prostate cancer with i brachytherapy peter d. grimm … · which these patients were treated, only...

PII S0360-3016(01)01601-7

CLINICAL INVESTIGATION Prostate

10-YEAR BIOCHEMICAL (PROSTATE-SPECIFIC ANTIGEN) CONTROL OFPROSTATE CANCER WITH 125I BRACHYTHERAPY

PETER D. GRIMM, D.O.,* JOHN C. BLASKO, M.D.,* JOHN E. SYLVESTER, M.D.,*†

ROBERT M. MEIER, M.D.,* AND WILLIAM CAVANAGH, B.S.*

*Seattle Prostate Institute, Swedish Medical Center, Seattle, WA;†Puget Sound Tumor Institute, Edmonds, WA

Purpose: To report 10-year biochemical (prostate-specific antigen [PSA]) outcomes for patients treated with125Ibrachytherapy as monotherapy for early-stage prostate cancer.Methods and Materials: One hundred and twenty-five consecutively treated patients, with clinical Stage T1–T2bprostate cancer were treated with125I brachytherapy as monotherapy, and followed with PSA determinations.Kaplan–Meier estimates of PSA progression-free survival (PFS), on the basis of a two consecutive elevations ofPSA, were calculated. Aggregate PSA response by time interval was assessed. Comparisons were made to anearlier-treated cohort.Results: The overall PSA PFS rate achieved at 10 years was 87% for low-risk patients (PSA< 10, Gleason Sum2–6, T1–T2b). Of 59 patients (47%) followed beyond 7 years, 51 (86%) had serum PSAs less than 0.5 ng/mL; 48(81%) had serum PSAs less than 0.2 ng/mL. Failures were local, 3.0%; distant, 3.0%. No patients have died ofprostate carcinoma. The proportion of patients with a PSA< 0.2 ng/mL continued to increase until at least 7–8years posttherapy. A plot of PSA PFS against the proportion of patients achieving serum PSA of less than 0.2ng/mL suggests a convergence of these two endpoints at 10 years. Patients treated in the era of this study(1988–1990) experienced a statistically improved PFS compared with an earlier era (1986–1987). This differenceappears independent of patient selection, suggesting that the maturation of the technique resulted in improvedbiochemical control.Conclusion: With modern technique, monotherapy with 125I achieves a high rate (87%) of biochemical andclinical control in patients with low-risk disease at 10 years. The decline of PSA following brachytherapy withlow-dose-rate isotopes can be protracted. Absolute PSA and PFS curves merge, and are comparable at 10 years.© 2001 Elsevier Science Inc.

Prostate cancer, Brachytherapy, Prostate-specific antigen, Radiotherapy.

INTRODUCTION

Radical prostatectomy, conformal external beam radiother-apy, and brachytherapy are the most common curative mo-dalities employed in the treatment of early-stage prostatecarcinoma. Current prostate brachytherapy techniquesevolved from an open laparotomy to a closed transperinealultrasound-guided approach (1). In recent years, more so-phisticated treatment planning and CT dosimetry have al-lowed further refinements in the transperineal technique.125I is the most common permanent radioisotope used forprostate brachytherapy.125I emits a low-energy (27-keV)photon and has a half-life of 60 days, which results in arelatively low initial dose rate of 7–10 cGy/h, at the pre-scription isodose contour.

This series represents the 10-year Seattle experience withtransperineal ultrasound-guided125I brachytherapy asmonotherapy for early-stage prostate cancer by an experi-enced team with techniques consistent with current implant

methods. The results are analyzed in relationship to PSAoutcomes and are compared to our early experience.

METHODS AND MATERIALS

Of patients presenting for evaluation of newly diagnosedprostate carcinoma at our institution between 1 January1988 and 31 December 1990, 126 received125I brachyther-apy as the sole form of treatment. The implant procedurewas a preplanned, ultrasound-guided method (1). Preloadedneedles, employing permanent125I sources (OncoSeed, Ny-comed-Amersham, Princeton, NJ), were used in every case.One patient was lost to follow-up before any postimplantPSA determinations; he is excluded from the study. Theremaining 125 patients comprise the basis for this study.

