This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Graefen, M. Articles by Kattan, M. W. Search for Related Content PubMed PubMed Citation Articles by Graefen, M. Articles by Kattan, M. W. Social Bookmarking                            What's this?

International Validation of a Preoperative Nomogram for Prostate Cancer Recurrence After Radical Prostatectomy

From the Departments of Urology and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, NY; Garvan Institute of Medical Research and St Vincent’s Hospital, Sydney, Australia; Cleveland Clinic, Cleveland, OH; University of Southern California, and University of California, Los Angeles, CA; University Hospital, Hamburg-Eppendorf, Germany; Louisiana State University Health Science Center, Shreveport, LA; and Department of Urology, Erasmus University, and Academic Hospital Rotterdam, the Netherlands.

Address reprint requests to Michael W. Kattan, PhD, Departments of Urology, Epidemiology, and Biostatistics, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021; email: kattanm{at}mskcc.org

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: We evaluated the predictive accuracy of a recently published preoperative nomogram for prostate cancer that predicts 5-year freedom from recurrence. We applied this nomogram to patients from seven different institutions spanning three continents.

METHODS: Clinical data of 6,754 patients were supplied for validation, and 6,232 complete records were used. Nomogram-predicted probabilities of 60-month freedom from recurrence were compared with actual follow-up in two ways. First, areas under the receiver operating characteristic curves (AUCs) were determined for the entire data set according to several variables, including the institution where treatment was delivered. Second, nomogram classification–based risk quadrants were compared with actual Kaplan-Meier plots.

RESULTS: The AUC for all institutions combined was 0.75, with individual institution AUCs ranging from 0.67 to 0.83. Nomogram predictions for each risk quadrant were similar to actual freedom from recurrence rates: predicted probabilities of 87% (low-risk group), 64% (intermediate-low–risk group), 39% (intermediate-high–risk group), and 14% (high-risk group) corresponded to actual rates of 86%, 64%, 42%, and 17%, respectively. The use of neoadjuvant therapy, variation in the prostate-specific antigen recurrence definitions between institutions, and minor differences in the way the Gleason grade was reported did not substantially affect the predictive accuracy of the nomogram.

CONCLUSION: The nomogram is accurate when applied at international treatment institutions with similar patient selection and management strategies. Despite the potential for heterogeneity in patient selection and management, most predictions demonstrated high concordance with actual observations. Our results demonstrate that accurate predictions may be expected across different patient populations.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
NOMOGRAMS, WHICH are tools that predict outcome probabilities for individual patients, can be useful for a variety of reasons. Although counseling the individual patient and treatment decision-making may be the primary uses, research purposes such as clinical trial eligibility are also important. There are numerous pitfalls to nomogram development, and chief among them is validation failure. A nomogram may not predict well when applied to future patients if the predictor variables are not reproducible or are confounded by other factors. If the sample size, follow-up, or both used for nomogram development are inadequate, the estimates may similarly be suboptimal. In addition, the statistical model behind the nomogram may lack fit, either in the derivation phase or in the validation sample. Therefore, validation on one or ideally on several cohorts represents an essential step before a nomogram may be safely implemented in routine clinical practice.¹

In 1998, a nomogram was published² that predicted the probability that a prostate cancer patient diagnosed with clinically localized disease would avoid disease progression within 5 years after radical prostatectomy (RP; Fig 1). The basis for the nomogram was a Cox regression model derived from 983 men treated by a single surgeon. The nomogram considered pretreatment serum prostate-specific antigen (PSA), clinical stage, and Gleason grade of the biopsy as predictor variables. Disease progression was defined as biochemical, clinical, administration of adjuvant therapy, or death from prostate cancer. The single surgeon performed all clinical staging, and a single pathologist read the biopsy slides for all patients. The Hybritech (San Diego, CA) PSA assay was used principally but not exclusively for both pretreatment evaluation and disease follow-up.

View larger version (42K): [in this window] [in a new window]   Fig 1. Preoperative nomogram based on 983 patients treated at Baylor College of Medicine, Houston, TX, for predicting freedom from recurrence after radical prostatectomy, adapted from Kattan et al.2 PSA, prostate-specific antigen.

