| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIALS |
From the Division of Surgical Oncology, University of Louisville, James Graham Brown Cancer Center, Louisville, Kentucky.
Correspondence: Address correspondence to: Kelly M. McMasters, MD, PhD, Division of Surgical Oncology, University of Louisville, James Graham Brown Cancer Center, 315 E. Broadway, Suite 308, Louisville, KY 40202; Fax: 502-629-3393; E-mail: kelly.mcmasters{at}nortonhealthcare.org
In this issue of the Annals of Surgical Oncology, Ribuffo et al.1 report their experience with reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of sentinel lymph nodes (SLN) for molecular nodal staging of patients with melanoma. Tyrosinase and MART-1 were used as the markers for RT-PCR analysis. The authors define a positive RT-PCR test as the presence of either tyrosinase or MART-1. In an analysis of 134 patients, 11% had microscopically positive SLN by routine histology or immunohistochemistry (IHC) evaluation, 52% had RT-PCR–positive SLN that were negative by routine histology and IHC, and only 37% were negative by all tests. Overall, with a median follow-up of 42 months, there were no recurrences in the RT-PCR–negative group, whereas the group with nodes positive only by RT-PCR had an intermediate disease-free survival rate between those with histologically/IHC-positive SLN and those negative by all tests. This is similar to the results reported from the Moffitt Cancer Center, the John Wayne Cancer Institute, and other groups in recent years (as cited in the article).
The sensitivity of the RT-PCR testing, as defined in this study, is excellent, since none of the patients with RT-PCR–negative SLN have recurred. The main problem, however, is one of specificity. Overall, 85 of 134 patients (63%) were found to have histologic or molecular evidence of nodal metastasis. This vastly overestimates the fraction of patients that would be expected to recur, or develop nodal metastases, in a population with a median Breslow thickness of 1.82 mm. In fact, there were five additional patients with tyrosinase-positive SLN who were excluded from the RT-PCR–positive group because histological analysis revealed benign nevus cells in the SLN. If these patients were included in the RT-PCR–positive group, fully two thirds of the patient population would have evidence of nodal metastasis! The real question, therefore, is how many of the RT-PCR–positive patients have benign nevus cells that were not detected histologically? Was this the source of such a high percentage of patients with RT-PCR–positive SLN? It does not seem appropriate to exclude these patients, post hoc, from the analysis, since any test will have to take into account the presence of occult benign nevus cells as a potential source of false-positive test results. It also would be instructive to know the patterns of recurrence in this patient population. How many had local or in-transit recurrence versus regional nodal or distant recurrence? Does an RT-PCR–only positive SLN predict a higher risk of nodal metastasis or is it a marker for risk of distant metastasis?
This study, while providing further evidence that RT-PCR analysis eventually may be a useful tool for staging of melanoma patients, also points out its limitations. A diagnostic test must be sensitive but not overly sensitive. Although a negative RT-PCR test may comfort some patients to know that their prognosis is likely to be excellent, at the present time this does not outweigh the harm that comes from telling over half of the patients that they have evidence of cancer cells in their lymph nodes, especially when the vast majority of them are likely cured of their melanoma. Therefore, in order for RT-PCR testing to be clinically useful, it must have greater specificity. The authors used the presence of tyrosinase or MART-1 as their definition of a positive test. MART-1 was used to add some specificity to the assay. However, it appears that nearly all patients positive for tyrosinase were also MART-1 positive and vice versa. Other markers may provide better discrimination between clinically relevant and irrelevant RT-PCR–detected messenger RNA (mRNA) expression.
Several important issues remain to be resolved before RT-PCR analysis can be used for clinical decision-making. First, the conditions and detection methods for RT-PCR analysis vary greatly from laboratory to laboratory. Standardization and optimization of techniques are necessary to compare results. Second, the best combination of markers to obtain the optimal balance between sensitivity and specificity has yet to be defined. It is especially important to determine the combination of markers that will exclude the possibility of a false-positive RT-PCR test based on the presence of benign nevus cells, which may or may not be seen histologically. For example, there is no reason that some patients with benign nevus cells might not also have the presence of melanoma cells. The test must discriminate between the two. Third, the value of quantitative or semiquantitative analysis of mRNA expression to add specificity to the assay must be further evaluated. Newer techniques, such as real-time quantitative PCR, may improve the results. Finally, the value of additional treatment for patients with RT-PCR–positive SLN (i.e., lymph node dissection and adjuvant therapy with interferon alfa-2b) has yet to be established. This is one of the goals of the ongoing Sunbelt Melanoma Trial,2 which should help answer such questions. Until such information is available, RT-PCR analysis of SLN should not be used outside of a clinical trial. Remember the words of your old medical school professor who advised you not to order a test unless you knew what you would do with the results. Right now, we do not know what to do about an RT-PCR–positive SLN.
Received for publication February 20, 2003. Accepted for publication March 2, 2003.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  H. N. Abrahamsen, B. Sandahl Sorensen, E. Nexo, S. J. Hamilton-Dutoit, J. Larsen, and T. Steiniche Pathologic Assessment of Melanoma Sentinel Nodes: A Role for Molecular Analysis Using Quantitative Real-Time Reverse Transcription-PCR for MART-1 and Tyrosinase Messenger RNA Clin. Cancer Res., February 15, 2005; 11(4): 1425 - 1433. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. N. Abrahamsen, E. Nexo, T. Steiniche, S. J. Hamilton-Dutoit, and B. S. Sorensen Quantification of Melanoma mRNA Markers in Sentinel Nodes: Pre-Clinical Evaluation of a Single-Step Real-Time Reverse Transcriptase-Polymerase Chain Reaction Assay J. Mol. Diagn., August 1, 2004; 6(3): 253 - 259. [Abstract] [Full Text]
|
|
|
|
 |
|
 K. J. Busam Advances in Molecular Staging of Melanoma Patients: Multimarker Analysis of Archival Lymph Node Tissue J. Clin. Oncol., October 1, 2003; 21(19): 3550 - 3551. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
