Ed. Crawford et al., The use of artificial intelligence technology to predict lymph node spreadin men with clinically localized prostate carcinoma, CANCER, 88(9), 2000, pp. 2105-2109
BACKGROUND. The current study assesses artificial intelligence methods to i
dentify prostate carcinoma patients at low risk for lymph node spread. If p
atients can be assigned accurately to a low risk group, unnecessary lymph n
ode dissections can be avoided, thereby reducing morbidity and costs.
METHODS. A rule-derivation technology for simple decision-tree analysis was
trained and validated using patient data from a large database (4133 patie
nts) to derive low risk cutoff, values for Gleason sum and prostate specifi
c antigen (PSA) level. An empiric analysis was used to derive a low risk cu
toff value for clinical TNM stage. These cutoff Values then were applied to
2 additional, smaller databases (227 and 330 patients, respectively) from
separate institutions.
RESULTS. The decision-tree protocol derived cutoff values of less than or e
qual to 6 for Gleason sum and less than or equal to 10.6 ng/mL for PSA. The
empiric analysis yielded a clinical TNM stage low risk cutoff value of les
s than or equal to T2a. When these cutoff values were applied to the larger
database, 44% of patients were classified as being at low risk for lymph n
ode metastases (0.8% false-negative rate). When the same cutoff values were
applied to the smaller databases, between II and 43% of patients were clas
sified as low risk with a false-negative rate of between 0.0 and 0.7%.
CONCLUSIONS, The results of the current study indicate that a population of
prostate carcinoma patients at low risk for lymph node metastases can be i
dentified accurately using a simple decision algorithm that considers preop
erative PSA, Gleason sum, and clinical TNM stage. The risk of lymph node me
tastases in these patients is less than or equal to 1%; therefore, pelvic l
ymph node dissection may be avoided safely. The implications of these findi
ngs in surgical and nonsurgical treatment are significant. (C) 2000 America
n Cancer Society.