Despite good performance of the nomogram in this separate data set, concerns related to the generalizability of the nomogram remained. These related to the homogeneity of the validation population as well as to its similarity on staging, diagnosis, and treatment processes. Further validation of the nomograms in different populations and using different diagnostic, staging, and treatment processes was clearly required. Herein we address that requirement. We submitted the original nomogram to further validity testing by using data sets from institutions across the United States and overseas.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Validation data representing men treated with RP were obtained from seven institutions: four in the United States (n = 3,869) and three overseas (n = 2,363). The following institutions provided data for this study: University of Southern California (USC), Los Angeles, CA (n = 1,501); Cleveland Clinic, Cleveland, OH (n = 1,168); University of California at Los Angeles, Los Angeles, CA (n = 617); Louisiana State University Medical Center, Shreveport, LA (n = 583); University Hospital Hamburg-Eppendorf, Germany (n = 1,134); Garvan Institute of Medical Research and St Vincent’s Hospital, Sydney, Australia (n = 754); and Erasmus University and Academic Hospital Rotterdam, Rotterdam, the Netherlands (n = 475). Table 1 lists the clinical and pathologic characteristics of patients included in this validation and those of the original patients used to develop the nomogram. Table 2 lists the follow-up status and characteristics for each institution’s patients. PSA failures were defined by each institution individually and ranged from a single value of 0.1 to 0.4 ng/mL followed by another value that was higher. Preoperative biopsies were either mainly performed in the respective institution (on-site) or outside the institution (off-site). Clinical recurrences concurrent with an increased PSA were classified as PSA failures. Patients with a pretreatment PSA of 100 ng/mL (n = 23) were excluded because the nomogram does not predict outcome for these patients. Patients with missing pretreatment PSA values (n = 180), clinical stage (n = 98), or biopsy Gleason sum (n = 221) were excluded from analysis. Clinical stage in all institutions was assigned by using the 1992 American Joint Committee on Cancer tumor-node-metastasis classification.

Biopsy Gleason Grade
Primary and secondary Gleason grades were available in 4,708 patients. In 1,524 patients (USC and Cleveland Clinic), the primary and secondary Gleason grade patterns could not be provided, and only the Gleason sum was available. To address this limitation, we recoded the Gleason sum data from these 1,524 men by creating a scenario with a higher primary or higher secondary Gleason grade for purposes of sensitivity analyses. For example, a Gleason sum of 2, 3, or 4 was always considered in the nomogram as a Gleason grade of < 3 + < 3, and a Gleason sum of 6 was considered as a Gleason grade of 3 + 3. Furthermore, Gleason sums of 8, 9, or 10 were all counted as > 3 + > 0. However, a Gleason sum of 5 could be considered as 3 + < 3 (high-primary scenario for biopsy Gleason grade) or < 3 + 3 (high-secondary scenario for biopsy Gleason grade). Similarly, a Gleason sum of 7 could be entered into the nomogram as > 3 + > 0 (high-primary scenario for biopsy Gleason grade) or < 4 + > 3 (high-secondary scenario for biopsy Gleason grade). Therefore, the scenarios that were based on the high-primary or high-secondary Gleason grade differed only when the Gleason sum was 5 or 7. We ultimately evaluated the nomogram in three ways: (1) using data only from the series that provided primary and secondary Gleason grades, (2) assuming the high-primary scenario for the Gleason sum 5 and 7 patients, and (3) assuming the high-secondary scenario for these patients.

Neoadjuvant and Adjuvant Hormonal Therapy
Separate analyses addressed patients with and without neoadjuvant hormonal therapy with the goal of assessing the impact of neoadjuvant hormonal therapy on nomogram accuracy. Patients who received neoadjuvant therapy did so in a nonstandardized fashion; treatment usually lasted up to 3 months. A few patients received hormones for longer than 3 months (n = 31). Neoadjuvant therapy included androgen-receptor blockade, luteinizing hormone–releasing hormone agonists, or a combination of both. In all analyses, patients who received adjuvant hormonal therapy were considered to have experienced treatment failure at the initiation of hormonal therapy in accordance with the nomogram derivation data set.

