No abstract

Objective. To develop diagnostic criteria for a familial form of antiphospholipid antibody syndrome (APS), identify families with >1 affected member, examine possible modes of inheritance, and determine linkage to potential candidate genes.Methods. Family members of probands with primary APS were analyzed for clinical and laboratory abnormalities associated with APS. Families with >2 affected members were analyzed by segregation analysis and typed for candidate genetic markers.Results. Seven families were identified. Thirty of 101 family members met diagnostic criteria for APS.Segregation studies rejected both environmental and autosomal recessive models, and the data were best fit by either a dominant or codominant model. Linkage analysis showed independent segregation of APS and several candidate genes.Conclusion. Clinical and laboratory criteria are essential to identify the spectrum of disease associated with APS. We believe a set of criteria was developed that can precisely define affected family members with APS. Modeling studies utilizing these criteria strongly support a genetic basis for disease in families with APS and suggest that a susceptibility gene is inherited in an autosomal dominant pattern. However, in these families, APS was not linked with HLA, Fas, or other candidate genes, including ␤ 2 -glycoprotein I, HLA, T cell receptor ␤ chain, Ig heavy chain, antithrombin III, Fas ligand, factor V, complement factor H, IgK, and Fas.

Between 1971 and 1984, 22 of 190 adult patients (11.6 per cent) with acute leukemia seen at the University of Arizona had hypocellular acute leukemia (HAL), defined as lymphoblasts or myeloblasts (plus atypical promyelocytes) of greater than or equal to 30 per cent, but marrow cellularity of the core biopsy or clot section of less than or equal to 50 per cent based on a 1000 point count. These 22 patients with HAL plus the 48 previously published patients with well documented HAL (combined series of 70 patients) were evaluated in detail with multivariate analysis. The median leukocyte count was 2700/microL, hemoglobin of 8.2 g/dl, and platelet count 63,000/microL. Circulating blasts were noted in 27 of 52 patients (52 per cent). Twenty-seven of 34 patients (79 per cent) had abnormal cytogenetics. The overall median survival was 8 months (range: 0.1-48). The median survival for the 22 patients managed with supportive care alone was 4 months, 6 months for the 16 patients treated with non-aggressive induction therapy, and 13 months for the 32 patients treated with aggressive induction therapy (p less than 0.02 versus other categories). Multivariate analysis confirmed that aggressive induction therapy was a major favourable prognostic factor (p = 0.016). Multivariate analysis of the aggressively induced patients revealed that younger patients (less than or equal to 65; p = 0.04) and patients with no AHD (p = 0.09) lived longer. Thus, aggressive remission induction can be attempted in HAL and appears to contribute to prolonged survival especially under age 65 years.

The authors have studied the natural history of 377 patients with Stage I cutaneous malignant melanoma followed at the Arizona Cancer Center, Tucson. Two hundred eight patients, or 55%, remained free of metastatic disease after a median follow-up of 30 months. The survival at 5, 8, and 10 years was 69, 65, and 63%, respectively. Natural breakpoints in Breslow thickness for survival occurred at 0.85, 1.95, and 4.00 mm. These are not significantly different from those found by other investigators. A step-down multivariate analysis using the Cox regression model yielded four factors as highly significant in predicting survival: Breslow thickness (P less than 0.001), an age/sex interaction (P = 0.0012), clinical ulceration (P = 0.0039), and a prophylactic node dissection (P = 0.019). No predictive value for a BANS or non-BANS location was detected. These results are discussed in reference to other large series which describe the natural history of cutaneous melanoma.

The survival history of 259 patients with Stage I cutaneous malignant melanoma who were at risk for developing regional nodal metastases (Stage II) were studied. Eighty-seven of 377 Stage I patients (23%) developed regional nodal metastases (Stage IIB) with 40% 5-year survival. Fifty patients had regional nodal metastases at presentation, with or without a known primary (Stages IIA or IIC, respectively), with a 42% 5-year survival. A step-down multivariate analysis using the Cox regression model revealed four risk factors as being highly significant for predicting a more favorable survival outcome: (1) thinner Breslow thickness (P = 0.0001), (2) pathologic Stage I disease (P = 0.004), (3) no clinical ulceration (P = 0.0004), and (4) being a woman younger than 50 years of age (P = 0.029). These results are discussed in reference to other series.