These data suggest that pre-existing interstitial changes detected by chest radiography or computed tomography and radiotherapy to the contralateral mediastinum (> 40 Gy) may predict the development of severe radiation pneumonitis.

We investigated the usefulness of serum pro-gastrin-releasing peptide (Pro-GRP) as a tumor marker for diagnosis, treatment monitoring and the prediction of relapse and prognosis in patients with small-cell lung cancer (SCLC). Serum samples were obtained from 127 patients with primary lung cancer (48 patients with small-cell carcinoma, 31 with adenocarcinoma, 36 with squamous cell carcinoma and 11 with large-cell carcinoma). The cutoff levels of serum Pro-GRP and neuron-specific enolase (NSE) were set at 46 pg/ml and 10 ng/ml, respectively. The specificity of Pro-GRP was significantly higher than that of NSE (Pro-GRP: 93.7%, NSE: 65.8%, p < 0.01). According to the histological type of lung cancer, the positive rates of Pro-GRP were 75% (36/48) in the small-cell carcinomas, 9.7% (3/31) in the adenocarcinomas, 5.6% (2/36) in the squamous cell carcinomas and 0% (0/10) in the large cell carcinomas. The median levels of Pro-GRP in limited disease (LD) and extensive disease (ED) patients were 199 and 295.5 pg/ml, whereas those of NSE were 14.8 and 29.3 ng/ml, respectively. The positive rates of Pro-GRP in LD and ED patients were 80.0% (16/20) and 71.4% (20/28), whereas those of NSE were 70.0% (14/20) and 89.3% (25/28), respectively. The positive rate of NSE tended to elevate with the progression of disease, whereas that of Pro-GRP was already high at an early stage. Among the 29 patients with SCLC who could be followed, the serum Pro-GRP levels of 18 responders were significantly decreased after treatment (p < 0.01), whereas those of the 11 nonresponders were not significantly different between before and after treatment (p = 0.72). In the 9 patients with SCLC who relapsed, the serum Pro-GRP levels were again elevated at the time of relapse. Seventeen patients whose ratio of the Pro-GRP level after treatment to the level before treatment was below 50% (taking the levels before treatment as 100%) survived significantly longer than did the patients whose ratio was over 50% (p < 0.01). The results of the present study suggest that serum Pro-GRP has high specificity and could be a useful marker of SCLC for treatment monitoring and prognosis.

CYFRA 21-1 is a new tumor marker using two different monoclonal antibodies which recognize the divergent epitope on the N- or C-terminal region of domain 2 of cytokeratin 19 fragment, respectively. In this study, we investigated the relationship between levels of CYFRA 21-1 and survival duration, as well as the efficacy of chemotherapy associated with changes in CYFRA 21-1. Serum samples were obtained from 87 patients with nonoperable lung cancer (35 cases with squamous-cell carcinoma, 33 with adenocarcinoma, 3 with large-cell carcinoma, and 16 with small-cell carcinoma). The cutoff point was set at 3.5 ng/ml. In a CYFRA 21-1 assay, significantly more patients with squamous-cell carcinoma and adenocarcinoma were positive compared to patients with small-cell and large-cell carcinomas (p = 0.0017). Following chemotherapy, blood levels of CYFRA 21-1 decreased significantly in responders versus nonresponders (p = 0.0246). A significant correlation was noted between survival periods and pretreatment levels of CYFRA 21-1 (p = 0.0036). The present study suggests that CYFRA 21-1 might be useful as a possible indicator of survival and therapeutic effect for lung cancer.

It is reported that the combination of cisplatin (CDDP) and carboplatin (CBDCA) is synergistic in vitro. The objective of this study was to evaluate the therapeutic effect and safety of the two platinum compounds in combination with etoposide in the treatment of non-small-cell lung cancer (NSLC). Forty patients were registered. Based on the results of a phase I study, patients were treated with CDDP (80 mg/m2 i.v. on day 1), CBDCA (280 mg/m2 i.v. on day 1), and etoposide (80 mg/m2 i.v. on days 1-3). Of the 40 patients, 30 were men and 10 women. Histology revealed adenocarcinoma(AC) (n = 20), squamous cell carcinoma(SCC) (n = 18), and large cell carcinoma(LCC) (n = 2). Staging: IIIA (n = 3); IIIB (n = 17); and IV (n = 20). A 32.5% overall response rate [13 of 40; 95% confidence interval (CI) 18-47%] was achieved. The response rates in patients with SCC and AC were 55.6 and 10.0% (p < 0.005), respectively. The median duration of response was 47.1 weeks and the overall median survival time was 57.1 weeks. Leukopenia and thrombocytopenia--World Health Organization (WHO) grade IV--occurred in nine and 11 patients, respectively. Nonhematological toxicities were mainly nausea, vomiting, and alopecia. In conclusion, further investigations of this regimen are warranted in the treatment of NSLC.

A combination chemotherapy consisting of a 72-hour continuous infusion of cisplatin (CDDP; 100 mg/m²/72 h) and 5-fluorouracil (5-FU; 3 g/m²/72 h) was conducted for inoperable non-small cell lung cancer (NSCLC). Sixty patients were accepted for this study between June 1988 and December 1990. Forty-seven patients were male (median age 68). Thirty-four patients had stage III and 26 had stage IV disease. The response rate was 25.0% (95% confidence interval, CI, 14.0–36.0%), median survival was 15.7 months. In squamous cell carcinoma, the response rate was 35.5% (95% CI, 18.7-52.3%) and median survival was 15.1 months. In non-squamous cell carcinoma, the response rate was 13.8% (95% CI, 1.2-26.4%) and median survival was 17.7 months. There was a significant difference in response rate (p < 0.01), but no significant difference in survival (p = 0.36). Grade 3 leukopenia was 11.7%, grade 3 and 4 thrombocytopenia was 13.3%. Grade 3 nausea and vomiting were 8.3%. One patient had grade 4 renal toxicity. However, there was no treatment-related death. This regimen was well tolerated. In multivariate analysis, the significant parameters were CEA, performance status and response. In conclusion, 72-hour continuous infusion of CDDP and 5-FU for treatment of NSCLC provides similar response and toxicity as previously reported regimens using CDDP.