• nomogram;
  • prediction;
  • prostate cancer;
  • prognosis;
  • outcome


  1. Top of page
  2. Abstract
  3. What Is a Nomogram?
  4. Conflict of Interest Disclosures
  5. References

Prostate cancer is a heterogeneous disease with a wide prognostic spectrum and a variety of treatment options. Such a complex clinical scenario has led to uncertainty in risk assessment and prediction of outcome. Nomograms have served as scientific formulas designed to maximize the predictive accuracy. The use of nomograms in prostate cancer has been applied to many clinical states and outcomes and has provided the most accurate predictions. Cancer 2009;115(13 suppl):3107–11. © 2009 American Cancer Society.

When a group of students was asked to rank items by weight, few guessed right but the collective average was almost exactly right. In The Wisdom of Crowds, James Surowiecki demonstrates through a series of anecdotes that the predictions of a large group of people are often better than those of any given individual, even if such a person be one of the elite few, and brilliant. In the medical literature, experiments comparing the performance of nomograms with the predictions of expert clinicians concluded that the former outperform the latter.1, 2 It is likely that, in a single experiment, 1 or 2 clinicians or experts will be more accurate than the nomogram or the collective average. But they will not be the same people each time.

In prostate cancer, the fundamental questions raised by Willet Whitmore continue to summarize the clinical dilemma of treating prostate cancer: “Is cure necessary when possible? And is it possible when necessary?”

If the Swedish randomized trial has provided evidence that surgical excision significantly reduces the risk of metastases (hazards ratio [HR], 0.60) and death from prostate cancer (HR, 0.56), establishing the overall superiority of treatment versus nontreatment in the population studied,3 long-term data have demonstrated that few men die from prostate cancer when treated by radical prostatectomy, raising the concern that perhaps many patients with low-risk disease are overtreated. Hence the need for accurate assessment tools to provide answers to Dr. Whitmore's questions: When is cure necessary? When is it possible?

Accurately recognizing patients based on risk groups has been challenging. One approach to categorizing low‒, intermediate‒, and high‒risk groups, based on serum levels of prostate‒specific antigen (PSA) and clinical stage derived from digital rectal examination and biopsy Gleason score, was found to correlate well with outcome for the cohort of patients in each group.4 Such a risk classification is very useful in daily clinical practice, but it is simplistic. Prostate cancer is not. Rather, prostate cancer is a complex heterogeneous disease in which risk spans a continuous spectrum, from totally indolent at 1 end to lethal at the other.

In recent years, research has helped determine clinical and pathologic prognostic parameters. Clinically, pretreatment serum levels of PSA correlate with tumor volume; clinical stage obtained through digital rectal examination and magnetic resonance imaging (MRI) provides information regarding the extent of the disease; and biopsy Gleason score provides insight on the cancer's aggressiveness. Pathologically, the presence of extracapsular extension (ECE), positive surgical margins, seminal vesicle invasion, and lymph node metastasis have been recognized as independent predictors of outcome. Separately, each of these prognostic indicators carries its own relative risk; the ability to assign a prognostic weight to these different parameters and then combine them, whether in a pretreatment or post-treatment setting, is the essence of the nomogram.

What Is a Nomogram?

  1. Top of page
  2. Abstract
  3. What Is a Nomogram?
  4. Conflict of Interest Disclosures
  5. References

Mathematically, a nomogram is a graphic calculating scale designed to provide an approximate computation of a function. The scale can be linear or logarithmic. In clinical practice, a nomogram is used as an algorithm to predict the probability of a given outcome. The accuracy of the algorithm is limited by the precision with which physical markings can be drawn. In his treatise on the analytical solution of linear and quadratic equations, Muhammad Al Khawarizmi (780-850), a Muslim mathematician and founder of algebra, was the first to describe the concept of algorithm (a Latin adaptation of his name). The utilization of nomograms in medical science is not new; to our knowledge, the first introduction of nomograms to medicine was J.L. Henderson's blood nomogram in 1928.5

Management of Prostate Cancer: Why Do We Need Nomograms?

