Predictors of radiation response in lung cancer. A clinico-pathobiological analysis.

From February 1972 to July 1975, 200 lung cancer patients were seen at the University of Rochester Cancer Center's Division of Radiation Oncology; 40% had squamous cell tumors and 87.5% had advanced disease localized to the thorax. Of the 160 patients who completed treatment, 101 were treated with continuous therapy schedules, and 59 were treated with split-course schedules, and 59 were treated with split-course schedules; 40 patients did not complete treatment because of early metastatic disease or death. Radiation therapy was very effective in local tumor ablation. To assess local tumor response, doubling times were obtained in measurable lesions prior to treatment. The doubling times (DT) were 25 days for small cell cancers and 192 days for adenocarcinoma. More than 50% tumor shadow regression was a good prognosticator of local tumor response; this increased as the mean DT decreased. The order of kinetic increase in tumor ablation per histology was the opposite of the one-year survival results because of the metastatic spread patterns of the different tumors. Survival rates in lung cancer emerge as simplistic and inadequate to explain local radiation effectiveness. Survival is conditioned by stage, histology, and modality of treatment, total dose delivered, and local tumor response. Although the most effective treatment dose seems to be over 6000 rads, the most efficient schedules were split-course delivering lower tumor doses. This modality of treatment is proposed as the optimal schedule to be combined with other forms of therapy with the goal of achieving better survival.