The relationship between the level of retinoblastoma protein (RB) expression and the survival of 113 newly diagnosed acute myelogenous leukemia (AML) patients was studied. Western blotting was used to determine the level of RB protein present in peripheral blood leukemia cells and results were confirmed in 26 patients by immunohistochemistry. The leukemic cells from 22/113 AML patients (19%) contained RB protein at levels that were equal to or less than the level of RB observed in the mononuclear cell fraction of peripheral blood from normal individuals (Low RB). Levels of RB greater than that of normal blood (Elevated RB) were seen in 91 patients (81%). The median survival of patients with low RB was significantly shorter than that seen in patients with elevated RB, 12 weeks versus 40 weeks (P = .02). Remission induction frequency was 36% in low RB patients compared with 68% in AML patients with elevated RB (P = .01). Multivariate analysis showed that low RB protein level was an independent prognostic factor predictive or poor survival after allowing for other known prognostic factors. These data suggest that a low level of the RB protein at the time of diagnosis is associated with shortened survival in AML patients because of inferior response to conventional therapy. Monitoring of the RB level could identify a subgroup of AML patients with an extremely poor prognosis when treated with chemotherapy alone, who would be eligible for alternative therapeutic strategies.

We administered a combination of rubidazone, cytosine arabinoside, vincristine, and prednisone (ROAP) to 91 patients with acute myelogenous leukemia who were 50 yr of age or older. These patients had been identified in previous studies to be a group with a relatively poor prognosis. One-third of the patients had an antecedent hematologic disorder prior to treatment. Forty patients (48%) obtained a complete hematologic and clinical remission. A history of an antecedent hematologic disorder, male sex, and absence of Auer rods were adverse factors for achieving remission in this older population. More than half of the patients achieved remission in one course. The major cause of failure to obtain a remission was death due to infection, 40% of which were caused by fungi. Resistance to chemotherapy, although uncommon, was noted more frequently in patients with an antecedent hematologic disorder. Univariate and multivariate prognostic factor analysis was used to compare these patients with a historical control group treated with a program in which adriamycin was used instead of rubidazone (AdOAP). No significant difference in remission rate was detected. Cyclocytidine was used as a maintenance agent in this study, and while the median remission duration was only 37 wk, 30% of patients are expected to be in remission for 2 yr. Chemotherapy programs combining an anthracycline with cytosine arabinoside, given to older patients in similar fasion to younger patients will achieve remissions in one-half of a group of older patients. These remissions are of comparable quality to those of younger patients. Mathematical models derived from analysis of prognostic factors are of use in identifying patients likely to fail these programs who are in need of innovative approaches to treatment.

The risk of development of CNS leukemia was investigated in 153 adults with acute lymphocytic leukemia (ALL) who received systemic combination chemotherapy without CNS prophylaxis. Overall, 31 patients (20%) developed CNS leukemia after a median of 6 months of therapy; the estimated 1-year incidence of CNS leukemia was 21% (SE, 3.9%). Characteristics significantly associated with CNS involvement included the presence of elevated hemoglobin creatinine, alkaline phosphatase, fibrinogen, and lactic dehydrogenase levels; B-cell leukemia; and high leukemic cell proliferative activity. Multivariate analysis identified lactic dehydrogenase levels of greater than or equal to 600 U/L and greater than or equal to 14% of cells in the S + G2M compartment to have independent additive poor prognostic significance. Patients were categorized into different risk groups for CNS leukemia with 1-year incidences ranging from 4% to 55%. While related to a high occurrence of CNS leukemia at diagnosis (33%) and subsequently (100%), the low incidence of B-cell disease excluded it from the multivariate analysis. The use of systemic chemotherapy containing multiple agents with good CNS penetration and in high doses (VAD regimen) in 90 patients was associated with a trend for lower CNS leukemia at 1 year (15% v 31%), especially in the low-risk category. We propose to develop future therapies for adults with ALL that include risk-oriented CNS prophylactic approaches.

The records of 292 patients with malignant lymphoma other than Hodgkin's disease, registered in our protocols from 1967 to 1977, were reviewed to identify those with central nervous system (CNS) involvement. Thirty-one patients were encountered with this complication, an incidence of 11%. Patients with a diffuse histology had a higher frequency of CNS recurences (27/174 = 16%) in contrast to only 4/118 (3%) for those with nodular types. However, if only patients with diffuse histology in CR are considered, the frequency of CNS relapse is 13.5% (13/98). The risk factors that predict for the development of this complication were studied using multivariate analysis. Diffuse poorly differentiated lymphocytic and diffuse undifferentiated lymphomas were found to be associated with a high risk of CNS relapse. Prior chemotherapy, bone marrow involvement, age less than 35, and extranodal disease were also identified as high-risk factors. Using the information generated by a logistic regression model, patients with malignant lymphoma of diffuse type can be classified into three categories when first seen: low-risk group, intermediate, and high-risk group. CNS prophylaxis is recommended for the intermediate and high-risk group, while only close follow-up is advised for the low-risk group patients who have one adverse characteristic.

based tool, System-Theoretic Process Analysis (STPA), the system was modeled at multiple levels of abstraction. Interactions between system components were modeled as control loops. The analysis was conducted by a multidisciplinary team with expertise in systems engineering, medical physics, MR physics, and clinical radiation oncology. Results: A high-level system model in the form of a safety control structure (SCS) to enable MRI simulation was developed. Areas of inadequate control and feedback were identified at multiple levels (Table 1), including those beyond the frontline. 317 unsafe control actions (UCAs) were identified. Conclusion: The development of a SCS revealed areas of inadequate feedback and enabled the identification of over 300 UCAs, highlighting the complexity of the sociotechnical system surrounding MRI simulation. The majority of UCAs surrounded RT simulation and planning, aspects of the system with greater complexity. By identifying UCAs prior to clinical implementation, this analysis enabled a proactive approach to MRI safety. By utilizing abstraction, the model can be applied to other centers. Future work is needed to mitigate circumstances leading to UCAs.