Description

Vincent et al developed a tool for predicting anemia induced by palliative chemotherapy given to a patient with advanced nonsmall cell lung cancer. This can help identify a patient who may benefit from closer monitoring or preventive therapies. The authors are from the multiple institutions in Canada and Europe as well as Ortho Biotech. Inc.


NOTE: Most these authors were involved in writing the prediction model used in the previous section.

 

Patient selection: advanced nonsmall lung cancer receiving palliative chemotherapy

 

Endpoint: anemia with hemoglobin <= 100 g/L (10 g/dL)

 

Parameters:

(1) hemoglobin prior to chemotherapy

(2) age in years

(3) body surface area (BSA)

(4) ECOG performance status

(5) current therapy for persistent or recurrent cancer

(6) about to receive platinum-based chemotherapy

(7) about to receive gemcitabine

 

Parameter

Finding

Points

hemoglobin prior to chemotherapy

 

-0.07 * (Hgb)

age in years

< 68 years of age

0

 

>= 68 years of age

0.5

body surface area

>= 1.97

0

 

< 1.97

2

ECOG performance scale

0

0

 

1

0.5

 

2 to 4

1

current therapy for persistent or recurrent cancer

no

0

 

yes

0.5

about to receive platinum-based chemotherapy

no

0

 

yes

1

about to receive gemcitabine

no

0

 

yes

0.5

 

total score =

= SUM(points for all 7 parameters) + 13

 

Interpretation:

• minimum total score: 0 (to achieve this would require a hemoglobin of 18.6)

• maximum total score: 15

• The higher the score the greater the risk for developing anemia.

• A risk score between 8.0 and 9.99 was associated with the highest Youden index, with approximately 26% developing anemia.

• A risk score < 6 was associated with a very low risk of anemia.

 

Performance:

• A cutoff of 8.0 - 9.9 had a sensitivity of 83% with specificity of 68%.

• A cutoff of >=10 had a sensitivity of 32% with specificity of 90%.


To read more or access our algorithms and calculators, please log in or register.