| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
Division of Surgical Oncology, John Wayne Cancer Institute at Saint John’s Health Center, 2200 Santa Monica Boulevard, Santa Monica, California 90404
Correspondence: Address correspondence and reprint requests to: Mark B. Faries, MD; E-mail: fariesm{at}jwci.org.
Imagine a striking report of a new treatment modality for patients with stage IV melanoma. This new therapy yields 100% complete pathologic responses, requires a median 7-day inpatient hospital stay, and more than triples the median overall survival. Such a report would certainly gain the attention of the national lay press and Wall Street. In this issue of the Annals of Surgical Oncology, Pawlik et al.1 present these results for carefully selected patients undergoing a therapy called resection for melanoma metastatic to the liver. However, it seems unlikely that this report will cause significant stock price fluctuations, because of the critical importance of patient selection and the very small number of eligible patients. Even so, we congratulate the authors for their results and strongly support their approach.
Until fairly recently, distant metastasis of any cancer was assumed to represent diffuse, incurable, systemic disease for which surgery could play only a limited, palliative role. This assumption was not unreasonable when operative mortality for hepatic resection exceeded 10% and metastatic melanoma was detected clinically by palpation of hepatomegaly,2,3 but it is now clear that surgical removal of distant disease, such as hepatic colorectal or pulmonary sarcoma metastases, may lead to long-term survival and possibly cure in carefully selected patients. Improved surgical techniques and perioperative care have diminished the risks of surgery, and advances in imaging have improved detection of early resectable metastatic disease. As a result, operative mortality for hepatic resection in the United States decreased by nearly half between 1989 and 2000.2 Recent large series of patients undergoing hepatic metastasectomy have reported average operative mortalities of <3%.4 Mortality in the series reported by Pawlik et al.1 is only slightly higher. Relative to many other therapies for stage IV melanoma, the toxicity of resection is also very acceptable; patients usually are discharged home within a week after the operation.
The principal challenge to the success of metastasectomy is patient selection. Although four large centers contributed clinical data from a 16-year period, the study of Pawlik et al.1 had only 40 patients. So on average, each center performed one metastasectomy every 19 months. The number of melanoma patients with liver metastases is not reported, but published data from the John Wayne Cancer Institute and Sydney Melanoma Unit indicate that only about 2% of melanoma patients with liver metastases undergo surgical exploration.5 Is it possible that other melanoma patients with hepatic involvement are candidates for resection? In metastatic colorectal cancer, population data suggest that many patients with potentially resectable disease do not undergo surgery, possibly because of physician or patient preference/bias.6 This may be even more likely in melanoma.
Extensive disease is the most common exclusion criterion for metastasectomy of melanoma. However, safer surgical techniques and the use of radiofrequency or cryoablation may have increased the number of patients eligible for removal or destruction of all evident disease.7 In addition, techniques such as isolated hepatic perfusion, which has a 62% reported objective response rate in metastatic ocular melanoma,8 may increase the number of patients with resectable disease. Surgeons and medical oncologists are often reluctant to consider resection of metastatic disease, possibly because they are not aware of the potential benefits of doing so. Retrospective analyses such as the multicenter study presented by Pawlik’s group1 reveal these benefits, but results should be interpreted cautiously.
In their review of hepatic resection for metastatic melanoma, Pawlik et al.1 report better outcomes in patients with ocular versus cutaneous primary disease sites. This is consistent with previous reports from the John Wayne Cancer Institute and other centers, which demonstrated a higher frequency of liver-only disease and a relatively indolent course in ocular compared with cutaneous melanoma.9,10 It is difficult to assign much significance to the 2-year survival difference reported by Pawlik et al.1 because the curves intersect before 2 years. But there does appear to be a larger difference at 5 years, albeit with fairly broad confidence intervals.
Among the patients with cutaneous melanoma, the authors noted a large difference in survival associated with systemic therapy. Because randomized studies of very aggressive biochemotherapy have demonstrated at best a marginal impact on overall survival,11 there must be another explanation for this difference. As the authors indicate, systemic therapy was most often administered before resection; generally, only those patients who responded to systemic treatment or who had long tumor-doubling times were eligible for resection. This selection process would have a large impact on survival.
Pawlik et al.1 also examined the potential impact of the extent of hepatic disease, surgical procedure, and margin status. Unilateral metastasis and smaller metastasis (<5 cm) tended to be favorable indicators (P = .13–.15), but margin status and number of lesions were not important (P = .4). Because previous studies have assigned significance to some of these factors,5 it is unfortunate that the study’s limited power prevents firm conclusions about predictive variables. Still, this type of study offers crucial information regarding which patients tend to have rapid progression versus extended survival after resection. For example, cutaneous melanoma patients with synchronous extrahepatic metastases had a very short disease-free survival (2.8 months) and a limited overall survival (7.2 months). Other series have shown a longer overall survival (15 months) for similar patients, with long-term survival in two (29%) of seven.5 Meaningful comparisons are difficult with such small numbers, and conclusions about absolute contraindications should be tempered, but this group of patients certainly deserves very careful consideration before resection is undertaken.
Despite the importance of patient selection, the combined experience of all reported series indicates only which patients are more likely to have long survival—not the impact of surgery on survival. If a patient lives 6 months without surgery but might enjoy 18 months of high-quality life after surgery, resection seems warranted. We cannot estimate the degree of benefit or even confirm its existence since, as the authors note, we do not have a reasonable control population. Statistical methods such as propensity score analysis, which attempt to control for prognostic variables, suggest that surgical resection of ocular melanoma metastasis improves survival even after accounting for the effects of patient selection.10 A randomized prospective trial would answer the question but is problematic in the setting of stage IV melanoma. However, the availability of such data could potentially convince the medical community of the benefits of surgical treatment in stage IV melanoma and make those benefits available to a larger number of patients.
Our previous work12 supports the conclusions of Pawlik et al.1: surgical resection has an important role in the treatment of stage IV melanoma.13 The role of computed tomographic scanning in calculation of tumor volume–doubling time and its relation to the results of surgical resection can now be evaluated for liver metastases, as previously defined for pulmonary metastases.14 Overall, current outcomes in this stage of disease are poor, and a more aggressive surgical approach may yield longer survival. For patients with isolated hepatic metastases from cutaneous or ocular melanoma, approaches that combine surgical, systemic, and regional modalities must be further investigated.
ACKNOWLEDGMENTS
Supported by grant CA76489 from the National Cancer Institute and by funding from the Nancy and Carroll O’Connor Foundation (Los Angeles, CA) and the Wayne and Gladys Valley Foundation (Oakland, CA).
Received for publication December 23, 2005. Accepted for publication January 5, 2006.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
