10.1245/ASO.2006.05.003
Annals of Surgical Oncology 13:1123-1129 (2006)
© 2006 Society of Surgical Oncology
A Phase II Trial of Isolated Limb Infusion With Melphalan and Dactinomycin for Regional Melanoma and Soft Tissue Sarcoma of the Extremity
Mary S. Brady, MD, FACS1,
Karen Brown, MD2,
Ami Patel, BS1,
Charles Fisher, CRNA3 and
Will Marx, MD3
1 Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021
2 Department of Radiology, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021
3 Department of Anesthesiology, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021
Correspondence: Address correspondence and reprint requests to: Mary S. Brady, MD, FACS; E-mail: bradym{at}mskcc.org.
 |
ABSTRACT
|
|---|
Background: Isolated limb infusion (ILI) is a minimally invasive technique of delivering regional chemotherapy in patients with advanced melanoma or soft tissue sarcoma of the limb. Reports from Australia of efficacy similar to that of isolated limb perfusion prompted us to conduct a phase II trial to evaluate the efficacy and safety of ILI.
Methods: Eligible patients had American Joint Committee on Cancer stage IIIB or IIIC melanoma or unresectable soft tissue sarcoma of the limb. Angiographic catheters were positioned just above the knee or elbow of the extremity. General anesthesia was performed, a proximal tourniquet was inflated, and a normothermic, low-flow, hypoxic infusion of melphalan and dactinomycin was circulated through the involved limb for 20 minutes. The tumor response was assessed by using standard criteria at 3 months. Morbidity was determined in the hospital and at 2, 6, and 12 weeks.
Results: Twenty-five patients were accrued to the trial, and 32 ILIs were performed (8 patients had 2 ILIs); 1 patient was not treated. Of the 22 assessable patients, 11 (50%) had a significant response at 3 months: 23% of patients had a complete response, and 27% of patients had a partial response. The median duration of complete response was 1 year (range, 6–32 months). Morbidity was acceptable. Peak morbidity occurred at 2 weeks and was considered moderate in most patients. Limb edema and erythema were common. No patient developed compartment syndrome or required amputation.
Conclusions: ILI is well tolerated. Half of the patients experienced a complete or partial response.
Key Words: Isolated limb infusion • Melanoma • Soft tissue sarcoma • Regional chemotherapy
|
INTRODUCTION
|
|---|
Up to 10% of patients with intermediate-thickness melanoma of the extremity will develop local or intransit recurrence after wide excision of their primary melanoma.1 These recurrences cause pain, bleeding, and psychological distress. At least half of patients with recurrent melanoma of the extremity survive for 2 years without developing distant metastasis.2 Although a curative approach to treatment is generally unrealistic, palliation of regional disease is an important therapeutic goal.
Current regional treatment options for these patients include surgical resection, intralesional therapy with sclerosing agents or immune adjuvants, isolated limb perfusion (ILP) with melphalan, systemic chemotherapy, or amputation. Surgical excision is a reasonable procedure when the lesion(s) can be encompassed without extensive skin grafting or are few in number. Intralesional therapy with bacillus Calmette-Guérin or interferon can provide relief for small lesions that are primarily dermal.3,4 Major amputation is uncommonly performed for recurrent melanoma or soft tissue sarcoma (STS), even when it is confined to the extremity. In a highly selected group of patients with melanoma limited to the extremity, however, amputation can be associated with a 5-year survival of 15% to 35%.5
Palliation of regional disease is a less common problem for patients with STS of the extremity. Excellent local control is usually achieved with a combination of limb-sparing surgery and radiotherapy.6 Select patients, however, present with primary or recurrent tumors not amenable to a conservative surgical approach. ILP with melphalan alone or in combination with other agents can be an effective treatment in these patients: 40% to 90% of these patients experience a complete response (CR) to the therapy.7,8
A major disadvantage of ILP is that it is a technically complex and highly invasive surgical procedure. It is performed by open surgical cannulation of the vessels at the root of the extremity. Isolation of the limb is achieved by ligation of vascular collaterals and placement of a proximal tourniquet. The limb is then placed on a pump oxygenator, which provides a high-flow, hyperoxic, hyperthermic circuit. Isolated limb infusion (ILI) is conceptually similar but much easier to perform.9 Infusion catheters are placed percutaneously by interventional radiologists. There is no need for a pump team or cardiopulmonary bypass circuit. John Thompson and colleagues, of the Sydney Melanoma Unit, have reported response rates to ILI similar to those seen with ILP.10 For this reason, we conducted the first North American trial of ILI in patients who would normally be candidates for ILP. Our goal was to determine the efficacy and morbidity of this innovative new procedure.
