This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Veltri, R. W. Articles by Partin, A. W. Search for Related Content PubMed PubMed Citation Articles by Veltri, R. W. Articles by Partin, A. W.

Ability to Predict Metastasis Based On Pathology Findings and Alterations in Nuclear Structure Of Normal-Appearing and Cancer Peripheral Zone Epithelium in the Prostate

¹ The James Buchanan Brady Urological Institute, The Johns Hopkins University School of Medicine, Baltimore, Maryland, and ² Quakertown, Pennsylvania

	ABSTRACT

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Purpose: Malignant transformation in the prostate produces significant alterations in glandular architecture (Gleason grade) and nuclear structure that provide valuable prognostic information. Normal-appearing nuclei (NN) adjacent to cancer may also have altered functions in response to malignancy. We studied NN adjacent to peripheral zone (PZ) prostate cancer (PCa), as well as the PZ cancer nuclei (CaN) using quantitative image cytometry. The nuclear structure information was combined with routine pathological findings to predict metastatic PCa progression and/or death.

Experimental Design: Tissue microarrays of normal-appearing and cancer areas were prepared from 182 pathologist-selected paraffin blocks. Feulgen-stained CaN and NN were captured from the tissue microarrays using the AutoCyte Pathology Workstation. Multivariate logistic regression was used to calculate quantitative nuclear grade (QNG) solutions based on nuclear morphometric descriptors determined from NN and CaN. Multivariate logistic regression and Kaplan-Meier plots were also used to predict risk for distant metastasis and/or PCa-specific death using QNG solutions and routine pathology.

Results: The pathology model yielded an area under the receiver operator characteristic curve of 72.5%. The QNG-NN and QNG-CaN solutions yielded an area under the receiver operator characteristic curve of 81.6 and 79.9%, respectively, but used different sets of nuclear morphometric descriptors. Kaplan-Meier plots for the pathology variables, the QNG-NN and QNG-CaN solutions, were combined with pathology to defined three statistically significantly distinct risk groups for distant metastasis and/or death (P < 0.0001).

Conclusions: Alterations in cancer or normal-appearing nuclei adjacent to peripheral zone cancer areas can predict PCa progression and/or death. The QNG-NN and QNG-CA solutions could be combined with pathology variables to improve the prediction of distant metastasis.

	INTRODUCTION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Over the past two decades there have been tremendous changes in the diagnosis and management of prostate cancer. The widespread use of serum prostate-specific antigen (PSA), along with transrectal ultrasound-guided prostate biopsy, and improved acceptance of early detection has produced a marked increase in prostate cancer incidence in the United States during the 1990s (1 , 2) . Improvements in surgical technique have substantially reduced perioperative morbidity (3) . Approximately 75% of men newly diagnosed with prostate cancer today present clinically with nonpalpable disease (stage T1c) and are regarded as potentially curable with definitive intervention (surgery or radiation). However, 25% of this population will experience biochemical recurrence over a 10–15-year period after treatment (4, 5, 6) . The majority of men that have a PSA recurrence will experience distant metastases and/or die from prostate cancer at intervals ranging from 1 to 13 years after PSA recurrence (6) . Improved methods to predict these intervals are greatly needed.

Image cytometry has provided one tool to detect cancer and assess recurrence and progression at the tissue architecture and cellular level (7, 8, 9, 10, 11) . Imaging systems, both flow and static, are able to perform DNA ploidy and quantify several nuclear morphometric descriptors (NMDs), which are useful to gain supplemental diagnostic and prognostic information from clinical specimens (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19) . Several investigators have applied nuclear morphometry to assess the precursor lesions (e.g., high-grade prostatic intraepithelial neoplasia), as well as cancer-adjacent normal nuclei, to occult but histologically not evident prostate cancer (14 , 20, 21, 22, 23) .

Our group used the Cell Analysis System (CAS-200; Ref. 12 ) to produce quantitative nuclear grade (QNG) solutions from a single 5-µm histological section or a cytology preparation for prediction of clinical, diagnostic, and prognostic outcomes in both prostate and bladder cancer (15, 16, 17, 18, 19) . The current study used prostate tissue microarrays (TMAs) to derive QNG solutions from the cancer area nuclei (CaN) and from normal nuclei (NN) collected adjacent to the peripheral zone prostate cancer area. These QNG solutions were used to predict, at the time of definitive surgery, the time to distant progression for 182 men with long-term follow-up who experienced PSA recurrence after radical prostatectomy.