Adjuvant Radiation Therapy
In three of the participating institutions, a subset of patients was treated with adjuvant radiation therapy before evidence of recurrence. These patients would be considered as having experienced immediate treatment failure according to the definitions used for the nomogram derivation. In the derivation data set at Baylor College of Medicine, treatment was considered to have failed in 3% of the men (n = 25) because of initiation of adjuvant radiation therapy. At the Cleveland Clinic and the Garvan Institute of Medical Research/St Vincent’s Hospital, 2.4% (n = 28) and 6.4% (n = 48), respectively, received adjuvant radiation therapy and were treated as having experienced treatment failure at the time of initiation of secondary treatment. Of 1,501 patients treated at USC, 420 patients (28%) received adjuvant radiation therapy according to local protocol designed for patients with unfavorable pathologic findings. However, radiation therapy use reflected institutional policy and did not represent treatment failure per se. We addressed this situation by performing two separate analyses. (1) Individuals treated with adjuvant therapy were classified as having experienced treatment failure at the time of adjuvant therapy in accordance with the original nomogram philosophy, and (2) adjuvant therapy was considered part of primary therapy and thus ignored as a failure event. Treatment failure was defined on the basis of evidence of clinical or biochemical recurrence. Time to treatment failure was calculated from the date of the prostatectomy.

Statistical Methods
We performed receiver operating characteristic curve analysis that compared the predicted 5-year probability of freedom from recurrence with the actual follow-up. Because the data were censored, the traditional area under the receiving operating characteristic curve (AUC) is problematic,³ and Harrell’s version was calculated.⁴ Nonetheless, its interpretation is similar. The AUC is the probability that, given two randomly drawn patients, the patient whose disease recurs first had a higher probability of recurrence. Note that the calculation assumes that the patient with the shorter follow-up has the first disease recurrence. If both patients’ disease recurs at the same time or if the patient without recurrent disease has a shorter follow-up, this pair of patients is excluded from the calculation. We first calculated the AUC for those patients who closely fit the derivation criteria (patients who had both primary and secondary Gleason grades assigned, did not receive neoadjuvant hormonal therapy, and were not treated at an institution with an aggressive adjuvant radiation therapy policy; n = 4,280). We then calculated the AUC for the various conditions described previously as well as for each institution individually.

Calibration of the nomogram was assessed by comparing its predicted probability of prostate cancer recurrence with actual recurrence. Computation of actual progression-free probability was performed with Kaplan-Meier analysis. To facilitate this analysis, population quadrants were defined according to nomogram-predicted probability of recurrence (0% to 25%, 26% to 50%, 51% to 75%, and 76% to 100%). Subsequently, using the Kaplan-Meier method, we determined the actuarial probability of recurrence at 5 years after RP. This allowed a comparison of the median nomogram prediction for each quadrant at 5 years with the 5-year Kaplan-Meier probability. Note that restricting the analysis to only those patients with 5-year follow-up would result in a biased estimation of recurrence,⁵ necessitating the use of an actuarial method such as Kaplan-Meier. All statistical tests performed were two sided.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Table 1 lists the preoperative serum PSA levels, clinical stages, and biopsy Gleason grades by institution and for the entire cohort of 6,232 men. Differences exist between institutions. For example, the pretreatment PSA level in United States patients was lower than in non–United States patients (P < .0001; Wilcoxon rank sum test). Differences among sites in the distribution of the predictor variables provided a more rigorous validation of the nomogram by testing it in populations with different baseline characteristics. If the hypothesis that the nomogram will predict across a range of patients is correct, then differences in the range of predictor variables between institutions should not be of consequence, because the recurrence nomogram uses these variables when making predictions and should adjust for any differences.

Similarly, crude treatment failure rates (Table 2) seem different, but these are unadjusted for the patient mix and follow-up and are therefore difficult to compare across series. The overall 5-year freedom from recurrence rate was 72% (95% confidence interval, 70 to 74) for the 4,280 patients who had primary and secondary Gleason grades assigned, did not receive neoadjuvant hormonal therapy, and were not treated at an institution with an aggressive adjuvant radiation therapy policy (Fig 2). These patients closely fit the derivation criteria of the nomogram, and receiver operating characteristic analysis resulted in an AUC of 0.75.

