| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Sydney Melanoma Unit (PL, AD, JFT) and the Department of Anaesthetics (PCAK), Royal Prince Alfred Hospital; and the Department of Surgery (JFT), University of Sydney, Sydney, New South Wales, Australia.
Correspondence: Address correspondence and reprint requests to: J. F. Thompson, MD, Sydney Melanoma Unit, Sydney Cancer Centre, Royal Prince Alfred Hospital, Camperdown, New South Wales 2050, Australia; Fax: 61-2-9550-6316; E-mail: john{at}mel.rpa.cs.nsw.gov.au
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Methods: The outcome in 135 patients treated by ILI was reviewed.
Results: The overall response rate in the treated limb was 85% (complete response [CR] rate 41%, partial response rate 44%). Median response duration response was 16 months (24 months for patients with CR). Median patient survival was 34 months. In those with a CR, the median survival was 42 months. CR rate and survival time decreased with increasing disease stage. Patients aged >70 years had a better overall response than younger patients. On multivariate analysis, factors associated with an improved outcome were a lower stage of disease, a final limb temperature >37.8°C, and a tourniquet time >40 minutes.
Conclusions: The frequency and duration of responses after ILI were comparable to those achieved by conventional ILP. The ILI technique is particularly useful for older patients who might not be considered suitable for conventional ILP.
Key Words: Hypoxia • Isolated limb infusion • Melanoma • Melphalan • Prognostic factors • Regional chemotherapy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
), and interferon gamma (90% CR).3 Although it is effective, the ILP technique unfortunately involves a complex and invasive operative procedure. A simplified procedure has been developed at the Sydney Melanoma Unit (SMU) and is called isolated limb infusion (ILI)4,5 to differentiate it from ILP. However, ILI is essentially a low-flow ILP performed via percutaneously inserted catheters, but without oxygenation. Despite the short duration of exposure of the tissues to melphalan during ILI (15–30 minutes), there is apparently adequate cellular uptake for tumor cell killing to be achieved in most patients.6 ILI has in a previous evaluation been shown to give overall response (OR) rates comparable to those observed after conventional ILP.5
Patients treated by ILI and ILP often have or develop systemic disease, and no increased survival rate has yet been shown. The main goal of both ILI and ILP is thus to achieve locoregional tumor control in the affected limb.
In this study we analyzed prospectively collected data for all patients who had undergone ILI for melanoma at the SMU over a 6-year period. Our aim was to identify prognostic factors that might allow better patient selection, explain treatment failures, and indicate intraoperative factors that could improve the outcome for patients undergoing this treatment. It has previously been suggested, for example, that a low temperature in the limb results in suboptimal perfusion of tumor deposits, particularly those that are cutaneous in their location.7 Therefore, one aim of the study was to test the hypothesis that hypothermia during the procedure is detrimental to outcome, with better results if higher temperatures are achieved. Another aim was to identify factors influencing limb toxicity after ILI. We also sought to determine whether a favorable response was associated with improved patient survival.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The technical details of the ILI procedures were as described previously.4,5 In brief, standard radiological catheters were inserted percutaneously into the axial artery and vein of the disease-bearing limb via the contralateral groin, and their tips were positioned so that they lay at the level of the knee or elbow joint (Fig. 1; tissues more proximally located in the limb, but distal to the level of the tourniquet, were perfused in a retrograde fashion via collateral vascular channels). The patient was then anesthetized and given heparin (3 mg/kg) to achieve full systemic heparinization. A pneumatic tourniquet was inflated around the root of the limb to be treated, and the cytotoxic agent or agents (in most cases, melphalan 5–10 mg/L of tissue ± actinomycin-D 50–100 µg/L of tissue, in 400 ml of warmed, heparinized normal saline) were infused into the isolated limb via the arterial catheter. For the duration of the ILI procedure (20–30 minutes), the infusate was then continually circulated by repeated aspiration from the venous catheter and reinjection into the arterial catheter by using a syringe attached to a three-way tap in the circuit. Limb temperature was increased by incorporating a blood-warming coil in the extracorporeal circuit and by wrapping the limb in a hot-air blanket. After 20 to 30 minutes, the limb was flushed with 1 L of Hartmann’s solution via the arterial catheter, and the venous effluent was discarded. The limb tourniquet was then deflated to restore normal limb circulation, and the catheters were removed. Subcutaneous and intramuscular limb temperatures were monitored continuously during the ILI procedure, and blood samples were taken at regular intervals for measurement of melphalan concentrations, blood gases, and pH. Postoperatively the serum creatine phosphokinase (CK) level was measured daily, and limb toxicity, systemic toxicity, and tumor response were assessed regularly.
