| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIAL |
From the M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Box 106, Houston, Texas.
Correspondence: Address correspondence to: Raphael Pollock, MD, M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Box 106, Houston, TX 77030; Fax: 713-792-4689; E-mail: rpollock{at}notes.mdacc.tmc.edu
The natural history of soft tissue sarcoma is well established, at least in its broadest strokes. Comprising more than 20 distinct histologic subtypes, it is known that these rare tumors of putative connective tissue origin metastasize readily as a function of size, grade, and primary tumor location. Eighty percent of metastases are pulmonary, and 80% of metastases occur within 2 years of initial treatment. However, an appreciable number of patients will fail late in their disease course, and sarcoma recurrence follows the path of an exponential decay curve with time as the x-axis representing an asymptote such that the risk of failure never entirely disappears.1
This issue of the Annals of Surgical Oncology contains excellent reports from two of the leading sarcoma centers in the world, the Royal Marsden Hospital in London and the Memorial Sloan-Kettering Cancer Center in New York. Although ostensibly different in their scope and focus, both articles touch on several common themes that may ultimately improve our understanding about the genes and their cognate proteins that drive the sarcomagenesis and recurrence processes. As a parenthetical aside, both reports are first authored by fellows in training, a hallmark of so many of the advances as has been reported in the sarcoma surgical oncology literature.
Ramanathan et al.2 retrospectively considered 110 patients with local only recurrent extremity soft tissue sarcoma referred to the Royal Marsden Hospital between 1989 and 1995. The impact of various considerations on local recurrence, distant metastasis, and disease-free survival was analyzed using standard regression analysis techniques. Recurrent tumor size, recurrent tumor grade, and time to recurrence after initial tumor treatment were identified as independent prognostic factors for subsequent distant metastases and disease-specific survival. A prognostic index was developed by which each factor was rated on a 1–3 scale and the three individual factor scores were then summed. One can haggle about an ordinate qualitative scale for continuous versus discontinuous variables. However, as the prognostic index value increased there were statistically robust comparable decreases in sarcoma-specific survival. On the basis of this finding, the authors suggest that the nature of the local recurrence, rather than its occurrence, is critical in generating meaningful prognostic information. The ease in application of this three-factor system is easy to apply, and the authors recommend that this algorithm be incorporated into staging-based decisions regarding specific therapeutic interventions being contemplated on behalf of patients afflicted with recurrent sarcoma.
This is an interesting recommendation, and speaks to the reality that the current AJCC sarcoma staging system is effective for the majority of sarcomas, which happen to be de novo extremity lesions, but works less well for recurrent extremity disease, as well as for primary sarcoma of the head and neck, retroperitoneum, intra-abdominal cavity, and to some extent even truncal soft tissue sarcoma. These discrepancies will be considered by the Soft Tissue Sarcoma Staging Committee of the AJCC, which I have the honor and privilege of chairing. Revisions of the current staging system, or even creation of new staging systems for nonextremity sites will be actively addressed by this Committee over the next several years. Such alterations will be contingent on accessing and integrating large institutional sarcoma databases such as those of the EORTC, the Memorial Sloan-Kettering Cancer Center, and the University of Texas M. D. Anderson Cancer Center. The power of such database resources should be sufficient relative to the challenging development tasks alluded to above.
Although this report offers a convincing and innovative approach to a problem that has traditionally been met with empiricism and dogmatism, before the recommendations can be accepted there are several caveats that must be borne in mind. The Royal Marsden Hospital is a major sarcoma referral center for Great Britain, and there may be considerable bias inherent in deciding to send a patient to such a quaternary treatment facility. Such referral-based bias may or may not be subtle and could impact on the strength of the recommendations, particularly the size of the recurrent tumor and the time to recurrence, where recurrence is ultimately a histopathologic rather than clinical determination. This issue can be addressed in a prospective validation of the Royal Marsden algorithm. It may also be pertinent to retrospectively examine much larger numbers of recurrent sarcoma patients whose clinical and pathological data could be retrieved from the large databases mentioned above.
Clary et al.3 report on a series of patients treated at the Memorial Sloan-Kettering Cancer Center who had either gastrointestinal stromal tumors (GIST) or leiomyosarcoma of the abdomen or retroperitoneum. Up to the relatively recent past, both diagnoses were clustered together under the broader rubric of leiomyosarcoma. These two entities can now be distinguished from each other because GIST usually stain immunohistochemically positive for the myeloid stem cell antigen CD34. GIST also have a much higher likelihood of expressing the c-kit proto-oncogene. Although the overall clinical outcome was similar for both patient groups, there were remarkable differences regarding the sex of the tumor-bearing host. GIST usually occurred in men, whereas leiomyosarcoma was predominantly a disease of women. GIST also had a predilection for metastasis to the liver and not the lung, whereas the converse was true of leiomyosarcoma. These phenotypic differences may be of more than academic interest in that the new "designer" drug STI 571, a potent tyrosine kinase inhibitor, has demonstrated remarkable activity in GIST, perhaps via ligation of the c-kit receptor, a member of the type III tyrosine kinase receptor family that also includes the platelet-derived growth factor receptor (PDGFr), which like the c-kit receptor is also associated with tyrosine kinase-mediated signal transduction and sarcoma proliferation. These distinctions are critical in that early clinical trials with STI 571 have shown >50% response rates in GIST, a percentage of response to systemic therapy never previously seen in this disease. Several larger scale phase II/III trials are currently under consideration.
What biological processes underlie sarcoma recurrence? We know that sarcomas are remarkably heterogeneous in both their macro- and microscopic appearance. The theory of clonal heterogeneity would suggest that all solid tumors consist of disparate clones of cells, some of which are much more capable of giving rise to the metastatic phenotype than other cells. Both metastatic and nonmetastatic cells are integral to the primary tumor. With serial cellular reproduction, those cells whose genotype is more prone to mutation may come to dominate, particularly in the metastatic or recurrent sarcoma, as has been shown by us.4 Such molecular factors are also important as mediators of sarcoma chemoresistance. In addition, not all sarcomas resected with microscopically positive margins fail as local recurrences. Unfortunately, the converse is also true, namely, that not all margin negative sarcoma resections result in a patient who remains free of disease.5 The ultimate driver here may not be the margin status per se so much as the natural biological nature of a given sarcoma, where the presence or postinitial therapy persistence of metastasis or recurrence-promoting cells is the critical factor regarding disease-free outcome. As a consequence of the above biology, issues such as recurrent tumor size, grade, and time to recurrence may actually just be surrogate markers of the undergirding molecular biology, a "poor man’s" Southern blot, as it were. As such, the utility of a proposed new staging sytem may only persist for as long as it takes cDNA microarray-based or proteomic-dependent molecular staging criteria of almost unfathomable precision to come "on line" as the ultimate staging criteria for this devastating illness.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  E. A. Perez, J. C. Gutierrez, F. L. Moffat Jr, D. Franceschi, A. S. Livingstone, S. A. Spector, J. U. Levi, D. Sleeman, and L. G. Koniaris Retroperitoneal and Truncal Sarcomas: Prognosis Depends Upon Type Not Location Ann. Surg. Oncol., March 1, 2007; 14(3): 1114 - 1122. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