|
METHODS
|
|---|
The trial was reviewed and approved by the Institutional Review Board of the Memorial Sloan-Kettering Cancer Center. Patients were accrued from September 1999 until August 2004. Eligible patients were those with American Joint Committee on Cancer stage IIIB or IIIC melanoma11 or patients with unresectable STS of the limb. Patients were considered for ILI if 95% of their disease was distal to the proximal third of the arm or leg and if they were reasonable candidates for angiographic catheter placement and general anesthesia. Patients with significant vascular disease or an absence of distal extremity pulses were excluded. In addition, patients were required to have a recent computed tomographic scan of the chest, abdomen, and pelvis to exclude metastatic disease. Patients who were previously treated with chemotherapy, radiotherapy, or immunotherapy were candidates as long as the ILI was performed at least 4 weeks after completion of prior therapy.
ILI was performed by using the same technique as had been previously described by investigators at the Sydney Melanoma Unit.9 Preinfusion biopsies were performed as necessary to document disease before treatment. The surgeon identified measurable lesions (index lesions) before infusion, and the size and location of these were recorded on a preinfusion data sheet. In patients with extensive small deposits of tumor, a description of the size and approximate number was sufficient for preinfusion documentation. Preinfusion photographs were used to facilitate documentation of disease burden and index lesion size and location. Deep lesions difficult to assess by surface inspection were measured and documented by using computed tomography or magnetic resonance imaging.
The following technique of ILI was used throughout the study and continues to be performed as described, except where indicated. On the day of infusion, by using the Seldinger technique, 6F multi–side hole angiographic catheters are placed from the contralateral groin. The catheters are positioned in the popliteal or brachial artery and vein of the involved limb, just above the knee or elbow joint. Angiography is performed to document the status of the runoff vessels. The position of the catheter tip within the artery and vein is then marked on the patient’s skin with indelible marker. The patient is given 5000 IU of heparin intravenously as a bolus and then begun on a constant infusion. Ten thousand units of heparin are diluted in 500 mL of normal saline. The constant infusion is administered into the arterial catheter via a pump at a rate of 18 IU/kg/ hour, but the minimum dose infused is 1250 IU/hour. The venous catheter is infused with normal saline at a rate that maintains catheter patency. The patient is then transported to the operating room holding area.
In the operating room, the patient is placed on a warming blanket, a proximal tourniquet is placed on the involved extremity, and temperature probes are placed on the skin and within the muscle of the limb. The limb is wrapped in cotton padding, and a pediatric warming blanket heated to 42°C is wrapped around the limb. Granisetron 1 mg and dexamethasone 20 mg are administered. This dose is effective prophylaxis against chemotherapy-induced nausea.