	MATERIALS AND METHODS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Patient Population and Creation of TMAs.
The mean age of the 182 men was 60 years (range = 42–75 years), and none had received neoadjuvant radiation or hormonal therapy. The mean follow-up time for the biochemical recurrence (PSA level > 0.2 ng/ml) and the biochemical recurrence with distant metastasis and/or death groups were 12.5 years (median = 13 years, range = 4–19 years) and 10.6 years (median = 6 years, range = 2–20 years), respectively. Of these 182 men, 88 (48%) experienced biochemical recurrence without evidence of distant metastasis (i.e., nonprogression), whereas 94 (52%) either progressed to distant metastases or died from prostate cancer (i.e., progression) after their initial PSA recurrence. All clinical, demographic, pathological, and follow-up data, as well as the pathological material, were collected under an Internal Review Board approved protocol that passes Health Insurance Portability and Accountability Act compliance. Informed consent was obtained prior to surgery and where appropriate for long-term follow-up.

Slides from all 182 cases were reviewed and mapped by a single pathologist and the normal-appearing and index tumor areas, usually defined as the largest and/or highest Gleason score were identified and marked on the slide for each case. Using these template slides marked for normal-appearing and diagnostic cancer areas, the tissue blocks were coordinately marked using the template slides, and 2-mm cores were punched from the normal-appearing and tumor areas and then transferred to recipient blocks. The TMAs were prepared for all 182 cases (both normal-appearing and cancer areas) using a Beecher MT1 manual arrayer (Beecher Instruments, Silver Spring, MD) in the Johns Hopkins University Specialized Programs of Research Excellence grant pathology core facility. Each TMA was constructed using four replicate 2-mm core tissue samples from the normal-appearing and cancer areas of each patient who had undergone radical prostatectomy for PCa. Each block had an orientation marker and contained internal controls placed in a pre-established pattern throughout each one of the blocks to assess quality of the Feulgen stain throughout the histological sections. A total of five blocks was created for the 182 cases and controls. Slides prepared from each of the TMA blocks were stained with H&E, and the normal-appearing and cancer areas for each patient’s core set were reviewed by a pathologist to ensure their diagnostic accuracy.

Preoperative Evaluation and Pathological Characteristics.
The men were staged according to the 1992 American Joint Committee on Cancer staging guidelines (24) , including digital rectal examination by a single surgeon and routine serum PSA studies (Hybritech and Tandem-R and EC Beckman Coulter, San Diego, CA and TOSOH, Tosoh Medics, San Francisco, CA). Preoperative pathological diagnosis of PCa was based on examination of prostatic tissue obtained by transrectal ultrasound or digital guided prostate biopsy or from transurethral resection of the prostate. On the basis of clinical and pathological evaluation, tumors were determined to be organ-confined, to exhibit extraprostatic extension without positive surgical margins, extraprostatic extension with positive surgical margins, or seminal vesicle involvement without nodal disease, or to have spread to pelvic lymph nodes (25 , 26) .

Patient Follow-Up and Definitions of Progression.
Postoperative follow-up was obtained through routine serum PSA assays and digital rectal examinations performed every 3 months for the first year, semiannually the second year, and annually thereafter. Isolated biochemical PSA elevation was defined as a serum PSA level 0.2 ng/ml. Radionuclide bone scans were performed at the time of biochemical recurrence and on a yearly basis thereafter unless performed earlier for symptoms suggestive of bone metastasis. A positive bone scan result or other radiographic or histological (lymph node biopsy) evidence was used for the diagnosis of distant metastasis.

QNG Determination.
The methods of QNG calculation have been reported previously (16, 17, 18) . Using 5-µm sections prepared from the TMA blocks, Feulgen staining was performed per instructions in the staining package insert (TriPath Imaging, Inc., Burlington, NC). Next, a minimum of 100 intact Feulgen-stained CaN and NN were captured from each case using all four 2-mm spots/case of the TMAs using an AutoCyte Pathology Workstation (TriPath Imaging, Inc.) and the QUIC-DNA software (16, 17, 18) . The QUIC-DNA software calculated a total of 60 NMDs, including nuclear size, shape, DNA content, and chromatin texture features (at a step size of one pixel), for each nuclei captured from both the CaN and NN cases. Next, the variance of each NMD for the CaN or NN captured from each case was calculated (16, 17, 18) .