Patient selectionThis study describes a subset of consecutively treated,

prospectively followed patients. During the time period in

Reprint requests to: Dr. Peter Grimm, Seattle Prostate Institute,Suite 1101, 1101 Madison Street, Seattle, WA 98104. E-mail:[email protected]

Received Dec 7, 2000, and in revised form Mar 5, 2001.Accepted for publication Mar 13, 2001.

Int. J. Radiation Oncology Biol. Phys., Vol. 51, No. 1, pp. 31–40, 2001Copyright © 2001 Elsevier Science Inc.Printed in the USA. All rights reserved

0360-3016/01/$–see front matter

31

which these patients were treated, only patients who pre-sented with low- to moderate-grade, early-stage diseasewere considered candidates for125I monotherapy. No pa-tient underwent medical or surgical intervention for thepurpose of diminishing hormone levels. No form of radio-therapy other than the implant was employed. No other localor systemic treatment was prescribed. The intention of theseed implant was curative.

All patients were evaluated according to the result ofmedical history, physical examination, digital rectal exam-ination, and serum PSA measurement. Clinical stage wasassigned by digital rectal examination, and, where clinicallyindicated, chest X-ray, bone scan, and/or computed tomog-raphy of the pelvis were performed. Pathologic lymph-nodestaging was not performed. Clinical stage was assignedretrospectively according to the 1992 American Joint Com-mittee on Cancer (AJCC) staging system (2). T stage wasassigned strictly by digital rectal examination findings. Alldiagnostic biopsies were reviewed by the pathology staff atNorthwest Hospital and graded by the Gleason sum meth-odology. The median patient age was 70 years.

TreatmentThe Seattle technique of transperineal prostate brachy-

therapy has been reported elsewhere in detail (1). For radi-ation planning purposes, a volumetric study was performedseveral weeks before the implant, using transrectal ultra-sonography to define the prostate. The treatment volumeincluded the ultrasonically defined prostate plus a discre-tionary margin of 2–5 mm. Only the proximal portions ofthe seminal vesicles were included in the treatment volume.General brachytherapy treatment-planning software, run-ning on a PC platform, calculated a three-dimensionalsource matrix that would attain a minimum prescriptiondose of 160 Gy (145 Gy TG-43) encompassing the treat-ment volume. Sources were distributed in a uniform patternacross the target volume. The implant procedure was per-formed on an outpatient basis, utilizing preloaded 18-gaugeneedles. Transrectal ultrasound was the primary imagingmodality used for needle guidance.

Follow-upPatients were monitored by physical examination and

serum PSA determinations at 3- to 6-month intervals duringthe first 5 years postimplant, and yearly thereafter. Fivepatients for whom follow-up PSAs were obtained at amaximum of less than 2 years are included. All 5 expiredwith no evidence of disease within 3 years of implant; theiraverage age was 75 years at the time of implant. They areincluded here to preserve the consecutive nature of thecohort.

Biopsy, bone scan, or other study was performed uponthe occasion of a steadily rising PSA to define the locationof putative tumor progression.

Analytical methodsBiochemical outcomes are summarized in two ways.First, to describe biochemical success rates, cumulative

survival functions are employed in which each patient in thecohort is represented as either a failure or as a censoredobservation. For the purposes of this analysis, failure is“PSA progression failure,” defined as two consecutive risesin serum PSA, manifest at the time of last follow-up. Pa-tients who exhibited temporary elevations of PSA (5), fol-lowed by a pattern of decline, were not considered failuresbased on the two-rise failure definition. Time to failure isdefined as the midpoint between the lowest serum PSAattained and the first of the elevated values. Thus, thedetermination of biochemical failure, in this series, is sim-ilar to that of the American Society for Therapeutica Radi-ology and Oncology (ASTRO) Consensus Conference def-inition (3), except that only two rises are required forfailure, instead of three. Censored observations are recordedat the time of last PSA follow-up.