When the skin temperature of the limb reaches 37°C degrees, 60 mg of papaverine is injected into the arterial catheter, and the tourniquet is inflated to 350 mm Hg. Melphalan and dactinomycin are rapidly infused into the arterial catheter. Through most of the trial, a patient weight–based calculation was used to determine the dosage. This was done because it was logistically easier than calculating the limb volume by using water displacement.11 Melphalan was used at a dose of .6 to .8 mg/kg body weight for the lower extremity and .25 to .4 mg/kg body weight for the upper extremity. Dactinomycin was given at a dose 6 to 10 µg/kg body weight for the lower extremity and 2.5 to 4 µg/kg body weight for the upper extremity. Ranges were chosen to allow for modification of dosing during the trial according to observed toxicity. In practice, we used the middle dose of the range in the great majority of patients who underwent an initial infusion. For patients who underwent ILI of the lower extremity, this was .7 mg/kg body weight of melphalan and 8 µg/kg body weight of dactinomycin. More recently, the trial was modified to be consistent with a proposed national trial of ILI, and volume-based dosage was instituted after protocol amendment. Melphalan was used at a dose of 5 to 10 mg/L limb volume, and dactinomycin was used at a dose of 50 to 100 µg/L limb volume. The most common initial dosage of melphalan currently used is 7.5 mg/L limb volume (maximum, 100 mg), and that of dactinomycin is 75 µg/L limb volume (maximum, 500 µg/ L). Limb volume was calculated by taking circumferential limb measurements from the distal extremity to the proximal extremity at 1.5-cm longitudinal intervals. The last proximal measurement was taken at the inferior aspect of where the blood pressure cuff would rest at the time of ILI. An Excel-based (Microsoft Corp., Redmond, WA) software program provided by Dr. Douglas Tyler, Duke University Medical Center, was used to calculate the cylindrical volume of the limb.
The chemotherapy is administered via a blood warmer set at 40°C and bubble excluder. Once the chemotherapy is administered, the infusion begins by extracting 60 mL from the venous catheter and reinjecting it into the circuit to pass through the blood warmer and back into the limb via the arterial catheter (Fig. 1
). The chemotherapy is circulated through the limb for 20 minutes. During this period, the infusate becomes quite dark (hypoxic), and resistance in the circuit increases. At the completion of the infusion, the limb is infused with crystalloid (approximately .800–1 L), and an equal volume of effluent is extracted from the venous catheter and discarded. The tourniquet is then deflated, the arterial and venous catheters are removed, and protamine is administered. Manual compression is applied to the puncture sites for 20 minutes.
View larger version (30K):
[in this window]
[in a new window]
|
FIG. 1. Schematic diagram of the isolated limb infusion circuit. Chemotherapy is rapidly infused via the arterial catheter after tourniquet inflation. Arterial inflow occurs via a blood warmer and bubble excluder. Venous effluent is then manually extracted from the venous catheter with a large-bore syringe and reinjected into the circuit. The infusion occurs for 20 minutes. The limb is then flushed with crystalloid, the tourniquet is deflated, and the catheters are removed. IV, intravenous.
|
|
The leak rate from the isolated limb into the systemic circulation was determined in the first 15 patients (19 procedures). This was performed by using autologous red blood cells labeled with technetium. Background counts were established by administering 500 µCi into a large-bore peripheral catheter after the tourniquet was inflated. Once a steady state was established, as determined by a precordial gamma probe, 5 mCi of technetium-labeled red blood cells in a volume of 1 mL was injected into the arterial catheter before chemotherapy administration. Leakage from the infusion circuit into the systemic circulation was then monitored throughout the infusion via a precordial gamma probe and graphed to determine the percentage leak.
Patients were hospitalized for approximately 2 to 3 days. Bed rest was maintained the evening of the ILI, and the limb was elevated. Ambulation was resumed the next day. Complete blood count, creatine phosphokinase, and liver function tests were drawn on postoperative days 1 and 2, and the limb was examined in the hospital for at least 2 days. Low-dose subcutaneous heparin was administered after ILI until discharge from the hospital. Limb morbidity was assessed by using the Wieberdink scale—standard criteria used for ILP.12 Additional toxicity was graded by using the National Cancer Institute Common Toxicity Criteria (version 2.0). Morbidity was assessed in the hospital and at 2, 6, and 12 weeks after ILI in the outpatient clinic.
Response assessment was performed 3 months from the day of the last infusion. Patients whose disease progressed or who had less than a CR were considered for a second ILI, as deemed appropriate by the treating surgeon. A CR was defined as the disappearance of all clinical evidence of disease by 3 months from the last ILI. A partial response was defined as a 
50% decrease in the sum of the largest diameters of the index lesions without the appearance of new lesions. A minor response was defined as a 25% to 49% decrease in the sum of the largest diameters of the index lesions. Flattened, pigmented lesions remaining after ILI underwent biopsy to assess for viable tumor cells. Any patient with less than a minor response or a response of <3 months from the last infusion was considered a nonresponder. Patients who had new lesions or who had an increase in the size of their index lesion or other lesions after ILI were considered nonresponders. Patients who did not return for evaluation or who commenced other treatment before final response assessment were considered nonassessable.
|
RESULTS
|
|---|
In general, the technique was performed without complication in the study patients. As mentioned previously, one patient was accrued to the trial having undergone a previous axillary lymphadenectomy. The venous catheter could not be placed as a result of thrombosis of the axillary vein. The median leak rate in the first 19 procedures (15 patients) was 7.5% (range, 1%–29%).