View larger version (11K): [in this window] [in a new window]   Fig 2. Freedom from recurrence in 4,280 patients; patients had both primary and secondary Gleason grades assigned, did not receive neoadjuvant hormones, and did not have an aggressive adjuvant radiation therapy policy (dotted lines, 95% confidence bands; numbers above months, patients at risk).

View larger version (17K): [in this window] [in a new window]   Fig 3. Kaplan-Meier analyses of freedom from recurrence; risk-group stratification was based on the predicted likelihood of freedom from recurrence per the nomogram. Median predicted 5-year recurrence-free probabilities were 87%, 64%, 39%, and 14% for the strata. Numbers below the x-axis indicate patients at risk.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Herein, we report the results of a validation study on external and heterogenous data that originated from three different continents, where detection, staging, and treatment may vary. Physicians from several institutions supplied their databases, and although still not perfectly representative of all prostate cancer patients who are diagnosed and eligible for surgery, these data sets are a significant improvement in the pursuit of generalizability. The comparison of the predicted outcome and actuarial recurrence demonstrated agreement differing by less than 5%. This demonstrates the general applicability of the nomogram, at least in the academic setting. Future research needs to examine applicability in the community setting.

When applied to the closely matching subset of patients (Fig 2), the nomogram had an AUC of 0.75. Across institutions, the AUC ranged from 0.67 to 0.83. When originally developed, the nomogram had a bootstrap-corrected AUC of 0.74 and a small validation data set AUC of 0.79. Reasons for the variability across sites remain unclear, because we were unable to recognize any specific pattern in predictor variables, PSA cutoff definitions, or follow-up status that may have contributed to an improved AUC. However, it must be emphasized that differences in the AUCs merely reflect differences in the ability of the nomogram to predict outcome in a given group of patients and not an actual difference in patient outcome. In other words, a high AUC means that outcome was predicted accurately by the nomogram, not that patient outcome was better than expected. It is interesting to note that the derivation data set was the only one to incorporate imaging results into the assignment of clinical stage, but validation accuracy did not suffer as a consequence.

A limitation of the nomogram is its decreased ability to account for adjuvant radiation therapy. Consideration of adjuvant radiation therapy as immediate treatment failure would be consistent with the original definition of recurrence used in the nomogram. However, in the derivation data set, only 25 patients (3%) received adjuvant radiation therapy. Although this percentage is similar to that in some participating institutions where adjuvant radiation treatment was performed (at the Cleveland Clinic and the Garvan Institute of Medical Research, 2.4% and 6.4% of the patients, respectively, received adjuvant radiation), 420 (28%) of 1,501 eligible USC patients were irradiated within the first 3 months after surgery on the basis of the policy of this institution. As we have discussed previously,² statistical considerations of these patients are complex. Any second treatment has the potential to mask a recurrence, and this would falsely decrease the recurrence rate. However, simply omitting these patients might eliminate a major fraction of all the recurrences, also biasing the recurrence rates. In addition to excluding the entire USC series from the initial analysis, we therefore decided to calculate the AUCs for two additional scenarios: in the first, patients receiving adjuvant radiation were included as having experienced treatment failure (AUCs ranged from 0.70 to 0.73). In the second scenario, these patients were observed and were considered to have experienced treatment failure once there was evidence of recurrence (AUCs ranged from 0.74 to 0.77). Considering adjuvant therapy as immediate treatment failure therefore consistently decreased the predictive accuracy of the nomogram when all institutions, including those with an aggressive second-therapy policy, were included. Observing these patients regardless of their second therapy led to an improved AUC. In either case, the AUC remained above 0.70. These results should not be interpreted as making any statement on the efficacy of adjuvant radiation therapy.