Second, a series of cross-sectional analyses was per-formed for the purpose of quantifying actual cohort PSA

Table 1. Clinical characteristics of the 125 patients

Numberof patients

% ofpatients

Gleason Grade2 10 83 17 13.64 33 26.45 47 37.66 18 14.4

Clinical StageT1a/b 10 8.0T1c 20 16.0T2a 77 61.6T2b 17 13.6T2c 1 0.8

Serum prostate-specificantigen (ng/mL)

0–4.0 54 43.24.1–10.0 42 33.610.0–20.0 19 15.220.0–30.0 5 4.0.30.0 4 3.2Unknown 1 0.8

Range 0.2–74.6Median 5.1Mean 8.1

Table 2. Clinical status at last PSA follow-up

Number (%) Median PSA

NED 109 (87.2) 0.1PSA progression only 8 (6.4) 3.8Local disease 4 (3.2) 17.4Distant disease 4 (3.2) 26.4

Abbreviations: NED 5 no evidence of disease; PSA5 prostate-specific antigen.

32 I. J. Radiation Oncology● Biology ● Physics Volume 51, Number 1, 2001

response. Intervals were defined as 6 months per interval forthe first 2 years, and 12 months per interval beyond 2 years.Patients were allowed only one PSA measurement per in-terval. Where two or more measurements were present, thelatest was considered, while the others were discarded. Ofspecial concern in examining the serum PSA datain totowere (1) a description of the effect of time upon response ofthe group as a whole, and (2) a description of the minimumPSA levels that may be reasonably expected following125Ibrachytherapy in this series.

In the case of all data presented, serum PSA valuessubsequent to the documentation of clinical failure are ex-cluded from analysis so that appropriate androgen depriva-tion for clinically evident recurrence does not influence thesubsequent PSA profile of these patients. The PSA value atthe time of clinical relapse is maintained as the last PSA ofrecord for that patient.

Clinical failure was defined as either distant failure, man-ifested as radiographically evident metastatic disease, or

local failure defined as either a positive biopsy or a positivedigital examination.

Cumulative survival functions were calculated by theKaplan–Meier method. Hypotheses regarding differences inbiochemical relapse-free survival functions were evaluatedusing the log–rank test. Linear 95% confidence intervalswere included where appropriate. The influence of multiplecovariates on PSA progression-free survival was estimatedby Cox regression; the influence of single covariates onPSA progression-free survival was evaluated using the log–rank test.

RESULTS

The presentation characteristics of the study group arecontained in Table 1. All patients presented with diagnosticbiopsies of 6 or less by Gleason sum methodology. Eighty-six percent (85.6%) of patients presented with tumors StageT2a or less by digital rectal examination. Seventy-sevenpercent (76.8%) of patients presented with serum PSA of 10ng/mL or less.

Table 2 documents the last clinical status of all 125patients. All patients who were diagnosed as either localfailures as confirmed by positive biopsy (n 5 4), or distantfailure as confirmed by positive bone scan or other imagingmodality (n 5 4), exhibited PSA progression of at least twosuccessive rises before the documentation of recurrent dis-ease. All 8 clinical failures were diagnosed with recurrentdisease before 5 years.

Fig. 1. Prostate-specific antigen progression-free survival for all 125 study patients.

Table 3. Last serum prostate-specific antigen (PSA, medianfollow-up, 81.4 months)

Serum PSA (ng/mL) No. (%)

#0.2 88 (70.4)#0.5 99 (79.2)#1.0 107 (85.6)#2.0 111 (88.8).2.0 14 (11.2)

33125I prostate brachytherapy● P. D. GRIMM et al.

Eight additional patients currently satisfy criteria for bio-chemical progression failure without demonstrable clinicaldisease at a median of 52 months. Thus, 16 of the 125patients (12.8%) meet our biochemical failure criteria in thisseries. Table 3 contains a summary of the last follow-upserum PSA measurement for all 125 patients.

Figure 1 depicts the Kaplan–Meier freedom from bio-chemical failure function using two successive elevations inserum PSA as the failure criteria. The Kaplan–Meier esti-mate of freedom from biochemical failure (FFBF) in thiscohort is 85.1% (95% confidence interval, 79.3–90.9%) at10 years.

