| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Correspondence |
The Sahlgrenska Academy at Göteborg University Göteborg, Sweden
To the Editor-in-Chief:
Recent studies have shown that mucoepidermoid carcinomas (MECs) of the salivary and bronchial glands are characterized by a recurrent t(11;19)(q21;p13) translocation, resulting in a MECT1-MAML2 fusion in which the cAMP response element binding protein (CREB)-binding domain of the CREB-regulated transcriptional coactivator MECT1 (also known as CRTC1, WAMTP1, or TORC1) is fused to the transactivation domain of the Notch coactivator MAML2.1, 2 The molecular consequences of the fusion are still largely uncertain. Earlier studies have indicated that MECT1-MAML2 can activate Notch signaling independent of ligand stimulation,1, 2 and recent studies have demonstrated that MECT1-MAML2 can activate transcription of multiple cAMP/CREB target genes in vitro and that the N-terminal CREB-binding domain is crucial for the transforming activity of the fusion protein.3, 4 Collectively, these studies indicate that MECT1-MAML2 is a multifunctional protein with effects on several signaling pathways.
Interestingly, recent studies have indicated that the MECT1-MAML2 fusion may not exclusively occur in MEC but may also be found in other types of glandular tumors. Thus, we have identified the fusion in two apparently unrelated benign tumor types: Warthin’s tumor (WAT) and clear cell hidradenoma of the skin.2, 5 In this respect, MECT1-MAML2 is reminiscent of the ETV6-NTRK3 fusion that has been encountered in tumors originating from multiple cell lineages.6
We read with interest the study by Martins et al7 in The Journal of Molecular Diagnostics in which they studied the MECT1-MAML2 fusion in 10 MECs and 7 WATs using fluorescence in situ hybridization (FISH) and/or reverse transcriptase-polymerase chain reaction. The fusion was detected in seven MECs but in none of the WATs. It should be noted that none of the classical WATs in this study had a t(11;19) translocation. However, one of their translocation- and fusion-positive MECs (case 10) was of particular interest because it was originally diagnosed and published as a benign WAT.8 This tumor was now reclassified as a MEC ex WAT. As indicated by the authors, it is known that MECs in rare instances may arise from or coexist with WAT. Case 10 in their series was suggested to represent such a case. The authors conclude that the MECT1-MAML2 fusion appears to be rare in WATs that do not contain concomitant evidence of MEC and that the fusion might be a useful diagnostic marker for MEC, especially in histologically challenging cases.
The observations by Martins et al prompted us to re-examine our previously published t(11;19)- and fusion-positive WAT2, 9 both histologically and molecularly. Analysis of tissue sections from multiple levels of this tumor confirmed the original diagnosis of a benign WAT. Despite extensive sampling, we found no areas reminiscent of MEC in any of the sections examined.
To further characterize this tumor, we also performed FISH on paraffin sections using a dual-color break-apart rearrangement probe for MAML2. Rearrangements of MAML2, indicating a MECT1-MAML2 fusion, were only detected in the morphologically benign tumor cells lining the duct-like structures and cysts of the tumor (Figure 1A)
. Split signals were not observed in the surrounding lymphoid cells, supporting the concept that these cells are non-neoplastic stromal cells.
|
. The surrounding lymphoid cells were mainly negative. Taken together, our results show that the present tumor represents a benign WAT with a MECT1-MAML2 gene fusion. Available data do, however, indicate that the fusion presumably is rare in WAT.7 In addition to the present case, there is only one previously reported WAT with a t(11;19)(q21;p13) translocation and one case with a possible variant translocation t(11;16)(q13–14;q23).11, 12 The fact that the MECT1-MAML2 fusion may also infrequently occur in WAT is important to consider when using the fusion as a diagnostic marker for MEC. This is particularly important in metaplastic variants of WAT with atypia and mitotic activity, which may simulate MEC.13
Our findings of the MECT1-MAML2 fusion in two unrelated benign tumor types are of particular interest in relation to recent data showing that MECT1-MAML2 preferentially occurs in low-grade MECs and that fusion-positive patients have a significantly lower risk of local recurrence, metastasis, or tumor-related death compared with fusion-negative patients.10 When considering tumor-related deaths only, the estimated median survival for fusion-positive patients was found to be >10 years compared with 1.6 years for fusion-negative patients. Taken together, these observations suggest that the MECT1-MAML2 fusion is preferentially associated with benign and low-grade malignant tumors of glandular origin and that the fusion may be a useful marker in predicting the biological behavior of MECs.10 Additional studies of larger series of tumors are, however, needed to further substantiate these observations.
Acknowledgments
This study was supported by grants from the Swedish Cancer Society and the IngaBritt and Arne Lundberg Research Foundation.
References
Instituto Português de Oncologia Francisco Gentil Lisboa, Portugal
National Cancer Institute and National Naval Medical Center Bethesda, Maryland
Authors’ Reply:
Winnes et al have previously reported the detection of MECT1-MAML2 expression in a case of Warthin’s tumor (WAT),1 and in the present correspondence they have restated their confidence in the histological diagnosis of this case and propose that the MECT1-MAML2 fusion oncogene is not specific to mucoepidermoid-like (MEC-like) tumors. Although we cannot comment on the diagnosis of an isolated case of WAT based on a single photomicrograph, we would like to make two points to clarify what we believe to be the best interpretation of the emerging biology of MECT1-MAML2-associated tumorigenesis.
First, we would like to emphasize that malignancies with a MEC-like histological pattern have been detected as primary lesions in a wide range of anatomical locations, including 1) each of the major and minor salivary glands, 2) the upper and lower respiratory tract, and 3) rare cases involving mandible, thyroid, breast, esophagus, stomach, pancreas, skin, and other sites.2 3 4 5 6 7 8 9 Although the majority of these cases of MEC-like tumors have not yet been tested for the presence of MECT1-MAML2 gene expression, it is quite remarkable that identical MECT1-MAML2-positive cases have already been reported in tumors arising from parotid, submaxillary, oral, and many different thoracic sites within the bronchopulmonary tree.10 11 12 In addition, we and others have noted the histological similarities between salivary MEC and certain cutaneous MEC-like tumors, which can include some cases of clear cell hidradenomas or eccrine acrospiroma.9 13 Therefore, the detection of the identical MECT1-MAML2 transcript in a cutaneous clear cell hidradenomas14 may suggest a common histological, genetic, or biological origin that serves to unify the concept of a MECT1-MAML2-associated tumorigenesis pathway.
Second, we would like to suggest that the otherwise histogenetically distinct entity called WAT is not associated with MECT1-MAML2 gene expression. We base this conclusion on the absence of MECT1-MALML2 expression in seven different WATs cases11 and on the absence of MECT1-MAML2 in 26 different WATs in a separate study by Okabe et al.15 Although it is partly semantic to argue the fine distinction between the reality of "no occurrence" versus "a very rare occurrence," we prefer the interpretation that an outlier sample of a MECT1-MAML2-positive non-MEC salivary gland tumor is more likely a case of misdiagnosis. This is particularly likely given the coincident presence of MEC and WAT within the same tumor samples.16
Finally, we agree that the ETV6-NTRK3 (EN) chimera represents an interesting example where the same EN fusion can result in either pediatric mesenchymal tumors or secretory breast cancer of adults and children.17 Accordingly, it is possible that MECT1-MAML2 could also be associated with histologically dissimilar tumors. Although much remains to be learned about the biology of MECT1-MAML2 tumorigenesis, the emerging data to date suggest that MECT1-MAML2 is etiologically linked with a subset of MEC-like tumors and not with WAT.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
