Hess et al used classification and regression tree (CART) analysis to evaluate prognosis for a patient with a metastatic carcinoma from an unknown primary site. One analysis starts from the whether or not lymph node metastases are present. The authors are from the University of Texas M.D. Anderson Cancer Center.
Parameters:
(1) lymph node
(2) histologic type
(3) number of sites involved
(4) liver metastases
(5) bone metastases
(6) adrenal gland metastases
(7) age of the patient in years
Node |
Histo |
Sites |
Liver |
Bone |
Adrenal |
Age |
Group |
pos |
S, C, N |
<= 2 |
|
|
|
|
E |
pos |
A |
<= 2 |
|
|
|
|
C |
pos |
|
> 2 |
|
|
pos |
|
C |
pos |
|
> 2 |
|
|
neg |
|
A |
neg |
N |
|
|
|
|
|
D |
neg |
S, C, A |
|
neg |
neg |
|
|
C |
neg |
S, C, A |
|
neg |
pos |
|
|
B |
neg |
S, C, A |
|
pos |
|
|
<= 61 |
B |
neg |
S, C, A |
|
pos |
|
|
> 61 |
A |
where:
• Histologic types: S = squamous; A = adenocarcinoma; N = neuroendocrine; C = carcinoma NOS
The CART analysis identified 9 groups, but I combined some of these to simplify the presentation.
Group |
Median Survival |
A |
5 months |
B |
8 months |
C |
12 - 13 months |
D |
24 months |
E |
45 months |
Specialty: Hematology Oncology, Surgery, general