7549 Background: Little is known about patterns of tx for elderly w/ LA NSCLC ( IIIa, IIIb w/o effusion). Combined modality tx (CMT) (chemotherapy, C and radiation, R) is indicated for “fit” patients (pts) and is potentially curative. We examined population based tx patterns. Methods: Medicare pts ≥66 years with LA NSCLC who did not undergo surgery from 1997 to 2002 were identified in SEER registries. Medicare claims provided tx-related data. Multinomial logit models examined tumor and patient characteristics associated with tx (none, C only, R only, C+R). Of those who initiated C + R w/in 6 mo, we determined whether it was sequential (either C then R or R then C: C><R) or concurrent (C/R). We evaluated effects of claims-based indicators of poor performance status (PS) (eg. hospital stay, O2 or wheelchair use) prior to diagnosis. Results: 31.7% of 6,325 LA NSCLC pts received no tx; 10.0% C only; 28.8% R only; 10.2% C><R and 19.3% C/R (29.5% C+R). Platinum regimens accounted for 84.4% overall, and 76.6%, 84.4% and 88.5% of pts in the C only, C><R, and C/R groups. Single, non-platinum drugs were more common in the C only group (16.4%) compared to the C><R (12.3%) and C/R (9.8%) groups. Multivariate analysis: sex, age, marital status, substage (IIIa vs. IIIb), CHF, CVD, poor PS, consultation with medical oncologist, and Medicaid were significant in choice of therapy. Increasing age was associated with R only and reduced use of C only or with R. Married pts or w/ oncologist visit were more likely to receive any tx, particularly C and C+R. Pts w/ prior CHF and CVD were less likely to receive C+R. Poor PS was associated with no tx. Pts w/ potential access issues, e.g. rural residence were less likely to receive R only. There was no influence of race or income. Over time there has been reduced use of R only, with increasing use of C only, C + R, and no tx. Conclusions: 1. Most elderly pts do not receive CMT. 2. Comorbidities result in single modality therapy. 3. Claims-based indicators of poor PS are significant predictors; further research is warranted. 4. Advanced age is independently associated with reduced use of C+R. While this may reflect unmeasured comorbidity, poor PS, or preference, it may also reflect physician bias not to treat the fit elderly. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration sanofi-aventis sanofi-aventis sanofi-aventis sanofi-aventis sanofi-aventis