Harold J. Wanebo

Malignant melanoma of the extremities: a clinicopathologic study using levels of invasion (microstage).

A clinicopathologic study was done in 151 patients with malignant melanoma of the extremities who were surgically treated in Memorial Hospital and had 5– to 9‐year followup. Microstaging was done according to the depth of invasion, as determined by Clarks levels as well as by direct measurement. This was related to treatment and to clinical course. There was a correlation between the depth of invasion by Clarks levels and the incidence of lymph node metastases in patients with Stage I melanoma who had elective node dissection. The incidence of nodal metastases was 4% for Level II, 7% for Level III, 25% for Level IV, and 70% for Level V. There was a correlation between Clarks level of invasion and survival after surgery. The 5‐year cure rate was 100% for Level II, 88% for Level III, 60% for Level IV, and 15% for Level V melanoma. The presence of nodal metastases augured a much worse prognosis than Clarks level per se. In patients with Level IV melanoma, the 5‐year cure rate was 82% in patients with negative nodes and 27% in those with nodal metastases after elective node dissection. Microstaging primary melanoma according to Clarks levels serves as a useful standard with which to compare surgical results. In this series of extremity melanomas there was no difference between local recurrence and lymphadenectomy for Level II melanoma. For Level III and Level IV melanoma, wide excision and lymphadenectomy gave higher cure rates than wide excision only, both at 5 and 9 years after surgery. The results were significant only for patients with Level III, however. Use of the measured depth of invasion added significant clinicopathologic information. The incidence of nodal metastases at elective node dissection was 5 to 9% for melanoma showing 0.6 to 2.0 mm of invasion, 22% for melanoma measuring 2.1 to 3.0 mm, and 39% for melanoma invading beyond 3.0 mm. The 5‐year cure rate was 100% for melanoma measuring less than 1.0 mm, 83% for melanoma invading 1.1 to 2.0 mm, 58% for lesions measuring 2.1 to 3.0 mm, and 55% for melanoma invading over 3.0 mm. The microstage technique combining Clarks levels and the measured depth of invasion has an important use as a prognostic index and as a standard upon which to select treatment for primary melanoma of the extremities.

Hepatic resection for metastasis from colorectal cancer.

SummaryTwenty-five patients who had hepatic metastases from carcinomas of the colon and rectum had resection for cure at Memorial Hospital, with a determinate five-year survival rate of 40 per cent, and 10-year survival rate of 28 per cent. Most of the hepatic metastatic lesions were solitary, small, and peripheral, and were treated with simple wedge resection. These favorable results justify an aggressive approach to solitary metastatic lesions in the liver.

Immunologic reactivity and prognosis in breast cancer.

The relationship of immune reactivity to the stage of disease and to prognosis within stages was studied in 202 women with breast cancer. Patients were staged according to the following system: those with primary operable breast cancer were classified according to the presence or absence of regional lymph node metastases; those with advanced breast cancer were classified according to the predominant site of metastases: (1) those with skin, soft tissue, or nodal metastases, (2) those with bone metastases, and (3) those with visceral organ metastases. A significant linear trend of decreasing DNCB reactivity was seen with increasing extent of disease (P < 0.01). The results of other tests of immune reactivity, including intradermal skin tests with microbial antigens, and, in patients with advanced breast cancer only, absolute lymphocyte counts, lymphocyte proliferative responses to mitogens and microbial antigens, serum immunoglobulin levels, and T and B‐cell counts, did not change significantly with increasing extent of disease. No significant difference in recurrence distributions was seen when all primary operable breast cancer patients with positive and negative DNCB responses were compared. However, when the presence or absence of lymph node involvement was taken into account, a trend to earlier recurrence for DNCB‐positive patients was seen (P = 0.03). When the survival distributions for all DNCB‐positive and‐negative patients were compared, the DNCB‐positive patients showed a significantly longer survival (P < 0.01). However, when the survival distributions for DNCB‐positive and‐negative patients with either primary operable or advanced breast cancer were compared separately, significant differences were not seen. Other tests of immune function, including intradermal skin tests with microbial antigens, absolute lymphocyte counts, and lymphocyte responses to mitogens in vitro, were not useful in distinguishing prognostically favorable groups among patients with advanced disease. Advanced breast cancer patients who were given an adequate trial of combination chemotherapy showed no difference in response rate or survival when DNCB‐positive and‐negative patients were compared. We conclude that, although DNCB reactivity is progressively impaired in patients with increasing tumor burden and correlates with survival in breast cancer patients in general, in our experience such tests do not provide prognostically important information above that given by careful clinical and pathologic staging.

