The predictive value of lymphoscintigraphy for nodal metastases of cutaneous melanoma.

Lymphoscintigraphy is used to identify ambiguous drainage patterns in cutaneous melanoma of the head, neck and trunk. This study evaluated the efficacy and reliability of lymphoscintigraphy to identify the drainage nodes in 51 patients undergoing both elective and therapeutic lymphadenectomy over a seven-year period. All 13 patients who had lymph node metastases during this follow-up period had the metastatic disease in the very lymph node basins identified by lymphoscintigraphy. Most significantly, none of the 51 patients had metastatic disease in lymphatic basins that were not previously identified by lymphoscintigraphy.

Safety and efficacy of isolated perfusion of extremities for recurrent tumor in elderly patients.

BACKGROUND: The treatment of bulky recurrent melanotic lesions of extremities with isolated limb perfusion with high dose chemotherapy offers palliation in a number of patients. However, the question is raised whether these major surgical procedures are too risky to warrant performing them in elderly patients. METHODS: Sixty-seven limbs were perfused in 60 patients with various drugs from 1976 through 1996 (35, imidazole carboxamide; 7, cisplatin; 20, carboplatin; 5, thiotepa). Among the 67 perfusions, 20 were in patients aged 70 years and older. Perfusion was performed for 16 upper extremities and 51 lower extremities by using the pump oxygenator for 1 hour. RESULTS: A total of 19 complications were noted after a total of 14 of the 67 perfusions (21%) (postoperative edema, 5; seroma, 4; wound separation or infection, 9; nonfatal pulmonary embolus, 1). The complications in 4 of 20 perfusions in the older patients (20%) were less than in 15 of 47 perfusions in the younger patients (32%). Among the 17 patients older than 70 years of age who were treated with perfusions for recurrent disease, four patients (24%) are alive with no evidence of disease (NED) for a median of 29 months (range, 16 to 80 months); one patient is now more than 6 years with NED after her third perfusion for repeated in-transit disease. Another 2 of 17 patients (12%) are alive with disease for a median of 89 months (range, 54 to 123 mos). The remaining 11 patients (64%) are dead of their disease. These data are comparable to the control rates in the group of younger patients in the study. Overall, half of all the patients (14 of 28) who died of their disease in both groups had maintained local control of their involved extremities. CONCLUSIONS: Aggressive treatment in selected patients with regional isolated perfusion of limbs for melanoma can lead to significant palliation of symptoms and salvage of limbs with adequate disease-free control and occasional survival benefit. This series of patients was associated with meaningful disease control and with few serious complications. Perfusions are tolerated well by patients in their 70s and 80s; therefore advanced age is not a contraindication to this procedure in carefully selected patients.

A phase II study of interferon-alpha 2b with dacarbazine, carmustine, cisplatin and tamoxifen in metastatic melanoma.

A phase II trial was conducted to determine the efficacy and toxicity of the addition of interferon-alpha 2b (IFN-alpha) to the chemotherapy combination of dacarbazine (DTIC), carmustine (BCNU), cisplatin and tamoxifen (DBCT), in patients with stage III or IV melanoma. Treatment consisted of DTIC 220 mg/m2 and cisplatin 25 mg/m2 intravenously on days 1-3, BCNU 100 mg/m2 IV on day 1 only, tamoxifen 20 mg orally twice daily and IFN-alpha 5 x 10(6) units/m2 subcutaneously on days 1-5. Cycles were repeated every 4 weeks. All patients received a loading dose of tamoxifen 100 mg orally twice daily for 5 days before the first course of therapy. Of the 24 patients treated, three (13%) achieved a complete response (CR) and six (25%) a partial response (PR), for an overall response rate of 38% (95% confidence interval, 17-58%). Two patients, one who achieved a clinical CR and one a PR, had pathologically confirmed complete responses. Severe myelosuppression occurred in 47% of cycles and constitutional symptoms were common. Overall, the addition of IFN-alpha to the DBCT regimen did not appear to enhance the response rate and may have increased toxicity.

