Background: ACT for NSCLC improves survival, but the benefits are modest and must be weighed against the harms of treatment. We determined the survival benefits judged necessary to make ACT worthwhile for patients with resected NSCLC, and their doctors (surgeons and medical oncologists).

Methods: A self-administered questionnaire used the time trade-off method to determine the minimum survival benefits judged sufficient to make ACT worthwhile in 4 hypothetical scenarios. Baseline survival times were 3 and 5 years, and baseline survival rates (at 5 years) were 50% and 65%. All tests were 2-sided and non-parametric. Baseline responses of the first 60 patients (before ACT) and 78 doctors are reported.
Results: Most patients were male (53%) with a median age of 66 years (range, 43-78 years), had a lobectomy (83%), adenocarcinoma histology (55%) and stage II disease (53%), and received 4 cycles (74%) of cisplatin/vinorelbine chemotherapy (75%). The median benefit judged sufficient (by 50% of patients) was 9 months beyond 3 years, 6 months beyond 5 years, and 5% beyond 50% and 65%. Preferences varied across the entire range of possible benefits: from an extra 0 days to 15 years (IQR, 1 month-1 year) and 0% to 50% (IQR, 1%-10%). Patients’ preferences were not strongly associated with any baseline characteristics. The median benefit doctors judged sufficient was 9 months beyond 3 years and 5 years and 5% beyond 50% and 65%. Compared to patients’ preferences, doctors’ preferences did not differ in the median benefit (p>0.1 for all 4 scenarios) and varied less (IQR, 6 months-1 year and 5%-10%).

Conclusion
Most patients and their doctors judged moderate survival benefits necessary to make ACT for NSCLC worthwhile. Larger benefits were judged necessary for ACT by patients with NSCLC (before ACT) than by patients with breast cancer and colon cancer who had experienced ACT, in previous studies.