Tumor Length in Prostate Cancer - Robin T. Vollmer, MD
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Anatomic Pathology / Tumor Length in Prostate Cancer
Tumor Length in Prostate Cancer
Robin T. Vollmer, MD
Key Words: Prostate cancer; Tumor length; Survival; Grade; Needle biopsies; Serum PSA; Prostate-specific antigen; Tumor size; Age
DOI: 10.1309/PJNRHT63TP6FVC8B
Abstract Since the 1990s, many have observed that the quantity
To evaluate the impact of tumor length and fraction of tumor in needle biopsy specimens of the prostate is impor-
of positive biopsy cores on overall survival, I used the tant for prognosis. For example, the amount of tumor in the
data for 526 patients with prostate cancer. Median biopsy specimens has been positively correlated with tumor
follow-up in patients not observed until death was more volume,1-21 with tumor stage (ie, pT stage),8,15,22-40 with nodal
than 6 years. In a Cox model analysis that included metastases,38,41,42 and with biochemical tumor failure mani-
age, serum prostate-specific antigen (PSA) level, grade, fest by a rising serum prostate-specific antigen (PSA) level
and fraction of positive cores, tumor length was the after treatment.22,24,29,32,33,36-38,40,43-49 These 4 outcomes can
most closely associated with overall survival time (P be seen as the first in a sequence of 6 outcomes in prostate
= 6 × 10–5); however, the impact of tumor length was cancer that include metastases, hormone-refractory tumor,
mostly for a subset of men with tumors measuring more and the event of death (overall survival).
than 20 mm. Patient age, serum PSA level, Gleason Although biochemical tumor failure is the outcome most
score, fraction of cores with tumor, and tumor length commonly studied, patients with failure can still be treated
were all significantly codependent variables. For with radiotherapy if their initial treatment was surgery, and
routine cases of prostate cancer, measuring tumor they can also be treated hormonally. Furthermore, after such
length in the needle cores may be unnecessary. Tumor secondary treatments, many of the patients with biochemical
length may assist studies of long-term outcomes or failure do well and die of other causes. The central question
treatment trials in prostate cancer by reducing baseline for many men at the time of diagnosis is whether the tumor
variance better than other prognostic variables. For the will shorten their life, and its answer will undoubtedly affect
few patients with unusually large amounts of tumor in their decisions about the choice and timing of treatment.
biopsy specimens, tumor length may provide a concise Thus, the outcome that may be of greatest concern to
indicator for the likelihood of an adverse outcome, these men is overall survival. Any observations made at the
especially when the values of other prognostic variables time of diagnosis and that can relate to the time of death
appear by themselves to be less ominous. should be important, and they are likely also to be useful for
deciding treatment. Nevertheless, I know of just 1 group that
studied the relationship between the quantity of tumor in the
initial needle biopsy specimens and overall survival, and they
had just 7 patients who were followed up until death.38
Finally, the optimal way to measure the amount of tumor
in needle biopsy specimens remains unclear. The quantity
of tumor in needle cores has been expressed as number of
positive cores, fraction or percentage of positive cores, overall
© American Society for Clinical Pathology Am J Clin Pathol 2008;130:77-82 77
77 DOI: 10.1309/PJNRHT63TP6FVC8B 77Vollmer / Tumor Length in Prostate Cancer
estimate of the percentage of tissue with tumor, total tumor zTable 1z
length in millimeters, greatest tumor length in 1 core, great- Characteristics of Study Population
est percentage of tumor in 1 core, and greatest percentage of Characteristic Result
high-grade tumor.35
No. of patients 526
To study some of these issues, I began, in 1997, to Mean age (range), y 64 (42-88)
prospectively record 2 measures of the quantity of tumor in Primary treatment
Surgery 314
needle biopsy specimens of the prostate: fraction of cores Radiation 159
with tumor and total tumor length in millimeters. In addition, Other 53
I retrospectively measured these 2 variables in 55 cases in Mean serum PSA level (range), ng/mL* 12.3 (0.0-99.6)
Mean total No. of biopsy cores (range) 7.3 (1-16)
which the patients underwent biopsy earlier and whose biopsy Mean biopsy Gleason score (range) 6.7 (4-10)
specimens were reviewed for other purposes. Herein, I report No. of patients observed until death 108
Mean follow-up of patients alive at last 6.1 (0.5-15.2)
how the fraction of cores with tumor and total tumor length follow-up (range), y
relate to overall survival in 526 men diagnosed with prostate Mean tumor length (range), mm 13.4 (0.02-135.8)
cancer and followed up until their death or for a median of * Prostate-specific antigen (PSA) levels are given in conventional units; to convert to
more than 6 years. Système International units (µg/L), multiply by 1.0.