Regional isolated perfusion of extremities for melanoma: a 20-year experience with drugs other than L-phenylalanine mustard.

Recurrent melanoma of the extremities can lead to bulky symptomatic lesions that become difficult management problems. Treatment of these tumors with isolated limb perfusion with high dose chemotherapy may offer palliation in a number of patients. Unfortunately, the most commonly used drug, L-phenylalanine mustard, has been known to have significant associated tissue toxicity. Therefore, during the years 1976-1995, we perfused 67 limbs in 60 patients with various other drugs: 36 with dimethyltriazeno imidazole carboxamide, 6 with cisplatin, 20 with carboplatin, and 5 with thiotepa). Perfusion was performed for 16 upper extremities and 51 lower extremities using the pump oxygenator for 1 hour. Among the 60 patients, 17 were treated prophylactically for high-risk melanoma, whereas 43 were treated for local and in-transit recurrences. The technique of perfusion successfully isolated the limbs from the systemic circulation: the median leaks over time were 0.5-1.6 percent for the upper extremities, and 0.2-7.5 percent for the lower extremities. Among the 43 patients treated with therapeutic isolated limb perfusion, 11 patients (26 percent) are alive with no evidence of disease for a median of 58 months (range: 8 months to 17 years 9 months), and another 5 patients (12 percent) are alive with recurrence for a median of 45 months (range: 27 months to 10 years 7 months). Four patients required two perfusions, and two patients required three perfusions (one patient has no evidence of disease 6 years after her third perfusion for recurring in-transit disease). There were 19 complications noted after 14 of the 67 perfusions (21%): postoperative edema, 5; seroma, 4; wound separation/infection, 9; and nonfatal pulmonary embolus, 1. In our experience, aggressive treatment in selected patients with regional isolated perfusion of limbs for melanoma has provided meaningful palliation and salvage of the limbs with adequate disease-free control, and occasional survival benefit. This regional treatment modality is associated with meaningful control and with few serious complications, especially when compared with studies using L-phenylalanine mustard. This series illustrates the safety of controlling limb recurrence with this technique, even with repeat perfusions in the same patient.

Immunomodulation and enhancement of antitumor activity by co-administration of 1,3-bis(2-chloroethyl)-1-nitrosourea and thymidine.

The antitumor activity of 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) has been shown previously to be enhanced markedly by the co-administration of pyrimidine deoxyribonucleosides (Lin and Prusoff, Cancer Res 47: 394-397, 1987). In the present study, we examined the cellular mechanisms underlying the augmentation effect of thymidine, one of the pyrimidine deoxyribonucleosides. It was found that thymidine did not increase the cytotoxicity of BCNU for B16/F10 melanoma cells in vitro. Instead, thymidine appeared to produce modulatory actions on the immune system of the tumor-bearing mice. More than 40% of the BCNU/ thymidine-cured mice specifically rejected secondary rechallenge with the B16/F10 tumor. Furthermore, these cured mice developed extensive depigmentation of their natural black hair, suggesting immune reactions to normal melanocytes. When spleen cells from normal mice were treated with BCNU alone, their response to T-cell mitogen phytohemagglutinin was suppressed markedly. This suppression was ablated by co-administration of BCNU with thymidine. Such BCNU/thymidine treatment also augmented the activity of tumor-specific cytotoxic T-cells in tumor-bearing mice. Taken together, these results suggest that the enhanced antitumor activity of combined BCNU and thymidine may result from the action of thymidine on the immune effector mechanisms, which facilitate the development of antitumor immune responses in the presence of immunosuppression induced by BCNU.

Prolonged survival in patients with advanced melanoma treated with neoadjuvant chemotherapy followed by resection.

