|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 23, No 9 (March 20), 2005: pp. 2108-2109 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.05.268
In Reply:Institute of Cancer Research, Sutton, Surrey, United Kingdom We welcome the comments by Drs Vincenzi, Santini, and Tonini related to our previously published study1 and acknowledge that patients with colorectal cancer (CRC) now have an unprecedented number of therapeutic options available, with clearly effective first- and second-line treatments.2,3 However, we disagree with the authors on the use of progression-free survival (PFS) over overall survival (OS) as a suitable end point for the assessment of biologic markers of prognosis in CRC patients with advanced disease. Our analysis was based on the relationship between thymidylate synthase (TS) expression and prognosis, and in the advanced setting did not have the ability to assess the role of TS expression in predicting outcome on the basis of any specific drug, per se. Although the adoption of PFS (or even response rates) as a primary end point in this setting would be entirely justified to investigate the ability of a biologic parameter to predict therapeutic outcome, this was not the case in our analysis. Moreover, to assess the utility of a molecular marker to predict therapeutic effect, a direct comparison of the relative hazards (or risk in case of response rates) for patients with the marker, stratified by presence/absence of therapy in question, compared with those for patients who do not possess the marker, is required.4 This was not within the scope of our analysis, nor was it possible given the data presented. However, we agree with the authors and reiterate our assessment that conclusions based on the pooled PFS data on patients with advanced disease should be interpreted with caution, given the small number of contributing studies and the heterogeneity observed, which might in part reflect the origin of tissue for TS analysis. The authors suggest that differences in therapy administered to patients between the two TS patient groups identified in each of the contributing data sets, particularly in the adjuvant setting, might have biased our findings. Such bias is, however, only likely to have been significant if treatments were nonrandomly administered to patients on the basis of TS status. This is therefore unlikely to have impacted greatly on our findings. Moreover, of the seven contributing studies based in the adjuvant setting, three5-7 were based on a mixture of colon or rectal carcinoma patients, and in two,5,7 the datapoints used were based on multivariate analysis. While some data suggest that sequence variants in the TS gene (TYMS) might possibly play a role in down-staging rectal carcinoma,8 these are extremely unlikely to have influenced our conclusions. TYMS genotypes investigated in this context are thought to be functional.9,10 Since TS level was the primary variable by which survival was stratified, the role of such genotypes as potential confounders is therefore moot. We do, however, acknowledge the more general issue that in contrast to meta-analyses of individual-patient data, those meta-analyses based on abstracted data are limited in their inability to fully explore the role of potential confounders. Our study has adequately demonstrated the marked methodologic heterogeneity between studies that have assessed the relationship between TS status and prognosis, in both adjuvant and advanced disease settings. We identified that high TS expression seems to be associated with a significantly poorer OS, but heterogeneity and publication bias contribute to this conclusion. Finally, we agree with the authors that further work is required before routinely integrating TS expression into treatment paradigms, and we reiterate our call for prospectively analyzed, blinded, sufficiently powered studies, utilizing consistent methodology, and nonbiased thresholds. Authors' Disclosures of Potential Conflicts of Interest The authors indicated no potential conflicts of interest. REFERENCES
1. Popat S, Matakidou A, Houlston RS. Thymidylate synthase expression and prognosis in colorectal cancer: A systematic review and meta-analysis. J Clin Oncol 22:529-536, 2004
2. Hurwitz H, Fehrenbacher L, Novotny W, et al: Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350:2335-2342, 2004
3. Cunningham D, Humblet Y, Siena S, et al: Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 351:337-345, 2004
4. Ribic CM, Sargent DJ, Moore MJ, et al: Tumor microsatellite-instability status as a predictor of benefit from fluorouracil-based adjuvant chemotherapy for colon cancer. N Engl J Med 349:247-257, 2003 5. Yamachika T, Nakanishi H, Inada K, et al: A new prognostic factor for colorectal carcinoma, thymidylate synthase, and its therapeutic significance. Cancer 82:70-77, 1998[CrossRef][Medline]
6. Allegra CJ, Parr AL, Wold LE, et al: Investigation of the prognostic and predictive value of thymidylate synthase, p53, and Ki-67 in patients with locally advanced colon cancer. J Clin Oncol 20:1735-1743, 2002
7. Edler D, Glimelius B, Hallstrom M, et al: Thymidylate synthase expression in colorectal cancer: A prognostic and predictive marker of benefit from adjuvant fluorouracil-based chemotherapy. J Clin Oncol 20:1721-1728, 2002
8. Villafranca E, Okruzhnov Y, Dominguez MA, et al: Polymorphisms of the repeated sequences in the enhancer region of the thymidylate synthase gene promoter may predict downstaging after preoperative chemoradiation in rectal cancer. J Clin Oncol 19:1779-1786, 2001 9. Kawakami K, Omura K, Kanehira E, et al: Polymorphic tandem repeats in the thymidylate synthase gene is associated with its protein expression in human gastrointestinal cancers. Anticancer Res 19:3249-3252, 1999[Medline] 10. Ishida Y, Kawakami K, Tanaka Y, et al: Association of thymidylate synthase gene polymorphism with its mRNA and protein expression and with prognosis in gastric cancer. Anticancer Res 22:2805-2809, 2002[Medline]
Related Correspondence
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