Materials and Methods Statistical Methods
To relate clinical and histologic variables to overall sur-
Study Patients vival time, I used Kaplan-Meier plots, the Cox proportional
The study patients comprised 526 men who were diag- hazards model, and the Weibull survival model.50 To relate
nosed with prostate cancer. All but 55 were diagnosed during continuous variables to the categories of Gleason score, I
the period from 1997 to 2006, and all of the diagnoses were used the nonparametric Kruskal-Wallis test. All analyses were
made on needle biopsy specimens. In addition, I recorded the done using S-PLUS software (MathSoft, Seattle, WA), and all
Gleason score, the number of cores with tumor, and the linear P values were 2-sided.
measurement of total tumor length in millimeters along the
axis of the biopsy. For tumor length, I used the same cali-
brated eyepiece micrometer I use to measure tumor thickness
Results
in cutaneous melanomas, and no other lengths were recorded.
In other words, throughout this article, tumor length refers to
the total length of the tumor. Survival and Quantity of Tumor in the Biopsy Specimens
Of the men whose data were included in the study, 444 zFigure 1z shows the distribution of tumor lengths in the
were patients in the Durham Veterans Affairs Medical Center study population. zFigure 2z provides a Kaplan-Meier plot
(VAMC), Durham, NC, and the remaining 82 were partici- of 5-year survival probability vs the fraction of biopsy cores
pants in a concurrent Cancer and Leukemia Group B study
for which I was a pathology reviewer. Of the 526 patients,
314 were treated with radical prostatectomy, 159 were treated
primarily by external beam radiation therapy, and the remain- 300
ing 53 were treated as needed by other methods, including
hormonal therapy. In general, information about clinical 250
stage was not uniformly documented in their medical records;
No. of Patients
200
however, I assumed that all patients treated by surgery or
radiation therapy had clinically localized tumor, because the 150
standard of practice routinely excluded these treatments for 100
patients with evidence of metastatic tumor. Furthermore, I
excluded men whose initial values of PSA exceeded 100 ng/ 50
mL (100 µg/L). I obtained survival data from hospital records 0
or public records of deaths (eg, see www.peoplefinders.com),
0 20 40 60 80 100 120 140
and the study was approved by the local VAMC Institutional
Tumor Length (mm)
Review Board. Other details about these study patients and
their tumors are given in zTable 1z. zFigure 1z Tumor length.
78 Am J Clin Pathol 2008;130:77-82 © American Society for Clinical Pathology
78 DOI: 10.1309/PJNRHT63TP6FVC8BAnatomic Pathology / Original Article
1.0 1.0
Probability of 5-year Survival
Probability of 5-year Survival
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
0.0 0.2 0.4 0.6 0.8 1.0 0 20 40 60 80 100 120 140
Fractions of Cores With Tumor Tumor Length (mm)
zFigure 2z Kaplan-Meier plot of the probability of surviving 5 zFigure 3z Kaplan-Meier plot of the probability of surviving 5
years vs fraction of cores with tumor. The analysis was done years vs tumor length. The analysis was done on data for the
on data for the subset of 308 patients with at least 5 years of subset of 308 patients with at least 5 years of follow-up or
follow-up or who died during this interval. The faint upper and who died during this interval. The faint upper and lower lines
lower lines give 95% confidence intervals. give 95% confidence intervals.