In most developed countries, prostate cancer is the most prevalent malignancy in men. The disease is often diagnosed at a clinically localized stage by means of a biopsy triggered by an abnormal digital rectal examination and/or elevated serum PSA levels. At the current time, prostate cancer cannot usually be visualized or diagnosed by way of cross-sectional medical imaging.

This state of affairs leaves us with an array of information (biopsy Gleason score, PSA levels, percentage of biopsy core involvement, clinical stage, patient's age and comorbidity status, and so on) from which we have to determine the ranking on the risk spectrum and match it with the best treatment choice, a choice that includes, among other options, active surveillance, surgery through either an open or a laparoscopic approach, and radiotherapy by means of brachytherapy, external beam, or a combination of both. However, to our knowledge there are no published results of randomized trials comparing the various modalities of definitive therapy for clinically localized prostate cancer. Such a complex scenario results in a large degree of uncertainty for the patient and the physician when assessing the prognosis of the disease and predicting the outcome of the treatment. Therefore, in counseling a man with prostate cancer, the physician is left with 4 options: 1) Deny the ability to predict at the individual level 2) Predict the outcome based on clinical judgment and experience 3) Predict the outcome for the general group or class that the patient lies in or 4) Assess risk and apply an algorithm or nomogram.

In terms of decision making, predictions based on an algorithm or nomograms have been shown to be more accurate than those based on clinical judgment and experience, and superior to risk groups as well.1, 2

The majority of prostate cancer nomograms are readily available through the Memorial Sloan-Kettering Cancer Center Web site (available at:

Predicting the Pathologic Stage

The goal of radical prostatectomy is to control the cancer with minimal impact on sexual and urinary function. The preservation of neurovascular bundles is associated with a greater likelihood of recovery of sexual function; however, indiscriminate preservation of neurovascular bundles in every man undergoing radical prostatectomy for prostate cancer will undoubtedly increase the risk of local recurrence and treatment failure. Predicting the pathologic stage and particularly the likelihood of ECE is very helpful in surgical treatment planning. In this regard, Partin et al developed statistical tables that provide a cumulative percentage of the probability of ECE, seminal vesicle invasion, and lymph node invasion.6 Although the Partin tables constitute a major contribution in the risk assessment and management of prostate cancer, their use in surgical planning is limited by the finding that the probability of ECE is not specific to location. The use of mapping biopsies through an increase in the standard number of biopsy cores obtained and detailed pathology reports including the number of positive cores and the percentage of cancer in the cores provide valuable information that increases the accuracy of predicting the pathologic stage. Ohori et al developed a nomogram to predict the probability of side-specific ECE based on serum PSA levels, biopsy Gleason grade, and clinical stage. The predictive accuracy of their nomogram increased from 0.788 to 0.806 when the percentage of positive cores and the percentage of cancer in the specimen were added to the model.7 The external validation of this model yielded a predictive accuracy of 0.84.8 Other nomograms predicting pathologic features such as seminal vesical invasion and lymph node invasion were developed. The nomograms of Koh et al9 and Gallina et al10 predicted seminal vesicle invasive with a concordance index ranging from 0.78 to 0.88. Cagiannos et al reported a preoperative nomogram that takes into consideration PSA, clinical stage, biopsy Gleason score sums, and the institutional specific prevalence of positive lymph nodes, with an improved prognostic accuracy of 0.76 when compared with the Partin tables (predictive accuracy of 0.74).11 Given the variability of the anatomic template of pelvic lymph node dissection and the recent data supporting the superiority in staging of the standard (extended) template, Briganti et al developed a pelvic lymph node invasion nomogram based on 602 patients who underwent standard (extended) pelvic lymph node dissection with a predictive accuracy of 0.76.12