|
Of the 135 patients in this study, 130 had locally recurrent or metastatic melanoma involving a limb, and of these, 17 also had known distant metastases at the time of treatment (5 had pulmonary metastases, 8 had subcutaneous or lymphatic metastases in the limb above the level of the tourniquet or in the groin, and 4 had other visceral disease). Five patients had advanced stage I disease considered unsuitable for primary surgical resection. The median follow-up time was 18 months (mean, 24 months). Disease status at the time of the initial ILI, according to the M. D. Anderson classification, was as follows: stage I (intact primary tumor), 4%; stage II (local recurrence), 15%; stage IIIa (in-transit metastasis), 42%; stage IIIb (in-transit metastasis and regional node involvement), 28%; and stage IV (distant metastasis), 11%. Patients with stage IV disease were treated if their systemic metastases were not considered to be life-threatening in the short to medium term, but they had seriously symptomatic disease in a limb. The median age of the patients undergoing ILI was 69 years (range, 36–91 years), and 59% (80 of 135) of the patients were women. Limb toxicity was assessed by using the scale proposed by Wieberdink et al.8
Possible prognostic factors were tested for their influence on response rates, duration of response, and survival. The 
2 test was used for comparison of frequency distributions. Survival and progression-free survival were analyzed with the Kaplan-Meier method.9 For continuous variables, the median was chosen as the cutoff point to avoid bias (because the data were nonparametric). The variation of time parameters was expressed as the interquartile range. The log-rank test was used for univariate analysis, and Cox regression (Cox’s proportional hazard model)10 with the stepwise backward method was used for multivariate analysis with various cohorts. A significant difference was assumed for a probability value of <.05. Statistical analysis was performed with the Statview 5.0 software package (SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The dose of melphalan (mean, 7.6 mg/L of limb tissue; median, 7.1 mg/L; range, 4.3–18.8 mg/L) was not correlated with local toxicity. However, the maximal serum CK value observed postoperatively was correlated with the concentration of melphalan measured in the limb blood after both 5 and 20 minutes (Table 1). Although we have previously shown that maximal serum CK levels in the postoperative period were correlated with toxicity after ILP,11 this was not the case after ILI. After conventional ILP in our institution, a CK level exceeding 1000 IU was usually associated with serious toxicity, but a number of the ILI patients exceeded this level and did not develop significant limb toxicity. However, all patients whose CK levels exceeded 1000 IU were treated with corticosteroids until their CK levels had decreased to <1000 IU again.
|
Response Rates
Responses were assessed according to standard World Health Organization criteria.12 These define a CR as the disappearance of all known disease, determined by two observations not <4 weeks apart, and a PR as a 
50% decrease in total tumor size determined by two observations not <4 weeks apart and no appearance of new lesions or progression of any lesion.
It was possible to evaluate responses after 128 of the 135 ILIs. Seven patients could not be assessed because they lived in remote regions or overseas. An OR rate of 85% was achieved. In 53 patients (41%), CR occurred, 55 patients (44%) had a PR, 15 patients (12%) had stable disease, and 5 (4%) patients had progressive disease. The overall median duration of response was 16 months (range, 6–66 months). The duration of CR was 42 months (range, 12–68 months), which was significantly higher (P < .001) than the duration of PR, which was 9 months (range, 4–20 months; Fig. 2). When stratified for stage of disease, the prolonged duration of response observed in the CR group remained significant (although the P value became .05). Of all 70 documented relapses in the treated limb, 67 (96%) occurred within 22 months. The median time between ILI and ultimate response was 1.3 months (range, 1 week to 9.5 months). Of the 69 patients who died during the study period, 33 (48%) did not have any progression in the limb at the time of death.
|
|
|
The time to progression within the infused limb was shorter in stage IIIb disease (11 months) and stage IV disease (9 months) compared with stages I, II, and IIIa (20 months). Patients with no regional lymph node involvement (stage IIIa) had a longer survival (48 months; range, 28 to >67 months; P = .004) than patients with involved lymph nodes (stage IIIb), in whom the survival was only 19 months (range, 10 to >72 months; Fig. 4). Patients with tumor infiltration beneath the deep fascia had a shorter duration of their response compared with patients with only cutaneous or subcutaneous tumor involvement. Patients with multiple lesions had a better prognosis than patients with a single lesion in the limb. Patients with tumor nodules <6 mm in diameter had a higher CR rate (57%) than those with nodules >6 mm in diameter (32%; P = .01).
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Most regional relapses (95%) occurred within 2 years of ILI. This is comparable to the ILP series of Di Filippo et al.,16 in which 100% of relapses occurred within 3 years. The median duration of CR after ILI (16 months) was comparable to the longest median response duration described after CR for ILP (19 months),17 so in that respect ILI was thus demonstrated to be of similar efficacy to ILP.