Figures 2 and 3 and Table 4 summarize the interval dataaccumulated for these 125 patients. A total of 1133 unique,nonhormonally influenced follow-up serum PSA data pointswere available from these patients, an average of 9.1 perpatient. Figure 2 depicts the protracted time required toachieve very low PSA levels following125I brachytherapy.Figure 3 demonstrates the PSA decline and lessening of thePSA variability with increasing time.

Figures 4 and 5, contain the Kaplan–Meier FFBF func-tions, stratified by stage and Gleason sum at biopsy. Figures

6 and 7 are the same functions stratified by serum PSA atpresentation.

Figure 8 contains the rates of success, based on FFBF forthe patients in this study, by the lowest serum PSA valuereached. No clear demarcation was evident for nadir valuesbetween 0.1 ng/mL and 1.0 ng/mL, with regard to PSAprogression-free survival. Figure 9 contains the Kaplan–Meier FFBF functions for the current data set, and for asecond consecutive cohort of 97 patients treated before thisseries, between January 1986 and December 1987, using thesame failure criteria employed for this series. A significantdifference in FFBF was noted in favor of the current series,compared to the 1986–987 series,p 5 0.0002 (Fig. 9)Summary data for the two cohorts are compared in Table 6,and demonstrate no significant difference in risk factorsbetween the two groups.

DISCUSSION

We report here on the PSA-based outcomes of a consec-utive cohort of prostate cancer patients treated via contem-porary, ultrasound-guided125I brachytherapy. During the

Fig. 2. Boxplot representation of prostate-specific antigen data by interval posttreatment.


interval in which these patients where treated, 1988–1990,125I brachytherapy (as monotherapy) was considered appro-priate treatment only for those individuals considered tohave organ-confined disease. For this reason, most of thepatients comprising the basis for this study occupy a cohortthat would be considered a “low” stratum of risk for cancerbeyond the confines of the prostate (4) and local treatmentvolume.

We present the follow-up serum PSA data accumulatedon these patients in two ways. First, we present the Kaplan–Meier freedom from biochemical progression function forthe entire cohort. Biochemical progression for the purposesof this report has been defined as two consecutive rises inserum PSA. In doing so, we deviate from the ASTROconsensus definition for PSA-based failure following radio-therapy (3), which requires three consecutive increases,separated by a minimum of 3 months.

The importance of the ASTRO definition cannot be under-stated when evaluating cohort response with early- to interme-diate-range follow-up following brachytherapy, especiallyconsidering the existence of temporary, benign elevations—or“bounces”—in serum PSA that have been described follow-ing low-dose-rate brachytherapy for carcinoma of the pros-tate (5, 6). However, observations of “bouncing” are typi-cally restricted to 18–48 months postimplant. Given thatthe median PSA follow-up time of this group stands at 82months, the likelihood of PSA bouncing at last follow-up issmall. Thus, a two-rise definition confers sensitivity to theoutcome measure without sacrificing specificity.

The Kaplan–Meier estimate of biochemical relapse-freesurvival (BRFS) in this entire cohort is 85.1% (95% confi-dence interval, 79.3–90.9%) at 10 years (Fig. 1). During theperiod of treatment, the importance of serum PSA at diag-nosis had not been realized to the extent that it is currently.As a result, a small group of patients (n 5 9 [7.3%]) weretreated who presented with serum PSAs greater than 20

Fig. 3. Percentage of patients achieving designated prostate-specific antigen by interval postimplant.

Table 4. Prostate-specific antigen (PSA) data bypostimplant interval

IntervalMonths

postimplantMedian

PSA

No. of patientsw/PSA (ng/mL)

n#0.5 #1.0 #2.0

0 0 5.1 1 4 20 1241 0.1–6.0 1.4 14 32 64 982 6.1–12.0 0.8 34 62 83 993 12.1–18.0 0.6 41 65 86 954 18.1–24.0 0.55 42 64 75 845 24.1–36.0 0.3 69 83 91 976 36.1–48.0 0.2 68 80 81 867 48.1–60.0 0.1 64 69 72 768 60.1–72.0 0.1 65 65 68 709 72.1–84.0 0.1 52 55 56 59

