| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIALS |
From the Department of Surgery, Duke University Medical Center, Durham, North Carolina.
Correspondence: Address correspondence to: Douglas Tyler, MD, Department of Surgery, Box 3118, Duke University Medical Center, Durham NC 27710; Fax: 919-681-8701; E-mail: tyler002{at}acpub.duke.edu
Isolated limb perfusion (ILP) is a technique that was first performed in 1957 to treat patients with regionally advanced cancers not considered surgically resectable.1 Although several waves of enthusiasm have occurred over the last five decades, including the addition of hyperthermia in the 1960s2 and tumor necrosis factor (TNF) in the 1960s,3 the underlying technique of ILP has not changed much since it was first introduced. In its current form, ILP is performed predominantly using melphalan. Controversy still exists to whether hyperthermia improves response rates. Although seeming to increase the complete response rate in small trials, TNF showed no benefit at the interim analysis of the recent American College of Surgeons Oncology Group randomized phase III trial, leading to that trial being stopped early. The future role of TNF in the regional therapy of melanoma is probably limited.
Which patient population benefits from ILP has also been controversial. Generally, ILP is used to treat patients with American Joint Committee on Cancer (AJCC) stage IIIB and IIIC disease. These patients have intransit disease, which is defined as tumor recurrence between the primary tumor and the regional lymph node basin. ILP has been studied in the adjuvant setting with excision of primary melanoma through a randomized clinical trial and found to decrease intransit recurrences but not to have any benefit on overall survival.4 This is not surprising because most patients with primary melanoma will not develop intransit disease and those who do (in the presentinel lymph node era) probably also have microscopic systemic disease, for which regional ILP will not help. Thus, whereas this study demonstrated that ILP clearly has an impact on micrometastatic disease in an extremity, it was probably not powered to detect a survival difference if one existed.
On the other side of the spectrum, patients with unresectable extremity melanoma, as defined by more than 10 lesions or a lesion greater then 5 cm, seem to have the best response rate (60% to 80%) and limb salvage rate with ILP compared with any other therapy.5 In this subset of patients, ILP has not been tested in randomized trials and compared with other types of treatment. Many believe that most of these patients have some degree of systemic disease, which makes ILP a form of palliative therapy. Interestingly, older studies that used amputation as a form of treatment for patients with advanced extremity disease showed an approximate 25% 5-year survival rate, suggesting that a small subset of these patients have disease that is truly indolent or confined to the extremity.6
A third group of patients who are frequently considered for ILP are those individuals with small volume intransit disease who can have ILP in a prophylactic setting. Prophylactic ILP is defined as regional therapy for an individual who has had an intransit or local recurrence failure but has no clinically detectable disease at the time of ILP. By extrapolating from the data above, some believe that if prolonged survival can be achieved in 25% of patients with advanced extremity melanoma using amputation, then perhaps a larger percentage could benefit from aggressive regional therapy with prophylactic perfusion in the setting of small volume intransit disease after it was removed. Whereas prophylactic ILP has been associated with a lower recurrence rate and a longer recurrence-free interval, it has not been associated with improvements in overall survival. In this issue of Annals of Surgical Oncology Noorda et al., in an article entitled "Isolated Limb Perfusion Prolongs the Limb Recurrence-free Interval after Several Episodes of Excisional Surgery for Locoregional Recurrent Melanoma,"7 try to identify a subset of these patients who may benefit from regional chemotherapy using ILP. They focus on the group of patients who had their first ILP after their third intransit recurrence. Not unsurprisingly, they found that ILP lengthens the local recurrence-free interval and decreases the number of lesions per recurrence. Although this article suggests that ILP might be a valuable adjunct for some patients with surgically resectable intransit disease, it does little to address two greater issues for regional therapy in melanoma, namely, is the spectrum of intransit disease changing in the era of sentinel lymph node biopsy and what are we doing to improve the efficacy and decrease the toxicity of regional therapy in the future?
Intransit melanoma in the presentinel lymph node era is reported as a pattern of failure in approximately 5% to 10% of patients in most large series. In most of these series, the median time to appearance of intransit disease is between 13 to 16 months, with 80% becoming apparent with 3 years of initial diagnosis.8 Distant metastatic disease develops with, or shortly after, the appearance of intransit disease, although approximately 25% of patients seem to have a more indolent pattern of extremity only disease. Some concerns have been raised that intraoperative lymphatic mapping and sentinel lymph node biopsy can alter patterns of recurrence and potentially increase the incidence of intransit disease. In a recently, published series from M. D. Anderson Cancer Center, involving a prospectively followed database of almost 1400 patients who had had a sentinel lymph node biopsy, 90 patients (6.4%) developed intransit recurrence at a mean follow-up of 3.5 years.9 For 85 of the 90 patients, the intransit disease was a component of their first recurrence and in 60 of those patients (71%), it was the only site of first recurrence. The presence of intransit disease was associated with a worse prognosis compared with individuals who did not develop intransit disease. Whereas these findings do not clearly suggest a marked difference in intransit disease with lymphatic mapping, additional follow-up is necessary to more clearly define the natural history of intransit disease in this patient population.