Prognostic Factors and Their Clinical Implications
Of potentially important practical significance in this study was the finding that response rates after ILI were higher in patients >70 years old. Such patients are often considered unsuitable for treatment by conventional ILP, because it is such a major undertaking with high associated morbidity. Yet when older patients have seriously troublesome recurrent melanoma involving a limb, their quality of life can be greatly impaired. Radical local surgery, radiotherapy, and systemic chemotherapy are unsatisfactory and often risky treatment options in this age group, and although limb amputation may be a realistic alternative, satisfactory rehabilitation after amputation is notoriously difficult to achieve in the elderly.
However, ILI is well tolerated by older patients, even those who have intercurrent medical problems. Higher response rates in this age group thus make it an even more attractive treatment option for those with serious limb disease that might otherwise be treated conservatively (and ineffectively) or not at all.
Prognostic Factors
That disease stage affected survival outcome in the study was not unexpected, with the most obvious decrease in survival between M. D. Anderson stages IIIa and IIIb, i.e., when there was disease outside the region treated by the ILI. The importance of the temperature achieved in the limb during ILI was confirmed by demonstrating a relationship between subcutaneous limb temperature at the end of the ILI procedure and response. This finding of a synergism between hyperthermia and melphalan is consistent with the results of others using ILP.18 However, Kroon7 has suggested that it is the avoidance of hypothermia that is of greatest importance, and it is noteworthy that the group at the Netherlands Cancer Institute has achieved excellent results after normothermic, rather than hyperthermic, ILP with melphalan. Similarly, with ILI it is highly likely that it is the avoidance of hypothermia that is important in achieving a successful outcome, because temperatures exceeding 38.5°C seldom are achieved. Because the infusion circuit has a limited heating capacity, it is therefore necessary to maintain the temperature of the limb with an external heating blanket as well. It is also of value to commence heating both patient and limb as soon as the patient arrives in the anesthetic room, because it is otherwise difficult to reach the desired subcutaneous temperature of at least 37.8°C in the limb by the end of the ILI procedure.5
Our finding that a tourniquet time longer than 40 minutes was associated with improved survival was unexpected, because the elimination half-life for melphalan is 15 to 25 minutes. Thus, during an ILI procedure, most of the drug has been eliminated from the limb circulation after 40 minutes. In vitro studies of melanoma cells indicate that the uptake of melphalan by the tumor is a rapid, active, and temperature-dependent process that reaches a plateau after approximately 10 minutes because of saturation of the uptake mechanism.19 Therefore, the beneficial effect of a long tourniquet time is unlikely to result from enhanced drug delivery from the circulation to the tumor cells. Instead, the effect of ischemia itself and the potential synergism of hypoxia with melphalan may explain this finding. Because the limb was perfused with drug for approximately 15 to 30 minutes in most patients, and washout time before tourniquet release was fairly constant (5–10 minutes), the time that varied most was the tourniquet time before drug perfusion commenced. The hypoxia and acidosis induced during that time might have been beneficial for the outcome. In vitro studies have shown that hypoxia enhances the cytotoxic effects of melphalan by a factor of approximately 1.5,20 whereas the combination of hypoxia and acidosis can increase the effect by a factor of 3. By using hyperthermia and administering glucose during ILP, intracellular pH in the tumor can be decreased, with a concomitant increase in the response rate.21 How the synergism between melphalan and hypoxia is mediated is presently unknown. It is tempting to speculate that producing deliberate hypoxia during conventional ILP might improve response rates, and there are some animal and human data (particularly for hypoxic isolated pelvic perfusion) to support this concept. However, we believe that further studies would be required before routine reduction of oxygen tensions during ILP could be recommended.
The finding that ILI for patients with only one tumor in the limb was associated with a poorer survival than ILI in patients with several tumors was also unexpected. However, the mean size of single tumors was considerably greater (30 mm) than the mean size of the multiple lesions (9 mm). This is in accordance with the finding that lesions smaller than 6 mm were easier to treat than larger lesions, which could be explained by a lower tissue penetration of drug into larger lesions that have a poor central microcirculation.
The regional toxicity reported in this article is within the range described after conventional ILP.22 A higher CR rate was observed in patients experiencing grade III or IV toxicity, compared with those who developed only grade I or II toxicity. This was in contrast to a recent analysis of ILP and toxicity which did not find any relationship between toxicity and effect.22 It is unlikely that further dose escalation for ILI would be possible without unacceptable toxicity. A tendency toward increased regional toxicity after a second ILI is also a deterrent to further increasing the dose. Because higher response rates were not achieved at higher melphalan levels, it is doubtful that dose escalation would yield better results. However, because addition of TNF to a melphalan regimen during ILP has been shown to increase CR rates to values as high13 as 90%, a study using melphalan and TNF administered by ILI would be of great interest.