10 84.1–96.0 0.1 37 38 40 4311 96.1–108.0 0.1 25 26 26 2712 108.1–120.0 0.1 13 13 14 1513 120.1–138.2 0.1 8 8 9 9


ng/mL. These patients have fared significantly worse thanthose with presentation PSAs less than 20 ng/mL (46% vs.87%, Fig. 7). Given that the cohort is heavily weightedtoward “low risk” patients, as the literature currently de-fines, this result (87%) may fairly be considered the 10-yearprobability of BRFS facing low-risk patients (PSA, 10,Gleason Score# 6, T1–T2b) who undergo contemporary125I brachytherapy. Serum PSA greater than 20 ng/mL wasalso significant in Cox regression on the BRFS endpoint,p 5 0.012. Otherwise, the rather uniform risk profile of thiscohort precludes meaningful multivariate risk stratification.

The fact that 48% of the patients in this series presentedwith disease interpreted as Gleason Score 2 through 4 bearsmention. A recent review of prostate carcinoma specimensat the Johns Hopkins Hospital revealed a much higher rateof Gleason 2–4 histology in the nonacademic setting (22%)than was observed within Johns Hopkins University (1%)(7). This phenomenon of “grade deflation” in the commu-nity setting has been noted by other authorities as well (8).Given that the patients in this series were diagnosed in acommunity setting a decade ago, the presence of under-grading among these biopsies seems likely. In terms of theoutcomes described here, patients with diagnostic biopsyGleason 2–4 have a nearly identical 10-year rate of BRFS,to those with original biopsy Score 5–6. These facts supportthe contention that the preponderance of patients in thisseries possessed moderately differentiated carcinomas.

For the purposes of fully describing the data that buildsthe basis for the BRFS estimates, we also present thefollow-up serum PSA data as aggregate data per interval

posttreatment. As Figs. 2 and 3 demonstrate, serum PSA forthese patients follows a course that may take several yearsto reach group stability. For example, although it takes only18 months for 80% of the cohort to reach 2.0 ng/mL, at least6 years must pass before 80% of the group reaches at least0.2 ng/mL (Fig. 2). This observation highlights the extremeimportance of significantly long follow-up when evaluatingseries of men treated for low-risk prostate cancer on thebasis of PSA “nadirs” alone.

As illustrated by Fig. 3, at between 8 and 10 yearsposttherapy, the proportions of the cohort reaching allthreshold values (2.0, 1.0, 0.5, and 0.2 ng/mL) approach arelatively stable value between 80% and 90%. Interestingly,these data approximate the 85% BRFS estimate at 10 years.In our view, the intersection of these independent methodsprovides strong evidence of the accuracy of the BRFSmethod of determining biochemical success post-brachythe-rapy, at least in scenarios with relatively long follow-up.

Figure 8 demonstrates that, the lower the posttreatmentPSA “nadir” reached in this study, the higher the probabilityof success. However, it is noted that patients with very lownadirs can fail the biochemical progression endpoint (1patient at 0.1 ng/mL; 1 patient at 0.2 ng/mL). Furthermore,the trend of the success line, with regard to nadir, illustratesthat a clear demarcation is not evident between 0.1 and 1.0ng/mL. It is, thus, difficult, based on this data, to posit anumerical PSA result between 0.1 and 1.0 ng/mL thatguarantees success to a degree significant enough to rule allthose above a particular value “failures,” as has been sug-gested by other investigators (9).

Fig. 4. Prostate-specific antigen progression-free survival by clinical stage.


Fig. 5. Prostate-specific antigen progression-free survival by Gleason Score.

Fig. 6. Prostate-specific antigen (PSA) progression-free survival by PSA (ng/mL) at presentation.