Patients with metastatic melanoma (AJCC stage IV) generally have a very poor prognosis (median survival, 6-8 months). Combination chemotherapy is often the treatment of choice. Despite the higher response rates of a number of new combination regimens, the median duration of response ranges from 6 to 9 months, with no significant survival advantage. To evaluate whether surgical resection of residual metastases after chemotherapy can improve survival, we conducted a retrospective analysis of all patients enrolled on various chemotherapy protocols for metastatic melanoma at Yale between March 1987 and March 1993. We identified 16 patients who underwent surgical resection of residual disease after receiving one to four cycles of combination chemotherapy. Sites of metastases included regional and distant lymph nodes, skin, subcutaneous tissue, lung, and liver. No patients had brain or bone metastases at the initiation of therapy and performance status (PS) was 0.1. Follow-up for these 16 patients ranges from 14 to 62 months (median, 35 months) from the start of chemotherapy. All 16 patients had either complete responses (CR = 3) or partial responses (PR = 7), or stable disease (SD = 6) after chemotherapy. Eleven patients are currently alive; 10 remain free of disease. A comparison group (control) of 14 patients was identified from the same retrospective analysis using similar clinical characteristics such as sites of metastases, PS, and cycles of chemotherapy. No patients underwent surgical resection either because of patient refusal or concomitant medical problems. None had evidence of disease progression (CR = 3, PR = 2, SD = 9) at the completion of chemotherapy. However, duration of survival in this group from the start of chemotherapy ranged from 4 to 45 months (median, 11.5 months). Twelve patients have died and 2 are currently alive, 1 of which has recurrent disease. Based on the comparison of these two groups, this report suggests that patients treated with neoadjuvant chemotherapy may benefit from surgical resection of residual metastases, and this approach should be studied further.

Single institution experience with recombinant gamma-interferon in the treatment of patients with metastatic renal cell carcinoma.

We report the clinical course of eight patients with metastatic renal cell carcinoma (RCC) who were treated with recombinant gamma-interferon (Immuneron) as part of a phase II-III study comparing the safety and efficacy of gamma-interferon with that of medroxyprogesterone acetate (Depo-Provera). There were no objective responders among the eight patients treated with recombinant gamma-interferon at an i.v. dose of 1 mg/m(2) daily for five days every other week for four weeks then 1 mg/m(2) three times a week given every other week until there was documented disease progression or complete response (CR). Overall median survival was 17.3 months (range 1.4 to 184). The major side effects of treatment included fever/chills (75%), mild anorexia and fatigue (75%), nausea/vomiting (80%), leukopenia (38%), and abnormal liver function tests (25%). There were no life-threatening side effects observed. At our institution, in a random cohort of eight patients with metastatic RCC, recombinant gamma-interferon when given at a dose of 1 mg/m(2) per day given three times per week on an every other week schedule yields no clinical antitumor activity. A review of the literature on the use of gamma-interferon for metastatic RCC suggests that low-dose combination therapy with other cytokines may yield the best response-to-side effect ratio. Higher doses yield more responses but an added cost of more toxicity.

Recombinant beta-interferon in the treatment of patients with metastatic renal cell carcinoma.

We report on the clinical course of 15 patients with metastatic renal cell carcinoma (RCC) who were treated with recombinant beta-interferon as part of a phase I-II study. There were no objective responders among the 15 patients treated with recombinant beta-interferon at an i.v. dose escalating from 90 X 10(6) U given three times a week until there was documented disease progression or complete response (CR). Overall median survival was 24 months. One patient refused further treatment after 7 weeks. The major side effects of treatment included cardiovascular events (20%), mental status change requiring cessation of drug (6.7%), and grade 3 headaches/myalgias (26.7%). There were no life-threatening side effects observed; however, cardiac events led to the termination of treatment in three patients. Other minor toxicities included fatigue (46.7%), proteinuria (60%), diarrhea (6.7%), nausea and vomiting (13.3%), persistent fever (6.7%) and transient visual disturbance (6.7%). Thus, at our institution, in a cohort of 15 patients with metastatic RCC, recombinant beta-interferon when given i.V. at a dose < or equal to 720 X 10(6) U three times per week, yielded no clinical antitumor activity. A review of the literature on the use of beta-interferon for metastatic RCC suggests that there may be some efficacy, but our experience with escalating i.v. doses < or equal to 720 X 10(6) U given three times a week does not support it. Moreover, at these doses, one may find serious cardiovascular events although further studies need to be done in order to clearly define dose-related side effects as well as optimal efficacy-to-toxicity ratio.