Effect on Survival
No form of locoregional treatment for recurrent melanoma, even limb amputation, has ever been shown unequivocally to influence survival. This is presumably because survival is principally dependent on the existence and rate of progression of distant metastases outside the treated region. The finding in this study that several parameters were independent prognostic factors for survival after ILI in the multivariate analysis was therefore unexpected, and this result must be interpreted with great caution.
Few randomized studies have assessed the effect of ILP on survival. One study showed a nonsignificant increase in survival in recurrent melanoma.23 In a very small single-center study that was terminated prematurely,24 a significant increase in disease-free survival time in patients who underwent prophylactic ILP as an adjunct to surgery was reported. However, a much larger multicenter randomized study did not show increased survival after prophylactic ILP plus surgery,25 although an increase in disease-free survival was observed in the subgroup of patients who had not undergone elective lymph node dissection.
In this study, the prolonged survival observed in the CR group can probably be explained by the higher proportion of patients with low-stage disease in the CR group, because patients with less aggressive disease seemed to respond better to ILI. That disease stage affected survival outcome in the study was not unexpected, with the most obvious decrease in survival between M. D. Anderson stages IIIa and IIIb, i.e., when there was disease outside the region treated by ILI.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Acknowledgments |
|---|
Received for publication April 17, 2001. Accepted for publication September 20, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. M. Kroon, M. Moncrieff, P. C. A. Kam, and J. F. Thompson Outcomes Following Isolated Limb Infusion for Melanoma. A 14-Year Experience Ann. Surg. Oncol., November 1, 2008; 15(11): 3003 - 3013. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. Moncrieff, H. M. Kroon, P. C. Kam, P. D. Stalley, R. A. Scolyer, and J. F. Thompson Isolated Limb Infusion for Advanced Soft Tissue Sarcoma of the Extremity Ann. Surg. Oncol., October 1, 2008; 15(10): 2749 - 2756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Beasley, R. P. Petersen, J. Yoo, N. McMahon, T. Aloia, W. Petros, G. Sanders, T.-Y. Cheng, S. K. Pruitt, H. Seigler, et al. Isolated Limb Infusion for In-Transit Malignant Melanoma of the Extremity: A Well-Tolerated but Less Effective Alternative to Hyperthermic Isolated Limb Perfusion Ann. Surg. Oncol., August 1, 2008; 15(8): 2195 - 2205. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. K. Augustine, Y. Yoshimoto, M. Gupta, P. A. Zipfel, M. A. Selim, P. Febbo, A. M. Pendergast, W. P. Peters, and D. S. Tyler Targeting N-Cadherin Enhances Antitumor Activity of Cytotoxic Therapies in Melanoma Treatment Cancer Res., May 15, 2008; 68(10): 3777 - 3784. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Yoshimoto, C. K. Augustine, J. S. Yoo, P. A. Zipfel, M. A. Selim, S. K. Pruitt, H. S. Friedman, F. Ali-Osman, and D. S. Tyler Defining regional infusion treatment strategies for extremity melanoma: comparative analysis of melphalan and temozolomide as regional chemotherapeutic agents Mol. Cancer Ther., May 1, 2007; 6(5): 1492 - 1500. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. F. Hegazy, S. Z. Kotb, H. Sakr, E. El Dosoky, T. Amer, R. A. F. Hegazi, and O. Farouk Preoperative Isolated Limb Infusion of Doxorubicin and External Irradiation for Limb-Threatening Soft Tissue Sarcomas Ann. Surg. Oncol., February 1, 2007; 14(2): 568 - 576. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. S. Brady, K. Brown, A. Patel, C. Fisher, and W. Marx A Phase II Trial of Isolated Limb Infusion With Melphalan and Dactinomycin for Regional Melanoma and Soft Tissue Sarcoma of the Extremity Ann. Surg. Oncol., August 1, 2006; 13(8): 1123 - 1129. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. H. Ko, T. Ueno, Y. Yoshimoto, J. S. Yoo, O. I. Abdel-Wahab, Z. Abdel-Wahab, E. Chu, S. K. Pruitt, H. S. Friedman, M. W. Dewhirst, et al. Optimizing a Novel Regional Chemotherapeutic Agent against Melanoma: Hyperthermia-Induced Enhancement of Temozolomide Cytotoxicity Clin. Cancer Res., January 1, 2006; 12(1): 289 - 297. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Tyler Editorial: Where are We going with Regional Therapy for Melanoma? Ann. Surg. Oncol., May 1, 2004; 11(5): 455 - 457. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