Predicting Disease Progression in the Preoperative Setting

In 1998, Kattan et al published what to our knowledge were the first of a series of nomograms relying on preoperative clinical information to predict the probability of freedom from biochemical recurrence at 5 years after radical prostatectomy.13 This nomogram, which had an externally validated predictive accuracy of 0.79, proved to be a remarkable tool for counseling men newly diagnosed with prostate cancer.14, 15 Recently, it was updated to extend the prediction of the probability of biochemical recurrence to 10 years. This version took into account the numbers of positive and negative biopsy cores, along with PSA, biopsy Gleason score (primary and secondary), and clinical stage, and provided an externally validated, well-calibrated, and robust tool that predicts the probability of recurrence at any point in time from 1 to 10 years after surgery.16

Building on the original Kattan nomogram by incorporating pretreatment plasma levels of interleukin-6 soluble receptor and transforming growth factor‒β1, along with preoperative clinical features, improved the concordance index to 0.83.17 Other successful combinations of nonclinical variables were demonstrated by the integration of gene expression signatures to improve the cancer recurrence predictive accuracy of the clinical model (concordance index, 0.89),18 or by the integration of imaging modalities such as endorectal coil MRI in adding incremental value to the clinically based nomogram's prediction of ECE.19

Predicting Disease Progression in the Postoperative Setting

Using pretreatment serum PSA levels along with pathologic variables such as Gleason score, degree of capsular invasion, surgical margin status, and seminal vesicle and lymph node invasion, Kattan et al developed in 1999 a postoperative nomogram to predict the 7-year probability of disease recurrence.20 This nomogram, built on a cohort of 996 men who underwent radical prostatectomy for clinically localized prostate cancer by a single surgeon, achieved a predictive accuracy of 0.73. Validation of the model on an international cohort yielded an accuracy of 0.80.21 Recently, Stephenson et al updated the postoperative nomogram by extending the predictions to 10 years after radical prostatectomy and allowing an adjustment of the predictions to the disease-free interval that a patient has achieved after surgery.22 This feature circumvents any static limitation of the nomogram and recalculates the 10-year progression-free probability after radical prostatectomy on the basis of the length of the disease-free interval achieved by a patient.

Predicting Metastases

The development of metastases is an ominous stage in the spectrum of prostate cancer clinical states and represents a point at which the risk of death from cancer exceeds that from other causes. Predicting metastatic progression is most useful in counseling patients with an increasing PSA after radical prostatectomy and ensuring the homogeneity of the patients' profiles while designing clinical trials.

Dotan et al developed a nomogram based on 414 bone scans of 239 patients who developed biochemical disease recurrence after radical prostatectomy and who were hormonal therapy-naïve.23 The model included the pretreatment PSA, surgical margin status, seminal vesicle invasion, Gleason score on the prostatectomy specimen, extraprostatic extension, and PSA slope and velocity calculated from the last 3 PSA values. This nomogram predicted the probability of a positive bone scan with a concordance index of 0.93. Although highly discriminating, this nomogram is limited to patients who have not received hormonal therapy and was based on nonuniform interpretation of bone scan results interpretation and schedules.

Slovin et al published a nomogram predicting metastatic progression based on a prospective database of 148 patients treated by either surgery or radiotherapy who were enrolled in a clinical protocol and followed at predetermined intervals with physical examination and imaging studies until metastatic progression.24 This nomogram took into consideration PSA kinetics in the form of PSA doubling time and predicted the time to radiographically detectable metastases with an accuracy of 0.69.

Predicting Toxicity After Radical Prostatectomy

The goal of modern radical prostatectomy is to excise all cancer with the least morbidity and full recovery of continence and potency. The outcomes of radical prostatectomy are sensitive to surgical technique and recent evidence has demonstrated prostate cancer patients treated by more experienced surgeons have better cancer control and lesser morbidity than patients who have less experienced surgeons.25-27

For a given surgeon, the challenge lies in the finding that the 3 aims of the procedure are inextricably linked and, at times, may appear to hinder 1 another in that improvement in the outcome of 1 may occur at the expenses of the others. Quoting the risk of 1 individual complication or adverse effect separately of the others does not adequately inform patients seeking treatment for prostate cancer. More often than not, patients may legitimately ask what their chances of being cancer free and at the same time returning to their preoperative ‘normal’ sexual and urinary functional state: the optimal outcome.28, 29