Multiple primary melanomas: data and significance.

Cancer statistic reports show that the incidence of melanoma has increased each decade. It is now estimated that approximately 5 percent of the patients with primary cutaneous melanoma will develop another primary melanoma in their lifetime. This report describes the information gathered from 27 patients at the Yale Melanoma Unit who have developed 59 individual primary melanomas; 22 of the patients developed a second primary melanoma, and 5 patients each developed three primary melanomas. In 8 patients (30 percent), the second primary melanoma was diagnosed within 1 month of the first malignancy and was considered synchronous. The remaining 24 melanomas in the 19 patients presented subsequently: 4 (17 percent) within the first year, 7 (29 percent) during the second year, and 13 (54 percent) beyond the second year of the first diagnosis. Although the thickness of the initial melanoma ranged from 0.2 to 6.0 mm, all subsequent melanomas were either in situ or less than 1.0 mm in thickness. This study shows that patients who developed more than one melanoma invariably had thin subsequent lesions. The implications of the multiple melanomas are not a poorer prognosis, but rather that the patients' prognosis is the same as that of the original, or thickest, melanoma.

Prognostic factors for survival in patients with metastatic renal cancer treated with biological response modifiers.

PURPOSE: Clinical characteristics prognostic of survival in patients with metastatic renal cell carcinoma treated with biological response modifiers are poorly understood. Understanding these prognostic features may help with better stratification of patients in clinical trials and define further appropriate treatment for each prognostic subgroup. MATERIALS AND METHODS: A retrospective study of 84 patients with recurrent or metastatic renal cancer was conducted to identify prognostic factors for survival in patients who received biological response modifiers (alpha-interferon, beta-interferon, gamma-interferon and interleukin-2). RESULTS: Univariate analysis identified Eastern Cooperative Oncology Group (ECOG) performance status (1 versus 0, p < 0.001), bone metastasis (p = 0.008), recent weight loss (greater than 10% of total body weight versus no loss, p = 0.028), history of nephrectomy (no versus yes, p = 0.025), recurrence at the renal bed (p = 0.043) and sarcomatoid histology (yes versus no, p < 0.001) as important prognostic indicators. Multivariate analysis of prognostic factors in this patient population indicated that ECOG performance status, sarcomatoid histology and bone metastasis were most significant, while other factors were less significant (p > 0.05) after adjusting for ECOG performance status and sarcomatoid histology. Based on the total positive number of 5 risk factors defined previously the study population separates into 3 risk groups, with a median survival from the low to high risk groups of 14.4, 10.9 and 1.3 months, respectively. Prognostic scores based only on ECOG performance status, sarcomatoid histology and bone metastasis allowed for stratification of our patients into 3 distinct groups with median survivals of 18.6, 8.4 and 3.8 months, which were also predictive of survival (p < 0.05). CONCLUSIONS: Risk factors of ECOG performance status, sarcomatoid histology, bone metastasis, history of nephrectomy, recent weight loss and recurrence at the renal bed are predictive of survival in patients treated with biological response modifiers. In addition to previous findings of prognostic factors in renal cancer patients treated with chemotherapy, we identified sarcomatoid histology as an important risk factor in patients treated with biological response modifiers.

Biochemical modulation of 5-fluorouracil with brequinar: results of a phase I study.