with tumor, and the plot shows that as the fraction increased, survival was greater than the associations between age or PSA
the survival probability decreased. Cox model analysis also level and survival. After controlling for the effect of tumor
demonstrated that this association was significant (P = .0004). length, the fraction of biopsy cores with tumor contributed no
zFigure 3z provides a similar Kaplan-Meier plot of 5-year sur- further prognostic information, nor did Gleason score, tumor
vival probability, this time, however, vs tumor length in mil- length divided by the number of positive cores, or tumor
limeters, and once again, the plot shows that as tumor length length divided by the total number of cores obtained.
increased the survival probability decreased. Cox model Finally, I used a Weibull survival model to estimate the
analysis also demonstrated that this association was also sig- impact of tumor length on survival probability 5 years after
nificant and with a smaller P value of 1.6 × 10–8. zTable 2z diagnosis. Using just the 2 significant variables of patient age
shows a multivariable Cox model analysis that tests for the and tumor length in Table 2, the Weibull model fit the data
importance of the quantity of tumor after controlling for the well (result not shown), and plots of the estimated 5-year
effects of patient age and serum PSA level. Gleason score was survival probabilities for tumor lengths ranging from 0.5 to
also examined. The results demonstrate that tumor length was 130 mm for the ages of 50, 60, and 70 years appear as shown
significantly associated with overall survival (P = 6 × 10–5), in zFigure 4z. The results demonstrate that tumor length has a
and the strength of association between tumor length and modest impact on overall survival until tumor lengths exceed
20 mm (2 cm).
Codependence Between Tumor Length and Other
zTable 2z
Cox Model Analysis* Variables
In this study, I investigated 5 prognostic variables, all
Variable Coefficient P
of which can be known at the time of diagnosis. These 5 are
Patient age 0.0411 .003 patient age, serum PSA level, Gleason score, fraction of cores
Serum PSA level 0.194 .08 with tumor, and tumor length, but I found that they provide at
Tumor length 0.0184 6 × 10–5
Fraction of positive cores — >.9 best overlapping information because they are not statistically
Gleason score — >.3 independent of one another. For example, many have demon-
Tumor length/positive cores — >.6
Tumor length/total cores — >.6 strated that the serum PSA level is related to patient age (eg,
see Vollmer51 and Grunkemeier and Vollmer52).
PSA, prostate-specific antigen.
* Coefficients are given for only the significant variables in a 3-variable model By using the data, I found that serum PSA level was also
analysis. Serum PSA was used with a natural logarithm transformation, and it, positively and significantly related to Gleason score (P = .001;
age, total tumor length, and fraction of positive cores were coded as continuous
variables. Gleason score was coded asVollmer / Tumor Length in Prostate Cancer
1.0 suggest that among the pretreatment variables of patient age,
serum PSA level, Gleason score, and 2 measurements of
Probability of 5-year Survival
0.8 tumor quantity, the total tumor length provides the most infor-
mation about overall survival. Of course, these 5 variables
0.6 Ag
e5 provide overlapping prognostic information because they are
0
Ag
not statistically independent. The question to consider then is
0.4 e6
0 whether measuring tumor length is worth the additional effort
Ag
e7
a pathologist must make to record this variable, especially
0.2 0
when so few routinely measure tumor length and when few or
no urologists use this measurement to decide treatment.
0.0
Despite the need for an eyepiece micrometer and the
0 20 40 60 80 100 120 1 or 2 minutes required to measure tumor length, this mea-
Tumor Length (mm) surement has advantages over several other estimates of the
amount of tumor present, such as overall percentage of tumor
zFigure 4z Estimated probabilities of 5-year survival vs tumor
and the fraction of cores with tumor. For example, it is more
length for 3 men at the respective ages of 50, 60, and 70
objective. It requires no subjective estimate of the number of
years. The estimates are from a Weibull model that used just
cores, when they frequently fracture into smaller pieces, nor
age and tumor length as explanatory variables.
