recurrence was observed at 60- month followup, indicating an excellent overall prognosis for pt0 prostate cancer without NHT. In the series by Bostwick and Bostwick most patients were diagnosed by TURP specimen examination with ct1a in 42%, ct1b in 45% and ct2 in 13%, and no recurrent cancer diagnosed at a mean 9.6-year followup. 14 In these cases absent residual tumor may be explained by TURP that would have curatively removed all cancer. The first report of pt0 cases belonged to the vanishing cancer phenomenon description. 4 Goldstein et al considered 11 minimal pt2a and 2 pt0 cancers that required special efforts by pathologists to be documented or diagnosed. 4 A mean of 79 prostate slides (range 34 to 248) were examined per specimen and there was extreme difficulty or impossibility of finding residual cancer despite extensive sampling. The prevalence of pt0 prostate cancer despite positive biopsy is 0.15% to 1.5%. 5,15 It was up to 2.9% in a cohort of screened patients, 6 suggesting that screening programs enhance the prostate cancer prevalence and modify the characteristics of detected cases. Descazeaud et al analyzed 11 cases to determine 4 preoperative characteristics predicting a pt0 finding. 16 In this single center ctic population the pt0 rate was 0.5% and patients were disease-free at 30-month followup. According to the curative biopsy theory focal cancer could be entirely removed by biopsy, explaining pt0 after biopsy proven cancer. 17 However, this theory is insufficient to explain findings in several patients in our series with multiple positive biopsies (patients 2, 5, 9, 23 and 30), a nodular prostate (patients 3, 5, 15, 20, 21 and 30) and greater than 2 mm cancer (patients 5, 10, 11, 15, 16, 17 and 30). In all of these significant preoperative cancer cases the tumor was expected to be greater than the biopsy width and could not entirely be removed at biopsy. DiGiuseppe et al defined tumors less than 0.1 cc as minute and 0.03 cc tumors as difficult to find, 18 while Goldstein et al reported a mean cc tumor volume. 4 However, to our knowledge the volume and percent of prostate tissue analyzed during routine (Stanford) and extensive protocols remain unknown in the literature, although this would enable calculation of the probability of finding minimal tumors and a percent to determine tumor volume in the prostate. A minimal prostate volume threshold should be processed to establish a pt0 diagnosis. In patients who have accepted the risk of RP and the threat of malignant disease pt0 is an unusual, concerning result that may generate some incomprehension. They may question the validity of preoperative biopsy, the relevance of surgical indications, the reliability of pathology reports and the certainty of the prognosis. By redounding on medical credibility a pt0 finding may have medicolegal repercussions, especially in patients with a long recovery and adverse surgical effects. 19 To avoid such conflicts a clearly defined protocol should be established to manage this particular oncological result. For specimens in which initial routine examination does not reveal residual cancer Mazzucchelli et al established a 9-step cancer searching protocol, including 1) diagnostic needle biopsy review, 2) surgical specimen slide review, 3) search for nonembedded tissue to be possibly processed in toto, 4) additional sections in positive core areas, 5) additional sections after block flipping, 6) immunostaining for p63 and -methylacyl-coenzyme A racemase, 7) immunostaining for cytokeratin CAM 5.2, p63 and PSA, 8) search for missing tissue for surgical or technical reasons, and 9) DNA specimen and biopsy analysis. 5 The pathology protocol decreased the number of pt0 cases from 8 (0.6%) to 2 (0.15%) and resolved more than half of the initially discordant pathologist conclusions. Cancer was found after extensive specimen resectioning in 5 cases and after immunostaining in 1. Thus, small tumors may be missed upon first specimen processing since their dimensions are less than the section thickness but Mazzucchelli et al mentioned evident regressive changes due to endocrine therapy in most patients. In our series no prior treatment interfered with RP specimen examination. The second pathologist review excluded any false-positive biopsies, making the over diagnosis theory also unlikely to explain the discordance. Specimen and biopsy reexamination included most of the protocol steps of Mazzucchelli et al except the DNA process. 19 Polymerase chain re-

