Role of Comorbidity on Survival after Radiotherapy and Chemotherapy for Nonsurgically Treated Lung Cancer

Background: Comorbidity, such as diseases of the cardiovascular, pulmonary, and other systems, may influence prognosis in lung cancer and complicate its treatment. The performance status of patients, which is a known prognostic marker, may also be influenced by comorbidity. Due to the close link between tobacco smoking and lung cancer, and because lung cancer is often diagnosed in advanced ages (median age at diagnosis in Denmark is 70 years), comorbidity is present in a large proportion of lung cancer patients. Methods: Patients with any stage lung cancer who did not have surgical treatment were identified in the Danish Lung Cancer Registry. Danish Lung Cancer Registry collects data from clinical departments, the Danish Cancer Registry, Danish National Patient Registry, and the Central Population Register. A total of 20,552 patients diagnosed with lung cancer in 2005 to 2011 were identified. Comorbidity data were extracted from the Danish National Patient Registry, which is a register of all in-and outpatient visits to hospitals in Denmark. By record linkage, lung cancer patients who had previously been diagnosed with comorbid conditions were assigned a Charlson comorbidity index. Initial cancer treatment was categorized as chemotherapy, chemoradiation, radiotherapy, or no therapy. Data on Charlson comorbidity index, performance status, age, sex, stage, pulmonary function (forced expiratory volume in 1 second), histology, and type of initial treatment (if any) were included in univariable and multivariable Cox proportional hazard analyses. Results: Treatment rates for chemotherapy and chemoradiation declined with increasing comorbidity and in particular increasing age. Women received treatment more often than men. In a univariable analysis of all patients combined, stage, performance status, age, sex, lung function, and comorbidity were all associated with survival. Apart from excess mortality among patients with unspecified histological subtypes (hazard ratio), there was no clear difference between the specified subtypes. When adjusting for the other factors, particularly age, sex, performance status, and stage proved to be robust while risk estimates for comorbidity were attenuated somewhat. When grouped by the three types of cancer treatment or no treatment, there was no influence of comorbidity on radiation therapy and modest influence on survival after chemotherapy and chemoradiation. In contrast, age remained a strong negative prognosticator after multivariate adjustment as did stage and performance status. Conclusion: Comorbidity has a limited effect on survival and only for patients treated with chemotherapy. It is rather the performance of the patient at diagnosis than the medical history that prognosticates survival in this patient group.