Biochemical modulation can increase the efficacy of 5-fluorouracil (5-FU). Pizzorno et al. have previously shown that brequinar, a de novo pyrimidine synthesis inhibitor, enhances the antitumor effect of 5-FU in vivo [Cancer Res 52: 1660-1665, 1992]. On the basis of their data, we conducted a phase I study of brequinar in combination with 5-FU in patients with refractory solid tumors. The initial dose (100 mg/m2) of brequinar was raised in 100-mg/m2 increments in cohorts of three assessable patients. The initial dose of 5-FU was 500 mg/m2, but escalation was allowed in patients who showed no significant toxic reaction. Brequinar was administered over 1 h and 5-FU over 2 h starting 18-20 h after the initiation of infusion of brequinar. Treatments were repeated weekly. Responses were evaluated after 4 weeks (one course) and then every 8 weeks thereafter. Pharmacokinetics of brequinar and determination of plasma uridine levels were performed in at least three patients at each dose level. Of the 25 patients registered in the study, 21 were assessable for toxicity studies. The dose of brequinar was escalated up to 600 mg/m2. In addition, the dose of 5-FU was increased to 600 mg/m2 as a result of a lack of a significant toxic reaction in the first nine patients. No objective responses were observed. One patient developed grade 3 stomatitis, and one developed grade 3 esophagitis at the 400 and 600 mg/m2 dose of brequinar, respectively. Brequinar produced a dose-dependent decrease in plasma uridine levels at doses up to 500 mg/m2. No additional decrease in plasma uridine occurred with higher doses of brequinar, thus suggesting a plateau effect. This observation prompted us to terminate the study before reaching the maximum tolerated dose of brequinar. Our data indicate that brequinar in doses > or = 400 mg/m2 results in significant biochemical modulation. The lack of toxicity seen at these doses of brequinar suggests that the initial dose of the effector agent 5-FU should be increased in future studies.

Correlation between lipid-associated sialic acid and tumor burden in melanoma.

Recent studies have suggested that lipid-associated sialic acid (LSA) may be a useful tumor marker for monitoring patients with melanoma, but the relationship between LSA and tumor burden has not been previously studied. We therefore examined LSA levels in 240 patients of whom 169 had no clinical evidence of disease (NED) and 71 had metastatic disease. There was a statistically significant difference in LSA levels in patients with NED compared with metastatic disease as well as those with high tumor burden compared with low or intermediate tumor burden. There was no difference between the groups with low and intermediate tumor burden. An LSA level of 25 mg/dl provided a positive predictive value of 70% and a negative value of approximately 80%. Our data show that LSA levels correlate with tumor burden in patients with melanoma.

Neuron-specific enolase as a tumor marker in metastatic melanoma.

Neuron-specific enolase (NSE) has been shown by some investigators to be a useful tumor marker for melanoma, but the relationship between NSE and tumor burden has not been extensively studied. We therefore examined NSE levels in 240 patients of whom 169 had no clinical evidence of disease (NED) and 71 had metastatic disease. There was no statistically significant difference in NSE levels in patients with NED compared with metastatic disease as well as those with high tumor burden compared with low or intermediate tumor burden. In addition, the mean absolute values of NSE, despite a slight elevation with tumor burden, were within the normal range (< 20 ng/ml). Our data suggest that NSE levels measured by the method used in our study are of no benefit in melanoma.

Role of computed tomography in the staging of primary melanoma.

PURPOSE: To evaluate the role of body computed tomography (CT) for the staging of patients with early melanoma. PATIENTS AND METHODS: A total of 151 new patients with American Joint Committee (AJC) clinical stage I, II, and III melanoma who received a CT scan of at least the chest and abdomen are the subject of this study. CT scans considered suspicious for metastases were reviewed again by one of the investigators (A.McB.C.). RESULTS: Of 151 patients, 63 had AJC clinical stage I, 61 stage II, and 23 stage III disease. In addition, one patient each had primary melanoma of the anal canal, esophagus, or vulva. Twenty-nine (19%) of 151 patients had a CT scan that was considered suspicious for metastases. The most common radiologic findings were single hepatic, and single or multiple pulmonary nodules. Of these 29 patients with suspicious scans, 24 subsequently proved to have benign processes by biopsy or follow-up studies, three had second primary tumors (well-differentiated lymphocytic lymphoma, Hodgkin's disease, and renal cell carcinoma), and only two were found to have metastatic melanoma. Of these two patients, one had regional nodal disease (unsuspected on physical examination) and one had distant nodal metastases. CONCLUSION: Body CT is not a useful imaging study in the detection of occult metastases in patients with primary melanoma. Although body CT commonly shows suspicious radiologic abnormalities in patients with early melanoma, these abnormalities most likely represent benign processes or a second primary tumor, rather than metastatic melanoma. The value of body CT in patients who present with nodal metastases needs further study.