Intralesional injection of the methanol extraction residue of Bacillus Calmette-Guerin (MER) into cutaneous metastases of malignant melanoma.

Twenty‐two patients with cutaneous metastases of malignant melanoma were treated with intralesional injections of the methanol extraction residue of bacillus Calmette‐Guerin (MER). The local reaction consisted of erythema and pustule formation followed by ulceration and tumor necrosis. Side effects included fever, chills, headache and malaise in the majority of patients; nausea, vomiting, cyanosis and hypotension occurred infrequently. Hypersensitivity reactions were not observed. Temporary abnormalities in liver function were seen in 11 of 19 patients tested. Reversible lymphopenia and thrombocytopenia developed in 7 of 17 and 7 of 18 patients, respectively. Immune function, as measured by skin tests for delayed hypersensitivity and the in vitro response of isolated lymphocytes to mitogens and microbial antigens, was not influenced by treatment with MER. Transient increases were observed in total hemolytic complement, complement components and the reduction of nitroblue‐tetrazolium by neutrophils. Eight of eighteen evaluable patients showed a complete disappearance of all injected lesions. We conclude that intratumoral injection of MER is effective treatment for cutaneous metastases of malignant melanoma, with a complete response rate comparable to that observed after intralesional injection of BCG. Cancer 42:2648–2660, 1978.

Surgical adjuvant therapy of malignant melanoma with Corynebacterium parvum

The authors previous surgical adjuvant trial in patients with malignant melanoma at high risk of recurrence has shown no difference in disease‐free interval or survival between patients randomized to surgery + BCG or surgery alone. Reported here is a subsequent nonrandomized trial in 30 similar patients who received surgery + Corynebacterium parvum (CP) 4 mg I.V. daily × 5, followed by 4 mg S.C. weekly for up to three years. After I.V. C. parvum, chills, fever, headache, and hypertension were common. After S.C. C. parvum, varying degrees of local induration, erythema, and pain were experienced. Dose reduction was necessary for 14 patients during I.V. treatment and for six patients during S.C. treatment. A marked decrease in absolute lymphocyte count and a decreased proliferative response of lymphocytes to common antigens in vitro was observed after 2–3 days of I.V. C. parvum. Lymphocyte reactivity to mitogens decreased, particularly with Con A. Marked increase in nitroblue tetrazolium reduction by granulocytes was seen in 20 patients. Although changes in delayed cutaneous hypersensitivity reactions to recall antigens followed no consistent pattern, reactivity to DNCB increased in 18 patients. In addition, median time to recurrence was 33 weeks, significantly shorter than in the previous trial, but the survival distribution was no different from before. It can be concluded, therefore, that the administration of C. parvum in this dose and schedule had essentially no effect on the outcome of these patients.

Reappraisal of surgical management of sarcoma of the buttock

In an anatomic and regional consideration of sarcoma of the buttock involving 71 patients, 52 had operable tumors and 19 were inoperable. Radical buttectomy was done in 36 patients and hemipelvectomy was performed in 16 patients. Seventeen patients in the buttectomy group and 14 in the hemipelvectomy group were operated on for recurrent sarcoma secondary to local resection done elsewhere. Patients requiring hemipelvectomy had more extensive tumors or had sarcoma deeply invading into bone or sciatic nerve. Survival was significantly prolonged in the buttectomy group with 21 patients living 5 years and 14 patients surviving 10 years. In contrast, only 3 of 16 hemipelvectomy patients survived 5 years and 1 survived 10 years. The rate of local recurrence was higher in the buttectomy group, but the incidence of pulmonary metastases was higher in the hemipelvectomized patients. In both groups, the presence of recurrent sarcoma was the most important factor influencing local recurrence, metastases, and survival after surgery. The high rate of recurrence after local excision necessitating buttectomy or hemipelvectomy suggests that the operation which may finally become necessary in the management of buttock sarcoma is the one which should be done in the initial management of this tumor.