Skin and muscle temperatures of the involved limb were 37°C before chemotherapy in most patients treated. The median temperature of the skin in the 32 procedures performed was 38.1°C (range, 33.1°C–40.0°C). The median muscle temperature was 37.9°C (range, 33.2°C–39.9°C). The median volume of infusate circulated through the limb was 1680 mL (range, 400–3300 mL) for the lower extremity (22 patients; 28 procedures). The two patients (four procedures) who underwent upper extremity ILI had a median infusate volume of 1365 mL (range, 150–2820 mL). A summary of the technical details of the procedure is presented in Table 1.
Efficacy of ILI
Twenty-five patients were accrued to the trial; 24 were treated, and 32 ILIs were performed. Of the 24 patients treated, there were 18 women and 6 men. The median age was 68 years (range, 44–85 years). Two patients had STS, and the remaining 22 patients had regional melanoma. Of the melanoma patients, 15 had stage IIIB disease, and 7 had stage IIIC disease. Eight patients had 2 ILIs, and the remaining 16 patients had 1 ILI. Two patients were treated with upper extremity ILI, and the remaining 22 patients had lower extremity disease.
Of the 24 treated patients, 22 were assessable. One patient with STS required amputation before 3 months after ILI because of pain that was present before the infusion and was, therefore, not assessable for response. One patient with regional melanoma underwent injection therapy before 3 months from ILI.
Eleven patients had a significant clinical response to the treatment. Of these, five patients had a CR, and six patients had a partial response. One patient had a minor response, and 10 patients were nonresponders. The median duration of CR was 12 months (range, 6–32 months). The median duration of partial response was 11 months (range, 6–43 months). Examples of patients with complete and partial responses are provided in Figs. 2 and 3, respectively.
View larger version (70K):
[in this window]
[in a new window]
|
FIG. 2. A 61-year-old woman with local recurrence 3 months after treatment of a deep primary melanoma of the ankle. The tumor is shown before infusion (left) and 3 months after the isolated limb infusion (right). The patient had a complete response to the treatment and remained without recurrence for 2.5 years.
|
|
View larger version (94K):
[in this window]
[in a new window]
|
FIG. 3. A 48-year-old man with recurrent melanoma of the right upper extremity and multiple dermal and subdermal metastases before treatment (left). The patient had a partial response to the treatment (middle), residual disease was excised, and he remains without evidence of recurrence 43 months after his second isolated limb infusion (right).
|
|
Morbidity
The morbidity of the procedure was quite acceptable, and no patient experienced long-term (>3 months) grade III morbidity. No patient required amputation. No patient required fasciotomy for compartment syndrome.
Morbidity assessment was performed in the hospital and at 2, 6, and 12 weeks after ILI. Table 2 summarizes the median morbidity scores at each interval, as well as the maximum score reported. In general, morbidity was quite mild in the hospital; most patients experienced some mild swelling and erythema. Thirteen patients (54%) reported nausea on the initial morbidity form, which was completed on postoperative day 1 or 2. The highest morbidity scores were noted at 2 weeks after ILI, and most patients experienced some pain or discomfort, swelling, and erythema (Fig. 4). This was resolved in most patients by 6 weeks, and there were very few residual side effects of the ILI by 3 months, with the exception of some increase in edema of the limb and hyperpigmentation of the extremity (Fig. 5). The median morbidity scores after ILI at each time point are listed in Table 2.
