Annals of Oncology Advance Access originally published online on April 19, 2006
Annals of Oncology 2006 17(9):1468-1469; doi:10.1093/annonc/mdl076
© 2006 European Society for Medical Oncology
letters to the editor |
Correlation between anemia and functional/cognitive capacity in elderly lung cancer patients treated with chemotherapy
1 Friedrich Schiller Universität, Department Haematology and Oncology, Jena
2 Ruhr-Universität, Department Geriatric Medicine, Herne, Germany
*(E-mail: Ulrich.wedding{at}med.uni-jena.de)
We respond to the report by Mancuso et al. [1]. They report on a correlation between anemia and results of a comprehensive geriatric assessment (CGA) and find a strong correlation prior to the first and the following chemotherapy cycles and for changes in Hb. In addition 14 of the 42 patients received erythropoietin and had an improvement in some items of CGA. Their subject is of major importance, but we have to make some comments on why the data should be considered with caution.
- (a) With 81% of their patients having a PS of 0–1 the patients are a selected group of patients and may be due to referral bias. In our geriatric oncology program only 54.9% of the patients aged 70+ had a PS of 0–1.
- (b) There are a number of methods for the reporting of results of co-morbidity measured by CIRS-G score: sum score, number of affected organ systems, number of affected organ systems level 3–4, etc. [2]. We miss an explanation, which method was used.
- (c) Hemoglobin values are part of co-morbidity assessment within the CIRS score. Therefore the reported correlation between Hb and CIRS is part of the kind of co-morbidity measurement used [3].
- (d) The authors report that 69% of the patients received first line chemotherapy, report functional impairment and cognitive impairment for the whole group, but conclude that chemotherapy-related anemia is associated with impairment of functional status and cognitive function. At least in a considerable number of patients, the anemia is tumor, but not chemotherapy related.
- (e) Re-scoring of co-morbidity after one or two cycles of chemotherapy does not measure co-morbidity but toxicity, which can be part of CIRS score as well, e.g. chemotherapy induced anaemia, renal toxicity.
- (f) To report correlation of n = 167 measurements in 42 patients means that results of nearly four measurements per patient in the course of time were included in the calculation of R. But as the results of repeated measurement are not independent, the information is as valuable as if they would have been reported per patient and not per measurement.
- (g) Changes in VAS for quality of life and for fatigue can be caused by response or non-response to treatment. Therefore, this should be included in a thorough analysis.
- (h) In 14 patients erythropoietin was given and in 18 patients not. What happened to the 10 patients not mentioned?
- (i) An obvious difference in the change of CGA variables depending on administration of erythropoietin is demonstrated with the calculation of mean and standard deviation. For the calculation of Spearman's correlation coefficient the patients are analyzed together. This implies the problem of spurious correlation due to unequal groups. A solution could be to calculate the partial correlation coefficient, this means correlation between Hb and CGA values adjusted for administration of erythropoietin.
- (b) There are a number of methods for the reporting of results of co-morbidity measured by CIRS-G score: sum score, number of affected organ systems, number of affected organ systems level 3–4, etc. [2]. We miss an explanation, which method was used.
The authors conclude that the results need to be confirmed in larger controlled trials. We consider our remarks as a help to improve this trial.
 |
References |
|---|
 Top References |
|---|
1. Mancuso A, et al. (2006) Correlation between anemia and functional/cognitive capacity in elderly lung cancer patients treated with chemotherapy. Ann Oncol 17:1146–150.
2. Extermann M. (2000) Measuring comorbidity in older cancer patients. Eur J Cancer 36:4453–471.[CrossRef][Web of Science][Medline]
3. Miller MD, et al. (1992) Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale. Psychiatry Res 41:3237–248.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||