Prognostic impact of comorbidity in elderly lung cancer patients: use and comparison of two scores

Background: Mean age of patients with lung cancer rises as a result of increasing life expectancy. So, the proportion of patients with serious comorbidity also increases [1,2]. Lung cancer treatment is characterized by a narrow therapeutic index. When life expectancy is short and therapeutic benefit is limited, it is of paramount importance to know the specific cause of death. Comorbidity is understood as a competing cause of death, and is the main exclusion criterion for lung cancer clinical trials. The aim of this study was to determine the prevalence of comorbidity in elderly lung cancer patients seen in an outpatient oncology department and to determine its correlation with survival.

Patients and methods: Between January 2006 and February 2008, 83 untreated lung cancer patients over the age of 70 years were enrolled in the study. Comorbidity was evaluated according to the Charlson comorbidity index (CCI) [3] and the simplified comorbidity score (SCS) [4].

Results: 83 patients (97.6% men, mean age 77 years) were studied. Comorbidities: tobacco consumption (94.6%), cardiovascular diseases (65%), and chronic obstructive pulmonary disease (COPD) (59%). Mean CCI was 3 (range 0-9). Mean SCS was 9 (range 4-19), and 47% of patients had an SCS>9. Comorbidity was fairly well correlated with age, ADL, IADL, and stage. Neither the CCI nor the SCS was related to survival (p: 0.47 and p: 0.24, log rank, respectively). Median survival was 326 days (95% CI, 259-393 days; or 10.8 months, 95% CI 8.6-13.1 months). Main cause of death was lung cancer disease progression (69.5%, 57 patients), with 20 patients (25%) dying of other non-neoplastic causes. Stage was significantly associated with survival (log rank: p<0.001).

Conclusions: Although there was a high prevalence of comorbidity in our population, comorbidity was not related to survival. Comorbidity is one of the main reasons for undertreatment of elderly lung cancer patients, but this study indicates that this undertreatment may not be warranted given that those comorbidities may not cause a patient's death. Our data generated more of a hypothesis than a conclusion. Comorbidity should be an impetus for treatment design instead of an exclusion criterion for oncologic treatment.