Statistical Analysis.
All data were analyzed with Stata version 8.0 statistical analysis software (Stata Corporation, College Station, TX). A nonparametric k-sample test for the equality of medians was used to evaluate differences in the nonnormally distributed ages and preoperative serum PSA levels, whereas Wilcoxon’s rank-sum test was used to test for distribution differences in the clinical stage and the biopsy and pathological Gleason scores. Pearson’s ² test was used to test for independence of surgical margin status, capsular penetration status, organ confinement status, seminal vesicle status, and lymph node status. Multivariate logistic regression (MLR) analyses were used to construct models and to calculate areas under the receiver operator characteristic curves, sensitivities, specificities, and accuracies for the differentiation of the two groups of men (i.e., biochemical recurrence only versus biochemical recurrence with distant metastasis and/or death). Cox proportional hazard ratios and Kaplan-Meier survival plots were created to demonstrate the ability of QNG and routine pathological variables, both separately and in combination, to predict patient outcome.

	RESULTS

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The patient sample demographics and the analysis thereof are shown in Table 1 . The analysis of differences between the two status groups demonstrated that clinical stage, postoperative pathological Gleason score, seminal vesicle status, lymph node status, and organ confinement status were all univariately statistically significantly different (P < 0.05), but notably, biopsy Gleason score was not significant. Table 2 illustrates the ability of QNG-NN and QNG-CaN solutions and routine pathology (that included pathological Gleason score, lymph node status, seminal vesicle status, surgical margin status, capsular penetration status, and the organ confinement status) to predict the binary outcome using MLR analysis in the entire patient cohort. The results show that QNG-NN and QNG-CaN can each predict the outcome significantly better than routine pathology (i.e., based on improved specificity and accuracy) even when pathology was evaluated at two different stringencies. Also, there was no statistically significant difference between the areas under the receiver operator characteristic for QNG-NN and QNG-CaN. The addition of the pathology variables to the QNG biomarkers significantly improved the MLR model results (², P < 0.02).

View larger version (20K): [in this window] [in a new window] [Download PPT slide]   Fig. 1. Ability of pathological Gleason score and lymph node status to predict time to distant metastasis and/or death in 182 men with biochemical recurrence. The dashed line delineates the median time to progression for the high-risk group.

View larger version (23K): [in this window] [in a new window] [Download PPT slide]   Fig. 2. A, ability of quantitative nuclear grade-normal-appearing nuclei to predict time to distant metastasis and/or death in 182 men with biochemical recurrence after radical prostatectomy. The dashed line delineates the median time to progression for the high-risk group. B, ability of quantitative nuclear grade-cancer nuclei to predict time to distant metastasis and/or death in 182 men with biochemical recurrence. The dashed line delineates the median time to progression for the high-risk group.

View larger version (20K): [in this window] [in a new window] [Download PPT slide]   Fig. 3. Ability of pathological Gleason score, lymph node status, and quantitative nuclear grade-normal-appearing nuclei to predict time to distant metastasis and/or death in 182 men with biochemical recurrence. The dashed line delineates the median time to progression for the high-risk group.

View larger version (20K): [in this window] [in a new window] [Download PPT slide]   Fig. 4. Ability of pathological Gleason score, lymph node status, and quantitative nuclear grade-cancer nuclei to predict time to distant metastasis and/or death in 182 men with biochemical recurrence. The dashed line delineates the median time to progression for the high-risk group.

	DISCUSSION

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Bartels and Wied (7) and Bartels (8) applied quantitative histomorphometry and nuclear chromatin texture measurements in oncology to demonstrate presence of cancer in an organ system using information extracted from noncancerous sites. Palcic’s group (10) demonstrated that chromatin texture could be measured reproducibly by image analysis to assess malignancy-associated changes (19) . Mairinger et al. (23) used an automated CytoSavant imaging system, and their technique required the dissociation of cell nuclei from two 50-µm formalin-fixed, paraffin-embedded sections that had been marked to identify the cancer and adjacent normal-appearing cells. Their imaging protocol used 114 NMDs to analyze nuclei dissociated from the archival specimen blocks of 63 benign prostatic hyperplasia and 173 PCa cases. They were able to differentiate cancer and benign prostatic hyperplasia cases with a sensitivity of 90% and specificity of 97% using four nuclear texture features (mean fractal dimension, SD, density light spot, and fractal area SD).

