Materials & Methods:35 patients with unresectable esophaqeal cancer were treated with radiotherapv between June 1985 and June 1986 at Hiroshima University‘Hoipital. During this short period, combination treatment of radiation and chemotherapy (CMP;cisplatinum 50mg/m2, methotrexate 40mq/m2, peplomvcin 5mgx5) was the first choice in case of-certain eligible criteria. Only 19 cases were-matched 'to -the criteria and were treated with radiotherapy and chemotherapy (CMP group). The mean numbers of CMP courses were 1.8. Out of uneligible cases for CMP , rapy (PM;peplomycin 5mgx6, mitomycin 10mg) and 6 were treated with radiotherapy and mild chemothe10 were treated with radiotherapy alone (non-CMP group). Mean total radiation dose were 56.3Gy for CMP group and 60.6Gy for non-CMP group.

A multi-institutional prospective study for the analysis of prognostic factors for patients with osseous metastasis was performed. From February 1986 through June 1988, a total of 216 patients were included in this study. Cox's regression model made it clear that the most significant overall prognostic factor was primary site (p = 0.0002). In the lung cancer group, performance status (p = 0.0036) and metastasis of organs than bone (p = 0.0105) were also significant prognostic factors. In the breast cancer group, no significant factors were obtained. In the hepatoma group, the values for alkaline phosphatase (ALP) (p = 0.0021), lactate dehydrogenase (LDH) (p = 0.0195), and sex (p = 0.0264) proved significant. In the group of other cases, the most significant prognostic factor was the value for urinary hydroxyproline/creatinine ratio (p = 0.0001), followed by the pain score of RTOG (p = 0.0018). These factors and actual survival periods obtained in this study will be useful for the future stratification of patients for individualized optimal radiation schedules.

A prospective randomized study for the analysis of optimal schedule for the patients with bone metastases was performed. From February 1986 through January 1987, a total of 131 patients entered on this trial. The data from 129 patients out of 131 were available for the analysis of prognostic factors by Cox's regression model. As a whole, it was clear that first significant prognostic factor was urinary hydroxyproline/creatinine ratio (p = 0.0001), second primary site (p = 0.0001), third pain score (p = 0.0005) and fourth other organ metastasis except bone (p = 0.0006). In the group of lung cancer, first significant prognostic factor was urinary hydroxyproline/creatinine ratio (p = 0.0007), second multiplicity of bone metastasis (p = 0.0113), and third pain score (p = 0.0270).

During the period between 1972 and 1978, 226 patients were operated on for carcinoma of the uterine cervix at Hiroshima University, and 91 patients were treated by postoperative external irradiation. Patients with lymph node metastasis showed markedly lower five-year survival rates stage I: 44.4%; stage II: 45.0%) than those without lymph node metastasis (99.3% and 84.0%, respectively). In view of benefits of postoperative radiotherapy in the treatment of cervical cancer, only to the node positive and tumor rest patients is recommended.