does it require a subjective estimate of the percentage of total
biopsy tissue that contains tumor. All one need do is to mea-
sure the length of each focus of diagnostic tumor in the units
(9.2 µg/L). I found that the fraction of positive cores was also of the micrometer and along an axis parallel to the needle
positively and significantly related to Gleason score (P ~ 0; biopsy specimen, sum these units, and convert the sum to
Kruskal-Wallis test). Whereas the median fraction of positive millimeters. If the eyepiece micrometer is already calibrated
cores for Gleason scores 4 through 6 was 0.29, the median to measure tumor thickness in cutaneous melanomas, then the
fraction for Gleason scores 7 through 10 was 0.5. Similarly, I same algorithm can be used to measure total tumor length in
found that tumor length was positively and significantly relat- millimeters for the prostate. Using an electronic calculator
ed to Gleason score (P ~ 0; Kruskal-Wallis test). Whereas the simplifies the task.
median tumor length for Gleason scores 4 through 6 was 2.9 Weakening the argument for measuring tumor length is
mm, the median length for Gleason scores 7 through 10 was the observation that this variable’s importance is largely lim-
13.1 mm. In addition, linear regression analyses demonstrated ited to a modest number of tumors. For example, more than
that fraction of positive cores and tumor length were signifi- 75% of this study’s patients had tumor lengths of less than 18
cantly and positively related to serum PSA level (P ~ 0) and mm. Examination of Figures 3 and 4 shows that in the range
to one another (P ~ 0). Finally, I found that fraction of cores of 0 to 20 mm, the expected 5-year survival should be high
with tumor and tumor length were significantly and positively despite the exact tumor length. In other words, tumor length
related to patient age (P ~ 0; linear regression). Thus, all 5 of does not impact survival over typical values obtained for
these variables are codependent. Although the number of core most men with prostate cancer. By contrast, tumor lengths of
biopsies varied among these patients, there was no association more than 20 mm affect survival more dramatically, but such
between tumor length and the number of cores (P > .2; linear tumors are not common, and they tend to have higher Gleason
regression). scores and larger fractions of positive cores.
This study demonstrates that total tumor length in biopsy
specimens of the prostate is an important prognostic variable
that is more closely related to overall survival than patient
Discussion
age, serum PSA level, Gleason score, and fraction of cores
The most important information derived from a set with tumor. Its measurement requires a modest extra effort
of prostate biopsy specimens is the diagnosis, that is, the for pathologists—an effort similar to measuring the Breslow
presence or absence of tumor. (For the moment, patterns thickness in cutaneous melanoma. At the same time, this vari-
of nondiagnostic atypia will not be addressed.) If tumor is able’s prognostic impact is limited for most men with prostate
present, then other important clinical observations made cancer and comes into play mostly for a subset whose tumor
by the time of biopsy include the patient’s age, presence of lengths exceed 20 mm. Altogether, the results suggest that
comorbidities, serum PSA level, and clinical stage. To these investigators obtaining data for large prospective studies of
variables, pathologists routinely add Gleason score and often outcomes in prostate cancer or for clinical trials should con-
an estimate of the quantity of tumor present. Now these results sider measuring and recording tumor lengths on their patients
80 Am J Clin Pathol 2008;130:77-82 © American Society for Clinical Pathology
80 DOI: 10.1309/PJNRHT63TP6FVC8BAnatomic Pathology / Original Article
because this measurement is now largely ignored and because 14. Egevad L, Norberg M, Mattson S, et al. Estimation of prostate
it may reduce baseline variance in such studies. For others, cancer volume by multiple core biopsies before radical
prostatectomy. Urology. 1998;52:653-658.
the routine measurement of tumor length should be optional,
15. Sebo TJ, Bock BJ, Cheville JC, et al. The percent of cores
unless one wants to call attention to particular cases with high positive for cancer in prostate needle biopsy specimens is
tumor lengths. In such cases, tumor length may be more likely strongly predictive of tumor stage and volume at radical
to reflect shortened overall survival than other measures. prostatectomy. J Urol. 2000;163:174-178.