Our research group has pursued several applications of the NMDs to predict prostate cancer stage, recurrence, and progression (15, 16, 17) . Recently, we replaced the CAS-200 instrument with the newer AutoCyte Pathology Workstation from TriPath Imaging, Inc. (17, 18, 19) . The current study investigated 182 of 304 men with biochemical recurrence previously investigated by Pound et al. (6) . We demonstrated that both the NN and CaN solutions (QNG-NN and QNG-CA) can predict the outcome; however, 58% of the selected NMDs were different (Table 3) . Notably, the QNG-CaN solution used several cancer-related nuclear features, including 3 additional size and shape factors and 8 chromatin texture factors not used in the calculation of the normal-appearing QNG-NN solution. Our results clearly demonstrate that nuclear morphometry measurements of normal-appearing nuclei near peripheral zone cancers may represent malignancy-associated changes. The observation that one can derive a QNG solution from this area (QNG-NN) that can predict distant metastases and/or death requires confirmation. Also, Cox hazard ratios for QNG-NN (2.59) and QNG-CaN (3.74) compared quite favorably with combined pathological Gleason score and lymph node status variables (2.99). Clearly, when QNG-NN or QNG-CaN was combined with pathology (Gleason score and lymph node status), the three resulting risk groups were significantly different and showed improvement over the use of each set of variables alone.

Today, new prognostic challenges exist in contemporary patients presenting with prostate cancer. In particular, pathological Gleason scores are predominantly 6 or 7, and positive lymph node status is only detected in 5% of men, reducing the prognostic value of routine pathology alone as a predictor of disease outcome (27 , 32) . In an attempt to address this dilemma, we also evaluated the ability of QNG-NN or QNG-CaN solutions, as well as all available routine pathological variables shown in Table 1 in a subset of 136 lymph node-negative patients from our 182-case TMA study. The QNG-NN or QNG-CaN solutions performed significantly better (² value was P < 0.02) than the multivariately significant routine pathology variables alone (Table 9) . These results demonstrate that the QNG biomarker in this challenging prognostic setting outperformed routine pathology variables alone under conditions, where patients do not possess overt evidence of advanced disease.

The demonstration of alterations in nuclear structure and DNA organization in normal epithelium adjacent to the cancer area may represent another measure of a field effect. The concept is not a new one and has been demonstrated by others using gene expression, as well as imaging techniques (10 , 20, 21, 22, 23 , 33) .

Although our studies support the clinical value of computer-assisted image analysis to assess nuclear structure alterations, we appreciate that there remain significant limitations associated with our study. We clearly recognize a need for significant expansion of our training set and a requirement for separate validation sets possessing similar patient demographics. Care must be taken to not overfit our MLR models, and we did perform drop-one analysis and cross-validation to assess the contribution of the NMDs retained in our models. However, ultimately a larger patient sample and a well-balanced validation set matched for distant metastasis and death statistics will be required. Other limitations include the use of archival pathological specimens from a noncontemporary cohort of patients, the need for an expert pathologist to select the most appropriate histological area of clinical importance, and finally, the requirement for an image cytometrist trained to capture the appropriate cell nuclei. As one approach to address the issue of reproducing our results, the AutoCyte cytometry hardware and software with training are commercially available from TriPath Imaging, Inc., and they can be configured to the exact specifications of the instrument and software containing all of the required equations we describe in our article. Finally, the ability of QNG to detect those patients at a high risk of developing distant metastases and/or dying from PCa, after radical prostatectomy, may allow better targeting of those men who would benefit from early available adjuvant therapy and possibly for election of participation in chemoprevention trials.

	FOOTNOTES

 
Grant support: NIH/National Cancer Institute Specialized Programs of Research Excellence Grant P50CA58236 and a generous gift from the Urological Sciences Research Foundation (Culver City, CA). The tissue microarrays were produced by Helen Fedor through the Pathology Core (Angelo De Marzo, Principal Investigator) of the Johns Hopkins University Prostate Cancer Specialized Programs of Research Excellence Grant.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Any repeat analysis of TMA cores was performed by Cameron Marlow van Rootselaar.