A separate analysis of patients with and without neoadjuvant hormonal therapy allowed us to judge the effect of this treatment strategy on predictive accuracy. The AUCs in patients with and without and regardless of neoadjuvant hormonal therapy differed by a maximum of 1%. This suggests that neoadjuvant treatment does not affect the accuracy of the nomogram. It is noteworthy to mention that neoadjuvant treatment was administered in a nonstandardized fashion. Most patients received treatment of less than 3 months’ duration, and we are not able to make any statement as to the efficacy of this therapy.

The performance of the nomogram in the high-primary scenario (AUCs ranged from 0.74 to 0.75) and high-secondary scenario (AUCs also ranged from 0.74 to 0.75) for the biopsy Gleason grade indicated that changes in the primary and secondary Gleason grade had no adverse effect on the predictive accuracy of the nomogram. This indicates the stability of the nomogram and suggests that it is applicable even to institutions where only the biopsy Gleason sum is available.

The results of our study—to our knowledge, the largest validation study on a nomogram predicting outcome after RP published to date—suggest that the recurrence nomogram is accurate and generalizable. It is interesting to note that the seven institutions spanned three continents and that the highest AUC was found in a European institution, perhaps further supporting generalizability.

This study has important implications for the practicing physician as well as for patients who need a prediction of surgical efficacy. It seems that the nomogram is an accurate prognostic tool for recurrence prediction. Its value is enhanced in the context of a broader decision analysis of the treatment options, where predictions of outcome after external-beam radiation¹² and brachytherapy¹³ are needed. Today, predictions from these and other prostate cancer nomograms are freely available in software for the desktop or handheld computer (http://www.nomograms.org).

In conclusion, our study demonstrated that the recurrence nomogram based on preoperative Gleason grade, PSA, and clinical stage was applicable to other academic institutions both within the United States and internationally. It allows accurate prediction of 5-year freedom of recurrence after RP even when minor variations of the input variables or patient selection are evident.

	ACKNOWLEDGMENTS

 
Supported by grants from the Deutsche Forschungsgemeinschaft (GR 1866/1-1) and Deutsche Krebshilfe. P.I.K. was partially supported by the American Foundation for Urologic Diseases, the National Cancer Institute of Canada, and the Medical Research Council of Canada. Support was also provided by the National Health and Medical Research Council of Australia (NHMRC), the New South Wales Cancer Council, the R.T. Hall Trust, the Freedman Foundation, and St Vincent’s Clinic Foundation and St Vincent’s Hospital Foundation. D.I.Q. was the recipient of an NHMRC Neil Hamilton Fairley Postdoctoral Fellowship and the Vincent Fairfax Family Foundation Fellowship from the Royal Australasian College of Physicians. Additional support was supplied by grant no. RPG-00-202-01-CCE, awarded to M.W.K. by the American Cancer Society.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
1. Justice AC, Covinsky KE, Berlin JA: Assessing the generalizability of prognostic information. Ann Intern Med 130: 515-524, 1999[Abstract/Free Full Text]

2. Kattan MW, Eastham JA, Stapleton AMF, et al: A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst 90: 766-771, 1998[Abstract/Free Full Text]

3. Begg CB, Cramer LD, Venkatraman ES, et al: Comparing tumor staging and grading systems: A case study and a review of the issues, using thymoma as a model. Stat Med 19: 1997-2014, 2000[CrossRef][Medline]

4. Harrell FE Jr, Califf RM, Pryor DB, et al: Evaluating the yield of medical tests. JAMA 247: 2543-2546, 1982[Abstract/Free Full Text]

5. Fisher LD, Van Belle G: Biostatistics: A Methodology for the Health Sciences. New York NY, John Wiley & Sons Inc, 1993

6. D’Amico AV, Whittington R, Malkowicz SB, et al: Pretreatment nomogram for prostate-specific antigen recurrence after radical prostatectomy or external-beam radiation therapy for clinically localized prostate cancer. J Clin Oncol 17: 168-172, 1999[Abstract/Free Full Text]

7. Snow PB, Smith DS, Catalona WJ: Artificial neural networks in the diagnosis and prognosis of prostate cancer: A pilot study. J Urol 152: 1923-1926, 1994 (5 pt 2)[Medline]

8. Graefen M, Noldus J, Pichlmeier U, et al: Early prostate-specific antigen relapse after radical retropubic prostatectomy: Prediction on the basis of preoperative and postoperative tumor characteristics. Eur Urol 36: 21-30, 1999[Medline]

