The prognostic significance of disease features recorded at the time of diagnosis was examined among 813 patients with Philadelphia chromosome- positive, nonblastic chronic granulocytic leukemia (CGL) collected from six European and American series. The survival pattern for this population was typical of “good-risk” patients, and median survival was 47 mo. There were multiple interrelationships among different disease features, which led to highly significant correlations with survival for some that had no primary prognostic significance, such as hematocrit. Multivariable regression analysis indicated that spleen size and the percentage of circulating blasts were the most important prognostic indicators. These features, and age, behaved as continuous variables with progressively unfavorable import at higher values. The platelet count did not influence survival significantly at values below 700 X 10(9)/liter but was increasingly unfavorable above this level. Basophils plus eosinophils over 15%, more than 5% marrow blasts, and karyotypic abnormalities in addition to the Ph1 were also significant unfavorable signs. The Cox model, generated with four variables representing percent blasts, spleen size, platelet count, and age, provided a useful representation of risk status in this population, with good fit between predicted and observed survival over more than a twofold survival range. A hazard function derived from half of the patient population successfully segregated the remainder into three groups with significantly different survival patterns. We conclude that it should be possible to identify a lower risk group of patients with a 2-yr survival of 90%, subsequent risk averaging somewhat less than 20%/yr and median survival of 5 yr, an intermediate group, and a high- risk group with a 2-yr survival of 65%, followed by a death rate of about 35%/yr and median survival of 2.5 yr.

In the United States, there are an estimated 5000 to 6000 new patients annually who might be candidates for major hepatic resection to treat their recurrent colon cancer. Since 1971, the program reported here has evaluated various factors that might influence the curative potential of such an approach. Sixty-five patients had a major hepatic resection from March 1971 through May 1982. Using a stepwise proportional hazard analysis, all data that had been stored in CLINFO (a data analysis system by Bolt, Beranek and Newman; Boston, MA) were evaluated for the effect of multiple variables on the survival of patients with resected hepatic metastases. Twenty-seven had a right hepatic lobectomy; 14 had extended right hepatectomy with one having the caudate lobe also removed; ten had left lobectomy, nine had left lateral segmentectomy; and five had a major hepatic resection with three-dimensional wedge excision of a metastatic deposit in the contralateral lobe. The 30-day operative mortality rate was 7% (4/58) for patients undergoing the standard major hepatic resection. It was 14% for seven patients in whom the isolation-hypothermic perfusion technique was used early in the series. In ten patients, wedge excision only was required to remove the tumor. Stage I disease is defined as tumor confined to the resected portion of the liver without invasion of major intrahepatic vessels or bile ducts. Stage II disease is regional spread and Stage III disease is metastasis to lymph nodes or extraregional sites. The 3-year survival estimate was 66% for the 37 patients with Stage I disease. The 3-year survival estimate for 13 patients with Stage II disease was 58%. Five of the nine patients with Stage III disease are presently alive from 3 to 23 months; one of the other four died at 35 months of disease. The stage of liver disease was the most significant variable in this survival analysis (p = 0.02); Dukes' classification of colorectal primary was significant at p less than 0.05. Those factors found not to be significant determinants of survival were: number of metastatic hepatic deposits, site of colon primary, age, sex, preoperative liver function tests, and CEA.

Confusion exists concerning the influence of pregnancy on survival in patients with malignant melanoma. To evaluate this problem a retrospective computer‐aided study was performed of women in the child‐bearing years treated for Stage I cutaneous melanoma at the Duke University Comprehensive Cancer Center. Fifty‐eight women were identified who had melanoma arise during pregnancy (Group 1) and 43 patients were noted who became pregnant within 5 years of diagnosis of their melanoma (Group 2). Appropriate control groups matched for the clinical variables of age, primary site, and stage of disease and the pathologic variables of Clark's Level, tumor thickness, ulceration, and histologic type were selected from the cohort of 2938 melanoma patients seen at Duke. Actuarial survivals for Group 1 and 2 patients did not differ from their respective controls, although the small number of deaths in each group resulted in wide confidence intervals. When actuarial disease‐free intervals were plotted, there was a significant difference beween women who had melanoma develop during pregnancy when compared to their controls (P = 0.04). In a multivariate regression analysis, after adjustment for the influence of the more significant prognostic factors for Stage 1 melanoma, including Clark's Level, ulceration, and tumor thickness, the effect of pregnancy on disease‐free interval became more apparent (P = 0.02). No difference in actuarial disease‐free interval was noted in the melanoma patients who elected to become pregnant within 5 years of diagnosis (P = 0.31). A multivariate regression analysis confirmed this finding. These data indicate that although an intercurrent melanoma during pregnancy has a worse prognosis than the control groups, once a woman has been diagnosed as having a cutaneous melanoma, a subsequent pregnancy has no effect on recurrence rate or survival. Cancer 55:1340‐1344, 1985.

