They devised and tested the tool, called the Cancer and Aging Research Group–Breast Cancer (CARG-BC) score, in two cohorts of patients aged 65 years or older with stage I-III breast cancer.
The area under the curve for predicting grade 3-5 toxicity was 0.75 in the development cohort and 0.69 in the validation cohort, for a combined AUC of 0.73.
The CARG-BC score outperformed both Karnofsky performance status (AUC, 0.50) and the Cancer and Aging Research Group Chemotherapy Toxicity Tool (AUC, 0.56).
CARG-BC risk groups were also associated with hospitalizations, dose modifications, and early termination of treatment.
Allison Magnuson, DO, of the University of Rochester (N.Y.), and colleagues described these results in the Journal of Clinical Oncology.
About CARG-BC
To calculate a patient’s CARG-BC score, the researchers added up points assigned to eight independent predictors of grade 3-5 chemotherapy toxicity:
- Planned anthracycline use (1 point)
- Stage II or III disease (3 points)
- Planned treatment duration longer than 3 months (4 points)
- Abnormal liver function (3 points)
- Low hemoglobin level (3 points)
- A fall in the previous 6 months (4 points)
- Limited ability to walk more than 1 mile (3 points)
- Lack of social support (3 points)
Patients with scores of 0-5 have a low risk, those with scores of 6-11 have an intermediate risk, and those with scores of 12 or above have a high risk of grade 3-5 toxicity.
Patient characteristics and results
There were 283 patients in the development cohort and 190 in the validation cohort. There were no significant demographic, disease, or treatment differences between the cohorts.
All patients had a mean age of 70.5 years, 36.2% had stage I disease, 42.9% had stage II, and 20.9% had stage III disease. Three-quarters of patients were non-Hispanic White, and 99.4% were women. Roughly a third of patients had received an anthracycline-based regimen.
Overall, about a quarter of patients had an unplanned dose reduction (24%), dose delay (26%), stopped treatment early (24%), or were hospitalized during treatment (23%). All of these occurrences were more likely in intermediate- and high-risk patients versus low-risk patients (P < .001).
In the development cohort, 19% of low-risk patients, 54% of intermediate-risk patients, and 87% of high-risk patients developed grade 3-5 chemotherapy toxicity.
Compared with the 93 patients in the low-risk group, the odds of toxicity was almost 5 times greater for the 159 intermediate-risk subjects, and 28 times greater for the 30 high-risk subjects.
In the validation cohort, grade 3-5 toxicity rates were 27% in the low-risk group, 45% in the intermediate-risk group, and 76% in the high-risk group.
This study had its limitations, including that a majority of subjects (72.2%) had a college education, and the validation cohort was accrued from the same 16 institutions as the development cohort.
“Further validation in a more diverse population should be considered,” the investigators wrote.