Table 5 demonstrates that selecting arbitrary cutoff pointsfor nadirs greater than a particular value will nearly alwaysresult in statistically significant differences in outcome.Certainly, patients who fail to attain very low serum PSAsmany years after brachytherapy are at a higher risk to failthan those who do. It may be important to consider, how-ever, that the word “nadir” implies “lowest point.” Theobservation that serum PSA may decrease over a period ofmany years following brachytherapy alone (11) prevents

one from necessarily knowing the “absoluteness” of any onedetermination. So, although it may be fair to conclude thatpatients whose serum PSA has decreased to the limits ofdetection possess a high probability of cure in the 10- to12-year time range, patients whose serum PSA is very lowbut not ablated (e.g., between 0.3 and 0.9 ng/mL) may nothave reached their ultimate nadir. In fact, of 47 individuals(38%) whose serum PSA measured 0.3 ng/mL at some pointposttherapy (median, 37.5 months), 38 (81%) exhibited a

Fig. 7. Prostate-specific antigen (PSA) progression-free survival by initial PSA greater than or less than 20 ng/mL.

Fig. 8. Rates of freedom from prostate-specific antigen (PSA) progression lowest serum PSA value reached.


subsequent decline, including 10 patients (21%) who exhib-ited a serum PSA of 0.3 ng/mL beyond 5 years. Thus, thesedata suggest that the practice of “failing” patients exhibitingserum PSA of 0.3 ng/mL, even 5 years posttreatment, re-sults in an unacceptably high rate of false-positive findings.

A recent report by Ragdeet al. (10) details the biochem-ical progression-free survival rate for a cohort of patientsthat overlaps the patients described in this study. In theRagdeet al.study, a 66% likelihood of 10-year disease-freesurvival was noted for patients with clinically localizedcarcinoma of the prostate treated by125I brachytherapyalone. The estimate in this current study of 85% would seemto be at odds with this former finding, even considering thatthe failure criterion in the Ragdeet al. paper was based onserum PSA. 0.5 ng/mL.

As demonstrated in Fig. 9, a strong dichotomy of out-

Fig. 9. Comparison of freedom from prostate-specific antigen (PSA) progression by interval implanted and risk stratum.

Table 5. Statistical significance at various “nadirs”

Nadir False True p exact

% success

#0.1 62.5 98.8 ,0.0005#0.2 51.7 97.9 ,0.0005#0.3 47.8 96.1 ,0.0005#0.4 36.8 96.2 ,0.0005#0.5 33.3 96.3 ,0.0005#0.8 33.3 91.4 ,0.0005#1.0 14.3 91.5 ,0.0005

Table 6. Comparison of125I monotherapy cohorts by timeinterval treated

Interval treated

p1/1/86–12/31/87 1/1/88–12/31/90

Low riskn 75 97BRFS 65 87 0.008Median F/U (months) 85.2 82.2 0.1Average PSA 3.9 4.5 0.2PSA , 4 65 56 0.34 , PSA # 10 35 44 0.3GS 2–4 59 47 0.2GS 5–6 41 53 0.2cStg. T1c 71 78 0.3cStg. T2a 19 12 0.3

Intermediate riskn 22 27BPFS 37 79 0.002Median F/U (months) 80.6 80.4 0.8Average PSA 21.8 20.8 0.810 , PSA # 20 67 70 0.9PSA . 20 33 30 0.9GS 2–4 62 52 0.6GS 5–6 38 48 0.6CStage. T1c 91 67 0.08CStage. T2a 46 19 0.06

Abbreviations: BRFS 5 biochemical relapse-free survival;F/U 5 follow-up; PSA5 prostate-specific antigen; GS5 Gleasonscore; CStage5 clinical stage.


comes exists between those patients treated before 1988 andthose treated afterward, as part of the “Seattle” prostatebrachytherapy experience. All of the patients in the Ragdeetal. study and this study were treated by at least two of theauthors (J.B., P.G.). It is important to note that the Ragdeetal. series included patients treated before 1988. Based onthe characteristics of the two groups (Table 6), selectionfactors do not appear to be a likely explanation. Whetherthis difference in outcomes can be attributed to an improve-ment in apparatus or other technology during this timeperiod, or to a “learning curve” for a new technique, isunknown. One might surmise that the practice of obtainingpostimplant dosimetry—based on postprocedure computedtomography studies commenced in early 1988—may haveoffered feedback that resulted in “better,” or more uniform,implants. It is our opinion that the results of this currentstudy of patients treated between 1988 and 1990 most

accurately depict the results of the contemporary techniqueof ultrasound-guided, preplanned,125I brachytherapy of theprostate, performed by a suitably experienced team ofbrachytherapists.