16. Noguchi M, Stamey TA, McNeal JE, et al. Relationship
From Laboratory Medicine, Veterans Affairs Medical Center and between systematic biopsies and histologic features of 222
Duke University Medical Center, Durham, NC. radical prostatectomy specimens: lack of prediction of tumor
significance for men with nonpalpable prostate cancer. J Urol.
Address reprint requests to Dr Vollmer: Laboratory Medicine 2001;166:104-110.
113, VA Medical Center, 508 Fulton St, Durham, NC 27705. 17. Sebo TJ, Cheville JC, Riehle DL, et al. Predicting prostate
carcinoma volume and stage at radical prostatectomy by
assessing needle biopsy specimens for percent surface area and
References cores positive for carcinoma, perineural invasion, Gleason
score, DNA ploidy and proliferation, and preoperative
1. Terris MK, Haney DJ, Johnstone IM, et al. Prediction of serum prostate specific antigen: a report of 454 cases. Cancer.
prostate cancer volume using prostate-specific antigen levels, 2001;91:2196-2204.
transrectal ultrasound, and systematic sextant biopsies. Urology.
18. Lewis JS, Vollmer RT, Humphrey PA. Carcinoma extent in
1995;45:75-80.
prostate needle biopsy tissue in the prediction of whole gland
2. Thomassen L, Tsuchiya K, True LD. Accuracy of needle tumor volume in a screening population. Am J Clin Pathol.
biopsies in predicting tumor distribution in radical 2002;118:442-450.
prostatectomies [abstract]. Lab Invest. 1993;68:70a.
19. Grossklaus DJ, Coffey CS, Shappell SB, et al. Percent of
3. Irwin MB, Trapasso JG. Identification of insignificant cancer in the biopsy set predicts pathological findings after
prostate cancers: analysis of preoperative parameters. Urology. prostatectomy. J Urol. 2002;167:2032-2036.
1994;44:862-867.
20. Kattan MW, Eastham JA, Wheeler TM, et al. Counseling men
4. Haggman M, Nybacka O, Nordin B, et al. Standardized in vitro with prostate cancer: a nomogram for predicting the presence
mapping with multiple core biopsies of total prostatectomy of small, moderately differentiated, confined tumors. J Urol.
specimens: localization and prediction of tumour volume and 2003;170:1792-1797.
grade. Br J Urol. 1994;74:617-625.
21. Ochiai A, Troncoso P, Chen ME, et al. The relationship
5. Dietrick DD, McNeal JE, Stamey TA. Core cancer length
between tumor volume and the number of positive cores in
in ultrasound guided systematic sextant biopsies: a
men undergoing multisite extended biopsy: implication for
preoperative evaluation of prostate cancer volume. Urology.
1995;45:987-992. expectant management. J Urol. 2005;174:2164-2168.
6. Cupp MR, Bostwick DG, Myers RP, et al. The volume of 22. Ravery V, Boccon-Gibod LA, Dauge-Geffroy MC, et al.
prostate cancer in the biopsy specimen cannot reliably predict Systematic biopsies accurately predict extracapsular extension
the quantity of cancer in the radical prostatectomy specimen of prostate cancer and persistent/recurrent detectable PSA
on an individual basis. J Urol. 1995;153:1543-1548. after radical prostatectomy. Urology. 1994;44:371-376.
7. Humphrey PA, Baty J, Keetch DA. Relationship between 23. Badalament RA, Miller MC, Peller PA, et al. An algorithm for
serum prostate specific antigen, needle biopsy findings, and predicting nonorgan confined prostate cancer using the results
histopathologic features of prostatic carcinoma in radical obtained from sextant core biopsies with prostate specific
prostatectomy tissues. Cancer. 1995;75(suppl):1842-1849. antigen level. J Urol. 1996;156:1375-1380.
