Deutsch
Nederlands
A new formula for prostate cancer lymph node risk
Many investigators have created tools for predicting extraprostatic disease and lymph node (LN) involvement. One widely used tool is a linear formula created by Roach et al. the Roach formula (RF), which defines the risk of pelvic LN as follows: (% pelvic LN risk = prostate-specific antigen [PSA])2/3 + (Gleason – 6))10). There has been significant stage migration in prostate cancer over the past decade since the creation of the RF. To provide clinicians with a practical approach to estimating LN risk that was developed from a population-based sample of patients who reflect the vast majority of patients diagnosed in the modern PSA era, and whose care reflects current patterns of care, we developed and validated a new predictive formula using the SEER database. A fast, accurate, and easy-to-use formula would be helpful in discussing LN risk with patients and in the conceptualization of LN risk for future clinical trials.
Many investigators have created tools for predicting extraprostatic disease and lymph node (LN) involvement. One widely used tool is a linear formula created by Roach et al. the Roach formula (RF), which defines the risk of pelvic LN as follows: (% pelvic LN risk = prostate-specific antigen [PSA])2/3 + (Gleason – 6))10). There has been significant stage migration in prostate cancer over the past decade since the creation of the RF. To provide clinicians with a practical approach to estimating LN risk that was developed from a population-based sample of patients who reflect the vast majority of patients diagnosed in the modern PSA era, and whose care reflects current patterns of care, we developed and validated a new predictive formula using the SEER database. A fast, accurate, and easy-to-use formula would be helpful in discussing LN risk with patients and in the conceptualization of LN risk for future clinical trials.
Research authors: James B. Yu, Danil V. Makarov, Cary Gross
Details
Formula
Study characteristics
Files & References
★★★
Model author
Model ID
699
Version
1.10
Revision date
2018-05-11
MeSH terms
Model type
Custom model
(Calculation)
| Formula |
|---|
Additional information
The SEER database was investigated for patients treated in 2004 through 2006. The SEER database is a large national cancer database that encompasses more than 26% of the United States population. Since 2004 it has included pretreatment PSA and pretreatment clinical stage, along with detailed pathologic information, as part ofits standardized data collection (13). We identified 150,012 prostate cancer patients diagnosed from 2004 to 2006 within the SEER database. We excluded patients if they did not receive a radical prostatectomy, if prostate cancer was not their first and only malignancy, if their histology was other than adenocarcinoma, if they received preoperative radiation, if theirclinical stage was recorded as cT2 without further specification, or if they had missing
or incomplete or discordant clinical or pathologic data. Patients for whom prostate cancer was not their first and only malignancy were excluded to prevent the erroneous attribution of non–prostate cancer positive LNs to the original prostate cancer diagnosis. We also excluded patients if they had clinical stage cT1a or cT1b, or had cT3 or cT4 or clinically metastatic disease, to reflect the most common clinical situations of cT1c, cT2a, cT2b, and cT2c disease
and to improve the accuracy of our potential formula.
Formula was made using 1460 patients to build it and 1470 patients to validate it.