Receptor-directed focusing of lymphokine release by helper T cells.

The interaction between helper T cells and B cells, leading to the production of antibody to thymus-dependent antigens, was the first cell interaction clearly defined in the immune system; it remains both paradigmatic and controversial. Two requirements of this interaction, that the helper cell (TH) and the B cell must recognize antigenic determinants that are physically linked, and that the TH and the B cell must share genes encoding major histocompatibility complex (MHC) class II molecules, led to the concept that TH-B interaction required an intimate physical association of the two cell types. But in vitro studies have shown that TH can be replaced by soluble, antigen-nonspecific factors, capable of activating any B cell to secrete antibody. We have previously proposed that the requirements for TH-B contact might result from TH cells releasing their lymphokines in a polar fashion directed at that portion of the cell membrane where T-cell receptor cross-linking is actually occurring. Using an artificial monolayer of a cloned helper T-cell line, we show that lymphokines are released preferentially over the area of receptor cross-linking under conditions of limited TH-cell activation. Thus, it appears that one important aspect of the specificity of TH-B cell interactions is the receptor-directed polar release of helper lymphokines.

The effect of hydrocortisone on the production of gamma-interferon and other lymphokines by human peripheral blood mononuclear cells.

Concentrations of hydrocortisone as low as 0.08 microgram/ml significantly reduced the yields of gamma-interferon (IFN-gamma) when phytohemagglutinin (PHA) or concanavalin A (ConA) were used as inducers; however, when staphylococcal enterotoxin A was utilized, higher concentrations (5.0 micrograms/ml) were required to achieve the same effect. Yields of interleukin-2 (IL-2) and lymphotoxin were also found to be sensitive to the effects of the steroids, but expressions of TAC antigen was not generally affected by these agents. In contrast to the effects of steroids on cell proliferation, lymphokine production remained suppressed after steroid withdrawal. Hydrocortisone appeared to influence the concentrations of cyclic nucleotides following lectin stimulation, but attempts to correct these alterations or to add exogenous IL-2 failed to restore lymphokine production to normal levels. Addition of the calcium ionophore A23187 partially restored IFN-gamma production. We conclude that the effects of corticosteroids on the yields of lymphokines, including IFN-gamma, are profound. The depression of lymphokine production appears to be associated with a number of alterations in the cell, including depression of protein synthesis, alterations in cyclic nucleotides, and diminution of the production of cofactors necessary for IFN-gamma production. Enhancement of the flux of calcium into the cell may restore some of the ability to produce IFN-gamma.

Lateral electrophoresis and diffusion of Concanavalin A receptors in the membrane of embryonic muscle cell.

A uniform electric field of 10 V/cm applied across the surface of embryonic toad Xenopus muscle cells results in the asymmetric accumulation of concanavalin A (Con A) receptors toward one side of the cells within 10 min, as visualized by postfield fluorescent Con A labeling. This field produces an extracellular voltage difference of 20 mV across these 20-microns wide cells. The effect is reversible in two respects: (a) Additional exposure of the cell to the same field of opposite polarity for 10 min completely reverses the asymmetric accumulation to the other side of the cell. (b) Relaxation occurs after the removal of the field and results in complete recovery of the uniform distribution in 30 min. Both the accumulation and the recovery movements are independent of cell metabolism, and appear to be electrophoretic and diffusional in nature. The threshold field required to induce a detectable accumulation by the present method is between 1.0 and 1.5 V/cm (corresponding to a voltage difference of 2-3 mV across a 20-microns wide cell). The electrophoretic mobility of the most mobile population of nonliganded Con A receptors is estimated to be about 2 x 10(-3) microns/s per V/cm, while their diffusion coefficient is in the range of 4-7 x 10(-10) cm2/s. Extensive accumulation of the Con A receptors by an electric field results in the formation of immobile aggregates. The Con A receptors appear to consist of a heterogeneous population of membrane components different in their charge properties, mobility, and capability in forming aggregates.