4 STAGE pt0 AFTER PROSTATECTOMY WITH PREVIOUS POSITIVE BIOPSY 961 action based microsatellite marker analysis determines whether biopsy and subsequent RP specimens are from the same patient 20 but in accordance with French legislation we could not perform retrospective DNA analysis without patient written consent. To date investigators who proceeded to this assessment for pt0 prostate cancer only noted 1 case in error of the 28 pt0 cases analyzed. 5,20,21 Patient identification errors exist in routine preoperative steps, usually due to slips or lapses of automatic actions, and they are difficult to prevent entirely. 19 In a few cases this first step in pt0 management could detect a tissue change. To our knowledge its prevalence remains unknown, and depends on local protocols and habits. We recommend DNA based tissue assessment to confirm pt0 identity. In our series only 9 cases met all criteria for nonsignificant tumor on the biopsy set. 22 Another 21 cases had significant tumor, of which 5 were classified as at intermediate risk for clinically localized prostate cancer (stage T2b, Gleason score 7 or PSA greater than 10 and 20 ng/ml or less), 23 highlighting that pt0 is an unpredictable finding. Whereas a higher recurrence rate could have been expected in significant or intermediate risk subgroups, no recurrence was noted in any group in our series. In our experience undetectable prostate cancer despite positive biopsy was associated with remission at 82- month followup. This was also found in patients without NHT in other studies, 3,4,13,14,16,21 suggesting that a pt0 finding restores an excellent prognosis. In a single case report Thwaini et al described recurrent bone metastasis in a pt0 case with nonsignificant tumor criteria diagnosed on prostate biopsy and without prior NHT but metastasis was not histologically confirmed. 24 Is it worth searching for focal cancer in an RP specimen in cases staged as pt0 after routine pathology examination? Without NHT the prognosis remains excellent despite severe preoperative criteria and a tumor finding upon resectioning. Moreover, cancer recurrence is an exceptional event that could not be predicted by the finding of focal cancer. For a pt0 finding after biopsy proven prostate cancer and without NHT, when possible we recommend DNA based identity comparison of the biopsy and the RP specimen with a blood sample as the referent. For example, under staging could be corrected if a possible specimen identification error was proved, avoiding the lack of chance in surveillance. If the first pathology analysis was done according to current guidelines with slides a maximum of 3 mm thick, there may be no benefit to specimen resectioning in patients. 14 The finding of focal cancer would probably not modify the standard clinicobiological followup that should be recommended to detect any biological recurrence after pt0, as for any other confirmed, locally confined prostate cancer. CONCLUSIONS No residual prostate cancer despite biopsy proven carcinoma is an unpredictable pathological finding that has medicolegal repercussions and requires codified management, including DNA based tissue analysis. Technical considerations such as specimen section intervals may explain failure to find a tumor with no significant criteria at biopsy but in some cases vanishing significant tumors remain inexplicable. Prognosis remains excellent and unaffected by significant positive biopsy criteria. There is no evidence that finding focal cancer after extensive prostate resection changes patient prognosis and postoperative treatment. APPENDIX Members