|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 23, No 16 (June 1), 2005: pp. 3863-3864 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.05.367
Pitfalls in the Use of RT-PCR As a Prognostic Indicator in MelanomaCancer Research Center of Hawaii, University of Hawaii, Honolulu, HI To the Editor: Kammula et al1 studied the prognostic significance of sentinel lymph node (SLN) staging using reverse-transcriptase polymerase chain reaction (RT-PCR) in patients with melanoma. They concluded that extended follow-up of patients, to a median duration of 67 months, rendered the RT-PCR status of histopathologically-negative SLNs irrelevant to patient prognosis. Although this study includes significantly longer patient follow-up when compared to many—though not all—earlier studies, the results of this important study should be interpreted with caution. First, 60 (54%) of the 112 patients studied underwent conventional hematoxylin and eosin (H&E) staining of their bisected SLNs, whereas the remaining 52 (46%) of patients' SLNs were assessed using serial-sectioning and H&E staining, followed by immunohistochemistry (IHC) staining with S-100 and HMB-45 when the H&E stains were negative for tumor cells. Nevertheless, the authors state that the incidence of positive nodes detected in each of these two groups of patients was statistically equivalent (15% v 12%). This is an unusual finding, in that IHC staining has been shown to upstage an additional 10% to 30% of melanoma patients' SLNs over H&E staining alone.2-4 Because an accurate assessment of the true histopathologic status of SLNs is critical in interpreting the prognostic value of SLN RT-PCR, the apparent absence of upstaging in the SLNs of patients evaluated with serial sections, H&E, and IHC (when H&E was negative) in this study is disconcerting. A second concern is the methodology used to perform the molecular analysis of the SLNs in this study. The authors state that the historical false-positive rate for the nested tyrosinase RT-PCR assay used in this study is 11%, which is quite high for a clinical RT-PCR study (a false-positive rate of 0% to 5% is standard). This high false-positive rate may arise, at least in part, from the use of a single RT-PCR marker, as well as the use of a nonquantitative RT-PCR assay and a gel-based PCR product detection system. The greater sensitivity, specificity, and reproducibility afforded by the use of quantitative real-time RT-PCR, when compared with the nonquantitative assay used in this study, has resulted in the abandonment of nonquantitative RT-PCR by the majority of laboratories working in this field. A related issue is this study's reliance on a single marker for the RT-PCR assay, and on tyrosinase in particular. The trend is clearly towards the use of multiple-marker RT-PCR assays, in view of heterogeneous tumor marker expression by human melanoma cells (observations of tyrosinase expression by melanoma cells, at a level detectable by nonquantitative RT-PCR, have ranged from 50% to 90%). A related concern is that tyrosinase is expressed by benign nevi cells as well as by melanoma cells. Among patients with melanoma, prior studies have shown that 4% to 28% of such patients will have detectable benign subcapsular nevi cells in their SLNs, resulting in a positive tyrosinase RT-PCR assay, even in the absence of metastatic melanoma cells.5-10 It is likely that the authors' high false-positive rate with their nested tyrosinase RT-PCR assay can be attributed to one or more of these limitations (contamination of samples upstream of the PCR reaction is also a common cause of false-positive results). A recent study of 308 paraffin-embedded SLNs, from 215 patients with clinically node-negative melanoma, compared intensive H&E/IHC analysis of serially-sectioned SLNs against a quantitative, real-time, multiple-marker RT-PCR assay, using the tumor-associated markers MART-1, MAGE-A3, GalNAc-T, and Pax3. Follow-up was in excess of 8 years for all patients. A total of 25% of patients had histopathologically positive SLNs (ie, positive by H&E or/and IHC). Among the patients with histopathologically negative SLNs, 30% were positive for at least one RT-PCR marker. The risk ratio for disease recurrence in the patients with histopathologically negative but RT-PCR-positive SLNs was 7.48 (P < .0001) when compared with patients with SLNs that were negative by both histopathologic evaluation and by RT-PCR. The risk ratio for decreased overall survival in the RT-PCR–positive group of patients (with histopathologically negative SLNs) was 11.42 (P = .0002). Finally, an increasing number of positive RT-PCR markers was associated with a decreased overall survival (P < .0001).11 These results comport well with other recent studies utilizing multiple-marker RT-PCR assays to evaluate the SLNs and blood of patients with melanoma, including patients with long-term follow-up.12 Despite these concerns, the authors are to be congratulated for their long-term study of patients with early stage melanoma, and for their important contributions to the emerging field of molecular tumor staging. Author's Disclosure of Potential Conflicts of Interest The author indicated no potential conflicts of interest. REFERENCES 1. Kammula US, Ghossein R, Bhattacharya, et al: Serial follow-up and the prognostic significance of reverse transcriptase-polymerase chain reaction-stage sentinel lymph nodes from melanoma patients. J Clin Oncol 22:3989-3996, 2004 2. Abrahamsen HN, Hamilton-Dutoit SJ, Larsen J, et al: Sentinel lymph nodes in malignant melanoma: Extended histopathologic evaluation improves diagnostic precision. Cancer 100:1683-1691, 2004[CrossRef][Medline] 3. Li W, Stall A, Shivers SC, et al: Clinical relevance of molecular staging for melanoma: Comparison of RT-PCR and immunohistochemistry staining in sentinel lymph nodes of patients with melanoma. Ann Surg 231:795-803, 2000[CrossRef][Medline] 4. Yu LL, Flotte TJ, Tanabe KK, et al: Detection of microscopic melanoma metastases in sentinel lymph nodes. Cancer 86:617-627, 1999[CrossRef][Medline] 5. Holt JB, Sangueza OP, Levine EA, et al: Nodal melanocytic nevi in sentinel lymph nodes: Correlation with melanoma-associated cutaneous nevi. Am J Clin Pathol 121:58-63, 2004[CrossRef][Medline] 6. Yan S, Brennick JB: False-positive rate of immunoperoxidase stains for MART1/MelanA in lymph nodes. Am J Surg Pathol 28:596-600, 2004[Medline] 7. Rimoldi D, Lemoine R, Kurt AM, et al: Detection of micrometastases in sentinel lymph nodes form melanoma patients: Direct comparison of multimarker molecular and immunopathological methods. Melanoma Res 13:511-520, 2003[CrossRef][Medline] 8. Biddle DA, Evans HL, Kemp BL, et al: Intraparenchymal nevus cell aggregates in lymph nodes: A possible diagnostic pitfall with malignant melanoma and carcinoma. Am J Surg Pathol 27:673-681, 2003[CrossRef][Medline] 9. Gutzmer R, Kaspari M, Brodersen JP, et al: Specificity of tyrosinase and HMB45 PCR in the detection of melanoma metastases in sentinel lymph node biopsies. Histopathology 41:510-518, 2002[CrossRef][Medline] 10. Bautista NC, Cohen S, Anders KH: Benign melanocytic nevus cell in axillary lymph nodes: A prospective incidence and immunohistochemical study with literature review. Am J Clin Pathol 102:102-108, 1994[Medline]
11. Takeuchi H, Morton DL, Kuo C, et al: Prognostic significance of molecular upstaging of paraffin-embedded sentinel lymph nodes in melanoma patients. J Clin Oncol 22:2671-2680, 2004
12. Robert A: Wascher, Donald L. Morton, Christine Kuo, et al: Molecular tumor markers in the blood: Early prediction of disease outcome in melanoma patients treated with a melanoma vaccine. J Clin Oncol 21:2558-2563, 2003
Related Reply
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
