|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2009.23.7644 on July 27 2009 © 2009 American Society of Clinical Oncology.
Gene Expression Profiling in Cancers of Unknown PrimaryDepartment of Gastrointestinal Medical Oncology, The University of Texas M. D. Anderson Cancer Center, Houston, TX To the Editor: Monzon et al1 should be congratulated for their multicenter study and validation of a 1,550-gene expression profile for tissue of origin determination for cancers of uncertain primary particularly since they targeted the difficult group of poorly differentiated and undifferentiated tumors. As physicians who take care of patients with cancer of unknown primary (CUP) and who believe that expression profiling promises to be an important tool for the management of these patients, we are particularly pleased at the continued maturation of this field. Although Monzon et al1 was not focused on CUP, and one has to be careful about extrapolating data from profiling of metastases of known cancers (even those with poorly differentiated and undifferentiated primaries) to patients with CUP, we felt it would be helpful to share our experience with CUP profiling and point out some of the pitfalls in applying this and other gene expression studies to patients with CUP. It is interesting to note that Monzon et al1 have chosen to steer clear of the CUP terminology. Gene expression profiling is an emerging diagnostic tool for uncertain cancers, though to include CUP in the category of "uncertain primary" is not entirely accurate since with CUP, despite extensive diagnostic testing, a primary cancer is usually not identified during the lifetime of the patient, and we remain uncertain whether there is something fundamentally different in CUP cancers versus those where the primary is obvious or somewhat difficult to pathologically define based on its unique features.2
Sample acquisition and processing for profiling studies has always been a challenge in the clinic, and Monzon et al1 emphasizes this issue. Excluding the 271 tumors for which the group had electronic microarray data, they had access to 371 tumor samples where Another issue that must be considered in developing any profiling algorithm is the choice of tissues for the training set and the validation study (Table 2 of Monzon et al1). We commend Monzon et al for using over 25 samples for each tumor type to represent their validation study, which is a significant improvement over other studies. One caveat from the CUP perspective is that the training set is over-represented by cancers that are not relevant to CUP (eg, prostate, lymphoma, thyroid, and melanoma) because with adequate pathology including robust IHC and molecular studies for lymphoma, these primaries are not commonly encountered in the CUP clinic. The training set is also over-represented by cancers for which there is reasonably reliable IHC available (eg, colon, breast). As CUP physicians, we struggle with the decision to order profiling outside a clinical trial where the IHC results are consistent, and the expression profiling results are expected to be concurrent (eg, a cancer that is cytokeratin 20+/cytokeratin 7- and CDX2+ suggesting a tissue of origin diagnosis of colon3). Conversely, if profiling results suggest a specific tissue of origin for a patient with CUP where IHC is not helpful, the information may be useful to plan initial management. However, we must be mindful of the fact that, at the moment, validation of the tissue of origin for patients with CUP can only be based on clinical parameters such as response to a specific chemotherapy regimen or on the patient's clinical outcome. Clinical trials that address these issues in patients with CUP are warranted and hopefully will help us understand the value of these tests in managing this challenging group of patients. Finally, as a general point, the cost effectiveness of these newer tests remains to be determined. Related to this issue is the need for new reimbursement models for diagnoses made solely by gene expression profiling tests, and for drugs that would be recommended off label based on profiling results. We are optimistic about these new tools and their ability to tailor therapy for our patients with CUP, though we also believe that oncologists have a responsibility to first understand whether any test we order will ultimately impact management decisions or the survival of our patients. We applaud these efforts that may help change the paradigm for management of CUP cancers, though we ask that physicians and patients use caution in the broad application of these results to patients with CUP since much more work needs to be done. AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a "U" are those for which no compensation was received; those relationships marked with a "C" were compensated. 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 or Leadership Position: None Consultant or Advisory Role: Gauri R. Varadhachary, Veridex (C) Stock Ownership: None Honoraria: None Research Funding: Gauri R. Varadhachary, Veridex, Rosetta Genomics Expert Testimony: None Other Remuneration: None REFERENCES
1. Monzon FA, Lyons-Weiler M, Buturovic LJ, et al: Multicenter validation of a 1,550-gene expression profile for identification of tumor tissue of origin. J Clin Oncol 27:2503–2508, 2009. 2. Pavlidis N, Fizazi K: Cancer of unknown primary (CUP). Crit Rev Oncol Hematol 54:243–250, 2005.[Medline] 3. Varadhachary GR, Talantov D, Raber MN, et al: Molecular profiling of carcinoma of unknown primary and correlation with clinical evaluation. J Clin Oncol 26:4442–4448, 2008.
Related Article
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2009 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
0.1 g of frozen tissue was available. Even with the availability of sufficient tissue, the drop-out rate was 26% (90 samples). The problem of gaining access to sufficient amounts of tissue becomes even more of a challenge when less tissue is available from patients who have a pathologic diagnosis made using small core-needle biopsies. This is particularly true with (necrotic) liver or bone metastases, which are not unusual presentations for CUP. Monzon et al feel that since their assay requires only 1 µg of total RNA, this quantity is obtainable from core-needle biopsies. Our experience suggests that acquisition of adequate tissue may still be a challenge. In practice a substantial part of the tumor block is utilized for routine pathologic stains and immunohistochemistry (IHC), and some tissue must be saved for directed IHC studies that may be necessary to validate the microarray profile–generated tissue of origin diagnosis. The need for adequate tissue for both IHC and expression profiling will probably not be completely eliminated in the foreseeable future, even as the expression array studies evolve. Therefore, the time and costs of repeat biopsy, which may often be needed where tissue is limiting, should be kept in mind. Recognizing the fact that many patients have a tissue diagnosis established before referral to a medical oncologist or a tertiary center and snap-frozen tissue sampling is not easily accomplished in clinical practice, we are anxious to review the Pathworks Tissue of Origin Test (Pathwork Diagnostics, Sunnyvale, CA) results on formalin-fixed paraffin-embedded tissue and how it compares with their current results. 