9. D’Amico AV, Whittington R, Malkowicz SB, et al: Clinical utility of the percentage of positive prostate biopsies in defining biochemical outcome after radical prostatectomy for patients with clinically localized prostate cancer. J Clin Oncol 18: 1164-1172, 2000[Abstract/Free Full Text]

10. Partin AW, Yoo J, Carter HB, et al: The use of prostate specific antigen, clinical stage and Gleason score to predict pathological stage in men with localized prostate cancer. J Urol 150: 110-114, 1993[Medline]

11. Blute ML, Bergstralh EJ, Partin AW, et al: Validation of Partin tables for predicting pathological stage of clinically localized prostate cancer. J Urol 164: 1591-1595, 2000[CrossRef][Medline]

12. Kattan MW, Zelefsky MJ, Kupelian PA, et al: Pretreatment nomogram for predicting the outcome of three-dimensional conformal radiotherapy in prostate cancer. J Clin Oncol 18: 3352-3359, 2000[Abstract/Free Full Text]

13. Kattan MW, Potters L, Blasko JC, et al: Pretreatment nomogram for predicting freedom from recurrence after permanent prostate brachytherapy in prostate cancer. Urology 58: 393-399, 2001[CrossRef][Medline]

Submitted December 5, 2001; accepted April 14, 2002.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				M. R. Cooperberg, J. M. Broering, and P. R. Carroll Risk Assessment for Prostate Cancer Metastasis and Mortality at the Time of Diagnosis J Natl Cancer Inst, June 16, 2009; 101(12): 878 - 887. [Abstract] [Full Text] [PDF]

				S. F. Shariat, J. A. Karam, J. Walz, C. G. Roehrborn, F. Montorsi, V. Margulis, F. Saad, K. M. Slawin, and P. I. Karakiewicz Improved Prediction of Disease Relapse after Radical Prostatectomy through a Panel of Preoperative Blood-Based Biomarkers Clin. Cancer Res., June 15, 2008; 14(12): 3785 - 3791. [Abstract] [Full Text] [PDF]

				S. F. Shariat, J. Walz, C. G. Roehrborn, A. R. Zlotta, P. Perrotte, N. Suardi, F. Saad, and P. I. Karakiewicz External Validation of a Biomarker-Based Preoperative Nomogram Predicts Biochemical Recurrence After Radical Prostatectomy J. Clin. Oncol., March 20, 2008; 26(9): 1526 - 1531. [Abstract] [Full Text] [PDF]

				A. S. Bjartell, H. Al-Ahmadie, A. M. Serio, J. A. Eastham, S. E. Eggener, S. W. Fine, L. Udby, W. L. Gerald, A. J. Vickers, H. Lilja, et al. Association of Cysteine-Rich Secretory Protein 3 and {beta}-Microseminoprotein with Outcome after Radical Prostatectomy Clin. Cancer Res., July 15, 2007; 13(14): 4130 - 4138. [Abstract] [Full Text] [PDF]

				A. J. Stephenson, P. T. Scardino, M. W. Kattan, T. M. Pisansky, K. M. Slawin, E. A. Klein, M. S. Anscher, J. M. Michalski, H. M. Sandler, D. W. Lin, et al. Predicting the Outcome of Salvage Radiation Therapy for Recurrent Prostate Cancer After Radical Prostatectomy J. Clin. Oncol., May 20, 2007; 25(15): 2035 - 2041. [Abstract] [Full Text] [PDF]

				L. Wang, H. Hricak, M. W. Kattan, H. N. Chen, K. Kuroiwa, H. F. Eisenberg, and P. T. Scardino Prediction of Seminal Vesicle Invasion in Prostate Cancer: Incremental Value of Adding Endorectal MR Imaging to the Kattan Nomogram Radiology, December 1, 2006; 242(1): 182 - 188. [Abstract] [Full Text] [PDF]