The experience of these patients provides a plausiblebasis for concluding that 87% of men will be free frombiochemical (PSA) failure, however defined, at 10 yearsfollowing contemporary125I brachytherapy for low-risk car-cinoma of the prostate. These results with125I closelymirror the results we have previously reported using103Pdbrachytherapy (11). It is anticipated that, with the technicalimprovements that have occurred over the past 10 years,these results will continue to improve. As such, a parity ofoutcomes is supported by this study for the low-risk patientacross all currently available treatment modalities (4, 12,13). The challenge for the near future is the identification ofimplant strategies that optimally maintain quality of life.

REFERENCES

1. Grimm PD, Blasko JC, Ragde H. Ultrasound-guided transperi-neal implantation of Iodine-125 and Palladium-103 for thetreatment of early-stage prostate cancer: Technical concepts inplanning, operative technique, and evaluation. In: Schellham-mer PF, ed. New techniques in prostate surgery. Philadelphia:WB Saunders Co; 1994. p. 113–126.

2. American Joint Committe on Prostate Cancer. In: Beahrs OH,Henson DE, Hutter RVP, Kennedy BJ, eds. Manual for stag-ing of cancer. 4th ed. Philadelphia: JB Lippincott; 1992. p.181–183.

3. Consensus statement. Guidelines for PSA following radiationtherapy.Int J Radiat Oncol Biol Phys1997;37:1035–1041.

4. Zelefsky MJ, Leibel SA, Gaudin PB,et al. Dose escalationwith three-dimensional conformal radiation therapy affects theoutcome in prostate cancer.Int J Radiat Oncol Biol Phys1998;41:491–500.

5. Cavanagh W, Blasko JC, Grimm PD,et al.Transient elevationof serum prostate-specific antigen following 125I/103Pdbrachytherapy for localized prostate cancer.Semin Urol On-col. 2000;18:160–165.

6. Critz FA, Williams WH, Benton JB,et al. Prostate specificantigen bounce after radioactive seed implantation followedby external beam radiation for prostate cancer.J Urol. 2000;163:1085–1089.

7. Steinberg DM, Sauvageot J, Piantadosi S,et al. Correlation ofprostate needle biopsy and radical prostatectomy Gleason

grade in academic and community settings.Am J Surg Pathol1997;21:566–576.

8. Iczkowski KA, Bostwick DG. The pathologist as optimist:cancer grade deflation in prostatic needle biopsies (Editorial).Am J Surg Pathol1998;22:1169–1170.

9. Critz FA, Williams WH, Holladay CT,et al. Post-treatmentPSA , or 5 0.2 ng/mL defines disease freedom after radio-therapy for prostate cancer using modern techniques.Urology1999;54:968–971.

10. Ragde H, Korb LJ, Elgamal AA,et al. Modern prostatebrachytherapy. Prostate specific antigen results in 219 patientswith up to 12 years of observed follow-up.Cancer. 2000;89:135–141.

11. Blasko JC, Grimm PD, Sylvester JE,et al. Palladium-103brachytherapy for prostate carcinoma.Int J Radiat Oncol BiolPhys. 2000;46:839–850.

12. Zelefsky MJ, Wallner KE, Ling CC,et al. Comparison of the5-year outcome and morbidity of three-dimensional conformalradiotherapy versus transperineal permanent iodine-125 im-plantation for early-stage prostatic cancer.J Clin Oncol1999;17:517–522.

13. D’Amico AV, Whittington R, Malkowicz SB,et al. Clinicalutility of the percentage of positive prostate biopsies in defin-ing biochemical outcome after radical prostatectomy for pa-tients with clinically localized prostate cancer (See com-ments).J Clin Oncol. 2000;18:1164–1172.


prostate cancer with i brachytherapy peter d. grimm … · which these patients were treated, only...

Documents