Where are we going in the future with regional therapies? In the absence of any systemic treatment that has significant and durable response rates, interest will continue in regional therapies for melanoma. Much of the focus will be to develop (1) ways to improve the therapeutic index of drugs given regionally and (2) regional strategies that can impart some beneficial systemic effects.
The Sydney melanoma group has recently published their experience with isolated limb infusion (ILI), which for a single institution has complete response rates of 41% and partial response rates of 44%.10 In contrast to ILP, ILI is performed through percutaneously placed catheters and the drugs (usually melphalan and Actinomycin D) are circulated at low flow rates in an hypoxic, acidotic environment. This is in contrast to ILP in which drug (usually melphalan alone) is circulated at high rates by a pump oxygenator, which maintains normal tissue oxygenation and pH. Limb infusion has the potential appeal that it is less invasive to the patients, appears to have a lower complication rate, and may be applicable to a wider range of patients (e.g., those with vascular disease or lower functional status) than might otherwise be required for ILP. Validation of the Sydney data regarding ILI in a multi-institutional phase II study through the American College of Surgeons Oncology Group is currently in development.
Drug pharmacokinetics and optimization of drug delivery to tumor tissue is another area for focus. Recent studies at Duke University Medical Center suggest that marked variations in plasma drug melphalan levels exist between patients using standard formulas to calculate drug dose.11 In this analysis, the ratio of estimated limb volume to the volume of the drug distribution (Vesti/Vss) was the most potent predictor of limb toxicity compared with other variables that have been associated with limb toxicity (e.g., peak melphalan levels) or area under the curve calculations). All significant toxicity was seen when Vesti/Vss ratios exceeded 4, suggesting that in certain patients limb volume estimates markedly overestimate the volume of drug distribution, leading to an overcalculation of the appropriate drug dose. Future studies should examine drug pharmacokinetics more carefully to determine if alternative drug dosing schemes should be developed for various patient populations.
Although melphalan is the current drug of choice for regional therapy, controversy remains on the best way to give it. Whereas many institutions feel hyperthermia improves response rates, many groups, including Noorda et al., still perform normothermic ILP. Although hyperthermia has never been conclusively shown to be beneficial in clinical trials, in an animal model, it has recently been demonstrated that hyperthermia—not the prevention of hypothermia—markedly potentiates melphalan’s effectiveness by significantly increasing tumor apoptosis.12 Actinomycin D, a drug frequently given in conjunction with melphalan in ILI, but not in ILP, may also augment tumor apoptosis.
Tumor acidosis is another strategy used to make melphalan more effective. Acidosis plus melphalan was recently shown to induce nitric oxide-mediated tumor regression in an ILP human melanoma xenograft model.13 Attempts to apply selective acidosis clinically have centered around the use of cold metaiodobenzyl guanidine (MIBG), which is preferentially picked up by melanoma cells and increases tumor acidosis intracellularly.14 This strategy, which was also shown to be effective in an animal model, is currently in development as a phase 1 clinical trial.
Finally, an appreciation of the role of chemoresistance to melphalan may also be a strategy to exploit in future melphalan-based trials. Glutathione is a tripeptide that binds to melphalan intracellularly and inactivates it. Whereas glutathione levels in muscle can be protective of toxicity, their presence in tumor makes the tumor less responsive to melphalan. Modulation of glutathione levels markedly improved tumor responses to melphalan-based regional therapies in an animal model, causing a marked improvement in the therapeutic index of melphalan.15 Chemotherapy resistance pathway modulation is also under development in the context of a phase I clinical trial.
New drugs can also help optimize regional therapy. Temozolomide will soon be coming out in an intravenous formulation. Unlike 5-(3,3-dimethyl-1-triazino)imidazole-4-carboxamide (DTIC), which is the single most active compound against melanoma, temozolomide does not require hepatic metabolism to generate its active form MTIC. Studies in animals suggest that this drug can be more effective than melphalan when used in a regional fashion and that temozolomide’s efficacy can be further improved by resistance pathway modulation much like melphalan.16
Until newer agents are proven effective, regional therapies should be considered palliative in most individuals, with current efforts focused on optimizing the therapeutic index of the drugs used to maximize tumor response and minimize toxicity. In addition, close follow-up of patients having sentinel lymph node biopsy needs to continue to make sure that this staging test does not alter the pattern of failure in a potentially detrimental way.
Received for publication March 9, 2004. Accepted for publication March 12, 2004.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