8. Peller PA, Young PC, Marmaduke DP, et al. Sextant 24. Huland H, Hammerer P, Henke R-P, et al. Preoperative
prostate biopsies: a histopathologic correlation with radical prediction of tumor heterogeneity and recurrence after radical
prostatectomy specimens. Cancer. 1995;75:530-538. prostatectomy for localized prostatic carcinoma with digital
9. Terris MK, McNeal JE, Stamey TA. Detection of clinically rectal examination, prostate specific antigen and the results of
significant prostate cancer by transrectal ultrasound–guided 6 systematic biopsies. J Urol. 1996;155:1344-1347.
systematic biopsies. J Urol. 1992;148:829-832. 25. Egan AJM, Bostwick DG. Prediction of extraprostatic
10. Loch T, McNeal JE, Stamey TA. Interpretation of extension of prostate cancer based on needle biopsy findings:
bilateral positive biopsies in prostate cancer. J Urol. perineural invasion lacks significance on multivariate analysis.
1995;154:1078-1083. Am J Surg Pathol. 1997;21:1496-1500.
11. Egawa S, Satoh T, Suyama K, et al. Significance of preoperative 26. Wills ML, Sauvageot J, Partin AW, et al. Ability of sextant
parameters for predicting tumor volume in nonpalpable biopsies to predict radical prostatectomy stage. Urology.
prostate cancer. Jpn J Clin Oncol. 1996;26:356-361. 1998;51:759-764.
12. Goto Y, Ohori M, Arakawa A, et al. Distinguishing 27. Egawa S, Suyama K, Matsumoto K, et al. Improved
clinically important from unimportant prostate cancers predictability of extracapsular extension and seminal vesicle
before treatment: value of systematic biopsies. J Urol. involvement based on clinical and biopsy findings in prostate
1996;156:1059-1063. cancer in Japanese men. Urology. 1998;52:433-440.
13. Arcangeli CG, Humphrey PA, Smith S, et al. Percentage of 28. Vargas SO, Jiroutek M, Welch WR, et al. Perineural
free serum prostate-specific antigen as a predictor of pathologic invasion in prostate needle biopsy specimens: correlations
features of prostate cancer in a screening population. Urology. with extraprostatic extension at resection. Am J Clin Pathol.
1998;51:558-565. 1999;111:223-228.
© American Society for Clinical Pathology Am J Clin Pathol 2008;130:77-82 81
81 DOI: 10.1309/PJNRHT63TP6FVC8B 81Vollmer / Tumor Length in Prostate Cancer
29. D’Amico AV, Whittington R, Malkowicz SB, et al. Clinical 40. Briganti A, Chun F K-H, Hutterer GC, et al. Systematic
utility of the percentage of positive prostate biopsies in assessment of the ability of the number and percentage
defining biochemical outcome after radical prostatectomy for of positive biopsy cores to predict pathologic stage and
patients with clinically localized prostate cancer. J Clin Oncol. biochemical recurrence after radical prostatectomy. Eur Urol.
2000;18:1164-1172. 2007;52:733-743.
30. Rubin MA, Bassily N, Sanda M, et al. Relationship and 41. Ackerman DA, Barry JM, Wicklund RA, et al. Analysis of risk
significance of greatest percentage of tumor and perineural factors associated with prostate cancer extension to the surgical
invasion on needle biopsy in prostatic adenocarcinoma. margin and pelvic node metastasis at radical prostatectomy.
Am J Surg Pathol. 2000;24:183-189. J Urol. 1993;150:1845-1850.
31. Grossfield GD, Chang JJ, Broering JM, et al. Under staging 42. Conrad S, Graefen M, Pichlmeier U, at al. Systematic sextant
and under grading in a contemporary series of patients biopsies improve preoperative prediction of pelvic lymph
undergoing radical prostatectomy: results from the Cancer of node metastases in patients with clinically localized prostatic
the Prostate Strategic Urologic Research Endeavor database. carcinoma. J Urol. 1998;159:2023-2029.
J Urol. 2001;165:851-856. 43. Ravery V, Schmid H-P, Toublanc M, et al. Is the percentage
32. Linson PW, Lee AK, Doytchinova T, et al. Percentage of of cancer in biopsy cores predictive of extracapsular disease
core lengths involved with prostate cancer: does it add to in T1-T2 prostate carcinoma? Cancer. 1996;78:1079-1084.
the percentage of positive prostate biopsies in predicting 44. Presti JC Jr, Shinohara K, Bacchetti P, et al. Positive fraction
postoperative prostate-specific antigen outcome for men with of systematic biopsies predicts risk of relapse after radical
intermediate-risk prostate cancer? Urology. 2002;59:704-708. prostatectomy. Urology. 1998;52:1079-1084.