Requests for reprints: Robert W. Veltri, Visiting Associate Professor, James Buchanan Brady Urological Institute, The Johns Hopkins University School of Medicine, Baltimore, MD 21287. Phone: (410) 614-6380; Fax: (410) 614-3695; E-mail: rveltri1{at}jhmi.edu

Received 11/24/03; revised 2/ 9/04; accepted 2/20/04.

	REFERENCES

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Carter HB, Partin AW. Diagnosis and staging of prostate cancer Walsh PC Retik AB Vaughan ED Wein AJ eds. . Campbell’s Urology, p. 3055-79, Elsevier Science New York 2002.
2. Polascik TJ, Oesterling JE, Partin AW. Prostate specific antigen: a decade of discovery—what we have learned and where we are going?. J Urol, 162: 293-306, 1999.[CrossRef][Medline]
3. Walsh PC. Anatomic radical prostatectomy: Evolution of the surgical technique. J Urol, 160: 2418-24, 1998.[CrossRef][Medline]
4. Stamey TA, Donaldson AN, Yemoto CE, McNeal JE, Sozen S, Gill H. Histological and clinical findings in 896 consecutive prostates treated only with radical retropubic prostatectomy: epidemiologic significance of annual changes. J Urol, 160: 2412-7, 1998.[CrossRef][Medline]
5. Freedland SJ, Presti JC, Jr., Amling CL, et al Time trends in biochemical recurrence after radical prostatectomy: results of the SEARCH database. Urology, 61: 736-41, 2003.[CrossRef][Medline]
6. Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD, Walsh PC. Natural history of progression after PSA elevation following radical prostatectomy. J Am Med Assoc, 281: 1591-7, 1999.[Abstract/Free Full Text]
7. Bartels PH, Wied GL. Automated image analysis in clinical pathology. Am J Clin Pathol, 75: 489-93, 1981.[Medline]
8. Bartels PH. Computer-generated diagnosis and image analysis: an overview. Cancer (Phila.), 69: 1636-8, 1992.
9. Shankey TV, Kallioniemi OP, Koslowski JM, et al Consensus review of the clinical utility of DNA content cytometry in prostate cancer. Cytometry, 14: 497-500, 1993.[CrossRef][Medline]
10. Palcic B. Nuclear texture: can it be used as a surrogate endpoint biomarker?. J Cell Biochem (Suppl.), 19: 40-6, 1994.
11. Boone CW, Lieberman R, Maringer T, Palcic B, Bacus J, Bartels PH. Computer-assisted image analysis-derived intermediate endpoints. Urology, 57: 129-31, 2001.
12. Bacus JW, Grace LJ. Optical microscope system for standardized cell measurements and analyses. Applied Optics, 26: 3280-93, 1987.
13. Diamond DA, Berry SJ, Umbricht C, Jewett HJ, Coffey DS. Computerized image analysis of nuclear shape as a prognostic factor for prostatic cancer. Prostate, 3: 321-32, 1982.[Medline]
14. Desok K, Charlton JD, Coggins JM, Mohler JL. Semiautomated nuclear shape analysis of prostatic carcinoma and benign prostatic hyperplasia. Analyt Quant Cytol Histol, 16: 400-14, 1994.
15. Veltri RW, Partin AW, Epstein JI, et al Quantitative nuclear morphometry, Markovian texture descriptors, and DNA content captured on a CAS-200 image analysis system, combined with PCNA and Her-2/neu immunohistochemistry for prediction of prostate cancer progression. J Cell Biochem (Suppl.), 19: 249-58, 1994.
16. Veltri RW, Miller MC, Partin AW, Coffey DS, Epstein JI. Ability to predict biochemical progression using Gleason score and computer-generated quantitative nuclear grade derived from cancer cell nuclei. Urology, 48: 685-91, 1996.[CrossRef][Medline]
17. Veltri RW, Partin AW, Miller MC. Quantitative nuclear grade (QNG): a new image analysis-based biomarker of clinically relevant nuclear structure alterations. J Cell Biochem (Suppl.), 35: 151-7, 2000.
18. Veltri RW, Miller MC, Mangold LA, O’Dowd GJ, Epstein JI, Partin AW. Prediction of pathological stage in patients with clinical stage T1c prostate cancer: the new challenge. J Urol, 68: 100-4, 2002.