Twenty-six patients with known or suspected medullary thyroid carcinoma (MTC) and 21 normal control subjects were tested intravenously on four separate days with calcium gluconate (CG), 2 mg Ca(++)/kg/1 min.; pentagastrin (P), 0.5 ug/kg/ 5 sec.; calcium chloride (CC), 3 mg Ca(++)/kg/10 min.; and a combination of calcium gluconate and pentagastrin (CG + P). Calcitonin (CT) levels were determined by radioimmunoassay on plasma collected before and immediately following each test infusion. In none of the 21 control subjects was there a clear increase in CT above 200 pg/ml following any of the four provocative tests. Conversely, in all 26 patients with known or suspected MTC, plasma CT levels were markedly increased (>300 pg/ml) following the combined infusion of CG + P. The peak CT response was greater with CG + P than with a) CG alone (22 of 24 patients, p < 0.002), b) P alone (25 of 26 patients, p < 0.002), or c) CC alone (17 of 17 patients, p < 0.002). Of 12 MTC patients with undetectable basal calcitonin levels, all had peak responses greater than 300 pg/ml following CG + P, whereas such responses occurred less often following CG alone (8 of 12) or P alone (8 of 12). The results demonstrate that the combined administration of pentagastrin and calcium gluconate constitutes a more effective and reliable stimulus for CT secretion from MTC cells than the use of either agent alone, and appears the most useful single screening test for the detection of occult MTC.

Eighty-eight patients with hormone-resistant Stage IV prostate cancer were treated with a five-drug chemotherapy program. Patient demographic data, prior therapy, symptoms, extent of disease, and laboratory studies were analyzed statistically to evaluate the association of these parameters with survival from the onset of chemotherapy. Factors associated with short survival included age greater than 65, severe bone pain, poor performance status, presence of soft tissue metastases, anemia, elevation of serum LDH, SGOT, alkaline and acid phosphatases, and prolactin, and hypoalbuminemia. Race, stage at initial diagnosis, prior radiation therapy, prior orchiectomy, and elevation of CEA had no prognostic association. We suggest that clinical trials of new therapies of hormone-resistant prostate cancer take into account the presence of these prognostic factors in the analysis of the results of therapeutic programs.

To determine the prognostic significance of histologic variables in oligodendroglial neoplasms, the presence and degree of 15 such variables were correlated with postoperative survival rates in 71 patients. By univariate analysis, prognostically significant factors, in order of decreasing importance, were mitoses (log), necrosis, nuclear cytologic atypia, vascular hypertrophy, and vascular proliferation. When studied by stepwise regression, necrosis and the number of mitoses contained all of the prognostically useful information. When each of the five variables significant by univariate analysis was tested in the Cox model by adding a variable to the model containing the other four, necrosis was found to be the only independently significant variable. There were significant positive pairwise correlations between each of the five significant histologic variables except for cytologic atypia with necrosis. The only histologic variable with a significant association with older age was the number of mitoses. These results suggest that necrosis and, to a lesser extent, the mitotic count are features that, in the appropriate setting, can be used to identify the "anaplastic" oligodendroglioma.

Cystosarcoma phyllodes is an unusual breast neoplasm that rarely metastasizes. Most series report chemotherapy, radiation, and hormonal therapy to be ineffective. Three patients were treated with cisplatin and etoposide combination chemotherapy with effective palliation in two patients. Radiation therapy was effective in controlling symptomatic metastasis in all three patients. Hormonal therapy was ineffective in two patients despite the presence of positive hormone receptors. Chemotherapy and radiotherapy may be more effective in the treatment of this tumor than has been reported, although there is no apparent role for hormonal therapy. Functional hormone receptors are probably not present.

No abstract