of the Prostate Committee, Cancerology Committee, French Urological Association are Philippe Beuzeboc, François Cornud, Nicolas Gaschignard, Pascale Grosclaude, Vincent Molinie, Pierre Mongiat-Artus, Michaël Peyromaure, Pierre Richaud, Laurent Salomon, Frédéric Staerman, François Rozet and Cyril Bastide. REFERENCES 1. Greene FL, Page DL, Fleming ID et al: American Joint Committee on Cancer Manual for Staging Cancer, 6th ed. New York: Springer 2002; p Kollermann J, Feek U, Muller H et al: Nondetected tumor (pt0) after prolonged, neoadjuvant treatment of localized prostatic carcinoma. Eur Urol 2000; 38: Herkommer K, Kuefer R, Gschwend JE et al: Pathological T0 prostate cancer without neoadjuvant therapy: clinical presentation and followup. Eur Urol 2004; 45: Goldstein NS, Begin LR, Grody WW et al: Minimal or no cancer in radical prostatectomy specimens. Report of 13 cases of the vanishing cancer phenomenon. Am J Surg Pathol 1995; 19: Mazzucchelli R, Barbisan F, Tagliabracci A et al: Search for residual prostate cancer on pt0 radical prostatectomy after positive biopsy. Virchows Arch 2007; 450: Postma R, de Vries SH, Roobol MJ et al: Incidence and follow-up of patients with focal prostate carcinoma in 2 screening rounds after an interval of 4 years. Cancer 2005; 103: Stamey TA, MJ, Freiha FS, Redwine E: Morphometric and clinical studies on 68 consecutives radical prostatectomies. J Urol 1988; 139: Heidenreich A, Aus G, Bolla M et al: EAU guidelines on prostate cancer. Eur Urol 2008; 53: Corica FA and Bostwick DG: Clinically unsuspected and undetected (clinical stage t0) prostate cancer diagnosed on random needle biopsy. Urology 1999; 53: Schulman CC and Sassine AM: Neoadjuvant hormonal deprivation before radical prostatectomy. Eur Urol 1993; 24: Noguchi M, Noda S, Nakashima O et al: No residual tumor in a radical prostatectomy specimen after neoadjuvant hormonal therapy for localized prostate cancer. Oncol Rep 2002; 9: 1075.

5 962 STAGE pt0 AFTER PROSTATECTOMY WITH PREVIOUS POSITIVE BIOPSY 12. Kollermann J, Hopfenmuller W, Caprano J et al: Prognosis of stage pt0 after prolonged neoadjuvant endocrine therapy of prostate cancer: a matched-pair analysis. Eur Urol 2004; 45: Prayer-Galetti T, Gardiman M, Sacco E et al: Finding of no tumor (pt0) in patients undergoing radical retropubic prostatectomy for clinically localized prostate cancer. Anal Quant Cytol Histol 2007; 29: Bostwick DG and Bostwick KC: Vanishing prostate cancer in radical prostatectomy specimens: incidence and long-term follow-up in 38 cases. BJU Int 2004; 94: Duffield AS and Epstein JI: Detection of cancer in radical prostatectomy specimens with no residual carcinoma in the initial review of slides. Am J Surg Pathol 2009; 33: Descazeaud A, Zerbib M, Flam T et al: Can pt0 stage of prostate cancer be predicted before radical prostatectomy? Eur Urol 2006; 50: Kommu S: A model to explain the vanishing prostate the curative biopsy theory. BJU Int 2004; 94: DiGiuseppe JA, Sauvageot J and Epstein JI: Increasing incidence of minimal residual cancer in radical prostatectomy specimens. Am J Surg Pathol 1997; 21: Suba E, Pfeiffer J and Raab S: Patient identification error among prostate needle core biopsy specimens: are we ready for a DNA time-out? J Urol 2007; 178: Cao D, Hafez M, Berg K et al: Little or no residual prostate cancer at radical prostatectomy: vanishing cancer or switched specimen? A microsatellite analysis of specimen identity. Am J Surg Pathol 2005; 29: Trpkov K, Yuan G, Hay R et al: No residual cancer on radical prostatectomy after positive 10-core biopsy: incidence, biopsy findings and DNA specimen identity analysis. Arch Pathol Lab Med 2006; 130: Epstein JI, Walsh PC, Carmichael M et al: Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA 1994; 271: D Amico AV, Whittington R, Malkowicz SB et al: Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 1998; 280: Thwaini A, Anjum F, Kalubac J et al: Vanishing prostate cancer in radical prostatectomy specimens: incidence and long-term follow-up in 38 cases. BJU Int 2004; 94: 1145.