Based on the experience of 2 surgeons who have performed more than 1000 radical prostatectomies each, Eastham et al have recently published what to our knowledge is the first nomogram to predict the likelihood of ‘optimal outcome.’ The model included the pretreatment PSA, clinical stage, pretreatment erectile function, biopsy Gleason grade, age at radical prostatectomy, and months from radical prostatectomy. The nomogram calibrated well and achieved a predictive accuracy of 0.77.

This nomogram did not take into consideration individual comorbidities that may influence erectile function and may not be applicable to other surgeons given the variability of outcomes among surgeons.30


The discovery of the biologic background of prostate cancer will help us decipher the heterogeneity of the disease and establish the prognosis for the individual patient. Until then, nomograms will continue to represent a reliable scientific way to integrate and converge multiple independent variables to provide an accurate prediction.

In this era of information technology, we no longer rely on memory for even the simplest tasks, such as performing calculations or remembering directions, telephone numbers, and addresses; therefore, ‘outsourcing’ our medical decision making to nomogram software may after all be just a sign of the efficiency of modernism.

Conflict of Interest Disclosures

  1. Top of page
  2. Abstract
  3. What Is a Nomogram?
  4. Conflict of Interest Disclosures
  5. References

Sponsored by ASTRA Zeneca and the European School of Oncology.