Adjuvant trial of levamisole in patients with squamous cancer of the head and neck: a preliminary report.

A preliminary report is made of a randomized double-blind study comparing levamisole with placebo as surgical adjuvant treatment of patients with squamous cancer of the head and neck. The study groups were comparable according to the prognostic factors of type (primary or recurrent), site (oral cavity, pharynx, larynx), and stage of disease and were also similar in pretreatment immune function and in radiation exposure. Of 65 evaluable patients, 31 were treated with levamisole and 34 received placebo. Although there was no difference in the distribution of the time to recurrence of the overall treatment group, there was a difference of borderline significance in favor of levamisole in the primary disease patients (P less than 0.006). Further analysis of subgroups in the primary disease category showed decreased recurrence rates in patients receiving levamisole who had cancer of the oral cavity P less than 0.01 or who had stage II disease, P less than 0.02. Considering site and stage simultaneously, the only significant difference was in stage II patients with oral cancer. The above results, though encouraging, must be viewed with caution due to small sample sizes and relatively short follow-up of most patients.

DNCB reactivity and prognosis in 419 patients with malignant melanoma

Delayed cutaneous hypersensitivity to 2,4‐dinitrochlorobenzene (DNCB) was tested in 419 patients with malignant melanoma to determine whether DNCB reactivity was associated with prognosis. At the time of definitive surgery, a positive DNCB test was seen in 82% of the patients in Stage I (regional lymph node histologically negative) and in 81% of patients in Stage II (regional lymph node positive). A positive DNCB test, obtained at the time of staging in patients with more advanced disease, was seen in 70% of patients in Stage III (metastases in, at most, one internal organ) and in 68% of patients in Stage IV (metastases in more than one internal organ). Using the x2 test, the trend of a decrease in DNCB reactivity from Stage I to Stage IV is significant (P < 0.04). In regard to DNCB reactivity and prognosis, the following statistical differences (P < 0.05) were obtained: In Stage II, the recurrence distribution for DNCB‐positive patients is more favorable than for DNCB‐negative patients (estimated median time to recurrence 16 vs. 7 months); the survival distribution again in Stage II was more favorable for DNCB‐positive patients than DNCB‐negative patients (estimated survival time 31 vs. 18 months). When the group of Stage II patients was divided into subgroups with different prognoses (state of primary at the time of lymphadenectomy), a survival advantage was seen in all substages but was statistically significant (P < 0.05) only in Stage II d (regional lymph node involvement diagnosed more than two months after resection of primary melanoma). The results of DNCB tests in patients with malignant melanoma correlate with stage of disease and, in Stage II, with prognosis.

Neoplastic disease and thyrotoxicosis.

The relationship of thyrotoxicosis and malignancy has been studied in a cancer center. There was a lower incidence of cancer in patients with hyperthyroidism (25%) than in admissions generally (59.8%) but this probably represents a sampling difference. Of 300 women with thyrotoxicosis, 83 had neoplastic disease. Of the women with hyperthyroidism and cancer (excluding thyroid and skin malignancies), 56% had breast primaries. This is significantly different from the 36% incidence of breast cancer among all women with malignant neoplasms at the center. The incidence of hyperthyroidism was only 0.25% in females with breast cancer but the diagnosis of both diseases was made in 21 of 40 women within the same year. Such a time‐linked occurrence is statistically significant. No effect on survival was demonstrated. The significance of these findings is discussed.

Severe hypoglycemia associated with terminal lymphomas: Report of five cases

Five patients with lymphoma developed severe hypoglycemia in the terminal stages of the disease. There were several possible causes for the hypoglycemia in these patients, including severe liver disease, malabsorption, inanition and gastro‐intestinal ulceration. Characteristics were present which were similar to nonpancreatic tumors associated with hypoglycemia. The authors suggest that hypoglycemia may be more frequent than commonly is recognized and should be searched for in patients with lymphomatous disease.