19. Potter SR, Miller MC, Mangold LA, et al Genetically engineered neural networks for predicting prostate cancer progression after radical prostatectomy. Urology, 54: 791-5, 1999.[CrossRef][Medline]
20. Bartels PH, Montironi R, Hamilton PW, Thompson D, Vaught I, Bartels HG. Nuclear chromatin texture in prostatic lesions. I. PIN and adenocarcinoma. Analyt Quant Cytol Histol, 20: 389-96, 1998.
21. Bartels PH, Montironi R, Hamilton PW, Thompson D, Vaught I, Bartels HG. Nuclear chromatin texture in prostatic lesions. II. PIN and malignancy-associated changes. Analyt Quant Cytol Histol, 20: 397-406, 1998.
22. MacAulay C, Lam S, Payne PW, LeRiche JC, Palcic B. Malignancy-associated changes in bronchial epithelial cells in biopsy specimens. Analyt Quant Cytol Histol, 17: 55-61, 1995.
23. Mairinger T, Mikuz G, Gschwendtner A. Nuclear chromatin texture analysis of nonmalignant tissue can detect adjacent prostatic adenocarcinoma. Prostate, 41: 12-9, 1999.[CrossRef][Medline]
24. Beahrs OH Henson DE Hutter RVP eds. . American Joint Committee on cancer staging manual, Lippincott Philadelphia 1992.
25. Gleason DF, The Veterans Administrative Cooperative Urological Research Group. Histological grading and clinical staging of prostatic carcinoma Tannenbaum M eds. . Urologic pathology: the prostate, p. 171-98, Lea and Febiger Philadelphia 1977.
26. Partin AW, Pound CR, Clemens JQ, Epstein JI, Walsh PC. Prostate-specific antigen after anatomic radical prostatectomy: the Johns Hopkins Experience after ten years. Urol Clin N Am, 20: 713-25, 1993.[Medline]
27. Khan MA, Han M, Partin AW, Epstein JI, Walsh PC. Long-term cancer control of radical prostatectomy in men younger than 50 years of age: update 2003. Urology, 62: 86-91, 2003.[Medline]
28. Mohler JL, Partin AW, Epstein JI, Lohr WD, Coffey DS. Nuclear roundness factor measurement for assessment of prognosis of patients with prostatic carcinoma. II. Standardization of methodology for histologic sections. J Urol, 139: 1085-90, 1988.[Medline]
29. Partin AW, Walsh AC, Pitcock RV, Mohler JL, Epstein JI, Coffey DS. A comparison of nuclear morphometry and Gleason grade as a predictor of prognosis in stage A2 prostate cancer: a critical analysis. J Urol, 142: 1254-8, 1989.[Medline]
30. Partin AW, Steinberg GD, Pitcock RV, et al Use of nuclear morphometry, Gleason histologic scoring, clinical stage, and age to predict disease-free survival among patients with prostate cancer. Cancer (Phila.), 70: 161-8, 1992.
31. Epstein JI, Berry SJ, Eggleston JC. Nuclear roundness factor: a predictor of progression in untreated stage A2 prostate cancer. Cancer (Phila.), 54: 1666-71, 1984.
32. Khan MA, Partin AW. Management of high-risk populations with locally advanced prostate cancer. Oncologist, 8: 259-69, 2003.[Abstract/Free Full Text]
33. Veltri RW, Miller MC, An G. Standardization, analytical validation, and quality control of intermediate endpoint biomarkers. Urology, 57: 164-70, 2001.[CrossRef][Medline]

This article has been cited by other articles:

				M. S.Q. Kortenhorst, S. Isharwal, P. J. van Diest, W. H. Chowdhury, C. Marlow, M. A. Carducci, R. Rodriguez, and R. W. Veltri Valproic acid causes dose- and time-dependent changes in nuclear structure in prostate cancer cells in vitro and in vivo Mol. Cancer Ther., April 1, 2009; 8(4): 802 - 808. [Abstract] [Full Text] [PDF]

				J. D. Debes, T. J. Sebo, H. V. Heemers, B. R. Kipp, D. A. L. Haugen, C. M. Lohse, and D. J. Tindall p300 Modulates Nuclear Morphology in Prostate Cancer Cancer Res., February 1, 2005; 65(3): 708 - 712. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Veltri, R. W. Articles by Partin, A. W. Search for Related Content PubMed PubMed Citation Articles by Veltri, R. W. Articles by Partin, A. W.