33. Freedland SJ, Csathy GS, Dorey F, et al. Percent prostate 45. D’Amico AV, Whittington R, Malkowicz SB, et al. Utilizing
needle biopsy tissue with cancer is more predictive of predictions of early prostate-specific antigen failure to optimize
biochemical failure or adverse pathology after radical patient selection for adjuvant systemic therapy trials. J Clin
prostatectomy than prostate specific antigen or Gleason score. Oncol. 2000;18:3240-3246.
J Urol. 2002;167:516-520.
46. D’Amico AV, Whittington R, Malkowicz SB, et al.
34. Freedland SJ, Csathy GS, Dorey F, et al. Clinical utility of Combination of the preoperative PSA level, biopsy Gleason
percent prostate needle biopsy tissue with cancer cutpoints score, percentage of positive biopsies, and MRI T-stage to
to risk stratify patients before radical prostatectomy. Urology. predict early PSA failure in men with clinically localized
2002;60:84-88. prostate cancer. Urology. 2000;55:572-577.
35. Bismar TA, Lewis JS Jr, Vollmer RT, et al. Multiple measures 47. Graefen M, Ohori M, Karakiewicz PI, et al. Assessment of the
of carcinoma extent versus perineural invasion in the prostate enhancement in predictive accuracy provided by systematic
needle biopsy tissue in prediction of pathologic stage in a biopsy in predicting outcome for clinically localized prostate
screening population. Am J Surg Pathol. 2003;27:432-440. cancer. J Urol. 2004;171:200-203.
36. Freedland SJ, Aronson WJ, Csathy GS, et al. Comparison of 48. Greene KL, Elkin EP, Karapetian A, et al. Prostate biopsy
percentage of total prostate needle biopsy tissue with cancer to tumor extent but not location predicts recurrence after
percentage of cores with cancer for predicting PSA recurrence radical prostatectomy: results from CAPSURE. J Urol.
after radical prostatectomy: results from the SEARCH 2006;175:125-129.
database. Urology. 2003;61:742-747.
49. Stephenson AJ, Scardino PT, Eastham JA, et al. Preoperative
37. Quinn DI, Henshall SM, Brenner PC, et al. Prognostic nomogram predicting the 10-year probability of prostate cancer
significance of preoperative factors in localized prostate recurrence after radical prostatectomy. J Natl Cancer Inst.
carcinoma treated with radical prostatectomy: importance of 2006;98:715-717.
percentage of biopsies that contain tumor and the presence
of biopsy perineural invasion. Cancer. 2003;97:1884-1893. 50. Hosmer DW, Lemeshow S. Applied Survival Analysis: Regression
Modeling of Time to Event Data. New York, NY: Wiley; 1999.
38. Lotan Y, Shariat SF, Khoddami SM, et al. The percent of
biopsy cores positive for cancer is a predictor of advanced 51. Vollmer RT. Predictive probability of serum prostate-specific
pathological stage and poor clinical outcomes in patients antigen for prostate cancer: an approach using Bayes rule.
treated with radical prostatectomy. Urology. 2004;171:2209- Am J Clin Pathol. 2006;125:336-342.
2214. 52. Grunkemeier NM, Vollmer RT. Predicting prostate biopsy
39. Guzzo TJ, Vira M, Wang Y, et al. Preoperative parameters, results: the importance of PSA, age, and race. Am J Clin Pathol.
including percent positive biopsy, in predicting seminal 2006;126:110-112.
vesicle involvement in patients with prostate cancer. J Urol.
2006;175:518-522.
82 Am J Clin Pathol 2008;130:77-82 © American Society for Clinical Pathology
82 DOI: 10.1309/PJNRHT63TP6FVC8BYou can also read