View larger version (49K):
[in this window]
[in a new window]
|
FIG. 4. The side effects of treatment peaked at 2 weeks after isolated limb infusion in most patients. The graphs depict the percentage of patients experiencing symptoms (y-axis) to within each degree category (x-axis) at each time point. Edema and erythema were the most common complications experienced. Mild to moderate disturbances in mobility also occurred. Limb pain at rest was uncommon, but mild pain with motion occurred in many patients.
|
|
View larger version (45K):
[in this window]
[in a new window]
|
FIG. 5. A 61-year-old man 5 months after his second isolated limb infusion for unresectable soft tissue sarcoma. Note the persistent hyperpigmentation of the treated extremity.
|
|
|
DISCUSSION
|
|---|
The administration of regional chemotherapy by using ILP was initially developed in New Orleans in the mid 1950s.13 The technique was based on the proposal that vascular isolation and perfusion of the extremity with chemotherapy would allow regional drug concentrations several orders of magnitude higher than could be attained with systemic administration. Today, ILP is performed at a few tertiary referral centers in Europe and the United States. Melphalan is the standard drug used in ILP, and it is given under hyperthermic conditions.14 ILP with hyperthermic melphalan results in CRs in up to 50% of patients with advanced regional melanoma and partial responses in a further 30%.14 ILP is technically complex, however, and associated with significant morbidity and cost. The procedure necessitates autologous blood donation or transfusion of allogeneic units. Vascular isolation of the extremity for ILP requires surgical exposure and open cannulation of the artery and vein to the extremity. The extremity is then placed on bypass, and this necessitates the presence of a perfusion team and a cardiopulmonary bypass machine in addition to surgical and anesthesia staff. Significant regional toxicity with skin necrosis, peripheral neuropathy, or both can occur. Vascular catastrophe necessitating arterial reconstruction or amputation is uncommon but may occur.15
ILI was developed at the Sydney Melanoma Unit by Thompson et al.9,16 as a simple alternative to ILP. Percutaneously placed catheters replace open surgical cannulation, and the chemotherapy is recirculated manually so that no pump oxygenator is needed. Operating room time is approximately 1 hour, compared with the 4 or 5 hours needed to perform ILP. Most importantly, Thompson et al. report that ILI has an efficacy similar to that of ILP. They reported significant clinical responses in 135 patients treated with ILI, with 41% complete and 44% partial responses.10 The morbidity of the procedure, in their experience, is low.9,16
We observed a lower response rate in our initial experience with ILI. Several factors may have contributed to this. Patient selection is likely to play a significant role in obtaining more consistent responses to the treatment. We did not have an established regional chemotherapy program in place when the study began, and patients referred from outside hospitals tended to have very advanced disease. Lindner et al. have reported a higher response rate in patients with a lower burden of disease.10 Many or most of our patients had a significant limb tumor volume. In addition, our general policy has been to use less invasive techniques, such as surgical excision, when the lesions are small and few. Regional control with ILI is more likely if patients are treated before their disease is bulky or extensive.
Another factor that may have contributed to the lower response rate observed in our initial experience with ILI when compared with that reported by Thompson et al.9 is the relatively short infusion and tourniquet time we used. We chose a 20-minute infusion because this was what was used by Thompson et al. when we began our study. This undoubtedly contributed to the relatively high median pO2 and pH recorded at the end of the infusion in the study patients (Table 1
). The investigators at the Sydney Melanoma Unit subsequently reported that response rates seemed to be associated with longer tourniquet times and modified the procedure to increase the length of the infusion and the overall tourniquet time.17 In addition, as we gained experience with ILI, we were able to achieve normothermia in the treated limb more consistently. Several of the early patients on our trial had suboptimal limb temperatures at the onset of treatment.
Our results represent the first North American experience with this new technique. We observed significant clinical responses in half of the patients treated. In addition, we confirmed the low morbidity of the procedure. ILI has effectively replaced ILP at our institution and is likely to gain increasing use in the United States and Europe. Because of its ease of administration and low morbidity, ILI will certainly facilitate testing new and promising regional chemotherapy agents individually or in combination.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank John Thompson of the Sydney Melanoma Unit for his invaluable assistance with this study. We thank Ethan Kavet for expert photographic assistance and Liza Marsh for editorial expertise. In addition, we thank Drs. Ann Covey, Lynn Brody, and George Gertrajdman of the Interventional Radiology Service, Department of Radiology, Memorial Sloan-Kettering Cancer Center. Finally, we thank Dr. Steve Larson and the staff of the Division of Nuclear Medicine, particularly Dr. Sam Yeh, for their assistance.
