|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 26, No 10 (April 1), 2008: pp. 1767-1768 © 2008 American Society of Clinical Oncology. DOI: 10.1200/JCO.2007.15.8212
In ReplyDivision of Oncology, Washington University School of Medicine, St Louis, MO
Cancer Research UK Clinical Trials Unit, Division of Cancer Sciences, Faculty of Medicine, University of Glasgow, The Beatson West of Scotland Cancer Centre, Glasgow, United Kingdom
UNC Institute for Pharmacogenomics and Individualized Therapy, University of North Carolina, Chapel Hill, NC
Department of Oncology, Imperial College London, Hammersmith Hospital, London, United Kingdom Branford et al correctly highlight that ethnic variability in allele frequencies can be a major source of variability between pharmacogenetic studies. Indeed, as we discuss in the original publication,1 population differences along with other differences between studies (eg, sample size, tumor type, and treatment regimen) are all possible causes for the lack of replication of previous findings. In the specific case highlighted by Branford et al concerning the lack of association with ABCB1 2677G>T/A and response, although the Gréen et al2 study was performed in a Swedish population, the allele frequencies between the two studies are very similar (Gréen et al3: G, 0.56; T, 0.42; A, 0.02 v Scottish Randomised Trial in Ovarian Cancer (SCOTROC1)1: G, 0.55; T, 0.42; A, 0.03) and reproducing the statistical analysis performed by Gréen et al3 did not identify any associations. It is unclear that population differences would play a role here, if ABCB1 2677G>A/T is indeed a functional polymorphism. Gréen et al4 have suggested that differences in patient treatment regimens may be one of the reasons for the lack of overlap between these studies. Race and ethnicity information was not collected for the SCOTROC1 trial. Although it would have been interesting to stratify the data and determine any population-specific associations, unfortunately no sample set will have the absolute ideal combination of data. To collect a large cross-section of the cancer population, trials such as SCOTROC1 are invaluable; however data collection in that study was necessarily focused on the clinical comparisons, which were the primary study objective. One of the main problems with previously published pharmacogenetics studies on taxanes is that the majority of studies are performed on fewer than 100 patients. The small sample sizes combined with low allele frequencies and multiple testing issues make it critical to validate such studies in large cohorts that represent the typical clinical setting. The majority of the polymorphisms selected for analysis were done so based on previously published associations, or using in silico or in vitro evidence that they were functional polymorphisms. In cases where previous associations are not based on true functional polymorphisms but rather markers in linkage disequilibrium with other functional variants, it is certainly plausible that population can play a large role. The amount of linkage can differ significantly between populations. Consequently, it is possible in some instances that a pharmacogenetic marker for one population does not translate to a different population. We agree that the pharmacogenetic study in the SCOTROC1 sample set was not applicable to identifying associations of small magnitude. However, we would argue that, although such associations would be interesting, their relevance in the clinical setting would be questionable. For a pharmacogenetic marker to have utility in the clinic, it needs to have both robust functional relevance, and it needs to be prevalent at a high enough frequency to warrant clinical testing. Moreover, allele frequencies for almost every clinically relevant polymorphism are not currently known in every single world population.5 Although it is well documented that population differences exist for many variants,6 it is inaccurate to assume global classifications of race are accurate reflections of allele frequencies. Even between populations that would fall under the same classification, significant differences can be seen, (eg, white populations and GSTT1*0 frequency).7 Consequently, to be truly applicable to the clinical setting, the focus needs to be on finding robust, functionally relevant polymorphisms that exert their effects independent of ethnicity. AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The author(s) indicated no potential conflicts of interest. ACKNOWLEDGMENTS Supported by NIH Exploratory/Developmental Research Grant Award (R21) CA113491, CRUK programme Grant No. C536/A2662, and the Pharmacogenetics Research Network Grant No. U01 GM63340. REFERENCES
1. Marsh S, Paul J, King CR, et al: Pharmacogenetic assessment of toxicity and outcome after platinum plus taxane chemotherapy in ovarian cancer: The Scottish randomised trial in ovarian cancer. J Clin Oncol 25:4528-4535, 2007 2. Gréen H, Soderkvist P, Rosenberg P, et al: Mdr-1 single nucleotide polymorphisms in ovarian cancer tissue: G2677T/A correlates with response to paclitaxel chemotherapy. Clin Cancer Res 12:854-859, 2006 3. Marsh S, King CR, McLeod HL, et al: ABCB1 2677G>T/A genotype and paclitaxel pharmacogenetics in ovarian cancer. Clin Cancer Res 12:4127, 2006 4. Gréen H, Peterson C, Soderkvist P: In Response. Clin Cancer Res 12:4127-4128, 2006 5. Marsh S, Van Booven DJ, McLeod HL: Global pharmacogenetics: Giving the genome to the masses. Pharmacogenomics 7:625-631, 2006[CrossRef][Medline] 6. Engen RM, Marsh S, Van Booven DJ, et al: Ethnic differences in pharmacogenetically relevant genes. Curr Drug Targets 7:1641-1648, 2006[CrossRef][Medline] 7. Garte S, Gaspari L, Alexandrie AK, et al: Metabolic gene polymorphism frequencies in control populations. Cancer Epidemiol Biomarkers Prev 10:1239-1248, 2001
Related Correspondence
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
