|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 25, No 9 (March 20), 2007: pp. 1141-1142 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.09.1801
In ReplyDivision of Surgical Oncology, Department of Surgery, James Graham Brown Cancer Center, University of Louisville, Louisville, KY
Department of Statistics, University of Kentucky, Lexington, KY
Advertek Inc, Louisville, KY We are pleased to respond to the intriguing queries raised by Quaglino et al regarding our recently published manuscript on reverse-transcriptase polymerase chain reaction (PCR) analysis of sentinel lymph nodes (SLN) and peripheral blood mononuclear cells (PBMC) for patients enrolled onto the Sunbelt Melanoma Trial.1 We have performed extensive additional statistical analysis of this patient population, now with a median follow-up of 54 months. We analyzed the data by individual markers and by the number of positive markers. There was little evidence, in any analysis, that the markers MART 1, Mage 3, or GP 100 added useful prognostic information. In the updated analysis, however, tyrosinase as a single marker demonstrated prognostic significance for prediction of disease-free survival (DFS) but not overall survival (OS), when the entire patient population was analyzed (Table 1).
In response to the specific queries, we first used the 2002 AJCC Staging System2 to stratify patients who underwent PBMC PCR analysis (N = 838) into stages I, II, and III. We found that for stage I (n = 353) and II (n = 266) patients, there were no differences in DFS or OS between patients with negative or positive PBMC PCR tests (Table 1). Stage III melanoma patients with a positive PCR test for tyrosinase demonstrated significantly worse DFS, but not OS, when compared with those patients with a negative PCR result (P = .0007). In addition, those stage III patients who had multiple positive PBMC markers (n = 7) had a worse DFS (P = .0002) and OS (P = .0012) when compared with patients with either zero or one positive marker, although the clinical significance of this finding may be limited considering the small number of patients who had more than one positive marker. Second, as demonstrated by Osella-Abate et al,3 when we analyzed patients with an initial baseline PBMC PCR result who also had at least one follow-up PCR result, there were no significant differences in DFS (P = .2674) or OS (P = .3484) between patients who entered the Sunbelt Melanoma Trial with evidence of circulating melanoma cells versus those who were initially negative and then became positive by PCR. Finally, we analyzed PBMC PCR for tyrosinase as a time-dependent covariate. As in the study by Quaglino et al,4 a proportional hazards model for DFS with time-dependent covariate was fit to the data set. As with their data, tyrosinase analyzed as a time-dependent covariate was significant (P < .001), but we also found sentinel lymph node status (P = .001), sex (P = .013), ulceration (P = .010), and Breslow thickness (P < .001) to be significant, whereas they did not. In our data, the risk ratio was 5.5 for tyrosinase as a time-dependent covariate, suggesting that a patient with an early positive reverse transcriptase (RT)-PCR result would have a better DFS rate than a patient who subsequently developed a positive RT-PCR result for tyrosinase. Possible explanations include: (1) the possibility that resection of the primary melanoma (and in some cases tumor-involved lymph nodes) eliminated the source of circulating tumor cells, (2) an effect of adjuvant therapy (which we can analyze when the final study results are available), or (3) a statistical phenomenon that has no biologic relevance. In conclusion, our updated analysis with a follow-up of 54 months demonstrates that a positive PBMC tyrosinase test is predictive of worse DFS (especially among stage III patients), whereas OS has not reached statistical significance. The finding of more than one positive PBMC marker is prognostically significant, although not of great clinical importance, because this applies to a very small number of patients. Additional investigation is necessary before PBMC PCR testing can be recommended as a routine staging procedure for patients with melanoma. AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Although all authors completed the disclosure declaration, the following author or immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors. Employment: N/A Leadership: N/A Consultant: Kelly M. McMasters, Johnson & Johnson Stock: N/A Honoraria: Kelly M. McMasters, Johnson & Johnson Research Funds: Kelly M. McMasters, Schering Testimony: N/A Other: N/A
REFERENCES
1. Scoggins CR, Ross MI, Reintgen DS, et al: Prospective multi-institutional study of reverse transcriptase polymerase chain reaction for molecular staging of melanoma. J Clin Oncol 24:2849-2857, 2006 2. Greene FL, Page DL, Fleming ID, et al: AJCC Cancer Staging Manual (ed 6). New York, NY, Springer, 2002 3. Osella-Abate S, Savoia P, Quaglino P, et al: Tyrosinase expression in the peripheral blood of stage III melanoma patients is associated with a poor prognosis: A clinical follow-up study of 110 patients. Br J Cancer 89:1457-1462, 2003[CrossRef][Medline] 4. Quaglino P, Savoia P, Osella-Abate S, et al: RT-PCR tyrosinase expression in the peripheral blood of melanoma patients: From research bench to clinical practice. Expert Rev Mol Diagn 4:727-741, 2004[CrossRef][Medline] Related Correspondence
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