				L. True, I. Coleman, S. Hawley, C.-Y. Huang, D. Gifford, R. Coleman, T. M. Beer, E. Gelmann, M. Datta, E. Mostaghel, et al. A molecular correlate to the Gleason grading system for prostate adenocarcinoma PNAS, July 18, 2006; 103(29): 10991 - 10996. [Abstract] [Full Text] [PDF]

				A. J. Stephenson, P. T. Scardino, J. A. Eastham, F. J. Bianco Jr., Z. A. Dotan, P. A. Fearn, and M. W. Kattan Preoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Natl Cancer Inst, May 17, 2006; 98(10): 715 - 717. [Abstract] [Full Text] [PDF]

				L. Wang, H. Hricak, M. W. Kattan, L. H. Schwartz, S. C. Eberhardt, H.-N. Chen, and P. T. Scardino Combined endorectal and phased-array MRI in the prediction of pelvic lymph node metastasis in prostate cancer. Am. J. Roentgenol., March 1, 2006; 186(3): 743 - 748. [Abstract] [Full Text] [PDF]

				M. S. Anscher PSA Kinetics and Risk of Death From Prostate Cancer: In Search of the Holy Grail of Surrogate End Points JAMA, July 27, 2005; 294(4): 493 - 494. [Full Text] [PDF]

				D. A. Loblaw, D. S. Mendelson, J. A. Talcott, K. S. Virgo, M. R. Somerfield, E. Ben-Josef, R. Middleton, H. Porterfield, S. A. Sharp, T. J. Smith, et al. American Society of Clinical Oncology Recommendations for the Initial Hormonal Management of Androgen-Sensitive Metastatic, Recurrent, or Progressive Prostate Cancer J. Clin. Oncol., July 15, 2004; 22(14): 2927 - 2941. [Abstract] [Full Text] [PDF]

				L. Wang, M. Mullerad, H.-N. Chen, S. C. Eberhardt, M. W. Kattan, P. T. Scardino, and H. Hricak Prostate Cancer: Incremental Value of Endorectal MR Imaging Findings for Prediction of Extracapsular Extension Radiology, July 1, 2004; 232(1): 133 - 139. [Abstract] [Full Text] [PDF]

				L. G. Horvath, S. M. Henshall, J. G. Kench, D. N. Saunders, C.-S. Lee, D. Golovsky, P. C. Brenner, G. F. O'Neill, R. Kooner, P. D. Stricker, et al. Membranous Expression of Secreted Frizzled-Related Protein 4 Predicts for Good Prognosis in Localized Prostate Cancer and Inhibits PC3 Cellular Proliferation in Vitro Clin. Cancer Res., January 15, 2004; 10(2): 615 - 625. [Abstract] [Full Text] [PDF]

				M. W. Kattan, S. F. Shariat, B. Andrews, K. Zhu, E. Canto, K. Matsumoto, M. Muramoto, P. T. Scardino, M. Ohori, T. M. Wheeler, et al. The Addition of Interleukin-6 Soluble Receptor and Transforming Growth Factor Beta1 Improves a Preoperative Nomogram for Predicting Biochemical Progression in Patients With Clinically Localized Prostate Cancer J. Clin. Oncol., October 1, 2003; 21(19): 3573 - 3579. [Abstract] [Full Text] [PDF]

				M. W. Kattan, M. S. Karpeh, M. Mazumdar, and M. F. Brennan Postoperative Nomogram for Disease-Specific Survival After an R0 Resection for Gastric Carcinoma J. Clin. Oncol., October 1, 2003; 21(19): 3647 - 3650. [Abstract] [Full Text] [PDF]

				S R J Bott, A J Birtle, C J Taylor, and R S Kirby Prostate cancer management: (1) an update on localised disease Postgrad. Med. J., October 1, 2003; 79(936): 575 - 580. [Abstract] [Full Text] [PDF]

				A. V. D'Amico Predicting Prostate-Specific Antigen Recurrence Established: Now, Who Will Survive? J. Clin. Oncol., August 1, 2002; 20(15): 3188 - 3190. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Graefen, M. Articles by Kattan, M. W. Search for Related Content PubMed PubMed Citation Articles by Graefen, M. Articles by Kattan, M. W. Social Bookmarking                            What's this?