  1. Top of page
  2. Abstract
  3. What Is a Nomogram?
  4. Conflict of Interest Disclosures
  5. References
  • 1
    Ross PL, Gerigk C, Gonen M, et al. Comparisons of nomograms and urologists' predictions in prostate cancer. Semin Urol Oncol. 2002; 20: 82-88.
  • 2
    Specht MC, Kattan MW, Gonen M, Fey J, Van Zee KJ. Predicting nonsentinel node status after positive sentinel lymph biopsy for breast cancer: clinicians versus nomogram. Ann Surg Oncol. 2005; 12: 654-659.
  • 3
    Bill-Axelson A, Holmberg L, Ruutu M, et al. Radical prostatectomy versus watchful waiting in early prostate cancer. N Engl J Med. 2005; 352: 1977-1984.
  • 4
    D'Amico AV, Whittington R, Malkowicz SB, et al. Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA. 1998; 280: 969-974.
  • 5
    HendersonJL, ed. Blood: A Study in General Physiology. Vol 3. New Haven, CT: Yale University Press; 1928.
  • 6
    Partin AW, Kattan MW, Subong EN, et al. Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multi-institutional update. JAMA. 1997; 277: 1445-1451.
  • 7
    Ohori M, Kattan MW, Koh H, et al. Predicting the presence and side of extracapsular extension: a nomogram for staging prostate cancer. J Urol. 2004; 171: 1844-1849, discussion 1849.
  • 8
    Steuber T, Graefen M, Haese A, et al. Validation of a nomogram for prediction of side specific extracapsular extension at radical prostatectomy. J Urol. 2006; 175( pt 1): 939-944, discussion 944.
  • 9
    Koh H, Kattan MW, Scardino PT, et al. A nomogram to predict seminal vesicle invasion by the extent and location of cancer in systematic biopsy results. J Urol. 2003; 170( pt 1): 1203-1208.
  • 10
    Gallina A, Chun FK, Briganti A, et al. Development and split-sample validation of a nomogram predicting the probability of seminal vesicle invasion at radical prostatectomy. Eur Urol. 2007; 52: 98-105.
  • 11
    Cagiannos I, Karakiewicz P, Eastham JA, et al. A preoperative nomogram identifying decreased risk of positive pelvic lymph nodes in patients with prostate cancer. J Urol. 2003; 170: 1798-1803.
  • 12
    Briganti A, Chun FK, Salonia A, et al. Validation of a nomogram predicting the probability of lymph node invasion among patients undergoing radical prostatectomy and an extended pelvic lymphadenectomy. Eur Urol. 2006; 49: 1019-1026, discussion 1026-1027.
  • 13
    Kattan MW, Eastham JA, Stapleton AM, Wheeler TM, Scardino PT. A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst. 1998; 90: 766-771.
  • 14
    Graefen M, Karakiewicz PI, Cagiannos I, et al. A validation of 2 preoperative nomograms predicting recurrence following radical prostatectomy in a cohort of European men. Urol Oncol. 2002; 7: 141-146.
  • 15
    Graefen M, Karakiewicz PI, Cagiannos I, et al. International validation of a preoperative nomogram for prostate cancer recurrence after radical prostatectomy. J Clin Oncol. 2002; 20: 3206-3212.
  • 16
    Stephenson AJ, Scardino PT, Eastham JA, et al. Preoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Natl Cancer Inst. 2006; 98: 715-717.
  • 17
    Kattan MW, Shariat SF, Andrews B, 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. 2003; 21: 3573-3579.
  • 18
    Stephenson AJ, Smith A, Kattan MW, et al. Integration of gene expression profiling and clinical variables to predict prostate carcinoma recurrence after radical prostatectomy. Cancer. 2005; 104: 290-298.
  • 19
    Wang L, Mullerad M, Chen HN, et al. Prostate cancer: incremental value of endorectal MR imaging findings for prediction of extracapsular extension. Radiology. 2004; 232: 133-139.
  • 20
    Kattan MW, Wheeler TM, Scardino PT. Postoperative nomogram for disease recurrence after radical prostatectomy for prostate cancer. J Clin Oncol. 1999; 17: 1499-1507.
  • 21
    Graefen M, Karakiewicz PI, Cagiannos I, et al. Validation study of the accuracy of a postoperative nomogram for recurrence after radical prostatectomy for localized prostate cancer. J Clin Oncol. 2002; 20: 951-956.
  • 22
    Stephenson AJ, Scardino PT, Eastham JA, et al. Postoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Clin Oncol. 2005; 23: 7005-7012.
  • 23
    Dotan ZA, Bianco FJJr, Rabbani F, et al. Pattern of prostate-specific antigen (PSA) failure dictates the probability of a positive bone scan in patients with an increasing PSA after radical prostatectomy. J Clin Oncol. 2005; 23: 1962-1968.
  • 24
    Slovin SF, Wilton AS, Heller G, Scher HI. Time to detectable metastatic disease in patients with rising prostate-specific antigen values following surgery or radiation therapy. Clin Cancer Res. 2005; 11( pt 1): 8669-8673.
  • 25
    Begg CB, Riedel ER, Bach PB, et al. Variations in morbidity after radical prostatectomy. N Engl J Med. 2002; 346: 1138-1144.
  • 26
    Vickers AJ, Bianco FJ, Serio AM, et al. The surgical learning curve for prostate cancer control after radical prostatectomy. J Natl Cancer Inst. 2007; 99: 1171-1177.
  • 27
    Bianco FJJr, Riedel ER, Begg CB, Kattan MW, Scardino PT. Variations among high volume surgeons in the rate of complications after radical prostatectomy: further evidence that technique matters. J Urol. 2005; 173: 2099-2103.
  • 28
    Bianco FJJr, Scardino PT, Eastham JA. Radical prostatectomy: long-term cancer control and recovery of sexual and urinary function (“trifecta”). Urology. 2005; 66( suppl): 83- 94.
  • 29
    Saranchuk JW, Kattan MW, Elkin E, Touijer AK, Scardino PT, Eastham JA. Achieving optimal outcomes after radical prostatectomy. J Clin Oncol. 2005; 23: 4146-4151.
  • 30
    Eastham JA, Scardino PT, Kattan MW. Predicting an optimal outcome after radical prostatectomy: the trifecta nomogram. J Urol. 2008; 179: 2207-2210, discussion 2210-2211.