Study Population
Total population size: 2930Males: {{ model.numberOfMales }}
Females: {{ model.numberOfFemales }}
Categorical characteristics
| Name | Subset / Group | Nr. of patients |
|---|---|---|
| Race | White | 2518 |
| Black | 244 | |
| American Indian/Alaska Native | 4 | |
| Other unspecified | 4 | |
| Unknown | 23 | |
| Age (y, median = 61) | <50 | 198 |
| 50-59 | 1085 | |
| 60-69 | 1327 | |
| 70-79 | 316 | |
| 80+ | 4 | |
| Gleason score | 6 and lower | 1079 |
| 7 | 1431 | |
| 8 | 210 | |
| 9 | 205 | |
| 10 | 5 | |
| Clinical stage | T1c | 2126 |
| T2a | 215 | |
| T2b/T2c | 589 | |
| Preoperative prostate specific antigen | 0-2.5 | 164 |
| 2.6-4.0 | 304 | |
| 4.1-6.0 | 1076 | |
| 6.1-10 | 900 | |
| 10.1-25.9 | 486 | |
| Pathological stage | Organ confined | 1936 |
| Extraprostatic extension | 725 | |
| Seminal vesicle invasion | 108 | |
| Lymph node involvement | 136 | |
| pT4 | 25 | |
| Number of lymph nodes examined pathologogically | 10-15 | 1868 |
| 16-20 | 575 | |
| 21-30 | 392 | |
| 31-56 | 95 |
Related files
No related files available
Supporting Publications
| Title or description | Tags |
|---|---|
| A new formula for prostate cancer lymph node risk. | Internal validation Paper Peer review Model coefficients |
The calculated probability of lymph node involvement is: ... %
{{ resultSubheader }}
{{ model.survival.PITTitle }}
{{ model.survival.YNETitle }}
Result
Note
Notes are only visible in the result download and will not be saved by Evidencio
The calculated probability of lymph node involvement is: %
{{ resultSubheader }}
Outcome stratification
Result interval
{{ additionalResult.min }} to {{ additionalResult.max }}
Conditional information
Result interpretation
Reported sensitivity is 39% and specificity is 94,9%.
Compared to the Roach Formula (RF) and Nguyen Formula (NF), our Yale Formula (YF) was able to classify the most patients in the high-risk (>15%) group while still maintaining good PPV and specificity. The YF had the best combination of sensitivity with high specificity and did not underestimate LN risk as the NF did. Underestimation of LN risk is potentially more harmful than overestimation of risk if patients are counseled to undergo prostate-only therapy based on an underestimation of risk, whereas they could potentially undergo pelvic LN irradiation with a chance at improving subsequent pelvic and cure rates. The utility of pelvic radiotherapy remains controversial, but it stands to reason that pelvic radiotherapy should be given more often to patients at higher risk for LN metastases than to those who are at lower risk.
Calculations alone should never dictate patient care, and are no substitute for professional judgement. See our full disclaimer.
Comments
Rating
Comment
Please enter a comment of rating
Comments are visible to anyone
Model feedback
No feedback yet 1 Comment {{ model.comments.length }} Comments
Recent Oncology Models
Recent Urology Models
Sign In
Sign Up
With an Evidencio Community account you can:
- Create and publish your own prediction models.
- Share your prediction models with your colleagues, research group, organization or the world.
- Review and provide feedback on models that have been shared with you.
- Validate your models and validate models from other users.
- Find models based on Title, Keyword, Author, Institute, or MeSH classification.
- Use and save prediction models and their data.
- Use patient specific protocols and guidelines based on sequential models and decision trees.
- Stay up-to-date with new models in your field as they are published.
- Create your own lists of favorite models and topics.
Please sign in to enable Evidencio print features
In order to use the Evidencio print features, you need to be logged in.
If you don't have an Evidencio Community Account you can create your free personal account at:
https://www.evidencio.com/registration
If you don't have an Evidencio Community Account you can create your free personal account at:
https://www.evidencio.com/registration
Printed results - Examples {{ new Date().toLocaleString() }}
Evidencio Community Account Benefits
With an Evidencio Community account you can:
- Create and publish your own prediction models.
- Share your prediction models with your colleagues, research group, organization or the world.
- Review and provide feedback on models that have been shared with you.
- Validate your models and validate models from other users.
- Find models based on Title, Keyword, Author, Institute, or MeSH classification.
- Use and save prediction models and their data.
- Use patient specific protocols and guidelines based on sequential models and decision trees.
- Stay up-to-date with new models in your field as they are published.
- Create your own lists of favorite models and topics.
Disclaimer: Calculations alone should never dictate patient care, and are no substitute for professional judgement.
{{ variable.title }}