| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Surgical Metabolism Section, Surgery Branch, Center for Cancer Research, National Cancer Institute, 10 Center Drive, Building 10, 4W-5940, Bethesda, Maryland 20892
Correspondence: Address correspondence and reprint requests to: H. Richard Alexander, Jr., MD; E-mail: richard_alexander{at}nih.gov
The article by Rivoire et al.1 in this issue of Annals of Surgical Oncology details the outcomes of 63 patients diagnosed with liver metastases from ocular melanoma from a cohort of 602 who were being evaluated by screening that consisted of serial physical examination, ultrasonography, and chest x-ray every 6 months after treatment of their primary neoplasm. Outcomes of the 63 individuals who developed liver metastases are detailed according to the nature of the treatment; 28 patients underwent surgical exploration with curative intent, yet only half (n = 14) were able to have complete surgical resection. The other 14 had incomplete resection with or without additional regional or systemic chemotherapy. The other 35, who had diffuse hepatic involvement on the basis of preoperative imaging, received systemic chemotherapy or best supportive care. I applaud the authors for such an informative study and find these data provocative in that they bring into focus many of the challenges and opportunities before us in combating this perfidious condition.
Early detection of recurrent cancer, particularly in the context of performing routine screening, can be considered useful only when there are simple and sensitive methods of early detection and when therapeutic options are available to control disease progression and improve survival. In the current report, approximately 10% of patients (63 of 602) developed liver metastases during follow-up and, after detection of metastasis, had a 15-month median survival. The authors chose not to use serial serum liver function test determinations because of the acknowledged limitations of these tests as a useful screening modality. The Collaborative Ocular Melanoma Study (COMS) group has recently reported the results of regular follow-up screening for patients enrolled on two random assignment trials in which liver function tests were used as the primary screening modality. When these test results were abnormal, they had >90% specificity but only 14% sensitivity.2 The futility of screening with liver function tests is reflected in the dismal outcome of individuals diagnosed with liver metastases in the COMS; the median survival was <6 months, and the 1- and 2-year mortality rates were 81% and 92%, respectively.
Certain features of primary tumors, such as location and size at the time of treatment for the primary neoplasm, have been shown to be associated with an increased risk of metastasis and decreased survival.3,4 The 5- and 10-year cumulative rates of metastasis in the COMS studies were 24% and 32%, respectively, and were significantly greater (38% 5-year metastasis rate) in patients with large primary neoplasms. This study demonstrates that primary tumor characteristics lose their prognostic relevance once liver metastases are diagnosed.
The overall median survival of 13 months after the diagnosis of liver metastases in this study is a sobering observation for those of us who have reported similar median survivals in patients after intensive hepatic directed regional therapies such as chemoembolization, intra-arterial chemotherapy, or isolated hepatic perfusion.5 These data support my perception that very little we do can meaningfully alter the natural history of the disease. The authors report that outcome was significantly longer in patients who could successfully undergo a complete resection of disease and in those with 
10 hepatic metastases (both had almost 60% 2-year survival). The authors report that 14 patients (24% of the population with metastatic liver disease) successfully underwent a complete resection and enjoyed a modest but respectable median 25-month survival: this is quite comparable to the results reported by investigators at the John Wayne Cancer Institute.6 However, I view the ability to completely resect disease as, more likely, a surrogate of low disease burden and indolent tumor biology and am not yet prepared to consign significant therapeutic benefit to surgical resection in this clinical setting on the basis of available data. We7 and others8 have shown that serum lactate dehydrogenase is an independent prognostic parameter for outcome in patients with liver metastases from ocular melanoma, and whether those selected for surgery had a more indolent disease reflected in normal preoperative lactate dehydrogenase determination is something we cannot learn from this study.
The main message I take from this important and informative study is that we, as surgeons, oncologists, and scientists, will make very little progress in advancing the field from the perspective of viewing the enemy while flying at 60,000 feet. We struggle to understand the biological events that will play out over years—even decades—by sifting through somewhat crude, albeit somewhat informative, clinical and pathologic parameters such as age and sex or primary tumor size or morphology at the time of initial diagnosis and treatment. Then, despite being well armed with an understanding of where the enemy is hiding (the liver in 
90% of patients with metastasis), we are virtually powerless using current state-of-the-science screening modalities such as serum or bio-marker assays or imaging studies to verify its presence until it has emerged from dormancy and initiated its inexorable assault. Even then, we may pick and win a few skirmishes with largely ineffective weaponry and always lose the greater struggle.
The biological behavior of a primary ocular melanoma that plays out over many years after the primary tumor is treated is very likely reflected in altered genetic profiles and aberrant cellular pathways within the primary tumor itself at the time of the initial diagnosis. Understanding these genetic and cellular alterations will help to identify new targets for more effective prophylactic or therapeutic interventions. Recently, Onken et al.9 reported remarkable findings showing that primary uveal melanomas cluster into two distinct molecular classes with very different clinical outcomes based on gene-expression profiling. They identified and validated a three-gene signature that predicted death from metastatic disease more accurately than any other clinical or pathologic prognostic indicator or cytogenetic analysis. The 92-month survival probability was 95% vs. 31% when the two groups were compared on the basis of gene-expression differences by using quantitative polymerase chain reaction. Moreover, we and others have used molecular techniques to identify new targets in melanoma. For example, expression of interleukin 1, an inflammatory cytokine with proangiogenic effects, in the tumor microenvironment has been associated with a virulent phenotype in different types of cancers, including melanoma. Overexpression of the interleukin 1 receptor antagonist inhibits melanoma xenograft growth and metastatic potential.10 Yang et al.11,12 have shown that interferon or low-dose angiostatin inhibits liver metastases in melanoma allografts. Taken together, these data provide some justifiable optimism that molecular profiling of primary (or metastatic) ocular melanoma will provide powerful insights into the biology of this disease, identify new targets for prophylaxis or therapy tailored for individuals according to tumor genotype, and ultimately translate into meaningful improvements in survival.
Received for publication February 23, 2005. Accepted for publication March 17, 2005.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