Received for publication May 10, 2005.
Accepted for publication January 20, 2006.
|
REFERENCES
|
|---|
- Karakousis CP, Balch CM, Urist MM, et al. Local recurrence in malignant melanoma: long-term results of the multiinstitutional randomized surgical trial. Ann Surg Oncol 1996; 3:446–52.[Abstract]
- Cascinelli N, Bufalino R, Marolda R, et al. Regional non-nodal metastases of cutaneous melanoma. Eur J Surg Oncol 1986; 12:175–80.[Medline]
- Cohen MH, Jessup JM, Felix EL, et al. Intralesional treatment of recurrent metastatic cutaneous malignant melanoma: a randomized prospective study of intralesional Bacillus Calmette-Guerin versus intralesional dinitrochlorobenzene. Cancer 1978; 41:2456–63.[CrossRef][Medline]
- von Wussow P, Block B, Hartmann F, Deicher H. Intralesional interferon-alpha therapy in advanced malignant melanoma. Cancer 1988; 61:1071–4.[CrossRef][Medline]
- Jaques DP, Coit DG, Brennan MF. Major amputation for advanced malignant melanoma. Surg Gynecol Obstet 1989; 169:1–6.[Medline]
- Yang JC, Chang AE, Baker AR, et al. Randomized prospective study of the benefit of adjuvant radiation therapy in the treatment of soft tissue sarcomas of the extremity. J Clin Oncol 1998; 16:197–203.[Abstract/Free Full Text]
- Schraffordt KH, Eibergen R, Oldhoff J, et al. Isolated regional perfusion in the treatment of soft tissue sarcomas of the extremities. Clin Oncol 1976; 2:245–52.[CrossRef][Medline]
- Hoekstra HJ, Schraffordt KH, Molenaar WM, et al. Results of isolated regional perfusion in the treatment of malignant soft tissue tumors of the extremities. Cancer 1987; 60:1703–7.[CrossRef][Medline]
- Thompson JF, Kam PC, Waugh RC, et al. Isolated limb infusion with cytotoxic agents: a simple alternative to isolated limb perfusion. Semin Surg Oncol 1998; 14:238–47.[CrossRef][Medline]
- Lindner P, Doubrovsky A, Kam PC, et al. Prognostic factors after isolated limb infusion with cytotoxic agents for melanoma. Ann Surg Oncol 2002; 9:127–36.[Abstract/Free Full Text]
- Balch CM, Buzaid AC, Soong SJ, et al. Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. J Clin Oncol 2001; 19:3635–48.[Abstract/Free Full Text]
- Wieberdink J, Benckhuysen C, Braat RP, et al. Dosimetry in isolation perfusion of the limbs by assessment of perfused tissue volume and grading of toxic tissue reactions. Eur J Cancer Clin Oncol 1982; 18:905–10.[CrossRef][Medline]
- Krementz ET, Carter RD, Sutherland CM, et al. Regional chemotherapy for melanoma. A 35-year experience. Ann Surg 1994; 220:520–34.[Medline]
- Brady M, Coit D. Regional chemotherapy for melanoma. In: Markman M, ed. Regional Chemotherapy. Totowa NJ: Humana Press, 1999.
- Taber SW, Polk HC Jr. Mortality, major amputation rates, and leukopenia after isolated limb perfusion with phenylalanine mustard for the treatment of melanoma. Ann Surg Oncol 1997; 4:440–5.[Abstract]
- Thompson JF, Hunt JA, Shannon KF, et al. Frequency and duration of remission after isolated limb perfusion for melanoma. Arch Surg 1997; 132:903–7.[Abstract]
- Lindner P, Doubrovsky A, Kam PC, Thompson JF. Prognostic factors after isolated limb infusion with cytotoxic agents for melanoma. Ann Surg Oncol 2002; 9:127–36.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
